The 1999–2000 Elections in Russia Their Impact and Legacy A distinguished collection of specialists analyze the critical...
17 downloads
453 Views
1MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
The 1999–2000 Elections in Russia Their Impact and Legacy A distinguished collection of specialists analyze the critical elections that ushered out the Boris Yeltsin era in Russia and ushered in the leadership of Vladimir Putin. These parliamentary and presidential elections are proving critical for the future of Russia and are highly enlightening to scholars and students who study electoral politics, party development, and democratization. Collectively, the expertise represented by these authors extends to all the important facets of electoral politics and party development in Russia. Contributors include Timothy J. Colton (Davis Center for Russian Studies, Harvard University); M. Steven Fish (Department of Political Science, University of California, Berkeley); Stephen E. Hanson (Department of Political Science, University of Washington); Andrei Melville (Faculty of Politology, Moscow State Institute for International Relations); Neil Munro (Center for the Study of Public Policy, University of Strathclyde); Mikhail Myagkov (Department of Political Science, University of Oregon); Sarah Oates (Department of Politics, University of Glasgow); Thomas F. Remington (Department of Political Science, Emory University); Richard Rose (Department of Government, University of Strathclyde); Richard Sakwa (Department of Politics and International Relations, University of Kent); Olga Shvetsova (Department of Political Science, Washington University); and Stephen White (Department of Politics, University of Glasgow).
Vicki L. Hesli is Professor of Political Science at the University of Iowa. Her most recent book is Medical Issues and Health Care Reform in Russia (1999). William M. Reisinger is Professor of Political Science at the University of Iowa. He is the author of Energy and the Soviet Bloc (1992), coauthor of Can Democracy Take Root in Post-Soviet Russia? (1998), and coeditor of Constitutional Dialogues in Comparative Perspectives (1999).
The 1999–2000 Elections in Russia Their Impact and Legacy
Edited by VICKI L. HESLI WILLIAM M. REISINGER University of Iowa
published by the press syndicate of the university of cambridge The Pitt Building, Trumpington Street, Cambridge, United Kingdom
cambridge university press The Edinburgh Building, Cambridge cb2 2ru, uk 40 West 20th Street, New York, ny 10011-4211, usa 477 Williamstown Road, Port Melbourne, vic 3207, Australia ´ 13, 28014 Madrid, Spain Ruiz de Alarcon Dock House, The Waterfront, Cape Town 8001, South Africa http: // www.cambridge.org C
Cambridge University Press 2003
This book is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2003 Printed in the United Kingdom at the University Press, Cambridge Typeface Sabon 10/12 pt.
System LATEX 2ε [tb]
A catalog record for this book is available from the British Library. Library of Congress Cataloging in Publication Data The 1999–2000 elections in Russia : their impact and legacy / edited by Vicki L. Hesli, William M. Reisinger. p. cm. Includes bibliographical references and index. isbn 0-521-81676-9 (hb) 1. Elections – Russia (Federation) 2. Presidents – Russia (Federation) – Election – 2000. 3. Russia (Federation). Federal§oe Sobranie. Gosudarstvenna¨e`ı Duma – Elections, 1999. 4. Political campaigns – Russia (Federation) 5. Political parties – Russia (Federation) 6. Russia (Federation) – Politics and government – 1991– i. Hesli, Vicki L. ii. Reisinger, William M. (William Mark), 1957– jn6699.A5 a17 2003 324.947 086–dc21 2002038845 isbn 0 521 81676 9 hardback
Contents
Contributors Acknowledgment section i: introduction 1. Parliamentary and Presidential Elections in Russia: The Political Landscape in 1999 and 2000 Vicki L. Hesli section ii: parties in the electorate 2. Television, Voters, and the Development of the “Broadcast Party” Sarah Oates 3. Parties, Voters, and Foreign Policy Stephen White, Neil Munro, and Richard Rose 4. The Foreign Policy Implications of the Elections Andrei Melville 5. Parties, Leaders, and Voters in the Parliamentary Election Timothy J. Colton section iii: the role of regional power 6. Elections and National Integration in Russia Richard Sakwa 7. The 1999 Duma Election in Russia: A Step Toward Democracy or the Elites’ Game? Mikhail Myagkov
page vii ix
3
29 51 73
90
121
142
v
vi
section iv: parties and institutions 8. Instrumental Democracy: The End of Ideology and the Decline of Russian Political Parties Stephen E. Hanson 9. The Impact of the 1999–2000 Parliamentary and Presidential Elections on Political Party Development M. Steven Fish 10. Resolving the Problem of Preelection Coordination: The 1999 Parliamentary Election as an Elite Presidential “Primary” Olga Shvetsova 11. Coalition Politics in the New Duma Thomas F. Remington section v: conclusion 12. The 1999–2000 Elections and Russia’s Prospects for Democracy William M. Reisinger Bibliography Index
Contents
163
186
213 232
261 275 299
Contributors
Timothy J. Colton Harvard University M. Steven Fish University of California Berkeley Stephen E. Hanson University of Washington Vicki L. Hesli University of Iowa Iowa City Andrei Melville Moscow State Institute for International Relations Neil Munro University of Strathclyde Mikhail Myagkov University of Oregon Sarah Oates University of Glasgow William M. Reisinger University of Iowa Iowa City Thomas F. Remington Emory University Richard Rose University of Strathclyde vii
viii
Richard Sakwa University of Kent Olga Shvetsova Washington University St. Louis Stephen White University of Glasgow
Contributors
Acknowledgment
The chapters in this volume were originally presented at the University of Iowa in March 2000 at a conference made possible by the Benjamin F. Shambaugh Memorial Fund, with additional funding from the university’s Center for Russian, East European, and Eurasian Studies. We are grateful to the Department of Political Science at the University of Iowa and to the center for their support.
ix
The 1999–2000 Elections in Russia Their Impact and Legacy
section i INTRODUCTION
1 Parliamentary and Presidential Elections in Russia: The Political Landscape in 1999 and 2000 Vicki L. Hesli1
The chapters in this book have been written in response to a question about what the effects of the 1999 parliamentary elections and the 2000 presidential election have been on the evolution of political institutions and democratic government in the Russian Federation. The sitting of the 1999 parliament and the inauguration of President Vladimir Putin marked the completion in Russia of three competitive rounds of presidential elections (1991, 1996, and 2000) and three cycles of parliamentary elections (1993, 1995, and 1999). In the aftermath of each of the previous electoral cycles, scholars have interpreted, analyzed, and judged the development of presidentialism, parliamentary government, political parties, public opinion, and national integration. In a sense, scholars have been keeping a running theoretical and empirical tally of Russia’s progress through the epic challenges of democratic consolidation in the world’s territorially largest state. This book continues that evaluative and prescriptive process as we watch the institutional structures and the political landscape of the Russian Federation begin to stabilize into a form that is partly uniquely Russian but also representative of political processes characteristic of industrialized countries more broadly. Early studies of Russia’s democratization process tended to be either reserved or negative in their assessments and conclusions. Russian democracy was described as “electoralist,” “formal,” and “unconsolidated” (Linz and Stepan 1996; Grey 1997; Remington 1997; Sorensen 1998). This was contrasted with more positive assessments of democratic consolidation in other post-Communist regimes, such as in Hungary or Poland. Russia’s 1999 parliamentary and 2000 presidential elections, and the subsequent changes in the configuration of legislative and executive power, call for a fresh assessment 1
The author acknowledges the help of Bogdan Nica and Loni Pham in writing this chapter. She is also grateful for the suggestions provided by William Reisinger and Stephen White. Assistance on the preparation of the full manuscript was provided by Wendy Durant, Daniel Morey, and Lauren Klein.
3
4
Vicki L. Hesli
of its post-Communist regime. Considering the process and the outcome of the 1999 and 2000 elections, how are we to assess the likely effectiveness and performance of Russia’s political institutions? Our focus here is on the presidency, the parliament, and political parties. Are these emergent institutions operating in ways that will contribute to the consolidation of democratic practices and procedures in Russia? The present volume presents a package of scholarly perspectives on the development of the Russian presidency and parliament, and on the role that political parties and elections are playing in Russia’s post-Communist transition. The contributing authors analyze Russia’s electoral institutions and political parties in terms of the existing literature on democratization and institutional choice. Chapter authors also provide new and insightful theoretical frameworks for understanding emergent political parties and political leaders as these shape and are shaped by the outcomes of the parliamentary elections of 1999 and of the presidential election of 2000. In sum, the 1999 and 2000 Russian elections provide excellent cases for the testing of a multitude of hypotheses derived from a rich set of theoretical frameworks. In this introductory chapter, I briefly overview the political setting in Russia by enumerating the key powers of the Russian president and parliament. I provide information on the results of prior elections and I introduce the main political parties and presidential candidates that competed in the 1999 and 2000 elections. I provide a short summary of the results of the two contests. From a theoretical perspective and within the broad rubric of democratic theory, I also consider briefly the key issues raised and the major themes discussed in each of the subsequent chapters.
the presidency I begin my discussion with a look at the Russian presidency. The office of the presidency was first instituted as a result of a March 1991 referendum, when Boris Yeltsin as Chairman of the Russian Congress of People’s Deputies placed the question of establishing the position before the people on the same day that they were asked by USSR President Gorbachev to support a reconstituted Soviet Union. After the referendum was approved, Yeltsin garnered 57 percent of the valid vote to defeat his nearest competitors, Nikolai Ryzhkov and Vladimir Zhirinovsky (with 17 percent and 8 percent of the vote respectively), in the first contested presidential race in Russia’s history (June 1991). Yeltsin became president of Russia without embracing a political party. He had publicly resigned from the Communist Party of the Soviet Union (CPSU) a little over a year earlier. In the 1996 Russian presidential election, Communist Party leader Gennadii Zyuganov emerged as a formidable opponent to the incumbent President Yeltsin. The elections were forced into a second round of voting
Parliamentary and Presidential Elections in Russia
5
when Zyuganov challenged Yeltsin in the first round with 32 percent of the vote (to Yeltsin’s 35 percent). Yeltsin emerged victorious in July 1996 after forming an electoral alliance with General Alexander Lebed. According to the constitution, elections for the presidency of the Russian Federation were due to be held again in the summer of 2000. In the aftermath of the December 1999 parliamentary elections, however, Boris Yeltsin resigned from his position as president of the Russian Federation and elections were pushed forward to March 2000. Vladimir Putin, who had been appointed prime minister in August 1999, assumed the position of acting president (while remaining prime minister). Once he became acting president, Putin was the clear favorite in the 2000 presidential elections, and the presidential election campaign narrowed to a two-man race between him and Communist Party leader Gennadii Zyuganov. Whereas in the 1996 presidential election, Zyuganov competed with considerable chances of success against Boris Yeltsin, in March 2000, he faced a much stronger competitor, in the person of Vladimir Putin. Throughout Yeltsin’s reign, Zyuganov had positioned himself as an anti-Yeltsin advocate, and likewise in the 2000 campaign, his rhetoric, while rooted in a socialist ideology, was directed against the legacies of Yeltsin’s rule – rather than against Putin. Putin’s own platform, as reflected in his “Open Letter to the Russian Voters” (February 2000), was more pragmatic than ideological. It consisted of a plan for economic revitalization; the continuation of liberal reforms; a strengthening of state authority and the rule of law; and the suppression of any secessionist drives within the Russian Federation. Other presidential candidates who had appeared to be significant players in the race before the Duma elections quickly faded. Opinion polls taken in January 20002 indicated that Putin’s popular support hovered around 60 percent, while that of his strongest opponent, Gennadii Zyuganov, was only 15 percent. Others among the final list of eleven presidential candidates, with significantly lower popular support and chances of winning, were Grigorii Yavlinsky and Vladimir Zhirinovsky. Grigorii Yavlinsky is the founder and leader of the democratic-oriented party Yabloko. Yavlinsky ran as a presidential candidate in 1996 and finished fourth (in the first round), with 7.3 percent of the vote. Vladimir Zhirinovsky, the founder and leader of the Liberal Democratic Party of Russia (LDPR), is a veteran of presidential races, running in all three Russian presidential elections. In the 1996 presidential elections, Zhirinovsky came in fifth (in the first round), with 6 percent of the vote. Whereas his main ideological position has invariably been one of extreme nationalism, his main targets throughout the 2000 campaign were the communist and democratic candidates rather than Putin. He advocates the strengthening of the Russian state, 2
Opinion polls: January 22–3, 2000 ROMIR; January and February 2000 VTsIOM Surveys; January 1–29, 2000 New Russia Barometer VIII.
Vicki L. Hesli
6 table 1.1 March 2000 Presidential Election Results (results are listed for the top five candidates only) Candidate
Percentage Vote
Vladimir Putin Gennadii Zyuganov Grigorii Yavlinsky Aman Tuleev Vladimir Zhirinovsky
52.94% 29.21% 5.80% 2.95% 2.70%
Source: Central Electoral Commission of the Russian Federation 2000 (b), 191.
especially of the army and security services, and is opposed to the growing autonomy and influence of regional leaders. The preelection support for Grigorii Yavlinsky and Vladimir Zhirinovsky, as registered in public opinion polls, hardly reached 3 percent. Meanwhile, Putin continued to enlist the support of various political forces as the election date approached. A first-round victory (should he win over 50 percent of the vote in the first round) seemed more likely. Among Putin’s supporters in the prepresidential election period were Unity, several other parties close to Kremlin (Fatherland and the Union of Right Forces [SPS]), as well as a growing number of regional governors. Indeed, one significant effect of the outcome of the 1999 parliamentary contest was the decision of Yevgenii Primakov, the leader of Fatherland-All Russia (OVR), not to compete for the presidency but rather to support Putin’s candidacy. The March 2000 presidential election registered a turnout of 68.74 percent, with 1.88 percent votes cast “against all candidates.” The results of the election were hardly surprising (see Table 1.1): acting president and Prime Minister Putin won the race in the first round, with a slim majority (52.9 percent), while the next best performing contestant, Zyuganov, managed 29 percent of the votes. Yavlinsky came in third, with almost 6 percent of the votes, while Zhirinovsky came in fifth, with 2.7 percent. Note that the Communist Party candidate, Zyuganov, gained 29 percent (nearly a third) of the votes cast. This portion of the electorate that voted for Zyuganov is uncomfortable with the moves that have been made by the Yeltsin regime toward a more promarket, individualistic, and Westernoriented economy. In turn, the main candidate of the democratic camp, Yavlinsky, fared worse than in the previous presidential elections of 1996 when, disposing of far fewer campaign resources, he managed over 7 percent of the vote (or 5.5 million votes in 1996 as compared to 4.3 million votes in 2000). The results of the March 2000 presidential election are crucially important for the future of Russia. The results did not represent the imposition of
Parliamentary and Presidential Elections in Russia
7
a new regime, but they did facilitate Russia’s first transfer of executive power through the electoral process. Because Boris Yeltsin was prohibited by the Russian constitution from running for a third term, the country was braced to experience its first presidential turnover in post-Soviet times. Yet, because Yeltsin resigned early, immediately after the parliamentary elections, the outcome of the presidential election was all but predetermined. By transferring presidential power to Vladimir Putin early, and with the presidential elections moved up to March rather than being held in June of 2000, Putin’s victory was all but guaranteed. Even without a regime change, the retirement of Yeltsin and ascension of Putin to the presidency has changed the nature of the political system of the Russian Federation. The reader will recall that both the Russian president and the Russian Congress of People’s Deputies (elected in 1990) remained in place after the dissolution of the Soviet Union in December 1991. The constitutional crisis of 1993, which pitted the Soviet-era Russian legislature against the promarket and anti-Soviet executive represented by President Yeltsin, led to a new parliament (elected in December 1993) and to a new constitutional structure (ratified in December 1993) by which presidential powers were significantly increased. The Russian presidency that Putin inherited is a formidable position with extensive powers and constitutional prerogatives. Articles 80 through 92 of the Constitution of the Russian Federation give the president control over the armed forces, foreign policy, and the military doctrine of the Russian Federation; and the power to dissolve the State Duma (under restricted conditions), call a referendum, sign federal laws, and issue decrees and directives. The constitution gives the president the authority to settle differences between federal bodies and the authorities of regional governments and to declare a state of emergency. Elected for a four-year term, the president is empowered to select the prime minister (to be approved by the parliament), nominate judges, and appoint presidential representatives to the regions. In institutional terms, the President of the Russian Federation emerges as dominant over the legislative organs of the Russian Federation (the Federal Assembly). Few scholars would argue against a characterization of the Russian system as one with a strong executive presidency. Indeed, the term “semipresidentialism” was coined primarily in reference to the period before the referendum on the 1993 Constitution of the Russian Federation. Since the adoption of the 1993 constitution, it is clear that the person who holds the position of president has extensive authority to shape the future of the Russian polity. Where scholars diverge is on the question of the implications of this strong presidency for democratic consolidation. Lijphart (1992), Sartori (1997), and Linz (1997) have cautioned that presidentialism is unlikely to contribute toward political stability in newly democratized countries, and that strong presidential powers are in fact associated with low
8
Vicki L. Hesli
democratic performance throughout the post-Communist world. According to Linz (1990), the fixed term of office and the winner-take-all feature of presidentialism contribute to rigidity of the political process, inadequate societal representation, a dual authority structure, and fragmentation of the party system. Other authors have linked the structure of the presidential system and the development of the party system (Mainwaring 1992; Shugart and Carey 1992). The “dual authority” structure of Russian presidentialism is identified as the major source of interinstitutional conflict (Linz 1997; Linz and Stepan 1997; White 1997). Shugart (1996 and 1997) terms the system of dual accountability of the cabinet to the Duma and to the president a major source of regime instability. Similarly, Sartori (1997, 139) characterizes Russia’s presidential system as “ill-conceived” given the tendency toward confrontation in the executive-parliament relationship. Yet there are proponents of presidentialism who argue that the system provides greater efficiency than parliamentary systems, and therefore promotes political stability. Shugart (2000) argues forcefully in favor of presidentialism under conditions of weak parties, although empirical studies (e.g., Ishiyama and Velten 1998) have found that the strength of the presidency throughout the post-Communist world, including Russia, is not associated with indicators of political stability. Nichols (1999) provides another defense of Russia’s presidential institutions. These evaluations of Russian presidentialism were all based upon the Yeltsin era (1991 through 1999). Observations of the acute and violent conflict between the legislature and the presidency that characterized Russia in 1992 and 1993 influenced much of the analysis that is referred to in the two preceding paragraphs. Although the constitution has not changed, the office of the president and relations between the president and parliament changed considerably when Valdimir Putin took over as acting president in December 1999. One of the most spectacular developments in the months between the December 1999 parliamentary elections and the March 2000 presidential elections was undoubtedly the ascendance of Vladimir Putin from obscurity to central stage. Putin’s popularity gives him distinct advantages in shaping the office of the president and also in shaping his relations with the Federal Assembly. The rapid turnover in the office of prime minister that characterized the end of the Yeltsin era has not been replicated in the beginning of the Putin era. Vladimir Putin has been able to benefit from the support of a stable, working majority in the Duma, which has in turn facilitated the government’s drive toward enacting and implementing policy reforms. This has in turn had a positive impact on the nature of legislative-executive relations under the new administration. In contrast to Boris Yeltsin’s personalistic style, Putin’s own image of being above ideology has left open for him the possibility of alliance with any of Russia’s political parties represented in the Duma. In April 2000, the State Duma and the Federation Council
Parliamentary and Presidential Elections in Russia
9
ratified the START-II nuclear arms reduction treaty. President Yeltsin had been unsuccessful in winning approval of the treaty by the parliament though it had been signed (by Yeltsin and the U.S. president) seven years earlier (and ratified four years earlier by the U.S. Senate). Putin’s ability to work with the political party Unity and with other factions within the Duma has allowed him to build and maintain a much more productive relationship with the legislative branch than had been the characteristic of the Yeltsin era. Thus, the presidential office has assumed a different profile under President Putin. The Russian parliament is likewise evolving. The 2001 spring session of the Duma was notable in its legislative achievements, which rested directly on the coalition of centrist forces from Unity, Fatherland-All Russia, People’s Deputy, and Russian Regions groups. In early 2002, Moscow Mayor Yurii Luzhkov’s Fatherland party, the All-Russia movement, and the Unity party all held congresses in Moscow at which delegates agreed to dissolve their respective political organizations in favor of forming a new party called Unified Russia. Thus, factionalism in the Duma is likely to continue to decline, while a pro-Putin majority becomes more normal. In the following chapters of this book, our contributing authors will offer further characterizations of the institutional and policy changes that are occurring under the leadership of Vladimir Putin. Before I turn to these, however, we first review the history and the importance of parliamentary elections in the Russian Federation.
parliamentary elections in russia The powers of the Russian parliament (the Federal Assembly) are specified in Articles 94 through 104 of the Constitution of the Russian Federation. The Federal Assembly has two chambers. The State Duma (the “lower house”) has 450 members, each of whom is elected for a four-year term. Two hundred twenty-five of these members are elected in single-seat constituencies (often referred to as single-member districts). The other 225 deputies are selected by proportional representation voting, with seats given proportionally to those parties receiving at least 5 percent of the popular vote. The 178 delegates to the Federation Council (the “upper house”) represent Russia’s 89 administrative regions (two delegates from each region). The method of selecting these delegates has changed since 1993: first, delegates were elected, then the heads of the executive and legislative organs of the region became ex officio delegates, and more recently delegates are selected by regional leaders. (Chapters 6 and 10 provide more details.) The Federation Council has the power to approve decrees of the president and to impeach the president – although the process of impeachment is very difficult. Federal laws are first adopted by the State Duma, then voted on by the Federation Council, and ultimately signed by the president. It takes a two-thirds vote by both chambers to overcome the rejection of a law by
10
Vicki L. Hesli
the president. The State Duma approves the nomination by the president for the prime minister, but in the absence of such approval, the Duma may be subject to dissolution. Voters who participate in Russia’s State Duma elections have two ballots: one to choose a candidate competing to represent their electoral district, and the other to choose a political party or an electoral organization that is running a list of candidates. In the single-member districts (SMD) elections, candidates are not required to have a party designation, but for seats filled by parties in proportion to the share of the national vote received if that share is over 5 percent (the proportional representation [PR] portion of the ballot), party names are the main component of the ballot (along with the names of the top three candidates on each party’s central and regional list). Given the PR portion of the ballot, political parties are central to the operation of Russia’s parliamentary elections. In Russia, the first post-Soviet parliament was elected in December 1993. As a result of the 1993 elections, pro-Communist forces and the nationalist opposition together gained more than 40 percent of the seats in the State Duma, while proreform (i.e., pro-Yeltsin) forces won 35 percent of the seats in the lower chamber (Rahr 1994, 32). For the 1995 election campaign, the political field was fragmented and disordered with forty-three organizations listed on the ballot. When the votes were counted, four parties crossed the 5 percent threshold and received representation through the party-list portion of the ballot: the Communist Party, Our Home is Russia (NDR), the LDPR, and Yabloko. The proreform vote and the centrist vote were split among numerous parties, and 49.5 percent of the votes cast were for parties that failed to win any seats through PR allocation. The Communists had the best showing among all the parties in both the SMD elections and the PR vote share and became the largest faction in parliament. Groups opposed to reform maintained majorities in the State Duma elections in December of 1993 and 1995. The resulting legislatures were rightly characterized as obstacles to the reform agenda that was advanced by President Yeltsin. In preparation for the 1999 elections, the Central Electoral Commission approved twenty-six parties as qualifying for the ballot. When all the votes were counted, the results showed that six parties had managed to win at least 5 percent of the party-list (PR) ballot and, therefore, passed the threshold for proportional representation in the Duma. These six parties divided 225 Duma seats among themselves according to their relative share of the vote. Out of the six, three (Unity, OVR, and SPS) did not exist (in their present form) or participate in the 1995 elections. The best performers in the PR race were the Communist Party and Unity (with 24.3 and 23.3 percent of the vote respectively). The OVR coalition pulled 13.3 percent of the vote, while the SPS, the Zhirinovsky Bloc, and Yabloko managed 8.5 percent, 6.0 percent and 5.9 percent respectively (see Table 1.2).
Parliamentary and Presidential Elections in Russia
11
table 1.2 Results of the December 19, 1999 Elections to the State Duma of the Russian Federation
Parties
PR List % Vote (number of seats)
Communist Party (CPRF) Unity (Yedinstvo) Fatherland-All Russia (OVR) Union of Right Forces (SPS) Yabloko Zhirinovsky Bloc (LDPR) Our Home is Russia (NDR) Other Parties Independents
24.3% (67) 23.3% (64) 13.3% (36)
Total Seats
224
Single-Member Districts Seats
Total Number Seats
46 9 31
113 72 66
8.52% (24)
5
29
5.93% (16) 5.98% (17) 1.2%
4 0 7 9 106
21 17 7 106
215
439
Turnout was 60.1 percent. Source: Central Electoral Commission of the Russian Federation 2000 (a), 121–2, 172.
Russian voters seemed to have engaged in strategic voting (they didn’t “waste” their votes on parties unlikely to pass the 5 percent threshold). The disproportionality that results from votes being given to parties that do not clear the 5 percent threshold was far less in the 1999 election than it was in 1995. The parties with the largest percentages of the vote, but with too few votes to gain representation through the PR lists, each captured a relatively small portion of the vote (2.2 percent, 2.0 percent, 2.0 percent, and 1.2 percent respectively). This contrasts with the 1995 election, where parties with 3 percent and 4 percent of the vote were excluded from representation in the Duma. Thus, the 1999 elections saw a significant reduction in the number of wasted votes (votes that were won by parties that received no seats in the Duma) and by definition, a concurrent increase in the representativeness of the institution. This bodes well for both the institutionalization of the Duma and the consolidation of the party system. Among those 225 deputies who were selected by direct election within electoral regions, the Communist Party and OVR were the only parties with a significant number of candidates winning SMD seats. The largest category of SMD candidates was that of independents, and they won half of the single-member seats in the 1999 elections. Unity performed the worst in this respect – indicating that the party was not able to build support for its SMD candidates in the regions. With the sitting of the 1999 Duma, most of the independent candidates formed or joined parliamentary factions. Three independents-based Duma factions emerged – the People’s Deputy Group, which aligned itself with the Unity bloc; the Agro-Industrialists, which are
12
Vicki L. Hesli
close to and controlled by the Communist Party; and Russian Regions (for more on the composition of the Duma factions, see Chapter 11). Given the central role that political parties play in the operation of democratic governance, the question of the role played by parties in the 1999 parliamentary election and the concurrent impact of the election on political party development emerges as a central theme of this book. Each of our contributing authors evaluates the operation of political parties either before, during, or after the 1999 and 2000 parliamentary and presidential elections. Timothy Colton (Chapter 5) provides an excellent overview of why wellfunctioning political parties are good for political systems. These include the presentation to citizens of principled choices on issues of importance as well as continuity and structure to politics that last beyond the life span of particular individuals. In addition, the nature of political parties and the party system shape patterns of representation and influence government effectiveness. Most scholars agree that a few well-structured, programmatic parties, with a solid support base, help foster both government effectiveness and accountability (Lipset 1994; Sartori 1997). Highly fragmented, unstructured, or polarized party systems, in contrast, may lead to unstable regimes and may induce a withdrawal of public support from democratic institutions. Fragmented, unstructured, or polarized party systems are also problematic for executive-legislative relations in systems of dual authority due to the increased likelihood of interinstitutional conflict and high cabinet turnover. The existing literature on this issue distinguishes between PR systems, which aim to increase accurate representation, and majority/plurality systems, which aim to increase effectiveness and clarity of responsibility (Powell 1989). The goal of Russia’s mixed system, employing both a PR ballot and an SMD ballot, was to achieve a balance between the two objectives of representation and responsibility. Given the achievements of the Putin government and the success of the 2001 and 2002 sessions of the Duma, one could argue that Russia’s electoral system is beginning to render these potential benefits – in contrast to evidence from the aftermath of the 1995 elections that seemed to indicate that the electoral system was impeding the development of wellfunctioning government. As was the case with the literature on presidentialism, the 1999 Russian parliamentary election provides an opportunity to update the existing literature on political parties and on parliamentary systems. Our contributing authors do just this. Each author agrees that the nature of the political parties that are emerging in Russia is having significant consequences for the prospect of democratic consolidation. Our authors, however, offer a variety of perspectives on whether these prospects have been enhanced or hindered as a result of the 1999 and 2000 elections. In preparation for the analyses that will be offered by the contributing authors in this volume, we can think of the major political parties that competed in the 1999 parliamentary election as falling into categories based upon two
Parliamentary and Presidential Elections in Russia
13
major criteria. The first criterion is longevity – whether the party had competed in the previous Duma elections of 1993 and 1995 – and the second criterion is what we label “insiders” and “outsiders.” Insiders participated in the Yeltsin-era establishment, while outsiders are seeking to gain control of the governmental and economic establishments. The resulting categories are given in Table 1.3. The Russian party system is hard to categorize. Left-right dimensions, proreform and antireform dimensions, pro-Communism versus antiCommunism dimensions, and proregion versus procenter dimensions are table 1.3 Major Political Parties Competing in the 1999 Parliamentary Elections in the Russian Federation
Government Orientation
Old Parties (competed in the 1993 or 1995 elections)
New Parties (created for the 1999 elections)
Our Home is Russia (NDR) Founded: May 1995 Leaders: Viktor Chernomyrdin and Vladimir Ryzhkov
Unity (Yedinstvo) Founded: September 1999 Leaders: Sergei Shoigu, Alexander Karelin, Alexander Gurov Fatherland-All Russia (OVR) Founded: August 1999 Leaders: Yevgeny Primakov, Yuri Luzhkov, Vladimir Yakovlev Union of Right Forces (SPS) Founded: August 1999 Leaders: Sergei Kiriyenko, Boris Nemtsov, Irina Khakamada
Insider Parties Led by 1999–2000 Government Team Led by Former Insiders
Parties Led by Outsiders
The Communist Party of the Russian Federation (CPRF) Founded: February 1993 Leader: Gennadii Zyuganov Yabloko Founded: October 1993 Leader: Grigorii Yavlinsky Liberal Democratic Party of Russia–Zhirinovsky Bloc (LDPR) Founded: March 1990 Leader: Vladimir Zhirinovsky
14
Vicki L. Hesli
also offered by authors of this volume as well as others as ways to understand the structure of the Russian party system. To make the situation even more complicated, relationships are different among parties during electoral campaigns than they are after elections when coalitions across parties need to be formed in order to pass legislation. The LDPR, for example, has historically presented itself as an alternative (oppositional) organization to the choices and decisions that were made by the Yeltsin presidency, yet according to roll-call analyses, the Zhirinovsky Bloc frequently voted for Yeltsin’s bills. Similar problems emerge with our longevity dimension. The SPS was technically a new bloc created in 1999, yet the bulk of its members and leaders came out of the party Russia’s (Democratic) Choice, which competed in 1993 and 1995. The fact that the new parties performed better on the whole than the established parties (the Old Parties in Table 1.3) in the 1999 election lends credence to those who see the party system in Russia as “weak” or “underdeveloped” (see Chapter 9). Arguably the newly created political parties were meant to function primarily as electoral machines for their leaders – rather than to offer packages of policy alternatives and program commitments to a constituency of voters. Some Russian political parties, however, such as the Communist Party of the Russian Federation (CPRF), do exhibit a developed organizational structure and ideological coherence. The CPRF has the most extensive organization of Russia’s parties as it directly inherited the structures of the Communist Party of the Soviet Union (CPSU). Heading into the 1999 electoral season, the CPRF was considered to be the strongest party as it had won the largest share of the party-list votes in the 1995 parliamentary election. As it had for the 1993 and 1995 parliamentary elections, the CPRF continued to stress in the 1999 election campaign, albeit in a more moderate fashion, the Marxist-Leninist ideals of social justice. It advocated a more temperate drive toward reform than was characteristic of the Yeltsin regime. The CPRF looks back to the glorious and proud days of the Soviet Union and seeks to return Russia to its former stature as a world super power. Yabloko (led by Grigory Yavlinsky) is similar to the CPRF in the sense that it participated in the 1999 Duma elections as one of just a few parties that had also contended in the first Duma elections in 1993 and the second Duma elections of 1995 (Table 1.3). Yabloko’s strategy, amidst a seasonal fervor of mergers and coalitions, was to run independently. This strategy was expected to deliver a calculated result: to maintain its core of “democratic” supporters. Throughout the campaign, Yabloko positioned itself as the only viable democratic opposition to the current regime. Prior to the election, the party consistently voted against the government’s proposed budgets and disassociated itself from the government’s economic policies. Yabloko advocates the country’s complete transition to a market economy, including the privatization of all state industries and the removal of subsidies for agriculture and fuel.
Parliamentary and Presidential Elections in Russia
15
Vladimir Zhirinovsky’s LDPR is also a competitor that survived through Russia’s first (1993) and second (1995) post-Soviet parliamentary elections. The LDPR was initially disqualified in 1999 by the Central Election Committee for improper income declarations of some of its members. This prompted its leader to form a new party, called the Zhirinovsky Bloc. Zhirinovsky’s Bloc, although known for an ultra-nationalist platform that calls for reincorporating former union republics into Russia, also made significant promises in the area of social welfare. This decade-long history of political party competition, as represented by the participation through three electoral cycles by the Communist Party, Yabloko, and Zhirinovsky’s organization, was supplemented in the 1999 parliamentary election campaign by the emergence of new political organizations. As the December 1999 Duma elections approached, new electoral associations were formed and existing political parties started exploring strategies for electoral alliances. Thus, the Fatherland-All-Russia coalition (OVR) coalition was formed in August 1999 – despite the fact that Fatherland and All-Russia were two separate organizations, distinct in ideology and electoral base – as a strategy to compete more effectively against the Communist Party and to ensure parliamentary representation for both parties. The organization called Fatherland, led by Moscow Mayor Yurii Luzhkov, was appealing to those discontented with Russia’s post-Communist economic reforms and, according to VTsIOM public opinion polls from the summer of 1999, was second in popularity after the Communist Party. AllRussia, founded in April 1999, brought together regional leaders, members of the State Duma, and directors of large enterprises. As a movement founded by influential governors and republican presidents, All-Russia advocated more powers for the regional authorities. Its own incentive for forming an alliance with Fatherland was the requirement that prohibited it from participating by itself in the party-list portion of the ballot, due to its late formation and registration as a party. The OVR centrist coalition supported market reforms, but also called upon the state to control and regulate the market. The party list for the OVR in the 1999 election was headed by former Foreign Minister and former Prime Minister Yevgenii Primakov. The OVR’s main electoral message was the creation of nationally oriented state capitalism. Shortly after the OVR alliance was formed, the Agrarian Party also joined the alliance, ending its coalition with the Communist Party. Regional electoral blocs, such as All-Russia, sought to improve the representation of regional interests in the State Duma. Regional presidents and governors who participated in these parties had been discontent with their limited influence based solely on those members of the Duma elected from SMDs. By forming their own parties to enter the electoral races on the party-list ballots, the regional leaders hoped to gain greater leverage on the federal government, through a stronger presence in the Duma. In turn, with Russia’s regional governors elected instead of appointed by the presidential
16
Vicki L. Hesli
administration, the major electoral blocs also competed to win the support of governors. Because governors and republican presidents have the power to influence voters in their respective regions, the support of regional leaders was an important asset for the electoral contenders (see Chapter 7). Thus, virtually all the major political forces courted and were in turn courted by regional elites. The Voice of Russia (led by Samara Governor Konstantin Titov), another electoral organization newly formed in the period before the 1999 election, was similar to All-Russia in the sense that it advocated a “new federalism” ` that would grant regions more authority vis-a-vis the federal government. These two regional associations contemplated the possibility of forming an electoral alliance, given the potential difficulty for each of them to alone pass the 5 percent threshold in the Duma elections. Although at the level of policy, both regional blocs sought increased powers for the regions, they differed on economic policy issues. Where All-Russia favored “state capitalism,” the Voice of Russia adopted a more liberal economic stance. The two parties eventually joined different alliances: All-Russia entered an alliance with Fatherland, while Voice of Russia joined the newly formed SPS. The Union of Right Forces (SPS), a center-right coalition, was formed in August 1999 from New Force (led by former Prime Minister Sergei Kiriyenko), Young Russia (led by former Deputy Prime Minister Boris Nemtsov), Common Cause (led by Irina Khakamada), Democratic Choice of Russia (led by former Prime Minister Yegor Gaidar), and Voice of Russia. Most senior members of this coalition were former government officials. Among its stated goals was a reduction in the decision-making power of the central authorities in Moscow. The coalition sought to enhance the representation of the country’s developing middle class, entrepreneurs, managers of small businesses, and private farmers. Their election campaign focused on calls for the protection of private property, on the idea of a professional army, and on anticorruption efforts. The last of the major “political parties” that competed in the 1999 parliamentary elections is the one that emerged as the second-place winner in those elections and now plays a central role in operation of the State Duma. This is the organization called Unity (Yedinstvo). Unity was formed only in September 1999; yet it emerged after the election as one of the strongest forces in the Duma. Unity was formed as a “party of power” and from its very inception it enjoyed the support of both President Boris Yeltsin and Prime Minister Vladimir Putin. Although formally not a member of the party, the latter attended its founding congress in October 1999 and explicitly endorsed the party in November before the elections. Unity’s major calculated impact on the preelectoral scene was to take away votes from the popular OVR bloc, although it spurred a realignment of potential candidates across all political formations. To summarize this section on the results of the parliamentary election and on the structure of the political party system, we can say that the
Parliamentary and Presidential Elections in Russia
17
December 1999 elections produced a new Duma, with a configuration distinct from that found in the preceding Duma. The election marked the shifting of power in the Duma away from the Communist-nationalist bloc toward a more centrist, proreform orientation. Although the Communist Party remained the largest single party in the State Duma after the 1999 parliamentary elections, its share of the Duma seats shrank considerably compared with the election of 1995. Although weaker, the Communist Party remained a force to be reckoned with in the Duma and the Federation Council. The Communist Party did well in the SMD races, which can be attributed to its superior organization and grassroots involvement. The party could also claim a “victory,” in the sense that it emerged as the sole representative of the “leftist” vote; indeed, none of the other leftist parties managed to come even close to CPRF’s performance (such parties include Spiritual Heritage, Pensioners’ Party, Communists and Workers’ Russia, and the Stalinist Bloc). In the configuration of the 1999 Duma, the Communists comprised two factions on the left side of the political spectrum – that of the Communist Party and of the Agro-Industrial Group. Unity, as Russia’s most steadfast pro-Putin political party, enjoyed a pivotal role in the post-1999 Duma. The fact that no single faction had a majority in the Duma conferred on the centrist forces (especially Unity) a position of leverage in coalition building. In contrast, Yabloko and Zhirinovsky’s LDPR each saw their share of the Duma seats reduced in the aftermath of the December 1999 voting. The nationalist message of the Zhirinovsky Bloc captured 7.6 percent of the party-list vote in 1999, as compared to 11.4 percent in 1995. Yabloko’s losses were due mainly to the strength of the newly created SPS, which competed with Yabloko for the “democratic” vote and performed better than expected, with 8.5 percent of the vote on the party-list ballot. Given its small share of the vote in 1999, Yabloko could aspire to be a significant player in the new Duma only by aligning itself with more dominant (centrist) factions. This move, however, would come at the expense of the party’s strong-held strategy of remaining true to its principles – the complete transition to market economy and liberal Western-style democracy. As Thomas Remington discusses later in this volume, the combination of alliances and factions in the Duma have significant implications for executive-legislative relations as well as for the government’s plans of reform. In terms of institutional effectiveness, factions within the Duma play a crucial role. Majority coalitions and power-sharing agreements are formed from factions. In Chapter 11, Remington studies in greater depth the distribution of seats in the 1999 Duma and notes that the progovernment forces are much larger in the 1999 Duma than in the 1993 and 1995 Dumas. Using a survey of candidates in the 1999 election, Remington is able to show that the members of the “parties of power” (Unity, NDR, and OVR) can generally be placed at the median of a policy dimension, which
18
Vicki L. Hesli
ranges from a pro-Communist orientation to a liberal, market-oriented position. Remington points out that because the “parties of power” are not distinctive in terms of left-right position, but rather campaigned on vaguer appeals “free of specific policy objectives,” they are relatively free ideologically to form majority coalitions with the more programmatic parties and ideologically distinctive parties – such as the Communists. Unity, in fact, joined a minimum winning coalition with the Communists and the People’s Deputies to select the speaker for the 1999 Duma. Unity initially aligned with the Communist Party to select the speaker and to allocate committee chairmanships, but subsequently realigned with liberal factions in voting on substantive policy issues. Thus, according to Remington, the strength of the Unity faction in the Duma (“it is the one indispensable member of any minimum winning coalition”), together with its lack of ideological rigidity, provides President Putin with a powerful legislative ally. As mentioned previously, President Putin was able to build a coalition to ratify the START-II treaty shortly after winning the March 2000 election. This capacity to build majorities within the Duma represents a positive step toward the effectiveness of government in the Russian setting. Another change associated with the December 1999 elections was a distinct increase in the representation of regional interests in the State Duma. Previous studies have already demonstrated that local administrators from Russia’s regions play a significant role in determining the outcomes of Russia’s elections (Badovskii and Shutov 1997, 50). The importance of regional politics and regional patterns of voting was once again clear in the context of the 1999 Duma elections. Preelection agreements that were negotiated between local authorities and national political elites were an important component of political maneuvering in the period prior to the 1999 and 2000 elections. Following up on this area of study, Mikhail Myagkov in Chapter 7 of this volume demonstrates the critical role played by regional governors in determining the outcome of the 1999 and 2000 elections. Myagkov notes that the level of support for the established parties (such as the Communist Party and Zhirinovsky’s LDPR) remained remarkably stable between the previous 1995 parliamentary election and the 1999 Duma election. The level of urbanization and the geographic location (“red belt” compared with northern and eastern regions of Russia) of a particular region still does help to explain the level of support for those parties that had participated in previous elections (such as the Communist Party), but these factors are not significant in explaining votes for Unity or OVR in the 1999 election. Rather the significant determinant of the size of the regional vote for Unity or OVR is whether the governor of the region was a member of one of these two “parties.” Thus, even though about 40 percent of the electorate appears to be developing relatively stable partisan attachments, a significant portion of
Parliamentary and Presidential Elections in Russia
19
the vote in the 1999 election appears to have been determined through the will and the influence of the regional governors. This means that in several regions of Russia, governors were able to control the outcome of their own local elections and thereby greatly influence the national election.
theoretical issues raised by the parliamentary and presidential elections Myagkov’s analysis raises a classic question about the effects that elections have on patterns of citizen representation and on the accountability of government. Also related to the questions of representation and accountability is the issue of voter identification with a specific political party. The traditional argument from the Western literature is that long-term, enduring attachments to political parties, characteristic of stable party systems, are conducive to democratic consolidation. Yet, earlier studies of the Russian electorate reported low levels of party identification among the electorate. Indeed, political parties were frequently identified by post-Soviet scholars as the least-trusted institutions of the new democratic system in Russia, which would make their task of structuring electoral choice difficult if not impossible. Kopecky (1995), for example, characterized post-Communist systems as having weak party identification among the electorate. At the time of his writing, parties were described as having loose electoral constituencies and party leaders as opposed to party members that played a dominant role in electoral outcomes. In characterizing the 1993 parliamentary elections, Sakwa (1995, 207) stated that “electors were faced with a choice not so much between parties and distinct programs as between lists of stars.” Yet, by the middle of the 1990s, researchers were seeing a link between political party preferences in Russia and issue preferences (White et al. 1995, 196), thereby establishing a relationship between political attitudes and the likelihood of identifying with certain political parties or candidates. Party attachments arguably are cultured over time. The question of whether party attachments are affecting electoral outcomes, therefore, needed to be readdressed on the basis of studies of the 1999 and 2000 elections. In Chapter 5 of this volume, Colton provides evidence that partisan attachments did affect voting behavior in Russia’s 1999 parliamentary elections. It is difficult to determine, however, at the level of the individual, how important party identification is in an abstract sense to the voting decision because parties in Russia tend to be so closely tied to the party leadership. Indeed, Colton presents evidence from surveys of Russian voters that the personal qualities of party leaders were very important to citizens when they made their decisions about the party for which they would vote in the 1999 parliamentary election. The program of the party, however, also enters prominently into the vote decision. Overall, Colton is able to determine that
20
Vicki L. Hesli
a little over 40 percent of the Russian electorate in 1999 could be described as partisans – in the sense of having an identifiable political party to which he or she is attached. Consistent partisans in Russia tend to be most numerous among the supporters of the Communist Party, Yabloko, and the LDPR. Among the main parties that participated in the 1999 election, Colton also demonstrates that both party attachments and assessments of the respective leaders of the parties affected the voting decisions of Russian citizens. Thus, if the question of party development is addressed from the perspective of partisanship in the 1999 parliamentary elections, we can hardly argue that partisanship is lacking or ineffectual in Russia. Tentative attachments to parties played a significant role in shaping the outcome of the 1999 election. The 1999 round of Duma elections also suggests that a structuring of partisan alignments along societal and ideological cleavages is possible, despite the persistent role of personalities in Russian politics. In Chapter 3 of this volume, Stephen White and his coauthors report on observed relationships between support for the main political parties and the foreign policy orientations of the individual voters. Among the major parties that contested the 1999 Duma elections, supporters of Unity, for example, most favored the conduct of the antiterrorist operation in Chechnya. Supporters of the Communist Party were most committed to closer integration with former Soviet states. They also find that Unity voters have few distinguishing attributes, while the Communist voters are more distinctive with clearer policy preferences. As would be expected given their political platforms, supporters of the SPS and Yabloko are the most pro-Western in their international orientation. The authors of Chapter 3 also provide for the reader an overview of public opinion on major questions of foreign policy facing the Russian Federation. In addition, White et al. provide an overview of the platforms of the major parties regarding foreign policy issues. In Chapter 4 of this volume, Andrei Melville discusses the foreign policy resources available to President Putin in the context of the major foreign policy orientations characteristic of the Russian populace. Through public surveys, Melville demonstrates that the majority of the Russian people long for a restoration of Russia’s greatpower status, but they also believe that Russia should develop mutually beneficial ties with the West. Through Vladimir Putin’s own political statements, Melville is able to show that Putin prefers a pragmatic approach to foreign policy. Putin’s foreign policy approach leans toward Westernism and pragmatic selectivism, which bodes well for Russia’s continued democratic growth. Evidence from both chapters, however, shows that domestic issues rather than foreign-policy issues are of primary concern to Russian voters. White et al. demonstrate that when selecting which political party to support, voters are mainly concerned with Russia’s political and economic transformation to a market economy – rather than with foreign-policy issues. On economic issues, the programs of all competing forces in 1999 were markedly more
Parliamentary and Presidential Elections in Russia
21
promarket as compared to the previous elections. This reflected a general consensus over the priority of revitalization of the economy. Even the Communist Party’s program shifted somewhat toward the center, to accommodate free market principles. Our contributing authors, however, note the lack of clearly articulated political platforms – particularly by the new political parties and notably by Unity. The remarkable showing of Unity in the 1999 Duma election (explained by Olga Shvetsova in Chapter 10) is not a result of an appealing party platform nor a result of charismatic leadership, but rather as the outcome of a coordination problem among Russian elites who needed to align themselves behind the winner of the 2000 presidential contest. Thus, the strategic connection between the parliamentary election and the presidential election explains the rush “to get on the bandwagon” of the newly created Unity party – even though the party had no clear platform or ideology and little organizational strength. Therefore, according to Shvetsova, serious debates about policy across the political parties were muffled by the focus on the likely victor in the presidential election. She argues that it is precisely because competition for parliamentary elections is not built around ideologies and policy issues, but rather around central political figures, that an institutionalized party system has yet to develop in Russia. Because of the importance of the presidency in the Russian system, parliamentary elections will likely continue to play the role of “elite primaries.” Shvetsova explains that given the focal power of the incumbency advantage in this coordination process, presidents in Russia have little incentive to build or strengthen political parties. She notes that new parties are likely to continue to be launched as vehicles for presidential bids, thereby continuing to undermine the consistency factor in party development as identified by M. Steven Fish in Chapter 9. M. Steven Fish decries the impact of the recent presidential and parliamentary elections of party development in Russia. In measuring the strength or development of party systems in general, he identifies the following criteria as “particularly telling”: magnitude, stability, differentiation, gravity, integrity, and reach. Neither magnitude, stability, nor integrity expanded or improved as a result of the elections. Differentiation of the party system, in terms of representation of a wide range of opinion, however, was a positive feature of political parties in Russia prior to the 1999 elections and remained so afterward. Likewise the geographical reach of political parties remained extensive after the 1999 elections. The gravity of the system, however, declined as even more candidates were elected in 1999 as compared with 1995 without party affiliations. Fish concludes, relying on his criteria, that no improvement in the strength or maturity of the Russian Party system occurred with the 1999 and 2000 elections, but rather that the effect may have been adverse. Fish explores possible explanations, such as the cultural and historical legacies of Russia and the superpresidentialism structure of the Russian
22
Vicki L. Hesli
government, for why elections in Russia have not encouraged party system development. Superpresidentialism, as characterized by the extraordinary ` power of the president vis-a-vis other potential sites of power within and outside the government, is identified as the primary obstacle to party development in Russia. This obstacle to party development is exacerbated by the timing of the elections (holding the parliamentary elections just six months before the presidential elections) and the nomination process that does not require party endorsement. Thus, the current system that schedules parliamentary elections close to the presidential election, together with the disproportionate power of the presidential office, is judged by both Shvetsova and Fish to be detrimental for strengthening the political party system in Russia. The timing of the parliamentary elections and the presidential election critically affects the outcome of both contests. Following just months after the December 1999 Duma elections were Russia’s presidential elections – held in March 2000. In addition, presidential elections in Russia have been won without the participation of parties – thereby further undermining the prospects for party building in Russia. Both Richard Sakwa and Stephen Hanson, however, provide institutional analyses in this volume that lead to more positive expectations about the future of party development in Russia. Even though party building has been delayed in Russia due to segmented regionalism (according to Sakwa) and uncertainty in the political and economic environment (according to Hanson), the stability and certainty associated with the stronger presidency of Putin as compared to Yeltsin is seen by these two authors as a positive development for Russia. In Chapter 6, Richard Sakwa defends presidentialism as a key to national integration – and national integration is required for building a democratic political community. He explores electoralism and the development of a political community in Russia and asserts that the presidential elections and the relatively high voter turnout in the competitive elections are evidence that a national political community is emerging in Russia. The very act of participation in elections legitimizes the polity and serves to structure a national political community. Prior to 2000, and as a result of the autonomy of regional executives, electoralism had actually undermined the political community. But the strong electoral mandate received by Putin in 2000 changed the balance of power in favor of the center. The limited reach of national political parties and the corresponding strength of distinctively regional parties that marked the Russian political scene in the late 1990s has now been overshadowed with the victory of Unity – which represents the victory of central authority in Russia. Since he became president, Putin has mounted a sustained assault on segmented regionalism – which had been undermining the development of a national party system in Russia. Thus, Putin’s approach to federalism and his willingness to confront regional demands for autonomy bode well for political reform in Russia.
Parliamentary and Presidential Elections in Russia
23
In Chapter 8, Stephen Hanson argues that the uncertainty characteristic of the Russian political setting severely limited the possibilities for successful national party building. Under uncertain conditions long-term incentives are lacking, and it is not instrumentally “rational” for leaders to invest their time and their efforts to build party organizations. Thus, the turbulent milieu of Russia during the 1990s explains individual decisions not to invest in the building of genuine and long-term party organizations. The value rational foundations for party building have also been lacking given the cultural cynicism that remains a legacy of the Leninist system and given the discrediting of alternate ideologies (such as liberalism and socialism) in Russia. With Putin’s ascension to the presidency, however, and given the relative economic stability in Russia since 1998, long-term political strategizing becomes more feasible for politicians. In other words, if we follow Hanson’s line of reasoning, the stronger and the more stable the Putin regime, the better are the prospects for institutional development in the Russian setting. In fact, Putin need not necessarily endorse a particular party for party organization to proceed. Clearly the elections that occurred in 1999 and 2000 provide valuable data for the study of political party development and democratic transitions more generally. Establishing electoral institutions and the holding of competitive elections are major and necessary steps in the implementation and consolidation of democracy (Lijphart and Waisman 1996 [a], 1996 [b]; Linz and Stepan 1997). Elections not only foster the legitimacy of the new democratic institutions, they are the essence of the democratic process. More is required for the existence of democracy, however, beyond the holding of regular competitive elections. In the concluding chapter of this volume, William Reisinger reviews key components of democracy in the context of the Russian setting. We note here that when assessing the performance of newly established democracies, key goals must be realized in several areas, including political parties, civil society, popular democratic support, and the rule of law, if a democratic regime is to become institutionalized, consolidated, and legitimate. According to Linz and Stepan (1997, 5), democracy is consolidated when “[it] becomes routinized and deeply internalized in social, institutional and psychological life.” A similarly well-accepted position in the field is that democracy becomes consolidated when major players accept the rules of the game, that is when democracy becomes the “only game in town” (O’Donnell and Schmitter 1986). In order for democracy to become “the only game in town,” the rules of the game must first be established. Institutional design is generally considered in the literature on democratization as playing a pivotal role in the establishment of a new order (Di Palma 1990; Przeworski 1995; Lijphart and Waisman 1996 [a], 1996 [b]; Linz and Stepan 1997; Remington 1997). Certain “legacies of the past,” however, are considered likely to undermine a successful transition to liberal democracy (Jowitt 1992). These include
24
Vicki L. Hesli
economic backwardness, the centrally planned economy, unfinished nation building, a political culture of intolerance, a weak state with illiberal traditions, and a weak civil society (Crawford and Lijphart 1997). Structural theorists consider the level of economic development, the nature of political culture, and also state-civil society relations as preconditions for democratization and consolidation (Reisinger 1997; Plasser, Ulram, Waldrauch 1998). Nonetheless, according to the neo-institutional tradition, consolidation entails “institutionalizing competitive politics” (Lijphart and Waisman 1996 [a], 3) and is therefore a matter of design rather than inheritance; legacies still matter (as constraining factors), but they are not overriding. The 1999 and 2000 elections indicate that, at least at the level of its electoral institutions, Russian democracy is working toward consolidation. Three rounds of parliamentary and presidential elections in Russia’s decadelong post-Communist history are in and of themselves highly significant for democratic consolidation. This is the case, despite considerable criticism regarding the procedural fairness of these elections and unabated allegations of fraud, misuse of campaign funds, and partisanship of state and private media in covering the electoral campaign (RFE/RL Russian Election Report, April 7, 2000). A major outcome of the media effect was the propelling of Vladimir Putin as a major contender for the presidential office. In fact, the biased media coverage of the elections provided the foundation for allegations of procedural unfairness. In Chapter 2, Sarah Oates provides evidence of the influences of the biased Russian media on presidential and parliamentary elections. Unity, for example, received a disproportionate share of news coverage across all the major Russian television channels in the period leading up to the 1999 parliamentary election. In contrasting traditional concepts of party development with Russia’s experience, Oates explains the growing success of the “broadcast party.” A broadcast party is defined as a “political movement that relies heavily on television for its creation and electoral success.” She cautions that although Russian officials have been relatively successful and proficient in creating broadcast parties, the manipulation of the media for political interests will lead to an erosion of public trust. Her analysis, therefore, raises concern about the ability of the public to access uncontrolled and unmanipulated information. When media freedom is restricted by the Russian state, democratic free choice as well as political party alternatives may be in jeopardy. Nonetheless, it is the political fluidity and societal chaos of the Yeltsin era that are interpreted by Oates as a reason why politicians rely more on short-term campaign appeals than on long-term accountability and party formation. Thus, the possibility for change in preelection tactics depends, in part, on the changing nature of the system itself. To conclude this introduction, I can say that Russia’s experience with democracy has been both convulsive and checkered from the beginning, but arguably no less checkered than the early histories of most established
Parliamentary and Presidential Elections in Russia
25
democracies. The potential for executive-legislative conflict that is, in a sense, embedded in the Russian constitution remains, but the composition of the 1999 parliament together with the personage of President Putin provide reasons for more optimistic expectations about the performance of Russian government. Likewise, Putin’s efforts to rein in regional autonomy bode well for national integration of the Russian state (his administrative reforms include provisions for the dismissal of regional authorities if lower-level legislation is not brought into line with the federal constitution). A cautionary note has been raised, however, in this regard. Given the characterization of Russia’s political arrangements as superpresidential under Yeltsin, Putin’s success at concentrating even more power into the position of the president raises issues of democratic accountability. And even though the political system has become more predictable and more stable, the party system is likely to remain slow to develop, given fragmenting institutional factors such as the nonsimultaneous presidential and parliamentary elections and the SMD section of the Duma ballot. The consolidation of democracy in Russia, as in any system in transition, is a complex process involving multiple dimensions and multiple tasks. The assessments of Russia’s transition provided within the pages of this volume are equally complex and involve a variety of theoretical approaches. It is, therefore, not surprising that we have somewhat different conclusions emerging from our diverse set of authors. In particular, they offer varying assessments of the degree to which the 1999 and 2000 parliamentary and presidential elections contributed to political party development and to democracy in Russia. They agree, however, that these parliamentary and presidential elections have proven critical for the future of a major world power. Insights into the future, which may be drawn from our contributing authors’ assessments, are explored further in Reisinger’s probing conclusion to this book.
section ii PARTIES IN THE ELECTORATE
2 Television, Voters, and the Development of the “Broadcast Party”1 Sarah Oates
Very few political parties have endured in the young Russian state since the collapse of the Soviet Union and one-party rule. What we can observe over time is an improved ability of those in power to create, market, and win election for political parties and presidential candidates. At times, this task was enormously difficult for the Russian president, as both Boris Yeltsin and Vladimir Putin encountered deep popular discontent with the economic and political problems of the post-Soviet state. Yet evidence ranging from the creation of the Russia’s Choice progovernment party in 1993 to the election of Putin as the new president in 2000 suggests that the Kremlin has learned important lessons and improved its ability to use elections to consolidate its political power. Notably, the Yeltsin administration discovered the importance of favorable media treatment, plausible candidates, and crafting campaign statements that appealed to the Russian heartland. While issues and government performance surely matter to Russian voters, the political fluidity and societal chaos have led to politicians, namely those in the Kremlin, relying more on short-term campaign appeals than on long-term accountability and plausibility for party formation. This chapter will argue that Russian politicians have learned these lessons well, leading to the creation of the “broadcast party,” that is, a political movement that relies heavily on television for its creation and electoral success – albeit not for its survival. Broadcast parties are a reflection of the larger political forces at work in a postauthoritarian political landscape, namely the relative lack 1
The study of Russian elections and television at the University of Glasgow was supported by ESRC Grant No. R000223133 “Building a New Democracy?: Television, Citizens and Voting in Russia.” A previous version of this chapter was presented at the Political Studies Association Conference held April 10–13, 2000, in London (United Kingdom). The author gratefully acknowledges the work of Russian Research Ltd. for taping the news and political advertisements in the 1999 Duma and 2000 presidential elections in Russia. In addition, much of the information for this article was gathered during a media monitoring mission by the European Institute for the Media (EIM), in which the author took part.
29
30
Sarah Oates
of fixed voting preferences, the ability of the state to manipulate television, and the strength of broadcasting in a political tabula rasa. While broadcast parties can allow for electoral victory, they inhibit the development of political parties with solid roots in the electorate. While consolidating the power of the elite, broadcast parties undermine the political strength of the masses. Although Russian society has generated some relatively strong and enduring party organizations, particularly the Communist Party of the Russian Federation, the party system is becoming increasingly deformed by the growing success of the media-based broadcast parties.
traditional concepts of party development and the russian case Most of the political science literature on parties focuses more on the gradual evolution of preferences within society than on the sort of political “big bang” experienced by Russia after the collapse of Soviet rule. However, there are still quite useful concepts that help in the understanding, up to a point, of how political parties have formed in Russia. For example, Lipset and Rokkan (1967) suggest that political parties develop along the traditional cleavage lines in society. For example, parties emerge to support peasant interests, parties champion the rights of workers, and parties protect Catholic interests. Certainly, recognizable “cleavages” appeared even within the relatively egalitarian Soviet system, including differences between those who lived in the cities and those who lived in the countryside, between workers and peasants, and between young and old. But the work of Lipset and Rokkan, while valuable for societies in which parties evolve over a relatively long period of time, becomes more problematic in societies such as Russia, in which an authoritarian system collapsed in a short span of time and civil society had little time to develop. At any rate, it is not enough merely to identify a sector of society and claim to be its representative. For example, studies of European parties have shown that they must change and reformulate as society changes (Dalton, Flanagan, and Beck 1985; Merkl 1988). Parties that purport to represent a large faction in society often fail, as Rose and Mackie (1988) show in their historical study of party success in Western Europe. In addition, the connection between a party and a key group of supporters within society is complicated, for support may shift considerably over time (Dalton et al. 1985). While the social cleavages described by Lipset and Rokkan matter, more recent work suggests that other factors also are important in explaining party success and failure at the polls. Authors in this volume have found some link between attitudes and social characteristics with voting patterns. For example, in Chapter 3 Stephen White, Richard Rose, and Neil Munro find that Russian voters are mainly concerned with Russia’s political and economic transformation to a market
TV, Voters, and Development of the “Broadcast Party”
31
economy. Some constituencies are more defined; White and his colleagues found that Communist voters had clearer policy preferences than Unity supporters. Early work has shown that voters have become more suspicious of a market economy, more nationalistic, and more concerned about law and order (Oates 1998 [a]). In Chapter 5, Timothy Colton provides evidence from a major survey of Russian voters to show that attachments to parties played a role in shaping the outcome of the 1999 elections – but partisanship is low when compared with the United States or Western Europe. It is difficult to discuss traditional left-right cleavages for Russia, particularly as there is disagreement over where to place the contemporary Russian communists on a traditional political spectrum. So if traditional cleavages do not provide solid foundations for parties in Russia, what is the primary foundation? Parties, after all, cannot merely announce that they represent impoverished farmers or wealthy urban entrepreneurs and wait to be voted in at the polls. Parties must create some sort of reputation or, failing that, at least an attractive image for the voters. Some explanation for this comes in another facet of the comparative political literature in the West, namely political communication writings, which deal with the “Americanization” of political parties. For example, Denis Kavanagh (1996) posits that the trend of U.S. parties to rely more on image than substance has spread to other countries, particularly Great Britain. This has come about as the strength of partisan identification has eroded on a personal level, in parallel with the weakening of traditional class cleavages and party lines in Britain. Kavanagh lists components of Americanization as the devotion by parties of more resources to media strategy and tactics; the adaptation of the campaign to the requirements of television; the idea of electioneering as political marketing; and the increase in negative (or attack) campaigning. Kavanagh points out that British parties are somewhat restrained from following the Americanization model by certain key differences from their U.S. counterparts. In particular, British parties have less money and opportunity for advertising, are stronger organizations, and attach more importance to policy and the candidates’ stands on the national party programs than U.S. parties. Other writers on political communication, such as Negrine and Papthanassopoulos (1996), suggest that it is not so much “Americanization” as “modernization” that is reflected in the more media-based political campaigns of the late twentieth and early twenty-first centuries. They argue that as the world becomes more modernized, that is, more educated; becomes less insular; and has broader communication systems to make more information available, voters no longer rely as much on passive cues for vote choice. As a result, media messages during elections become more important. Whether it is thought of as Americanization or modernization, it is useful to note that political parties in Russia are not alone in their focus on media over
32
Sarah Oates
substance. However, unlike their counterparts in the United States, Great Britain, or other advanced industrial democracies, most Russian parties lack roots either in the past or in the consciousness of the voters. Thus, while Western parties may be abandoning some of their traditions and history in a media-based push for the large segment of voters in the center of the political spectrum, most Russian parties have little political history to abandon. As discussed in the introduction to this volume, most Russian political parties come and go with astonishing speed. In 1993, thirteen parties ran on the ballot for the parliamentary elections to the Duma. By the 1995 Duma elections, forty-three parties appeared on the ballot, which dropped to twenty-six parties in 1999. Most of the parties are small, poorly funded, and tend to be organized around an individual or small group of people. Even relatively well-organized and wealthy parties such as Russia’s Choice often self-destruct. The strongest party organization in Russia remains the CPRF, the successor to the Communist Party of the Soviet Union. Although fieldwork to discover actual party organization elements, including membership, offices, and contributions, is difficult in Russia, the Communists have perhaps the best claim to a permanent party organization throughout the regions and in the center as well as an official party platform. Both Yabloko and the Liberal Democratic Party of Russia (LDPR), which are described in the book’s introduction, have also survived to run in five elections, but generally are not characterized as having as strong an organizational base as the Communists.2 Thus, the Communist Party, the LDPR, and Yabloko are the “troika” of Russian politics, the only electorally successful parties to have lasted since the 1993 elections. The other dominant players have been parties created by powerful leaders or institutions, either “parties of personality” or “parties of power.” In each election, the Kremlin has promoted a different “party of power” with various leaders to forward its interests. So far, those parties have never managed to maintain their ties with the Kremlin until the next parliamentary election. In 1993, the party of power was Russia’s Choice, in 1995 it was Our Home is Russia (NDR), and in 1999 it was Unity. Evidence presented in this chapter suggests that the Kremlin has done a better job in each election in creating a political party from scratch, from the selection of leaders to the crafting of the party ideology to the marketing of the party via television. Although the Kremlin’s party of power received fewer party-list votes and fewer single-member constituency seats in 1995 than in 2
In the 1999 Duma elections, the LDPR ran under the name Zhirinovsky Bloc due to a legal challenge that charged the party organizers with a failure to report all of their assets. The complaint was settled and withdrawn on the eve of the elections. As Vladimir Zhirinovsky is widely recognized as the leader of the LDPR and the party is still referred to by its original name, it will be called the LDPR in this chapter.
TV, Voters, and Development of the “Broadcast Party”
33
table 2.1 National Television Channels in Russia, 2001 Daily Viewership (percent of population)
Channel Name
Ownership
1
Russian Public Television (ORT)
82
2 3
Russian Television and Radio (RTR) TV-Center
51 percent owned by state, rest owned by mix of public and private corporations State-owned
15
4
NTV
5
Culture
6
TV-6
Funded primarily by City of Moscow Commercial station, now controlled by state interests State-owned cultural channel created by presidential decree in 1997. Only channel that does not carry advertising. Commercial station
69
51 8
19
Source: Author’s research; daily viewership is from April 2001 survey cited in chapter.
1993, it can be argued that its ability to succeed even when the presidential administration was extremely unpopular (as in 1995) showed increasing skill in campaigning. In addition, Unity managed to run almost even with the Communist Party in the 1999 Duma elections, providing a smoother path for Putin’s victory in the presidential race a few months later.
the broadcast media and broadcast parties What features of these Russian elections suggest that some parties were “broadcast” parties? The first element that needs to be considered is the role of television in Russian society. Russia has a mixture of state-run and commercial television (Table 2.1). Both state and commercial channels, except for the state cultural channel broadcast on Channel 5, accept paid advertising, including political advertising. The three major television channels are state-controlled Channel 1 (Russian Public Television or ORT), state Channel 2 (Russian Television and Radio or RTR), and the commercial NTV on Channel 4.3 While both ORT and RTR are state run, only RTR 3
The abbreviations come from the Russian names – Obshchestvennoe Rossiskoe Televidenie or ORT and Rossiskoye Televidenie i Radio or RTR. NTV (Channel 4) was thought by many observers to stand for Nezavisimoye Televidenie in Russian or “Independent Television,” but the station officials claimed in an interview with the author in December 1999 that the initials do not stand for “independent television.”
34
Sarah Oates
is still 100 percent owned by the state. TV-Center (Channel 3) is funded by public money as well, but in this case the station is supported primarily by the Moscow City administration. The ORT channel remains the broadcast flagship of Russia and the most-watched channel for Russian citizens.4 The major commercial television stations are NTV (Channel 4) and TV-6 (Channel 6). The NTV channel was founded and controlled by a large media holding company (Media Most) and had been relatively committed to fair news coverage. However, Vladimir Gusinsky, the majority shareholder in the media group controlling NTV, was arrested by the Russian government on charges of fraud in 2001. Eventually, Gusinsky was forced to give up his share in NTV. At the time, many of the famous news personalities in the station left to set up a news team on TV-6, which has less penetration than NTV. TV-6 was then controlled by media mogul Boris Berezovsky, once a close presidential adviser, who had become a powerful political figure in his own right. Berezovsky was in turn forced out. Television, as elsewhere, is an enormously important medium in the postSoviet political sphere. Developed aggressively as a propaganda tool by Soviet leaders (Mickiewicz 1988, 1999), television reaches roughly 99 percent of the far-flung population of Russia. According to the 2001 survey, about three-quarters of Russia’s roughly 140 million residents can receive both the major government and commercial channels. About a quarter of Russia’s viewers, mostly those in rural areas, cannot receive NTV, which was significant when NTV often challenged the presidential administration’s spin on news before its takeover in 2001.5 Politicians, media owners, and journalists apparently are now reconciled to the idea that programs on television channels, especially news and analytical shows, will reflect the political agenda of those who control the station.6 The two powerful state-run television channels, ORT and RTR, are controlled by the presidential administration, significantly reducing 4
5
6
The survey was of 2,000 people across Russia. It was conducted by Russian Research Ltd. (Moscow and London) from April 10 to April 26, 2001, under the direction of the author, Professor Stephen White, and Dr. John Dunn (also of the University of Glasgow). A notable exception to this was when NTV chose to support Yeltsin’s campaign for president in 1996 due to fears of a Communist victory. This decision and the way NTV chose not to report on key factors in the campaign, such as Yeltsin’s failing health, are not a part of this study although an interesting overall part of media development in Russia (see Mickiewicz 1999). I base these comments not only on objective studies by groups such as the EIM that showed marked media bias in elections, but also on interviews with broadcasting directors, journalists, and analysts during the 1999 Duma campaign. Although often reluctant to admit it openly, a general air of understanding was that those that “sponsored” the station were able to dictate the nature of important political coverage. For details of this research, see Monitoring the Media Coverage of the December 1999 Parliamentary Elections in Russia: Final Report, March ¨ 2000, Dusseldorf, Germany: European Institute for the Media.
TV, Voters, and Development of the “Broadcast Party”
35
the amount of unbiased information available to viewers throughout the Russian Federation. In addition, it is clear that the content on NTV is now more pro-Kremlin than it was before its takeover in 2001, although there has been little chance for a systematic study of any changes in content after the switch. The print media and radio report some problems of governmental influence as well, although they are not as powerful as television and as a result are under less pressure from political influences. Freedom of speech is technically protected by the Russian constitution, but financial tactics, political pressure, and even violence are commonly used against journalists who probe too deeply into sensitive matters. According to the 2001 survey, 77 percent of Russians watch national television daily,7 compared with just 12 percent who read national newspapers every day. Nationwide, ORT remains the most popular television channel. When asked to state their favorite television channels, 37 percent of the respondents picked ORT, 26 percent picked NTV, and 14 percent picked RTR. However, when viewers were given a choice between state-run television and the flagship commercial channel, NTV was more popular. In areas that received both ORT and NTV, 33 percent chose NTV as their favorite, followed by 31 percent who selected ORT. The survey showed that NTV is more popular among urban, wealthier Russians who live in Moscow or St. Petersburg. The survey also revealed two reactionary trends in attitudes toward the mass media in a post-communist society. First, national state television remains immensely well trusted and, in fact, it was selected as “the most unbiased and reliable” source of information by 65 percent of the respondents in the 2001 survey. Only 18 percent picked national newspapers as the most unbiased and reliable source of information, with the same level of confidence in national commercial television shared by just 13 percent of those surveyed. At the same time that many Russians are generally untrusting of commercial television, because they fear that it pursues only the interests of its owners, they are uncomfortable with the notion of television as a societal watchdog. In addition, the survey revealed a clear relationship between television habits and preference on vote choice. For example, those who picked ORT as their preferred station were more likely to vote for the pro-Kremlin broadcast party Unity in 1999. In addition, Unity voters were more likely to say (61 percent versus 43 percent) that national state television was the most important influence on their vote choice. There were discernible differences in voting patterns based on a preference for state-run or commercial television as well: While 23 percent of those who preferred ORT voted for Unity, only 18 percent of those who favored NTV supported this pro-Kremlin party. 7
In fact, the respondents reported an even higher level of television viewership when specific stations were mentioned, as 82 percent claimed to watch the main state channel (ORT) daily.
Sarah Oates
36
The effect was even more marked in the presidential elections, in which ORT fans were even more likely (62 percent) to vote for Putin than NTV fans (53 percent). In addition, Putin voters were more reliant on state television as the most important source for vote choice, as 63 percent of the Putin supporters chose state television as the most important source of information in the campaign, compared with 51 percent of those who didn’t vote for Putin. Many Russians may be content with television coverage in contemporary Russia, but the news it carries isn’t particularly objective. All the major television channels had biased coverage of the 1999 Duma elections, ac¨ cording to a study by the EIM in Dusseldorf, Germany. The institute hired Russian researchers to time the coverage of political parties and major political figures on all the programs on the major channels during the elections. The report found that television channels had different patterns of coverage, each favoring a particular party or parties (Figure 2.1). For example, although ORT and RTR as state broadcasters had what the EIM report termed a “particular responsibility to provide impartial and fair information” (EIM March 2000, 2), bias was clear in new programs and through attacks on political opponents in the chatty news analysis shows. The ORT
80
Percent of Time on News
60
40
ORT
20
NTV TV-Center
0
Unity
Fatherland Communists
LDPR URF
Yabloko
Party figure 2.1 News Coverage of Political Parties in the 1999 Russian Duma Campaign. Source: EIM, from data gathered from November 28, 1999 to December 18, 2000.
TV, Voters, and Development of the “Broadcast Party”
37
channel devoted more than a quarter of its election news coverage to Unity. While this news coverage tended to be positive, the coverage of rival party Fatherland-All Russia (OVR) tended to be negative in tone (EIM March 2000, 2). At the state-funded television stations, the message was clear: Support the government branch that funds the station. While ORT officials declined to meet with EIM media monitors during the 1999 campaign and a top RTR editor claimed his station had no bias,8 the quantitative monitoring makes it clear that the pro-Kremlin party, its leaders, and its interests received more and favorable treatment on Channels 1 and 2 during the 1999 Duma campaign. At Channel 3, where TV-Center is funded by the Moscow administration, the political party led in part by the Moscow mayor (Fatherland) received extensive, positive political coverage. Even stations that have, from time to time, expressed a commitment to objectivity were supporting particular political causes. For example, NTV is a commercial station, yet it gave preferential treatment to Fatherland in 1999 because it was being ignored or covered negatively by Channel 1.9 How can political parties use television to promote themselves? Russian parties have essentially three different ways to campaign using television: through their free-time allotment on state-run channels, paid advertising, and editorial coverage such as news and analytical programs. Until the 1999 Duma campaign, free time was distributed under a highly equitable system as each party received an equal amount of free time with no regard for its actual or projected amount of electoral support.10 Paid advertising is technically open to all; in fact Russian law requires media companies to sell it to all parties at the same rate. Most political parties, however, cannot afford to advertise on national television, where ad rates reached as high as $40,000 a minute during prime time on ORT during the 1999 Duma elections. Very few political parties received any serious editorial attention. To be fair to the journalists, most Russian parties are so small and new that they have negligible news value as well as little ability to interact professionally with the media. However, a far more worrying problem for the future of Russian democracy is the sort of overt bias shown to various political parties on Russian television (Figure 2.1), as well as unprofessional mud slinging on talk shows such as The Sergei Dorenko Show on Channel 1 (ORT). Thus, unless a party was supported by the station’s political
8 9 10
Interview by author, Moscow 1999. Interview with NTV’s Deputy Editor-in-Chief Vladimir Kulistikov, Moscow, December 1999. During the Duma campaign in 1999, the Russian electoral commission announced that any party receiving less than 2 percent of the popular vote would be asked to reimburse media outlets for its “free” time. As a result, some small parties chose not to use their free time, although there was some ambiguity about how the electoral commission could force insolvent parties to pay their bills.
38
Sarah Oates
backers, it was unlikely even to get its fair share of coverage on national television. In this situation, how does a Russian party create itself over the television airwaves? There are two sets of criteria for the success of a broadcast party over a conventional party. Certain preconditions must exist in a society, conditions normally found in a postauthoritarian state, before the media can essentially take over functions of political party aggregation. These preconditions would include: 1. Relatively unfixed voting preferences on the part of the electorate. If partisan allegiances are formed relatively early in life and are fairly enduring, fewer “floating” voters would be available to vote for a new party. Hence, Ross Perot’s run for the U.S. presidency was relatively limited in 1992 despite a large amount of financing, paid political advertising, and a platform that appeared to appeal to many American voters. But in Russia, as in some other post-communist states, a large percentage of people still make up their minds during the campaign. For example, in the 2000 presidential elections, 48 percent of the voters made up their minds in the two months before the vote and a surprising 22 percent of those who voted claimed that they waited until the last week or later to decide on their vote choice.11 While there is evidence of some links between certain attitudes and social characteristics and vote choice in Russia (e.g., Chapters 3 and 5 in this volume), both the voters and the parties remain far more in flux than in the United States or Western Europe. 2. A chaotic and rapidly changing political situation in which political parties have little real power. Political parties were robbed of a great deal of potential power by the 1993 Russian constitution. Not only is the executive branch strong enough to stifle the Duma (in which only half of the seats are elected directly through parties), but the office of the president has remained relatively free of party affiliation. Thus, the link between party support and real political power is quite tenuous. 3. The lack of an independent media. In Russia, the two largest state television networks have been controlled by the government at the time of elections. The president has the ability to hire and fire the heads of the stations (a power he has exercised freely in the past). Perhaps even more importantly, there is no tradition of journalistic freedom or protection in Russian society. Journalists must cooperate if they want to keep their jobs or, in some recent cases, their liberty or even their lives. While there are commercial television stations in Russia, they 11
Figures from a March 2000 survey of 1,600 respondents in Russia, commissioned by Stephen White and Richard Rose. For more results from this survey, please see Chapter 3.
TV, Voters, and Development of the “Broadcast Party”
39
either have been overtly controlled by owners (as in the case of TV-6) or have tended to take sides in political battles and fail to always meet the standards of objectivity (as in the case of NTV). It should be noted that NTV was still a commercial news outlet relatively independent from government control at the time of the elections under study in this volume. If the preceding are preconditions, then the parties must exploit those conditions in order to succeed as broadcast parties. The parties must have: 1. The ability to influence news coverage and content on at least one national television channel, preferably ORT (Channel 1). It is only strong television presence that allows the party to develop an image rapidly before the voters, particularly as broadcast parties are founded within months of the Duma elections. 2. Sufficient money not only to produce professional advertisements, but to buy airtime on national television. While editorial influence is no doubt more important, paid political advertising allows for short, regular exposure for the viewers. In addition, parties are at liberty to control the content and message in the paid advertising. 3. No track record or burdensome past to explain. This means the party has more flexibility in developing a media image when it is quite new. The 2001 survey in Russia showed that new parties could have a special appeal. Although only 4 percent of the respondents said that they voted for a particular party in the Duma race in 1999 because it was a “new political force,” 15 percent of the Unity voters chose this pro-Kremlin party for this reason. 4. An institutional structure that can be used for the mechanics of the campaign. This can range from the actual Russian governmental apparatus with its far-flung regional offices to the powerful Moscow administration to perhaps a large business concern. Not only does this provide ready manpower and office resources, but it also gives the party a cadre of motivated political supporters whose livelihoods are linked to the political power of their bosses. 5. The ability to target policies and platforms, even if quite vague, at the relative center of the voting preference spectrum. This was a problem with the government’s first attempt at the broadcast party, that is, Russia’s Choice in 1993. Russia’s Choice was trying to promote promarket, Westernized ideas to a population that was experiencing grave financial hardship because of the transition from a socialist economy. In addition, Russia’s Choice was often critical of the Soviet past and of nationalist ideas, both of which still have great positive resonance for much of the electorate. Thus, a party must be at least partially in touch with the mood of the population to succeed even as a broadcast party.
40
Sarah Oates Yet there is no onus on the broadcast party to deliver on any promises – from those involving the market to foreign defense – once it is in office. Instead, the party will fade and another will appear to promise popular policies in the next election. Yelena Rykovtseva, a Moscow columnist who writes on television, coined the phrase “yefernaya partiya” (airwaves party), pointing out that parties that don’t exist outside of a stream of flickering electronic images are, in the end, merely a reflection of the popularity of television rather than a real political phenomenon.12
There are several items missing from the preceding list that one would expect to find necessary for party development. In particular, broadcast parties do not need an ideology or even a particularly coherent set of policies, but they do need good slogans and good sound bites for use on “their” media outlets. That is not to say that ideology does not play an important role in elections and even for the relatively ephemeral broadcast parties. In fact, it is an important point that the broadcast parties have arguably become more sensitive to the will of the electorate, changing from an unpopular message that pushed a market economy to one that emphasizes concerns over Russian nationalism, security, and social guarantees. The problem remains the lack of accountability. Even if the broadcast party now espouses an ideology that better reflects the desires of the electorate, the nature of the party is such that there is no guarantee of delivery on broad ideology, promises, or policies in office. It must be emphasized that not all parties in Russia are broadcast parties. As discussed previously, the Communist Party, Yabloko, and the LDPR have all existed through several elections in Russia, winning party-list seats, gaining seats through single-member district (SMD) contests, and working as a group (with some defections) in the parliament. The point of this chapter is to point out how broadcast parties have been surprisingly successfully compared with these more institutionalized parties. Thus parties that have some or many of the characteristics of Western-style parties exist side by side with the more ephemeral broadcast parties. What is important is how well the broadcast parties have done at the expense of the more institutionalized, structured party organizations in Russia. In addition to the preconditions listed previously, there are other elements of a traditional party system that are not necessary for broadcast parties. Broadcast parties don’t need grassroots organizations, party regional offices, incumbency in office, or a political track record. In fact, any of those elements could get in the way of an effective television campaign, taking away flexibility from political party image makers as they construct broadcast advertising 12
Rykovtseva made this comment during an interview with the author in Moscow in December 1999.
TV, Voters, and Development of the “Broadcast Party”
41
and news campaigns. It could not even be argued that the party needs telegenic or newsworthy leaders as the ability of the party to get positive coverage on the nightly news and other programs is more important than coverage based on personality or charisma.
the development of the broadcast party in russia To an observer from the West, the concept of a broadcast party might seem untenable. How could people vote for a party that has just sprung into existence, a party for which one could not even readily define a central ideology or set of policies? Yet the evidence is strong that the broadcast party has become a successful political entity in Russia, particularly as those in the position to create such parties have learned from previous mistakes. Another argument about this theory is that it may be difficult to delineate between a “party of power” and a broadcast party. However, it can be argued that the concept of a broadcast party could provide more understanding of the critical elements of this particular political phenomenon – as well as make it easier to export the concept to elucidate the relationship between television and political parties in other countries. In the first Russian Federation parliamentary elections, the use of television by political parties generally was not characterized by an “Americanized” style (Oates 1998 [a]). Rather, with little experience in political marketing, most parties opted for a Soviet-style focus on the leader and speeches. The results of the 1993 election were surprising, particularly as the Kremlin had confidently expected its “party of power” (Russia’s Choice in this election) to sweep the contest. In the end, it was the nationalist LDPR and the Communists that did particularly well, although progovernment groups turned in a respectable showing. Performances by political parties were noteworthy for various reasons. Russia’s Choice used Westernized techniques in commercials, trying for modern images. Meanwhile, Russian politicians such as outrageous ultra-nationalist Vladimir Zhirinovsky went in an opposite direction, using his fervent speaking style to whip up support for his proSlav policies using both his free time and a large amount of paid television advertising. The Communists had virtually no presence on television and no paid advertising, according to an EIM report on the elections (EIM 1994) and the author’s own research (Oates 1998 [a]; Oates and Roselle 2000). Russia’s Choice was the first attempt to build a broadcast party. It met all of the criteria listed. It existed in a situation of little party preference, the country was experiencing chaotic change, the Kremlin had fairly good influence over Channels 1 and 2, and it had ample resources to pay for political advertisements on television. In addition, NTV had yet to set up its national network, so there was no serious challenge from commercial coverage to the government’s line on television. Russia’s Choice pursued the most aggressive media campaign in the 1993 elections, spending a great deal
42
Sarah Oates
of money on paid advertising.13 In the end, though, the LDPR won almost a quarter of the party-list vote, with Russia’s Choice running second with just 16 percent. The Communists, despite virtually no television presence, received 12 percent of the vote. What had Russia’s Choice done wrong? In hindsight, there were several obvious problems. Most importantly, it seems that the Russian government, in formulating a liberal, promarket party, failed to realize that such positions were far from those of the many voters who were deeply disenchanted with the notion of a market economy as they lost their jobs, struggled to survive after the freeing of many price controls in 1992, and dealt with steep inflation. In addition, the party had missed the importance of the allure of national pride and the emotional appeals to nationalism. No matter how much news time was devoted to party leader and prime minister Yegor Gaidar and to the party itself, it was not an attractive political product to many voters. Thus, despite its support among urban, wealthier, more educated voters and an extensive campaign, the party did not manage the overwhelming victory it had envisioned. It is clear that the Kremlin learned from its mistakes and produced an arguably more palatable party by the 1995 elections in the form of NDR. The party, despite an early marketing misstep,14 presented a far slicker and more professional image than Russia’s Choice. There were several key factors in the change. First, the party was again led by a prime minister, but this time the more popular and telegenic Viktor Chernomyrdin. The party policies were no longer overtly Western and promarket, although certainly more liberal in some respects than the more nationalist and communist party groupings in the country. Their ads now were more Red Square than Madison Avenue, the main spots using Soviet images such as a rocket launch, sparks flying in a factory, and Russian children marching off to their first day of school.15 What did continue from 1993 was the lopsided domination of television by this progovernment party, as NDR had almost a quarter of both the advertising time and the editorial coverage in the 1995 elections.16 The party won about 11 percent of the party-list vote in 1995, less than Russia’s Choice in 1993. However, it could be argued that the party was more successful, in the sense that there were more than three times as many parties in the race than in 1993, Yeltsin’s popularity had plummeted, and the government was pursuing 13
14
15 16
According to the EIM report (1994), Russia’s Choice bought 281 minutes of advertising in 1993, or about 30 percent of the total campaign advertising time, and received almost a third of all editorial coverage during the campaign. The party had adopted the slogan and image of a “roof” as its party motif until amused reporters at a press conference let them know that the word was Russian Mafia slang for protection money. From the author’s study of political advertising in the 1995 Duma elections. Monitoring the Media Coverage of the 1995 Russian Parliamentary Elections: Final Report, ¨ February 15, 1996, Dusseldorf, Germany: European Institute for the Media.
TV, Voters, and Development of the “Broadcast Party”
43
a then-unpopular war in Chechnya. The victors in the 1995 contest were the Communists, while the LDPR’s support was cut roughly in half from 1993. It could be argued that the triumphant performance of the progovernment Unity party in 1999 was a culmination of all the lessons learned in creating a successful broadcast party. The Kremlin was willing to continue to retool the party – from the way the leaders presented themselves to advertising images even to its underlying policies – in order to win votes. Certainly nothing is wrong or unhealthy about a party reinventing itself. After all, this is a process that parties such as the Conservatives in the United Kingdom in the 1970s and the U.S. Democrats in the 1990s have undergone. Yet Unity was not ultimately concerned with a better fit with the public’s wishes. Rather, Unity wanted to consolidate the power of the Yeltsin administration by using the media both to shape public preferences and to convince the electorate it was the best party to take care of their needs. The Interregional Movement “Unity” was created in 1999 to contest the Duma elections as NDR and its leaders had long since broken with the Kremlin. This time the party image makers chose a bear as the party’s symbol – partly because the first letters of the party’s long name in Russian spell out the Russian word for “bear” and no doubt partly because the bear is a popular symbol for Russians. In addition, the party looked slightly further afield than the Prime Minister’s office for leaders and found three personable, newsworthy individuals – Minister for Emergency Situations Sergei Shoigu, famous Greco-Roman wrestler Alexander Karelin, and Alexander Gurov, who headed the effort against organized crime in the Russian Ministry of Interior. Each iteration of the progovernment broadcast party didn’t completely disappear. Russia’s Choice reappeared under a slightly different name (Russia’s Democratic Choice) and stripped of its government backing to run in the 1995 Duma elections. It failed to reach the 5 percent vote barrier. A remnant of the original Russia’s Choice joined with a successful broadcast party in 1999 (discussed in the following text). The NDR party also ran in the 1999 Duma elections, but failed to get any party-list seats. As both parties no longer were useful to the Kremlin interests, they had only a fraction of the attention on the nightly news of each successive, new broadcast party. By 1999, the party image had shifted even more noticeably away from the slick, Westernized ideas used by Russia’s Choice in 1993. The imaging focused on the three leaders, particularly Shoigu’s activities in Chechnya. The party’s slogan of “Strength, Honesty, Worthiness” was backed up by images of Shoigu among the troops of Chechnya, Karelin throwing wrestling opponents to the ground, and Gurov chasing down criminals. In fact, the image makers avoided words as much as possible, using pictures of the three men respectively as soldier, victor in a wrestling match, and policeman. One
44
Sarah Oates
five-minute free-time spot for the party, aired on ORT in November 1999, is almost text free. The spot shows the men in action in their various roles with stirring music (a different theme tune for each), allowing them each to utter merely a single sentence about strength and worthiness in the final seconds of the ad. Once again, the “party of power” was given much attention on television. According to a study of the 1999 Duma campaign by the EIM, Unity received 16 percent of the news coverage across all the major Russian television channels. In addition, coverage became much more polarized in these elections. It was clear that various stations had political interests to foster: The state-run channels promoted parties close to the government line, Moscow-funded TVCenter supported the Moscow mayor’s Fatherland party, and the commercial NTV championed the Kremlin opponents (Figure 2.1). In fact, Unity had a fair amount of negative coverage, according to the EIM study. But where Unity really benefited was from extensive and virtually uncritical coverage of the invasion of Chechnya on state television. With Unity’s link to the government and the now-popular war effort there (especially Shoigu’s involvement), there was a powerful psychological advantage in the marketing effort for the party. In fact, Unity was able to break through a very important barrier for the 1999 elections and found a broader base of support than either Russia’s Choice in 1993 or NDR in 1995. Unity was relatively popular among rural as well as urban voters, according to postelection surveys by the Public Opinion Foundation.17 This was the first time a pro-Kremlin party had managed to garner such a large base of support from outside the urban centers, although the constituency for NDR in 1995 was not as urban as that of Russia’s Choice in 1993. The Kremlin was not alone, of course, in attempts to use the mass media to build support for new parties. In the past three Duma elections, most of the parties were new and needed the attention of the mass media to gain any sort of support. Two other parties in addition to Unity, Fatherland-All Russia, and the Union of Right Forces (SPS) also relied quite heavily on the media in the 1999 elections, albeit on different television channels. The Fatherland-All Russia party was created by urban and regional leaders, notably Moscow Mayor Yuri Luzhkov and Governor of St. Petersburg Vladimir Yakovlev. Fatherland was an attempt by powerful leaders to challenge the Kremlin’s domination and it had substantial media resources, particularly in the form of Moscow-funded TV-Center television on Channel 3. In addition, not only did Luzhkov enjoy a great deal of popularity, but the party recruited popular former Prime Minister Yevgenii Primakov as 17
For results of Public Opinion Foundation surveys, see the archive on their Web site at www.fom.ru.
TV, Voters, and Development of the “Broadcast Party”
45
coleader. Luzhkov, Yakovlev, and Primakov all could command media attention in their own right as prominent Russian politicians. Yet instead of a dialogue between leaders and a discussion of party platforms, Russian television stations wound up mostly favoring “their” politicians and parties rather than engaging in real news coverage in the 1999 elections. Quantitative monitoring by the EIM makes it clear that Unity, its leaders, and interests received more and favorable treatment on state-run Channels 1 and 2. For example, on ORT news programs, Unity was the leading election news story, receiving 28 percent of the news coverage. Also, ORT covered government figures quite heavily on the news: Prime Minister Putin and Shoigu dominated the news on key figures, with 42 percent and 19 percent respectively. In fact, Putin gave the “presidential” address that traditionally comes on the eve of Russian parliamentary elections even though he was not appointed acting president for another two weeks. Meanwhile, Moscow Mayor Luzhkov had just 4 percent of the coverage on the ORT news, while at his Moscow-funded TV-Center (Channel 3), Fatherland received almost three-quarters (71 percent) of the political party coverage on the news. On TV-Center, major political players such as Unity, Yabloko, and the CPRF were barely mentioned. It is certainly unsurprising that the main political figure covered on TV-Center was Luzhkov, with 37 percent of the coverage compared with just 15 percent for Putin. In addition, a new nastiness emerged in the 1999 campaign. Fatherland encountered formidable opposition from the Kremlin authorities, who were obviously displeased at Luzhkov’s attempts to broaden his Moscow popularity into a national power base. Luzhkov in particular was savagely attacked on a popular Sunday night news/talk show hosted by Sergei Dorenko. In a classic example of the new Russian style of kompromat (an abbreviation for “compromising material”), Dorenko presented a series of rumor and innuendo about Luzhkov’s alleged financial malfeasance. The show, the most popular on Russian television at the time, featured a string of poorly documented allegations as well as aerial photos of Luzhkov’s admittedly lavish estate outside Moscow just one week before the elections. In fact, Luzhkov successfully sued ORT and Dorenko for libel, winning 150,000 rubles (about $5,600) during the campaign, but the negative reports about him continued on Dorenko’s program.18 When the dust settled after election day on December 19, 1999, it was clear that the Moscow-grown Fatherland lacked the popularity of its Kremlin counterpart. Unity won 23 percent of the party-list vote, widening the constituency base of a “party of power,” while Fatherland garnered a respectable, albeit smaller, 13 percent of the party-list votes. Although also operating as a broadcast party, Fatherland was not as effective a broadcast 18
Radio Free Europe/Radio Liberty Newsline, December 6, 1999, archived at www.rferl.org.
46
Sarah Oates
party as Unity. Certainly Fatherland was able to use Moscow’s Channel 3 as a venue for virtual infomercials on the nightly news, but it did not meet all the criteria for a successful broadcast party. Because of its leadership, Fatherland was linked with the large cities, an unattractive prospect for the large percentage of voters who live in rural areas in Russia. In fact, surveys by the Public Opinion Foundation found that the party did not do well among rural voters. Fatherland was hampered from creating a Russia-wide political movement, even though it did relatively well with candidates in some SMDs. Perhaps the final “broadcast” success of the 1999 campaign could be attributed to the Union of Right Forces, a new party that earned almost 9 percent of the party-list vote in the 1999 elections. The party, headed by youthful former Prime Minister Sergei Kirienko, was noticeably more liberal in outlook, taking under its party wing Gaidar and the reformulated Russia’s Choice (now Democratic Russia’s Choice) as well as other relatively promarket politicians. Kirienko was not a particularly popular prime minister and he lasted only a matter of months. However, the party did advertise relatively heavily and received friendly treatment from the government. In a bizarre twist, Putin appeared with Kirienko in a paid political advertisement on the eve of the elections.19 The SPS party did not exist until just before the 1999 elections. Yet through the use of advertising, endorsement by the prime minister, and relatively favorable news coverage on the main television channels, the new party was relatively successful in 1999. One could argue that liberal, promarket voters turned to the SPS as a fairly pragmatic choice and its strength came from actual voter preferences. But it is difficult to see how many voters could have received much concrete information about the party during the campaign. Similar patterns were clear in the 2000 presidential campaign. This time, compromising material on state-run television channels was aimed at moderate candidate Grigorii Yavlinsky.20 But this was a minor violation of journalistic ethics compared with the fact that Putin received an overwhelming amount of news coverage during the campaign, close to 50 percent of the total for all national television channels together. Communist contender Gennady Zyuganov, although he ran a respectable second in the race with about 30 percent of the vote, received just 11 percent of the news coverage on the major Russian television channels. Coverage on the major state channels (ORT and RTR) continued to be biased in favor of progovernment forces, in this case Putin. Other channels, including the commercial NTV, TV-6, and the Moscow-funded TV-Center, were more balanced in their coverage, according to the EIM report on the elections. 19 20
Thus violating Russian law that states Russian government officials cannot use their positions to campaign for political parties. According to the EIM report on the 2000 presidential elections.
TV, Voters, and Development of the “Broadcast Party”
47
a model for creating successful broadcast parties Based on the preceding evidence, what did the Russian government learn about the creation of broadcast parties and candidates from 1993 to 2000? This chapter suggests the following points: 1. Political marketing in post-Communist states should be sensitive to the cultural values and norms of the post-Communist people. This seems quite simple and logical, but it was not universally accepted or understood in early post-Communist elections. Instead, parties such as Russia’s Choice in 1993 were more inspired by Western models than by an extension of the communist/postcommunist stock of imagery. It took the success of Zhirinovsky, with his own particular blend of showmanship and nationalist rhetoric, to show what sorts of political styles might appeal in Russia. In addition, it was important to understand that many Russian citizens have deep pride and attachment to the Soviet past. Although the Soviet regime had notable failures, particularly in the area of human rights, it had quite tangible achievements as well. To disregard or, even worse, to denigrate the Soviet past proved an unpopular tactic with much of the Russian electorate. 2. If you are going to try to influence the voters through television, use relatively subtle tactics to make sure the voter feels as if he or she were making a real choice. As Mickiewicz (1999) has pointed out, the post-Soviet viewer is a canny and careful viewer. He or she is bound to be unmoved by a copious amount of meaningless coverage, free time, or advertising. Rather, it is important to have the progovernment party or candidate persistently framed in a positive light on the television news and in other programs. This made the popular second invasion of Chechnya immensely useful to the progovernment Unity party in 1999 as well as to Putin’s presidential campaign in 2000. Both Unity leaders and the acting president could be presented as brave protectors of the Russian state. This also fits in well with a secondary theme to elections: Nationalist appeals are often useful. Although their claims were not as far-flung as some of those made by Zhirinovsky, the progovernment parties and candidates are now unafraid to denigrate the West while promoting pro-Russian national pride. 3. You need a relatively docile and universal television system for success. If one is going to market a new political party or leader to the people in a short time frame, then little time is available to build a reputation or to answer negative news reports. Instead, it is important to pursue a pervasive, “feel-good” campaign with positive news coverage (albeit trying to avoid outright propaganda) and easily digested political advertising. Thus, you have the “feel-good” advertisements by Unity and the decision by Putin not to irritate the voters with any free
Sarah Oates
48
time or paid political advertising. Putin did not need advertising. The primary state television stations already were providing ample footage during news and other programming. At the same time, it is helpful to make your opponents seem less credible through vague, yet damning reports of titillating personal or financial scandal, through the use of kompromat.
implications for the future What do the advent of the broadcast party, the relatively strong showing of Unity, and the victory of Putin mean for the future of Russian politics? Several points could be considered, but perhaps the most important one is that television should not replace real party organization. The ability of political parties to spring up just before elections, put across a “message” to the voters, and be voted into office is worrying. What responsibility do these parties have to their voters? What long-term stability can they offer? Isn’t the popularity of the political party little more than a reflection of the popularity of the television station that lends it support? When other political institutions are weak, the role of television becomes even more critical and prey to manipulation. It is easy to imagine that if the U.S. president were given a choice, he might well try to use direct appeals to the voters using television to circumvent problems with Congress. In fact, U.S. presidents consistently use television to appeal to the electorate, both during elections and for support for their policy initiatives. However, the U.S. president’s ability to effect direct change in legislation through media appeals is limited. In Russia, lacking stable traditions of the creation and passage of legislation, popular appeals through television can circumvent and even damage nascent political institutions. Some scholars, including a selection of those in this volume, are concerned about the direction of party development in Russia to the point that they perceive Russian elections as even detrimental to the development of a democracy. For example, Stephen Fish (Chapter 9) suggests that recent elections actually may have weakened party structures in Russia. In addition, Olga Shvetsova (Chapter 10) argues that the Duma elections really serve as a primary for the upcoming presidential elections in Russia, which undermines the true purpose of legislative elections. Stephen Hanson (Chapter 8) posits that Russian politics lack value-rational principles that provide organizers with long-term incentives and, as a result, parties are used for short-term gain rather than to build enduring political institutions. This chapter joins in the pessimism about Russian political parties as useful political organizations that aggregate the interests of the electorate. Rather, the success of broadcast parties suggests that parties often are used in Russia to promote the interests of a narrow group of elites, a subversion of the intended nature of political parties as tools of democracy.
TV, Voters, and Development of the “Broadcast Party”
49
To compound the problem, the concept of an unbiased media that serve the needs of the public has little respect in Russia. Rather, the media – and particularly television – more or less openly serve their political support base. Thus, ORT and RTR represented the needs of the Kremlin while TVCenter served the needs of the Moscow administration during the 1999 elections. Now that NTV and TV-6 are no longer controlled by different political forces, it is likely these two major commercial television channels in Russia will join the pro-Kremlin chorus. While television is unusually powerful in its ability to influence voter choice, at the same time television is presenting slanted news coverage. Most people in Russia have little ability to obtain objective information with which to make informed electoral choices. If citizens become more critical of the lack of objective information on television, and post-Soviet viewers are highly sensitive to propaganda efforts, trust in television will erode. In the long term, the Russian government, political parties, and leaders will lose their ability to communicate meaningfully with citizens via the most popular medium. While using television to promote political interests, such as the election of Putin or Unity, may pay off in the short term, it erodes a very important link with the public. Meanwhile, as television is focusing in on the politics of personalities, the voters are not getting involved in the really important debates of the day, in particular the future direction of the Russian economy. On a more sinister and cynical note, it could be argued that the broadcast party is more valuable to Russian leaders. A well-rooted party organization with regular meetings, rules, members, and other institutions should be able to control the leadership, at least to a degree. But if parties do not become institutionalized, and if broadcast parties come and go (most go quite quickly), leaders are free to forget about even the most vague campaign promises and to rule unfettered by any party constraints. Thus, Russian leaders have little impetus to build permanent party organizations. Essentially, the Russian government has used television in ways certainly not anticipated by the British model, which maintains relatively careful distance between the government and publicly funded channels. Nor do Russian viewers benefit from the type of media competition found in the United States, where the media is driven to a degree by commercial interests yet adheres to the broad concepts of plurality and objectivity in the news. Rather, the heady media freedom of the young Russian state has been replaced by service to political masters rather than its duty to provide information to the public. Over time, Russian officials and politicians have become much better at crafting broadcast parties, certainly as seen by the improvement in their party “product” from 1993 to 1999. In fact, the success of Unity and its campaign strategy was an important step in Putin’s successful bid for the presidency. In the longer term, this ability to control and manipulate television will be
50
Sarah Oates
an important part of Putin’s ability to maintain his support. At the same time, however, this type of control of television for certain political interests undermines its ability to inform and enlighten the public about real political choices and alternatives. Eventually, Russian citizens could be isolated further from their leaders if trust in television erodes and the ability to mobilize the electorate dwindles.
3 Parties, Voters, and Foreign Policy Stephen White, Neil Munro, and Richard Rose1
For Russians, it was not only their state that collapsed in 1991. It was also their international environment. It had begun to give way at least two years earlier, in 1989, when Mikhail Gorbachev made clear that the Soviet government would not intervene to prevent the election of non-Communist governments throughout Eastern Europe. Speaking to the Central Committee in December 1989, Gorbachev professed to welcome the “positive changes” that were taking place in the region, presenting them as a further stage in the “renewal of socialism” (Materialy 1989, 18–19). But in any case, he asked the 28th Party Congress in the summer of the following year, what was the alternative? “Tanks again?” (XXVIII s”ezd 1990). The international organizations that had held these states together collapsed during 1991, when the Council for Mutual Economic Assistance (Comecon) and the Warsaw Treaty Organization agreed to dissolve themselves. When Gorbachev resigned as USSR President on Christmas Day at the end of that year, he was the last Communist leader in Europe. The Russian Federation, with the agreement of the other post-Soviet republics, was the USSR’s successor state: it acceded to the USSR’s 15,000 treaties, to its international debt, and to its seats in the United Nations and within its Security Council. But it was scarcely a geopolitical successor. Post-Communist Russia was still the world’s largest state, but it was only three-quarters of the territorial extent of the USSR. Its population was only half as numerous. By 2000, after a decade of near-continuous contraction, Russia’s Gross Domestic Product (GDP) was still the twelfth largest in the world, but it had dropped behind Brazil, China, and South Korea as well as the larger Western democracies. For years, complained Kommersant, Russia 1
We acknowledge the support of the U.K. Economic and Social Research Council under grants R238107 to Rose, White, and Munro for “Consolidating Russian Democracy?” and ¨ L213252007 to White, Margot Light, and John Lowenhardt for “The Outsiders: Russia, Ukraine, Belarus, Moldova and the New Europe.”
51
52
Stephen White, Neil Munro, and Richard Rose
had been described as an “Upper Volta with rockets” (January 21, 1989, 1); now it seemed to have become a reality. There were still more complicated tasks of psychological adjustment. Russians, for a start, had to accept that they were no longer a superpower, but at best a regional presence in the international system. In effect, Russia had lost the Cold War. Their ideology was discredited. They had been marginalized in parts of the world in which their influence had once been considerable, including the Middle East and the Balkans. There was a real danger, as acting president Putin warned in his New Year address at the end of 1999, that the whole country could slide into the second, or even the third, rank of the members of the international community (Rossiiskaya gazeta December 31, 1999, 5). Speaking to the Federal Assembly after his election, Putin warned that demographic contraction, caused by falling birth rates and rising death rates, was threatening the survival of the nation itself (Rossiiskaya gazeta July 11, 2000, 3). Based on existing trends, Russia would lose a seventh of its population over the coming fifteen years. Our concern, in this chapter, is to examine the responses of Russia’s parties and politicians to these changing circumstances, using survey data that were gathered in January 2000 as part of the New Russia Barometer VIII (for details of the questionnaire and sample see Rose 2000). Did Russians, at the start of a new century, identify with what their first post-Communist foreign minister Andrei Kozyrev had called their “natural allies” in the Western democracies? Or were their views still colored by the Cold War, and perhaps still older antagonisms? Beyond this, we examine the positions of parties and politicians in the elections of 1999–2000, and use discriminant analysis to probe the strength of the association between foreign policy opinion and electoral behavior. As Almond (1950) had argued, ordinary citizens had an “extremely limited” grasp of international affairs, but foreign issues could become prominent when there was an “immediate threat to the normal conduct of affairs” (pp. 71, 80). At least one vital issue during the period of Russia’s two elections, the Chechen conflict, had obvious international implications; another was the relationship between Russians and other citizens throughout the former Soviet territory. How salient were such issues to Russian voters at the end of a turbulent decade in which their external environment had completely changed? And what difference did issues of this kind make to their electoral choices?
friends and foes in a post-communist world Foreign policy issues are normally not difficult to define. In post-Communist Russia, some are less straightforward. Chechnya, formally, was one of the subjects of the Russian Federation, and its attempted secession an entirely domestic matter. But the “war against terrorism” that resumed in August 1999 had obvious implications for Russia’s relations with other states and
Parties, Voters, and Foreign Policy
53
with international organizations such as the Council of Europe and the Organization for Security and Cooperation in Europe (OSCE), and accordingly it forms part of our discussion. A little further from home were those issues that affected relations with the Commonwealth of Independent States (CIS), or what Russians call the “Near Abroad.” Formally, again, these were foreign states, but at least 25 million Russians found themselves living there when the USSR collapsed, and relations with “compatriots” were an important part of the domestic agenda. Further still from home were issues that are uniquely the subject of state-to-state relations, such as perceptions of geopolitical threat, military alliances such as North Atlantic Treaty Organization (NATO), and membership of transnational bodies such as the European Union (EU). We conceptualize these sets of issues – internal affairs with an international dimension, the Near Abroad, and relations outside the former Soviet space – as three concentric circles of external influence on public opinion. For a start, did Russia have any friends or allies at the beginning of the post-Communist twenty-first century? There was no longer a “socialist community,” or a network of client states in Eastern Europe whose loyalty could be assumed – or, if necessary, imposed. Relations with still-communist Cuba, China, North Korea, and Vietnam were more distant than they had been in the days of proletarian internationalism. But there was, and still is, a Slavic community among Soviet successor states (Russia, Ukraine, and Belarus), and between peoples. There were more than eleven million selfidentified Russians living in Ukraine when the last all-union census took place in 1989, and more than six million in Kazakhstan. Altogether, more than 18 percent of all ethnic Russians lived outside their own republic. Relations with Belarus were the closest of all those with a foreign state in the post-Soviet period, cemented by a series of treaties that were intended to establish a union with a common citizenship and currency. Relations with Ukraine were more distant, reflecting disputes about the Black Sea fleet and territorial boundaries, although a friendship and cooperation treaty had been signed in 1997. Did these moves toward a closer Slavic union reflect a popular consensus, the tactical maneuvering of politicians, or collective national interests? The New Russia Barometer asked first of all what view our respondents took of closer relations with the other former Soviet republics. In each case, were they in favor, opposed, or indifferent? There was an implicit hierarchy among our responses that almost exactly mirrored the pattern of state-tostate relations (Table 3.1). Closest of all were the two other Slavic states, with more than two-thirds in favor of a closer relationship and virtually no dissent. Then came Kazakhstan and Moldova, again with very similar ratings; in the case of Kazakhstan this is likely to reflect its substantial Russian-speaking population, and in Moldova the existence of a predominantly Russian enclave in Transdniestria that has been defended by Russian arms. In both
Stephen White, Neil Munro, and Richard Rose
54
table 3.1 Attitudes Toward Closer Relations with the other Former Soviet Republics
Belarus Ukraine Kazakhstan Moldova Georgia Uzbekistan Kyrgyzstan Armenia Estonia Latvia Lithuania Azerbaijan Tajikistan Turkmenistan
In Favor
Against
Indifferent
69% 68% 55% 55% 49% 49% 48% 47% 47% 46% 46% 45% 45% 45%
5% 6% 9% 9% 12% 11% 10% 12% 15% 15% 15% 13% 13% 13%
26% 26% 36% 36% 39% 40% 42% 41% 38% 39% 39% 42% 42% 42%
Source: New Russia Barometer VIII, conducted by the AllRussian Center for the Study of Public Opinion (VTsIOM) fieldwork January 19–29, 2000, n = 1,940; for fuller details see “A Note on Data.”
cases just over half of our respondents favored a closer relationship; just under half favored a closer relationship with all the other former Soviet republics, with little difference between Central Asia and the Baltic. While the wish for closer relations was generally strongest where the proportion of Russians in the local population was the highest, this was not so in the case of the Baltic states, where up to a third of the population in the 1989 census were postwar Russian settlers without Baltic citizenship rights. How closely did Russians associate themselves with what was happening elsewhere in post-Soviet space? And what would they do if fellow nationals were under threat in another of the former Soviet republics? There were family links, for a start: 36 percent had one or more relatives in at least one other member country of the CIS. People had traveled to other CIS states: between 12 and 15 percent had visited at least one of the Baltic republics, which was more than had visited the East European countries that were formerly under Communist rule – just 3 percent had visited Hungary or Czechoslovakia, and 5 percent had visited Poland. Positive views of a closer and more formal association among the former Soviet republics, accordingly, were widely supported. Half of our respondents thought it would be a good idea if all the countries of the CIS united in a single state, 17 percent disagreed, and 33 percent had no opinion. The USSR remains a popular concept, with up to 85 percent deploring its demise more than a decade after it had taken place (Novye izvestiya January 30, 1999, 1). There is, however, no significant
Parties, Voters, and Foreign Policy
55
table 3.2 Support for Various Forms of Action in Defense of Russians in Other CIS Countries
Negotiations Resettle in Russia Economic Pressure Noninterference Military Intervention
Completely Agree
Mostly Agree
Mostly Disagree
Completely Disagree
66% 27% 27% 13% 7%
28% 44% 37% 23% 13%
4% 22% 29% 37% 38%
2% 7% 7% 27% 42%
Source: New Russia Barometer VIII, fieldwork January 13–29, 2000, n = 1,940.
correlation between having relatives in other former Soviet states and views on unification. What if it cost money to establish a new union state, or to bring the living standards of its prospective members closer together? And what if Russians were asked to take some kind of action to defend their fellow nationals in the other newly independent states? Understandably, there was much less enthusiasm. Only 6 percent thought levels of support to the other CIS countries should be increased, and 27 percent thought existing levels of support should be maintained. Two-thirds thought Russian support should either be reduced (37 percent) or eliminated altogether (30 percent). And there was some reluctance to take decisive action if fellow nationals in the other former Soviet republics were under threat (Table 3.2). Fully 80 percent of our respondents were against military intervention on behalf of Russians in other CIS states. Fifty-four percent supported economic sanctions and 71 percent favored allowing ethnic Russians in other CIS states to resettle in Russia. The most popular option of all was to negotiate with the governments concerned as a means of addressing problems of interethnic relations. Faute de mieux, geography assigns Russia west of the Urals to the European continent, but Russian attitudes toward Europe have always been ambiguous (Baranovsky 2000). Half of the New Russia Barometer respondents thought of themselves as Europeans and half did not. And 84 percent supported the proposition that Russia should develop in accordance with its “own traditions” rather than “along the lines of the countries of Western Europe.” The EU is an unknown entity to a majority of Russians. When the New Russia Barometer asked if Russians could identify which of a list of five major European capitals was the EU headquarters, only 31 percent were able to name Brussels, and 60 percent readily admitted they had no idea. Questions asking about relations between the Russian Federation and the EU generated a disproportionate number of “don’t know” responses, indicating that most of these answers were casually considered rather than a reflection of settled thought. Thus, 21 percent approved of the aims of the EU compared to 10 percent disapproving – but 22 percent said they didn’t know enough to
56
Stephen White, Neil Munro, and Richard Rose
comment and 47 percent found it hard to say. Would Russia benefit from a closer association? About half guessed that it would, 12 percent took the opposite view, and 38 percent thought it would make no difference. In the focus group discussions that helped to shape our survey questions, we found a very hazy impression of the implications of an enlargement of the EU that would include Poland, but leave Russia outside. These implications are likely to include the imposition of a tariff barrier to the East and visa requirements along an external border between an enlarged EU and the Soviet successor states. The Kosovo crisis of spring 1999 had driven a wedge between the Russian government and its Western counterparts. The NATO bombing of Serbia, without the explicit sanction of the United Nations, was for Boris Yeltsin an act of “undisguised aggression” that risked a still wider Balkan conflict (Rossiiskaya gazeta March 26, 1999, 2); Russian Foreign Minister Igor Ivanov spoke of “unconcealed genocide against the peoples of Yugoslavia” (Diplomaticheskii vestnik no. 4, 1999, 18); and relations with NATO were suspended. Press coverage, in the immediate aftermath of the bombing, was sharply anti-Western; and Russians themselves were the most firmly opposed to the bombing in a seventeen-nation survey conducted in late March and early April 1999 (The Economist April 24, 1999, 48). Less than a year later, however, these passions had subsided. Less than a fifth of respondents identified with the Serbs, who were Russia’s traditional allies in the region and part of the Orthodox Christian community; an insignificant number took the side of the Kosovar Albanians. The median respondents (42 percent) had no sympathy for either side, and 36 percent were “not at all interested.” Most Russians, in fact, had favored a joint attempt to find a peaceful solution to the crisis, and once it had passed they expected their relations with the West would return to normal (Monitoring 1999, 60). Three-quarters of Russian respondents did not expect a military attack in the coming five years, while 24 percent thought it very likely or probable. But if there was an external threat – and both the national security concept, approved in January 2000 (Sobranie 2000), and the military doctrine, approved three months later (Rossiiskaya gazeta April 25, 2000, 5–6), assumed there was – where did it lie? In Europe, most obviously – if previous history was any guide – in Germany? In China, with its rapidly increasing population in areas that immediately adjoined the Russian border? Among the Islamic states, including rogue regimes like Iraq? Among other members of the CIS? Or did it still emanate from the United States, in spite of the end of the Cold War and the establishment of what both sides described as a “partnership” between them? Of all the options we presented, Ukraine was the least likely to be seen as a threat to Russian security. A substantial majority saw no threat at all from Germany, in spite of the wartime experience (Table 3.3). Large majorities saw little or no threat from China, in spite of its population and level of military preparedness. The countries of the EU are seen as no threat by half of the
Parties, Voters, and Foreign Policy
57
table 3.3 Threat Perceptions
United States Non-Russians in Russia China European Union Iraq Germany Ukraine
Great Threat
Some Threat
Little Threat
No Threat
17% 15% 5% 5% 5% 2% 2%
32% 29% 19% 18% 20% 13% 7%
24% 23% 22% 25% 23% 22% 15%
27% 33% 54% 52% 52% 63% 76%
Source: New Russia Barometer VIII, fieldwork January 13–29, 2000, n = 1,940.
respondents, as is Iraq, one of the USSR’s oldest treaty partners. But nonRussians living in Russia – most obviously Chechens – were seen as a threat by just under half of respondents. In spite of the ending of the Cold War, 17 percent saw the United States as a great threat and 32 percent as posing some threat. However, it was also the country with which the largest proportion of Russians thought it important to have good relations; and attitudes became still more favorable in the aftermath of the September 11 outrages. The Putin government’s initial military operations in the Caucasus were popular; 72 percent supported the campaign entirely or in large part; 20 percent had many criticisms; and only 8 percent definitely opposed it. Boris Yeltsin was blamed a great deal for the war by 50 percent – about as much as the Chechens (47 percent) and “extremists from Middle Eastern countries” (44 percent). A significant minority (13 percent) thought much of the blame should be attributed to the United States. A large majority showed no willingness to take part in the military action, or to allow family members to do so. Just 12 percent were prepared to take part in the campaign themselves, or to allow close relatives to do so on their behalf, and 18 percent insisted that their age or state of health made it impossible for them to become involved. The remaining 70 percent simply refused to contemplate the possibility. Opinion on the longer-term future of Chechnya was ambiguous. How, for instance, would the war come to an end? A total of 39 percent thought it would conclude with the defeat of the Chechens and the reincorporation of the rebel republic. But nearly as many (35 percent) expected a protracted war that might spread across the entire North Caucasus. The median group (20 percent) expected a stalemate, as in 1996, with heavy losses on both sides, and the remainder expected the partition of Chechnya at the Terek River. Nor was there popular commitment to the position of the federal government to maintain the territorial integrity of the Russian state by whatever means were necessary. Some (15 percent) thought Chechnya had, in practice, already left the federation and 18 percent said they would be “delighted” if it did so.
58
Stephen White, Neil Munro, and Richard Rose
Around the same number (21 percent) were unconcerned either way; and 23 percent were against separation in principle, but prepared if necessary to accept it. Only 23 percent were in favor of resisting Chechen independence by all available means. Respondents had a better idea of what NATO was than of the EU: 60 percent correctly identified it as an American-led military alliance that included Western Europe and Canada. But they were relatively relaxed about the expansion of the alliance, both the expansion that had taken place and the further enlargement that appeared to be in prospect. Was there, for instance, any threat to Russian security from the adhesion of the Czech Republic, Hungary, and Poland that had taken place in spring 1999? Fewer thought there was a threat (28 percent) than took the opposite view (36 percent); but 36 percent had no idea. And what about a further expansion that included the Baltic republics, which would then become the first former Soviet republics to join? The central government in Moscow is vehemently opposed, but no more than a third of our respondents thought it would be a big threat (14 percent) or some threat (22 percent) to Russian security; once again the median respondents (34 percent) had no opinion. There was even some support among respondents for Russia itself to become a member of the alliance. While a third of respondents thought it would make no difference, slightly more (38 percent) thought it would enhance Russia’s security, while 29 percent took the opposite view. What were the implications of these views for Russian military spending, and the continuing process of disarmament? When respondents are asked which of a range of spending priorities they favor, social protection and health receive more support than defense. Nevertheless, our survey showed that in spite of the ending of the Cold War and great difficulty in paying pensioners, teachers, and nurses, higher levels of military spending received support: 27 percent would “spend a lot more,” and another 33 percent would spend “a little more”; only 15 percent wanted to spend any less and the remaining quarter wanted to leave spending at the same level. Any suggestion that Russia should unilaterally relinquish its nuclear weapons was strongly opposed. More than a third (37 percent) disagreed somewhat, a further half (46 percent) disagreed entirely, and only 17 percent offered some measure of support. Both of these views coincided with the official position of the Putin government, which had increased the military budget and lowered the nuclear threshold in its new national security concept to allow for a threat to the existence of the Russian state as well as for a nuclear attack from outside.
parties and their foreign policy preferences In principle, elections in Western nations offer voters an organized choice between policies, including foreign and security policy (cf. Holsti 1986). To
Parties, Voters, and Foreign Policy
59
what extent did the 1999 Duma election – the primary focus of our analysis – offer Russian voters a choice of foreign policies? Considering first the promarket parties, the Union of Right Forces (SPS) made no reference to foreign policy at all in its manifesto, apart from a commitment to the regularization of foreign debt. Creditors, the SPS suggested, should understand that debts would be repaid only when the economy had recovered and the sociopolitical situation had stabilized. In the first instance, the debts inherited from the Soviet period should be written down by 75 percent; and then debts that had been accumulated since that time should be restructured so that they became repayable over thirty years or more, with advantageous terms over the first five to seven years. Otherwise, it could be inferred, the SPS had no particular disagreement with the broadly pro-Western policies that were being followed by the Yeltsin administration (Pravyi 1999, 27–8). Yabloko, as befitted a party one of whose leading members was chair of the Duma foreign affairs committee, gave more detailed attention to Russia’s international environment. The country’s weak economy, they believed, meant that there must be modest but realizable aims in dealings with other countries. Russia should “strive for the formation of an organised and civilized multipolar world.” The party opposed the “claims of the USA and NATO to unipolarity and to a monopoly in the economic, military and political aspects of international affairs,” and placed more emphasis upon the role of the United Nations and its Security Council. Particular attention was to be given to the development of relations with the CIS countries, starting with a “firm union of Russia and Belarus” that was “open for the membership of all the states of the former USSR” (Bezopasnost’ 1999, 44–6). Russia, more generally, should integrate itself more closely into the world economy, but in a manner that respected the country’s national interests and avoided the “humiliating dependence on the International Monetary Fund (IMF)” that had developed in recent years (Yabloko 1999 [a]). The party leadership expressed muted criticism of the tactics being used in Chechnya, but did not campaign against the war. Unity, at the center of the political spectrum, acknowledged that it was “Putin’s party,” and its program was in effect a sympathetic recapitulation of current government policies. It was, for instance, in favor of the Chechen campaign – not a “war” so much as a legitimate offensive against “terrorist bases.” Elsewhere Russia had no constant friends or enemies, but only constant national interests. Russia, for instance, could be the partner of the United States in everything that concerned the strengthening of peace, and threats from other countries. But there had been too much copying of Western experience, forgetting that Russia’s Asian neighbors also had much to teach them (Voprosy 1999, 7–8). Unity was more generally in favor of the closest links with other nations, but only where this was in the interests of Russia and its citizens; and it was prepared to intervene in defense of the
60
Stephen White, Neil Munro, and Richard Rose
interests of Russians anywhere in the world, particularly if their rights were being violated (Tezisy 1999, 4–5). Although in opposition, Fatherland-All Russia (OVR) made the same commitment to “firmly defend the rights and interests of our citizens living beyond the borders of Russia,” including the right, if they wished, to return to their country of origin (Manifest 1999, 8). The OVR, however, offered a “new foreign policy for Russia,” one that reflected the views of the man who headed its federal list of candidates, former Foreign Minister Yevgenii Primakov. Russia, they pointed out, had been losing its international influence. Hopes of a “common European home” had proved illusory. Russia had been marginalized, the country’s legitimate interests ignored. It had no obvious enemies, but at the same time it had almost no allies. Russia, suggested OVR, should begin to act like a “great country” in ways that reflected its rich traditions, its status in the United Nations, and its formidable military strength. Its first priority should be relations with the former Soviet republics, and with Belarus, Ukraine, and Kazakhstan in particular. The countries of the EU, and of East-Central Europe, were Russia’s most important economic partners. At the same time there should be more attention to Russia’s Asian neighbors, including India and China, and to the Asia-Pacific region as a whole. But there should be no return to the “politics of confrontation” in relations with the United States and its NATO allies; the guiding principle, in all such matters, should be a “sensible national egoism and economic pragmatism” (Programma 1999, 112–20). The Communist Party, appealing for the nationalist vote, concentrated upon its economic proposals and upon social objectives such as employment, health, and education (Putem 1999). Its “fifteen steps to victory” concluded with two foreign policy objectives: the removal of all obstacles to the “unification of Russia, Belarus and Ukraine in a single union state,” and the revival of the country’s defensive capacity (CPRF 1999). The party program, adopted at its fourth congress in 1997 and still valid at the time of the election, spoke of the denunciation of the Belovezh agreement that had ended the USSR, and of the gradual, voluntary restoration of a single union state. Its foreign policy, the party promised, would be “independent,” and of a kind that would “strengthen the international authority of the Russian state” (Programma 1999, 23–4). Party leader Zyuganov was particularly scathing about the “monster called NATO” that had launched a Balkan war, and that was now seeking to extend its “fascism” even further. The task that stood before Russia’s patriotic forces, including the Communists, was to develop a “global anti-NATO alliance” in response to these developments, and which would be based in the first instance on Russia, Belarus, and Yugoslavia (Zyuganov 1999a, 26–30). The Liberal-Democratic Party of Russia (LDPR), strongly nationalist but often pragmatic supporters of government in the Duma, were led by the flamboyant Vladimir Zhirinovsky. In common with other parties,
Parties, Voters, and Foreign Policy
61
the LDPR opposed any form of expansion of NATO to the east or any weakening of its armed forces that would make Russia an “easy prey for other states.” And like the other parties, it also opposed any attempt by the United States to claim a dominant position in a world in which it was the only acknowledged superpower. The LDPR offered a distinctive interpretation of the place of Russia in the world between the “western Christian” civilization, and the “eastern Christian” civilization of which Russia was a part. The central development at the end of the 1990s was that the western world was seeking to extend its influence over Eastern Christianity, first by breaking up the USSR, and then by undermining the Russian Federation. Russia had been separated from its traditional allies toward the west and the south; and with the incorporation of Central and Eastern Europe into European structures, it found itself in a “geopolitical blockade.” The Liberal-Democrats called for the revival of a powerful and internationally respected Russian state within its “natural frontiers,” which included Belarus, Ukraine, Kazakhstan, and predominantly Russian parts of the Baltic republics. The party called for closer relations with states such as Iran, Libya, Cuba, Iraq, Vietnam, and India, allowing it to establish a counterbalance to Western hegemony. Nearer to Russia there should be a “buffer zone,” headed by governments that were friendly or at least neutral, and the “firmest measures of an economic and political character” should be taken to protect the legitimate interests of Russians in the other former Soviet republics. Further abroad, the Liberal-Democrats took an equally cautious view of relations with the United States, pointing out that this was the “main anti-Russian power” (Programma 1998, 25–33).
foreign policy and patterns of electoral support Across the parties, Yabloko and Union of Right Forces voters were the most “pro-Western” in their attitudes (Table 3.4). They were the best informed in matters of foreign affairs, and most likely to feel they would benefit from membership of the EU. Yabloko voters were, however, less likely than Right Forces voters to feel European and, alongside other categories of voters, less likely than Right Forces voters to oppose CIS integration. Relative to Right Forces voters and most other categories, Yabloko voters were relaxed about possible threats from the United States. Unity voters had few distinguishing attributes. Support for the Chechen campaign was high across all categories of voters, and lowest among nonvoters. The lack of distinguishing characteristics of Unity voters corresponded to the character of the party electorate, which drew evenly from all sections of society (see Rose, Munro, and White 2001). Unity’s opponents, FatherlandAll Russia voters, were somewhat more likely than average to feel European and somewhat less likely than average to perceive an Asian threat from China
63
62
table 3.4 Foreign Affairs and Reported Vote in the 1999 Duma Election
Yabloko
Unity
Eta
.14 .14 .13
.11 .09 .08 .08 .06 .06 .06 .06 .05 .05 .05
.04
.04 .03 .03
a
FatherlandAll Russia
Support government actions in Chechnya Has visited one or more Baltic countries Have knowledge of Europe: both Brussels and NATO Important to have good relations with the United States Russia would benefit from joining EU If Russians threatened in CIS: favors negotiation Feels European Attack on Russia likely in next five years Russia should increase its defense spending Consequences of CIS unification favorable Russia should help CIS states Perceive threat from Iraq Perceive threat from China Perceive threat from the United States If Russians threatened in CIS: favors economic pressure Perceive threat from non-Russian nationalities in Russia Perceive threat from Germany Favor Russian unilateral nuclear disarmament If Russians threatened in CIS: favors military action total n
Union of Right Forces
Against All
No Vote
Total
85
86h
74
70
75
78
70
62b
72
50h
25
15b
35
24
22
19
16
22
43h
29
20b
35
25
37
30
23
27
86
85
79
84
81
88h
82
72b
78
62h
52
56
59
41
62h
32b
42
50
97
95
91
93
94
99h
91
90b
94
44b 11b
56 19
54 16
65h 23
55 30h
58 20
50 25
46 26
53 25
57
61
52b
69h
68
59
61
54
60
45
52
44
50
63h
29b
44
44
50
22b
28
32
27
32
27
36h
34
33
14 22 32b
19 20 51
32h 31 46
13b 14b 50
26 22 61h
26 22 54
31 30h 52
25 25 42
25 24 49
71h
64
70
56b
66
69
66
59
64
37b
40
45
45
39
47
50h
44
44
10
12
20
5b
17
7
24h
16
15
14
16
34h
20
12
9b
13
20
17
13
19
27h
15
22
8b
15
21
20
68
287
71
134
278
98
57
767
A statistic of correlation, suitable for a nominal dependent variable. 180 respondents voted for minor parties and are included in the total column. Source: New Russia Barometer VIII, fieldwork January 13–29, 2000, n = 1,940.
b
Communist Party
% of voters (h = highest figure in row, b = lowest figure in row)
a
.19
LDPR
1940b
64
Stephen White, Neil Munro, and Richard Rose
or from Iraq. Unity and OVR voters were neither more nor less likely than the average Russian to perceive a threat from the United States. Communist voters were distinctive in two respects: they favored closer integration with the CIS and were most likely to perceive a U.S. threat to Russian security. However, their anxiety about a possible military attack on Russia in the following five years differed little from that of the ordinary Russian. Overall, there was the least variation across the entire sample in support for negotiation as a response to threats to fellow Russians in other CIS countries, and the most variation in the likelihood that respondents had visited a Baltic country, or that they favored closer integration among the CIS member countries. Because foreign policy is only one of the policy areas that are of interest to the electorate, we hypothesized that issues of this kind to voting would be modest – although it was likely to be much greater in issues, such as Chechnya and the position of Russians in the other post-Soviet republics, that crossed the boundary between domestic and foreign policy. Overall, it was already clear that voters would be most likely to line up according to their opinions – or lack of opinions – on basic questions of political economy (Rose, Munro, and White 2001). Voters who described their political outlook as “promarket,” for instance, would tend to support the SPS in relatively greater numbers, as would voters who believed Russia would benefit from joining the EU. As we were concerned with foreign policy, we took a domestic policy model as the starting point and then added foreign policy variables. The resulting “domestic + foreign policy” model showed how foreign policy issues fitted into the framework of domestic policy issues and the extent to which the constituency of voters lining up behind each party had distinctive views and characteristics. We present only the combined model in detail (see the last section of this chapter for descriptive statistics of the variables used). Discriminant function analysis has multiple advantages in identifying influences on voting when there is a multiplicity of candidates (Klecka 1980). Two points must be grasped in any analysis of this kind. First, voter choice is a nominal variable with multiple categories. Secondly, discriminant function analysis identifies combinations of attitudes, known as “functions” or “dimensions,” which push and pull voters in different directions. Each variable in the analysis loads on each function according to its relative importance. The variables that load highest on a function characterize it. Paralleling our findings using domestic policy variables only, the combined “domestic + foreign policy” model identified three functions that together correctly classified 54 percent of the choice of the electorate. The first dimension, explaining almost two-thirds of the variance, concerned whether voters were on balance for or against the transition to the market. The second dimension, explaining a seventh of the variance, divided voters who held no particular political outlook from those who held a political outlook with
Parties, Voters, and Foreign Policy
65
programmatic aims on the economy, whether pro- or anti-market. The third dimension, explaining a tenth of the variance, distinguished a constituency of cosmopolitan voters from a constituency of less well-traveled voters in the countryside. Discriminant function analysis produces no measure of statistical significance. As we are concerned here with characterizing foreign policy variables, however unimportant or important they are, we report the coefficients of all variables in descending order of their largest absolute correlation with each discriminant function (Table 3.5). The most important function, explaining 67 percent of the variance, differentiated between those who were in favor of the transition to a market economy and those who opposed it. The “pro- versus anti-market” function was dominated by a dichotomous variable indicating whether or not respondents identified with a “promarket” political outlook. The function also loaded on blaming the former communist regime for economic problems, favoring private over state ownership of enterprises, preferring an economy where goods are available at high prices to a shortage economy with low prices, disliking the former communist political system, and evaluating Yeltsin favorably. It was characteristic of younger people. In comparison to domestic policy variables, foreign policy influences on the first function were weak. The strongest foreign policy influence on the pro- versus anti-market function was a respondent’s attitude toward joining the EU; but the coefficient was barely one-fourth as strong as that for identifying with a promarket political outlook. Each party can be positioned along each dimension (or function) according to the characteristics of its vote. The numeric position of each party along each function lets us know how far its supporters vary from those of other parties in relation to a particular combination of issues. Finally, because the functions are independent of one another, each function can be given a unique interpretation. This means that each party can be positioned anywhere on each function, depending on the attitudes of its supporters. Sometimes, a party may be at an extreme, and sometimes it may be close to the center. Parties with well-defined positions on a particular dimension will appear at the extreme, while “fuzzy focus” parties will appear in the middle of each dimension. When group means of each party on each dimension are plotted, zero on the y-axis represents the position of the typical or average case in the analysis. In the case of the first dimension, it represents the whole-sample mean on a scale representing the strength of the pro- or anti-market syndrome (Figure 3.1). Overall, the reported vote of respondents was skewed toward the “promarket” end of the function. However, three-fifths of the votes went to Unity, Fatherland, Yabloko, and the Zhirinovsky Bloc – parties that were closer to each other than to either extreme. The Communists, with 30 percent of the reported vote for the six biggest parties, were the only party scoring well below zero on the function. Right Force voters, however, were forty-four
66
Stephen White, Neil Munro, and Richard Rose
table 3.5 Domestic and Foreign Influences on Choice of Party Functions of Party Choicea Pro- vs. Anti-Market
Ideological vs. “Fuzzy Focus”
Cosmopolitan vs. Inward-Focused
Parties at extreme of each function:
Right Forces/CPRF
Right Forces/Unity
Yabloko/Right Forces
Variance explained:
67%
13%
11%
68 38
37 −16
−33 16
37
06
12
34
−11
22
31 −31 −31
−13 22 14
−00 23 −24
18 −12
−03 02
17 02
07
−05
07
−15
52
−33
17 −09
−44 −23
−24 −16
−21
−22
−09
04
−19
11
−02
15
−05
01
14
−02
−01 02 21 −12
−07 06 39 03
03 43 39 −31
21 −11
04 03
26 −22
Promarket Blame former Communist regime for economic problems Prefer private to state enterprise ownership Against pre-perestroika political system Pro-Yeltsin Age Prefers lots of goods in shops to low prices Benefit if Russia joined EUb Increase Russian defense spending Good relations with the United States important Identifies with a political ideology Pro-Putin CIS Russians threatened: military action Consequences of CIS unification Pro-unilateral nuclear disarmament CIS Russians threatened: economic pressure CIS Russians threatened: negotiate Consider self European Has visited a Baltic country Town size Threat to security from the United States Education Threat to security from Germany
Parties, Voters, and Foreign Policy
67 Functions of Party Choicea
Parties at extreme of each function: Knowledge of Europe Support Russian actions in Chechnya Threat to security from China Attack on Russia likely in next five years Threat from non-Russians in Russia Russia should help CIS states
Pro- vs. Anti-Market
Ideological vs. “Fuzzy Focus”
Cosmopolitan vs. Inward-Focused
Right Forces/CPRF
Right Forces/Unity
Yabloko/Right Forces
11 09
12 −17
19 −18
00 −07
03 06
−17 −08
03
03
−06
−03
−03
−06
a
Variables ordered by absolute size of correlation within function. Boxed coefficients are those loading highest on one function at twenty-five or above. Two decimal places, decimal point omitted. b Variables concerned with foreign affairs are italicized. Source: New Russia Barometer VIII, fieldwork January 13–29, 2000, n = 1,940.
points further out from zero than the Communists. The size of the statistical gap between them was 288 points. The second most important function explained 13 percent of the variance. We called it the “ideological versus fuzzy focus function.” The term “fuzzy focus” describes “catchall” parties that have no strong ideologically determined positions and do not seek to represent any particular social group, but aim to attract votes from people with divergent views or to represent the “whole nation.” The function differentiated between those who accepted a political outlook, including both Communists and promarket identifiers, and those who classified their political outlook as “none of the above.” Those who had a favorable opinion of Putin’s performance gravitated toward the “fuzzy focus” end of the function. Being in favor of military intervention if Russians were threatened in the CIS was also associated with the “fuzzy focus” end of the function. Perceiving favorable consequences of CIS unification was the second strongest foreign policy influence, and was again associated with the “fuzzy focus” end of the function. Both of these foreign policy variables exerted only a weak influence, their coefficients being less than half as strong as the coefficient for choosing “no ideology.” The reported vote is fairly evenly distributed along the second function: the Zhirinovsky Bloc and Unity – parties scoring below zero – took 39 percent of the reported vote for the six biggest parties. The three parties scoring above zero took 47 percent. Although closer to the “ideologically
Stephen White, Neil Munro, and Richard Rose
68 200
Pro- vs Anti-Market Ideological vs "Fuzzy Focus" 166
Right Forces
64 Yabloko 45 Unity
0
Cosmopolitan vs Inward-Focused
76 Right Forces 77 Yabloko 64 Fatherland 39 Yabloko 24 Communist
02 Zhirinovsky 01 Fatherland Fatherland -15 -16 -20 -33 Zhirinovsky -38 -45 Unity
Communist Unity Zhirinovsky Right Forces
-122 Communist
-200 figure 3.1 Placing Party Choice on Dimensions of Support. Source: New Russia Barometer VIII, conducted by the Russian Public Opinion Research Centre (VTsIOM), fieldwork January 19–29, 2000, n = 1,940; for fuller details see “A Note on Data.”
focused” end of the function than three other parties, Communist voters were closer to the typical respondent than to the extreme. The Union of Right Forces stood out as the party whose supporters were least likely to be ideologically fuzzy: its supporters were further from zero than supporters of Unity. The size of the statistical gap between the extremes was 121 points.
Parties, Voters, and Foreign Policy
69
The third function explains 11 percent of the variance. We call it the cosmopolitan versus inward-focused function. It loaded strongly on two variables: whether the respondent had traveled to one or more of the Baltic States, and town size. It also loaded on education and on the perceived threat from the United States. Support for the Russian campaign in Chechnya had its strongest coefficient on the function, but the coefficient was low, suggesting a weak effect. The reason the Chechen campaign had only a limited effect on the distribution of the vote was not that it was thought to be unimportant, but because there was little disagreement on its importance across those who had voted for the major parties. Seventy-nine percent of the vote for the six biggest parties went to the Communists, Unity, Right Forces, and the Zhirinovsky Bloc – parties scoring below zero on the cosmopolitan versus inward-focused function. Yabloko supporters were more than three times further above zero than Right Forces supporters were below it, and Fatherland-All Russia supporters were only a little less cosmopolitan than those of Yabloko. The size of the statistical gap between the extremes was 115 points. It is worth noting that there was a certain amount of cross-loading between functions. Identifying with a promarket outlook was of primary importance on the first function, but also pushed voters on the second function away from the “fuzzy focus” toward the “ideological” end, making them more likely to vote for Right Forces than for Unity. Living in an urban area cross-loaded on the second and third functions. On the second function, it pushed voters away from Unity in the direction of Right Forces, and on the third function away from Right Forces in the direction of Yabloko. Identifying with a political ideology also cross-loaded on the second and third functions. Foreign policy issues, overall, made only a minor contribution to the distribution of voter preferences in the 1999 Duma election. As we have pointed out, this was not necessarily because foreign policy issues were unimportant to voters, but because those issues that mattered most – including Chechnya and the fate of Russians in the CIS – were generally the least divisive. Otherwise, voters lined up according to their views on the political and economic transformation to a market economy. A second dimension – the difference between the ideologically committed and the voters for “fuzzy focus” parties – allows only a slight influence from views and experiences on international issues. The third function, accounting for the least variance, does however reflect the influence of at least one foreign policy issue, the perceived threat from the United States.
conclusions and implications The findings we have presented offer, in the first place, an indication of the foreign policy preferences of ordinary Russians as they become accustomed to a new world of “friends” and “foes” in a very different international
70
Stephen White, Neil Munro, and Richard Rose
environment. Their “friends” are the other Slavic states, Belarus and Ukraine – their neighbors, the states with which they formerly shared a union state and Orthodox Christianity as a state religion. Equally, these are the states in which they were most likely to have family members, and to have visited. The collapse of the USSR is widely seen as a source of economic difficulties, and accordingly a closer association with at least the more developed of its former members would appear to offer some hope of an improvement. A decade after the end of the Cold War, the biggest threat is still the United States, followed by non-Russians within Russia, in particular Chechens (the war was still very popular at the time of our survey). The parties that contested the Duma election had a variety of approaches to foreign relations. Unity, a party that had been formed so quickly that it had not yet held its founding congress at the time of the elections in December 1999, attracted supporters who favored the conduct of the Chechen “antiterrorist operation” – but the war was popular across the electorate as a whole. Communist supporters were the most likely to see the United States as the country’s main security threat, and they were the most firmly committed to the closer integration of CIS member states. Supporters of the Union of Right Forces and Yabloko, on the other hand, were the best informed about foreign affairs in general, and the most pro-Western. But overall, there were high levels of “don’t knows,” and a large majority lacked even the most basic information about international organizations such as the EU. Our discriminant function analysis suggested that the most divisive foreign policy issues, across the electorate as a whole, were CIS integration and membership of the EU. Under present circumstances EU membership is scarcely conceivable. Russia, for a start, is not a European state in the terms of the Treaty of Rome. But a closer association among the member states of the CIS, including a form of union between Russia and Belarus, is as much an objective of the Putin administration as of its predecessor. If further moves in this direction are taken for economic or geopolitical reasons, they may sharpen differences between the supporters of broadly “Western” and of “Soviet” or “Russian nationalist” foreign policy orientations. By contrast, there is relatively little dispute about the need for higher levels of defense spending alongside other priorities, and for a better relationship with the United States. Foreign policy issues are in turn closely related to Russia’s economic future, given the extent to which existing policies involve the export of energy for the hard currency on which the Russian state continues to depend. In addition, given its geopolitical location, Russia is bound to play an important role in any coordinated action against international terrorism. Western nations, for their part, have welcomed the closer relationship that has developed since September 11 but have continued to press for a closer adherence to the rule of law and market economics. Questions about the relationship between Russia and the West are themselves part of a much older Russian
Parties, Voters, and Foreign Policy
71
dialogue about the nature of their own society. Should it strive to remain different from the West, as Slavophiles have argued for many centuries, or should Russia adopt the strategy of Peter the Great and open its doors to Europe and the world beyond? On the evidence of the elections of 1999– 2000, Russians and their politicians would do their best to embrace both at the same time.
a note on data Our survey, the eighth in the series of New Russia Barometers, was conducted by the Russian Public Opinion Research Centre (VTsIOM) between January 19 and 29, 2000. The sample size was 1,940 and was representative of the adult (18+) resident population of the Russian Federation (Table 3.6). For full details of questions and answers, and a report on the sample, see Rose (2000). Further information on the New Russia Barometer, conducted by the Centre for the Study of Public Policy at Strathclyde University since 1992, is available at www.cspp.strath.ac.uk.
table 3.6 Descriptive Statistics of Independent Variables Used Mean
Standard Deviation
0 no 1 yes 1 blame not at all 4 blame very much
0.18 2.51
0.38 0.93
1 state++
4 private++
1.98
0.96
−100 worst 1 worst 18 1 low prices++
100 best 10 best 89 4 lots of goods++
36 2.99 44 2.56
52 2.03 17 0.98
1 much worse off 5 much better off
3.49
1.13
1 much worse off 5 much better off 1 decrease++ 5 increase++
3.48 3.68
0.92 1.13
1 not important
4 very important
3.31
0.72
0 no
1 yes
0.35
0.48
1 worst 1 oppose++
10 best 4 favor++
5.78 1.77
2.46 0.88
Minimum Promarket Blame former Communist regime for economic problems Prefer private to state enterprise ownership Pre-perestroika political system Pro-Yeltsin Age Prefers lots of goods in shops to low prices Consequences of CIS unification Benefit if Russia joined EU Increase Russian defense spending Good relations with the United States important Identifies with no political outlook Pro-Putin Favor Russian unilateral nuclear disarmament
Maximum
(continued)
72
Stephen White, Neil Munro, and Richard Rose
table 3.6 (continued)
If CIS Russians threatened: support military intervention support economic pressure support negotiation Consider self European Town size Has visited one or more Baltic countries Threat to security of Russia from the United States Threat to security of Russia from Germany Education Support Russian actions in Chechnya Knowledge of Europea Threat to security of Russia from China Attack on Russia likely in next five years Russia should help CIS states Threat from non-Russians in Russia a
Minimum
Maximum
Mean
Standard Deviation
1 oppose++
4 support++
1.85
0.90
1 oppose++ 1 oppose++ 1 not at all 0 village 0 no
4 support++ 4 support++ 4 very much 10 > 1 million 1 yes
2.83 3.59 2.51 4.77 0.22
0.91 0.65 0.99 3.79 0.42
1 no threat
4 big threat
2.39
1.06
1 no threat
4 big threat
1.53
0.78
1 elementary 1 oppose++
10 postgraduate 4 support++
5.06 2.94
2.53 0.92
0 none 1 no threat
2 knows both 4 big threat
0.92 1.76
0.78 0.94
1 not at all likely
4 very likely
2.11
0.72
1 should not help 1 no threat
4 more than now 4 big threat
2.08 2.26
0.89 1.07
A three-point scale derived by summing correct answers to two questions on what NATO is and where are the headquarters of the EU.
4 The Foreign-Policy Implications of the Elections Andrei Melville
The options available for a particular political regime in the sphere of international relations result to a large extent from the interplay between its available resources or capabilities and the intentions and choices of the major political actors within that regime. Particular choices made by key political actors are especially important during periods of regime change and transition. This chapter shares an actor-oriented approach to the issue of democratic transitions and reviews the foreign-policy aspects of the Russian transition; the dramatically decreased foreign policy resources available to the regime of Yeltsin’s successor, Vladimir Putin; and the evolving foreignpolicy orientations of the major groups within the new Russian political class and Russian public. The 1999 and 2000 elections mattered for Russian foreign policy because the change from the Yeltsin administration to the Putin administration provided an opportunity for political elites to revise their perceptions of the outside world, to rethink Russia’s place in it, and to adapt Russia’s foreign policy to the new realities. The elections also mattered because they put the formidable resources of the Russian presidency into the hands of Putin. Understanding his foreign-policy views is critical, not only for insight into Russian foreign behavior but to shed light on an important facet of regime change and democratic transition. Despite Putin’s programmatic support for the idea of democracy and his pro-Western slogans, substantial uncertainty about his foreign-policy course prevailed in 2000 and 2001. It remains to be seen whether Putin’s new (“post–September 11”) foreign policy, with its focus on integration with the West, will prove to be a new foreign-policy strategy that definitely breaks with the previous uncertainty.
choice and uncertainty in protracted transitions Two and a half decades’ experience with “third-wave democratizations” provides rich data in support of a previously only theoretical argument found 73
74
Andrei Melville
in the literature on democratic transition and consolidation – the argument against the “teleological temptation” (Diamond 2000) (i.e., an intellectual inclination to view these transitions as necessarily moving toward some natural democratic end state). Some authors even started to argue against describing self-proclaimed “new democracies” as being in transition as that implies that one knows a society’s starting point, where it is today, and where it is heading. Instead they suggested that the term “transformation” is more appropriate for describing divergent and often distorted trajectories of political developments in those countries (Rose, Mishler, and Haerpfer 1998). This argument has a generic character and may apply to postauthoritarian cases as well as to post-Communist (and post-Soviet) ones. Indeed, transitions of the third wave have demonstrated that the formal inauguration of democracy (i.e., of formal democratic institutions and procedures of “electoral democracy” – though not “liberal democracy”) (Diamond 1996; Diamond, Plattner, Chu, and Tien 1997; Zakaria 1997) in no way guarantees democratization. This empirical observation was widely conceptualized in an analytical distinction between two major phases of the process of democratization: (1) transition per se, that is, formal inauguration of democratic procedures and institutions, and (2) its consolidation (Mainwaring, O’Donnell, and Valenzuela 1992; Gunther, Diamandouros, and Puhle 1995; Linz and Stepan 1996). Consolidation of democracy (or democratic consolidation) is usually referred to as ranging from the “minimalist,” or procedural level, in which all formal democratic structures and institutions are in place (Di Palma 1900) to the level of comprehensive consolidation (Merkel 1998). Democratic consolidation depends on a variety of endogenous factors (the nature of the preauthoritarian democratic experience, if any; the character and type of the rejected nondemocratic regime; the conditions of the authoritarian collapse; and the strategies chosen by leading political actors) and exogenous factors (the external environment; the extent of the country’s inclusion in major international structures; and the international support systems available to it). An important addition to this general argument about democratic consolidation was made by those authors who had pointed out some other structural prerequisites of comprehensive democratic consolidation, including the role of civic culture and civil society (Dahl 1997; Gill 2000; Inglehart 2000) and of socioeconomic development, eventually leading to society’s affluence, economic growth with moderate inflation, and declining inequality (Przeworski, Alvarez, Cheibub, and Limongi 1997). Often these dimensions (or levels) tend to be understood as particular temporal sequences or stages in the process leading to the final democratic consolidation (Merkel 1998). And equally often this understanding unwittingly gets us back to the above-mentioned “teleological temptation” – this time in a form of a perception that the transition is continuous (i.e., as if it progresses steadily from the stage of nonconsolidated democracy to different
Foreign Policy Implications of the Elections
75
and interdependent stages of democratic consolidation). While there is merit in the developmental view of democracy (Sklar 1996; Diamond 2000) as emerging in fragments or parts, it may not be feasible everywhere at every time. “Windows of opportunity” for democratic development may sometimes open and then close. In fact, this not only reinforces the argument about the threat of “slow deaths” on nonconsolidated democracies as the period of transition extends beyond certain limits (Schledler 1998) but also raises the question of whether seemingly democratic formal procedures may in fact perform nondemocratic functions. In contrast to the analysis of preconditions and stages of democratic consolidation, in the existing literature there is much less conceptual emphasis (and available empirical data) on patterns and consequences of the nonconsolidation of democracy. While at the end of the century some experts voiced an enormous enthusiasm about the prospects of global democratization (Karatnycky 2000), others were much more cautious. And what is more, probably the major accomplishment of the third wave – the spread of “electoral democracies” (i.e., formal democratic procedures and institutions like elections in the first place) – may need a closer and deeper inspection. Democracy, in principle, is not about elections – at least not about just any old elections (Diamond 2000). As it turns out, not only can undemocratic games can be played “inside” the formally democratic institutions (O’Donnell 1997), but such institutions may be used as a smoke screen or a fac¸ ade behind which a caesaristic, plebiscitarian executive regime or some other nondemocratic division of power may emerge and thrive. In other words, formal electoral procedures may in certain cases represent not a crucial feature of “electoral democracy” understood as an intermediate stage in transition toward democratic consolidation, but a quite different phenomenon: the transition from one type of nondemocracy to another. Needless to say, this may (and sometimes does) result in a perfectly consolidated “new” autocracy. This is especially relevant to the issue of postCommunist and particularly post-Soviet transitions and transformations, which demonstrate not only a rich variety of starting points and initial conditions but also a remarkable diversity of preliminary stages and end results. Protracted transitions (such as the Russian one), often characterized by political stalemate and uncertainty, reinforce the importance of strategic choices and decisions. According to the mainstream theories of democratic transition and consolidation, political stalemate and uncertainty after the collapse of the authoritarian regime usually lead to the institutionalization of the democratic choice. However, this may not always be true. McFaul had already argued that in the Russian transition the stalemate, uncertainty, and the absence of strategic choice actually played a negative role (McFaul 2001). Stumbling in the stalemate and uncertainty may indeed eventually lead to the conservation of an elite balance of power with the imitation of democratic procedures (Solnik 1999) and result in a “derailed” transition
76
Andrei Melville
(Rupnik 2001). Unless choices are made and decisions implemented, options for successful transitions may not be available forever – windows of opportunities may close, and rather unexpectedly. These general observations lead us to some more concrete questions: What is the trajectory of Russia’s current transition/transformation, including its relationship with the outside world? What domestic economic and political factors influence Russia’s foreign policy? What are the prospects for Putin’s “post–September 11” foreign policy?
russia’s foreign-policy capabilities During the 1990s, the resources and capabilities available to Russian foreignpolicy makers diminished dramatically. To a large extent this was due to the circumstances in which the Soviet Union dissolved. However, the economic, political, and social policies of the Yeltsin regime contributed to the decline. Despite Moscow’s official rhetoric (at least prior to Putin), the resources and capabilities available in Russia today do not qualify it for superpower or even great-power status. During the Gorbachev and Yeltsin eras, Russia lost a large part of its Cold War–based geostrategic, economic, military, political, and ideological potential. Russia’s reduced geostrategic position in the world is in part reflected in the loss of territory associated with the breakup of the USSR. Russia not only became smaller, it also found itself in a new geopolitical position. Having lost its claim to be the traditional “Eurasian” country that provided a “balance” (or a “bridge”) between the West and the East, Russia is now, in a way, a “Finnish-Chinese” country, pushed into the North-Eastern part of the continent. Moreover, in the era of globalization, Russia’s territory and population per se no longer determine its role in the world as the great power – as was the case in the past. Beyond this, Russian capabilities have declined in several key ways. Russia’s state power has eroded. One of the manifestations of this erosion is the increased international activities of the regions, primarily in the economic sphere. This may have unprecedentedly diminished essential institutional resources of Russian foreign policy. One of the main foci of President Putin’s reforms is his effort to secure control over regional centers of political and economic power (Bunin, Makarenko, and Roslavlev 2000). The state of the economy is another painful wound for “postreform” Russia.1 This has particularly significant foreign-policy implications as economic strength 1
Russian gross domestic product in 1999, when Vladimir Putin assumed power, was only one-half that of the Soviet Union in 1990, a year when the economic crisis was already deep. Russia’s gross national product was only 10 percent of that of the United States and 20 percent of China’s (Putin 1999). The modest economic growth registered in Russia in 2000–1 is largely a product of high oil prices in the world markets. In this context one may be tempted to draw analogies with the early Brezhnev era with its high oil prices and lost opportunities for domestic modernization during the 1970s.
Foreign Policy Implications of the Elections
77
is a crucial element of status and position in the world of international affairs today. Most dangerous is the long-term structural trend in the Russian economy toward reliance on the extraction of natural resources, the sale of raw materials, and the production of the energy sectors (Graham 1999; Rogov 2000 [b]). The extent of Russia’s economic dependence on the West (primarily in terms of loans, technology, and investments) together with the domestic and international consequences of that dependency have yet to be fully appreciated by Russia’s new political elite. In this context, managing its foreign debt has become one of the major national goals of the Russian Federation. Russia is simultaneously facing dangerous demographic trends: declining life expectancy, a dramatic deterioration of the public health, and problems associated with the state of social anomie. Even more so, as a consequence of the country’s economic, social, and political problems, the Russian military has lost its former prestige and capacity to project power.2 The nuclear arsenal remains the only legacy of Russia’s previous superpower status. According to some experts’ estimates, however, this arsenal may pose a danger of self-destruction within ten to fifteen years largely because funds for adequate maintenance are not available. Last but not least, it is important to face Russia’s loss of a national identity and a sense of national purpose without which its foreign policy will be missing one of its major resources. In addition, Russia’s foreign-policy establishment is not in good shape. It bears the wounds of institutional disintegration in the final years of the Soviet Union and of competition among different interest groups during the Yeltsin era: the Ministry of Foreign Affairs, the Ministry of Defense, the Presidential Administration, the Security Council, the security forces, and others. The so-called “family” (the inner circle of Boris Yeltsin’s confidants, including his daughter Tatiana Diachenko) that was so important for Yeltsin even seemed to influence at least some of Putin’s decisions during his first months in power. Key foreign-policy documents produced by rival organizations sometimes differ dramatically not only in tone but in general worldview, the definition of external threats, and preferred strategies: the new National Security Doctrine, produced by the Security Council; the draft of the new Military Doctrine, produced by the Ministry of Defense; and the new Conception of Russian Foreign Policy, produced by the Ministry of Foreign Affairs. For a variety of reasons, therefore, the Russian state has decayed, and this has weakened its foreign-policy options. However, despite this obvious continuity of problems and challenges, the general political landscape of Russia changed significantly after the 1999 parliamentary and the 2000 2
Putin’s assertive policy in Chechnya was widely considered to be the last hope for the Russian military to save what remains of its former prestige. Probably the major part of Putin’s ` “semicharisma” has to do with his decisive actions vis-a-vis the Chechen separatists. However, the stalemate in Chechnya put this hope in question.
78
Andrei Melville
presidential elections, creating a more favorable domestic context for Putin’s foreign policy.
domestic political context for russian foreign policy In politics, legacies of the past often shape the problems of today. One of the dramatic shortcomings of the Yeltsin era was the first Russian President’s inability (or rather his unwillingness) to build and fully develop democratic political institutions as well as social institutions supporting the market economy. Instead, he relied on a personalistic style in politics (Breslauer 1998). The state, which partly disintegrated and partly was destroyed during the transition period, was not restored. Yeltsin did not create fully democratic political institutions (such as fair elections, an independent judicial system, and a developed party system) that could support the economic reforms, nor did he create the institutions that could support a market economy and a social security system. Old Soviet elites successfully reproduced themselves in the new Russian ruling class. Rechtsstaat, the rule of law as the only antidote against powerful and corrupted private interests and clientelistic politics, was never established. Instead, a system of oligarchic capitalism emerged, with a noninstitutionalized center of influence (the above-mentioned “family” of relatives and close advisers who surrounded President Yeltsin) playing a dominant role. By the end of the 1990s, dramatically shrinking domestic and international resources had rendered Yeltsin’s regime quite dysfunctional. The upper echelons of the new Russian ruling class (the oligarchs and the “family” in the first place) were troubled by continued political polarization, growing political distrust, calls for redistribution of property and power, and clear signs of the growing success of the political movement Fatherland-All Russia (OVR). Their major goal became to preserve privatized state property and provide guarantees for their personal economic and political security while putting forward somebody with a fresh political image who would be able to perform the requested function. As is well known, Putin issued a special decree following Yeltsin’s resignation that guaranteed full securities to Yeltsin and to his closest entourage (some even speculated about a special secret memorandum of understanding between Putin and the “family”). The resignation of Boris Yeltsin did not fit the spirit (although perhaps the form) of democratic procedures. He left his post not because of illness or inability to perform his duties (though this was the official rhetoric). The motivation was to guarantee a semidynastic transfer of power to the chosen one in the interests of the “family.” Putin’s emergence from nowhere was the product of efforts by the same “old team.” But that team could not be sure whether – as in the famous fairy tale – the artificially created “Golem of Clay” would eventually become independent from his creators. The crucial question was if, and to what extent, Putin would be able to distance himself
Foreign Policy Implications of the Elections
79
from the “family” and other oligarchs. The dilemma he faced was that, on the one hand, he had to play according to the rules of his “demiurges,” while, on the other, he needed to play his own game in order to win popular support and create his own political machine. There were Russian commentators who had even offered warnings of the end of public politics and of a new Pinochet coming to power (Petrov 2000). However, such judgments were not well grounded, even if some of Putin’s actions raised doubts about his understanding of democratic governance: his attitude toward freedom of the press (e.g., consider the closing of the oppositional NTV and TV-6 channels) and the style and at least parts of the contents of his administrative reforms that some analysts considered to be not fully constitutional (Ivanchenko, Ryzhkov, and Salmin 2000). While Putin was not able or willing to break with the Russian tradition of personification and concentration of power at the expense of independent loci of power, it is important to mention that only a few indications of a powerful thrust toward a new autocracy are apparent at either the elite or popular levels. Practical governance and executive manipulation seem to be the key words in the Russian ruling class’s vocabulary. The balance of forces in the new State Duma seem to confirm this general political direction. The major election result that is relevant for the foreign policy of Putin’s regime is probably the following: the new Russian President and the executive branch in general face an unprecedented (compared to the period of Yeltsin’s reign) domestic political context. The State Duma elected in December 1999 is no longer a bastion of political opposition to presidential foreign policy (including Putin’s change of foreign-policy direction after September 11). After the strong showing in the Duma elections of the propresidential Unity, Putin clearly has more resources of control over this third State Duma than Yeltsin did over the two previous ones. The Duma’s Committee on International Relations, for example, agrees with the Ministry of Foreign Affairs much more often than in the past. As Committee Chair Dmitry Rogozin (2000) stated in one of his first interviews in this capacity, “Foreign policy should not be the object of inter-party bargaining. There should exist an absolute, all-national consensus on this topic. The Parliament should support the actions of the Ministry of Foreign Affairs and, naturally, of the foreign-policy line of the Russian President.” As a result, Putin has the domestic political resources to pursue a rather broad spectrum of options. Putin, with the help of Unity (which at the end of 2001 had merged into a “super-party” together with the OVR), can expect almost complete loyalty from the oligarchs. However, the creation of a team of “yes men” does not seem to be Putin’s ultimate goal. Signs of his dissociation from this group emerged early in the months following the presidential election. “Not a single clan, not a single oligarch should be allowed to be close to regional or federal power,” stated Putin in his Open Letter to Russian Voters (Putin 2000).
80
Andrei Melville
In his public-relations events, Putin has tried to stress his distance from the oligarchs (not to mention the “family”) which actually had paved his road to power. However his “antioligarchic” policy seems to be, indeed, pretty selective – the Kremlin fights Russian oligarchs not per se, but rather according to some special list that leaves some old-timers of the “family” almost immune. This double standard leads some commentators to speculate about the emergence of a quasi–“oligarchic monopoly” enjoying special privileges within the Kremlin. More likely, though, it is just more evidence that Putin, while prepared for compromise, does not tolerate independent centers of political influence and ambitions. Thus, Putin faces a variety of formidable domestic and international challenges. Is he prepared to break with the uncertainty of the previous stages of the Russian transition? Is he ready to make strategic choices and implement them in concrete decisions? Does he really plan to undertake radical change and reform in his domestic and foreign policy? If so, he must implement these reforms against the will and the interests of the powerful groups within the new ruling class that paved his road to power. Which social and political groups will provide Putin with the necessary support?
public opinion Popular support is usually one of the important resources of foreign policy (Lippmann 1952; Rosenau 1961). The most important dimension of public political orientations in Russia today is the longing for order. Public opinion polls (VTsIOM 2000 [a]) find that 72 percent agree with the statement that “order is more important even if one needs to limit democratic principles and individual rights,” while a mere 13 percent believe that “democracy is preferable even if it paves the road to destructive and criminal elements.” This does not mean, however, that the public wants a dictator to emerge. There is significant uncertainty in the public mood about the interplay between “democracy” and “order” (Table 4.1). The phenomenon of Vladimir Putin should be understood within the context of dominant public expectations. On the eve of the 1999 parliamentary elections, there were clear indications of optimism about the possibility for future changes for the better. Fifty-five percent of respondents to one poll hoped that in the year 2000 they would start to live better (“definitely better” – 6 percent; “hopefully will be better” – 49 percent). Twenty-seven percent (versus 16 percent in 1994) said they experience hope (VTsIOM 1999). It is also critical to acknowledge that these expectations are directly linked to the emergence of a strong leader of the country – 62 percent believe this (ROMIR 1999). The ideal personal qualities of such a strong leader would be that he take care of the interests of the people and be decisive and ready to take responsibilities. His professional competence is of secondary importance (ROMIR 2000 [a]). What more specifically does the public most
Foreign Policy Implications of the Elections
81
table 4.1 Perception of the Concept of “Order” among the Russia Populace in 1994 and 2000 What, in your opinion, is “order”?
1994
2000
Political and Economic Stability in the Country Strict Rule of Law Limitations on Democratic Rights and Freedoms
37%
45%
31% 1%
35% 3%
What, in your opinion, is “democracy”?
2000
Strict Rule of Law Order and Stability Shallow Talking
29% 28% 10%
Source: All-Russian Center for the Study of Public Opinion (VTsIOM 2000 [a]).
table 4.2 Public Approval of Leadership Performance 2000
2001
May June Aug. Nov. Dec.
Jan. March May Aug. Sept. Oct.
Putin 61% 73% 60% 70% 68% 76% 75% 71% 74% 73% 75% Kasyanov 45% 49% 42% 47% 43% 47% 49% 45% 46% 50% 48% Government 34% 38% 38% 38% 38% 43% 40% 38% 39% 40% 42% Source: All Russian Center for Public Opinion (VTsIOM 2001 [b]).
expect from the new president? The public expects a new president to return Russia the status of a great, respected power (VTsIOM 2000 [b]). The new president was clearly expected to be the healer of the wounds of national humiliation of the past decade. The revival of the army seems to be one of the required recipes: the Russian military (especially after Chechnya) is now seen by 60 percent of the public as capable of defending the Motherland, compared to 40 percent in 1998 (VTsIOM 2000 [c]). The miracle of Putin’s popularity is much more than a public-relations event – it is as if he serves as some sort of a national psychotherapist. In other words, in the public mind Putin has become a “teflon” figure, accumulating the most general hopes for the future and not subject to proof by concrete performance. This can be illustrated by his approval ratings in contrast to those of Prime Minister Kasyanov and government in general (Table 4.2). Beyond the very high general approval ratings shown in this table, the same poll shows that 59 percent of the respondents rate Putin’s performance as president as being adequate to what they had expected before his election, while 49 percent of the respondents believe that the current signs of some improvement of the general situation in Russia results directly from Putin’s
82
Andrei Melville
table 4.3 Public Evaluations of Russian Policy by Sphere Percent Pleased with the Situation in the Following Spheres International Relations Countries of the “Far Abroad” Countries of the “Near Abroad” Russian Politics Russian Economy
2000 3% 2% 1% 4% 2%
Percent Displeased with the Situation in the Following Spheres
2000
International Relations Countries of the “Far Abroad” Countries of the “Near Abroad” Russian Politics Russian Economy War in Chechnya
7% 4% 7% 21% 43% 51%
Source: All Russian Center for Public Opinion (VTsIOM 2000 [d]).
performance. However, the concrete accomplishments of Putin’s team are graded much lower (social policy – 22 percent approve; foreign policy – 21 percent; economy – 19 percent; defense – 15 percent; state management – 10 percent; education and science – 9 percent; and industry – 7 percent). It seems that at the core of public perceptions of the new Russian messiah is a peculiar duality: People fail to grasp the gist of Putin’s views, values, and positions. Instead, they try to implant in his image their own preferences. Public opinion polls prove that the majority of Russians lack any clear political or ideological image of their president – for liberals he is a market reformist, for nationalists he is next to a patriotic stalwart, and so on. The desire for a miracle, associated with Putin in the public mind, becomes even more impressive when viewed within the context of public assessments of that with which Russians are pleased or annoyed (Table 4.3). Almost no one is happy with Russian policy in any sphere. The low percent of respondents reporting either pleasure or displeasure with foreign-policy spheres in Table 4.3 does not indicate Russian indifference to the outside world. As the responses in Table 4.4 make clear, the Russian public considers the main source of threats to Russia to be internal. Nonetheless, foreign economic threats are also of concern. A different poll found that the proportion of people who believe that Russia faces an external military threat increased from 33 percent in 1998 to 47 percent in 2000 (VTsIOM 2000 [c]). Russians are well aware of the weakening of Russia’s position in the world – 60 percent of respondents worried about this situation and are looking for a change (ROMIR 2000 [b]). This change is largely perceived as a desire for a restoration of order (which, as noted previously, is not
Foreign Policy Implications of the Elections
83
table 4.4 Public Perceptions of the Main Sources of Threats to Russia in 1996 and 2000 What, in your opinion, presents a major threat to Russia? Increased Unemployment Growth of Prices Economic Collapse, Russia’s Default Growth of Crime Plunder of National Resources by Foreign States and Companies Increased Dependency from the West, Russia’s Transformation into a Third-Grade Country
1996
2000
30% 30% 30% 26% 16%
35% 33% 32% 29% 28%
11%
19%
Source: All Russian Center for Public Opinion (VTsIOM 2000).
table 4.5 Russians’ Attitudes Toward NATO, 1996–2000 Which of the following items, in your opinion, best reflects Russian interests? Russia Should Become a Member of NATO Russia Should Establish Cooperation with NATO Russia Should Form a Defense Alliance Against NATO Russia Should Not Be Member Of Any Alliance
1996
1997
1999
2000
10%
10%
10%
9%
22%
23%
23%
28%
16%
13%
19%
17%
23%
24%
25%
28%
Source: All Russian Center for Public Opinion (VTsIOM 2000 [c]).
the antithesis of democracy). Fortunately, Russians’ longing for great-power status and their perception of foreign military threat do not directly translate into public xenophobia, anti-Westernism, or a desire for a Communist restoration. Quite the opposite: 68 percent of the public in 2000 agreed that Russia should develop mutually beneficial ties with the West, up from 61 percent in 1999 (VTsIOM 2000 [a]). Even on the very controversial issue of NATO, Russian public sentiments are clearly split (Table 4.5), with over a third preferring cooperation with or even membership in NATO. At the same time, there are worrisome indications of a new syndrome of “wounded exceptionalism” with isolationist overtones. Thus, in March 2000, 66 percent of survey respondents agreed with the statement that “Russia has always been perceived as an adversary by other countries and even today nobody wishes us well.” The same poll recorded the following distribution of public preferences concerning the preferred “historical path
84
Andrei Melville
of Russia”: adhere to the contemporary world path of Western civilization, 15 percent; return to the path of the Soviet Union, 18 percent; and follow a special Russian destiny, 60 percent (cf. Gudkov and Dubin 2001). In sum, the Russian mass public sends mixed messages on foreign-policy issues to their designated ruler/messiah. To what extent Putin will feel constrained by public opinion is, of course, a different question. Russia’s Foreign-Policy Aims Elite understandings and aims shape the formulation and execution of any country’s foreign policy. At the moment, foreign-policy attitudes among Russia’s elite are far from consensual. Actually, the lines of cleavage differ somewhat depending on which aspect of the foreign-policy agenda is being debated: understanding the megatrends of world politics, Russia’s role in the world, its goals and priorities, perceptions of enemies and allies, and concrete foreign-policy interests. Nonetheless, the foreign-policy orientations of Russia’s political elites can be grouped into five competing foreign-policy approaches: neo-isolationism, restorationism, Westernism, multipolarism, and pragmatic selectivism. While some isolationists are aggressive as well (e.g., Zhirinovsky’s Liberal-Democratic Party of Russia (LDPR), others are politically moderate intellectuals. Restorationists, such as Zyuganov’s Communist Party (CPRF), seek to return Russia to the leading role played by the Soviet Union. Pro-Western orientations are associated most clearly at present with the Union of Right Forces (SPS) party. Multipolarism acknowledges that Russia cannot be the other part of a bipolar world order but denies U.S. primacy in all spheres and believes that Russia can be an important power in some spheres. It became Russia’s official foreign-policy doctrine under Foreign Minister Primakov and remained so (with some variations) under his successor, Igor Ivanov. Until recently, it represented a majority view of Russia’s mainstream politicians and experts and remains influential still. It was found in the foreign-policy platforms of the liberal Yabloko party and of the centrist Fatherland-All Russia (OVR) party (before its absorption by Unity at the end of 2001). Interestingly, Unity, which emerged from nowhere with incredible success using the organizational and financial resources of an official “party of power,” expressed essentially no particular foreign-policy views except for a clearly ritualistic (and very formal) brief reference to multipolarity and a strong appeal to protect the rights and interests of Russian citizens: “Russia will protect them wherever they are” (Unity 2000). Neo-isolationism, restorationism, Westernism, and multipolarism have all received much discussion in analyses of Russian foreign-policy thinking. The final approach – pragmatic selectivism, or selective engagement – is less well known. Because it is moving toward a more central position, though, it deserves elaboration. It began to emerge in the election campaigns of 1999 and 2000. Its basic ideas resonate with the pragmatic (self-interested)
Foreign Policy Implications of the Elections
85
orientations of the foreign-policy platforms of both Primakov’s OVR and Putin’s Unity. This new orientation is based on disillusionment with the West, acknowledgment of the limited sphere of Russian responsibilities in the world arena, concentration on key foreign-policy directions, and a preference for a pragmatic and nonideological approach (Fedorov 1999). Pragmatic selectivism promotes the concepts of “enlightened national self-interest” and “maximum streamlining of Russian foreign policy” (Kokoshin 2000). Another important element of this orientation is its criticism of the previously dominant “multipolarist” approach as a confrontational and outdated revival of a nineteenth-century world view (Trenin 2000; Zagladin 2000). The pragmatic selectivism/selective engagement approach assumes that Russia will continue to lag behind the industrial countries of the West and the East for at least the next fifteen to twenty years. This results, on the one hand, in a trend toward Russia’s isolation from the West, and, on the other hand, in a trend toward Russian self-isolation (largely due to the disenchantment of a large strata of Russian liberals with the West). Moreover, large segments of the new Russian economic class actually fear foreign competition and thus do not really welcome the creation of an economic climate favorable to foreign investments. In the words of one of the proponents of this approach (Karaganov 2000), “Russia’s major foreign-policy problem is the still unresolved question of what is more important: economic development and well-being, laying the groundwork for future influence, or present-day prestige – which is, moreover, frequently purely ephemeral.” Accordingly, one is tempted to compare Russia today to postwar Germany and Japan, countries that consciously preferred economic revival in the interests of public prosperity over greatpower ambitions. The foreign-policy implications are clear and simple: to avoid confrontation while preserving effective nuclear capabilities, to attract foreign investments, to avoid tough rhetoric, and, in short, to be pragmatic and selective. This pragmatic selectivism/selective engagement approach has not yet been fully represented in the foreign-policy programs of any particular political group or party. It is still, strictly speaking, largely an intellectual position. However, it is important to mention that different elements of this position can be traced to specific political platforms, views, and statements, including a few by Vladimir Putin himself. It would certainly be premature at this point to make predictions related to this particular foreign-policy orientation or to the foreign-policy directions of the Putin regime. Nevertheless, this pragmatic selectivism/selective engagement approach bears watching. Putin’s Foreign-Policy Perspective Although some analysts have claimed that Putin’s foreign-policy doctrine is already basically clear and his foreign-policy line quite predictable
86
Andrei Melville
(Nikonov 2000), his swift change of policy after September 11 disproved that. Even so, Putin’s programmatic statements, as well as his foreign-policy actions since the elections, deserve attention as they may offer a glimpse of Russia’s emerging and evolving foreign-policy line. In his few programmatic statements, Putin has stressed some general philosophical observations that seem to be central in his worldview. He is definitely not on the side of those who like neo- or self-isolationism for Russia. On the contrary, he looks forward to Russia’s integration in global processes. Europe in this respect is a special object of Putin’s interests. He cannot see Russia “isolated from Europe and the rest of civilized world” because “Russia is a part of European culture” (Putin 2000). In his “Millennium article,” Putin (1999) makes another remarkable contrast: he speaks of the seventy years of Soviet rule as an impasse and considers it an important accomplishment that Russia has finally chosen “the mainstream of the Western civilization.” This may mean that in this respect Putin at least partly sides with a moderate version of Russian “Westernism.” Putin does not idealize globalization. In his “Millennium article” he points at its negative effects, such as ecological damage, unequal distribution of its fruits, and the emergence of “new rich” and “new poor” on the international level. He almost seems to say that globalization is not global and that Russia badly needs to catch up with the “front-runners.” Otherwise, says Putin, Russia may easily and pretty soon find itself among the second- or even third-rate states. One may appreciate that these statements, implying a de facto recognition of Russian decline, come from an acting President in the zenith of his preelection popularity. “Who is to blame?” “What is to be done?” To these traditional Russian questions Putin gives different answers. He is not (yet?) in a position to point concretely at those who are to be blamed. Instead, he prefers to see the general and hardly disputable decline of the Russian state and the loss of will as the source of national problems. In his Open Letter, Putin (2000) is blunt: the main problem in Russia today is the loss of will. The will of the State in the first place is weak, but the popular will has also declined. Putin would also add that Russia needs a resolute and decisive leader (and, as noted previously, Russia today would definitely agree). Of course, a strong state (especially if one takes Russian tradition into account) may be harmful. Yet there is little doubt that a weak, fragmenting, and declining state (especially when it has been “privatized” by an oligarchic group) cannot guarantee democracy and reforms at home, and effective integration into the world community abroad. It is therefore important to note that, at least once (in the “Millennium article”), Putin (1999) tried to clear up the picture and stressed that the power of the state today depends not so much on the military component but rather on its ability to be a leader in new technologies and to provide well-being for the people. Among Putin’s areas of priority, the Commonwealth of Independent States (CIS) was initially considered his highest. This referred not only to the “core”
Foreign Policy Implications of the Elections
87
CIS countries that are regarded as being politically close to Russia (like Belarus, Armenia, Kazakstan, Kyrgyzstan, and Tajikistan) but to Ukraine, Georgia, Azerbaijan, Uzbekistan, and Turkmenistan as well. Two additional points regarding Putin’s policy toward CIS countries are important to mention. The first is Putin’s stress on pragmatic and nonaltruistic economic cooperation when rhetorical reverence is not enough and when Russia expects the bills for oil and gas to be paid. The second is his hint of possible preemptive strikes against terrorists in foreign countries neighboring the members of the CIS (i.e., Afghanistan) because such strikes threaten Russian security and the security of Russian allies in the region. A European orientation is extremely important for Putin’s foreign policy (e.g, as shown by the geography of the Russian President’s foreign visits). It may even have been seen as a counterbalance to Russian-American relations when the latter were stumbling. Through the end of the Clinton era and despite the summit in 2000, Russian-American relations seemed to be restricted largely to the not-very-promising sphere of arms control. Unfortunately, the incoming Republican administration’s resumed confrontational spirit exacerbated this coolness – at least until Bush and Putin’s Ljubljana summit. Finally, the East represents another important focus of Putin’s attention as a source of threat (from Muslim extremism, first of all) and of foreign-policy priorities. Purely pragmatic considerations seemed to be guiding Russia’s foreign policy in the region (e.g., reflected in Russian policy toward the economically and geopolitically important countries of Iran and North Korea). At the same time, Russian-Indian and Russian-Chinese relations are pretty nebulous – something that eventually has to be addressed, given the strategic importance of these Eastern powers. From the very beginning, something else was striking in Putin’s foreignpolicy statements: his avoidance of tough multipolarity and great-power rhetoric, even in comparison to the centrist OVR or liberal Yabloko. Instead, he definitely prefers a businesslike, pragmatic tone and approach. Putin does not dramatize Russia’s loss of empire. He considers this to be a positive development (unlike the aggressive isolationists and restorationists in Russia’s elite): “We do not dictate our will upon others, but instead we have time and resources for ourselves” (from Putin’s Open letter). Like Russian Chancellor Gorchakov, “concentration” is one of Putin’s key words. However, at least at this moment, Putin’s “concentration” does not imply international behavior “as if Russia continues to be a Great Power” (which, actually, was Primakov’s ideological foreign-policy recipe). Rather, “concentration” for Putin means focusing on domestic problems and a very pragmatic and selective approach to foreign policy as a tool for serving pressing domestic, primarily economic, needs. He does not seek grandiose foreign-policy goals, stable military-political alliances, and other impressive international schemes. Instead, he is very selective (one may be tempted to say, like his Chinese colleagues).
88
Andrei Melville
In particular, since his first days in office President Putin has emphasized that Russia’s economic interests should be considered the primary diplomatic goal. Putin’s idea of Russia’s integration into the world economy stresses three points: (1) state support of Russian businesses abroad; (2) state measures against Russia’s discrimination in the world markets; and (3) Russia’s participation in international trade organizations. Putin’s pragmatic and selective approach with a strict economic focus veers close to being a new austerity principle: “If Russia is being urged to address global concerns costing a lot of money . . . we must weigh up our capabilities and perhaps wait a little. . . . Only the country’s real interests, including its economic ones, should be the law for Russian diplomats” (Putin 2000). While Putin’s world view and his general approach to the international arena give him substantial foreign-policy flexibility, the events of September 11 forced him to make choices. After the terrorist attacks, his prompt reaction of support for the United States in general and for President George W. Bush, Jr. in particular could have been seen initially as just a gesture of compassion. However, as it turned out very soon, it in fact heralded a chain of unexpected foreign-policy moves and departures from past practice. They may even constitute the embryo of a new “Putin doctrine.” Not only is Putin personally involved in this foreign-policy reorientation but it originated directly from the President’s office in the Kremlin and was only imperfectly coordinated with other institutions (including the Ministry of Foreign Affairs and the Ministry of Defense). At present the more-or-less visible elements of the embryonic “Putin doctrine” are (1) a deliberate pro-Western political and ideological thrust that is much more radical than the “Westernist” views he had previously espoused; (2) de facto acceptance of American use of its military power right next to Russia’s borders (that is, in Afghanistan); (3) recognition and toleration of the projection of American military power into regions traditionally considered to be Russia’s spheres of influence (Central Asia and the Caucasus); (4) reorientation in Russia’s policy toward at least some members of the CIS, accepting greater flexibility of their domestic (including attitudes toward separatist movements, like in Moldova and in Georgia) and foreign-policy options; (5) a “softening” of Russia’s positions on some key arms control issues (like the Anti-Ballistic Missile (ABM) treaty); and (6) indications of changing priorities in Russia’s Middle-East policy, becoming less pro-Arafat, for instance. In a way, Putin’s pragmatism and his general ideological and programmatic flexibility, which were discussed previously, facilitated this reorientation while preserving at least some continuity with the pre–September 11 policies. Still, the obvious changes in Putin’s foreign policy produced further divisions within the Russian political class as well as among the public. Quite indicative was the result of Putin’s discussion with the leaders of factions of both chambers of the Federal Assembly of how Russia should react to
Foreign Policy Implications of the Elections
89
U.S. actions against the Taliban: twenty-one of them supported his policy, while one backed the Taliban and eighteen voiced support for Russian neutrality – that is, nineteen de facto opposed the President’s new foreignpolicy line (Yanov 2002). A sharp division of opinions is clear. This division also exists among analysts, with some calling for a new domestic consensus in favor of the “Western choice” (Shevtsova 2002) and a “pro-Western foreign policy” (Adamishin 2002), and others deploring Russia’s “geopolitical retreat” (Migranian 2002). Public opinion is split as well, but with a large proportion of the public disagreeing with Putin’s shift. In September 2001, 69 percent were against providing military bases in Central Asia to the United States during the antiterrorist operation; 48 percent disapproved the American plans to conduct antiterrorist actions in Afghanistan; 73 percent insisted that the United States should seek the UN Security Council approval before taking any military actions in Afghanistan; and 54 percent believed that Russia should remain neutral (VTsIOM 2001 [a]). Despite Putin’s very high approval ratings, fewer than half of Russians (48 percent) agreed with his October 2001 statement that U.S. actions in Afghanistan were “justified and balanced” (VTsIOM 2001 [b]). Moreover, while in March 2002 Putin reacted to Georgia’s invitation of the U.S. military specialists with demonstrative calmness, almost two-thirds (61 percent) of Russians felt this directly affected Russia’s interests. Fifty-five percent were blunt – this threatened Russia (Public Opinion Foundation 2002). Putin’s resources for shaping Russian foreign policy are formidable. If his new, post–September 11 foreign-policy posture is to develop into a “Putin doctrine” with a coherent foreign-policy strategy and a uniformity of practice, it must gain strong support from the Russian political class and the public. His challenge, then, is to overcome the elite and mass divisions that arose in response to his post–September 11 moves.
5 Parties, Leaders, and Voters in the Parliamentary Election Timothy J. Colton
Students of regime change trenchantly argue that the more election campaigning and voting behavior are pegged to political parties, and the less to candidates, the brighter the prospects for nurturing democracy (e.g., see Aldrich 1995; Mainwaring 1999). This claim rests on two pillars. The first is the observation that, as collective actors with professional staffs, parties are better suited than solo leaders to presenting citizens with principled choices. A party-based mass politics will tend to be richer in issue content than a leader-based politics revolving around personal virtues and vices, populist rhetoric, and clientelistic networks. It should enable citizens to make more judicious and more value-impelled voting decisions and to be more forceful in demanding governmental accountability to them after election day.1 The second rationale for the superiority of parties has to do with their perseverance over time. Parties and the loyalties they engender are capable of persisting for many more years than the political career or the life span of a single president, prime minister, or party chairman. Specialists on the microfoundations of voting come away, in the words of a classic text, “impressed by the degree to which the emotional attachments formed by voters toward specific political parties . . . become remarkably durable . . . as well as pervasive in their effects, lending a predictable coloration to all manner of political evaluations” (Converse and Pierce 1986, 68). If democratic consolidation is ultimately about entrenching popular sovereignty in stable institutions and norms, it is easier to imagine parties than political personalities emerging as the resilient bricks and mortar of the new system. Human mortality, the 1
“In fluid systems [with weak parties] the electorate votes more according to personality or clientelism than party; antiparty politicians are more able to win office. Thus populism and antipolitics are more common with weakly institutionalized systems. Personalities rather than party organizations dominate the political scene. Given the propensity toward personalism and the comparative weakness of parties, mechanisms of democratic accountability are usually weaker; consequently, the quality of democratic practice tends to be lower” (Mainwaring 1999, 37–8).
90
Parties, Leaders, and Voters in the Parliamentary Election
91
vicissitudes of the struggle for recognition and votes, and, in some nations, statutory term limits cap a public figure’s period in the limelight at a decade or two on the outside. In the right hands, parties can outlast their creators, their luminaries of the moment, and for that matter the voters they cultivate, updating and renewing the terms of the association as they proceed. In not a few countries, major parties match the longevity of the democratic setup that sanctions multiparty competition. The prescriptive case for effective parties is as cogent in the Russian Federation as in any transitional polity. Russian politics and governance would benefit immensely from the construction of an orderly party system. The rudiments of such a system would already appear to be in place. Associations dubbed parties and displaying their essential characteristics, having debuted during the twilight of Soviet power at the turn of the 1990s, are conspicuous players on the post-Soviet stage. They have built organizational machinery from scratch, taken part in all national elections since 1993, and been ascendant in the State Duma, the paramount lower chamber of the federal legislature (see Smith and Remington 2001 and Chapter 11). In numerous respects, just the same, Russia’s parties have fallen woefully short of the ideal type drawn from the experience of the older liberal democracies – the smooth-running mechanism that manages the competition for power and aggregates key societal interests. Not without reason, skepticism about the communication skills, programmatic consistency, and staying power of Russia’s fledgling parties is prevalent in the Western literature (e.g., see Rose 2000, and Chapters 8 and 9).2 Many analysts go on to assert that the combination of underachieving parties and the uncertainty rampant in the transitional milieu has bred the egregious personalization of Russian politics – an evil that the comparative scholarship on democratization urges be avoided. Richard Sakwa (1996, 115), for example, wrote that because support for Russia’s immature parties is “friable and susceptible to rapid changes,” voters fixate on individuals as opposed to party labels or ideological platforms. Boris Yeltsin’s landslide victory in the presidential election of 1991, on the eve of the Soviet collapse, and the subsequent rise of the demagogic Vladimir Zhirinovsky bear witness, Sakwa said, to “the degree to which support for charismatic personal leadership is structured into Russian politics” (Sakwa 1996, 115). Jerry F. Hough, in the same vein, beams the critique of personalization at the parties themselves, portraying the first batch of them in Russia as “highly personalistic and ephemeral” (Hough 1998 [b], 688). Domestic discourse echoes these foreign voices. One Russian pundit, Susaina Pshizova, sketched what had become the regnant interpretation, also focusing on personalism within 2
For background discussion on Russian parties, see Fish 1995 (a), 1995 (b); Urban 1997; ¨ White, Rose, and McAllister 1997; Hough 1998 (b); Lowenhardt 1998; Wyman, White, and Oates 1998; and Colton 2000.
92
Timothy J. Colton
the parties: “It is understandable that, given the absence in the postsocialist countries of old parties with firm reputations, the leadership factor is strengthened many times over. In Russia only the CPRF [Communist Party of the Russian Federation] has a more or less firm reputation . . . so that a voter can orient himself mainly to the party and not to the leader. In all other cases the authority, charisma, or reputation of concrete leaders defines not only the success but the very existence of the organization” (Pshizova 1998, 106). Sweeping declarations about the relative weakness of parties and the relative strength of leaders in protodemocratic Russia have seldom been buttressed by hard evidence about mass perspectives, as distinct from the conduct of the elite players. The little opinion-based research dealing with leadership influences on the vote mostly addresses elections for president (see especially Colton 2000, ch. 6).3 Yet presidential contests as they are now orchestrated in Russia supply the feeblest hope for sustained party development. Any systematic comparison of the roles of parties and personalities best belongs in the parliamentary setting. My aim in this chapter is to extract some clues pertinent to the puzzle from fresh data generated by a panel survey of a large probability sample of the Russian electorate. Michael McFaul, Russian colleagues, and I conducted the survey before and fast on the heels of the December 1999 election to the State Duma. A sifting of the data shows that citizen evaluations of party leaders do have effects on electoral choice, that it is possible to disentangle them from budding affinities for parties, and that these effects are approximately equal in intensity to the effects of protopartisanship. The implications for current and future Russian politics are mildly encouraging.
some preliminary considerations Neither the rules of the game nor the wider political context have stimulated party involvement in Russian presidential elections. The procedures for electing a president tolerate participation by the parties in nominations and campaigning but do not mandate it. Although some parties, notably the Communist Party of the Russian Federation (CPRF), Yabloko, and the Liberal-Democratic Party of Russia (LDPR), have repeatedly proposed presidential candidates and beaten the drum for them, the winners in the three elections to date have all followed the nonpartisan route. Yeltsin, the architect of the post-Communist state, set the mold. Once out of the Communist Party of the Soviet Union (CPSU) in 1990, he staunchly refused to father his own political party or to throw in his lot with anyone else’s. Elected president in 1991 as a freelance crusader for reform, Yeltsin won a second term 3
See also the thoughtful and well-researched article by Miller, Reisinger, and Hesli (1998). It does not deal with voting.
Parties, Leaders, and Voters in the Parliamentary Election
93
through the same formula in 1996, his papers filed by an eclectic coalition of more than two hundred groups.4 Vladimir Putin mimicked the strategy in crafting his election triumph in March 2000. As prime minister and anointed successor to Yeltsin in the autumn of 1999, Putin had conferred a backhand endorsement (“as a private citizen”) on Unity, the new pro-Kremlin bloc in the Duma election.5 But, like Yeltsin, he accepted nomination for chief executive from a conglomeration of parties, allied organizations, and celebrities. On the campaign trail he stood serenely above the partisan fray. It remains to be seen whether Putin will lend his active support to, or perhaps take charge of, efforts to upgrade Unity into a coherent “party of power” (Colton and McFaul 2000). Elections to the Russian parliament are more conducive to checking out the effects of partisan and personalistic factors. Unlike presidential elections, campaigns for the State Duma by design privilege political parties. The hybrid rules and regulations on the books in one version or another since 1993 – stipulating that voters select half of the 450 deputies by proportional representation (PR) from among national candidates’ lists submitted by registered parties or quasi-parties, and that the parties may also nominate candidates in the territorial districts – were drafted with the explicit intent of catalyzing party formation (Remington and Smith 1996; Hough 1998a). To go by the roster of the twenty-six parties entered in the third Duma election, in December 1999, that objective has to an appreciable extent been met. The six organizations whose lists hurdled the 5 percent threshold for admission into the Duma were markedly heterogeneous in political outlook: the socialistic CPRF; the pro–status quo, progovernment Unity; OVR (for OtechestvoVsya Rossiya, or Fatherland-All Russia), a left-of-center grouping anchored in parts of the political establishment disenchanted with Yeltsin’s rule; two liberal formations, SPS (Soyuz Pravykh Sil, or the Union of Right Forces) and Yabloko; and the Zhirinovsky Bloc, a proxy for Zhirinovsky’s nationalist LDPR, which had been disqualified for violations of procedures in nominating its list.6 4
5 6
Yeltsin did take the lead in the formation of the progovernment Our Home Is Russia (NDR) movement, headed by Prime Minister Viktor Chernomyrdin, in 1995, and most of its supporters were to vote for him in the 1996 presidential election. But Yeltsin refrained from explicitly endorsing the party in 1995, and showed no interest in entrusting his reelection campaign to it in 1996. Putin also telegraphed approval of the liberal SPS slate, stressing the attractiveness of their economic program. Fish and Hanson, in Chapters 8 and 9, rightly point out that there were many contradictions in the formal programs of a number of the parties entered, and that Unity did not bother to release a written program at all. My reading is, nonetheless, that the parties, including Unity, projected quite different images of how they would govern. Individual-level evidence from surveys of the electorate shows that many voters discerned some significant differences among the parties on “the issues,” and that consequently it is legitimate to group the parties by programmatic group or family.
94
Timothy J. Colton
If this had been a parliamentary election in a well-anchored liberal democracy, we would have anticipated that citizen attitudes toward the parties as such would have done much to drive the vote in 1999. And yet, this prediction need not hold true in transitional societies, where the chaotic political environment and the amateurism of the bulk of the participants depart so much from the Western archetype. The qualitative sophistication and capacity of Russian parties, as their detractors have duly noted, have lagged dramatically behind their quantitative proliferation. Most of the dozens of organizations that call themselves political parties or movements show no signs of life other than a tiny cadre of part-time activists, a Moscow office suite, and, for the most striving of them, a Web site and a financial benefactor. The barriers to entry and exit being extraordinarily low, parties in Russia are often flashes in the pan, rarely making much of a stir as they come and go. Three of the six entities that gained Duma seats through PR in 1999 – Unity, OVR, and SPS, which among them took 44.6 percent of the party-list votes cast – were launched in the summer and early autumn of 1999, with election day looming. The other three, with 36.4 percent of the total vote – the CPRF, Yabloko, and the LDPR/Zhirinovsky Bloc – are grizzled veterans in this company, dating back to the early 1990s.7 Altogether, but eight of the twenty-six parties and near parties that contended in 1999 had been around for the previous Duma election in 1995; four (the CPRF, Yabloko, the LDPR, and Women of Russia) were in on the inaugural post-Soviet election in 1993. Many parties in the race, and many of those too lame to make it onto the ballot, had weathered recurrent reorganizations, intramural feuds, and splits. In so volatile a situation, one cannot help but wonder what kind of “predictable coloration” parties can conceivably impart to electoral decisions. Absent established parties, it is not far-fetched to visualize citizens resorting to personalities as an alternative and legible guide to civic choice. In a partisan parliamentary election waged according to PR rules, the relevant personalities do not stand aloof from the parties, as may be the case in a presidential election where candidates have the option of soliciting votes as nonpartisans. Rather, they are in the thick of things, the official heads of the very same parties on whom the election hinges. The law governing elections to the Russian Duma in 1999 entitled any participating party to inscribe the names and vital statistics of up to three of its leaders, in order of precedence, on the PR ballot slip. Only one of the entrants (the lilliputian Social-Democratic Party of Russia) declined to take advantage of this clause. It is also standard practice in Russia for parties to showcase their leaders in 7
The CPRF and Yabloko were both founded in 1993 and won seats in the 1993 and 1995 Dumas. The LDPR was born in 1990 and was also represented in both the 1993 and the 1995 Dumas. Its leader, Zhirinovsky, ran for president in 1991. The CPRF can, of course, trace its lineage back to the CPSU, which Yeltsin banned by decree in 1991.
Parties, Leaders, and Voters in the Parliamentary Election
95
their television commercials, billboards, and printed handouts, normally depicting them in an action shot or a studio interview as striking a pose of statesmanlike concern for their compatriots. This was very much so in the 1999 campaign. Had Russian voters in 1999 zeroed in on designated leaders and not on the parties and quasi-parties as organizations, they would have discovered politicos of considerable stature, and sporting very diverse styles, occupying the senior spots on the candidates’ lists of all the prominent parties. Each had been a household name long before wading into the latest Duma campaign. Gennadii Zyuganov, the droning secretary-general of the neo-Communist CPRF, had picked up twenty-four million first-round votes as runner-up to Yeltsin in 1996. Sergei Shoigu of Unity was Minister of Emergency Services in the federal cabinet for most of the 1990s. This made him the ubiquitous dispenser of relief to victims of the natural and man-made disasters in which Russia abounds, and the beneficiary of the attendant publicity. Yevgenii Primakov of OVR and Sergei Kiriyenko of SPS had both served as prime minister of Russia. Primakov had done stints as internationalrelations adviser to Mikhail Gorbachev, director of Russia’s spy service, and Foreign Minister. The boyish-looking Kiriyenko, half Primakov’s age, had earned his spurs in business and political organizing in the provinces. Grigorii Yavlinsky of Yabloko, a mainstay of Russia’s perennial debate over economic reform, and Vladimir Zhirinovsky, the flamboyant initiator and mouthpiece of the LDPR, were long-standing party leaders and parliamentarians, fixtures on the evening newscasts for the past decade. Of the six, only Zhirinovsky plainly qualifies as charismatic, if we construe the charismatic leader, to paraphrase Max Weber, to be “a prophet, not merely a personality”(Ansell and Fish 1999, 284). The others, with the possible exception of Yavlinsky, embody “noncharismatic personalism,” the “dominant but uninspiring leader” who presides over a variegated political movement (Ansell and Fish 1999, 310). Arriving at judgments as to the actual importance of parties and personalities – and not speculation as to why either might hypothetically be important – is, for several reasons, more vexing in Russia than in many countries. It so happens that the more enduring of the Russian parties have stuck with the leaders who founded them and piloted them to a modicum of success. For parties headed since birth by one person, even attentive and conscientious voters could be hard put to sever their image of the organization from their image of the leader. The same holds for the analyst seeking to decipher Russians’ voting behavior. Muddying the waters further is the trend among Russian parties toward collegial leadership and balanced tickets. The practice originates in mergers, opportunistic alliances, and the exigencies of wooing varied constituencies. A loophole in Russia’s electoral legislation as of 1999–2000 has exacerbated the problem. To be eligible to contest a Duma election, parties have
96
Timothy J. Colton
had to register with the Ministry of Justice by twelve months before the vote. But preexisting parties, collaborating if they wish with other state-accredited public organizations, are permitted to band into electoral “blocs” at any time until the close of nominations. The product is typically a political hydra, with the star leaders of the constituent organizations yoked in an awkward partnership. Three of the six winners in 1999 conform to this type. The upstart Unity was hatched by two obscure parties (the National Patriots, composed of Afghan war veterans, and the Russian Christian Democratic Party) and a handful of other sponsors, including the politically well-connected “oligarch,” Boris Berezovsky. Sergei Shoigu, an electoral novice, was recruited to head up the Unity list because he had the credentials to act as a broker among its internal factions and between them and the Russian government.8 The OVR coalition, which finished third, put forward Yevgenii Primakov as its top spokesman, yet the men ranking second (Mayor Yurii Luzhkov of Moscow, who took the first steps to assemble the group in 1998) and third (Governor Vladimir Yakovlev of St. Petersburg) were scarcely less known to the national audience than Primakov. The leadership troika of the fourthplace SPS consisted of the heads of three of the extant liberal movements that spawned it (Sergei Kiriyenko, Boris Nemtsov, and Irina Khakamada). Even the venerable Yabloko, chaired for its entire history by Grigorii Yavlinsky, listed Sergei Stepashin, Yeltsin’s next-to-last prime minister, as its number two candidate in 1999. These and other subtleties of the party scene in Russia make disciplined study of the interrelationships among leadership perceptions, attitudes toward parties, and voting decisions a daunting task. Headway, nonetheless, can be made.
introspective analysis of the problem In asking if parties or personalities sway electoral choice in Russia, one clear-cut approach would be to inquire of the voters what they make of their own experience. There is utility in laying bare how citizens explain their decisions, although, granted, we cannot be hostage to these accounts. The firsthand information they provide should be of some intrinsic interest and will furnish leads that can be followed up through other methods. Respondents in the second (postelection) wave of our panel survey of Russians (see the Appendix to this chapter for technical details) were asked 8
Unity packaged itself as an insurgent bloc, despite its closeness to the presidential apparatus and to the political faction grouped at the time around Berezovsky and President Yeltsin’s daughter Tat’yana. Besides Shoigu, the second and third persons on the Unity ticket – Aleksandr Karelin, a world titleist in Greco-Roman wrestling, and Aleksandr Gurov, the former head of the organized crime division of the Soviet police ministry – were public figures of some repute.
Parties, Leaders, and Voters in the Parliamentary Election
97
table 5.1 Considerations Reported as Very Important to the Respondent in Deciding on Party-List Vote (Percentages)a By Party Voted For Consideration Work in the last Duma Program Personal qualities of the leader Defense of the interests of people like you Always vote for the party Weighted N
Overall
CPRF
Unity
OVR
33 54 60
61 66 59
11 47 70
19 49 68
57
69
55
31
64
1,403b
434
SPS
Yabloko
LDPR
13 45 57
44 54 56
49 63 66
47
47
49
61
7
12
10
41
9
369
184
130
98
57
a
Question asked in postelection wave. Excludes 37 cases where respondent voted against all parties and includes 133 cases where respondent voted for a party other than the six given in columns two through seven. Source: 1999–2000 Colton-McFaul election survey. For details, see Appendix. b
a series of closed-ended questions about whether, how, and on what basis they had voted on December 19, 1999. Two of the queries submitted to respondents who had turned out to vote for a party on the PR ballot shed light on our topic.9 The first invited respondents to gauge the importance of five considerations related to the party or bloc of choice to their voting decision: approval of its work in the last State Duma, its announced program, the personal qualities of its peak leader, its ability to defend the voter’s interests, and the habit of always voting for the party. The second query was a counterfactual question asking if the respondent would have voted for the same party or bloc had it been led by a different person. Table 5.1 presents the results of the first debriefing. As can be spied in the first column, they lend credence to the proposition that leadership perceptions and party-list voting are somehow linked. Sixty percent of our respondents, marginally more than for any other category of explanation, said the leaders’ personal qualities were “very important” to their decision to vote for the party; 21 percent opted for the adjacent rating, “important,” on the scale. A trifling 3 percent felt that the leader’s qualities were “completely unimportant” to them. This is by no means to say that the leadership factor, in the eyes of Russian voters, has the field to itself. Aside from leadership, more than half of the electors we polled attached the utmost importance to the program of the party they supported and to its putative defense of the interests of people 9
Some of our survey respondents did not vote, and a few of them voted for a candidate standing in their local Duma district but not for one of the party lists.
98
Timothy J. Colton
like them. And about one-third said the work of the party they fancied in the most recent Duma or the fact that they “always” vote for the party or bloc was very important. The vast majority of our respondents offered more than one reason for their decision. Eighty-two percent of all respondents who gave a “very important” response went on to adduce one or more others; 26 percent averred that all five considerations were very important to their decision. Of those respondents who said the personal qualities of the party leader were a very important impetus, 88 percent gave at least one other stimulus as being very important; 32 percent insisted that all five considerations were very important to them. As the second through seventh columns of Table 5.1 show, the parties vary some in the salience ascribed to justifications for supporting them. The reasons for voting on which the survey responses are most scattered are the two – work in the most recent Duma and customary voting – that should not have been germane to the three neophyte parties in 1999. Whereas in the vicinity of 40 to 65 percent of supporters of the three older parties (the CPRF, Yabloko, and the LDPR) said they had been strongly steered by these considerations, only about 10 to 20 percent of those who voted for the three newcomers (Unity, OVR, and SPS) said the same. That subset of the supporters for the newer parties betrayed some confusion and, in all likelihood, wishful thinking and backward projection from voting preference.10 For personal qualities of the leader, party program, and empathy with the voter, citizen assessments of the six core parties in 1999 manifest less variation across the parties. The proportion of those polled who proclaimed that the personal traits of the party leader were very important to their voting choice ranges from 56 percent (for Yabloko) to 70 percent (for Unity). For CPRF supporters, leadership qualities are, by a slender margin, the least frequently tendered pretext for the vote; for the other five parties, leadership is the most frequently reported factor. Answers to the polling question about willingness to vote for the same party or bloc if it had been led by someone different (see Table 5.2) add some depth to the picture before us, although the payoff is circumscribed by the absence of counterfactual items that would delve into other dimensions of the voting decision.11 All in all, more voters said they would stand pat with their party of choice after a leadership change (37 percent) than said they 10
11
For example, persons who said they “always” voted for SPS may have been thinking of past votes for one of the liberal organizations that set SPS up in 1999. SPS propaganda made much of the cooperative spirit in which its founders had acted. Likewise, respondents who cited the Duma role of Unity, OVR, or SPS may have been inspired by memories of individual members of their 1999 slates performing in the Duma or in other political posts. We do not know, for example, how many citizens think they would have changed their vote if their party of choice had refashioned its program or stopped defending the interests of ordinary people. To pursue such leads across the board in a survey would eat up interview time with tedious queries and run the risk of alienating respondents.
Parties, Leaders, and Voters in the Parliamentary Election
99
table 5.2 Responses to Question about Whether Individual Would Have Voted for Same Party if Led by a Different Person (Percentages)a By Party Voted For Response
Overall
CPRF
Unity
OVR
SPS
Yabloko
LDPR
Yes No Depends on the other leaderb Don’t know
37 21 33
61 9 24
19 31 39
12 30 49
29 25 38
34 19 36
25 39 28
8
6
11
9
9
11
9
a
Question asked in postelection wave. Weighted Ns same as in Table 5.1. Volunteered response. Source: 1999–2000 Colton-McFaul election survey. For details, see Appendix.
b
would switch (21 percent); one voter in three opined that it would depend on who the alternative leader was. The differences among the parties by this yardstick are much starker than on the first survey question. The CPRF stands apart from the crowd, as its supporters were the only voters for whom “yes” was the modal response (61 percent of CPRF supporters would unflinchingly have stood by the party if Zyuganov had ceased to be its leader). At the opposite pole, no more than 12 percent of OVR voters said they would have voted for it if somebody other than Primakov had headed its ticket.12 It is worth noting that amenability to deserting the party of one’s choice in the event that its leader was replaced is not the only way sensitivity to leadership can be expressed. The most leader-centric respondents in our postelection survey, as indicated by the belief that the personal qualities of the party leader were very important to their choice of party list, proved to be only slightly less likely than the average individual to say they would have voted for the party of their choice even under a new leader; they were a bit more likely than average to give the agnostic “depends on the leader” response to the second question.13 Faith in the primacy of leadership, in other words, can find outlet in two very different behaviors – a readiness either to fall in behind the existing leader or to search for alternatives to that person. The fruits of this self-scrutiny are moderately instructive and thought provoking. Most Russians, we have seen, were convinced in 1999 that their personal assessments of the party leaders had made an impact on them in the 12
13
Reluctance to shift one’s vote in response to a leadership change was perceptibly greater among older voters in 1999–53 percent for persons over the age of sixty-nine, versus 23 percent for those younger than thirty – but this is largely a reflection of the much higher rate of support for the CPRF among older Russians. Thirty-two percent of party-list voters in 1999 who said they thought the leader’s qualities to be very important to their choice would have voted for the same party even if its leader were someone different; 24 percent would not have done so; and 37 percent would have waited to see who the new leader was.
100
Timothy J. Colton
voting booth. A substantial minority also conjectured that they would have voted otherwise if leadership of their approved party had changed hands. Differences among the parties spark curiosity about why this might be so. And divergence in the response patterns to the two questions underlines the sensitivity of citizen assessments to question wording and more generally to the way the issue of leadership effects is framed. This having been said, the evidentiary value of these regularities is dimmed by respondents’ faulty recall, by their proclivity for concurring with equal enthusiasm in multiple interpretations of their conduct, and, it is fair to presume, by the inherent limitations of individuals operating as “lay scientists” probing the attitudinal bases of their own behavior. Those limitations in mind, citizens’ testimony about the role of leadership evaluations in sculpting their electoral choices probably tells us less about those private deliberations per se than about popular notions about mass political behavior from which they reflexively borrow in formulating their self-analysis.14 Our safest bet, hence, is to take the first-person depositions of our survey informants as an intriguing start down the path to understanding the problem at hand. To push forward, it is best to capitalize on the more conventional and more authoritative technique of amassing evidence about citizen attitudes from survey interviews and, working in the third person, calibrating it with the political outcome through careful statistical modeling.
evaluations of parties and party leaders As an embarkation point, we need discriminating indicators of how Russians evaluate their parties and party leaders. These, plus measures of the voting choice, will be the grist for the statistical analysis. In our postelection survey in 1999–2000, in which we executed verbatim the election module of the Comparative Study of Electoral Systems (CSES), interviewers asked all respondents to rate the half-dozen parties that made it over the 5 percent threshold, their principal leaders, and several other politicians on an affect scale going from zero (least liked) to ten (most liked). Respondents were more confident sizing up leaders than parties, but not by much. Seventy-two percent volunteered scores on the scale for all six of the parties, with an average of 5.26 evaluated; 78 percent could size up all six leaders, or a mean of 5.52 leaders per respondent.15 The summary evaluations of the parties are 14
15
“Actors’ insights into the causes of their behavior are best regarded as inferences rather than as privileged or ‘direct’ observations of the workings of their mental machinery. . . . Frequent errors in the self-ascription of emotions and attitudes would appear to be inevitable, given the inadequacy and inaccuracy endemic to causal explanations of one’s own behavior” (Nisbett and Ross 1980, 223, 227). Similar questions about ten parties in 1995 yielded a greater discrepancy between parties and leaders – an average of 5.8 parties recognized and evaluated and an average of 7.0 leaders recognized and evaluated (Colton 2000, 186).
Parties, Leaders, and Voters in the Parliamentary Election
101
highly correlated with evaluations of their top-most leaders. In our survey sample, the correlations (Pearson’s r ) run from .64 (for the correlation between Unity and Shoigu) to .86 (for the CPRF and Zyuganov), averaging .73 for the six dyads. These questions about generalized affect are akin to the U.S. National Election Studies’ “feeling thermometer” questions (asking survey respondents to rate on a 101-point scale the Democratic and Republican parties, their presidential nominees, and candidates for the House of Representatives and Senate). The catch, as analysts of American politics have pointed out, is the possibility that measures like these will be compromised by endogeneity, that is to say, by feedback from the act of voting. Respondents may have warm feelings toward a political party or a leader precisely because they voted for, or intend to vote for, that party or leader. This prevents us from using their warm feelings to elucidate why they vote as they do.16 The use of more fine-grained attitudinal measures, possessing greater semantic and logical separation from the voting outcome, will minimize the ` danger of endogeneity. Vis-a-vis political parties, I measure “transitional partisanship,” which can be interpreted as a benchmark of psychological attachment to a political party or quasi-party, a precursor in a democratizing polity to the socialized partisan identification in the Western democracies. I classify a transitional partisan in 1999 as a respondent who in the preelection wave of interviews (1) either indicated that there was a party or political movement he (or she) considered to be “my party,” (2) or, failing that, agreed that there was a party or movement that reflected his interests, views, and concerns better than the others, and (3) was also able, unprompted, to name the party or quasi-party with which he felt the affinity. Forty-three percent of our survey sample qualified as partisans in late 1999, leaving a majority, 57 percent, as nonpartisans.17 The CPRF basked in the allegiance of the largest number (17 percent out of the 43 percent), followed by Yabloko (6 percent), OVR (5 percent), Unity (4 percent), the LDPR (3 percent), and SPS (2 percent); minor parties accounted for another 7 percent. For the three older parties, there is evidence of some temporal continuity in partisan sentiment. Eighty percent of those CPRF partisans who recollected voting in the Duma election of 1995 said they had voted for the CPRF then; 73 percent 16 17
See Miller and Shanks (1996, 389). They cite the revealing discussion in Rahn, Aldrich, and Borgida (1994). Of the 43 percent, 14 percent were what I call strong partisans (individuals who had “their” party and said it reflected their interests and concerns completely), 11 percent moderate partisans (persons with a favored party which, they said, reflected their interests only partially), and 18 percent weak partisans (who answered the “my party” question in the negative but said there was nevertheless a party or quasi-party that reflected their interests and concerns more than any other party). See the background discussion in Colton (2000, ch. 4).
102
Timothy J. Colton
of Yabloko partisans and 64 percent of LDPR partisans said they had voted for the selfsame party in 1995.18 Concerning party leaders, I begin with cognizance of specific personal qualities, not with an omnibus measure of affect. The 1999 postelection interview contained a battery of items about the six major party leaders and certain other national political figures. The interviewer recited a list of desirable leadership attributes, all of them much discussed in the comparative literature on political behavior, and asked how the respondent would appraise each leader in terms of each attribute: intelligence, strength, integrity, and empathy.19 The allowable responses were “yes,” “probably yes,” “probably no,” and “no.” The distribution of attitudes toward the six top party leaders is laid out in Table 5.3. Popular perceptions, as is evident, run the gamut from the dismissive to the reverential. All in all, positive responses about the leaders outnumber negatives. Even for the black sheep in the group, Vladimir Zhirinovsky, there are more positives than negatives on one of the four character traits (intelligence). As captured by the categorical “yes” responses, Yevgenii Primakov of OVR led the others in popularity, garnering more accolades than his five peers for intelligence and integrity. Sergei Shoigu of Unity nosed out Primakov for strongest leader, and Shoigu and Gennadii Zyuganov of the CPRF finished in a dead heat for most laudatory rating on empathy with ordinary people. As would be expected, citizen assessments of the discrete character traits of particular party leaders are closely interlinked (see the top panel in Table 5.4). The average of the mean interitem correlations for the six candidates was .63 in 1999. Chances are that a survey respondent who harbored a good opinion of, say, Primakov’s intelligence would also give him a relatively generous rating on integrity or empathy. What would not necessarily be expected are the modest but significant associations uncovered when we peruse indicators of particular character traits across the roster of candidates (see the bottom panel of Table 5.4). Although less than one-third the average magnitude of the associations among 18
19
The prior voting record of partisans of the three newly minted parties was extremely diverse. Thirty percent of SPS partisans in 1999 said they had voted in 1995 for Russia’s Democratic Choice, the main predecessor of SPS, and the rest for a smattering of smaller parties. Twentyseven percent of Unity partisans said they voted for the CPRF in 1995, 23 percent for the liberal Yabloko party, 18 percent for the government party NDR, and 11 percent for the LDPR. Among OVR partisans, 27 percent had voted for the CPRF, 27 percent for Yabloko, and 25 percent for NDR. These four items, plus a fifth on the politician’s vision of the future, were administered to respondents in the survey William Zimmerman and I did after the 1996 presidential election. The questions related to presidential candidates only. We did not ask analogous questions about party leaders at the time of the 1995 parliamentary election, and to the best of my knowledge no other researcher did this, either. See Colton (2000, ch. 6).
Parties, Leaders, and Voters in the Parliamentary Election
103
table 5.3 Assessments of Personal Traits of Six Party Leaders (Percentages)a Trait and Assessment
Zyuganov Shoigu Primakov Kiriyenko Yavlinsky Zhirinovsky (CPRF) (Unity) (OVR) (SPS) (Yabloko) (LDPR)
Intelligence Yes 47 Probably yes 30 Probably no 10 No 5 Don’t know 9
42 31 3 2 21
57 31 4 2 6
42 30 8 5 14
42 32 8 5 13
26 25 20 21 8
Strength Yes Probably yes Probably no No Don’t know
32 28 20 7 13
35 30 7 3 25
34 29 21 6 11
14 24 30 13 19
13 24 31 12 20
19 21 26 22 12
Integrity Yes Probably yes Probably no No Don’t know
24 25 19 11 21
28 35 5 2 29
30 31 15 7 17
14 31 18 10 28
14 32 18 8 28
4 9 33 34 19
Empathy Yes Probably yes Probably no No Don’t know
24 23 20 15 19
24 30 8 5 33
20 28 20 13 18
10 24 23 14 28
10 26 24 12 27
5 9 30 38 18
N = 1,846 weighted cases. Source: 1999–2000 Colton-McFaul election survey. For details, see Appendix.
a
different dimensions of character for a single candidate, the fact that these correlations exist at all connotes the presence of an underlying dimension of something like “respect for leaders” that predisposes citizens to look kindly or unkindly on the personal qualities of all politicians, over and above the esteem in which they hold any one politician. This subliminal tendency should be kept in mind in exploring linkages with voting behavior.
associations with the vote It is immediately apparent that there are bivariate associations between partisan self-image and leadership evaluations, on the one hand, and voting in Russian parliamentary elections, on the other. As Figure 5.1 shows vividly,
Timothy J. Colton
104 table 5.4 Average Interitem Correlations for Leadership Evaluations Leader or Leadership Trait
Average Correlation (Pearson’s r )a
By Leader Gennadii Zyuganov (CPRF) Sergei Shoigu (Unity) Yevgenii Primakov (OVR) Sergei Kiriyenko (SPS) Grigorii Yavlinsky (Yabloko) Vladimir Zhirinovsky (LDPR) all leaders
.67 .66 .57 .62 .61 .62 .63
By Trait Intelligence Strength Integrity Empathy all traits
.26 .16 .17 .18 .19
a
Percent Who Vote for the Party
Missing values deleted pairwise. Source: 1999–2000 Colton-McFaul election survey. For details, see Appendix.
90 80 70 60 50 40 30 20 10 0 KPRF
OVR Partisanship
figure 5.1 Partisanship and the Party-List Vote, 1999.
Yabloko
Parties, Leaders, and Voters in the Parliamentary Election
105
Percent of Voters for the Party Who Are Partisans of the Party
60 50 40 30 20 10 0 KPRF
Yedinstvo
OVR
SPS
Yabloko
LDPR
Party
figure 5.2 Partisans in Relation to All Voters for the Six Main Parties, 1999.
self-described partisans voted in droves for the parties they professed to identify with in December 1999. For only the two smallest parties of the six returned to the Duma (Yabloko and the LDPR) was the proportion of avowed partisans who ended up voting for the party less than 60 percent; for two parties (Unity and SPS) it exceeded 80 percent. The variation across the parties is much more pronounced when we examine the share of these transitional partisans in each party’s pool of supporters (see Figure 5.2). Partisans of the allotted party comprised anywhere from 13 percent of voters for the party (for Unity) to 54 percent (for the CPRF). These consistent partisans were twice as plentiful in the electorates of the three old-hand parties (the CPRF, Yabloko, and the LDPR, for which they averaged 43 percent of the total) as in the electorates of the three new parties (Unity, OVR, and SPS, for which they averaged 21 percent).20 The reason surely is that by 1999 the holdover parties had already enjoyed a number of years to court and cement a partisan following. In the majority of cases their partisans in 1999 had voted for the party’s list in the Duma election of 1995. Figure 5.3 likewise conveys a zero-order association between leadership evaluations and the party-list vote in 1999. The leadership term utilized is the mean of the grades assigned by the respondent to each politician for the qualities of intelligence, strength, integrity, and empathy, rounded to the 20
These are unweighted averages. The contrast is rather greater if values for the parties are weighted by the parties’ share of the popular vote.
Timothy J. Colton
106 80
70
Percent Who Vote for the Party
60
50
40
30
20
10
0 1
2
3
4
Evaluation of Party Leader (scale 1 to 4) KPRF
Yedinstvo
OVR
SPS
Yabloko
LDPR
figure 5.3 Leadership Evaluations and the Party-List Vote, 1999.
nearest integer (with missing values of each component coded at the sample mean). On its own, this indicator makes a decent predictor of voting preference, notwithstanding possible interference from mass attitudes toward secondary leaders on the party lists.21 For example, only 2 percent of voters who gave Zyuganov the lowest score for personal qualities (one 21
Unfortunately, we did not collect the systematic information about attitudes toward leaders other than the number one person on the slate (the likes of Yurii Luzhkov, Boris Nemtsov, and Sergei Stepashin) that would have allowed us to dig into the question of their role.
Parties, Leaders, and Voters in the Parliamentary Election
107
80
Percent Who Vote for the Party
70
60
50
40
30
20
10
0 1
2 3 Evaluation of Party Leader (scale 1 to 4) CPRF
Unity
OVR
SPS
Yabloko
4
LDPR
figure 5.3 (continued).
point out of four) cast their ballots for the CPRF, whereas 72 percent of those who awarded him the highest score did so, for a gap of 70 percentage points. For Shoigu and the Unity vote, the corresponding gap is 43 percentage points, for Primakov and OVR 28 points, for Kiriyenko and SPS 30 points, for Yavlinsky and Yabloko 29 points, and for Zhirinovsky and the LDPR 54 points. Satisfying as these associations may seem, so simple a juxtaposition of several variables can hardly be the last word about the sources of voting choice. For one thing, the quantities of interest to us in this chapter, protopartisanship and leadership evaluations, are liable to overlap, in addition to being correlated with the voting outcome. Pains must therefore be taken to tease out their separate effects. For another thing, a streamlined bivariate analysis excludes the many other independent variables that we know or suspect to be associated with Russians’ voting preferences. Omitting them would be a recipe for producing biased estimates of effects on the dependent variable. What is more, the bivariate statistics, as would be the case with any single-stage analysis, leave no room for the intricacy of the causal processes involved, which the most polished contemporary works on voting treat as having their way through multiple layers or stages. Tables 5.5 and 5.6 encapsulate the results of a multistage statistical analysis of the 1999 vote for the four top-finishing Russian parties. The analysis accommodates the preceding objections by incorporating other variables that pertain to the voting decision and by doing so in stages in which the explanatory variables are introduced in blocs of causally equivalent factors. The
Timothy J. Colton
108
table 5.5 Elaboration of the Impact of Partisanship on the Party-List Vote (Differences in Predicted Probabilities)a Estimationb
CPRF
Unity
OVR
SPS
Bivariate – partisanship for given party only
.64∗∗
.56∗∗
.53∗∗
.76∗∗
Multivariate – partisanship for given party, controlling progressively for: Social characteristics .59∗∗ .51∗∗ .57∗∗ ∗∗ ∗∗ Also current conditions .57 .50 .56∗∗ ∗∗ ∗∗ Also other partisanship (but not issue opinions) .53 .42 .51∗∗ ∗∗ ∗∗ Also issue opinions (but not other partisanship) .48 .48 .58∗∗ Also issue opinions and other partisanshipc .45∗∗ .40∗∗ .52∗∗ ∗∗ ∗∗ Also retrospective evaluations of incumbents .44 .38 .52∗∗ ∗∗ ∗∗ Also party leaders .35 .32 .40∗∗ Also prospective evaluations of parties .12∗∗ .21 .20∗∗
.72∗∗ .74∗∗ .73∗∗ .67∗∗ .66∗∗ .66∗∗ .67∗∗ .72∗∗
Sample N = 1,412. Variables described in Appendix. c Gray-toned cells give the total effect of partisanship. ∗∗ p .01 Source: 1999–2000 Colton-McFaul election survey. For details, see Appendix.
a
b
table 5.6 Elaboration of the Impact of Evaluations of Party Leaders on the Party-List Vote (Differences in Predicted Probabilities)a Estimationb
CPRF
Unity
OVR
SPS
Bivariate – evaluation of given party’s leader only
.70∗∗
.41∗∗
.28∗∗
.32∗∗
Multivariate – evaluation of party leader, controlling progressively for: Social characteristics .57∗∗ .39∗∗ .38∗∗ Also current conditions .53∗∗ .40∗∗ .38∗∗ Also partisanship (but not issue opinions) .43∗∗ .49∗∗ .37∗∗ ∗∗ ∗∗ Also issue opinions (but not partisanship) .41 .40 .42∗∗ ∗∗ ∗∗ Also partisanship and issue opinions .34 .47 .39∗∗ Also retrospective evaluations of incumbents .35∗∗ .44∗∗ .38∗∗ c ∗∗ ∗∗ Also other party leaders .46 .51 .43∗∗ ∗∗ ∗∗ Also prospective evaluations of parties .11 .27 .38∗∗
.29∗∗ .29∗∗ .24∗∗ .21∗∗ .19∗∗ .19∗∗ .25∗∗ .10∗∗
Sample N = 1,412. Variables described in Appendix. c Gray-toned cells give the total effect of evaluations of party leaders. ∗∗ p .01 Source: 1999–2000 Colton-McFaul election survey. For details, see Appendix.
a
b
basic statistical estimation is done by means of multinomial logit, a maximum likelihood algorithm suitable to the analysis of a dependent variable consisting of discrete, nonordered categories (for a description of multinomial logit, see Maddala 1986, ch. 2; Long 1997, ch. 6). The parameters in the tables are
Parties, Leaders, and Voters in the Parliamentary Election
109
not the raw regression coefficients, which are of no substantive interest, but estimated first differences in the predicted probability of the respective outcomes occurring obtained, along with confidence intervals, from a statistical simulation.22 Each first difference represents the shift in the predicted probability of the citizen voting for the party list in question that is associated with a shift in the value of the independent variable from its minimum to its maximum, holding the specified control variables constant at their medians. Independent variables influence the dependent variable directly and indirectly, by transmission through other variables intermediate between them and the dependent variable. When, for any independent variable, the control variables encompass all causal variables that are causally prior and equivalent to the quantities of interest, I refer to the measure of effect (following Miller and Shanks 1996) as the “total effect” of the independent variable.23 The total effect may be thought of as “an approximation of the overall extent to which differences between voters on that variable were in fact responsible for ‘producing’ differences between them in their vote” (Miller and Shanks 1996, 196). A crucial premise of the statistical analysis is knowledge of the order in which explanatory variables enter into the causal flow culminating in the voting decision. Especially in a polity in flux, and where electoral studies are necessarily in their infancy, unprovable assumptions about causality are inescapable.24 I assume that partisanship joins the process of determining Russians’ votes at the third of the six causal stages and that evaluations of party leaders join in at the fifth stage. Other indicators, which are not of urgent interest in the present chapter (see the Appendix for a rundown), are assumed to click in at appropriate causal stages: social characteristics of the citizen at the first stage; the citizen’s perceptions of current economic and political conditions at a second stage; issue opinions simultaneous with protopartisanship at the third stage; retrospective evaluations of incumbent officeholders (president and prime minister) at the fourth stage; leadership assessments at the fifth stage; and prospective evaluations of the parties’ performance in office at the sixth stage. I thus posit that partisanship conditions leadership evaluations, but not the other way around. This is a plausible way to conceptualize the process, and one I am comfortable defending, although it cannot be validated with finality.25 22
23 24 25
The coefficients are also very sensitive to the selection of the excluded category in the regression. This problem dissolves when the analysis moves to predicted probabilities, which are constrained to sum to one, and to differences in predicted probabilities, which sum to zero. To do the simulation I employed the Tomz, Wittenberg, and King (1998) package, which is used in conjunction with the STATA package. I first adapted this approach to post-Soviet conditions in Colton (2000). Miller and Shanks used linear regression to estimate their model of voting in U.S. presidential elections. For justification of the causal sequence used here, see Colton (2000, 22–7). Most observers of Russian and post-Soviet politics would probably agree that this approach applies perfectly well to the largest party in Russia, the CPRF. Bringing it to bear on some
110
Timothy J. Colton
Tables 5.5 and 5.6 are best read row-wise, from top to bottom. The statistics in the first row summarize the bivariate relationship between the explanatory variable (partisanship in Table 5.5, leadership evaluations in Table 5.6) and the behavioral outcome (probability of voting for the party named in the column heading). The parameters in successive rows of the tables estimate the independent variable’s effect as variables in the various stages of the model are larded into the estimation one bloc at a time. The total effects are given in the gray-toned rows. Residual effects, measuring the impact of each explanatory variable that remains once all other variables in the model have been assimilated, are in the bottom row of each table. The tables record effects on the voting outcome for only those explanatory variables that directly relate to the category of the dependent variable (party voted for in December 1999) dealt with in the column. I have computed the effects of cognate variables, but leave them out of the tables for the sake of parsimony. In Table 5.5, for example, the parameters for the CPRF estimate the influence of partisanship for the CPRF (invariably a positively signed statistic) on the likelihood of voting for the CPRF. The table does not show the parameters (invariably a negatively signed statistic) estimating the influence of partisanship for any other party on the likelihood of voting for the CPRF. Likewise, the estimates in the CPRF column in Table 5.6 are for effects of evaluations of Zyuganov on the CPRF vote. The effects on the likelihood of voting CPRF of opinions of Shoigu, Primakov, and other party leaders are not reported here. A headline message of tables 5.5 and 5.6 is that we do not need to choose once and for all between partisan and personalistic explanations of the vote in Russia. In the Duma election of 1999, both nascent partisan attachments and assessments of the respective leaders of the parties materially affected the chances of voting for the candidates’ lists of all four of the largest parliamentary parties. These effects were independent, large, and statistically significant. The most compelling estimates of influence are the total effects, controlling for the confounding influences of causally antecedent or concurrent determinants. Ceteris paribus, the fact of being a partisan of one of the four main parties was enough to boost the likelihood of a citizen’s voting for that party by amounts ranging from .40, or 40 percentage points (for Unity), to .66 (for SPS). Separate from that effect, subscribing to the most glowing net evaluation of the leader of one of these parties, as compared to the most disparaging evaluation, upped the chances of voting for the party by values ranging from .25 (for SPS) to .51 (for Unity). By and large, the contributions of partisanship and leadership assessments to the party-list vote in 1999 were other parties would be more controversial. But the causal sequence cannot be adjusted ad hoc from one party to the next, as effects on the behavioral outcome must be estimated in a single master model.
Parties, Leaders, and Voters in the Parliamentary Election
111
roughly comparable in potency. For the four big parties, the unweighted means of the total effects are .51 for partisanship and .41 for leadership evaluations, meaning that partisanship had the edge over leadership.26 The outlier of the four parties is the SPS bloc, for which the estimated total effect of partisanship (.67) dwarfs an anemic effect for leadership evaluation (.25). An educated guess would be that this aberration stems from SPS’s federative structure and from the general impression, fostered by the Moscow media and by parties hostile to SPS, that Sergei Kiriyenko was a lightweight and a front man for the real movers and shakers in the bloc behind the scenes.27 Tables 5.5 and 5.6 make engrossing reading against the backdrop of the introspective accounts of our survey respondents. So far as leadership goes, objective analysis by the disinterested observer is not always congruent with the subjective self-analysis of our respondents. Citizens’ autobiographical reports did profitably alert us to the gross importance of the leadership factor. Beyond that, however, they can be quite misleading. The introspective evidence, for instance, does not really prepare us for the SPS anomaly. A more important example concerns backers of the CPRF. Responses to the question about voting the same if the party were to change its leader (Table 5.2) hinted that CPRF voters placed a lower premium on leadership than the supporters of other parties. But the statistical analysis (Table 5.6) argues otherwise. Whatever they believed about their personal motivations, or were prepared to divulge about them to an interviewer, the data suggest that supporters of the CPRF very much were instigated in forging their electoral decision by the personal qualities of Gennadii Zyuganov. Indeed, the total effect of the leadership-evaluation variable is somewhat greater for CPRF voters than the average for the electorate writ large.28 26
27
28
Were we to put partisanship at the same causal stage as leadership evaluations and redo the logit analysis, the estimated effects of partisanship on the vote would change to .34 for the CPRF (down from .45), to .32 for Unity (down from .40), and to .39 for OVR (down from .52), and would remain unchanged at .66 for SPS. The unweighted mean of the estimates for partisanship would become .43, which is still faintly larger than the average estimated effect for leadership evaluations. The most powerful figures in SPS were members of Russia’s Democratic Choice, the political party headed by Yegor Gaidar, the author of Russia’s economic shock therapy of the early 1990s. The party ran in the 1993 and 1995 Duma elections on its own, finishing second in the party-list vote to the LDPR in 1993. The exceptionally unpopular Gaidar was deliberately listed low on the SPS slate. Anatolii Chubais, an ally of Gaidar’s and the head of the national electrical utility – a formidable source of campaign funds and credit – probably had more influence over SPS than Gaidar or Kiriyenko, and he did not run at all in 1999. Kiriyenko led the SPS caucus in the Duma for only several months in 2000 before accepting an administrative appointment from President Putin. A reverse disparity concerns OVR. Only 12 percent of its voters, the lowest proportion of any party, claimed in introspective mode that they would have still supported it if its present leader (Primakov) had not been at the helm (see Table 5.2). But in the regression analysis
112
Timothy J. Colton
Finally, the analysis illuminates the ways partisanship and leadership evaluations fold into the whole multifaceted process of shaping Russians’ electoral decisions. As is the rule with this kind of analysis, the estimates of the impact of partisanship and leadership assessments mostly dwindle as other causal variables are injected into the analysis. Up to the total-effect stage, the estimates decline because the variable of interest (partisanship or leadership evaluations) to some extent transmits the influence of the variables causally upstream of it. As those variables progressively enter the estimation, the observed effect of partisanship or leadership subsides. After the total-effect stage, variables causally downstream to partisanship or leadership (and conditioned by them) enter the analysis. The upshot, once they are in the blend, is usually to depress the manifest effect of the variable of interest. Notice that for leadership evaluations (Table 5.6) total effects are actually larger than the effects given in the next cell up. This curious spike in the measure comes out of a complex interaction of factors. Attitudes toward the leaders of the remaining parties and blocs, which are taken into the reckoning in the parameter for the total effect of leadership evaluations, are negatively correlated with a decision to vote for the citizen’s party of choice but positively correlated with attitudes toward that party’s leader. Two cases – in the second and third columns of Table 5.6 – deviate sharply from the general mold. For both Unity and OVR, the effect of evaluations of the bloc’s leaders, Shoigu and Primakov, moves up rather than down as the analysis advances from the bivariate to the total-effect stage. The change is most arresting for Yevgenii Primakov of OVR. In bivariate analysis using only evaluations of the leader as the predictor, appraisals of the ex–prime minister’s personal qualities had about two-and-a-half times less influence on the likelihood of voting for OVR than appraisals of Gennadii Zyuganov’s qualities had on the likelihood of voting for the CPRF. By the total-effect stage, Primakov and OVR are in a virtual tie with Zyuganov and the CPRF – the parameter for Zyuganov having declined gradually after the bivariate stage and the parameter for Primakov having increased gradually. To some extent, Zyuganov’s personality ratings are mediators for other causal factors.29 But for Primakov, and rather less for Sergei Shoigu of Unity, citizen ratings of their personalities are more autonomous of other considerations, such as perceptions of current conditions, partisanship, and issue opinions. When the scope of the analysis widens and these other considerations are
29
the influence of leadership evaluations on the tendency to vote for OVR was lower than the analogous statistics for the CPRF and Unity. Some sense of this mediation at work can be gleaned from row-wise comparisons. The larger the decline in estimated effect between successive rows in the CPRF column of Table 5.6, the more the party-evaluation variable mediates between the factors introduced at that causal stage and voting choice.
Parties, Leaders, and Voters in the Parliamentary Election
113
pulled into it, the electoral effect of leadership evaluations is not deflated, and in fact increases somewhat.30
conclusion Analysts have frequently asserted that the development of normal political parties in the post-Soviet Russian Federation has been stunted and that, as a consequence, electoral decision making pivots on the personalistic appeal of individual leaders. The survey evidence put forth in this chapter suggests that the first claim is correct but should not be exaggerated and that the second claim is only partially true. A decade into Russia’s ragged democratization, its parties and quasiparties suffer from myriad internal shortcomings and infirmities, and are subject to myriad external constraints, that keep them from claiming the commanding position in mass politics and the governmental process that democratic theory maintains is their just deserts. As of the winter of 1999– 2000, a little over 40 percent of the Russian electorate evinced partisan sentiments, putting it in the league of Latin American countries such as Chile, Argentina, and Brazil, where electoral democracy is a recent creation, and far below the levels of party identification documented in the United States and Western Europe (Mainwaring 1999, 30–1). In a society so late to break with dictatorship, this finding should surprise no one. But it does not follow that in mass politics parties in the new Russia are completely ineffectual. On the contrary, the survey data testify that tentative attachments to parties, old and new, played a significant role in shaping the parliamentary election of December 1999. Nor should we be under the mistaken impression that leaders’ personal reputations, charismatic and noncharismatic, either dominate civic choice in Russia or satisfactorily compensate for the defects of the parties. The observed impact of leadership evaluations on the vote for the four biggest parties in 1999, while hefty, was a sliver less than the electoral impact of ` partisanship vis-a-vis those same parties. Moreover, party by party there was no demonstrable relationship between the effects of the two categories of causal variable: the magnitude of the effect of one on voting decisions implies nothing in particular about the effect of the other. There are, in sum, no grounds for alarm or for euphoria about the interplay among parties, leaders, and voters in Russia. Looking to the near horizon, it is an open question whether parties will strengthen, remedying some of the flaws instilled in them this past decade, or if mass politics will degenerate into a full-blown personalism, reducing the parties to the merest decorations or possibly dispensing with them altogether. For now, Russians 30
The same phenomenon was visible for the presidential candidacy of General Aleksandr Lebed in 1996. See Colton (2000, 197–9).
114
Timothy J. Colton
at the grass roots are at ease with the country’s mix of political styles. They will not of their own accord stampede to a radical alternative to it. The choice, however, will to some degree be removed from their hands if elites move to reengineer the terms of the electoral game. The straws in the wind today, with Boris Yeltsin in retirement and Vladimir Putin ensconced in the Kremlin, are contradictory. Some of the older parties seem poised to go out of business or to shatter into parts, while in other quarters, old and new, political entrepreneurs talk boldly of overcoming fragmentation and building commodious organizations for fighting elections, representing citizens, and implementing a legislative program. Putin, the player with the strongest cards, blows hot and cold, speaking one day of fortifying and invigorating the parties, starting with Unity, and acting like a solitary gunslinger the next. The issue will not disappear any time soon from the Russian political agenda or from the research agenda of scholars of post-Soviet affairs.
appendix The panel survey on which this paper relies was supervised by Michael McFaul of Stanford University and the author and was carried out by the Demoscope group at the Institute of Sociology of the Russian Academy of Sciences, headed by Polina Kozyreva and Mikhail Kosolapov, in association with Michael Swafford. Funding was provided by the National Science Foundation and the National Council for Eurasian and East European Research. A total of 1,919 Russian voters were interviewed in the first wave of the survey between November 13 and December 13, 1999, and 1,842 of them were reinterviewed in a second wave after the Duma election, between December 25, 1999, and January 31, 2000. This is a high reinterview rate by Western standards but is attainable in Russia with well-trained and well-motivated personnel. Respondents were selected in a multistage areaprobability sample of the voting-age population, with sampling units in thirty-two regions of the Russian Federation. Sampling procedures were similar to those given in Colton (2000, 236–8) (for analogous survey work done in 1995–6), with the one minor difference that Novosibirsk region in Siberia replaced the nearby Khanty-Mansiisk region. Interviewers were forbidden to make substitutions for designated respondents unavailable for interviews. A third wave of the survey was done in April–May 2000 to investigate the Russian presidential election. All univariate and bivariate statistics in the chapter are weighted by the number of adults in the dwelling unit, to correct for the overrepresentation of individuals in small households due to the sampling procedure. The weights are not employed in the multivariate analysis.
Parties, Leaders, and Voters in the Parliamentary Election
115
Survey Questions and Variables Considerations Important to Voting Decision. The question reads: “Various views exist about why people vote for one party or another. Using a scale where 1 means ‘completely unimportant’ and 5 means ‘very important,’ please tell me how important it was to your decision to vote for this particular party or bloc the fact that . . . (1) You approve of the work of the party or bloc in the last State Duma, (2) You like the program of this party or bloc, (3) You like the personal qualities of the leader of this party or bloc, (4) This party or bloc can defend the interests of people like you, (5) You always vote for this party or bloc.” This question was asked in the postelection interview. Would Have Voted for Party if Leader Changed. The question reads: “Would you still have voted for this party or bloc if it had been led by a different person than it is now?” Question asked in the postelection interview. Generalized Affect for Party and Leader. Respondents were told the interviewer wanted “to know what you think about a number of our political parties and blocs.” “After I read the name of a political party or bloc, please rate it on a scale from 0 to 10, where 0 means that you strongly dislike that party or bloc and 10 means that you strongly like that party or bloc. If I come to a party or bloc you haven’t heard of or you feel you do not know enough about to evaluate, just say so.” The names read out were those of the six parties that cleared the 5 percent threshold. The question was repeated for ten politicians, including the top leader of the six winning parties, President Boris Yeltsin, Prime Minister Vladimir Putin, Yurii Luzhkov, and Anatolii Chubais. These questions were asked in the postelection interview. Partisanship. Respondents were asked if there is “any one among the present parties, movements, and associations about which you would say, ‘This is my party, my movement, my association’?” Those who answered affirmatively were asked to name the party. Respondents who gave a negative response to the first question or were unable to answer were asked if there was “a party, movement, or association which more than the others reflects your interests, views, and concerns” and, if so, to name it. Partisans are individuals who answered “yes” to either question and were able to name a party or quasiparty with whom they felt the tie. This series of questions was part of the preelection interview. Personal Qualities of Leaders. Interviewers read out the statement: “When they talk about politicians, people use various words and characterizations.” For a list of eight politicians (the six main party leaders, plus Putin and Luzhkov), respondents were asked: “Would you say about [the individual] that he . . . (1) Is an intelligent and knowledgeable person, (2) Is a strong
116
Timothy J. Colton
leader, (3) Is an honest and trustworthy person, (4) Really cares about the interests of people like you.” The set responses were “yes,” “probably yes,” “probably no,” and “no.” These questions were also asked in the postelection interview. Dependent Variable in Logit Regressions Choice in the party-list vote (as determined in the postelection interview), with nonordered categories for vote for the CPRF, Unity, OVR, SPS, and for any other party or against all parties (an allowable choice in a Russian election). Independent Variables in Logit Regressions Social Characteristics. (A) Age group: six-point index (values for eighteen to twenty-nine years, thirties, forties, fifties, sixties, and seventies and older). (B) Education: six-point index (values for none or elementary education, incomplete secondary, secondary, secondary specialized, incomplete higher, higher). (C) Male sex (binary measure). (D) Former member of the CPSU (binary measure). (E) Community size: five-point measure (by quintile within the survey sample). (F) Ethnic Russian (binary measure). (G) Southern location: latitude less than 54 degrees (binary measure). (H) Eastern location: in Siberia or the Russian Far East (binary measure). (I) Good life: nine-point index of consumption (count of ownership by respondent or a member of his or her family of an automatic washing machine, a personal computer, an automobile, a dacha or second dwelling, a telephone, access to the Internet, travel abroad, travel to “a capitalist country”). (J) Gained in 1990s: “In general, did you win or lose as a result of the reforms carried out in the country in the 1990s?” Responses are that respondent lost, mostly lost, mostly won, or won, with “Won some and lost some” as a volunteered response. All this information was gathered in the preelection interview. Current Conditions. (A) Pocketbook economic welfare: change in the “material situation” of the respondent’s family over the past twelve months (values for worsened a lot, worsened a little, no change, improved a little, improved a lot). (B) State of the national economy: changes over the past twelve months (same categories as for pocketbook welfare). (C) State of democratization: “On the whole, are you fully satisfied, satisfied, dissatisfied, or completely dissatisfied with how democracy is developing in Russia?” All these questions were in the preelection interview. Partisanship. Dummy variables for partisanship for the CPRF, Unity, OVR, SPS, Yabloko, the LDPR, and any other party. Questions asked in the preelection interview (see previous).
Parties, Leaders, and Voters in the Parliamentary Election
117
Issue Opinions. (A) Left-right: self-placement on an eleven-point left-right scale. (B) Preference for political system: “What kind of political system, in your opinion, would be most appropriate for Russia? (1) The Soviet system we had in our country before perestroika, (2) The Soviet system, but in a different, more democratic form, (3) The political system that exists today, (4) Democracy of the Western type.” Both questions asked in the postelection interview. Retrospective Evaluations of Incumbents. Approval of activity of President Yeltsin and of Prime Minister Putin on a five-point scale (fully approve, approve, approve some and disapprove some, disapprove, completely disapprove). Both questions given in the postelection interview. Leadership Evaluations. Four-point index, constructed as the mean of the scores assigned to each politician in the postelection interview for intelligence, strength, integrity, and empathy (see previous), rounded to the nearest integer. The regression includes indicators for the top leader of each of the six parties to clear the 5 percent threshold. Prospective Evaluations of Parties. The introductory question reads: “Various problems have to be resolved in our country. I am going to read out a list of some of these. I would ask you please to tell me which party or bloc in Russia you think would do the best job of pursuing that objective . . . or would there be no particular difference between the parties.” The problems listed were providing social guarantees to the population, improving the Russian economy, safeguarding human rights and democratic freedoms, Chechnya, promoting Russia’s international interests, and combating crime and corruption. The regression includes scores compiled for each of the six parties, with one point awarded for each mention and with scores of three and over rounded to three. The questions were asked in the postelection interview. Missing Values Several of the tables refer to nonresponses to attitudinal questions as “don’t know” responses. These include two situations noted by the interviewer on the questionnaire form: “finds it hard to say” and, less frequently encountered, “refusal to answer.” In the multivariate analysis, I have substituted the weighted mean of the distribution for missing values, except in cases where a nonresponse can obviously be construed as a negative response (e.g., in the questions that ask about partisanship).
section iii THE ROLE OF REGIONAL POWER
6 Elections and National Integration in Russia Richard Sakwa
This chapter focuses on the spatial aspect of Russian electoral democratization. The concept of space will be interpreted broadly not only to encompass the territorial aspect but also to reflect on the way that political space is structured and represented. The central argument will be that electoralism in Russia has become one of the main devices forging a political community, and is thus at the heart of the nation-building enterprise in the new postCommunist state. This electoralism, however, is flawed and fractured, and the differentiated application of electoralism fragments the nascent political community. In this context, this chapter will focus on three central issues. First, the ethical basis for the federal separation of authority has been undermined by spontaneous processes of segmented regional devolution (sometimes called “asymmetrical federalism,” although some of what goes under this term has little to do with federalism as it is usually understood). This segmented regionalism undermined the development of a national party system, something that the authorities under President Vladimir Putin sought to overcome. Second, and related to the first, the presidential regime was much weakened in the late Yeltsin years, and the regional executives sought to fill the vacuum of central authority. The main drama of the 1999–2000 electoral cycle focused on the struggle between the center and the regions, a struggle that was over power but also about the nature of Russian federalism and of the national political community. In the event, the Kremlin was able to repulse the efforts of the regions to take over the center. However, under Putin the resurgence of the central state is torn between two forms. The first, compacted statism, uses the rhetoric of the defense of constitutional norms and the uniform application of law throughout the country but threatens the federal separation of powers. The second, pluralistic statism, defends the unimpeded flow of law and individual rights while respecting the diversity of civil society and federalist norms. Pluralistic statism challenges the segmented regionalism that had emerged under Yeltsin, and fosters the development of 121
122
Richard Sakwa
a more coherent federalized regionalism that sustains national political integration while enhancing the autonomy of the subjects of federalism. The debate over federalism and regionalism is at the same time a debate over the type of political community that will be established in Russia. Will it be one where local regimes exert considerable autonomous authority over individuals and everything else in the given region, or does the creation of a national political community take priority? If the latter, then the reassertion of central authority in the name of the unimpeded writ of the constitution and other normative acts represents not just an act of recentralization, but the defense of a particular vision of democracy. Third, electoralism in Russia has carried a double charge. On the one hand, it has allowed regional regimes to gain legitimacy through the ballot box, often through the manipulation of the electoral process in both regional and national elections; but at the same time, electoralism challenges the self-perpetuating nature of these regional regimes while, however opaquely, confirming the centrality of a national political community. Two different logics underpinned the dynamics of regional politics. One was based on electoral politics, where the regional leader sought to radiate a popular public image to gain electoral advantage. The other focused on power maximization (including the struggle for economic resources) in an intraelite struggle of the most basic traditional sort. It is the intersection between these two strategies that kept Russian regional politics away from an equilibrium point. Caught between segmented regionalism and Putin’s reconstruction of the state in its compacted form, federalism as the legal separation of powers in the spatial context was under threat; only the pluralistic reconstitution of the state could sustain the development of a democratic national political community.
forging a national political community When we think of previous experiences with the constitution of a state, two sharply divergent patterns of development characterize Western Europe. While France and Britain were established as states long before the emergence of nations, Germany and Italy existed for centuries as nations, long before nineteenth-century nationalists insisted that they were in search of states. Russia, in contrast, stands as both an uneasy state and an uncertain nation.1 Russia’s imperial evolution as a state affects the shape of contemporary nation-building efforts, while the historical Soviet multinational entity endowed Russia with a hybrid territorial and ethnic federalism. Our concern in this chapter is not on the external boundedness of Russia but rather on the development of a viable domestic political community. Just as in the European Union (EU) much discussion revolves around the need to establish 1
The imperial identity in prerevolutionary Russia has recently been explored by Hosking (1997).
Elections and National Integration in Russia
123
a “demos” so, too, in Russia in 1991 a polity was born before the emergence of a structured political community.2 Nationality and federal issues intersect and sustain, at one level, segmented regionalism, while the cultural and political resources available to the central authorities to create a national political community have been mobilized by Putin’s leadership. There is a danger, however, that the response to segmented regionalism could provoke the development of the “compacted statism” mentioned previously, where the federal separation of powers is denigrated and the institutions of a national political community, for example political parties, are imposed from above. Parties would then become not a reflection of the concerns of civil society but a department of government acting as a substitute for the political community. A democratic political community is characterized by a number of features. One of the most important is the homogenization of political space, governed by a single constitutional and legal order. This is, as it were, the political community in a passive mode. The political community in action is expressed above all by its electoral voice. National elections give shape and coherence to the political community, and are usually accompanied by political parties appealing to sections of society. Elections are the most important iterative process whereby a nation is forged into a political community, and however sectionalized or broad the appeal of particular parties, the very act of participation in the electoral process legitimizes the polity. In addition, a modern political community is characterized by the relative uniformity of administration where the state bureaucracy provides, either directly or indirectly, a range of services and public goods. In Russia, of these four – a single constitutional and legal space, elections, a national party system, and administrative uniformity – we can say that only the electoral process has to date been able to enhance the infrastructural resources of the nascent political community, and that with many caveats. Contemporary Russian politics are characterized by the tension between fragmentation and the establishment of a single coherent political community. At communism’s fall Ekiert (1992, 287–8) had warned that the establishment and consolidation of democratic orders was only one of a number of possible political outcomes. In particular, he warned of two main dangers: the “formation of new non-democratic political regimes,” not Communist but authoritarian; and of the “anarchization” of society within the framework of formally democratic and capitalist systems. In the latter, the 2
The notion of a political community entered into the discussion of constitutionalism in late eighteenth-century Germany and was used by state theorists to enhance the moral authority of the state in addition to its formal legal and procedural foundations; by the same token, it is not a term central to classical debates of Anglo-Saxon constitutionalism, which focused more on individual rights and contractarianism (Vincent 1987, 27). From our perspective, the term is useful as it suggests both a collective identity and constitutional rights. A recent discussion of the concept can be found in Schecter (2000, cha. 4).
124
Richard Sakwa
institutions of democracy and the market have only a weak binding effect on the behavior of the members of these societies (Ekiert 1992, 310). In the 1990s it appeared that Russia had simultaneously fallen prey to both of what may appear superficially to be opposed outcomes, authoritarianism and anarchization. Russia’s road to the democratization of the polity, the marketization of the economy, and the liberalization of society has indeed been uneven. As for democratization, while progress has been contradictory, it has nevertheless been relatively solid. At the institutional level, however, a multiple bifurcation of legitimization structures has taken place. At its most basic, this was reflected in the emergence under President Boris Yeltsin of a hybrid system torn between, on the one hand, the normative legal-constitutional order represented by the state and the political system (government in the abstract) and, on the other hand, the informal practices focused preeminently on the personalized practices of presidential leadership, what I have called elsewhere the regime system (Sakwa 1997; see also Klyamkin and Shevtsova 1999 [a]). The tension between system and regime is also one between formal and informal political relationships, between law and politics, and between the institutionalization and personalization of political authority. This duality was reproduced in relations between the center and the regions. At the same time, there was a tension between the legitimacy represented by central bodies (above all, the presidency and parliament), and regionally based representative and executive bodies. If under the Soviet system a “party-state” had emerged, where the Communist Party of the Soviet Union (CPSU) exercised leadership over the state and prevented the state from gaining political and institutional autonomy, then we can describe the system that emerged under Yeltsin as a “regime-state,” where the regime focused on the presidency exerted extraconstitutional authority over the political system as represented in the institutions of the state. Political practices that were once associated with “the party” were now exercised by “the regime.” The result was the continued debilitation of the state, unable to assert the principles of the constitutional autonomy of the political system vis-`a-vis the regime. This is one reason why federalism in Russia was so compromised, and allowed the development of segmented regionalism, reproducing at the local level regimes relatively unconstrained by the normative legal order of either the national or the regional state. We stress, however, that in the long run the normative resources of the constitutional state appear to be far greater than those of the personalized regime. From the very first days of his presidency Putin drew on constitutional resources to reaffirm the prerogatives of the state vis-`a-vis segmented regional regimes. We may, literally, see the reconstitution of the state. Elections in this process are a necessary but far from sufficient condition. Elections establish the political and sociological contours of the nascent political
Elections and National Integration in Russia
125
community to which, ultimately, the government and regime are accountable. At present this accountability link is tenuous in Russia and, in the case of parliamentary elections, indirect. As parliament does not form the government, Duma elections (like congressional elections in the United States) are less a mechanism to establish an oversight chamber than to fulfill certain representative and legislative functions distinct from the process of government. Presidential elections, however, are the direct source of legitimacy for presidential authority and provide the fundamental terrain for the development of regime-type politics.
electoralism, parties, and the political community Although parliamentary elections may decide little in terms of the formation of the government, they provide crucial structuration to the national political community. Russian Duma elections are perhaps less important as a mechanism of popular representation than as an instrument for the selection of political forces that will participate in elite politics. There remains a gulf between the political authorities and representatives of civil society. Russian political parties have not been able to develop the role of gatekeepers, aggregating and representing the demands of society in the decision-making bodies of the political system, for the simple reason that the political system was subordinated to informal regime-type relations. Much rule making takes place outside of the normal legislative process. Traditional corporatist patterns of the direct representation of lobbyistic concerns bypass parliaments, but in Russia the emergence of the presidentially based regime (see Chapter 9) allows policies and programs to be devised outside the framework of representative bodies (above all, parties and parliament). Modern elections everywhere are becoming more capital intensive rather than labor intensive. In Russia it is only the Communist Party of the Russian Federation (CPRF) that still relies on traditional methods of agitation, including the widespread use of leaflets, posters, stickers, and doorstep politicking, accompanied by the mass meeting. Other electoral associations rely on the media, above all its electronic forms, to put across their message (see Chapter 2). At the same time, traditional party forms of representation are overshadowed by the predominance of conglomerate pseudo parties such as Our Home is Russia (NDR), Fatherland-All Russia (OVR), and Unity. When first constituted, Unity was neither a modern political party nor a mass movement. It was perhaps the best example of a political association made to order by power elites, to act as the simulacrum of a competitive political organization and to occupy the space where political parties should belong. Alexander Veshnyakov, the chair of the Central Electoral Commission, in the wake of the December 1999 elections argued that a revised law on political parties was required to make more transparent their financing and organization, and in return only genuine political parties would be allowed
126
Richard Sakwa
to participate in parliamentary elections (Nezavisimaya gazeta, December 28, 1999, 1, 3), and a law on political parties reflecting these views was sent to the Duma by Putin in December 2000 and adopted in July 2001. The potential significance of such a reform cannot be exaggerated as in time it will eliminate the pseudoparty electoral associations. No democratic national community is complete without political parties enjoying extensive regional representation. While Yabloko had high name recognition and a leader (Grigorii Yavlinsky) known to the whole country, its attempts to establish a national party organization met with indifferent results. Only in St. Petersburg, Omsk, and Moscow did it have serious regional party organization. The CPRF, of course, as the main successor party of the old CPSU inherited much of its organizational network. The third party that deserves to be so designated is Vladimir Zhirinovsky’s Liberal-Democratic Party of Russia (LDPR) with a relatively disciplined organization and generous financial resources. Unlike a classical European-style party, the LDPR lacked a defined program; only the CPRF enjoyed the combination of a coherent program and an extensive organization. The source of Russia’s peculiar parties clearly lies in the disjunctured and fractured emergence of mass politics in Russia. Whereas parties from the late nineteenth century sought to attract each newly enfranchised segment of the population, in Russia the mass franchise predated the parties; we have an electorate that for over a decade has been in search of credible parties. Not only did this encourage the development of catch-all parties but it reinforced the tendency toward programmatic amorphousness and the politics of what Viktor Sheinis calls the “quasi-center” (Nezavisimaya gazeta, December 29, 1999, 8), a centrism based not on principle but on the politics of the marais, bending with the tides of power while seeking to exploit opportunities provided by the weakening of that power. While the age of mass parties may everywhere be in decline, the political party as a representative institution remains central to any model of democratic institutionalization. The problem, however, is that the cleavage structure (Rokkan 1997) of post-Communist Russia is highly fragmented. Cleavage structures can be divided into identity, ideological, and interest conflicts, but in Russia these do not line up and reinforce each other as they did in the heyday of industrial civilization. Identity conflicts are particularly pronounced in the ethnic republics, but they are reinforced neither by interests nor by ideology and hence have remained relatively marginal. The development of a national party structure has been inhibited by the fact that mechanisms of patronage and political clientelism have been divorced from the party system, and thus one of the main factors that promoted party development in Western Europe, the United States, and in developmental states is absent. Although the CPRF does enjoy the advantages of a national organization, it is far from penetrating the governing life of peripheral, or even “red belt,”
Elections and National Integration in Russia
127
regions. Other than in a petty way the CPRF is not the source of patronage or advancement, even where its governors identify with it. A good example of this is the way that the Communist governor of Voronezh, Ivan Shabanov, insisted that “No party will have any privileges in the administration.” He admitted that local CPRF officials had expected to fill all the key posts in the regional administration, and were disappointed when Shabanov refused to do this (EastWest Institute, Russian Regional Report [henceforth RRR], 5[9], March 9, 2000). Similarly, governor Nikolai Maksiuta of Volgograd oblast’ struggled to free himself from the demands of the regional CPRF organization after his election in December 1996, thus finding himself at odds with the Communist-dominated local legislature. Parties are out of the loop of power and patronage. The emergence of Unity as the putative presidential party had the potential to transform the Russian party scene by bringing together power and electoral maximization strategies of governing elites while provoking the countermobilization of oppositional parties on a national scale. Indeed, just six months following the 1999 elections, merger talks between Yabloko and the Union of Right Forces (SPS) were well advanced. As Putin noted at the Second Congress of Unity on May 27, 2000, “We all know that the influence of political parties on regional elections is at present very limited. Candidates strive to distance themselves from parties, apparently for good reason.”3 He also warned of the opposite danger; of candidates falling under the influence of “clans and groups.” The aim now, as a “state task in the fullest sense,” was “to move beyond pocket parties toward the creation of “parties that actively form power, advancing from their ranks outstanding leaders, outstanding people, able to create effective teams responsible to society and the state.” Whether Unity, as a “system-forming party,” as Putin put it at the founding congress of Unity on February 26–7, 2000, would be little more than a reflection of compacted statism rather than representative of civil society remained to be seen.4 Elections have become an inalienable part of political life throughout the Commonwealth of Independent States (CIS), but it is still “too early to argue that the CIS countries have been able to convert democratic elections from a political slogan into a legal institution regulating relations between the authorities (vlast’) and society” (Kurtov 2000, 9). In Russia, though, an active process of electoral iteration has been at work, with attempts to learn from the shortcomings of earlier elections. Not surprisingly, governors favored a revision of Russia’s electoral law to increase the number of deputies elected directly from local constituencies, something that would allow them greater leverage than the party-list system. In the event, the changes introduced after 1995 were modest. Russia’s dual electoral system (party-list and singlemember constituencies) relates in different ways to the demos (see Chapter 1). 3 4
See http://www.president.kremlin.ru/events/35.html. See http://www.putin2000.ru/02/11.html.
128
Richard Sakwa
The combination of what are in effect two systems means that the dynamics of electoralism have a differential impact: party-list infringements, for example, affect the whole country, whereas in the majoritarian part they affect only the electors of a given constituency. The two distinct dynamics in elections had profound institutional and behavioral consequences. Two types of electoral behavior were promoted. While the general verdict of the 1999 elections was that they were free but not fair (Sakwa 2000 [b]), in the regions electoral practices developed in a far more ambivalent way. The Russian General Procurator found that already by 1997 there were 1,471 infringements of electoral laws, a figure that by 1998 had risen to 2,760, many of them concerning the issuance of illegal normative acts in Russia’s regions (366 in 1997 and 935 in 1998) (Kurtov 2000, 12). Only a handful of cases of electoral abuse were pursued in the courts.5 The autonomy of regional executives to run their own affairs was legitimated by the fact that since late 1995 they had been elected, and thus, paradoxically, the extension of electoralism undermined the integrity of a single political community, let alone democracy, and legitimized segmented regionalism.6 Moreover, the electoral practices of the regions were homologous to those pursued by the federal authorities, and the two fed on each other. This was Yeltsin-style regime politics at its starkest and its most destructive of the establishment of the rule of law and the uniform application of the constitutional norms of the state. The emergence of a regime parasitic on the state at the center was replicated in the regions. The strong electoral mandate received by Putin in the March 2000 presidential elections substantively changed the balance of power between the federal authorities and the regions. The regions were placed on the defensive by Putin’s division of the country into seven federal districts, each supervised by a “governor-general,” and by the adoption of legislation that made it easier to dismiss governors infringing the constitution and changing the way that the Federation Council was constituted (instead of governors and the head of regional legislatures automatically becoming members, they nominated full-time representatives). The center had numerous levers at its disposal to ensure compliance to its will, if not to the normative values of 5
6
More broadly, the whole issue of variance between federal and regional normative aspects can be seen as part of the attempt to grope toward a genuine federalism, and thus “asymmetry” in this respect, as in so many others, was to be welcomed. This, at least, was the argument of Lyalya Alekhicheva, “Slovo v zashchitu regional’nogo zakonotvorchestva,” Nezavisimaya gazeta, March 2, 2000, 8. A central aspect of the ambiguities of Yeltsinism is awareness that the extension of electoralism is not necessarily synonymous with the extension of democracy. The point is reflected in Liliya Shevtsova’s (1999 [a], 55) arguments in favor of the postponement of the regional elections due in December 1991. As she puts it: “Despite the superficial non-democratic character of the decision to rely on the appointment by the president of heads of local government, it was in those circumstances justified.”
Elections and National Integration in Russia
129
the constitutional order. These range from fiscal instruments, the threat of renegotiation of the bilateral treaties, the use of the natural monopolies and Sberbank, and the appointment and removal of law enforcement officials.7 In all of this, the tension between democratic and compacted statism was evident. The electoral process is an instrument applied by the regime to influence regional affairs. Federal support for a candidate in gubernatorial elections can often be crucial. Here a whole subset of levers are available to the federal authorities: negative coverage on state television channels (ORT and RTR); delaying financial transfers, thus building up arrears of wages and pensions; cutting off energy supplies; and supplying Kremlin resources and electoral experts to other candidates. A classic case is the way that the Kremlin threw its weight behind the challenger to Vladimir Yakovlev’s incumbency in St. Petersburg. Valentina Matvienko had clearly been the Kremlin’s favored candidate in the gubernatorial contest until it became clear that she stood little chance of winning in the vote on May 14, 2000. Yakovlev had been one of Anatolii Sobchak’s deputies (the other had been Putin) until he challenged and won the mayoral post (later renamed the governorship) in 1996. In addition to this perceived disloyalty, Yakovlev had been one of the leaders of the oppositional “party of power,” OVR, in the 1999 parliamentary elections. It was clear that Putin’s Kremlin wished to punish Yakovlev for both offenses. The case also demonstrated the limits of the federal ability to influence electoral outcomes, and thus, per contra, the strength of segmented regionalism. To date there is little coherent organization of a single regional lobby; neither is there evidence of stable regional cleavages giving rise to specific political representation. The limited interregional political coordination that exists is mainly ad hoc and conjunctural. Even the obvious line between the twenty-one ethno-federal republics and the ordinary oblasts has not provoked a stable pattern of electoral or party affiliation. Kremlin politics, moreover, have operated according to classical principles of divide and rule, in the sense of offering incentives to one region while withholding them from another one, encouraging a selective pattern of engagement with the federal authorities. On budgetary matters, as Treisman has argued, it is not simply a matter of rewarding loyal regions but, on the contrary, the Yeltsin system was based on offering incentives to fractious regions to buy their acquiescence, if not loyalty (Treisman 1996). This pattern of selective incentives is reminiscent of the court politics of an earlier age where the monarch played off one boyar against another. This was regime politics at its starkest where the personalized politics at the center were replicated in the regions. A new type of regional leader emerged described as a “krepkii khozyaistvennik” (“strong manager,” 7
Russian Regional Report, vol. 5 (1), March 1, 2000.
130
Richard Sakwa
“hands-on practitioner”). Such a leader is described by Mikhailenko (1999, 82) as “in the ideational-political sense, an unprincipled conformist, authoritarian and undemocratic.” This was most dramatically in evidence in Primorskii Krai under the governorship of Yevgenii Nazdratenko where the tension between electoral and power maximization politics was reduced to a minimum. This helps explain Nazdratenko’s extraordinary political longevity until persuaded by Putin to resign on February 5, 2001. In Tatarstan the tension between electoral and power maximization politics was particularly interesting as neither predominated but each colored the other. President Shaimiev manipulated the electoral process throughout the 1990s, but he was never able entirely to subvert it, unlike some of his colleagues in other regions. In the 1999 parliamentary elections Shaimiev personally vetted the State Duma candidates on the OVR regional list, and chose who would be nominated from OVR in Tatarstan’s five single-member seats. Elections were held simultaneously to the republic’s legislative body, the State Council, and once again the “presidential list” of 130 candidates favored by the administration (the same number as seats available) received extraelectoral support, a large proportion of whom indeed were largely removed from electoral competition as they were heads of local administration appointed by the president of the republic.8 Opportunities for electoral and political countermobilization, however, remained strong, especially because the federal center could be activated as a resource in intraregional political struggles. In particular, the emergence of quasi-democratic regimes in the regions that undermined the development of a single political community provoked groups that appealed to the normative process of national state building. For example, an open letter prepared jointly by a number of oppositional groupings in Tatarstan took Putin at his word when he claimed to stand in defense of the constitution. The letter expressed concern that “Tatarstan’s constitution, laws, and political leadership violate the Russian constitution,” and warned that the residents of Tatarstan no longer felt that they were residents of Russia. In particular, the letter argued that the republican legislative branch had no authority at the local level as local legislators were controlled by mayors directly appointed by Shaimiev.9 The electoral consequences of this sort of mobilization against the local regime have yet to be seen. What is already clear, however, is that as Putin dismantled the Yeltsin-type political regime at the center, local regimes soon felt the pressure from the resurgent state. 8
9
Of the 130 favored candidates, 53 were heads of administration and 4 deputy heads, while 8 were state and trade union officials, including the republican prime minister, Rustam Minnekhanov, and the head of the presidential administration. The list also included 37 industrialists and 4 bankers. Midkhat Faroukshin, “Tatarstan President Knows Who His Deputies Will Be,” EastWest Institute, Russian Regional Report, vol. 4 (47), December 14, 1999. Russian Regional Report, vol. 5 (4), February 2, 2000.
Elections and National Integration in Russia
131
The 1999 elections served to diminish the influence of regional executives at the federal level, a trend confirmed by Putin’s rise to power. With a relatively strong propresidential party in the Duma, the regions would no longer be able to exploit tensions between parliament and the presidency. Putin’s own popularity meant that he was no longer dependent on them to “deliver” votes in the March 2000 election. A central element in the Yeltsinite “deal” between the center and the localities was the delivery of votes in elections. In the presidential election of 1996, for example, the presidents of Bashkortostan and Tatarstan had used their informal networks to shift the vote to Yeltsin in the second round.10 However, while the balance may have tilted against the regions after Putin’s rise to power, the regions were far from having become a negligible political force. Their institutionalized power remained strong in the Federation Council (although now by proxy), and no amendment to the constitution could be approved without their agreement (requiring two-thirds majorities in both houses of parliament and approval by at least two-thirds of regional legislatures). The governors had won direct representation in national politics through the creation of the State Council. The regional authorities, moreover, in the absence of strong party affiliations, could exert considerable influence over how the deputies from their regions voted in the Duma. Thus, the electoral processes at the interface of federal and regional politics represented a complex multilevel bargaining process where informal pressures were as important as official policy shifts and negotiations.
the regions attempt to storm the kremlin: regional blocs Let us now step back and examine the regional aspect of electoralism. In the 1995 Duma elections regional leaders were relatively passive, but with Yeltsin’s prolonged absences and illnesses a perceived vacuum opened up at the federal level and regional leaders sought to seize the opportunity to influence and shape power at the center. As noted, after 1995 governors were popularly elected and thus more independent than had been the case during the 1995 elections, when most were Kremlin appointees. In a paradoxical way, the weakness of “the center” and the various centrist political organizations (above all NDR) was to be compensated by renewal from the periphery. The spate of regional party formation in the run-up to the December 1999 parliamentary elections illustrated how fractured the political field was in 10
Presidents Shaimiev of Tatarstan and Murtaza Rakhimov of Bashkortostan openly boasted that “their” voters loyally voted as instructed. In 1996 in Bashkortostan Yeltsin trailed significantly behind Zyuganov in the first round (34.2 percent to 41.9 percent), while in the second round he soundly trounced Zyuganov (51.0 percent to 43.1 percent), a result that cannot be explained by normal voter “swing.” Similarly, in Tatarstan, the two candidates tied in the first round (38.3 percent to Zyuganov’s 38.1 percent), while Yeltsin shot ahead in the second round (61.5 percent to 32.3 percent).
132
Richard Sakwa
Russia, and how far Yeltsin’s own organized political base had fragmented, unable even to pretend to put forward a single quasi-presidential “party of power.” The regional executives sought to exploit the weakness of the federal government and of the relative vacuum in the sphere of party development. Regional executives believed that regional interests had been poorly represented at the federal level: what they had not been able to achieve through the party system, they now sought to do directly. Yurii Luzhkov (mayor of Moscow) was the first off the mark to establish a governors’ bloc, with the organization of his Otechestvo (Fatherland) grouping in December 1998. Luzhkov’s aim was to forge a coalition in support of his presidential bid, one that would serve to extend his support beyond the capital city. Few regions, however, were willing to join a bloc that would serve in effect to give political form to Moscow’s economic preeminence. The creation in January 1999 of Golos Rossii (Russia’s Voice) led by Konstantin Titov, governor of Samara oblast’, sought to bring together the liberal wing of Russian politics. Whatever limited support this grouping may have had largely evaporated with the establishment in April of Vsya Rossiya (AllRussia) led by Tatarstan president Mintimer Shaimiev and St. Petersburg’s governor Vladimir Yakovlev. This was a “centrist” agglomeration, but the nature and policies represented by this centrism remained unclear. Attempts to create a leftist governors’ bloc, Vozrozhdenie (Revival) and Unity, by Kemerovo governor Aman Tuleev, the head of the Agrarian Party Mikhail Lapshin and that of Popular Power, Nikolai Ryzhkov, together with the leader of the Spiritual Heritage organization, Aleksei Podberezkin, proved stillborn. Golos Rossii later became part of the SPS liberal alliance. We see reiterated in this attempted regionalization of Russian politics a second process at work that characterized party formation more generally: the creation of pseudo-party vehicles to enhance the reputation and national political visibility of their leaders. At the same time, a national persona was liable to bolster a local candidate’s chances in regional elections. We see this process adapted to the 2000 presidential electoral campaign when governors such as Tuleev of Kemerovo and Titov of Samara sought to use the national electoral process to reinforce their claims to regional and, indeed, national politics. Tuleev, who had stood down in the 1996 election in favor of communist leader Gennadii Zyuganov, now stated that he would throw his support to Putin in the second round as a man of “higher qualities” than Zyuganov, particularly in areas such as “professionalism, statehood, power and personality” (ORT, Vremya, March 5, 2000). Tuleev would have challenged Zyuganov for the leadership of the left following the presidential elections if the latter’s performance had been worse; while Titov already during the campaign called on Yabloko leader Grigorii Yavlinsky to withdraw from the race to allow the liberal vote to consolidate around himself. In both cases a long-established national elite figure was challenged by a regional leader, something that reflected the mood of the time.
Elections and National Integration in Russia
133
To return to the 1999 parliamentary elections, regional leaders sought to use their electoral blocs as ways of gaining greater influence in the State Duma, above all through the constituency seats and possibly in alliance with more established parties in the proportional part of the elections. As Tatarstan president Shaimiev put it at the founding congress of Vsya Rossiya on May 22–3, 1999 in St. Petersburg, “Our intention is to form a constructive majority in the State Duma which will stand for the interests of the regions” (Whitmore 1999). The fact that no single “party of the regions” emerged once again testified to the political and economic fissures within the regional “lobby.” The various parties of the regions, moreover, lacked a coherent policy platform other than regionalism: that is, the attempt to aggrandize the powers of subnational units (what we call segmented regionalism); whereas federalism represented the legal separation of powers between the center and the constituent parts of the federation. Fatherland under Moscow mayor Luzhkov could at least advance a consistent economic line of “competent capitalism.” The city of Moscow for most of the 1990s had been a relatively independent but basically loyal bastion of support for Yeltsin’s regime, but now it moved to a position where it sought to supplant that regime. Moscow’s shift into opposition to the Kremlin indicated the depth of the crisis facing the regime in Yeltsin’s final period. Moscow has an enormous concentration of wealth, with the city in 1998 accounting for 13 percent of Russia’s gross national product (GNP), a quarter of the incomes of all Russians, 40 percent of the federal budget, and 75 percent of the country’s financial capital. Whoever controlled Moscow could influence events in the country. Luzhkov built up an impressive war chest in the run-up to the 2000 presidential election. Even before openly declaring his candidacy in September 1998 he created an agitprop empire formally managed by a holding company called Metropolis that spanned the print and electronic media. He controlled Center Television and the newspapers Rossiya and Literaturnaya gazeta, the latter one of the most respected cultural forums in the country. By late 1998 he had succeeded in taking over the Moskovskaya pravda printworks, which printed not only Moscow’s most popular daily newspaper (Moskovskii komsomolets) but a range of other titles as well (40 journals and 128 newspapers). In not so subtle, but effective, ways Luzhkov ensured favorable coverage in his media empire, and was not averse to ensuring a rent rise or some other unpleasantness (including frequent threats to sue) for concerns that criticized him. Luzhkov’s press officials, as The Moscow Times noted, “openly divide media into two camps: those friendly to the mayor and those hostile to him” – and this particular paper fell firmly into the second category (The Moscow Times, August 14, 1998, 9). Clearly, a Luzhkov presidency would have challenged, if not threatened, freedom of speech – although whether he was any more of a danger than any other regional leader is open to debate. From a period of anarchic profusion in the very late Soviet and early democratic period, the bulk of the
134
Richard Sakwa
Russian media is now owned by five financial or industrial groups. As Anne Nivat puts it, “The problem is not the lack of plurality but of independence” (Transitions, June 1998, 47). On August 4, 1999 Fatherland formed an alliance with All-Russia to create the OVR bloc, a powerful challenge to the president, all the more so after August 17, 1999 when Evgenii Primakov, who had been foreign minister from January 1996 and the popular prime minister between September 1998 and May 1999, became their lead candidate. However, OVR was driven by divergent policy perspectives: Luzhkov and Primakov favored the recentralization of authority in Moscow, while All-Russia leader Shaimiev championed the rights of the regions and greater decentralization. The OVR bloc enjoyed the support of regional governors and basked in the reflected popularity of Primakov. Because the commanding position of OVR posed a real threat, the Kremlin in September 1999 sponsored the creation of the Unity governors’ bloc, headed by the popular Sergei Shoigu, the longtime head of the Ministry of Emergency Situations (MChS). There then followed an intense attempt by the Kremlin to undermine OVR, focusing their attacks above all on Luzhkov, whose presidential aspirations had long been advanced (Jensen 1999). While the various parties of the regions sought to gain greater influence over central politics, not all were distinctively regional parties. A study of the party lists presented by the various blocs in the 1999 elections found that the 240-long OVR list contained 110 candidates from the city of Moscow.11 Out of the crucial eighteen top names, only six were nonmetropolitan, and overall only 54.17 percent of the OVR list came from the regions. Of the other major groupings that broke through the 5 percent representation barrier in the 1999 elections Yabloko scored best in the regional representation stakes, with 133 out of 165 names listed from the regions (80.61 percent), followed closely by the SPS at 77.42 percent out of 155 candidates, and the CPRF with 197 out of 257 (76.65 percent), while Unity had 126 out of 173 names listed from the regions (72.83 percent). The Unity list was also heavily weighted toward mid-level officials with few independent political or professional resources, something that characterized the forty governors that initially supported Unity as a group. The success of Unity in the elections depressed the weight of the regions in the Duma, especially because their success, among other things, was at the expense of NDR, the party that had the highest proportion of regional representation (87.2 percent of its 211 candidates). The establishment of regional blocs represented an unaccustomed politicization of senatorial politics in the Federation Council. It suggested that existing institutions and procedures had been inadequate to represent regional interests at the federal level, revealing perhaps the inadequacies of the existing constitutional settlement, despite the emergence of tailored solutions 11
Russian Regional Report, vol. 4 (41), November 4, 1999.
Elections and National Integration in Russia
135
through the development of the ad hoc procedures of asymmetrical federalism. In particular, regional leaders complained about their minimal influence on the adoption of the federal budget, as they only became involved after the budget had passed four readings in the Duma and when nothing much could be changed. Another long-standing grievance was the organization of the road-building budget, which was concentrated in the hands of the federal authorities even when intended for the building of regional roads. More specifically, regional executives had been angered by Yeltsin’s dismissal of a string of prime ministers. This made forward planning difficult, and in general Yeltsin had resolutely excluded governors (and, of course, not only them) from the adoption of political decisions. There was, thus, a constitutional factor at work in the creation of the regional blocs – the underdevelopment of effective participatory mechanisms to reflect regional interests. There was also, of course, a more urgent instrumental logic at work, whereby the regions sought to fill the vacuum left by the dissolution of previous “parties of power.” The actual experience of these blocs, as we shall see in more detail in the following text, revealed yet a third element at work, something that we may call the boyarization principle.12 With their own budgets, mini-constitutional orders, economies, and political privileges, Russia’s governors, and even more the presidents of the republics, mimicked the state-forming process of the Russian state (Sakwa 2000 [a]). In Kalmykia, for example, President Kirsan Ilyumzhinov ruled his domain as his personal fiefdom with little regard to the laws of Russia. However, there were limits to the “medievalization” (that is, the fragmentation of economic space and political authority) of Russian politics, above all reflected in budgetary dependence on the center. Like the boyars, moreover, the regional executives were constantly locked in battle with each other and thus it proved impossible to construct a single united front. Luzhkov’s leadership ambitions, for example, had been mistrusted by other regional leaders not only because of the policies he represented but also because he was perceived as yet another centralizing Muscovite, even though temporarily at odds with the existing leadership in the Kremlin.
electoral ambiguities The underdevelopment of national political parties that could act as the focus of national integration across urban and rural, civil and military, and interest and ideological constituencies was to a degree compensated by the electoral system. The proportional part of the elections treated the whole country as one single giant constituency and thus, explicitly, as a single 12
The analogy with the boyars was drawn, among others, by Aleksandr Yusupovsky of the Federation Council’s National Security Department, EastWest Institute, Russian Regional Report, vol. 4 (38), October 14, 1999.
136
Richard Sakwa
political community. The single-mandate part of the elections, however, emphasized the fragmentation of political space. In 1993, 134 independents won seats (59.5 percent of the 225 available). While in 1995 this declined to 77 (34.2 percent), in 1999 this once again rose to 105 (46.6 percent). The high proportion of independents reflected the limited reach of national political parties, although the whole complex and multifaceted process of candidate selection needs to be taken into account (Ishiyama 1999 [b]), as does the use by individual candidates of minor electoral blocs as vehicles for election in the single-member districts (SMDs). In 1995, some sixty deputies were elected in this way, but in 1999 the smaller parties won only seventeen seats. Although some 6,000 candidates stood in the 1999 elections, clearly something was still missing as 3.6 percent of the proportional vote was cast “against all” candidates. In the single-mandate seats an average of 11.1 percent voted “against all” the 3,000-odd candidates. Only fifty candidates won more than 30 percent of the vote, while the great majority won with only 15 to 20 percent of the vote (Vestnik Tsentral’noi izbiratel’noi komissii, 1/91, 2000). In eight constituencies the elections were declared void and new elections called, something that had not occurred before (Nezavisimaya gazeta, February 8, 2000, 8). Elections were not held at all in Chechnya. One constituency is national but also increasingly regionalized – the military. The decline in central resources allocated to the military encouraged local military units to forge economic ties with regional political authorities. The latter have often made up for central budgetary shortfalls with local resources, as in the city of Severomorsk, the home of the Northern fleet. As Eva Busza has noted, these economic ties can well translate into political loyalties: “The new relationships are likely to play themselves out in local elections with military units increasing in importance as a potential voting bloc for incumbents seeking re-election” (1999, 2). If relations between the center and the regions were to take on crisis forms, this would no doubt be accompanied by a struggle for the loyalty of the armed forces. Already some of the most sensitive military forces have been withdrawn from Tatarstan, including most stocks of military mat´eriel, to avoid a repetition of the Chechen scenario, where Russian forces withdrew in late 1991 and early 1992 and left behind not only immovable items but also much mobile hardware. Nonetheless, the December 1999 parliamentary elections revealed a sea change in the orientation of the military vote toward the regime. If in the 1993 and 1995 elections it had been strongly oriented toward the opposition, in these elections the vote swung firmly toward the emerging party of power, Unity. The military vote numbers some six million if families are included, some 10 percent of the total participating electorate. Against the background of the Chechen war, some 90 percent of this electorate turned out, giving their vote primarily to the Putin-endorsed Unity (Nezavisimaya gazeta,
Elections and National Integration in Russia
137
December 20, 1999). For the first time in the post-Soviet period, the military electorate gave the plurality of its vote to the “party of power”: if one in four of the electorate at large voted for Unity, then among the military it was about half (Nezavisimaya gazeta, December 25, 1999, 3). The problem of how economic power was translated into a political resource in regional politics cannot be overemphasized here. The major oil companies are tied symbiotically to extractive regions, while Gazprom enjoys almost direct power in some gas-producing regions, and not only in them. The former governor of Krasnodar krai, Nikolai Kondratenko, for example, was known to enjoy Gazprom’s backing, a major economic force in the region, and in turn Kondratenko lobbied on behalf of Gazprom. The economic independence of Russia’s regions had increased considerably: by 1999 regional and republican governments accounted for 50 percent of tax raising and 70 percent of government spending and budget spending (Stewart 2000). Regional differentiation had increased dramatically, with great disparities in standards of living, although nominal wages did not vary greatly. Regions able to exploit raw materials (like oil- and gas-rich Tyumen) or that acted as gateways for Russia’s booming trade with the world economy (such as Moscow) thrived, while those dominated by agriculture were locked in deepest depression. Only about a dozen of Russia’s eighty-nine components were net contributors to the federal budget, although not all the others were net recipients. As noted, federal subventions were used as incentives, targeted as much toward opposition-dominated regions (to gain support) as to proreform areas. Significant turnover in regional elections suggested that Russian electoralism does have a certain bite. Out of the total of 104 elections of regional executives between 1995 and the end of 1999, 55 (52.9 percent) held on to their posts.13 If we discount the early elections, when Yeltsin appointees faced the electorate for the first time, the proportion of incumbents reelected rises sharply. The trend, therefore, was for incumbent governors to marshal their resources to greater effect in reelectoral races: in 1999 in twelve of the seventeen contests the incumbent won (71 percent); and of the forty-six elections held in 2000 to May 2001 challengers won in only eleven (76 percent reelection rate). The fact that there was some turnover indicates that all the advantages of incumbency are sometimes not enough to guarantee reelection; and what applies to the regions may well one day apply to the center as well. 13
In 1995 ten out of fourteen were reelected; in 1996 twenty-two of forty-six; in 1997 seven of seventeen; in 1998 four out of ten; and in 1999 twelve out of seventeen; see http//:www.fci.ru/elections/isp 99.htm for results up to October 1999. For an overview of the advantages of incumbency, and a different angle on the figures, see Robert Orttung,“Resourceful Governors Able to Counter the Kremlin,” EastWest Institute, Russian Regional Report, vol. 5 (10), March 15, 2000.
138
Richard Sakwa
the regions repulsed The relatively poor showing of Fatherland-All Russia (OVR) in the 1999 elections reflected the broader failure of the ambitions of regional executives to take over central politics. For most of 1999 it appeared that the collapse of central “parties of power” opened up a unique opportunity for regional alliances to influence the parliamentary elections, and indeed to control a large part of the new Duma if not to occupy the presidency. However, the emergence of Unity, bringing together at first some of the most marginal and most dependent of the regional executives with a strong political leader at the center (Shoigu and his allies in the Kremlin and the White House), upset these calculations. The OVR, representing some of the most ambitious regional leaders, appeared at first that it might become an oppositional “party of power” but was cut down to size, and the success of all other independent regional associations were limited – Fatherland failed to break out of its Moscow enclave, while Russia’s Voice atrophied in alliance with SPS. The political ambitions of the regional executives had been curbed. At the regional level, the 1999 elections demonstrated that the influence of regional leaders was very uneven. In some places (above all in Bashkortostan, Tatarstan, Ingushetiya, Mordovia, and Moscow) they clearly dominated the electoral process.14 According to the EastWest Institute, of the 152 SMDs in which it was possible to identify candidates supported by the governor, the favored candidate won in eighty-nine (59 percent).15 As they note, “There is almost a direct correlation between the success of a governor’s candidate and the strength of the governor in the region.” Other than in places where the regional executive was exceptionally strong, support for OVR tended to be unsuccessful (as in St. Petersburg, where Vladimir Yakovlev’s voice – even though he was third on OVR’s list – counted for little). The backing of Moscow oblast’ governor Anatolii Tyazhlov and Novosibirsk governor Vitalii Mukha was ineffective, and in the event they were unable even to get themselves reelected. Governors who supported Unity had a better chance of success than those who supported OVR or the Communists, suggesting that it was not the backing of governors that propelled Unity’s victory but Putin’s soaring popularity. Unity had brought into the national electoral process governors who had hitherto remained aloof, such as Alexander Rutskoi in Kursk, Aleksander Nazarov in Chukotka, Evgenii Nazdratenko in Primorskii Krai, and Kalmykia’s Kirsan Ilyumzhinov, but the election demonstrated how uneven their influence could be, with the electorate voting against establishment 14
15
In Bashkortostan, for example, OVR managed to win 39.39 percent of the partylist vote, beating the CPRF into second place (23.01 percent) and Unity into third (13.38 percent). In single-mandate seats all six went to Rakhimov-backed candidates (Tacis,“The European Union’s Project for Capacity Development in Election Monitoring,” Daily Briefing, December 21, 1999). Russian Regional Report, vol. 4 (48), December 22, 1999.
Elections and National Integration in Russia
139
candidates, for example, in Primorskii Krai.16 While power structures could influence the electoral process in many regions, they were in most cases, particularly in single-mandate seats, unable simply to dictate results. The performance of the twelve avowedly communist governors was little better. Even they in some places (e.g., Voronezh) were unable to deliver deputies in local constituencies, and in the party-list vote traditional “red belt” regions such as Kemerovo, Rostov, Volgograd, and Voronezh saw Unity surge to victory. As many commented at the time, “holes have appeared in the red belt” (e.g., Izvestiya, December 21, 1999). As noted, the lack of the support of regional executives for Yabloko clearly inhibited the party’s electoral breakout from its metropolitan heartlands. Although the elections did reveal the importance of the regions and their leaders, especially because in contrast to 1995 most were now popularly elected and thus no longer directly dependent on the Kremlin, the 1999 elections showed how this dependence had changed its forms but not its substance. The poor showing for OVR revealed the extreme heterogeneity of the interests of the governors and other figures who had made up the association in the first place, banking above all on expectations that it would become the next party of power. Success for the OVR would have meant greater sovereignty and autonomy for the republics and regions of Russia, whereas the victory of Unity represented the victory of central authority in Russia. The elections once again demonstrated that there is no single predominant cleavage of the “center” against the “regions,” but a series of crosscutting and situational alliances in a multiactor environment. The question can be formulated in terms of “collective action” problems (see Chapters 7 and 10). Following the elections the regions hurried to make peace with the acting president, fearing that the whole edifice of post-Communist federal relations was threatened by a new and more effective recentralizing trend that could jeopardize the power-sharing treaties and all the other effectively extraconstitutional privileges of the regions. Bashkortostan and Tatarstan soon faced recriminations and economic pressure, such as threats to deprive their national oil companies (Tatneft and Bashneft, respectively) of access to oil export pipelines. In response, Shaimiev complained that federal officials were engaging in “political pressure” (RFE/RL, Newsline, January 18, 2000). The reassertion of the Kremlin’s authority over governors led some to argue that “The process of regionalization has come to an end.”17 Whether this would lead to a revision of the forty-six power-sharing treaties (the one with Moscow city in July 1998 was the last of the Yeltsinite treaties) remained 16
17
In protest against the governor’s manipulation of the contest in the Vladivostok (no. 50) single-mandate constituency, which forced the deregistration of former Vladivostok mayor and Nazdratenko opponent, Viktor Cherepkov, 20 percent of the vote was cast “against all,” with the top candidate gaining only 11 percent, forcing a rerun of the election. Andrei Miroshnichenko in Russian Regional Report, vol. 5 (4), February 2, 2000.
140
Richard Sakwa
unclear. Tatarstan was particularly vulnerable to this, because its bilateral treaty of February 1994 gave it uniquely generous rights, in particular in its budgetary relations with the center. More broadly, the case demonstrated the way that a miscalculation in electoral strategy can cost a region dearly. As Midkhat Faroukshin put it, “Shaimiev’s main headache now is to figure out how to win Putin’s forgiveness for supporting OVR in the State Duma elections and secure his goodwill in the future” (RRR, 5 [3], January 26, 2000). The simplest way for regional executives to make amends for electoral waywardness was to join the Unity bandwagon, and this the great majority proceeded to do. As there was very little to choose between the regional blocs programmatically, the transfer of allegiances for most was not an intellectually traumatic one. Putin, as noted, had suggested that Unity could become the kernel of a “system-forming” political party. Whether this would be the core of a Gaullist-type Rally of the Republic party in a competitive two- or three- party system, or a party patronage network typical of the Italian era of the partitocrazia, or even of the Mexican Party of the Institutionalized Revolution (PRI) type (a form reminiscent of what we call compacted statism), remains to be seen.
conclusion: electoralism and democratic consolidation John Curtice and John Miller note that “Democracy did not come to Russia with a blank sheet, but rather was grafted onto a system of electoral administration and social mores inherited from the old regime. And old habits die hard” (1998, 141). Beginning with a system where the electoral process was primarily a means of legitimization rather than a mechanism reflecting political choice, gradually in the 1990s through repeated electoral cycles and the iteration of electoral procedures, Russian elections became more robustly democratic. Elections have been conducted with greater professionalism and there is now a greater distance between the electoral machinery and the administrative system of the presidential regime. However, while the process may be more openly democratic, it remains embedded in a society and political system that has replicated many of the practices of the Soviet past. What is radically new is the emergence of a national political community and its electoral voice as the source of legitimacy of the new order. As we have shown, however, this nascent political community is fragmented and unable to exercise adequate democratic accountability over the political regime at the center or over regional executives. The weak development of a national party system is both cause and consequence of the etiolation of mechanisms of representation and accountability. Representative institutions are unable to aggregate interests nationally and there remain fundamental disagreements over the balance to be drawn between regional devolution, the federal separation of powers and the role of the state.
Elections and National Integration in Russia
141
The creation of a number of “regional parties” in the late 1990s was an attempt to fill the vacuum in the context of a debilitated central leadership. Yet, the Duma elections of December 1999 and the presidential elections of March 2000 represented significant defeats for the regional “lobby.” In these elections the initiative passed to the federal authorities, preparing the way for a sustained assault against segmented regionalism. It remains to be seen whether the reconstitution of the state will take the form of compacted or pluralistic statism, but what is clear is that the regional factor in Russian politics remains fundamentally important, even though its political dynamics are changing. The 1999 election results showed that there could no longer be talk of a single national political culture. Instead, a rich variety of regional political cultures emerged to reflect and to promote a diverse range of regime types and electoral practices. The presidential elections demonstrated, however, the enormous desire for national integration, something that presidential systems, despite all the criticisms leveled against them in recent years, are uniquely able to do. National integration, however, is not necessarily the same thing as state reconstitution, especially because the latter in Russia is torn between two forms, compacted and pluralistic statism. Compacted statism under Putin is as much of a threat to genuine federalism as segmented regionalism was under Yeltsin. Nevertheless, the rudiments of a national political community are emerging in which electoral practices are beginning to converge, and thus the first steps toward the consolidation of democracy have been taken. Competitive elections by definition fragment political space, but the very process is a powerful force creating a political community. It allows parts of that community to speak with a common voice, and thus gradually to forge the bonds of loyalty to the polity.
7 The 1999 Duma Election in Russia: A Step Toward Democracy or the Elites’ Game? Mikhail Myagkov
The results of Russia’s 1999 parliamentary election and the composition of the new Duma seem to contradict what scholars had been claiming over the past several years (McFaul and Fish 1996; Myagkov et al. 1997): that despite all the problems and hardships that reforms brought to the Russian people, the country’s electorate remained remarkably stable. Analysis of the 1993 and 1995 Duma elections had showed that Russian voters’ behavior could be relatively well explained by a unidimensional “pro-reform/againstreform” issue space (with some intrusion of nationalism) and also that the distribution of voter preferences remained relatively unchanged over the course of at least seven years. On the surface, the outcome of the latest balloting, and the resulting distribution of Duma seats, might look like a serious blow to those theories of stability. In fact, as will be shown later in this chapter, the two major political parties that were created during the 1999 election campaign period, Fatherland-All Russia (OVR) and Unity, seem to be absolutely “issue independent” and their support cannot be explained with existing theories of vote choice for the Russian setting. Their electorate is not distributed along the “urban-rural” divide, nor do they rely on reformers or antireformers. Yet, the combined voting score of OVR and Unity was about 40 percent of the total popular vote, which is about one-half of all the ballots and about 80 percent of what the “over the 5 percent threshold parties” received in 1999 altogether. This chapter attempts to reconcile seemingly contradictory points of view about the results of the latest Duma election by focusing on the following set of questions: Why did so many people vote for the parties that were formed just shortly before the election and which did not take clear positions on the major political issues? Which side of the political spectrum is represented by the electorate of the new parties? 142
The 1999 Duma Election in Russia
143
Have Russian voters stopped voting along the proreform/antireform line? What are the possible long- and short-term political implications that arise from these results? In a nutshell, I will argue that the answers to these questions are found primarily in the behavior of the leaders of Russia’s regions (called governors) who played a critical role in the formation and in the eventual electoral support for OVR and Unity. These governors achieved the same level of success in their new enterprise as they had previously been able to enjoy in dealing with local elections, presidential elections, and referenda, thus monopolizing the last piece of Russian politics still up for grabs.
background A simplistic explanation that the Kremlin and its administrative and propagandistic resources propelled Unity to its phenomenal victory leaves important questions unanswered. If the Kremlin is so powerful, then why were they unable to prevent their main enemy, OVR, from getting almost 15 percent of the vote? Why did the previous government’s attempts to place the socalled “party of power” in the Duma fail miserably? In 1993 Yegor Gaidar’s Russia’s Choice party (RC) received only 5 percent of the popular vote and in 1995 Prime Minister Viktor Chernomyrdin’s Our Home is Russia party (NDR) received 10 percent. A look at the 1999 results from a different angle seems to make the puzzle even more complicated. The level of support (Table 7.1) for the established parties remained surprisingly level compared to 1995, thereby supporting the stability theory. Indeed, the two major Communist parties table 7.1 Comparison of 1995–1999 Voting Results for Major Electoral Groups Group
1995
1999
Left
CPRF + Communists of the USSR 26.8% Yabloko + Russia’s Choice + Ahead Russia + Common Cause 13.5% LDPR 11.1% 52%
CPRF + Communists of the USSR 26.5% Yabloko + SPS
Right
Nationalists total
14.5% LDPR 6% 47%
Source: TSIK (Central Election Commission of Russian Federation) official election results. Publications of 1996 and 2000.
144
Mikhail Myagkov
(Gennady Zuganov’s Communist Party of the Russian Federation [CPRF] and Viktor Ampilov’s Communist Party of the Soviet Union [CPSU]) received 26.5 percent in 1999 compared to 26.8 percent in 1995. Liberal reformers (Grigorii Yavlinsky’s Yabloko party and the Union of Right Forces [SPS]) scored 14.5 percent last December maintaining almost the same number as the same wing had four years ago: 13.5 percent (Yabloko, RC, Boris Fedorov’s “Ahead Russia,” and Irina Khakamada’s “Common Cause”1 ). Only the nationalist Vladimir Zhirinovsky saw his support decline from 11 percent to 6 percent, which is still enough to keep him (and his faction) among the major players for another several years. In addition to the fact that these parties kept their voters, their electoral support can also be explained by the same basic independent variables (rural/urban residence, age, education, residence in particular regions of the country) as before, and the old pro-/anti-reform issue line remains important (as I show in the following text). Therefore, if one neglects for a moment the presence of OVR and Unity, then the conclusion can be made that Russian voter preferences remained frozen for the last several years, thus making the future elections rather predictable. The (nearly) 40 percent of the popular vote that Unity and OVR received in 1999 cannot be explained by any of the standard references to basic social cleavages, such as the rural/urban divide. Moreover, the supporters of these parties can be found in the very center of the issue space between Communists and reformers. The only significant factor that does help to explain individual support for these two parties is whether or not the governor of the voter’s home region supported the party. This, combined with additional evidence, suggests that the two new parties are in fact parties of governors, and that their electoral success should be viewed as governors bringing in the vote of their own localities to their own parties. In fact, it has long been suspected that governors play a significant role in any election in Russia. Solnick (1998) shows evidence that officials at subnational levels of governments had begun to use the rules of “democratic” competition for their own purposes. For example, in many regions where local governors were running for reelection, the runoff rules were abolished. In other words, the gubernatorial elections became a single-round game instead of two rounds with a runoff. These changes had been made by regional legislatures to benefit their local “czars.” That move, ultimately, saved a number of incumbents. McFaul and Petrov (1997) observed that “strong- khozyain”2 1
2
Leaders of the Russia’s Choice, “Ahead Russia,” and “Common Cause” formed the SPS party in 1999. Thus the SPS party is considered an “old” party for the purposes of this paper analysis, as opposed to the Unity and OVR – two brand new parties of governors. The words “Strong-Khoziain” (or “Krepkii Khoziaistvennik”) are usually used to refer to a local boss who pays much attention (or at least pretends to pay attention) to such issues as road construction, building renovation, local utilities, and many others related to keeping the local infrastructure up and running. Moscow mayor Yurii Luzhkov is an example of such a
The 1999 Duma Election in Russia
145
candidates won in nine of the fourteen proreform regions and “of seven controlled regions in which incumbents run, the Ust’-Orda Buryat Okrug was the only place where the incumbent lost.” Much has also been written about the fact that it was “the weakness of central authority that gave carte blanche to regional leaders” and allowed those leaders to exert a powerful influence over what was officially reported as voting patterns of their constituencies (Lukin 1999; see also Wedel 1996 and McFaul 1997). The issue of election fraud in Russia was also linked to the motives of regional and local officials to participate in electoral manipulations (see, e.g., Myagkov and Sobyanin 1995 and Fillipov and Ordeshook 1997). Beryozkin, Myagkov, and Ordeshook (1999) further explore the issue of local elites’ influence on the election outcome. They focus on the extent to which urban-rural patterns of voting can be explained by the influence of regionally centered political elites versus the extent to which we must continue to relegate the explanation of this pattern to ideas such as “proximity to reform” or “cultural differences.” The main conclusion of their paper is that “there is another variable that needs to be considered – the influence of regional and local political bosses on officially reported election returns.” One thing that unites almost all of the previous research on the topic of local elites’ role in electoral outcomes is the focus on how the local elites use their executive and administrative power to push for a particular voting outcome on either national issues such as a presidential election or constitutional referendum or local issues such as a gubernatorial election. The Duma elections have to date been considered to be a “fair play” in the sense that neither the federal government nor the local elites would assert too much power in their efforts to reach a particular outcome. The December 1999 campaign, however, brought a whole new aspect into the Russian politics: two of the parties competing for seats in the Duma were formed (or explicitly supported) by a number of local governors. The OVR was considered by many as the Kremlin opposition’s first bid for the presidency. It was led by the leaders (“czars”) of three of the Russia’s biggest and most important regions – the city of Moscow, the city of St. Petersburg, and the Republic of Tatarstan – with a number of other significant local bosses on board. Their main opponent and the Kremlin’s political baby, Unity, had thirty-nine local governors signed on to support it. In the sections that follow, I first analyze voting results gathered at the raion (county) level to establish that neither OVR’s nor Unity’s voting patterns can be explained by the urban-rural division of Russia’s electorate, while the voting patterns of those who supported the Communists and liberal reformers can still be explained by this preexisting urban-rural division. A number of socioeconomic independent variables are also considered at the boss. McFaul and Petrov use their subjective expert evaluation to code governors according to this variable.
146
Mikhail Myagkov
regional level to determine if they can explain the electoral support of the two newcomer parties. Again, the answer is no. Third, a factor analysis of raion-level returns is performed that reveals the new spatial construction of Russia’s electorate. Unlike in previous elections, the main dimension now is “pro-/anti-Kremlin,” which puts OVR and Unity at the extremes. At the same time, the old “Communists versus liberal reformers” line becomes much less important. A dummy variable that signals whether a region’s governor supported one of the two parties turns out to be the only significant independent predictor of electoral support. This evidence for the power of governors is further supported by the peculiar (or irregular) voting patterns that have occurred in some of the “suspect”3 regions.
‘‘old’’ models of russian electoral behavior and the 1999 duma election It has long been conventional wisdom to characterize Russian voting patterns along a geographical dimension – the so-called “red belt.” The “red belt” refers to a number of regions, located in the southwest of the country, which tend to vote Communist most of the time.4 In addition, an urban-rural divide has been apparent within the Russian electorate. Clem and Craumer (1995) studied the 1993 Duma election and found that “relatively more urbanized, better educated, and younger populations tended to vote in favor of Yeltsin and the reform parties, whereas rural, agricultural areas with older populations typically aligned with the anti-reform parties or positions.” McFaul (1997) found the same patterns in the 1996 presidential contest: “the urban-rural divide helped determine the second round of the presidential vote.” Solnick (1998) noted that “at the national level, the urban vote has been consistently more pro-administration than the rural vote.” The emergence of the two new major parties naturally leads to the question of whether or not the urban-rural divide is still a factor, and if it applies to all the major parties as in the previous elections. To answer the question about the urban-rural divide, a raion-level data analysis that included 2,648 observations from eighty-eight of eighty-nine of Russia’s regions (excluding only Chechnya) was conducted according to 3
4
“Suspect” regions are ones where one of the two parties (OVR or Unity) received many more votes than the national average for the party, and the local governor was known to openly favor that party. There is no strong definition in the literature as to what the “red-belt” regions are. Most observers associate with the “red belt” such monster regions as Krasnodarskii Krai, Stavropolskii Krai, Rostov Oblast, Voronez Oblast, Lipetsk Oblast, and many other regions located in the southwest of Russia. It has been noted that the more south you go, the more Communist people vote. Various explanations of this phenomenon include demographic characteristics (people are older there), economic infrastructure (most people rely on collective farms), and political traditions (local bosses have always been strong there).
The 1999 Duma Election in Russia
147
table 7.2 OLS Estimates of the Urban-Rural Effect in Russian Electoral Behavior Party
Coefficient
Standard Error
T-Statistic
SPS Yabloko CPRF OVR Unity
0.042122 0.04058 −0.07336 −0.02917 −0.00524
0.002166 0.001305 0.006107 0.008111 0.006542
19.44989a 31.10255a −22.1126a −2.59639 −0.80033
a
Significant at the 95 percent level. Sources: TSIK (Central Election Commission of Russian Federation) official election results. Publication of 2000 and “Goskomstat” 1990 census data.
the following simple model: %P i = A + B %U i + ei where %Pi is the percentage of the vote that a particular party received in a raion i,and %Ui is level of urbanization of the same rayon. The analysis was performed separately for each of the following parties: SPS, Yabloko, CPRF, OVR, and Unity. The Ordinary Least Squares (OLS) estimates of the model’s parameters are presented in Table 7.2. The data in the table clearly indicate two things: first, support for the “old” parties (those listed in the top three rows) is significantly related to the ruralurban nature of the community, consistent with what the previous Duma elections produced. The coefficients are positive and statistically significant for the reform parties and negative (and also significant) for the CPRF. Thus, the numbers indicate that, on average, a 10 percent increase in the level of urbanization adds 8 percent to the reform vote and subtracts about 7 percent from the communist electorate. Second, the data show that the new parties’ patterns of electoral behavior are unrelated to the standard urbanrural explanation. The coefficient and t-statistic are virtually zero for Unity, and they are also very small for OVR. It is important to note that OVR would naturally have some bias toward more urban areas because two of its leaders represent the two biggest cities in Russia: Moscow and St. Petersburg. Another conventional-wisdom explanation of Russia’s past elections rests on the idea of the red belt. Is the red-belt effect apparent for the 1999 parliamentary elections? Do the new parties tend to draw more support from a particular geographic region? Answers to such questions can be obtained from the following model: %P i = A + Bx Xi + B y Y i + ei where Xi and Yi are the longitude and latitude, respectively, of a particular raion. The presence of a “red-belt” effect would make coefficients Bx and
Mikhail Myagkov
148
table 7.3 Geographic Patterns in the 1999 Duma Elections (the “Red-Belt” Effect). Regression of Party Support on Geographic Location Party
Coordinate
Coefficient
Standard Error
T-Statistic
SPS
X (longitude) Y (latitude) X (longitude) Y (latitude) X (longitude) Y (latitude) X (longitude) Y (latitude) X (longitude) Y (latitude)
0.000627 0.22629 0.00524 0.144971 −0.05086 −1.06036 −0.07053 −0.13791 0.067643 0.154293
0.002757 0.016402 0.001831 0.010896 0.006593 0.039227 0.009857 0.058654 0.00787 0.04683
0.227632 13.79624a 2.861684 13.30445a −7.71416a −27.0313a −4.1548a −2.35124 7.594654a 3.29473a
Yabloko CPRF OVR Unity a
Significant at the 95 percent level. Sources: TSIK (Central Election Commission of Russian Federation) official election results. Publication of 2000 and “Espar-Analyt” (Moscow-based political geography consulting firm) research data.
By statistically significant. Data in Table 7.3 show the OLS estimates of the preceding model’s parameters. The “Y” coordinates (latitudes) for the “old” parties are statistically significant at a very high confidence level. Therefore, it is possible to conclude that the “red-belt” effect is still present within the Communist electorate and that it harms the reform parties. On the other hand, the low significance level of “Y” coordinates for OVR and Unity suggests that the red-belt phenomenon only weakly influences their supporters’ voting patterns. It should be noted that “X” coordinates (longitudes) are significant for most of the parties, implying that a new east-west divide might be emerging in Russia’s electorate. Thus, in the aggregate at least, the conventional-wisdom variables do not adequately explain why many voters have chosen to vote for OVR or Unity. As another attempt to find explanatory variables behind the new parties’ election results, I will focus on a number of socioeconomic factors, which measure quality of life in Russia’s various regions. Such data, however, are only available on the regional level. The model is as follows: %P i = A + B H H i + B R Ri + BG Gi + B F F i + B D Di + ei , where Hi = the per capita amount of housing built in a region during the past year Ri = the per capita volume of retail sales in a region Gi = the per capita amount of unleaded gasoline sold in a region Fi = the price of a standard “food basket” in a region Di = the per capita amount of salaries debt to state employees
The 1999 Duma Election in Russia
149
table 7.4 Socioeconomic Variables and Voting Patterns in the December 1999 Election
Party
Variable
B (the Coefficient)
Standard Error of B
T-Statistic
SPS
Housing Retail Sales Gasoline Sales Food Prices Salaries Debt Housing Retail Sales Gasoline Sales Food Prices Salaries Debt Housing Retail Sales Gasoline Sales Food Prices Salaries Debt Housing Retail Sales Gasoline Sales Food Prices Salaries Debt Housing Retail Sales Gasoline Sales Food Prices Salaries Debt
−471.9 16.4 24.2 −1.1 7.0 −471.9 12.4 290.2 −0.1 0.2 1,540.3 −36.5 487.3 −3.0 8.8 1,850.8 4.6 −2,587.7 4.2 −44.1 −1,530.4 −9.9 514.5 3.9 −37.3
297.0 4.2 316.7 1.0 12.0 214.4 3.0 228.6 0.7 8.7 664.4 9.4 708.4 2.3 26.9 1,065.9 15.1 1,136.5 3.751 43.2 757.8 10.7 807.9 2.6 30.7
−1.5 3.9 0.0 −1.1 0.5 −2.2 4.0 1.2 −0.2 0.02 2.3 −3.8 0.6 −1.3 0.3 1.7 0.3 −2.2 1.1 −1.0 −2.0 −0.9 0.6 1.4 −1.2
Yabloko
CPRF
OVR
Unity
Sources: “Goskomstat” of Russia’s publication “Regions of Russia,” Moscow 1999.
The estimates presented in Table 7.4 show that, once again, these variables perform poorly in explaining support for either Unity or OVR. At the same time, the amount of per capita retail sales is correlated with all of the three “old” parties. It is not a substantial correlation, but in all three cases, the coefficients are statistically significant. The conclusion to be drawn from this analysis is that “old” approaches are less than satisfactory for explaining the voting patterns of the “new” parties. The “rural-urban divide” model shows no effect at all on voting patterns for the “new” parties, and election returns are not correlated in a systematic fashion with any of the five socioeconomic variables considered. With regard to a geographical explanation, we see an east-west divide in support for new parties, but not the more traditional north-south divide.
Mikhail Myagkov
150
To put it simply: we have yet to produce a decent explanatory model of why those Russians who decided to cast their ballots for either Unity or OVR had chosen to do so. At the same time, all the factors that worked before to explain support for the “old” parties seem to be working again in 1999. In addition, some signs of the economy’s influence on support for the old parties’ can be seen in Table 7.4. Unlike most of their predecessors, Unity and OVR are not issue-based parties in the pro-/anti-reform issue space. But what are they? Where are they positioned in Russia’s new voting environment?
the spatial character of the 1999 duma election The question of why so many people voted for Unity and OVR remains unanswered. In this section I approach the problem from a different angle – by looking at the spatial distribution of support for the major political parties in the latest Duma election, and by comparing this new spatial picture with the one obtained after the 1995 election. The methodology behind this approach is well documented. A spatial (Euclidean) construction, in which issues are portrayed as real-line intervals, candidates’ positions as points on these intervals, and voters as rational decision makers with well-defined issue preferences who vote on the basis of their distance from candidate positions, is a conventional representation for a formal mathematical analysis of elections and candidate strategies (Ordeshook 1997). We begin with a two-factor repetition of analyses previously done on data from the 1995 Duma election (Myagkov and Ordeshook 1998) . The raw data used in this factor analysis are about 2,400 raion-level aggregated election returns. Thus, the factors that we extract as a result of this procedure can be viewed as different issues that separate (or unite) various political parties. Admittedly, the interpretations of various factors are an “ad-hoc” procedure, which is mostly based on general information about these parties’ platforms. For example, if a factor puts CPRF and Russia’s Democratic Choice (DVR) on its extremes then we can assume that this is a “pro-/anti-reform” factor. Table 7.5 shows eigenvalues and percent of variance explained by each of the factors in the case of the 1995 election. table 7.5 1995 Duma Election Factor Analysis
Factor
Eigenvalue
Percent Variance
Cumulative Eigenvalue
Cumulative Variance
1 2
2.736926 1.194729
39.09894 17.06756
2.736926 3.931655
39.09894 56.1665
The 1999 Duma Election in Russia
151
0.8 Our Home is Russia
0.6
"Other" Factor
0.4
Agrarians
USSR Communists
0.2
0.0
Russia's Choice
Yabloko
-0.2 CPRF
-0.4
-0.6 -0.8
LDPR
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
Left-Right (Antireform / Proreform)
figure 7.1 1995 Duma Election.
Figure 7.1 portrays the spatial distribution of parties’ positions as produced by the factor analysis (factor loadings).5 All the “antireform” parties are clustered on the left-hand side of the figure and all the “proreform” parties are on the right-hand side. Thus, this first factor has a natural interpretation. However “the other” factor is far more difficult to interpret (see the following text). The spatial distribution of parties should not come as a surprise for anybody familiar with that election. As the figure shows, the 1995 election was largely a competition between “left” and “right.” Reformers Grigorii Yavlinsky (Yabloko) and Yegor Gaidar (RC) represented the “right,” while the Agrarians (APR), CPRF, and Ampilov (USSR Communists) were associated with the “left” end of the spectrum, with Yeltsin’s stand-in party, Chernomyrdin’s NDR, being the most centrally located reform party. On the other dimension, NDR’s position suggests that the vertical dimension has less 5
Relative values of eigenvalues show relative importance (amount of explanatory power) attributed to different factors. Absolute values of eigenvalues show how many independent variables can be “replaced” by a particular factor after the influence of more significant factors had been taken into account. “Percent of variance explained” refers to the percent of the total amount of variance of all the variables in the system that can be explained by a particular factor after the influence of more significant factors had been taken into account.
Mikhail Myagkov
152 table 7.6 1996 Duma Election
Factor
Eigenvalue
Percent Variance
Cumulative Eigenvalue
Cumulative Variance
1 2
2.454403 1.620344
40.90671 27.00573
2.454403 4.074746
40.90671 67.91244
1.0 LDPR
Unity
Anti-Pro Federal Government
0.6
0.2
Our Home is Russia CPRF
SPS Yabloko
-0.2
-0.6 OVR
-1.0 -0.8
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
Left-Right (Antireform / Proreform)
figure 7.2 1999 Duma Election.
to do with nationalism per se than with something more like the “party in power versus everyone else.” Myagkov and Ordeshook (1998) point out that “this second dimension serves essentially as a way to differentiate supporters and opponents of the current regime, which was then represented by Chernomyrdin – something like “either you believe that Chernomyrdin is the least risky alternative, or you prefer change, left or right.” This second dimension is called “other factor” in Figure 7.1. Let us turn our attention now to the 1999 Duma election. Focusing again on the major “old” parties plus the two newcomers OVR and Unity, Table 7.6 reports eigenvalues and explained variances, while Figure 7.2 gives the associated spatial positions. The almost extinct Our Home is Russia (NDR) is left in this analysis to make things consistent with 1995.
The 1999 Duma Election in Russia
153
The first thing to notice is that the second factor became much more significant in 1999 than what it was in 1995. In fact, the first factor accounts for about 40 percent of the variance, whereas the second factor in 1999 explains about 28 percent of the variance compared to only 16 percent in 1995. More interesting, though, is the spatial configurations portrayed in Figure 7.2. Here we finally see something that might shed light on the main question of the paper. The first – and the most important – dimension (it is portrayed on the vertical axis on Figure 7.2) has all the “old” parties, but the Liberal-Democratic Party of Russia (LDPR), sitting together in the center. At the same time OVR and Unity occupy opposite extremes in the bottom and on top respectively. Thus the first factor primarily differentiates among OVR, Unity, and everybody else. What could split all parties into such three groups? The answer comes naturally – their relationships with the Kremlin from support of Luzhkov’s OVR to neutrality toward all the “old” parties to love and support for the Kremlin’s own baby, Unity. The LDPR’s position on the picture comes as no surprise at all – Zhirinovsky has always been one of the power’s most vocal supporters. A consideration of the second dimension (portrayed on the horizontal axis of the Figure 7.2) reveals that parties are ranked from antireform on the left to proreform on the right. This horizontal axis represents what used to be the first, principal dimension in the “good old days” – attitudes toward reforms. But it now accounts for only 28 percent of the variance compared to 40 percent in 1995. It is interesting to notice that OVR is positioned in the very middle of the pro-/anti-reform line (horizontal axis), while Unity is leaning toward the antireform side of the spectrum. The main difference, therefore, between the 1995 and the 1999 Duma elections is that the pro-/anti-government dimension became the dominant one in 1999, while the pro-/anti-reform dimension dominated the election of 1995. Furthermore, the two new parties (Unity and OVR) led by the governors are the driving force of that change.
ovr vs. unity: anomalies within regional patterns of returns Equally interesting is the inference we can draw from Figure 7.2 as to whose preferences we see on the first dimension. Do we capture the effect of anti-/ pro-Kremlin governors delivering the vote to their parties, or do we observe voters’ attitudes, which have nothing to do with governors’ preferences? The answer to this question is the cornerstone of this chapter, and can be obtained by looking at distributions of voter preferences across regions, and by taking into account which of the regional bosses supported Unity and which supported OVR. After we have discovered that the vote for either Unity or OVR was related to voters’ attitudes toward the federal government, it remains to be
Mikhail Myagkov
154
table 7.7 Patterns of Electoral Support in Three Groups of Regions Regions whose governors belonged to:
Unity
OVR
SPS
Yabloko
CPRF
Unity OVR Unaffiliated Governors
30.36 15.9 27.56
7.15 36.9 7.30
7.65 5.33 7.90
5.82 4.36 5.17
22.6 22.6 27.3
seen if those attitudes had anything to do with the positions that particular regional bosses took on the same issue. If we find that supporters of both parties are more or less uniformly distributed across the country or if there is no correlation between a local governor’s support for one of the two parties and that party’s return in that region, then we can conclude that voters voted their genuine preferences, while if the support for a party in a region is related to the local governor’s support for the party, then we should infer that local elites can influence an election’s results. To do the analysis we split all the regions into three groups: The first group is composed of regions whose governors were members of Unity (thirty-nine). The second group contains those regions whose governors were members of OVR (seven). The final group contains all other regions. The average election returns within each of the three groups are reported in Table 7.7. The data show strong support behind the hypothesis that voting results were a subject of regional control. Within those regions whose governors supported OVR (the second group of regions) almost 40 percent of the voters supported OVR. In the other regions, only 7 percent of voters supported OVR. Unity, on the other hand, received 30 percent and 28 percent in the first and third groups of regions, implying that almost all governors would work hard for Vladimir Putin, unless they had been on OVR’s rolls. On the other hand, other parties received almost uniform support across all three regional groups, with only CPRF suffering a 5 percent loss in Unity and OVR regions. This most likely happened because some of the CPRF electorate decided to vote with the “party line” of the governor. Notably, it turned out to be a relatively small portion of all Communist voters. With these numbers in mind, we can infer that it was not genuine voter preferences, but rather the preferences of the governors and their influence that transformed the electoral construction of Russia in 1999. It is beyond the scope of this chapter to discuss how the governors were able to influence the distribution of the vote in their regions. Rather we will present two graphs just for illustrative purposes. They plot raion-level voting returns in two of the regions (the republics of Kalmikia and Tuva) whose governors belonged to Unity. The two regions were selected to illustrate the governors’ role in election outcomes because Unity scored much above average in these regions, and
The 1999 Duma Election in Russia
155
42
Unity Absolute Support
36
30
24
18
12 52
56
60
64
68
72
Actual
Turnout
Would be regular
figure 7.3 Kalmikia; Actual = −36,748 + 0.98 x + eps.
because the local governors of these regions are notorious for their ability to rule as dictators. Kirsan Ilimzhinov, the president of Kalmikia, was accused of being a godfather of an organized criminal ring involved with illegal fishing and exporting of black caviar. He later threatened to declare Kalmikia an independent state. That demand was dropped only after the Kremlin did away with a criminal investigation of Ilumzhinov. It is not hard to imagine that Ilumzhinov would try to do all he could to show his loyalty to the likely new federal boss to avoid such charges in the future. Figures 7.3 and 7.4 show the absolute support (as a percentage of all eligible voters) for Unity as a function of the turnout in a raion. In a regular case, this should be an increasing function with the slope being determined by the party’s average support. In other words, if a party’s average support is 40 percent and an extra 10 percent came to vote then we should expect that the party would get about 4 percent increase in its absolute support. The “would be regular” voting patterns (the lower lines and square data points) are presented in Figures 7.3 and 7.4 along with the actual returns for the republics of Kalmikia and Tuva. The actual voting patterns (the upper lines and round data points) show raion-level returns being around 45-degree lines, implying that every 1 percent increment in turnout results in the same 1 percent increase in the Unity absolute (with respect to all eligible voters) support.6 This means that additional voters cast their ballots for Unity. Such a voting pattern is consistent 6
The sum of all the parties’ absolute support numbers is equal to the turnout number.
Mikhail Myagkov
156 90 80
Unity Absolute Support
70 60 50 40 30 20 10 0 55
60
65
70
75
80
Turnout
85
90
95
100
90Actual 95 Would be regular
figure 7.4 Tuva; Actual = −33,507 + 1,265 x + eps.
with a conjecture that local election officials simply added extra ballots to the ballot boxes, and all these “additional” papers were marked for the Unity. Notably, these patterns predict that even if turnout numbers were equal to zero, Unity would nevertheless receive 33 percent of the “popular vote” in Tuva and 38 percent in Kalmikia. Thus, in addition to the election fraud conjecture, the two illustrations also suggest that governors may have had control over turnout numbers. The patterns uncovered in the previously mentioned two regions do not have to be present in all other “suspect” regions. There could have been different strategies that governors used to get the votes, and there could have been different mechanisms of counting and aggregating the ballots. Some of the “manipulation” strategies are briefly discussed in the concluding section.
conclusions The analyses offered here produced a number of important insights about what happened in the December 1999 Duma election and in Russian politics in general. First, the support for the “old” parties (CPRF, Yabloko, SPS) remains intact. Because electoral support for these parties in 1999 can be explained by the same models as before, we can conclude that the voters who supported the “old” parties in 1995 (and before) have not changed
The 1999 Duma Election in Russia
157
their preferences. Thus, about 40 percent of Russia’s electorate remains stable from one election to another. Second, consistent with previous elections, socioeconomic variables (measured at a regional level) do not fully explain the voting patterns of any parties. Third, the success that Unity and OVR enjoyed in the last election was largely due to the “hard work” of member governors (both OVR and Unity) and marginally due to the Kremlin’s endorsement (Unity). The spatial picture of Russian elections has changed. It is not the pro-/anti-reform issue that dominates the voting space. Instead, it is an attitude toward the federal center, which comes as a function of the local boss’s preferences, that plays a crucial role in determining the spatial location of support for a party. Finally, in the regions where the governor worked for OVR, the Kremlin could do nothing to prevent OVR from achieving a landslide victory. This is a surprising result, given the amount of propaganda and resources the federal government used in that election. Several plausible explanations may be considered as to how the governors did that. First, we can assume that election fraud did play a more significant role in Russian elections than most international observers believe. Indeed, Russia is a big country with relatively poor transportation and communications infrastructure. It would probably take an independent observer about a month of hiking to get to the raions of Tuva and Kalmikia (see Figures 7.3 and 7.4) and many other regions without the explicit consent and help from a regional boss. Add to this the observer’s chances of encountering a wild animal while on the way, and it becomes clear why we do not get many election fraud reports from Russia’s remote areas. On the other hand, regional bosses have no problems controlling these not-so-accessible places. In many of them, ruling elites have been friends of the governors for decades. Thus, a simple phone call from the governor might be enough for locals to “get the idea.” Secondly, one should not underestimate the “persuasiveness” that most of the governors possess. Despite that fact that many regions are economic and social disasters, people there tend to believe that they should do what the boss says, or things could get worse. There is a psychological theory that argues that when people are taken hostage, they start feeling sympathy toward whoever took them hostage as time passes. Most ordinary Russians are in fact “hostages” of regional elites: people cannot move, because their apartments are worthless. Pensioners depend on local governors for their pensions. Local businessmen cannot go to other regions because markets there are controlled by organized criminal groups connected to the governor, thus making it almost impossible to enter. Therefore, the governor’s control over how people vote might well be a psychological phenomenon. Third, the regional media is controlled by the governors. Televisions and newspapers are very important sources of electoral decision making in Russia because most people do not have a long history of a party affiliation. Many Russian voters believe that whatever TV says is the right thing to do. Governors not only control virtually every single local media source, but also
158
Mikhail Myagkov
they can affect the preelection campaigns through other means. For example, if a candidate wants to have a campaign event at a big factory, then the governor would decide if he (the candidate) is allowed to do so. There are other ways as well that the governors can control the election outcome, but it is far beyond the scope of this chapter to list them all. Instead, the main conclusion that can be made is that whatever power the governors had, they used it successfully. Finally, it is important to address a potential alternative explanation of the phenomenon, which might appear to be the most obvious one, especially to students of American politics. Namely, what if the governors were elected in the first place because they appealed to the electorate in particular ways and that the electorate simply reaffirmed its preferences by its choice of a particular party? The key word in the previous sentence is “preferences.” There are two possible ways of looking at this issue – preferences for an issue-based position and preferences for an individual. First, one can assume that the voters liked the position that their governor takes on various issues, and, therefore, they transpose that preference to the newly formed political party that the governor is associated with. One can compare this with an American voter who votes Democrat because he/she likes the local governor who is a Democrat as well. This hypothesis can be easily rejected by noticing that both OVR and Unity had no position on the main issue line – “pro-/ anti-reform” (see Figure 7.2) – and that both parties included governors with various political preferences and beliefs. Thus, the analysis in this chapter does not support a theory that voters vote for parties simply because they liked the political positions of their governor. The second interpretation of the word “preferences” can be a personal one. In other words, one can argue that the voters liked their governor personally and were ready to vote for whatever party he would ask them to support. This approach is consistent with the empirical results presented previously in this chapter and does not contradict our main conclusion that “governors can deliver votes, if asked for it.” The fact that the voters “liked” the governor as a person can be well explained by the same things that we used to argue how the votes could be delivered – local propaganda, personal connections, and even election fraud. It is important to underline, however, that the analysis has demonstrated that popular support did not come as a result of that governor’s position on a particular issue. A major implication for predicting the future of Russian democracy is that local governors have shown themselves to be much more powerful than they appeared before. Perhaps in future elections the Kremlin might try to ensure that only one party of governors appears on the ballot. If so, one party could potentially receive close to 50 percent of the popular vote. This would result in a “monopolistic democracy,” in which every electoral outcome would depend completely on what the president and a handful of important governors want. Such a political system would look like a real democracy
The 1999 Duma Election in Russia
159
from the outside, because the outcomes do come as a result of a popular vote. Or, worse, the Kremlin might decide that powerful governors are not exactly what they need, and may attempt to reduce their power through legislative initiatives. In that case, some strong governors might decide to “fight,” thus putting Russia’s integrity in jeopardy. Finally, it is important to note that the power the governors have is not limited to their control of national elections. They can (and do) also control their own elections; and as McFaul and Petrov (1997) note, the governors are very successful in that. Thus, the bottom line of the chapter is that voters have very little say in where Russia goes and who rules it. As privatization of the economy gave decision-making power to a dozen or so “oligarchs,” political reform resulted in about the same number of people being engaged in a game, in which the only preferences that matter are their own.
section iv PARTIES AND INSTITUTIONS
8 Instrumental Democracy: The End of Ideology and the Decline of Russian Political Parties Stephen E. Hanson1
The results of the parliamentary and presidential elections of 1999–2000 represented a severe setback to the process of formation of genuinely representative political parties in the Russian Federation (Petrov 2000). In the race for the State Duma in December 1999, the big winner was the hastily thrown together Unity party, a hodgepodge of pliable state bureaucrats and opportunist regional governors that explicitly organized its campaign around a promise to represent no ideology whatsoever.2 Capitalizing on the wardriven popularity of Prime Minister Vladimir Putin, Unity attracted nearly a quarter of the party-list vote and an additional nine seats in single-member districts (SMDs) – despite the fact that its policy positions were kept almost entirely vague before (and after!) election day. The performance of an equally opportunistic electoral alliance, the Fatherland-All Russia (OVR) bloc, was surely well below the expectations of its founders Yevgenii Primakov and Yurii Luzhkov – but OVR still took third place with 13.3 percent of the party vote and twenty-nine SMD seats in regions loyal to the party leadership, further demonstrating the electoral advantages of relying on local political machines rather than appeals to political principle. The fourth-place showing of the promarket Union of Right Forces (SPS), which received 8.5 percent of the party-list vote, represented the coordination of a larger number of liberal voters than in 1995. Yet the SPS, too, benefited greatly from its close connections with the Kremlin, which actively supported its electoral campaign. After Putin’s election as president, the party continued to function as a “party of power” with direct ties to the most important economic policy makers 1
2
The author would like to thank John Dunlop, Vicki Hesli, Ilya Prizel, William Reisinger, Michael Urban, and the anonymous reviewers of this volume for helpful comments on earlier drafts of this chapter. Thus, in rejecting a proposed alliance with the former “party of power,” Our Home Is Russia (NDR), Unity member Yevgeny Nazdratenko, the governor of Primorskii Krai, declared that “the ‘ideology’ of Unity is the absence of any ideology whatsoever” (Nezavisimaia Gazeta, October 2, 1999).
163
164
Stephen E. Hanson
in the Putin administration. Meanwhile, Grigory Yavlinsky and Vladimir Zhirinovsky – two political entrepreneurs who have consistently devoted themselves to building up genuine nationwide party organizations, albeit with diametrically opposed ideological projects – saw their organizations barely squeak past the 5 percent barrier for Duma representation; both men, and their parties, emerged severely weakened as a result. Thus, a decade after the collapse of the USSR, the Communist Party of the Russian Federation (CPRF) – as reconstituted and ideologically reoriented by its nationalist leader Gennadii Zyuganov – was still the only genuine mass party in Russia. The advantages of party organization clearly played a role in the CPRF’s rather good showing in the parliamentary contest, where it received a full 24.3 percent of the party-list vote and forty-seven district seats. Yet even this party’s status as an opposition party was murky. Signs of growing dissatisfaction among party hard-liners with Zyuganov’s leadership were obvious well before the Duma elections, as figures such as Ilyukhin, Makashov, and Podberezkin ran as representatives of splinter parties opposed to the CPRF. Then, in early 2000, the strategic alliance between the Unity party and the Communists to monopolize key committee posts in the new Duma (see Chapter 11) symbolically confirmed the CPRF’s de facto status as a proregime party – despite its continuing antiregime rhetoric. Zyuganov’s subsequent loss to Putin in the first round of the March 2000 presidential elections, by over 20 percent of the vote, further diminished his credibility as a serious opposition figure.3 Indeed, immediately following the 1999 Duma elections, the disarray of Russian political parties became almost farcical. The OVR bloc immediately lost a third of its Duma representatives to other parliamentary factions (see Chapter 11). A boycott of the Duma by Yabloko, SPS, and OVR to protest the Unity-CPRF deal ended quickly once figures from each of these movements were given a few largely symbolic Duma posts. Mayor Luzhkov publicly declared that those remaining in the Fatherland party were free to support Putin in the presidential campaign if they so chose. Other parties ranging from the SPS to the Liberal-Democratic Party of Russia (LDPR) (despite Zhirinovsky’s own entry into the campaign) joined in the chorus of support for Acting President Putin, helping to ensure his electoral victory in March.4 As election day approached, Putin announced that he saw the future of Russian politics as involving a competition among a small number of “system-forming” parties, mentioning in particular Unity and 3
4
Indeed, by the summer of 2000 erstwhile party loyalist Gennady Seleznev was actively attempting to set up a “Rossiya” movement as a “center-left” alternative to Zyuganov’s CPRF; he claimed to have already attracted 114,000 members (Radio Free Europe/Radio Liberty Newsline: Russia, July 18, 2000). Thus Zhirinovsky, in an interview with a French journalist, insisted that he was “ready to help [Putin] by joining his government” – where he claimed would play the role of Jorg Haider (Radio Free Europe/Radio Liberty Newsline: Russia, March 8, 2000).
Instrumental Democracy
165
the Communists (Rossiiskaia Gazeta, February 29, 2000) – never mind that these two groups were formally allied in the new Duma, and that the primary attraction of such a bipolar division of the Russian political spectrum for Putin was that it would ensure perpetual electoral victory for his side. As Putin began his first formal term as Russia’s president, the Russian Federation appeared to have established at best a “delegative democracy,” characterized by an “elected monarchy” and the near absence of genuinely representative political organizations that might oppose the will of the executive (Kubicek 1994; O’Donnell 1994; Shevtsova 1999 [b]). Why was the building of representative national-party organizations in Russia during the first post-Soviet decade so difficult? Mainstream political science theory is hard-pressed to explain this important feature of postcommunist Russian politics. Neo-institutionalist analysts initially expected that over time, repeated elections in a system with at least half its seats reserved for party organizations would surely generate electoral organizations and parliamentary alignments of a Western type (Fish 1995 [a]; Remington and Smith 1995). Closer scrutiny of the perverse incentives generated by Russia’s particular form of a mixed proportional representation (PR)-SMD ballot, it is true, has called into question earlier optimistic assessments of the effects of this electoral system on party formation (Smyth 1998; McAllister and White 1999). Yet the organizational weakness of Russian parties is apparent across a wide range of electoral contexts with very different institutional rules (Moser 1999 [a]). Other neo-institutional analyses blame Russia’s weak party system on its “superpresidentialism” (Linz and Stepan 1996 [a]; Chapter 9), which tends both to marginalize the legislative branch and to turn the Duma campaign into a de facto presidential primary (Chapter 10). Yet as Easter (1997) has argued, the fragmentation and dissensus of post-Soviet politics has been as much a cause as a consequence of presidentialism, which in several former Soviet republics – including the Russian Federation – was established only after the breakdown of a formally parliamentary government. Meanwhile, presidential and semipresidential systems such as those in the United States and France have proven to be quite compatible with relatively strong national parties. Given this evidence, the idea that well-organized and representative Russian political parties can eventually be generated solely through further constitutional, legal, and/or electoral reform seems hard to defend. Theorists emphasizing the impact of Russia’s political culture, too, appear to be flummoxed by the poor performance of Russian parties. For modernization theorists who considered Gorbachev’s USSR to be an emerging “civil society” of the Western sort, based upon the rapid urbanization, industrialization, and mass education implemented by the Soviet regime (Lewin 1988; Lapidus 1989; Fukuyma 1993), the absence of Western-style party organizations a decade after perestroika is a major anomaly. Public opinion polls show that Russian support for political parties is remarkably low – and
166
Stephen E. Hanson
diminishing over time (Rose, White, and McAllister, 1997; Chapter 5). Apparently, Stalin’s peculiar type of “modernization” – or “mismodernization,” as Sakwa (1996, 357) has put it – has directly hindered popular identification with representative institutions in the post-communist period. This is an outcome that calls into question the adequacy of unidirectional theories of modernization more generally. Indeed, in direct contrast to the early optimism of modernization theory, some analysts have begun to analyze Russian political culture as decisively “non-Western,” and to explain the poor quality of Russian democratic institutions as a reflection of this (Huntington 1996; Bova 1998; Eckstein 1998 [a]). But without identifying the precise mechanisms generating cultural change or continuity over long periods of historical time, this “civilizational” approach risks becoming tautological: “civic” cultures tend to generate strong political parties, so the weakness of Russian political parties must be a sign of the country’s antidemocratic cultural traditions! Thus, neither the “optimistic” analyses of Russian political culture generated by modernization theorists nor the more “pessimistic” analyses typical of those emphasizing Russia’s cultural distinctiveness appear to provide a satisfactory account of the dynamics of post-Soviet Russian-party development. In this chapter I will argue that the fragmentation of almost all new Russian political parties in the post-communist social milieu can be explained more successfully using Weberian sociological theory – an approach that combines the methodological individualism typical of most neo-institutionalists with the emphasis on ideological and ideational factors characteristic of the political culture approach.5 The absence of successful 5
An alternative explanation for the comparative weakness of Russia’s party system in the broad Weberian tradition has been advanced by Herbert Kitschelt and his collaborators, who argue that it can be explained as a result of the legacy of its particular “patrimonial” form of com´ 1999). According to this view, those munism (Kitschelt, Mansfeldova, Markowski, and Toka post-communist countries that experienced either “national-accommodative communism” (where Leninist elites made early compromises with national traditions) or “bureaucraticauthoritarian communism” (where preexisting strong state bureaucracies could be utilized to impose Leninist totalitarian rule) inherited both well-defined social cleavages and coherent bureaucratic apparati, generating a political environment relatively conducive to the project of party formation once communism collapsed. “Patrimonial” communism in the Soviet Union, by contrast, was built upon “vertical chains of personal dependence between leaders in the state and party apparatus and their entourage, buttressed by extensive patronage and clientelist networks” (Kitschelt et al. 1999, 23). The post-Soviet states (with the exception of the Baltics) have inherited an atomized social milieu still largely controlled by former members of the communist nomenklatura, posing nearly insuperable organizational obstacles to national party formation. Surely Kitschelt et al. are correct that the particularly debilitating institutional legacies of Soviet Leninism – in particular, seven decades of one-party dictatorship and six decades of inefficient industrial planning and agricultural collectivization (Jowitt 1992 [a]; Hanson 1995; Bunce 1999) – have exacerbated the difficulties of building genuine mass parties in the Russian Federation and other former Soviet republics. Yet Kitschelt’s coding of Leninist
Instrumental Democracy
167
national-party building during the first decade of democratization in the Russian Federation can be largely explained as the result of short-term instrumentally rational decision making on the part of almost all key political actors, generating a vicious cycle of uncertainty that has made it almost impossible for organizations providing public goods – political parties included – to emerge. Contrary to rational choice institutionalism, however, I argue that this environment of universal instrumental rationality is highly unusual in comparative historical perspective. Specifically, Russia’s “instrumental democracy” appears to be the product of two mutually reinforcing factors: the pervasive cultural cynicism engendered by decades of Leninism and the simultaneous discrediting of liberal, socialist, and fascist ideological alternatives in the post-Soviet context. I conclude with some speculations concerning the possible future trajectory of political-party formation in the Russian Federation under Putin.
weberian methodological individualism Of all the great social theorists of the nineteenth and early twentieth centuries, Max Weber has arguably been the most widely misunderstood. Paradigmatic debates among structural-functionalists, rational choice theorists, and neo-Marxists have generally ignored Weber’s own contributions to social science theory and methodology (Kalberg 1994). Thus it is important to set out some of the key features of Weber’s sociological theory before applying it to the post-communist Russian political system. Weber argues that the proper starting point for comparative political analysis is an understanding of the nature of domination, which he defines subtypes, which groups together such communist countries as Yugoslavia, Romania, Bulgaria, and the Soviet Union in the “patrimonial” category, appears more persuasive in retrospect than it would have during the communist period. After all, Yugoslavia was widely considered one of the most economically advanced and administratively rational countries in the bloc (Zaninovich 1968; Denitch 1976; Horvat 1976), and many leading scholars thought of the USSR as having too much state bureaucracy rather than too little (Meyer 1965). Nor would democratic theorists in the 1970s have agreed that the degree of bureaucratic rectitude in communist Czechoslovakia under Gustav Husak placed the regime in a fundamentally different sociological category than both the USSR and neighboring Central European states. Indeed, careful research has shown that the communist successor party in the Czech republic has been one of the least adaptive party organizations in the post-1989 period, precisely because of its continuing ties to the party bureaucracy of the unreformed Czechoslovak regime (GryzmalaBusse 2000). Even the emerging party system in Bulgaria, which suffered under a particularly stultifying form of “patrimonial” communism, is arguably performing much better than one might predict on the basis of Kitschelt et al.’s approach to analyzing Leninist legacies (Ganev 1997) – particularly when it is compared to the nearly nonexistent party systems of most of the former Soviet republics. Thus, while the debilitating legacy of the communist past is clearly an important factor in explaining the widespread weakness of political parties in the region, Kitschelt et al.’s framework is less successful in accounting for Russian party builders’ especially poor performance among post-communist democracies.
168
Stephen E. Hanson
simply as “the probability that certain specific commands (or all commands) will be obeyed by a given group of persons” (Weber 1968, 212). From this point of view, if subordinates are 100 percent likely to obey the commands of a superior, then domination is total; if the likelihood of obedience drops below a certain point, domination ceases to exist. In political life, “legitimate domination” refers to the subjective belief by the “staff” of the state or ruling group that those who give orders have the authority to do so. Weber argues that subjective belief in the legitimacy of commands tends to increase the probability they will be carried out (Weber 1968, 213). Note that this way of looking at domination focuses analytic attention on individuals within elite-staff relationships, rather than on the state-society relationships often studied by scholars interested in problems of legitimation. Thus, from a Weberian point of view, a state can be highly legitimate among its staff while being simultaneously hated by much of its subject population. For Weber, there are three main types of “legitimate domination” (Weber 1968, 215). Pure charismatic authority rests upon the acceptance by followers of a leader’s claim to possess miraculous, superhuman, and extraordinary powers. Charisma thus stands opposed to all “routine” bureaucratic or communal ways of life. Traditional authority rests upon the legitimacy of historical customs, familial bonds, and accepted practices. Rational-legal authority rests upon the legitimacy of impersonal rules and procedures governing individual behavior. Finally, there is the nonlegitimate type of instrumental domination in which elites sustain themselves in power through sheer coercion or material payoffs. Like charismatic authority, however, instrumental domination is fragile, because it inspires no loyalty among those who act as its agents of enforcement, and who are therefore likely to defect the moment immediate payoffs or penalties are withdrawn. Weber did not intend for these ideal types to be utilized in reified, static terms, as complete and accurate descriptions of actual political regimes. In fact, according to Weber, no pure charismatic, rational-legal, traditional, or even instrumental form of domination can ever exist. Instead, Weber sees his ideal types of collective order as built out of the aggregated social action of individuals. When the actions of many individuals are patterned along standardized lines over time and space, we can say that an organization exists. When such standardized patterns of behavior are seen as authoritative, we can say that the organizational order has become legitimate. But all of these designations are subject to change as the bases of individual action toward others in society change. To say that an organization exists, in this sense, is in effect to assert that there is a high probability that individuals within a given social location will act toward others in accordance with a particular set of formal and informal norms. If the probability of action conforming to such norms decreases below a certain point, the organization collapses.
Instrumental Democracy
169
table 8.1 Weberian Types of Social Action Degree of Deliberation
Degree of Time Salience
Low (sacred) High (profane)
Low (emotional)
High (rational)
affectual habitual
value rational instrumentally rational
In order to approach the study of organizations – and in particular states, which are the organizations that monopolize the legitimate use of physical force within a given territory (Weber 1946, 78) – we must begin by studying the bases of individual social action. Here, too, Weber outlined four basic types: instrumental-rationality, value-rationality, affect, and habit (Weber 1968, 24–5). Instrumentally rational action involves the calculation of the most effective or efficient means to obtain certain ends. Value-rational action involves engaging in behavior that is seen as an end in itself, regardless of payoffs. Affectual action is behavior motivated out of a sense of emotional loyalty or antipathy to others. Finally, habitual social action involves engaging in established ways of interacting with others that have become ingrained or unquestioned. Weber’s attempt to link these four types of individual action explicitly to the four ideal types of domination, it must be admitted, was quite unclear. To deal with this problem, my colleague Arista Cirtautas and I have developed a four-cell matrix to illustrate how one can present Weber’s four types of domination as built quite explicitly on the four types of social action (Cirtautas and Hanson, n.d.). In doing so, we have expanded somewhat upon Weber’s own theoretical presentation. We have, however, remained essentially faithful to the original tenets of Weber’s sociological method. In this matrix (Table 8.1), Weber’s four types of social action are seen as derived from the possible combinations of two variables: whether the salience of temporal context for the action is high or low, and whether the degree of deliberation associated with the action is high or low. A high degree of salience means that the motivation of action is perceived as firmly embedded in the context of concrete events connecting past, present, and future. A low degree of salience means that the action is perceived as being governed by, or oriented toward, timeless principles or loyalties. Following the work of Mircea Eliade (1959), we term action motivated by time-bound concerns as “profane” and actions where time is seemingly unimportant as “sacred.” A high degree of deliberation means that the principles of an action are evaluated, pondered, and processed at length; at some point, intense deliberation becomes simply thinking. A low degree of deliberation, by contrast, means that the motivation for action is more directly connected to impulses or
170
Stephen E. Hanson
instincts. At the risk of oversimplification, one can understand highly deliberate action as “rational” and less-highly deliberate action as “emotional.” Weber’s four types of social action can then be derived deductively from the previously mentioned axes. Along with most contemporary social scientists, Weber uses the term instrumental rationality to refer to highly deliberate, highly time-bound action. In other words, instrumentally rational action involves a careful evaluation of how to pursue one’s interests in a given time period. This is the classic form of means-ends rationality focused upon in modern economic theory and rational choice models of social organization. Instead of seeing instrumental rationality as universal or taking it as axiomatic, however, Weberian theory holds that this type of human social action exists empirically only where actors really do subjectively process information in a deliberate and goal-oriented manner. Weber terms highly time-bound action that does not involve much deliberation “traditional.” To avoid confusion with the traditional type of legitimate domination, we label this form of social action habit. Habitual action is motivated very much by day-to-day concerns, but without the actor being clearly aware of the principle according to which he or she is acting or the precise costs and benefits involved. The decision to do what one has done on a regular basis in the past does not involve much deliberation; nor is it likely to require any explicit religious or philosophical justification. In fact, as explorations of “bounded rationality” in economics and political science have shown, decision making based upon “rules of thumb” or “standard operating procedures” is empirically much more common than rational utility maximization (Simon 1982; Jones 1994, 1999). Weber’s third type of social action is value rationality, that is, highly deliberate action that does not take temporal context into account. Value rational action thus involves the deliberate decision to uphold ethical ideals or philosophic maxims that are perceived to be timelessly binding, without regard to the potential future payoffs or costs posed by the concrete social situation. This is perhaps the rarest form of individual action, but despite its infrequency, it can exert a disproportionate influence on social outcomes. Finally, action that does not involve much deliberation, but which is experienced as being out of the ordinary flow of events in time, can be understood as affectual. The classic example of affectual action is the behavior of people who are newly in love, who may sacrifice significant material interests with little if any serious deliberation in response to their sense that their personal relationship is a matter of “destiny.” Of course, affectual action can also take forms we would generally judge as negative, as in the case of action motivated by a desire for revenge or an addiction to mindaltering substances. Such behavior differs from value rational action due to the absence of any abstract ideological, philosophical, or religious principle involved, but can produce a similar feeling of connection to the realm of the “extraordinary.”
Instrumental Democracy
171
In short, then, Weber’s four types of social action can be thought of as synonymous with profane rationality (instrumental rationality), profane emotion (habit), sacred rationality (value rationality), and profane emotion (affect). Building upon this theoretical starting point, the ideal types of legitimate and illegitimate domination can be understood as involving the successful institutionalization of different combinations of individual instrumental rationality, value rationality, habit, and affect in the pursuit of political ends. Note that this procedure is precisely analogous to the search for the “microfoundations” of macropolitical phenomena within rational choice theory. The range of possible types of individual action that are seen as significant for explaining collective outcomes is simply expanded. Charismatic domination can be seen as based on the combination of value rationality and affect, the two forms of “sacred” or timeless social action. Thus, individuals involved in charismatic movements subjectively believe that the affectual bonds among members of the group are wholly consistent with quite self-conscious and explicit principles of justice, morality, or theology. At the same time, such movements must oppose both instrumental rationality, which is typically seen as introducing egoistic calculation capable of poisoning the purity of the group’s affectual bond, and habit, which is seen as eroding the clarity of deliberation necessary to grasp the group’s principles of faith. As a result of pure charisma’s basis in forms of social action for which temporal outcomes have very low salience, charismatic movements tend to be highly unstable, forced to rely on the “extraordinary” economics of charity, booty, or war. Sometimes, however, charismatic forms of domination undergo the process Weber termed “routinization” (Weber 1968, 246). From our perspective, routinization can be understood as the gradual replacement of whichever sacred category is less important in the original formation of a charismatic movement with the profane form of action diagonally opposed to it on the preceding chart. If value rationality is less important than affect, as in the case of a charismatic leader who attracts followers primarily due to the influence of her powerful personality, followers may over time begin to obey that leader (or her successors) primarily out of habit, though still preserving the affectual core of the charismatic bond (Madsen and Snow 1991). In this case, charisma has been routinized in a traditional direction. If affect is less important than value rationality, as in the case of a religious or ideological sect in which the charismatic leader’s authority is seen primarily as derived from her doctrine, followers may over time begin to pursue their mundane interests instrumentally, though still within the ethical framework mandated by that doctrine. In this case, charisma has been routinized in a rational-legal direction. The two stable forms of legitimate authority – the rational-legal and traditional types – thus tend to emerge initially out of charismatic movements. Traditional domination is based on the combination of affect and habit, and
172
Stephen E. Hanson
tends to repress or institutionally limit both value-rational and instrumentally rational social action. Rational-legal domination is based on the combination of value rationality and instrumental rationality, and tends to repress or institutionally limit all emotional forms of social action, whether affectual or habitual. Because each of these forms of domination welds together a “sacred” principle and a “profane” principle, they are capable of providing both an unchanging focus of identity and a mechanism for channeling the time-bound interests of those who find that focus of identity subjectively compelling. Rational-legal authority and traditional authority are thus relatively stable compared to the charismatic and instrumental types of domination. However, it is important to emphasize that even these more stable types of legitimate domination endure only contingently, as human beings in all societies act on diverse motivations and tend to resist strict institutional control.6 Moreover, the dynamic that produces the routinization of charisma does not end with the creation of rational-legal or traditional domination. Instead, the gradual erosion of the sacred bases of social action within any routinized context continues after stable institutions are established. Thus, the process of routinization is followed by a process of corruption, defined here as the decay of the dominant principle of legitimation – affect or value-rationality – within a routinized social order. Under traditional authority, this means the decay of the affectual bond of the staff with the leader. Under rational-legal authority, corruption involves the decay of the value-rational principle of legitimacy that generates the rules for order following in a given bureaucratic organization. Thus different forms of individual behavior are considered corrupt in different political settings; using one’s office to benefit family members, for example, is corrupt from the point of view of rational-legal norms of bureaucratic rectitude, but may be perfectly legitimate within the context of traditional authority based upon communal affective ties. Indeed, the upholding of legal procedures at the expense of personal loyalties can be considered “corrupt” by those fully enmeshed in traditional norms of legitimation. Nevertheless, corruption within both traditional and rational-legal organizations ultimately leads to the same end result: domination becomes illegitimate, or instrumental. This form of domination combines habitual and instrumentally rational bases of social action. In other words, instrumental domination occurs when an elite continues to pursue its worldly interests within roles to which they are habituated. Such an elite cannot possibly inspire any “timeless” value-rational or affectual loyalty. Therefore, 6
By no means should Weber’s ideal types of rational-legal and traditional domination be confused with Parsons’ analysis of traditional and modern social “systems,” which wrongly assumes that the legitimating principles governing elite-staff relationships reflect underlying cultural norms and values common to entire societies (Parsons 1951).
Instrumental Democracy
173
instrumental domination lasts only as long as those who might oppose it are coerced into obedience or bought off. The corruption of a routinized legitimate form of domination into purely instrumental domination, if unchecked, thus leads ultimately to the final breakdown of organizational order. Yet it is precisely in environments of large-scale organizational breakdown that new charismatic movements, which appear to reinfuse social life with sacred meaning as well as promise solutions to seemingly intractable social problems, tend to have both cultural power and material success. The cycle of charisma, routinization, and corruption thus continues indefinitely, or at least as long as human social action is governed by deliberation and time perception.
a weberian explanation for post-communist instrumentalism The form of methodological individualism previously outlined provides the basis for an interpretation of the obstacles to political party formation in post-communist Russia that recasts the central findings of neoinstitutionalist and culturalist approaches. Thus, a Weberian explanation of individual decisions not to invest in the long-term building of national political organizations overlaps substantially with a straightforward rational choice account: to do so would simply have been irrational for individual actors located in the turbulent social milieu of Russia during the 1990s (Hanson 1998). After all, genuinely representative political parties represent public goods both for their members and for those they represent (Aldrich 1995). And as we have known at least since the publication of Mancur Olson’s classic account of the problem of collective action, instrumentally rational individuals will not contribute to public goods unless there is some system of selective incentives that makes doing so individually worthwhile (Olson 1965). Given the extreme uncertainty characteristic of Russian political institutions since the collapse of the Soviet Union, and the relatively easy gains available to well-placed “free riders” within Russia’s half-reformed economy (Hellman 1998), the fact that individuals have rarely sacrificed shortrun payoffs for the potential long-run rewards of party building should not surprise us. It is perhaps worth emphasizing that nothing in Weberian theory prevents the utilization of formal models built around assumptions of instrumental rationality. On the contrary, Weber declared that “for the purposes of a typological scientific analysis it is convenient to treat all irrational, affectual elements of behavior as factors of deviation from a conceptually pure type of rational action” (Weber 1968, 6). Yet from a Weberian perspective, the fact that so much of Russian political life seems to be governed by the starkest forms of individual utility maximization is a phenomenon of considerable sociological interest. After all, the crisis-ridden environment
174
Stephen E. Hanson
of post-Soviet society is the sort that has historically generated powerful charismatic movements. Yet at the dawn of the twenty-first century, no charismatic leader plays an important role in the Russian political arena. Indeed, as we have seen, avowedly “pragmatic” politicians dominate the Russian state to an even greater degree than before. What appears to be absent in Russian politics today, in short, is any powerful value-rational principle that might motivate individual obedience despite short-run incentives to defect from the formal political organization to which one belongs. Affectual loyalties and hatreds, of course, still exist; but affect alone generates only cliques, never institutionalized parties. Those most emotionally attached to Soviet institutions and the daily life patterns associated with them have attempted to continue habitual forms of action. But as the turbulence of post-communist society made most standard operating procedures learned in the Soviet period untenable, and as life prospects became more threatened, more and more Russian decision making has been governed by simple instrumental rationality (Zisk 1997). The effects of short-term instrumental rationality in the post-Soviet context have, however, generated a whole series of negative economic and political consequences largely unanticipated by both Russian elites and their Western economic advisers. Thus a central reason for Russia’s institutional difficulties has been the high degree of uncertainty about the future that has confronted Russian elites and ordinary people alike since the breakup of the USSR. From the Gorbachev era until the end of the 1990s, Russia’s constitution, foreign policy, economic orientation, and geopolitical boundaries were almost perpetually under dispute. In the face of such high uncertainty, long-term investments of every sort, both economic and political, became instrumentally irrational. Unfortunately, a society in which individual decision making is overwhelmingly instrumental and short term in nature was bound to generate repeated institutional crises – which had the effect of making the future even more uncertain. This vicious cycle of uncertainty has had deleterious effects on both the economic and political levels. Economists have identified several economic reasons for Russia’s unexpectedly poor economic performance during the first decade after communism’s collapse: comparatively low levels of foreign and domestic direct investment, a very limited degree of enterprise and farm restructuring, asset stripping, capital flight amounting to tens of billions of dollars per year, severe problems with tax collection, and, as a result of all these factors, an increasing dependence on unsustainable short-term debt to finance basic government operations. Yet every one of these problems flows directly from the decision-making process of rational actors in a society facing extraordinarily high levels of uncertainty. The paucity of large-scale direct investment in the Russian economy during the 1990s, by foreign and domestic actors alike, is thus quite easily explained: given the political and social
Instrumental Democracy
175
turbulence of the past decade, only a fool would sink substantial resources into projects that might require years to yield positive returns. That the Soviet Union’s economic infrastructure was already crumbling and technologically outdated exacerbated this dilemma, because as a result even light industry and consumer goods production often required significant initial capital outlays that no rational actor would make. It is understandable, then, that Russian entrepreneurs in this period concentrated on sectors with low startup costs, like “shuttle trading” (buying goods abroad and selling them on the streets) and the export of raw materials through old Soviet distribution networks. As Western economists began to understand how the warped incentives of an uncertain society could undermine economic reform, increased emphasis was placed on the need for “good governance” during market transitions. Governments receiving IMF and World Bank loans were now told to observe and implement the “rule of law” in order to reduce the uncertainty facing economic actors concerning the sanctity of contracts and the inviolability of property rights. A well-functioning judiciary and stable electoral system were seen as crucial institutional foundations for this. Above all, bureaucratic corruption was somehow to be brought under control so that economic policy might be made consistent and credible. Unfortunately, such well-meaning advice also ignored the vicious cycle of uncertainty. In turbulent societies, rational actors are no more likely to invest in long-term political institution building than in long-term economic projects. While everyone would undoubtedly be better off in a society with stable legal procedures, truly representative democratic institutions, and uncorrupted public officials, no single individual can contribute anything significant to ensuring such an outcome. Indeed, in a society where everyone assumes that the government may collapse within a few years or even months, individuals who respect the rule of law inevitably lose out: they get neither the benefits of a law-governed society nor the payoffs from corruption. Under highly uncertain conditions, then, rational citizens should ignore laws when they can and organize only when directly threatened, rational judges should worry more about pleasing powerful patrons than about upholding abstract legal principles, and rational politicians should fight for control over state assets rather than pay the high costs of developing genuine national organizations to represent ordinary people. Attempts to create special government bodies to control corruption in such an environment just recreate the problem on a higher level, as every watchdog agency is always, by definition, in an even better position to extract bribes. In short, under conditions of high uncertainty, general and perpetual corruption is precisely what standard economic theory would predict. The vicious cycle of uncertainty thus explains the sad fate of various attempts to create “parties of power” in order to mobilize popular support for the Russian authorities. In each parliamentary election since the
176
Stephen E. Hanson
Soviet collapse, such parties have attracted initial pledges of support from self-interested regional politicians and state bureaucrats with a wide range of personal political views, only to disintegrate as party “activists” began to hedge their bets in the face of continuing political uncertainty. This dynamic was already evident in the first post-communist elections of December 1993, when Yegor Gadiar’s neo-liberal Russia’s Choice party self-consciously chose to position itself during the campaign as the “party of power,” filling the party’s electoral lists with high-ranking officials in the Yeltsin government and displacing many long-time “democrats” who had previously supported the president. With little reason for party loyalty in the face of such instrumental treatment, erstwhile activists from the Democratic Russia movement soon defected from the organization, joining various liberal splinter parties (McFaul 1998, 129–31). The result was an arrogant, ineffective campaign run by inexperienced state officials – and a disastrous fractionalization of the liberal electorate. By 1995, Gaidar’s party – now trying unsuccessfully to portray itself as a party of ideological principle – was unable to attain even the 5 percent of the party-list vote necessary for PR seats in the new Duma. Indeed, these second post-communist elections illustrated the consequences of instrumentalism for Russia’s party system even more starkly. By this time self-interested political entrepreneurs had discovered that running as “party leaders” on the PR ballot gave one strategic advantages, such as name recognition and increased advertising exposure, in local SMD races. This was one major reason why a remarkable forty-three separate parties were registered for the party-list ballot (Smyth 1998). Unsurprisingly, hardly any of these “parties” managed to attract committed activists – or even tried to do so. Only eight of the forty-three were on the ballot again in the 1999 campaign (see Chapter 5). Moreover, the two “parties of power” set up by President Yeltsin in 1995 – Ivan Rybkin’s bloc and the Our Home is Russia party led by Prime Minister Viktor Chernomyrdin – found themselves subject to the same dynamics of instrumental defection that had torn apart Gaidar’s party. In the case of Ivan Rybkin’s bloc, proposed by Yeltsin as a “moderate left” party to replace the hard-line CPRF, collective action by potential activists never really materialized at all; the party was dead on arrival, attaining only 1.1 percent of the party-list vote and three SMD seats. In comparison, Chernomyrdin’s party had several organizational advantages, including the fact that its leader had been the incumbent prime minister for three years, had continuing good connections at the natural gas monopoly Gazprom, and entered the relatively less-crowded “moderate right-wing” part of the political spectrum. Nonetheless, the party’s vague appeals to “pragmatism” and “centrism” attracted very few committed activists; indeed, despite the overwhelming financial support and media exposure given to NDR by the presidential administration, the party managed to attain just 10 percent of the party-list
Instrumental Democracy
177
vote and, even more strikingly, only ten seats in SMDs. Predictably, once Chernomyrdin was dismissed as prime minister in March 1998, the instrumentally rational regional governors and state officials who had been the party’s primary supporters began to turn their attention elsewhere. Like Gaidar in 1995 – but with even less credibility – new party leader Vladimir Ryzhkov tried during the 1999 campaign to reposition NDR as a party of “conservative” principle (Makarkin 1999). This strategy produced a partylist vote of 1.2 percent – almost exactly what Rybkin’s party had obtained in 1995 – along with eight SMD representatives who quickly defected to the winning Unity party in the new Duma, effectively eliminating NDR as an organization. As discussed previously, “parties of power” performed relatively well in terms of their vote share during the 1999 Duma elections. However, this in no way changed the underlying incentives for rational actors to defect from such parties once membership became inexpedient. The immediate fracturing of the instrumental OVR coalition of Primakov and Luzhkov was the most obvious evidence of this. Not only did most governors who made the “wrong bet” in the fall of 1999 and backed OVR quickly declare their support for the victorious Unity party, but even the two party leaders made increasingly public efforts to reestablish good relations with President Putin. By the time Putin was officially elected in March, it appeared highly unlikely that Fatherland would survive as an organization to compete in future elections. Meanwhile, fueled by the incentive of rational Russian politicians to be on the winning side, Unity attracted some 120,000 members by the spring of 2000 (Lambroschini 2000). But despite enormous efforts during the winter and spring of 2000 to present Unity as a key pillar of Putin’s planned new strong state – including an ostentatious and lavish party congress in May held in the same Kremlin hall that once served as the site for congresses of the Communist Party of the Soviet Union (CPSU) – the party’s dependence upon Putin’s own political standing could not be ignored. Indeed, in many regions both incumbent governors and their most dedicated political opponents declared themselves staunch Unity supporters, despite their continuing efforts to eliminate one another. In the Kirov Region, for example, the local “oligarch” Oleg Valenchuk did his best to outdo governor Vladimir Sergeenkov with professions of party loyalty: On the outside wall of the Central Hotel, [Valenchuk] has strung a huge banner emblazoned with a big black bear, the logo of Unity. . . . He has pasted a smaller bear on the main door through which tax police pass on their way to scrutinize his books. “They should know that I’m the one who really supports Putin, not the governor,” he says (Wall Street Journal, July 10, 2000).
The long-term solidity of such “support” for Putin’s new party, however, could obviously be questioned. Even Putin seemed quite aware of the
178
Stephen E. Hanson
long-term fragility of artificial parties of power. As he declared in his state of the union address in July 2000: Russia needs parties that enjoy mass-scale support and stable authority. Russia does not need another bureaucrats’ party that sucks up to the authorities, and even more so, tries to replace the authorities. Experience has shown – we know this from the past several years – that such formations die instantly like hot-house plants as they find themselves out in the cold, among other competitors (BBC Monitoring, July 8, 2000).
Yet it was hard to see on what basis Putin could prevent Unity from suffering precisely this same fate, were the future of his own administration to become uncertain or his personal patronage of the party to be withdrawn.7
the end of ideology in post-communist russia The analysis thus far logically leads to the critical question: where do long time horizons come from? Under what social circumstances do vicious cycles of uncertainty end, and individual actors begin to make investments in long-term payoffs – whether economic or political? This is where Weberian and rational choice versions of methodological individualism most clearly diverge. Most neoclassical economic models deal with the problem of time horizons by assuming actors have identical “discount rates,” that is, the rates at which they are willing to forgo present payoffs for future payoffs. A very high discount rate implies a focus on the short run; a low discount rate implies a higher willingness to risk present assets for potentially greater future rewards. Sophisticated analysis has demonstrated that the provision of public goods can be instrumentally rational when a sufficiently large number of individual actors have sufficiently long time horizons (Taylor 1987). In an environment of high social uncertainty, however, where most people think only about the near-term future, adopting a low discount rate would seem to be individually irrational. Indeed, extant rational choice theory has little to say about just why actors have the discount rates they do. Generally, due to the mathematical complexities involved in allowing discount rates to vary, econometric modelers prefer to assume that all actors have the same exogenously determined time horizons. The Weberian approach set out in this chapter implies a different solution to the problem: namely, that low “discount rates” in turbulent environments are typically generated by individual action in conformity to value-rational principles. Under conditions of high uncertainty, those acting out of habit 7
A hint of possible future struggles in this regard was provided when prominent “oligarch” Boris Berezovsky responded to Putin’s consolidation of power by announcing in the summer of 2000 his intention to form yet another new party in collaboration with prominent regional politicians – with the explicit intention of taking members away from Unity and factions allied with it in the Duma (Kommersant, July 8, 2000).
Instrumental Democracy
179
discover that rapidly changing circumstances constantly undermine old rules of thumb and uproot once-predictable patterns of daily life. Those acting out of affectual ties or hatreds may generally promote family and friends or struggle against sworn enemies – but affect alone cannot provide any consistent guide to dealing with strangers in an unpredictable national political economy. Those attempting to be instrumental, as we have seen, will rationally defect from all proposals for long-run public goods provision as long as the future remains uncertain. In turbulent societies, then, only those who act strictly in accordance with abstract principles considered to be timelessly binding will behave consistently over the long run. This helps to explain why ideologues have a strategic advantage over pragmatic party builders in uncertain social environments (Hanson 1998). Ideology can be understood as a specific type of value-rationality that explicitly sets out the “timeless” criteria held to define the proper boundaries and membership of a political community. Politicians who propose new ideologies in turbulent societies, then, are in effect attempting artificially to elongate the time horizons of potential ideological converts, by convincing them that history is on their side. Ideological conversion immediately provides a designated group of “trustworthy” individuals with whom it is one’s duty to cooperate. For individuals with few affectual ties and disrupted daily habits, it may even become instrumentally rational to “gamble” on conversion to an ideologue’s proposed definition of political membership, because in a social environment where few institutional paths for improving one’s life chances are reliable, charismatic groups may be uniquely capable of providing public goods to their memberships. Such dynamics help to explain the relative early success in national party building of the explicitly antiliberal ideologues Zhirinovsky and Zyuganov relative to their more pragmatic competitors, despite the fact that the latter often possessed far greater state support and financial backing. While both men certainly took advantage of the resources made available to them through connections with the security services (in Zhirinovsky’s case) and the former CPSU nomenklatura (in Zyuganov’s case), it is easy to forget in retrospect just how marginal these two politicians were in the early post-Soviet days of mass enthusiasm for Yeltsin and the “democrats.” Even in the opposition to Yeltsin within the Congress of People’s Deputies during 1992–3, neither Zhirinovsky nor Zyuganov was very influential. Yet by 1993, both had forged relatively powerful nationwide party organizations – a factor that played an important role in the LDPR’s and CPRF’s electoral performance both in 1993 and in 1995. Zhirinovsky’s vocal advocacy of a new “superimperial” Russia that would dominate the threatening, unruly Turkic and Muslim “South” and eliminate all “special” rights for non-Russian minorities, and Zyuganov’s equally explicit and consistent calls to rebuild the Soviet state within the territory of the former USSR in order to reclaim Russia’s historical greatness and to prevent the triumph of global
180
Stephen E. Hanson
“cosmopolitan liberalism,” appeared to demonstrate to their early supporters that these politicians were oriented toward value-rational principles that would not shift with the changing political winds (Hanson and Williams 1999). In particular, the fact that both men had adopted these extreme viewpoints prior to the collapse of the USSR made it hard to accuse them, at least initially, of simple opportunism. However, as the Russian Federation enters the Putin era, there is no denying that the original charismatic appeal of the antisystem ideologies articulated by Zhirinovsky and Zyuganov during the period of the Soviet collapse has rather rapidly routinized and corrupted. To be sure, the rhetoric of both men remains remarkably consistent with that characteristic of their earlier political careers. Zhirinovsky’s speeches and writings during the 1999 Duma campaign (1999 [a], 1999 [b]) continued to focus on supposed threats to destroy Russia emanating from the West and the South, on the LDPR’s opposition to communism and liberalism alike, and on the need to revamp Russian federalism to eliminate the special status of ethnic republics. Zyuganov (1999 [b]), meanwhile, continued to maintain that the CPRF was the only genuine opposition party in Russia willing to stand up to the threat of Westernization. Yet neither Zhirinovsky nor Zyuganov was willing to go beyond the articulation of antisystemic visions to advocacy of genuinely revolutionary action; indeed, both men claimed to reject the ideological “extremism” they ascribed to their political adversaries. By the end of the 1990s, the failure of either antisystem party ever to act resolutely against the system – combined with their particularly slavish initial devotion to the Putin regime – had all but eliminated Zhirinovsky’s and Zyuganov’s chances to lead a genuine revolt against the formal constitutional framework established under Yeltsin. Perhaps more remarkably, since the collapse of the Soviet Union in 1991 not a single Russian political entrepreneur has managed to articulate a persuasive new ideology around which a powerful new party organization might crystallize – despite the increasing sense among Russian elites that the absence of a consensual state ideology was debilitating for the performance of state institutions (Breslauer and Dale 1997). Well-publicized attempts by various advisers to Yeltsin to discover a new “national idea” by means of a contest in Rossiiskaia Gazeta predictably came to naught (Urban 1997). By 1999, savvy Russian elites had concluded that association with even an ersatz “ideology” was a political liability. As Putin declared in an open letter to Russian voters in early 2000, “I am convinced that the main feature of the new century will be not a battle of ideologies, but keen competition for a higher quality of life” (Izvestiia, February 25, 2000, quoted in Johnson’s Russia List #4138). To explain why party building in post-communist Russia has been so difficult, then, we need first to explain why value-rational principles have had so little political impact since the collapse of communism. Here two factors
Instrumental Democracy
181
seem particularly important: the deep political cynicism generated in Russian society by decades of Marxism-Leninism, and the thorough discrediting of ideological alternatives to communism in Russia during the post–Cold War era. I will deal with each of these points in turn. To begin with, the general rejection of “timeless” political principles as a motivation for short-term personal sacrifice in Russia cannot be understood without reference to the remarkably pervasive and alienating form of ideological indoctrination practiced by the Soviet regime. The ironic effect of decades of promises about the “shining future” of socialism has been to inculcate an instinctive distrust of all ideological principles – not only among Russians, but throughout the post-communist world (Rose, Mishler, and Haerpfer 1998; Howard 2000). Thus Russia’s instrumental democracy is a cultural reflection of the failure of the overtly ideological regime that preceded it. In this sense, the decline and fall of the Soviet system and its turbulent aftermath can be understood as a particularly complex example of the routinization, corruption, and disintegration of an unusual type of charismatic order – one that emphasized rational-legal proceduralism to a far greater extent than is typical of charismatic movements. As Ken Jowitt has argued, Leninist regimes, with their characteristic emphasis on simultaneous heroism and discipline, can be classified as “charismatically impersonal,” based upon political and economic institutions that uniquely synthesized the ethos of professionalism with a revolutionary call to smash the existing global order (Jowitt 1992). In terms of the four Weberian types of social action, Leninism combined an essentially affectual appeal to party loyalty, the promotion of a value-rational duty to uphold revolutionary theory and practice revolutionary discipline, and a system of formal institutional incentives that made outward conformity with Leninist political norms instrumentally rational for most individuals. Like Leninist parties, with their simultaneous emphasis on charismatic time transcendence and rational time discipline in the conduct of “professional revolutionaries,” Stalinist economies were designed to synthesize heroism and efficiency in everyday productive labor. The innovation of plan target overfulfillment, combined with “planning from the achieved level,” generated a sort of “final-exam economy” in which it became individually rational for workers and managers to behave as “revolutionaries” smashing “bourgeois” production targets – but not in a “leftist,” undisciplined manner. Thus the Stalinist economic system, like the Leninist political system, can be classified as based upon “charismatic-rational” principles of legitimation (Hanson 1997 [b]). The one type of social action that could not be formally accommodated within the formal legitimating principles of Leninist regimes was simple habit, which was ritually condemned as “time-serving,” “opportunism,” “bourgeois formalism,” and so forth (Gellner 1994). Because habit is in some ways the most common and natural form of social action, the Leninist regime
182
Stephen E. Hanson
was forced to generate perpetual revolutionary struggles against existing social realities in order to maintain its own formal principles of legitimation. After Joseph Stalin’s death, “class struggle” by means of mass terror was officially rejected, yet subsequent Soviet leaders continued to invoke revolutionary phraseology in support of their policy positions. Thus Nikita Khrushchev proposed the “full-scale construction of communism”; Leonid Brezhnev introduced the “scientific-technical revolution”; and Mikhail Gorbachev called perestroika the revolutionary “direct sequel to the great events started by the Leninist Party in the October days of 1917” (Gorbachev 1987, 19). Such calls to revolutionary action can be understood as efforts – in all three cases, probably sincere – to rekindle both sacred emotional bonds with the party leadership and value-rational principles of revolutionary discipline with supposedly “metahistorical” validity. As Weber (and Lenin) understood, however, no form of domination can endure on the basis of appeals to affect and philosophical principle alone; the satisfaction of the mundane interests of at least some key social groups is also crucial. By the late 1980s, the booming capitalist economy of the West, combined with Gorbachev’s naive dismantling of coercive institutions he saw as a “brake” on the expression of genuinely socialist popular sentiment, made continued obedience by the Leninist “staff” instrumentally irrational. Once this point was reached, Steven Solnick’s description of the resulting rapid breakdown of Soviet institutions as a “bank run” becomes perfectly apt (Solnick 1998 [b]). Sensing the near-term collapse of Soviet central authority, instrumentally rational officials had every incentive to steal state resources and to mobilize their connections in order to refashion themselves as independent “businessmen” – and the more officials adopted this strategy, the more rational it became for others. The resulting political and economic crisis generated by the wholesale defection of much of Leninist officialdom tended also to undermine residual affectual and value-rational motivations for fidelity to Leninist institutional norms among the elite – although not entirely, as the later memoirs of such figures as Yegor Ligachev, Nikolai Ryzhkov, and Gorbachev attest (Ligachev 1992; Ryzhkov 1992; Gorbachev 1996). Among the general population, of course, the notion that the party possessed a miraculous and unique ability to transcend time and transform history had been dead for a very long time. There were thus very limited reserves of social support for elite attempts to restore the basic institutional framework of Soviet Leninism after its de facto disintegration under perestroika. Indeed, the Russian Federation came into existence in a social environment where political fidelity to value-rational principles in general was, for most people, associated with oppressive state control over private life. This brings us to the second reason for the absence of successful national parties in the Russian Federation in the first decade of post-communism: the general discrediting of all established political ideologies in Russia at
Instrumental Democracy
183
the start of the twenty-first century. For post-Soviet Russia, at least, Francis Fukuyama’s famous argument that the collapse of Communism marked the “end of history” in the sense that a single ideology – Western liberal capitalism – had won a final victory, looks remarkably prescient (Fukuyama 1989; Lieven 1998). Indeed, there appears to be only one vision of the longterm political and economic future that anyone in the post-communist world finds particularly realistic, and that is integration with “West.” Certainly for those post-communist countries closest to Western Europe, widespread calculations about the feasibility of eventual membership in NATO and/or the European Union (EU) did much to extend time horizons and make cooperative political and economic behavior individually rational. In East-Central Europe, early individual decisions to sink resources into long-term investment and to build national democratic institutions generally paid off. A widespread desire to be accepted as a part of “Europe” thus helped to moderate both short-term corruption and long-term extremism in almost every post-communist European country – even those with comparatively disadvantageous sociological starting points, such as Bulgaria and Romania. In addition, most countries of the former Soviet bloc share a strong cultural antipathy toward the old imperial core; for this reason, nationalists in the post-communist region were ironically often some of the strongest supporters of neo-liberal policies that ended or reduced economic dependence on the Russian Federation. In Russia, however, the disappointments of the first decade of marketization have tended to undermine popular confidence in the long-term enforcement of “Western” liberal norms. Even the consummate pragmatist, Putin, publicly insisted that “Russia will not soon, if ever, become a copy of the USA or, say, Britain, where liberal values have deep historical roots” (Putin, quoted by Reuters, December 28, 1999). Nor can the difficulties of reform in the Russian Federation be as easily blamed on occupation by an alien power; instead, pride in Russia and its history tends to soften contemporary criticism of the Soviet past. Putin’s own ambivalence about Russia’s proper geopolitical place in world is evident from the contradiction between his reassurance of Western audiences that he cannot imagine a Europe without Russia – and his early insistence to domestic supporters that [y]ou and I live in the East, not the West. We have to decide for ourselves what we want: to get credits to buy lollipops or agree to an annexation of our territory from the Black Sea to the Caspian Sea (Putin, quoted in New Republic, November 29, 1999).
Such sentiments to some extent represent the flip side of Western policies of NATO and EU expansion, which have made the future less predictable for Russians at the same time as they have made Central Europeans feel more secure.
184
Stephen E. Hanson
Indeed, with the dream of reintegration into greater Europe increasingly remote, one might expect the emergence of powerful antiliberal ideologies that might catalyze the formation of mass opposition parties. Yet the peculiarities of Russia’s historical situation have also undercut the emotional and philosophical appeal of the two most widespread ideological alternatives to Western liberalism: communism and fascism. The disintegration of Marxism-Leninism has served to discredit, for all but the elderly pre–World War II generation, not just Soviet style but every other variety of “socialism.” Meanwhile, the nearly universal loathing of “fashisty” – a word still associated directly with the German soldiers who nearly destroyed Russia only a few decades ago – has severely limited the appeal of movements explicitly or implicitly emulating Hitler. Finally, the dogmatism, corruption, and seeming complacency of both Zhirinovsky’s and Zyuganov’s parties, especially since 1995, have ironically also contributed to the marginalization of both fascist and Communist ideology in post-Soviet Russia. Russians thus confront a domestic environment in which appeal to ideological principle is politically almost impossible. Russians know from personal experience the disastrous, genocidal results of the antiliberal forms of value rationality advocated by twentieth-century communist and fascist ideologues. Figures who appeal to pre-revolutionary Russian conceptions of duty and self-sacrifice, like Aleksandr Solzhenitsyn, appear sadly anachronistic. And as we have seen, liberal ideology, outside some groups within the urban intelligentsia, now seems equally uninspiring to most sectors of Russian society. Unless some currently marginal political entrepreneur emerges with a genuinely charismatic new ideological vision that might inspire long-term cooperation to build new Russian state institutions, it seems likely that ordinary Russian citizens will continue to assume that politics is a realm of pure instrumentality, and that genuinely compelling value-rational principles to guide individual action must be sought outside the political realm.
conclusion: the future prospects of instrumental democracy in russia If the argument of this chapter is correct, the decline of Marxist-Leninist ideology, and the weakness of value-rational motivations for political action more generally, have greatly exacerbated the difficulties of would-be political party builders during the first decade of post-Soviet Russian politics. To build truly representative national parties involves formidable collective action problems that, as long as post-Soviet Russian politicians act primarily according to the logic of short-term instrumental rationality, can never be overcome. Meanwhile, the main forms of ideological value-rationality that have played critical roles in party formation in Western democracies (Panebianco 1988) have been short-circuited in the Russian Federation. This is, of course, not necessarily such a bad thing, given the authoritarian and
Instrumental Democracy
185
xenophobic character of the ideologies that did prove most successful for organizing Russian parties during the 1990s. Indeed, the weakness of Russian political parties during the first decade of post-communist transition is one of the most salient differences between the Russian Federation and another uncertain postimperial democracy – that of Weimar Germany (Hanson and Kopstein 1997). Thus the absence of compelling ideological commitments that might have elongated the time horizons of fascists and communists has arguably played a positive role in preserving Russia’s formal constitutional order. With the election of Vladimir Putin to Russia’s powerful presidency, however, the continuing absence of powerful party ideologies around which committed activists might rally could prove increasingly problematic for democratic consolidation. On the one hand, the relative stabilization of the Russian economy after August 1998, combined with Putin’s far less mercurial approach to the presidency, appeared at least temporarily to reduce the level of social uncertainty in Russia, making longer-term political strategizing more feasible for instrumentally rational politicians. On the other hand, none of the parties represented in the Duma during Putin’s first term – with the possible exception of Yabloko – could any longer credibly promise to represent their social constituencies in opposition to the will of the Kremlin. Putin’s new law on political parties, introduced in 2001, promised to codify this state of affairs for the foreseeable future by eliminating parties not represented in at least half of Russia’s eighty-nine regions and by denying state funding to parties falling below the 5 percent vote threshold. Clearly, the Putin administration hoped in this manner to maintain the current alignment of political forces in the Duma after the 2003 elections and thus sustain its initial dominance of parliament. As long as the current economic rebound continues and relative international peace is maintained, such a strategy might suffice to preserve Russia’s formal presidential democracy. But the strong possibility remains that new geopolitical, economic, or social crises may erupt in Russia in the years to come – especially given the new global uncertainties after the September 2001 terrorist attacks on the World Trade Center and Pentagon. In that case, proposals for antidemocratic “solutions” to such crises may appear increasingly attractive to Russian elites. Under such circumstances the instrumental nature of support for existing Russian political parties, and for Russian democratic institutions more generally, may seriously weaken social resistance to authoritarian rule. The fate of Russia’s unconsolidated democracy may therefore ultimately depend on the success or failure of appeals to act according to political principle – however instrumentally “irrational” this may be under the circumstances. Thus renewed theoretical attention to the social sources of noninstrumental action is warranted in the West as well.
9 The Impact of the 1999–2000 Parliamentary and Presidential Elections on Political Party Development M. Steven Fish
Despite widespread fears that they might be postponed or even canceled, the parliamentary and presidential elections of 1999 and 2000 were carried out in an orderly fashion, and voter turnout in each election matched the impressive levels seen in the 1995 and 1996 elections for the legislature and the president. The presidential election was held about three months ahead of original schedule due to the sudden resignation of Boris Yeltsin on December 31, 1999. The vote conformed fully to an onerous constitutional provision that severely limits the time allowed for holding new elections in the event of the president’s resignation or death. Balloting was carried out within three months and the constitution was not infringed. Even as war raged in Chechnya, no major incidents of terrorism or other organized violence disrupted the elections. As in the 1995 parliamentary election, a bewildering number of parties competed in the party-lists portion of the balloting for the Duma in 1999, but only a handful cleared the 5-percent threshold for representation in parliament (see Table 1.1 in Chapter 1). The performance of parties created some appearance of continuity between elections, with three of the four parties that cleared the 5-percent barrier in 1995 doing so again in 1999. What is more, the percentage of the vote “wasted” – that which went to parties that failed to secure the 5-percent minimum – declined steeply, from 49 in 1995 to 19 percent in 1999. A core of several major organizations took the lion’s share of the vote. Thus, the effects of the noise created by marginal organizations were reduced and disproportionality in the allocation of seats in parliament fell between the 1995 and 1999 elections. The result would seem to suggest some movement toward consolidation and maturation of the party system. In the new Duma, parties continue, as they did previously, to furnish some structure for political competition, even if alliance patterns in the early months since the elections have provided some surprises (see Chapter 11; also Haspel, Remington, and Smith 1998). The notion that Russia is a virtually party-free polity, or that parties are 186
Impact of the Elections on Political Party Development
187
utterly insignificant in Russian politics, was not sound before the most recent round of elections, nor is it valid after them (Fish 1995 [a]; Moser 1999 [b]; Petrov 2000 [b]). Yet political parties and the system they constitute are scarcely hale and vigorous, and the elections of 1999 and 2000 did not strengthen them. In post-Soviet Russia, elections have generally provided only a modest stimulus to party development. The most recent elections, by re-creating essentially the same conditions that prevailed in previous contests, did not invigorate Russia’s party system but rather reinforced its overall weakness. The enduring debility of the party system holds important implications for democratization in Russia. Strong parties serve as a linchpin in state-society relations in democracies of all types. They enjoy a degree of legitimacy in the polity that interest associations, by virtue of their explicit aim to advance the cause of a particular, circumscribed group of citizens, cannot pretend to attain (Schattschneider 1948; Lipset 1960; Key 1961; Sartori 1976; Kitschelt 1992 [a]). Political parties advance the representativeness of government by furnishing voters with distinct choices and by lending discipline, structure, and predictability to competition in the organs of government. They thus have the potential to hold political elites accountable to constituencies. Without bona fide, reasonably strong programmatic parties that are capable of structuring campaigns, serving as coherent sources of identification for candidates from various locales and for various offices, and furnishing some constancy from one election to the next, elections may not be meaningful, even if they are freely contested (Katz 1980). Parties not only conduce the accountability of rulers to the ruled; they also may facilitate social control and thus sustain governability. Because parties organize, channel, and control societies’ demands on the state, they may help rulers maintain control without resort to coercion (Huntington 1968). During the past several decades, political scientists have debated the trend in parties’ roles and influence in established industrial democracies. Some analysts maintain that the importance of parties has waned relative to other organized structures of interest intermediation, such as interest groups or social movement organizations (Schmitter 1981; Schmitter 1992). Other scholars have argued that parties have indeed changed in significant ways but have nevertheless retained a central and indispensable place, even in highly corporatist democracies in which interest groups have assumed crucial roles in government (Lehmbruch 1984; Mair 1994). However one assesses the merits of these arguments, the centrality of parties during periods of the establishment and consolidation of democracy in the older industrial democracies, as well as the gravity of parties at least during certain stages and moments in neodemocracies, is beyond dispute. That the strength and character of parties and party systems are highly relevant to democracy and democratization is not in question among the vast majority of students of regime change, democracy, and political organization (O’Donnell and Schmitter 1986; Fish
188
M. Steven Fish
1995 [b]; Tok´es 1996; Huang 1997; Ishiyama 1997; Maravall 1997; Fleron, Ahl, and Lane 1998; Collier 1999).
evaluating russia’s party system In order to assess the effects of the elections on the party system in Russia, it is helpful to establish some criteria for the strength or development (the two terms will be used interchangeably) of party systems in general. Many criteria could be used, and a multitude may be found in the voluminous literature on parties. Here I use a half-dozen indicators that I regard as particularly telling. Analogues of some of these indicators may be found in other writings on parties. Some other of these indicators are not used, or are not used extensively, in the existing literature. Taken together, they represent a departure from previous work. All criteria may be treated as continuous, though in some instances I adduce hypothetical cases of “high” and “low” scoring in order to furnish simple, transparent, and concrete denotation of the indicators. The first indicator of party-system development is the overall size of political parties taken together, which may be understood as of the proportion of the voting-age population that belongs to a political party. The meaning of party membership of course varies from polity to polity. In some, such as the United States, membership is exceptionally easy to attain and imposes little obligation. In many European countries membership requires some sacrifice of time and dues as well as participation in activities beyond voting in elections. Membership numbers do not always reflect a party’s strength in society. Some small parties may be influential, and larger organizations often lose elections to smaller ones. Still, below a certain threshold of membership, it is difficult for a party to communicate its message and generate the resources needed to influence public opinion and policy (Duverger 1978, 61–132; Bartolini 1983; Fish 1997). An expansion in membership on the part of an established party has often been associated with a subsequent improvement in that party’s political performance. An example is found in the conversion of the German Christian Democratic Party (CDU) from a party of local notables to a mass-based, deeply rooted organization in the 1970s and 1980s (Paterson 1987; Clemens 1989; Clemens 1994). However one assesses the relationship between party size and performance, the extent of overall membership in political parties reflects the weight that parties carry in a given society. I refer to this factor as magnitude. The second indicator is the degree of constancy of party presence in electoral competition over time. If each election features more or less the same lineup of parties, provided the identities of those parties do not shift radically from one election to the next, then the party system in one sense may be regarded as robust. On the other end of the spectrum, if few or none of
Impact of the Elections on Political Party Development
189
the same parties appear in successive elections, the party system is highly fluid. This factor I call stability. The third aspect of party-system development is found in political diversity, which may be labeled differentiation. If viable parties represent a wide spectrum of opinion, then the system is highly differentiated; if they do not, the system is poorly differentiated. The fourth indicator, gravity, refers to the importance of party affiliation for candidates for public office. Very high gravity would be in evidence when all or nearly all politicians take a keen interest in party affairs. Very low gravity would be found in a polity in which politicians ignore parties and regard party affiliation as inconsequential. The fifth indicator is the internal coherence of parties. This variable refers to how well the major parties hold together from election to election. All parties can and must change over time. The composition of leadership and the predominance of this or that faction or tendency are often in flux. But one may distinguish between parties that maintain a reasonable degree of coherence over time and those that are prone to unraveling. This factor is designated integrity. The sixth component of the extent of party-system development, which I term reach, regards the scope of geographical coverage. A system in which all major parties are national organizations, in the sense that they compete against one another on the national level, have organizations everywhere, and enjoy identities that are basically the same from region to region, would score high on this dimension. A party system that is dominated by regional parties that merely fight out interregional battles in the capital city would score low. Magnitude, stability, differentiation, gravity, integrity, and reach thus provide a set of criteria for assessing the extent of the development of party systems. One can readily imagine other factors that might also represent important aspects of party-system development. But taken together, these six criteria furnish reasonably good coverage and they supply a firm basis for thinking about Russia’s party system. I make no attempt to weigh the relative importance of these criteria in the overall development of the party system, though doing so may provide an interesting avenue for further research. These criteria, like all variables, are meaningful only in a comparative context. As my interest is in assessing the effects of the 1999 and 2000 elections on party-system development in Russia, the primary field of comparison will be cross-temporal and within-country, though I also adduce other countries as comparative referents. How might one evaluate the Russian party system in light of these six indicators, and what difference, if any, did the elections make? In terms of magnitude, Russia commenced the electoral season with one of the lowest rates of party membership in the post-Communist region. In a recent study based on data gathered in 1998, Howard (1999) reported that Russia rated
190
M. Steven Fish
second to last among the eleven major post-Communist countries surveyed in terms of percentage of the population that belongs to political parties. Membership in parties per capita is far lower in the post-Communist region than in major postauthoritarian polities in Latin America, Southern Europe, and East Asia, as well as in the long-established democracies of Europe and North America. Yet, even by post-Communist standards, Russia rates very low. Less than 2 percent of the population belonged to a political party. Only Ukraine ranked lower, and the difference between the two countries was slight (Howard 1999, 94; see also White, Rose, and McAllister 1997, 134–8). There is no evidence that parties’ memberships expanded during the election campaign. Unity (Edinstvo), the party backed (but not joined) by the then-prime minister, Vladimir Putin, may eventually grow into a mass membership organization. If it does so, its birth will turn out to be these elections’ main contribution to the growth of popular participation in political parties. During the campaign and in its immediate aftermath, however, Unity was little more than an arm of the presidential administration. Because Unity did put candidates forward in the parliamentary election, it carried out what counts as a party’s minimum function and therefore qualified as a party, but it did not seek members in broader society. The Fatherland-All Russia party (Otechestvo-Vsia Rossiia [OVR]) similarly put candidates forward for election, but also did not recruit members outside of its own candidates for public office. These parties were not only not mass-membership parties; they were not even “membership parties” (see also Chapter 2). Whether or not Unity will become a real membership party, as of this writing, remains to be seen. The only mass-membership party in Russia, the Communist Party of the Russian Federation (CPRF), with its half-million members, does not appear to have either grown or shrunk as a result of the elections. The country’s theretofore second-largest party, the Liberal-Democratic Party of Russia (LDPR), experienced a surge of internal friction on the eve of the elections. A faction within the party opposed the party’s maximum leader, Vladimir Zhirinovsky, who subsequently ran not as the LDPR candidate but as the chief of the Zhirinovsky Bloc. The Yabloko party, until the elections the country’s leading liberal organization, showed no signs of growth and saw its vote total inch downward, continuing a trend in evidence since 1993. The Union of Right Forces (Soiuz Pravykh Sil [SPS]), which brought together liberals who had earlier been scattered in smaller parties such as Democratic Choice of Russia (DemVybor Rossii [DVR]) and Common Cause (Obshchee Delo [OD]), mounted a relatively impressive organizational effort. The SPS showed a degree of unity that had previously eluded liberals outside of Yabloko. But like Unity and OVR, the SPS was not a membership party but rather an electoral alliance of several hundred aspirants for public office. In sum, there is little evidence that the elections spurred a rise in the membership magnitude of political parties (McFaul, Ryabov, Petrov, and Reisch 1999; Petrov 2000 [b]).
Impact of the Elections on Political Party Development
191
The constancy of the composition of the party system, referred to here as stability, likewise did not receive a boost in the elections. Three parties that were present in the previous Duma (the CPRF, the LDPR/Zhirinovsky Bloc, and Yabloko) all secured places in the new Duma as well. But three of the six parties that cleared the threshold in the list vote did not exist six months before the parliamentary election. The three new parties together captured 45 percent in list voting, compared to 36 percent for parties that were represented in the previous Duma. Our Home Is Russia (Nash Dom Rossiia [NDR]), the party led by the former prime minister, Viktor Chernomyrdin, virtually disappeared in the 1999 vote, receiving just 2 percent. Though the elections did not transform the party system entirely, neither did they induce or evidence stabilization. The differentiation of the party system was extensive before the elections and remains extensive after them. In this respect, the post-Soviet Russian party system consistently has been impressive. The full spectrum of public opinion, from far left to far right, and from virulently nationalistic to highly cosmopolitan, has been well represented in the party system. The 1999 parliamentary elections might have represented a slight advance in the party system’s degree of differentiation, because the success of the SPS diversified the liberal/cosmopolitan portion of the deputies’ corpus of the Duma. On the other hand, the waning of NDR might be seen as weakening the political center and thereby reducing the richness of the party system’s differentiation. Precisely what constitutes “centrism” in the Russian context and which parties embody it is the source of discussion and debate. It is difficult to assess whether Unity will play the role of a genuine centrist party. It claims such a label for itself, but its political orientation will become clear only during the first several years of the current decade (McFaul 1993 [b]; Fish 1995 [a]; Petrov 2000 [b]). On the whole, the 1999 and 2000 votes did not appear to induce a major change in the differentiation of the party system, which was already highly advanced even before the elections. Gravity, or the importance of party affiliation for candidates for public office, certainly did not increase as a result of the 1999 and 2000 elections. Naturally parties remained important for candidates who ran on the party lists; the list component of the Duma elections assures a certain minimum level of significance for parties in all elections for the national legislature. But beyond this aspect of the election, which was constant through the 1993, 1995, and 1999 contests, gravity seems actually to have declined. Evidence is found in two areas. The first is the importance of parties in the single-member district (SMD) voting. The proportion of candidates who are unaffiliated with parties is difficult to assess precisely. Some candidates run as independents but nevertheless are informally affiliated with a party; some are put forward by a party but are not even members of it. Be that as it may, the number of candidates elected in the SMDs with no party affiliation actually grew in 1999, rising to about 46 percent from roughly 40 percent in 1995 (Petrov
192
M. Steven Fish
2000 [b]). Once in the Duma, most candidates elected as independents form a party (or “faction” in Russian legislative parlance) or join an established one (see Chapter 11). But the percent of candidates who win their seats without party affiliation is nevertheless of some significance for the character of the subsequent Duma, because many independents subsequently form or join politically amorphous and programmatically vacuous factions. In the current Duma these include Russia’s Regions and People’s Deputy, two groups with appropriately inane appellations that together comprise roughly the same number of deputies as Yabloko, OVR, and the SPS combined. The other indicator of gravity is the part that parties played in the presidential election. Parties played some role in organizing support for the various candidates in the 1996 presidential contest. But the winner, Boris Yeltsin, scorned any affiliation with a party, as he had from the time of the nascence of the democratic movement in the late 1980s. The third-place finisher, Aleksander Lebed, was only loosely associated with a party, and he with one that carried little weight either in his campaign or in Russian politics more broadly. Parties were of similarly little significance in 2000. The winner, the then-acting president, Vladimir Putin, was unaffiliated with any party, although Putin, like Yeltsin, had made his preference for a particular party known before the parliamentary elections that preceded the presidential contest. Putin, however, shunned closely associating himself with any party. What is more, Putin owes no party anything. Like Yeltsin in 1996, he did not even need the organizational assets of a party, but relied for resources exclusively on the presidential and governmental apparatuses. Putin’s first year in office did not reveal clearly how intensively he intends to engage in party politics. Like Yeltsin, Putin insists that he has nothing to do with the behavior of the party he endorses. Putin even claimed to have had no say in the most important action that Unity took during the opening of the new Duma, which was to create a condominium with the CPRF and divide committee chairmanships between the two parties and their close allies. The action virtually excluded other factions – in a sharp break with tradition – from a share of these positions. Putin’s denial of involvement seems highly dubious, but his pretense honored the tradition established by Yeltsin of posing as the popular tribune who flies above the indecorous business of party politics. The elections did not advance the coherence, or integrity, of political parties. Yabloko maintained roughly the same, reasonably high, level of internal coherence that it had before the elections. The decline in its performance – from 7 to 6 percent of the list vote and from fourteen to just four SMD seats – between 1995 and 1999, however, reduced Yabloko’s salience in Russian politics. The CPRF also maintained its substantial level of organizational cohesion through the parliamentary election. In the presidential contest, however, some prominent party leaders abandoned their leader, Gennadii Zyuganov, to support Putin. Indeed, fissures in the party
Impact of the Elections on Political Party Development
193
spread and deepened between the legislative and presidential elections. A substantial portion of the Komsomol, which claims 34,000 members and which in recent years has sometimes had strained relations with its parent organization, the CPRF, peeled off and joined Molodoe Edinstvo (Young Unity), the youth group that formed to support Putin.1 For its own part, Unity came together just months before the parliamentary contest and has little organizational identity apart from its professed loyalty to Putin. As long as Putin retains his formidable popularity, the party will have a rallying point, but whether it possesses the requirements for organizational integrity remains to be seen. That OVR did not possess such requirements became clear immediately following the parliamentary elections. Within a week after the vote the party began coming apart.2 The LDPR experienced internal discord on the eve of the parliamentary election. The SPS, while serving as an impressive organizational forum for amalgamating previously dispersed liberal forces, began coming unglued on the eve of the presidential contest. Some leaders abandoned their earlier commitment to Putin in favor of backing the head of Yabloko, Grigorii Yavlinsky, while other elements of the SPS stuck with their party’s original pro-Putin stance.3 Whether the SPS will become a coherent liberal party remains an open question. The geographical reach of Russia’s major parties has always been extensive and remains so in the aftermath of the elections. The 5-percent threshold in party-list balloting, as well as the absence of concentrations of ethnic or national minorities that number more than a few million people in any given geographical area, has helped give rise to a system devoid of substantial regional parties. As in any polity, some parties fare better in some regions than others. Yabloko’s organization and followership are most substantial in the largest cities, while the CPRF boasts its strongest organizations and most impressive electoral performance in central and southern European Russia. But all of the major parties are national in reach and reputation. None are concentrated exclusively in a single region or aim to represent regional interests. The Russia’s Regions faction in the Duma brings together independent deputies from various regions who do not have a party affiliation. Its pretensions to represent “regional interests” do not imply that it is a “regional” party in any meaningful sense, as its members hail from virtually everywhere. The party is analogous to a hypothetical party in the ` United States that represented “the states” – not “states’ rights” vis-a-vis the national government or any particular bloc of states, either of which might 1 2 3
“Stalinists Buddy Up to Unity’s Bear Cubs.” The Moscow Times, January 26, 2000. “U Primakova otniali ‘Vsiu Rossiiu’.” Segodnya, December 23, 1999. “A za spinoi – devich’i kolenki.” Vecherniaia Moskva, March 24, 2000. “Liberals Line Up Behind Yavlinsky.” The Moscow Times, March 23, 2000. “SPS Jumps on Putin’s Bandwagon.” The Moscow Times, March 15, 2000.
194
M. Steven Fish
mean that the party had some programmatic content, but just “the states.” Such a party would make no sense in the American setting. Russia’s Regions makes no sense in the Russian setting. Unsurprisingly, the faction’s history of discipline in voting in the Duma is unimpressive (Haspel, Remington, and Smith 1998; Remington 1998). There exists no more of a simultaneously unified and geographically universal “regional” interest in Russia than there does in the United States or for that matter any other country. Thus even the organization that calls itself “Russia’s Regions” is national in scope. There is no substantial organization that aims to represent the interests of, say, Siberians, Muslims, Tatars, far easterners, peoples of the Caucasus, or Muscovites. Many polities that are as multiethnic as Russia and that, like Russia, have regional concentrations of ethnic minorities have substantial ethnic parties. Furthermore, Russia’s sheer geographical size and the erosion of central power during the 1990s may have conduced the formation of parties based on ethnicity or region. The absence of such parties is a remarkable characteristic of Russian politics, and the 1999–2000 elections did not alter this feature. In sum, in terms of the third and sixth gauges of party-system development, differentiation and reach, the party system was impressive before the elections and the elections had no appreciable effect. With respect to the first and second indicators, magnitude and stability, the party system did not rate high before the elections and the elections did not spur any improvement. In terms of the fourth and fifth indicators, gravity and integrity, the elections actually appear to have instigated deterioration or backward movement. On balance, the elections did not improve the strength and maturity of Russia’s inchoate party system. They might have even had the opposite effect. Russia has had five national contests – three parliamentary and two presidential elections – since the demise of the USSR. And yet, elections have not induced party-system development as strongly or surely as they have in many other postauthoritarian, including post-Communist, polities. Why not? The following two sections investigate several possible answers. They focus particularly on historico-cultural, structural, and institutional explanations for the failure of elections to impel party development.
russia’s underdeveloped party system: historico-cultural and structural explanations The belief that ordinary Russians are too socially and culturally backward to participate effectively in politics, including in political parties, is widely held among Russian intellectuals. Such a viewpoint makes a sharp, if often unspoken, distinction between “them” (“the people,” or the narod in Russian) and “us” (the intellectuals). In such thinking, scornful condescension is often joined with empirical inaccuracy. For example, on the eve of Putin’s triumph, Evgenia Albats, a leading Moscow journalist, wrote: “Now
Impact of the Elections on Political Party Development
195
the time has come for an ordinary man who is neither part of the Soviet elite club, nor a democrat. In this way, Putin truly represents the country he is going to run . . . he is the man who truly reflects a sick, perplexed and lost Russian society.” Elsewhere in her analysis, Albats held that the revolution of 1991 was “imposed from above with no support on the ground.” She stated that “just a few hundred Muscovites” participated in the defense of the White House during the abortive putsch attempt of August 1991 (Albats 2000). Albats’s memory and understanding of her own people strike one who stood in crowds of hundreds of thousands in the capital and tens of thousands in other Russian cities during 1989–91, and who recalls that over 100,000 people defended the White House, as a bit odd. Her point of view nevertheless merits mention because it is so widespread among opinion leaders in Moscow. It figures prominently in many Russians’ explanations for why their country’s political society, including its party system, suffers from underdevelopment. Auspiciously, some analyses that stress the importance of culture and history reflect more incisive thinking. White, Rose, and McAllister (1997, 136– 51), after undertaking sophisticated survey work on partisan identification, find the bulk of the Russian electorate to divide into “uncommitted, untrusting voters” (40 percent) and “antiparty voters” (33 percent). The authors find that (non)identification with parties is due less to socioeconomic variables than to attitudinal factors, including authoritarian attitudes, general mistrust in institutions, and lingering attachment to the Communist Party, which of course has a long-established identity not enjoyed by other parties. Remington (1999, 171) concurs with White, Rose, and McAllister, and notes “the deep alienation of ordinary Russian citizens from the play of party politics.” In analysis published in the early post-Soviet period, Sakwa (1993) suggests that Russians may feel distaste for the open expression of discord between parties and criticism of powerholders. In later work, Sakwa (1995 [b], 958) finds in Russian opinion an “asocial subjectivity [that] has rejected politics.” Cultural and historical explanations figure prominently in sophisticated writings by journalists as well. In an influential book, Lieven seeks, in his own words, “to demolish the all-too-influential view which sees the Russian nation as culturally unchanging over the centuries.” He also selfconsciously dismisses crude historico-cultural determinism. While steering around this pitfall, Lieven offers a historical explanation of his own that emphasizes the Soviet past. In his view, recent history – and especially traumatic episodes such as Stalinism – produced major discontinuities whose effects weigh heavily upon the present. He holds that “it is because Stalin shattered and atomized Russian society” that the latter has, under post-Soviet circumstances, “proved largely incapable of generating mass democratic politics” (Lieven 1999, ix, 5, 162). In addition to history and culture, socioeconomic structure has been adduced to explain the weakness of party development in Russia. In my own
196
M. Steven Fish
study of the nascence of political society during the Gorbachev years, I pointed to the relatively undifferentiated structure of Soviet society and the absence of a market economy to help explain the slow growth of parties and other independent societal organizations (Fish 1995 [b]). The legacy of the command economy undoubtedly still affects both party formation and democratization more generally. Like many other countries in the postCommunist world but unlike many countries of Southern Europe, Latin America, and East Asia, Russia did not undertake its transition with a sizable, sturdy middle class already in place (Bunce 1998). Such a class, according to many scholars, provides a bulwark of support for open government (Moore 1966). Even if the presence of a substantial middle class does not always necessarily conduce democratization, its absence or weakness may limit the growth of political parties of certain types (Fish 1995 [a]). The extreme dependence of citizens on the state both as employees and consumers during the Soviet period, moreover, left a legacy that is not necessarily auspicious for rapid party formation or collective self-organization more broadly (Crowley 1994; Fish 1995 [b]). The dynamics of the transition processes, by deepening inequality and in many respects failing to broaden economic opportunities for ordinary citizens nearly as rapidly as they have for elites, may well reinforce the effects of the legacy of the command economy (Kryshtanovskaya and White 1996; Solnick 1998). Indeed, they may further deepen popular alienation and subsequently check the expansion of mass participation in public life. At this point structural and cultural perspectives are joined, with attitudinal and socioeconomic factors reinforcing one another – and in a manner that may work against the growth of broad and vigorous popular participation, including in political parties (Fleron, Ahl, and Lane 1998; Kelly 1999; Kullberg and Zimmerman 1999). Explanations that see Russian history as having exerted a powerful impact, through the creation of culture or of a particular socioeconomic structure, unquestionably have considerable merit. A spirit of antipolitics and antiparty sentiment, and distrust for political organizations in general, does pervade Russian society, and may well retard the growth of political parties. Even if Russians carry no gene for political passivity, there can be little doubt that the Soviet experience, and perhaps also pre-Soviet Russian history, left corrosive legacies of social atomization and inexperience in nonstate political participation. The legacy of the command economy, moreover, does not create ideal socioeconomic conditions for the emergence of interest-based political parties. And yet, historico-cultural and socioeconomic-structural factors do not necessarily furnish full, or even necessarily optimal, explanations for the weakness of parties in Russia. Cross-national, within-region comparisons reveal the possible limitations of these explanations. All the countries of the region bear the scars of the Soviet legacy, including cultures of suspicion and habits of withdrawal into private-regarding activities at the expense
Impact of the Elections on Political Party Development
197
of civic engagement (Rupnik 1989; Jowitt 1992 [b]; Tismaneanu 1992). Political party membership is lower in post-Communist countries than in other postauthoritarian countries, but there is still substantial variation in the post-Communist region. Per capita membership in parties is much higher in Bulgaria and Slovakia than in Russia. Latvia and Lithuania also boast higher levels of membership magnitude. Romania – a country that, unlike Russia, labored under full-blown Stalinism until the end of its Communist period – scores much higher in terms of participation in political parties. Roughly 12 percent of Romanians belong to a party. Only Ukraine trails Russia, and not by much. The highest level of membership magnitude in the region is found in Mongolia, which experienced the full brunt of Stalinism and was sovietized as thoroughly and for as long as any country in the USSR (Fish, 1998 [a]; Howard 1999). Such within-region differences cannot easily be accounted for in historical or cultural terms. Similarly, socioeconomic-structural explanations potentially provide plausible statements on why party development has been slower in the postCommunist region than in other areas, but do not necessarily account for cross-national divergence in party development within the region. It is extremely difficult to understand social stratification, and particularly problematic to analyze class formation, in the post-Communist world (Kaufmann and Kaliberda 1996; World Bank 1996). For example, the size of the “middle class” in post-Soviet Russia is the subject of a vigorous debate. On the one hand, Sovietism in Russia did what it set out to do and what it did everywhere in the region: eliminate classes. To be sure, the legacy of Sovietism was not complete social equality, but it was a social structure that did not include classes in the same sense that they are understood and exist in the West or most of the Third World. On the other hand, however, class formation has occurred at an explosive pace throughout the former Communist world since the demise of the partyocracies. Russia is no exception. Some analysts have caricatured social change in Russia, claiming that the vast majority of society during the Yeltsin period sank into a miserable swamp of undifferentiated poverty, while a miniscule elite, centered entirely in Moscow, acquired the country’s entire capital stock (Cohen 1999). More sober analyses place the size of the middle class – or, perhaps more aptly stated, the middle classes or strata – in Russia at somewhere between one-fifth and one-third of the population (Zaslavskaya 1995; Rutkevich 1997; Balzer 1998). According to a major recent study, moreover, the prevalence of class-related identities is on the rise in Russia and other post-Communist countries (Miller, Klobucar, Reisinger, and Hesli 1998). Simple observation supports such findings. As of the early 2000s, to say that there is no sizable middle class in Russia, or that the middle class was completely wiped out in the 1998 financial collapse, is to say that one has not visited Russia in recent years. To say that the only sizable nonpoor segment of the population in Russia lives in Moscow is to say that one has not traveled in Russia outside of Moscow recently. Given the
198
M. Steven Fish
recentness of the demise of Sovietism, Russia’s middle strata do not of course “look” or “act” precisely like long-established middle classes in Western industrial democracies – though one may question whether middle classes in the West ever really had quite the cohesion or reliable political orientations or played precisely the roles ascribed to them in some of the classical social science literature on socioeconomic modernization. At any rate, it would not appear that socioeconomic change or structure finds clear or predictable expression in Russia’s party system. Parties representing middle-class interests have not really taken hold. Yabloko, perhaps the closest thing Russia has to a bona fide “middle-class” party, is a diminutive organization of several thousand members whose vote total hovered at around 7 percent during elections in the 1990s. To the extent that one can identify cross-national differences in socioeconomic conditions, moreover, they do not appear to reflect many of the most significant and glaring differences among party systems. From the perspective of a socioeconomic approach, would one expect Bulgaria to serve as the site of the most formidable liberal right-center party in East Europe? Bulgaria is poorer – and certainly does not have a proportionately larger middle class – than, say, the Czech Republic, Hungary, or Russia. Yet Bulgaria’s Union of Democratic Forces (UDF) became the most puissant “bourgeois” party in the region (Fish and Brooks 2000). Some recent cross-national empirical analyses have found that the influence of social structure on citizens’ party preferences and voting behavior has been tenuous and indirect (Kitschelt, Mansfeldova, Markowski, ´ 1999, 291–302). and Toka Membership magnitude is not the only dimension on which Russia lags behind in within-region cross-national perspective. For example, Mongolia, Romania, Bulgaria, and Slovakia, despite their harsh Soviet legacies, also have party systems that rate higher than Russia’s in terms of gravity. Parties in all four of these countries play a more important role in politics than they do in Russia. Elections in all four have stimulated party formation and development. The notion of a serious politician eschewing party affiliation entirely is inconceivable in these polities. Party systems also rate higher in terms of stability than they do in Russia (Fish 1998 [a]; Fish 1999 [b]; Fish and Brooks 2000). Indeed, with the exception of differentiation and reach, where Russia’s parties score high, Russia generally lags behind other countries of the region in terms of the indicators used here. Excluding polities such as Belarus or Uzbekistan where pluralism and party competition are banned, Russia’s party system is, in comparative terms, especially underdeveloped. Russian parties seem to be exceptionally resistant to the opportunities for growth and maturation furnished by competitive elections. The relative underdevelopment of Russia’s party system suggests the necessity of looking beyond cultural, historical, and socioeconomic factors, many of which are shared broadly by Russia’s neighbors. The following section offers an alternative approach. It is not necessarily incompatible with the historico-cultural
Impact of the Elections on Political Party Development
199
and socioeconomic-structural explanations reviewed here. Indeed, there is no logical reason why such explanations and an institutional approach, such as the one I offer in the following text, need be regarded as mutually exclusive (Fleron, Ahl, and Lane 1998; Hesli, Reisinger, and Miller 1998). I regard the latter approach as enjoying distinct advantages, however, and in the interests of making these clear, I treat the approach as a separate, rival explanation.
an alternative explanation: people and the institutions they make “Institution” has become the most overused and amorphous concept in political science. Specifying precisely what one has in mind is therefore crucial in any analysis that places causal weight on institutions. By institutions I mean formal institutions, and specifically the constitutional provisions that define the regime and the electoral rules that govern access to public office. I aim to specify a particular way of analyzing the resistance of the party system to the potentially invigorating effects of electoral competition and to draw out the logic of specific causal connections. Some of the institutional architecture affecting parties in Russia has undoubtedly advanced party development. Four institutions in particular have helped ensure that parties have at least established a presence in political life. The first is the party-lists portion of balloting for the Duma, which allots half of the body’s 450 seats according to proportional representation (PR). The PR provision greatly enhances parties’ gravity. Without it, the stimulus to party development would be even weaker than it is in fact (Moser 1995). Russia would very likely resemble Ukraine, where the party system is even more anemic than in Russia. In Ukraine, the successor organizations to the old Communist Party are the only groups with a real national presence. The only major liberal party is associated with Ukrainian nationalism and lacks reach, because its presence is substantial only in the Western parts of the country. In 1998, in a self-conscious imitation of Russia and with the explicit goal of spurring party development, Ukraine adopted a system that provided for a PR component in its parliamentary elections (Way, Shaiko, and Tucker 2000). The PR portion of voting for the Duma exerts unequivocally positive effects on party development. The influence of the other three institutions that also aid parties actually may cut in several directions. Their effects are less clear-cut and are potentially controversial. The first is the 5-percent threshold for representation in the Duma. One could argue that this barrier is too high and produces too many “wasted” votes. But it has a sorely needed consolidating effect within the party system (Moser 1999 [a]). It helps weed out insubstantial parties and encourages affiliation with a major one. The strong performance of Unity in the 1999 legislative election shows that a
200
M. Steven Fish
party need not have a long history or a firmly established reputation in order to clear the threshold for parliamentary representation. But the 5-percent rule does give parties an incentive to develop and maintain a reasonably constant identity from one election to the next, which politicians in the CPRF, Yabloko, the LDPR, and several other parties have indeed endeavored to do. The rule therefore may promote stability and integrity. The third institution is the use of a single national list. One could argue that using local or regional party lists, as is done in some other postCommunist countries in East Europe, would encourage party development on the local level (Ordeshook and Shvetsova 1997). But treating the entire country as one large district, as the rules do in Russia, does enhance parties’ reach. Combined with the 5-percent barrier, it inhibits the rise of regional parties that represent particularistic geographical or ethnic interests, and helps explain why such parties do not enjoy a major presence in Russian politics. Finally, two rules that govern internal procedures in the Duma fortify parties. One is the provision that establishes a high minimum – thirty-five deputies, or nearly 8 percent of the entire corpus – for status as a bona fide “faction,” meaning a political party (Remington and Smith 1995). The rule encourages deputies elected as independents to affiliate with some established party or to found their own. It retards wild party switching and peeling off of deputies from established parties. Such problems are not unknown in the Duma, but they would be worse without the rule. The presence of an inducement to belong to a party at all in the Duma is ensured by the procedure of distributing party chairmanships on the basis of party. The thirty-five-deputy threshold and the allotment of committee chairmanships by party provide a fillip to parties’ gravity and integrity. Thus, parties in Russia do not face a phalanx of uniformly inhibiting institutions. Some rules governing elections and the Duma’s internal operation may have auspicious effects. These institutions, however, are countervailed by others that check the growth and maturation of parties. The most pernicious institution is superpresidentialism (on Russian superpresidentialism, see Roeder 1994; Colton 1995 [a]; Colton 1995 [b]; Huskey 1995; Stavrakis 1996; Fish 2000). The following features characterize superpresidentialism: an apparatus of executive power that dwarfs all other state agencies in terms of size and the resources it consumes; a president who enjoys decree powers; a president who de jure or de facto controls most of the powers of the purse; a relatively toothless legislature that cannot repeal presidential decrees and that enjoys scant authority and/or resources to monitor the chief executive; provisions that render impeachment of the president virtually impossible; and a court system that is controlled wholly or mainly by the chief executive and that cannot in practice check presidential prerogatives or even abuse of power. Superpresidentialism is a type of regime. The chief executive does not enjoy total power and is subject to bona fide,
Impact of the Elections on Political Party Development
201
periodic challenge in national elections. Superpresidentialism therefore differs from dictatorship, such as is found in Azerbaijan since 1993; Kazakhstan and Belarus since the mid-1990s; Serbia until 2001; and Turkmenistan and Uzbekistan since the demise of Communist regimes. But superpresidentialism also differs markedly from moderate presidentialism and semipresidentialism, which are found in, among other countries, the United States, France, Lithuania, Poland, Romania, and Mongolia. In the post-Communist region as of the end of the beginning of the 2000s, superpresidentialism is found in Kyrgyzstan and Armenia as well as in Russia. Ukraine is an ambiguous case; its regime includes several but not all of the components previously listed. In most respects it is superpresidential; in some, moderate presidential. Several countries have constitutions that are semipresidential or even parliamentary in form, but also presidents who exploited crises and constitutional ambiguities in a manner that enabled them to establish essentially superpresidential regimes. Croatia under Franjo Tudjman and Albania under Sali Berisha exemplify this phenomenon. Superpresidentialism chills party development in part by holding down incentives for important political and economic actors to invest in parties. Incentives to engage in party building depend vitally on the power of the national legislature. Parliaments and parliamentary elections are the fora in which parties typically draw public attention, form identities and reputations, and find their voices (Linz 1994; Fleron, Ahl, and Lane 1998). Parties play a substantial role in Russian parliamentary elections and structure competition in the Duma between elections. But the weakness of parliament compared to the executive branch deadens societal actors’ and many politicians’ interests in pursuing their goals by building political parties. Working on – lobbying, usually bribing – the official or handful of officials in the executive branch who are responsible for policy in one’s area of concern is typically much more effective than contributing to political parties or launching organizations that lobby parties and legislators. The superpresidential system therefore animates the formation of small, closed, compact societal organizations that are skilled in targeting and pressuring officials in executivebranch agencies, both in the presidential apparatus and the ministries. Thus, highly personalistic, well-heeled coteries representing narrow business interests – rather than political parties – constituted the main growth sector in the realm of sociopolitical organization in Russia in the 1990s (Shevtsova 1995; McFaul 1997; Fish 1998 [b]). One can, of course, imagine situations in which both such interest groups and political parties play large roles in politics; the contemporary United States and Germany offer two obvious examples. But in the United States and Germany, interest groups can and must exert their muscle, at least in part, through the major parties. Indeed, as the free flow of money from wealthy interest groups to party coffers in both countries shows, this is precisely how politics works. Many citizens justifiably feel that this stream of cash corrupts electoral competition. Still,
202
M. Steven Fish
the necessity to influence parties if one intends to affect policy does provide an inducement for well-endowed societal actors to aid parties and helps keep parties flush. The incentive that private interests have to invest in parties in the United States and Germany grows not only out of the presence of powerful legislatures, but also out of parties’ authority over nominating procedures. This observation illuminates the second major institutional obstacle to party development in Russia: rules that allow nonparty associations to nominate candidates for office and that do not afford parties special advantages in nomination. If parties enjoyed de jure a monopoly on nominating authority, or if they did not have such authority de jure but nevertheless enjoyed it de facto, the gravity – as well as the integrity, stability, and magnitude – of parties in Russia would be greater. If candidates for SMD seats in parliamentary elections were required, for legal or at least practical reasons, to rely on parties for nomination, parties would certainly be more important to politicians than they are in fact. They would also be more important to actors who seek to influence politicians. Recent changes in electoral rules, moreover, though prima facie rather minor, further weakened party influence in SMD races. Until 1999, candidates were required to gather a certain minimum number of signatures in order to register for participation in the election. The rule did not ensure a large role for parties, but it did require a certain degree of organization and mobilization in order to collect the required signatures. It sometimes gave candidates who enjoyed the backing of a party a bit of an advantage. In the 1999 elections, however, candidates were granted the right to register by putting up a security deposit in lieu of a list of signatures. In previous elections, candidates did in fact rely on money to fulfill the registration requirements. They sometimes supplied their signature-gatherers with stacks of small-denomination notes to hand out to passersby on the street who agreed to endorse the candidate’s registration. But at least candidates had to have a small team working for them to gather signatures. In 1999, however, candidates could simply circumvent the signature-gathering task if they could put up the money. As Turovsky (1999) has noted, this change boosted the proportion of wealthy businesspersons among candidates and diminished the importance of parties in the SMD races. If presidential candidates had to rely on parties, particularly given the supremacy of the presidency in Russia, parties’ gravity would be much greater. Political parties in the United States and Chile are consequential organizations not only because legislatures hold considerable power in the moderate presidential systems of both countries, but also because parties control the nomination of candidates for the presidency. The presence of a powerful parliament undoubtedly advances party development. Still, one can imagine a situation in which parties, if granted a monopoly over nominating authority for election to the presidency, may become strong organizations
Impact of the Elections on Political Party Development
203
even in regimes with weak legislatures. Such circumstances would encourage the rise of two predominant parties, because competition for the presidency normally encourages organizational amalgamation and a reduction in the number of parties. Still, parties in two-party-dominant systems may be and frequently are powerful and vigorous. Parties in the United States are sometimes regarded as weak and amorphous in comparison with their counterparts in some West European countries. But one can scarcely imagine the election of a president not affiliated with a party in the United States. In Congress, a total of two legislators are not members of one of the two major parties, and one of them is a leader of a third party. In recent American history, just two governors have been elected as independents. Because parties control nominating procedures in American elections, they are weighty and potent organizations. A starkly different situation obtains in Russia, where presidential candidates need not even associate themselves closely with, much less join, a political party. Indeed, parties have yet to play a major role in Russian presidential elections. Yeltsin’s and Putin’s own attitudes toward parties have been of considerable importance. Yeltsin could have chosen to build a party organization but consistently declined to do so. He always preferred the posture of the representative of “all the people,” unencumbered by partisan allegiance. Putin continued the tradition in 2000. The enervating effects of superpresidentialism on parties were thus intensified by the virtually nonpartisan character of presidential elections. In this matter, the personal preferences of the dominant and victorious candidates, as well as rules that allowed candidates to shun party affiliation, aggravated the influence of superpresidentialism. This state of affairs shows the aptness of Remington’s (1998, 214) assertion that “we should not assume that the impersonal laws of functionalist theory fashioned political parties by an invisible hand. Parties are politicians’ vehicles for solving a variety of tasks.” Neither Yeltsin nor, as of this writing, Putin chose to rely on parties to advance his own political ambitions. People, as well as the institutions they make, are of crucial importance to understanding the state of the party system. Fortunately, after many years in which human volition and intentions have been ignored or dismissed derisively as “voluntarism,” some scholars are coming to realize that human will and agency have a life of their own and influence political outcomes. Particularly because institutions are themselves artifacts – meaning human creations – an institutional approach is scarcely in conflict with one that also takes human agency seriously (Burgess 1997; Chhibber and Torcal 1997; Moser 1998; Fish 1999 [b]; Samuels 2003). In fact, Russian leaders’ own antiparty predilections were reinforced by the institutional environment in which they operated. Both Yeltsin and Putin could rely exclusively on the formidable resources of the presidential and governmental apparatuses; they did not really need an organizational weapon in society. Neither did they need one in dealing with other branches
204
M. Steven Fish
of government. As Ordeshook and Shvetsova (1997, 40) have argued, the presidency’s powers are so extensive that the president’s “relationship with all other branches of government is likely to be based simply on the constitutional authority of his office, which can only stifle the development of an integrated party system.” The president does not really need parties, even for election; nor does he need to engage in the kind of interbranch bargaining characteristic of executive-legislative relations in moderate presidential systems. Thus, the Russian presidency, in its character and its effects, differs from the U.S. presidency. The difference is one of character and type – not merely scope – of power. As Ordeshook and Shvetsova (1997, 36, 40) remark, the U.S. president enjoys relatively modest formal power, while the Russian president “has not only strong executive powers but extensive legislative and judicial powers as well.” Indeed, Articles 85 and 90 of the constitution grant the president powers of legislation and judicial control, while Article 91 furnishes immunity from prosecution. In sum, the superpresidency, combined with rules that do not afford parties anything resembling a functional monopoly over the nomination of candidates for public office, including the presidency, means that societal actors and leading politicians alike need take little interest in political parties. The timing of elections further exacerbates the debilitating effects of superpresidentialism. Specifically, the holding of parliamentary elections just six months before presidential elections casts a long shadow of the future over the former contests. Under a regime with a weightier legislature, such scheduling would not necessarily be of great moment. But under the superpresidential system, the schedule largely reduces the parliamentary contest to a first heat for the presidential election (see Chapter 10). All politicians, and indeed the media and the mass public, regard the parliamentary elections essentially as a prelude to the contest that really counts. After the party that he endorsed – although he did not even associate with it closely – fared extremely well in the parliamentary elections, Putin was assumed to be the front-runner for the presidency. Similarly, the poorer-than-anticipated performance of OVR knocked the party’s leaders, Yurii Luzhkov and Evgenii Primakov, out of the presidential race.4 As a first heat for the presidential race, the legislative election encourages intracamp rather than intercamp competition. In practice, the leaders of the major parties who harbor presidential ambitions do not have a strong motive to maximize their parties’ ` vote share vis-a-vis parties whose ideological orientation is starkly different from their own. Nor do they have reason to engage in organizational consolidation or cooperation with like-minded parties, nor even to maximize their own party’s vote as a share of the total vote. Instead, the timing of elections encourages party leaders who have presidential ambitions to maximize 4
“U Primakova otniali ‘Vsiu Rossiiu’”; “Vladimir Putin stal glavnoi politichesoi figuroi.” Nezavisimaia gazeta, December 22, 1999.
Impact of the Elections on Political Party Development
205
their own party’s vote share within a particular camp, even to the extent of undermining potential allies. The situation is illustrated clearly by the behavior of Yabloko’s Yavlinsky, whose sole aim since entering politics has been winning the presidency (Fish, 1997; Rutland 1999). In the 1995 and 1999 parliamentary campaigns, the invective that Yavlinsky directed against other liberal parties was far more shrill and uncompromising than that to which he subjected parties that were programmatically more distant from Yabloko. In 1995, Yavlinsky spent the better portion of the campaign denouncing the DVR and Forward, Russia (Vpered, Rossiia [VR]), two liberal parties that had a great deal in common with Yabloko and that were logical allies of Yabloko in the legislature. Yavlinsky hardly criticized blatantly antidemocratic parties. He was positively reverent in his relations with the Communists, many of whom would favor hanging him if they ever came to power. Yavlinsky’s aim was transparent: to emerge as the only viable liberal candidate for president after the parliamentary elections. The DVR and VR were led by figures who also had national stature and presidential ambitions, Egor Gaidar and Boris Federov, respectively. Thus, maximizing Yabloko’s share of the vote – after the point at which the party crossed the 5-percent barrier – was far less important to Yavlinsky than ensuring that neither the DVR or VR cross the 5-percent barrier and thereby establish a substantial foothold in parliament. Despite Yabloko’s mediocre performance (its total in the list vote fell from the 8 percent it received in 1993 to 7 percent in 1995), the result suited Yavlinsky well, since neither the DVR nor VR cleared the 5-percent barrier. The day after the 1995 parliamentary vote he announced that the field for the 1996 presidential election had now been definitively established. It consisted, he proclaimed, of the leaders of the four parties that cleared the 5-percent barrier, meaning himself, the LDPR’s Zhirinovsky, NDR’s Chernomyrdin, and the CPRF’s Zyuganov, as well as Yeltsin (Fish 1997). Yavlinsky was indeed the lone liberal candidate to stand for the presidency in 1996 (unless one also counts Yeltsin as a liberal). In the event, he was sorely disappointed, finishing with roughly the same percentage of the vote that his party won in the parliamentary elections. Yavlinsky pursued the same strategy in 1999, again lashing out primarily against other liberals, though this time Yabloko was bested by another liberal party, SPS. Yavlinsky has also dared not risk letting other leaders of his party upstage him and detract from his quest to raise his own profile on the eve of the presidential vote. In the 1995 parliamentary campaign, virtually no other leader of Yabloko ever spoke for the party in the electronic media, though Yabloko had a half-dozen other well-known leaders who might have enhanced the party’s appeal. In 1999, Yavlinsky brought Sergei Stepashin, a former prime minister, and Nikolai Travkin, a rough-hewn hero of the Gorbachev-era democratic movement, into Yabloko. According to Stepashin, Yavlinsky’s decision represented an attempt to shed the party’s reputation for organizational
206
M. Steven Fish
amateurishness and to “move from words and criticism to action.”5 But the campaign and its aftermath showed that Yavlinsky could not cede the limelight to his comrades. They were scarcely visible, either in the party’s advertisements or on the campaign’s endless talk shows. As in 1995, Yavlinsky spoke for the party. Yavlinsky’s behavior throughout the 1990s did not, to understate the matter, enhance the coherence of liberal forces in Russia.6 His refusal to cooperate closely with other prodemocratic parties, however, is understandable from the standpoint of his own presidential ambitions. As long as parliamentary elections are held six months prior to presidential elections, and as long as the superpresidential constitution remains in place, party leaders will face the urge to wreak dissension in the camp to which their own party belongs. They will want to thwart alliances between or amalgamations of like-minded parties that could threaten their pretensions to primacy within their own camp. Indeed, just as Yavlinsky has divided and weakened the liberals, so too has Zhirinovsky split and fractionalized the nationalists. Though many observers in the mid-1990s regarded Zhirinovsky as Russia’s great nation¨ alist godfather and Fuhrer-to-be, he actually has refused to allow the LDPR to cooperate with other nationalist parties. Like Yavlinsky, Zhirinovsky has viewed parliamentary elections as first heats for the presidential contest and has acted accordingly. Arguably, he has had the same enervating effect on the nationalists that Yavlinsky has had on the liberals. In short, the timing of the national elections, and in particular the holding of parliamentary elections just six months before the presidential contest, deadens party-system development. By encouraging intracamp rather than intercamp competition, it works against the combination of like-minded parties into larger, more substantial parties, thus counteracting stabilization of the party system and perhaps damaging party magnitude as well. By fostering personalism within parties, it increases parties’ dependence on the whims and fortunes of a single individual and may therefore check the growth of internal coherence, or integrity, of parties (Mainwaring 1998). Staggering parliamentary and presidential elections in a manner that scheduled the contests two years apart would help prevent the conversion of the former into a mere first heat for the latter. Such scheduling, in the absence of some shift of power from the president to the parliament, might not make a large difference in party development. Still, it might encourage both politicians and voters to think of the parliamentary elections as events in and of themselves and in so doing might discourage the type of behavior described previously 5 6
“Principle Still Yabloko’s Priority.” The Moscow Times, December 22, 1999. “Principle Still Yabloko’s Priority”; “Kiriyenko Once Again Brings Political Surprise.” The Moscow Times, December 22, 1999; “Ch’i vkusnee pirogi.” Tribuna, March 24, 2000; “Grigorii Iavlinskii: ‘Rossiiskoe ekonomichechoe chudo vozmozhno’.” Tribuna, March 24, 2000; Fish 1997.
Impact of the Elections on Political Party Development
207
and contribute to intracamp solidarity rather than dissension. Holding parliamentary and presidential elections simultaneously might also be superior to the current system. According to Ordeshook and Shvetsova (1997), holding simultaneous elections in Russia might enhance the significance of parties for presidential and legislative, as well as perhaps gubernatorial, races. Either staggering elections at wide intervals or holding them at the same time would help mitigate the pathology of the current system, in which parliamentary elections are reduced to first heats for the presidential contest. To sum up: The superpresidential regime is the institutional root of all mischief in Russia’s underdeveloped party system. It is the source of the party system’s immunity to the potentially felicitous effects of electoral competition. Russia provides ample support for Linz’s (1994, 35) thesis that pure presidentialism “is structurally in conflict, if not incompatible” with partysystem development. American-style moderate presidentialism, however, is not to be confused with Russian-style superpresidentialism – a distinction to which Linz does not always pay adequate heed (Linz 1994; Linz 1996). Electoral rules that do not furnish parties with a functional monopoly over the nomination of candidates, as well as the schedule of legislative and presidential elections, exacerbate the baneful effects of superpresidentialism in Russia. The choices of several highly influential politicians also cannot be ignored: Neither Yeltsin nor Putin, the dominant political figures of the postSoviet period, took interest in party building. Their decisions cannot be attributed to institutional factors alone, as both or either could have chosen to build a party for the sake of equipping themselves with an organizational weapon in broader society. Even here, however, the superpresidency exerted its effects. As the incumbent in 1996, Yeltsin could rely on the sprawling executive apparatus and its overwhelming financial and organizational resources in his reelection bid, and therefore probably did not regard the strong support of a party as a necessity. Putin, by virtue of his status as Yeltsin’s chosen successor and acting president, faced a similar situation in 2000. The present analysis illuminates the fallaciousness of the assumption that a presidential system of any type, with its “winner-take-all” logic, necessarily provides a stimulus to party “consolidation.” In the language of this chapter, “consolidation” would be associated with magnitude, stability, and integrity. Russia’s superpresidentialism, however, has contributed to none of these three qualities. Presidential elections have provided a “consolidating” effect in Russia only insofar as they have included the participation of candidates from fewer parties than have competed in parliamentary elections. The “consolidating” effect of presidential elections may be understood strictly as a reduction in the number of organizations, not a stimulus to organizational maturation and growth through amalgamation and integration. Politicians do have some incentive to abandon smaller organizations and groupings after the first round of elections and stand behind one of the two leading candidates in the event of a second round, such as took place in the 1996
208
M. Steven Fish
presidential contest. Naturally, being associated with the winning side can be beneficial to politicians and other actors who seek wealth and power. But under the conditions of the Russian-style superpresidency, a desire to be on the winning side, or part of the winning machine, does not necessarily involve, much less require, membership or activism in a particular party. Yeltsin did not apportion a major share of top positions in his government after coming to power in 1991 to leaders of the parties and proto-parties who had been his staunchest allies (Colton 1992, 18–21; Fish 1995 [b], 201–10). Nor was leadership or membership in Unity in any way a prerequisite to gaining high office in the executive branch or currying the favor of the chief executive following Putin’s election. The institutionalist explanation presented here stands up well in comparative perspective. The other countries in the post-Communist region that have superpresidential constitutions, Kyrgyzstan, Armenia, and, arguably, Ukraine, also have exceedingly weak party systems, even by postCommunist standards (Fuller 1996; Collins 1997; Dudwick 1997; Huskey 1997; Prizel 1997; Way, Shaiko, and Tucker 2000). Even countries that have suffered leaders and governments with arbitrary and authoritarian tendencies, but that do not have superpresidential constitutions, including Slovakia, Bulgaria, Georgia, and Macedonia, have fared better in terms of party development than has Russia (Perry 1997; Fish 1999 [a]; Fish 1999 [b]). In other regions as well, superpresidentialism stunts the development of parties. Three paradigmatic cases of superpresidentialism, Peru, Brazil, and the Philippines, have feeble parties, and the superpresidency has clearly undermined the possibilities for parties’ growth despite democratization (Mainwaring 1998; Levitsky 1999; Montinola 1999). Even if the link between superpresidentialism and party-system weakness drawn out here is sound in logical and empirical terms, one could argue that superpresidentialism is the regular, predicable product of some other factor; that is, it should be regarded as “endogenous.” The one most frequently adduced by Russians as well as by many Western scholars is the desire for a strong hand, a yearning that is supposedly deeply ingrained in Russian political culture (Brovkin 1996). One need not subscribe to a crude culturalist perspective to regard superpresidentialism as consistent with Russian political culture. Such a notion is, of course, hard to disprove. But two big facts oppose it. The first again arises out of comparative analysis. It is extraordinarily difficult to generalize reliably about the origins of constitutions. What cultural factor can account for why the Kyrgyz Republic, Russia, and Armenia adopted superpresidentialism, while Moldova ended up with moderate presidentialism and Mongolia with semipresidentialism? Are the Kyrgyz, Russian, and Armenian cultures really less favorably disposed to a division of power among central state agencies than the Moldovan and Mongolian cultures? An affirmative answer is not immediately apparent. Nor can superpresidentialism be seen merely as a functional response to the
Impact of the Elections on Political Party Development
209
breakdown of state agencies and a vacuum of power. Some leading scholars have suggested that executives were invested with overwhelming powers, as in Russia and Kyrgyzstan, or with dictatorial powers, as in Uzbekistan and Belarus, due to the functional need for an agency capable of reestablishing some semblance of a state and public order amid conflict-ridden societies and crumbling institutions (Easter 1997; Frye 1997). But the breakdown of the state was no less traumatic or extensive in Moldova and Mongolia than in Kyrgyzstan, Russia, and Armenia. In fact, constitutional choice in the post-Communist region has depended on a host of factors. Examination of constitutions and the conditions under which they were adopted show that it is extremely difficult to generalize about constitutional choice in the post-Communist region. The usual theories about the primacy of culture, the condition of the state, or the interests of politicians do not hold up well under close scrutiny. Highly concentrated executive power also cannot be readily attributed to the weakness of civil society, though the idea that the dispersion of power reflects or results from a potent civil society at the time of regime change is intuitively appealing. One might expect that power would be dispersed where strong pressures from below demanded that it be dispersed, or when contending portions of a mobilized society clashed over fundamentals and compromised by dividing power among agencies over which rival forces intended to exercise control. The Polish case furnishes some support for this notion. Polish society was mobilized and politicized at the time of regime change, and Poland did end up with a regime that separated powers. But other societies that also ended up with constitutions that separated power did not have well organized or highly mobilized societies at the time of regime change. Romania, Moldova, and Mongolia certainly did not. Societies in these countries, while restive and resurgent, were arguably even less articulate and muscular than Russian society at the time of regime change, and yet the former three countries ended up with constitutions that separated power. In fact, it is important to remember that superpresidentialism in Russia was created by the prodemocratic movement, not by people yearning for stronger hands or more ferric fists. Its consequences were not intended by many of those who supported it. To be sure, the institution enjoyed widespread public backing. The event that created the Russian presidency was a referendum held in March 1991, and the one that codified the superpresidency was a referendum held in December 1993. But in the first instance, support for a presidency was primarily a vote against the Soviet system and in favor of a more open regime. Voters knew well in the spring of 1991 that Yeltsin’s popularity and authority were unmatched in Russia and that a vote for establishing a presidency was a vote for the then-leader of the anti-Soviet movement. All of Russia’s democratic movement organizations supported the creation of a presidency. In the second referendum, citizens
210
M. Steven Fish
made their choice in the wake of a bloody confrontation between President Yeltsin and the Soviet-era legislature that Russia inherited from the pre-1991 period. The struggle had pitted Yeltsin and his government against a group of parliamentarians who were led largely by politicians associated with nationalism and nostalgia for the Soviet regime, backed by demonstrators in the streets who showed no compunctions about resorting to violence. To note this fact is not to excuse Yeltsin’s high-handed behavior in the weeks that led up to the tragic clashes in Moscow. It is merely to point out that the voters who supported Yeltsin’s favored draft of the constitution, which was put to a yes-or-no vote simultaneously with the December 1993 parliamentary balloting, were not necessarily endorsing overweening executive authority. It was not at all clear at the time that a “yes” vote represented a blessing for a father figure or a desire for a strong hand. The issue at hand in late 1993 was not a straightforward case of institutional choice. Above all it involved taking sides in an ideological war.
prospects and prescriptions Russia’s parties seem to be caught in a vicious circle. Politicians and powerful societal actors will not take a serious interest in parties until parties become more important, but parties cannot become more important until politicians and powerful societal actors take a more serious interest in them. Russia’s parties are indeed stuck in this quandary. But it is not difficult to imagine how certain institutional changes could alter the prospects for party development. If the parliament had more power and elections were held exclusively on the basis of PR, every relevant political and economic actor in Russia would find himself or herself deeply interested in parties overnight. Were ` the Duma invested with more power vis-a-vis the presidency, many political leaders and economic actors would remain intensely interested in parties, but the constitution that went into effect at the end of 1993 depressed the growth of incentives for investment in parties. Proportional representation can and does enhance elite interest in parties, but if the legislature is not powerful, party-list elections for parliament alone will not necessarily spur party development. According to the logic of this chapter’s argument, three institutional changes would ensure that the elections of 2003 and 2004 provide a stronger stimulus to party development than did previous contests. Unfortunately, none of these changes appears to be in the offing. A major crisis might be necessary to create the possibility for the needed reforms. The argument pursued in this chapter, however, also leaves open the possibility that human agency has a life of its own, and sometimes is not fully explicable in terms of historico-cultural, socioeconomic-structural, or even institutional determinants. Therefore, the president’s own unpredictable actions may alter conditions.
Impact of the Elections on Political Party Development
211
Institutional change of course would provide the surest way to brighten the prospects for party development. The first auspicious reform would be the conversion of the superpresidency into a moderate presidential system or some other form of regime. Movement to a moderate presidential or semipresidential regime would require empowering the legislature, circumscribing the president’s decree powers, reducing the executive’s control over the public purse, creating mechanisms for parliamentary oversight over the executive branch, and enhancing the autonomy of the courts. Such changes would require altering the constitution. Prospects for such reform appear to be more remote at the beginning of the 2000s than they were in 1998–9. Putin’s election greatly diminished public and elite interest in curtailing presidential powers. Controversy over reducing presidential prerogatives gave way to debates over whether and how to expand the president’s powers. As of this writing, relief at having a vigorous president has quieted discussion over strengthening the legislature. Granting parties a monopoly over the authority to nominate candidates for public office similarly is not a topic of extensive discussion, though some of Putin’s subordinates have hinted at the possibility of reform in this area. Neither is restructuring the political season to hold legislative and executive elections simultaneously, or to stagger the contests, rather than holding the former on the eve of the latter, a substantial part of public discourse. The near-term prospects for reform seem slight. Paradoxically, the best hope for reforms might lie with the president. Putin probably will not seek to reduce the powers of his own office. Even though he would benefit immensely in strictly personal terms from a system that reduced his own formal powers and thus enabled him to share blame and responsibility with the legislature, he cannot be expected to recognize this fact. Despite all evidence, chief executives rarely see that, beyond a certain point, more is actually less when it comes to the power of their own offices. Occasionally one might glimpse this reality. The vast majority of chief executives, however, regardless of their political complexions at the outset of their tenure in office, ultimately act more like arational gluttons for power than like rational utility maximizers or even level-headed optimizers or satisficers. Why this is the case poses an important challenge for rational choice approaches to politics and for political psychology. At any rate, Putin will most likely act like Yeltsin and most other leaders and will seek to aggrandize his own office even if doing so compromises his reputation, popularity, place in history, personal security, and ability to govern effectively. But even if Putin follows the well-worn and self-destructive path of blind self-aggrandizement, it is conceivable that he will pay parties more heed than did his predecessor. Putin took pains during his electoral campaign and in his early months in office to distance himself publicly from the specific actions of Unity, the party he endorsed in the parliamentary elections. Yet he may come to see the value of building up the organization. Yeltsin’s refusal
212
M. Steven Fish
ever to build a party of his own left him with no vehicle in society for defending himself and burnishing his reputation when his popularity began to decline and with no organizational structure outside the agencies of the central government to bolster implementation of his policies. Putin currently enjoys two assets that Yeltsin lacked at the time of the change of presidents: extensive public support and intimate personal control over the agencies of state security. But Yeltsin also enjoyed immense popularity early in his tenure, and Putin is well aware that astronomical approval ratings are rarely durable. Putin may see Yeltsin’s neglect of parties as the mistake that it was. If so, he might regard a powerful party as a potentially useful instrument in his arsenal of organizational weapons. In such a case he may be amenable to reforms, such as granting parties a functional monopoly over nominating authority, which would advance the development of the party system. One certainly cannot yet assume that Putin sees great value in party building. But his decision early after the parliamentary elections to bless a de facto condominium arrangement between Unity and the CPRF showed that Putin does know a serious party – in this case, the CPRF – when he sees one. He may even recognize parties’ potential significance for the future of his own program for change.
10 Resolving the Problem of Preelection Coordination: The 1999 Parliamentary Election as an Elite Presidential “Primary”1 Olga Shvetsova
The following seemingly shocking commentary appeared four weeks after the 1999 election to the Russian parliament, the State Duma, in one of the country’s leading elite periodicals. Yet, what it was saying by then had become the conventional wisdom and could hardly surprise any readers. The newspaper recounted the story of one of the main contestants of the 1999 election the following way: As recently as three months ago, no one had any doubts that OVR [Fatherland-All Russia] would receive a majority in the State Duma. But now, even though that alliance has managed to get into the Duma, the doubts about its ability to form an independent faction there are well justified. (Sanko 2000, 3)
The fate of Fatherland-All Russia (OVR) which was, indeed, an unquestionable front-runner as late as three months prior to the election is but one example of the sharp turns that the fortunes of parties and political personalities are known to take in Russia. With more than ten years of experience in electoral politics, however, should we not expect more predictability and continuity in the political system? Or is there a good explanation for the persistent alterations of the political landscape and the unpredictable party realignments? The answer to the puzzle lies in adjusting our perception of the objectives of the players behind parties. I suggest that the conduct of the parliamentary campaign and the process of elite endorsement of the major contestants were driven by a complex mix of goals in which subsequent access to the presidency was a weighty component. Much of the political work in the1999 election to the Russian Duma focused on resolving what I describe in the following text as a coordination problem among the Russian elites in anticipation of the presidential election scheduled to follow the 1
This research was in part supported by a grant from the National Science Foundation (SBR9730275) and by the Center in Political Economy at Washington University in St. Louis.
213
214
Olga Shvetsova
parliamentary race. The game played by the non-Communist political elites with regard to filling the all-important presidential post had multiple possible “equilibria,” and coordination was required to converge on one of them. When assessed as a step toward the presidency, shifts and realignments in the parliamentary campaign, which so heightened the uncertainty regarding the prospects of individual parties and politicians, make sense. But such a conceptualization links the observed volatility in party labels, leadership, and electoral outcomes with the institution of the presidency, suggesting that political volatility in Russia may be a condition that will persist in the long run and seriously affect the developing party system and other political institutions.
multiple elections and coordination That there should be a strategic connection between presidential and parliamentary selection has long been acknowledged in the literature, mostly with the focus on the presidency’s impact on party fragmentation. Shugart and Carey (1992) suggest that when the two elections are held concurrently, the institutional inducements created by one (presidential, conducted in a single-member district [SMD]) influence political behavior in the other, altering the relationship between parliamentary electoral rules and party systems. Specifically, consistent with Duverger’s law (1954), Shugart and Carey (1992, 220–2) expect presidentialism to lower the number of political parties generally, thus to lower their number in parliamentary elections. This proposition is put to a rigorous testing by Neto and Cox (1997, 158–64) and Cox (1997, 210), who do find support for the hypothesis “that legislative elections that are highly proximal to presidential elections should have a lower effective number of parties” (Cox 1997, 213). Yet in the post-Communist wave of democratic transitions, the expanded list of the world’s presidencies has offered evidence that the discernible impact of presidential elections on parties was not necessarily consolidating. Filippov, Ordeshook, and Shvetsova (1999) argue that popular presidential elections contribute to fragmentation in transitional party systems and confirm this on the entire cross-section of post-Communist democracies; they also observe that presidential elections held concurrently with or close in time to parliamentary elections do not reduce post-Communist multipartism. Instead, they find some indications that the concurrency variable may have initially had the opposite effect in those countries.2 2
The findings that the consolidation effect is not operational in post-Communist presidential systems may result from the use of the majority runoff mechanism rather than the plurality rule. Carey (1997, 72) finds the candidates’ vote to be relatively more fragmented in first rounds of majority runoff elections compared to plurality presidential elections. All popularly elected post-Communist presidents to date had been elected by a majority runoff, and a relatively higher electoral fragmentation could be expected.
Resolving the Problem of Preelection Coordination
215
Regardless of the direction of the presidency’s effect on party fragmentation, the Russian case strongly suggests that looking forward to a presidential election may change the strategies and the logic of the parliamentary campaign: when the presidency ultimately is at stake, rational politicians may behave inconsistently with their stated parliamentary goals. Thus, elites in Russia in 1999 rushed to get on the bandwagon of a newly created party with no policy platform, ideology, or organizational strength – once they were convinced that the party was associated with a strong presidential contender. Because the 1999 parliamentary and 2000 presidential elections were scheduled in close proximity and a strong possibility existed that the presidential election could be called even earlier due to Yeltsin’s illness, it was not surprising that major players viewed the two campaigns as a continuous process. Presidential calculations modified strategies of candidates and elites who supported them enough to make the Duma race less about policy or ideology of parliamentary parties and blocs and more about the likely presidential candidates – former Prime Minister Yevgenii Primakov, Moscow Mayor Yurii Luzhkov, and Prime Minister Vladimir Putin. The parliamentary election turned into an elite presidential primary. Pro-Kremlin and pro-Putin forces, on their part, did not conceal that their goal in the Duma campaign was to inflict maximal damage not on their sworn ideological enemy the Communist left, but on their closest ideological ally party of Primakov and Luzhkov, OVR.3 The main target of negative campaigning in 1993, 1995, and 1996, in 1999 the Communists were ignored by the Kremlin-controlled media, which concentrated instead on trashing OVR.4 Key observers clearly saw the fight as entirely about the presidency, admitting that there were no ideological or policy issues dividing OVR and Unity (Tsipko 2000);5 and the “war of kompromats” between the two blocs had stopped as soon as the fate of the presidency was sealed in the election. According to television personality Sergey Dorenko, “The preelection show had succeeded and Luzhkov and Primakov could not be considered viable presidential candidates any more” (ORT, December 19, 1999, 3
4
5
Vyacheslav Nikonov, who worked as a political consultant for Yeltsin’s 1996 presidential campaign and for Fatherland in 1999, compared methods used against Fatherland-All Russia with the “dirty tricks” in the Kremlin’s campaign against the Communists in 1996 (Federal News Service, December 12, 1999). According to the European Institute of the Media (EIM), the campaign coverage by ORT and RTR, state broadcasters with the greatest audience reach in the country, “was heavily biased against the Fatherland-All Russia alliance and biased in favor of the pro-government Unity.” The European Institute of the Media. 1999. Preliminary Report on Monitoring of Media Coverage during the Parliamentary Elections in the Russian Federation in December 1999. Available at http://www.eim.org/ru-prelims.html. Content analysis by Carnegie Moscow Center of economic programs of Fatherland-All Russia and Unity over ten issues and two dimensions found absolutely no differences between the two parties (Dmitriev 1999).
216
Olga Shvetsova
http://www.intergrumworld.com). Primakov left the presidential race shortly after the Duma election. While many expected the parliamentary vote to give presidential contenders an indication of where they stood in the race, in reality, it did much more, effectively deciding the outcome of the presidential election. Immediately after the vote, most of the regional leaders associated with OVR resolved that it was time to announce their loyalty to Putin, and All Russia broke away from the presidency-bound Fatherland. Yaroslavl Governor Anatolii Lisitsin was most frank in explaining the reasons for the split. “We governors are pragmatic people and will never support a hopeless presidential candidate who can get as little as ten percent of the vote,” he told the reporter of Kommersant-Daily four days after the parliamentary ballot (Tregubova 1999 [a]). A host of unique circumstances strengthened the connection between the two campaigns: Yeltsin was ill and no obvious figure had emerged as a presidential incumbent, the Chechen war enabled Putin to cash in on his growing popularity, and some fearful business “oligarchs” were frantic to prevent Primakov from winning. However, two institutional factors, strong presidency and weak parties, make the strategic connection between parliamentary and presidential elections an essential characteristic of the Russian political system. It is their combination that makes presidential office the ultimate political prize. Sartori (1997, 138–9) places Russia at “the extreme edge” of the semipresidential category, while others characterize the Russian model as superpresidential, dominated by the executive branch (see Fish 2000),6 where weak and fragmented political parties are unable to influence the president or affect his reelection.7 Because parties lack political influence and organizational resources for mobilizing a voting coalition broad enough to win the presidency, serious presidential candidates must form ad hoc alliances with political actors who can provide campaign resources and mobilize the electorate. Thus, it is widely believed in Russia that business elites and regional leaders are among those who control the media and influence the vote. They were credited with the success of the 1996 presidential campaign when Yeltsin was successfully reelected despite scoring below 5 percent in a popularity poll just six months prior to the ballot (McFaul 1996). 6
7
A well-established tradition distinguishes parliamentary, presidential, and semipresidential regimes based on the constitutional description of formal presidential prerogatives (see Chapter 1 for a detailed account of the powers of the Russian president). Formal prerogatives are easier measured than the informal powers, but they do not capture the office’s role in full, especially in a democratic transition when institutions are weak and undergo adjustments (Frye 2000). After the 1993 election, Sakwa (1995 [a], 194–5) described Russia’s party system as “atomized, with every category contended by numerous groups in a constant process of formation and separation,” reflecting “not a profound social cleavage (although there is no shortage of these) but the struggle of a relatively small elite.” The description remained largely accurate for the 1999 campaign.
Resolving the Problem of Preelection Coordination
217
But in 2000, business and regional elites themselves encountered a problem of coordination when they sought to concentrate their efforts on the same (and, therefore, the strongest) candidate. The problem of coordination is quite common in multicandidate races. It arises, for example, when trying to avoid wasting a vote on a losing candidate. When in a majoritarian race multiple candidates participate in the first round, and then a runoff is held between the two front-runners, from a strategic point of view, a vote in the first round for anyone other than those two is wasted. In the Anglo-American plurality system no second round is held, and the coordination problem is even more apparent. Indeed, one can interpret party primaries as a mechanism for precoordination on the two front-runners that precedes the actual vote and prevents the vote fragmentation that could take place if candidates from the same party ran against each other on the actual ballot. With regard to the Russian presidency (where the runoff is used), the problem of coordination among elites was not so much about wasting the votes or preventing fragmentation in the non-Communist camp. It centered on supporting the “right” candidate, recognizing from the start the future presidential winner. And even though after the fact the outcome of successful coordination (in this case, President Putin) might appear obvious, it did not until fairly late in the campaign, and, in fact, the efforts to engender coordination to that end were substantial.
the game of presidential coordination There are potentially severe penalties that an eventual winner of Russian presidency can impose on those who backed his election opponents (even though the fortunes of all candidates including the winner depend on the support of business and regional elites). Placing one’s bets the wrong way could be even costlier politically than to be in the opposite ideological camp (see Chapter 6 on the levers the Russian president has to ensure elite compliance). Divided as they may be in their preferences over possible presidential candidates, elites first and foremost want to support the future winner, whoever he turns out to be.8 A generic example of such a situation is given by the 8
Ironically, once he is the object of coordinated electoral support, a presidential candidate becomes independent from individual members of the elite organizations that helped propel him to that status. Thus, after the non-Communist “nomination” was clearly won by Putin, even the most powerful regional bosses found themselves in total dependence on the Kremlin. It was enough for an acting president Putin to tell journalists about his intention as the elected president “to declare war on the legal chaos in the regions” on January 31, 2000 (Segodnya, February 1, 2000) to cause Bashkortostan President Murtaza Rakhimov, one of the most influential regional leaders, to sign a decree four days later establishing a special commission to coordinate Bashkortostan’s laws with the federal constitution. Later, on March 23, Bashkortostan accepted another blow – it agreed to funnel republican budget money through the federal treasury (Russian Federation Report by RFE/RL, February 9, 2000 and March 29, 2000. Available at http://www.rferl.org/russianreport/).
Olga Shvetsova
218 Majority of Other Governors Support Primakov
Majority of Other Governors Support Putin
To Back Primakov
A positive payoff
A negative payoff
To Back Putin
A negative payoff
A small positive payoff
figure 10.1 Individual-Level Problem in Coordination on a Presidential Candidate.
game of the Battle of Sexes where two players disagree about the most desirable plan of action but definitely want to anticipate the actions of each other and coordinate on the same outcome (Fudenberg and Tirole 1991, 18–20). From the point of view of an individual decision maker, for example, a governor, the situation in the Battle of Sexes with many players appears as in Figure 10.1. Assuming that the decision maker prefers Primakov to Putin, his choice of whom to support would depend entirely on what he believes the other players are going to do. If a majority of them back Primakov, he, as well, would want to back Primakov. If a majority of others support Putin, he would support Putin. Theoretically, when the others are divided evenly and his choice is decisive, the governor would follow his own preference. A clear-cut “tie,” however, is unlikely under these circumstances because there are many players and because the process described in Figure 10.1 is not about the actual vote for the candidates but about the elite support for him in the campaign. A governor whose decision problem is described by Figure 10.1 knows that everyone else in the game faces an essentially similar choice except for a possible difference in their preferences over the two candidates; selecting between the resulting two equilibria amounts to coordinating with other players. Shelling (1960) argues that in games with multiple equilibria any signal that tends to focus players’ attention on any one equilibrium may lead them to select that equilibrium. Such a signal becomes a focal point of coordination. Thus, in order to shift from one equilibrium (e.g., “Primakov”) to another (“Putin”), all that is needed is a change in the beliefs about the intended actions of the others – some signal that would tip over coordination in the direction of a new winner. When obvious focal points for directing the vote exist (such as some commonly available information), coordination can be attained in a predictable way. For example, the Communist Party of the Russian Federation (CPRF) has a focal advantage among the Russian left of being an “old” party with
Resolving the Problem of Preelection Coordination
219
an extensive network of active local organizations. A swing of the bulk of the leftist vote to a different label, thus, is unlikely for as long as the CPRF remains intact. New entry is discouraged and, if it occurs, would not lead to major volatility within the voting block. Consequently, with a dominant player in the picture, the problem of coordination on a single candidate is not very pronounced among the left elites as nomination battles are largely contained within the party organization. Another possible focal point is the incumbent status of a politician. In 1996, unpopular as he was in the beginning of the campaign, as an incumbent Yeltsin was an obvious choice for coordination, and the Russian elites quickly agreed among themselves to support him in the presidential election despite his low public-approval ratings, history of frequent illnesses, and erratic behavior. But in 1999 there was no incumbent figure for the forthcoming presidential election. In the absence of an incumbent, the parliamentary campaign prior to the presidential race provided the institutional setting to achieve necessary coordination. The parliamentary campaign served a dual purpose in its role as a presidential “primary.” It became an arena for business elites and regional leaders who sought to support the “party of power” to identify such a party and its candidate. It also gave elites an occasion to formally join a coalition in support of the future winner – by means of lining up behind his parliamentary organization. To the candidates, in turn, this created an opportunity to contest the presidency by trying to sway in their direction a critical mass of elite supporters, specifically, the governors, knowing that once that was accomplished, the rest would follow. The need for coordination among the political elites and their attempts to resolve the problem of coordination was what has led to an all-or-nothing pattern in party performance, volatile swings in candidates’ fortunes, and opportunistic shifts of elite resources to backing whoever emerges as the more likely winner within the group. In this light, the next section discusses how looking forward to the 2000 presidential election influenced the course and the outcome of the 1999 parliamentary campaign.
the circumstances and the players While quite a few post-Communist democracies have party systems affected by both parliamentary and presidential institutions (e.g., Georgia, Moldova, Poland, Romania, and Ukraine), Russia in 1999 was a particularly interesting case of presidential incentives intruding into the parliamentary race. Several circumstances combined to provide a uniquely crisp correspondence of the case with the elite coordination model. The parliamentary campaign opened without an incumbent figure for the subsequent presidential contest. Since 1998, it had been clear that Yeltsin would be unable to compete in 2000. And though the public welcomed
Olga Shvetsova
220
table 10.1 Estimated Runoff Vote for Leading Potential Presidential Candidates Against Zyuganov
Primakov Luzhkov Stepashin Yavlinsky
Against Primakov
Percent of Vote for Centrist Candidate
Percent of Vote for Zyuganov
Percent of Vote for Other Candidate
Percent of Vote for Primakov
50 46 38 36
30 35 36 36
– 38 27 27
– 41 47 48
Source: VTsIOM poll of June 18–22, 1999. The numbers do not sum to 100 because of the “cannot answer” responses.
each new prime minister as his most likely successor, too many cabinets had changed in the course of the two years before the election to make anyone an unquestionable front-runner. No focal point existed upon which coordination to a single non-Communist candidate could be based. No “party of power” was in sight at the start of the race. Second, a known challenger, Communist Party leader Gennadii Zyuganov, was practically guaranteed a place in the runoff. Opinion polls indicated that the block of voters who sympathized with the Communists was both stable and sizable (around one-third of the vote) so if there were to be a runoff, a Communist candidate would be in it. Third, due to the stability of ideological divisions in the electorate, a Communist in the second round could be expected to be defeated by any of several centrist politicians.9 Moreover, should they get to the second round, centrist candidates would likely fare better in competition with a Communist than with a fellow centrist. In July of 1999, a poll by the Russian Center for the Studies of the Public Opinion (VTsIOM, July 23–8, 1999) showed that even at that early date at least four centrist politicians were capable of beating Zuganov, in a two-candidate race (see Table 10.1). Those were Moscow Mayor Yurii Luzhkov, former Prime Ministers Yevgenii Primakov and Sergei Stepashin, and the leader of an opposition proreform party Grigorii Yavlinsky. Among those, Primakov was also winning against any other centrist. Interestingly, when Vladimir Putin was appointed as a prime minister in August of 1999, he also had a shot against Zuganov, even though he was completely unknown as a politician. VTsIOM polls in August (VTsIOM, August 20–4, 1999) showed that only 2 percent of voters were willing to vote for Putin in the first round (rising to 4 percent in September [VTsIOM 1999]). 9
In other words, a Condorcet looser (a candidate who is defeated by any other in a pairwise comparison [Myerson and Weber 1993, 112]) was to participate in the runoff.
Resolving the Problem of Preelection Coordination
221
However, if paired against Zyuganov in the runoff, Putin would have been supported by 30 percent of the voters while 54 percent would have supported Zyuganov. The fact that many centrist politicians were predicted to be able to beat a Communist implied that the future of the Russian presidency depended on which one of them would become the recognized candidate of the non-Communist majority. This created fragmenting incentives for the non-Communist parties (see Filippov et al. 1999) and a clear coordination problem among the non-Communist elites. Because the level of electoral uncertainty with regard to the presidential race was so high, a number of the centrist politicians who announced their intentions to run for the presidency were considered serious contestants at the onset of the parliamentary campaign. That prompted the launch of new parties and blocs built around presidential hopefuls (see Chapter 9)10 and impeded differentiation among contestants on major issues. The proliferation of contestants on the “supply” side of the electoral equation was initially accompanied, consistent with the coordination hypothesis, by the relatively uniform spread of support for them among regional and economic elites. Most regional leaders were divided between three blocs – OVR, Our Home is Russia (NDR), and Voice of Russia. Prior to the campaign, Voice of Russia had claimed to have the support of thirty-three governors, NDR had twenty-one governors as members of its national committee, and OVR counted twenty-seven governors (Grigorieva 1999; Mulin 1999; Ulianova 1999). The support for Unity, which entered the picture at the end of September, fluctuated between ten and fifty-one governors. Most governors and republic presidents moved at some point between blocs or endorsed more than one bloc at a time. Almost all of them had been active members of the previous “party of power,” NDR.11 While many abandoned NDR, several of its members remained active in it and even joined its federal list while also publicly supporting the alternatives (e.g., Vyacheslav Pozgalev of Voronezh, Dmitrii Ayatskov of Saratov, Gennady Igumnov of Perm, and Vladimir Elagin of Orenburg). Some governors demonstrated remarkable flexibility in their search for potential allies in the future government. For example, the governor of Perm, Gennadii Igumnov, was active in all real and potential “parties of power” and never left any of them. First, he participated 10
11
Politicians easily circumvented the formal restriction that prevents parties registered less than a year before the election from running on the ballot by turning to older but formerly insignificant parties. This is how Unity was created; a bloc was registered by bringing together several obscure old parties under a new name. The most influential among them were the governors and republican presidents Anatolii Tyazhlov (Moscow Region), Anatolii Lisitsin (Yaroslavl Region), Gennadii Igumnov (Perm Region), Anatolii Yefremov (Arkhangelsk Region), Yurii Spiridonov (Komi Republic), Murtaza Rakhimov (Bashkortostan), Aslan Dzharimov (Adygea), and Ruslan Aushev (Ingushetia).
222
Olga Shvetsova
in NDR, then, additionally, in both Voice of Russia and All-Russia, was also active in forging a coalition between All-Russia and Fatherland, and, finally, joined the organizers of Unity.12 Because regional leaders and business elites could not afford not to endorse a future president, they were constantly searching for signals, “focal points,” that would tell them whom to back in coordination with each other. According to several public accounts (Ofitova 2000; Pravosudov 2000), governors repeatedly asked the Kremlin to select one party as a new “party of power” and, most importantly, to identify a presidential candidate whom the Kremlin would support. Thus, according to the president of Bashkortostan, Murtaza Rakhimov, ever since 1998, the governors kept offering to help the Kremlin create a new “party of power.” Initially, the administration had seemed to like the idea, as the cabinet minister Sergei Shoigu took active interest, but “suddenly, shortly before the election, Shoigu had stepped aside” and regions were left without a clear guidance. By Rakhimov’s account, regional leaders attempted to discuss the problem with Oleg Sysuev and Alexander Voloshin (Heads of Yeltsin’s administration) five times, and finally got an approval to proceed with the creation of the All-Russia bloc. “And now,” explained a disappointed Rakhimov on the eve of the presidential election, “we ended up in such a stupid situation,” supporting the losers, Luzhkov and Primakov (Ofitova 2000, 1). The former Prime Minister Sergei Stepashin, who ended up running for parliament on a Yabloko ticket, argued that the alliance between Fatherland and All-Russia would not have been created in the first place if the Kremlin could have guaranteed cabinet stability to the governors. Stepashin recalled that, while meeting with Yeltsin and Voloshin during the summer, eighteen governors, mostly All-Russia members, agreed to be “under” Stepashin in the forthcoming election, but wanted to know for how long Stepashin would keep his office. No answer was provided. Nonetheless, Stepashin’s attempted coalition, The Club of Governors, was, according to Stepashin, endorsed by fifty regional executives (Pravosudov 2000, 1). Meanwhile, by August 4, Moscow mayor Yurii Luzhkov and his Fatherland movement had won to their side All-Russia – an organization consisting of a number of influential governors – by promising to form an electoral alliance with the popular former Prime Minister Yevgenii Primakov. Five days later, quite possibly for the failure to prevent that alliance, Yeltsin dismissed Stepashin and nominated Vladimir Putin as prime minister. With Stepashin out and Putin in and with the advent of a Duma campaign, a coordination game arose that effectively constituted presidential primaries among the Russian elites. It can be described in a stylized way 12
According to Alexei Golovkov, the head of Unity campaign headquarters, governors had a tendency to instruct one of their deputies to affiliate with Fatherland and another one – with Unity (Victorova 1999).
Resolving the Problem of Preelection Coordination
223
as follows: Based on a combination of their popularity in the polls, access to the resources and the apparatus of the state, and reputation among the political elites, Primakov and Putin were the likely non-Communist frontrunners. Political elites (governors, but also minor parties, analysts, and the media) had to decide whom to support. The candidate with the stronger elite support would prevail over the other, because elite support translated into their backing of the candidate by all means at their disposal. While “electoral technologies” have been known to perform miracles in Russia (see Chapter 2 on how and why the media in Russia was effectively a tool to control the vote and Chapter 7 on the estimated extent of the impact of the governors on the regions’ vote), these mechanisms by themselves could not pick out the winner of the presidential race.13
the unity bloc shoots up like a rocket The seeming great puzzle of the parliamentary campaign was how a new electoral bloc, Unity, created just ninety days before the vote, with an unclear program and unknown candidates, could essentially win the election. This question, though, is answerable based on coordination theory: Unity became a focus of the non-Communist support because elites were looking ahead to backing Putin for the Presidency, and Unity was created to be a vehicle for that. And, consistent with theory, once Unity became clearly associated with Putin, its popularity among the prospective voters skyrocketed: from 4.5 percent on November 2 (VTsIOM, October 29–November 2, 1999) and 8 percent on November 22 (VTsIOM, November 19–22, 1999) to 18 percent on November 29 (VTsIOM, November 26–29, 1999) and 23 percent on the election day, December 19. Such a rapid change of fortune did not result from the party’s political program, its candidates’ personal popularity, or vote rigging, but reflected the victory achieved by Unity and those behind it in swaying the preelectoral game of coordination. The bloc was successfully presented as the “party of power,” the party that was most likely to be associated with the future Russian President. When Putin was named prime minister in early August, the Kremlin was facing a serious problem. Fatherland, as an opposition party, in coalition with All-Russia, was recognized as the favorite for the forthcoming parliamentary and presidential elections by all observers, was supported by the most influential governors from populous and prosperous regions, was sponsored by such leading Russian corporations as AFK Systema, Most, LUKoil, Gasprom, TatNeft, and Magnitogorsk Steel Company, and was joined 13
According to one observer (Igor Malashenko), “[media tycoon] Boris Berezovsky said many times (there are many witnesses; I heard it myself) that he sincerely believes that in Russia anybody can be elected president with adequate financial support and control of the TV . . . Anybody” (Paddock 2000). Boris Berezovsky is generally believed to have been behind Yeltsin’s reelection in 1996 and the ascendance of Putin in 1999.
224
Olga Shvetsova
by the Former Prime Minister Primakov, Russia’s most popular prospective presidential candidate. The Kremlin had repeatedly attempted to launch an alternative to Fatherland that the governors would support.14 These attempts continued to fail – mostly because the governors wanted guarantees before they made a commitment that the proposed bloc would indeed become the “party of power.” That created a dilemma – only a strong party with a good chance of success could attract the governors away from Primakov and Moscow Mayor Luzhkov, but without the prior commitment of a sufficient number of governors such a strong party would be impossible to create. Apparently, the solution was to entice some governors to join in a loose alliance that would not require any commitments, formal membership, ideology, policy platform consensus, or even a leader, and to hope that once a sufficiently large number of them participated the others would follow. According to Alexey Golovkov, the head of Unity campaign headquarters, the idea to create a new bloc of governors as the opposition to Fatherland belonged to the financier and media tycoon Boris Berezovsky and Minister for Emergency Situations Sergei Shoigu who proposed it to then President Yeltsin (Victorova 1999). While Yeltsin expressed no interest, new Prime Minister Putin fully supported the idea and took an active role from the very beginning. Previously, as the first deputy head of the presidential administration, he had been in charge of relations with the governors and, in his own words, had “maintained good personal contacts with many of them.”15 Just four days after his appointment as prime minister, Putin went to the Siberian city of Tomsk, accompanied by fifty federal officials, to participate in a meeting of the regional association of nineteen Siberian governors, The Siberian Accord. The move pleased the governors; the last time a prime minister had participated in the work of the association was in 1993 (Babaeva 1999; Izvestia, August 14, 1999; Kommersant-Vlast, August 24, 1999; Latyshev 1999). A month later, thirteen of the nineteen governors of The Siberian Accord were among the founders of Unity – a new progovernment regional bloc. For recoordination, all that the creators of Unity needed was to switch enough players from a “Primakov” to a “Putin” equilibrium. Thus, they did not ask for more from the governors than their mere presence. All that was required to join the bloc was to sign an open letter. The letter was distributed during the first fall session of the Council of Federation, at which all governors were present. It was such an indefinite statement that some of the governors (e.g., Viktor Ishaev of Khabarovsk, Yegor Stroev of Orel, 14 15
For example, “Russia,” The United Right-Centrist Bloc, The Centrist Bloc (Tregubova 1999 [b]). Quoted from Putin’s interview to the television station TV6 on November 28, 1999. English transcript is available in the BBC Summary of World Broadcasts, November 28, 1999. Available at http://www.lexis-nexis.com/universe).
Resolving the Problem of Preelection Coordination
225
Gennady Igumnov of Perm, Mikhail Prusak of Novgorod, and Nikolai Kondratenko of Krasnodarsky Krai) later said it was more like they had signed an appeal for fair elections than entered an electoral coalition.16 Indeed, the letter mentioned no electoral blocs nor names of presidential candidates and its signers did not have to forgo their prior affiliations.17 The letter was signed by forty governors (the media named it “the letter of the 39”) and was initially published on September 20, 1999; four days later, eleven other governors, including the Speaker of the Council of Federation, endorsed it.18 Because most of these now fifty-one governors were not yet ready to make an exclusive political commitment to Unity, on September 27, 1999, Vladimir Putin invited them all to the Kremlin to assure them that he personally liked and respected Sergei Shoigu and strongly supported the new electoral bloc. At the same time, Putin stressed that “Fatherland could not be supported” (Vremya MN, September 28, 1999). Based on this, thirty-two governors (the list was never published) agreed to sign yet another letter, saying that they had “made a principled decision to help the leader of Unity, Sergei Shoigu, win the December elections.” It was emphasized that the bloc did not expect the regional leaders to participate actively in the campaign. They could merely endorse the bloc from their positions of respectability and authority. Moreover, the governors retained the opportunity to leave the bloc or to join it even a few days before the election, or even join and leave it several times (Markov 1999).19 16 17
18
19
See reports in Parlamentskaya Gazeta, September 25, 1999; Sovetskaya Rossiya, September 28, 1999; Segodnya, October 5, 1999; and Nezavisimaya Gazeta, October 13, 1999. It is also possible that some of the governors were simply tricked into supporting Unity. Originally it was assumed that a new bloc of governors would merge with NDR (Tregubova 1999b). There were even reports (in Berezovsky-controlled media) that the deal had been sealed (e.g., Nezavisimaya Gazeta, September 30, 1999). That would explain why several influential governors – active supporters of the NDR – signed the initial letter. But at the press conference on October 4, Viktor Chernomyrdin declared that the NDR could not “accept the conditions” that Unity offered. He slammed the movement saying that Unity “[did] not have any ideology whatsoever; instead, it has Boris Berezovsky alone” (Interfax, October 4, 1999. Available at http://www.lexis-nexis.com/universe). The full text of the letter was distributed by Interfax Services (Interfax, September 20, 1999) and reprinted by several regional newspapers. Moscow-based newspapers published an abridged version of the appeal and mentioned only the names of the best-known signers. The full list of forty initial signers also appeared in regional sources. See, for example, the September 21, 1999 issue of KomiInform (Komi regional daily news service) or the Vladimir regional daily Molva from September 23, 1999. The names of seven governors who joined the group on September 21 were reported by the National News Service (NNS, September 21, 1999. Available at http://nns.ru/archive/chronicle/1999/09/21.html). There is a fine line between rushing to support the likely winner and yielding to the pressure from the presidential administration. There were rumors that both the governors and the business elite came under a strong pressure to support Unity and Putin. In fact, some suggested that among the first to join were governors having problems with law and who
226
Olga Shvetsova
The governors at that point did not rush to campaign on Unity’s behalf. In fact, until late November, fewer than a dozen governors had actively supported it. It is not surprising, then, that the ratings remained stagnant. While Putin’s ratings as a presidential candidate reached 42 percent by mid-November, public support for Unity was at 8 percent.20 When, on November 24, Putin invited governors to the annual meeting of Shoigu’s Ministry for Emergency Situations, where he endorsed the bloc publicly, as “a party for which I will vote as a citizen” (Izvestia, November 25, 1999), this announcement completed the recoordination of the game to an alternative equilibrium by bringing together the critical mass of “preassembled” elite supporters and a strong potential candidate. Next week, the polls showed a jump of ten percentage points in Unity’s favor. It had for the first time surpassed OVR in popularity.
fatherland-all russia: the quick demise of a front-runner The proposition that the parliamentary campaign was treated by major players as an elite-level primary for the non-Communist presidential nomination is also consistent with the dramatic decline of the initial favorite of the campaign, the electoral bloc OVR, whose story provided the opening for this essay. Former Prime Minister Primakov, one of the bloc’s two leaders, was Russia’s most popular non-Communist politician in the summer of 1999 (VTsIOM, August 20–4, 1999). All major parties, including the Communists, publicly invited Primakov to join, and Luzhkov had actively sought an alliance with him since April though the two disagreed on who would become a presidential candidate. On August 17, in a compromise move, they had announced that they would postpone that decision until the parliamentary election, and Primakov formally accepted Luzhkov’s invitation to lead the new bloc. The OVR immediately became the favorite for the Duma and was widely expected to produce a winning presidential candidate in the summer election of 2000. Primakov’s popularity, Fatherland’s organization, and the supporting governors’ political machines were a powerful combination. The union between Fatherland and All-Russia was not based on any strong programmatic foundation. Observers of the Russian elections seem to agree that were Primakov not the most likely presidential winner, the alliance would not have been possible. Moscow Mayor Luzhkov and his Fatherland had very little in common with the republic leaders of All-Russia
20
were, therefore, vulnerable to blackmail by a law-enforcement organization (Petrov and Makarkin 1999). A regionally based political movement, “For equal rights and fairness” of Ruslan Aushev (President of Ingushetia), at its Congress characterized methods used to create Unity as the methods of the KGB (Ulianov 1999). VTsIOM, November 18, 1999.
Resolving the Problem of Preelection Coordination
227
in terms of ideology, their favored economic policies, or views on the future of the Russian federalism. Indeed, the OVR alliance ended almost as soon as Primakov’s presidential victory was ruled out. Encouraged by its early success, OVR inadvertently contributed to the rise of Unity by overlooking the possibility of rapid elite recoordination. Instead of concentrating efforts on winning the elite “primaries,” it aimed its campaign directly to maximize its chances of victory against the Communist Party candidate in the presidential election, failing to notice until it was too late that the Kremlin (or others) could succeed in launching a new entrant and win the presidency if it found a way to refocus the support of governors and businesses. As early as August 1999, confident in its parliamentary success and in order to strengthen OVR’s position in the presidential race by attracting leftist voters, the Luzhkov-Primakov bloc adopted a centrist ideological platform and even tried to move further in the direction of their would-be competitor in the presidential runoff, the CPRF (Bocharov 1999). That strategy was optimal for reducing CPRF’s electoral base. In fact, Fatherland successfully managed to ally with two left-of-center parties formerly associated with the CPRF: The Agrarian Party and Spiritual Heritage. The leaders of OVR (Luzhkov and Primakov) and its activists were presenting it as a party of centrist and social-democratic values, sensitive to the concerns of the Left and of the Russian “patriots,” but also as a party of successful pragmatists and professionals.21 The flaws of the strategy of position taking and converging with the CPRF on issues became obvious only later, when, at the end of September and with only three months before the election, Unity entered the race as a new player for the Kremlin. Then Unity went on to win in the game of elite coordination, while adopting a platform vaguely to OVR’s right. As a result, OVR found itself not only deprived of its “focal” advantage, as Unity turned into the major player, but also, squeezed on policy issues on both sides with very little room for electoral maneuver. The speed of OVR’s demise was a striking manifestation of the equilibrium-selection nature of the preelectoral political discourse, taking place in the course of weeks if not days. But even in late December 1999, after being an object of an unprecedentedly negative campaign in the media and after his party fell behind in the polls, Primakov was third in popularity after Putin and Zuganov, thus remaining a plausible presidential candidate (VCIOM, December 31, 1999–January 4, 2000). The only person Primakov could not beat face-to-face in the popular ballot at that time was Putin, but 21
For example, when during a much publicized preelectoral debate one of OVR’s leaders was asked to articulate at least three distinctions between her bloc and the Communists, she named only one – their attitudes toward private property. While admitting that their programs were similar, she emphasized that the real difference was in the leaders. Indeed, content analysis of parties’ economic programs confirms this observation (Dmitriev 1999).
228
Olga Shvetsova
only one of them could advance to the second round and either one would have been sufficient to defeat the Communists. The coordination game thus was still present, even though with favorites reversed: in the six months left before the presidential election the situation could have developed in a way that prevented Putin from running or winning. Yeltsin’s resignation on December 31 and his passing of the presidential reigns to Putin served to seal the outcome of the coordination game. It added the status of an incumbent to Putin’s focal advantage and brought the presidential election three months closer. Primakov’s withdrawal from the presidential race reflected the realization that the elite coordination on the prospective winner had taken place and was about to become a self-fulfilling prophecy.
eclectic electoral incentives and party development in russia Because the process of elite coordination for presidential purposes has dominated parties-based political competition, the mixture of parliamentary and presidential incentives under the Russian semipresidential constitution is problematic for the development of an institutionalized party system. Considering the importance of the presidency, one can fully expect that presidential objectives will continue to weigh heavily in determining the parliamentary strategies of major parties. The consequences can be profound for democratic consolidation, as both the institutionalization of the party system and the stability and legitimacy of the country’s formal institutions are affected. For the party system to institutionalize, repetitive electoral outcomes are essential (Mainwaring and Scully 1995; Linz and Stepan 1996 [b]). This repetitiveness requires not only reasonably stable preferences in the electorate but also some continuity in elite behavior. The recurrence of electoral outcomes is necessary for party politicians to form stable expectations about others’ actions (O’Donnell 1994, 59). Stable expectations, in turn, enable players to adopt what can be described as stationary political strategies (Shvetsova 2001), meaning that elites stay in the same party, appeal to the same voters on a generally consistent overtime political platform, and keep political allies for longer than one electoral period. The pursuit of presidential goals has the potential to prevent this continuity in elite behavior – by turning interactions among parties into a strategic environment with multiple equilibria where electoral outcomes strongly depend on the difficult-to-predict outcomes of a coordination process. It potentially destabilizes expectations about the parties’ future performances by making parties’ pasts only marginally relevant for their futures. When parliamentary campaigns become occasions for the elites to converge on a likely presidential front-runner and to line up behind him, victorious new parties can be created along the way as organizational
Resolving the Problem of Preelection Coordination
229
fronts for the presidential bandwagon. As coalitions of convenience, parties and blocs created for presidential purposes poorly fit the demands of parliamentary existence. Indeed, in Russia, the parties and blocs with seats in the Duma quickly begin to lose their parliamentary membership and on occasion even their leaders as soon as the election is over. The initial dynamics in the 1999 Russian parliamentary campaign depended heavily on the absence of a presidential incumbent who could be a natural focal figure in the race. However, even after the selection of a new president was accomplished in March 2000, the prospects for partysystem stability did not improve. The winner of the elite “primaries” and subsequently of the presidential election, once in office, had few or no incentives to strengthen political parties. The weaker the parties, the less coordinated remains the electoral process and, other things being equal, the stronger, through its tremendous “focal” power, is the incumbency advantage. Even in the presence of a strong incumbent, as long as success in the presidential race is a function of ad hoc elite coordination on a particular candidate, there always remains a possibility that in the next election someone else will become the “focus” of elite support (e.g., due to timely external shocks to the political system). The very example of drastically reversed fortunes of Primakov and Putin provides a clear illustration of such a possibility. Given that the prospects of presidential hopefuls will remain highly volatile, in the next election, many different politicians may be potentially viable as candidates for the office. Of course, they would all be viable only up to the point when the coordination takes place – after that, in modern Russia, there is room for only one viable candidate to enjoy the universal backing of the elites. Nevertheless, it is rational for each potential candidate to build his own political organization in order to present himself as a reasonable alternative and to help him win the coordinated elite endorsement. By this logic, by the time of the next parliamentary election, even with a strong incumbent, quite a few new parties could be launched by visible public figures in preparation for their presidential bids and as a strategy for building support.22 Such a scenario (though always possible) is, of course, the more likely to occur the weaker the presidential incumbent. Also, one would not expect it to unfold until closer to the end of the electoral cycle, with parties and politicians pursuing parliament-appropriate strategies of cooperation and coalition building in the interim. Still, this danger of new parties 22
An interesting question is why presidential hopefuls could not connect with existing parties rather than create new parties. In the West, party leaders come and go while parties stay. In Russia up to this point, it appears, parties come and go while politicians stay. In fact, several visible parties do stay, but their own leaders harbor strong presidential aspirations and would hardly welcome an outsider as an internal competitor for the party’s presidential nomination.
230
Olga Shvetsova
creating a political lottery even for an incumbent contributes to understanding the willingness of Putin to support the new law on political parties, which makes creation of new parties substantially more difficult. But the problem for the development of the party system remains: the multiplicity of potentially viable presidential candidates competing for elite endorsement implies sustained long-term divisions among existing ideologically similar parties. On the other hand, it is both logical and feasible for the newly elected president to transform the very game of presidential selection in order to eliminate its current propensity to generate multiple equilibria and to leave the elites with only one choice – the incumbent. Winning office through the game of coordination gives the incumbent fundamentally different incentives with regard to preserving the constitutional regime than does winning through the institutionalized interpretation of expressed popular preferences. Such an officeholder does not owe his success to the status-quo political institutions and may detest the very manipulability of the process that brought him the victory. Instead, he may opt for a system of rules that produces deterministic outcomes and further strengthens the incumbent by making him independent of the swings among the elite supporters. Of course, an alternative to this scenario always exists, to strengthen the role of democratic institutions and elections. Regardless of its stated purpose and the true intentions behind it, the federal reform initiated by Putin shortly after assuming office, on May 7, 2000, greatly reduced political influence of regional politicians by granting the president an unprecedented degree of control over them (the reform includes his right to fire or suspend governors and city mayors, includes his right to remove the governors from the Federation Council, erodes their influence in the appointment and dismissal of local Heads of the Interior, and it also de facto enlarges the primary territorial units of Russia by combining them into seven federal districts). Similarly, major financial groups that have failed to support the “right” side in the presidential game (e.g., Gazprom, Lukoil, MediaMOST, Interross, Tyumen Oil Company, AvtoVaz., and Norilsk Nickel) recently came under attack. If successful, such assaults on the “oligarchs” and their financial empires would boost the government’s ability to control their political affinities. The mixed electoral motives in 1999 held back the development of Russian democracy in one other important regard. Because of the extreme competitiveness among the political elites prior to the resolution of the coordination problem and the lack of competitiveness once that coordination took place, the presidential campaign failed to generate a meaningful political discourse. From the beginning, at the parliamentary stage, all sides engaged in negative campaigning. Then, as elite convergence on Putin was accomplished, he ran effectively unopposed in the presidential race and the campaign largely ignored policy issues. As a result, voters did not have to
Resolving the Problem of Preelection Coordination
231
confront the need to make a political choice and, unless they supported the Communist Party candidate, were left with no other alternative but to approve a candidate about whom they knew little, a candidate who had never run for public office before, who presented no political or economic program, and who, for all practical purposes, faced no scrutiny in the campaign because he was unanimously supported by the regional and business elites and the media.
11 Coalition Politics in the New Duma1 Thomas F. Remington
That keen observer of Russian politics, Viktor Chernomyrdin, is supposed to have commented that “no matter what party we try to create, it always turns out to be the CPSU.” Presumably he meant that Russian political parties invariably turn out to intertwine themselves with the state, drawing on the state for political resources instead of cultivating a base of independent mass support. Russia has a “floating party system,” according to Richard Rose. Lacking roots in the electorate, the parties are instruments by which politicians keep themselves in power but avoid accountability to the voters. Electoral parties appear and disappear, and voters cannot reliably judge a party by its past performance or trust its promises for the future. The demand for parties, Rose argues, is not matched by a corresponding supply (Rose 2000; Rose, Munro, and White 2001). Political scientists postulate that modern democracy requires political parties to organize and structure political conflict, channeling the ambitions of politicians into open political competition, and defining publicpolicy alternatives that enable the electorate to make meaningful choices (Schattschneider 1960; Aldrich 1995). When it is working effectively, a competitive party system is an indispensable means of solving crucial collective action problems both for political elites and the general public. The continuing weakness of political parties in Russia, as several of the authors in this volume observe, fatally undermines Russia’s formally democratic political institutions. Yet in the arena of parliamentary politics, Russia’s parties play a major role in the policy process, determining the issue agenda for legislative deliberation and structuring voting alignments on the floor. Because parties influence the 1
I am grateful to the organizers of the conference, Vicki Hesli and William Reisinger, for their valuable comments on the initial draft of this paper, and likewise to Gerhard Loewenberg and Steven S. Smith for their suggestions. The paper arises from a larger project on Duma politics conducted in collaboration with Professor Smith.
232
Coalition Politics in the New Duma
233
ability of the Duma to resolve its own collective dilemmas, they determine the Duma’s ability to serve as a counterweight to the presidency. Despite their epiphytic character, the parties within the Duma have forced the president and government to engage parliament as a serious negotiation partner in the legislative process. Deputies differ in the nature of their electoral mandate – half represent single-member districts (SMDs), half party lists – and in the degree to which they are oriented programmatically or parochially. Yet the system of party factions in the Duma enables members within factions to coordinate on their legislative priorities and levels of voting cohesiveness. Collectively, the factions run the Duma, so that to the extent their members can speak with a single voice, interfactional bargains can forge majority coalitions for legislation and enable the Duma to form a collective will. Following the December 1999 election, the pro-Kremlin party Unity became the dominant force within the Duma and relations between executive and legislative branches became far more harmonious than they had been at any time in the 1990s. December’s parliamentary elections were followed shortly afterward by President Yeltsin’s resignation and Vladimir Putin’s accession to the presidency. Together, these developments have produced a striking new model of coalition politics within the Duma and a new pattern of legislative-executive relations. These changes have the potential to produce lasting and significant changes in relations among parliament, government, and the president. An early indication of the changes was the procedure used to distribute leadership positions in the Duma in early 2000, which differed sharply from the way in which the Dumas of 1994 and 1996 had decided the same questions. In those two previous convocations, faction leaders produced a powersharing agreement in which all factions and registered groups won leadership positions in proportion to their floor strength and no factions were excluded from receiving a share of the spoils (Remington and Smith 1995; Remington and Smith 1998). In the Duma of 2000, three factions comprising a bare majority of seats formed an alliance and divided up nearly all the leadership posts – including the chairmanship of the chamber and the chairmanships of the committees – among themselves. Several features of this arrangement deserve closer attention. First, the majority coalition resembled a “minimum winning coalition” in that it comprised the smallest number of factions that could jointly control a majority of seats. But it formed with little regard to likeness of policy preferences. Second, the Communist Party gained many more leadership posts than either their actual strength in the Duma or their relationship with the Kremlin would have dictated, while the Union of Right Forces (SPS), which had supported Putin (and the Chechen campaign) in the election, was shut out of the power game. Third, acting President Putin took an active hand in forming the winning coalition by directing the parliamentary strategy of a propresidential faction, whereas in the past, Boris Yeltsin had done little to construct
234
Thomas F. Remington
interfactional majority coalitions. Yet the actual balance of advantages and disadvantages that this new set of arrangements conferred differed from those that many observers feared.2 Specifically, the Communist Party’s influence was less than originally appeared to be the case, while the influence of the right forces was greater. As a result, President Putin gained much greater influence over the outcomes of legislative decision making in the Duma than Yeltsin had. To explain this striking new pattern of coalition politics in the Duma, we need to consider the influences – electoral, partisan, and ideological – that motivated the behavior of Russia’s elected representatives in setting up the arrangements for governing themselves. As we will see, there were some systematic differences among parties in how these motivations are combined among their candidates. An election such as the 1999 parliamentary election therefore decided not only the relative strength of the winning parties and ideological camps, but more generally the distribution of goals among the members of the newly convening Duma. This time another significant factor was operating as well. President Yeltsin’s decision shortly after the December election to resign and to allow Prime Minister Putin to take over as acting president created an opening for the Kremlin to craft a new kind of legislativeexecutive relationship by dictating the coalition behavior of the new pivotal faction, Unity. In this chapter we will briefly review two features of the constitutional and electoral environment in Russia that affect parties’ and deputies’ stakes in the organization of a new Duma. Then we will turn to the results of a survey of candidates running for the Duma in 1999 that shed some light on the way deputies’ calculations about Duma governance were affected by their electoral, partisan, and ideological interests. Third, we will examine the sequence of actions that the newly elected parliamentary leadership took in allocating power in the chamber. Finally, we will ask what the new pattern of leadership in the Duma may mean for future coalitions in the Duma and for legislative-executive relations.
russian parties and the state duma Parties’ aims in capturing seats in the Duma in 1999 were similar to those in the previous Duma elections in two respects. First, in varying degrees, parties intended to use the legislative process to shape public policy. Electoral parties differed both in their policy orientations and in the degree to which they had distinct policy orientations. As in previous elections, parties’ political tendencies in 1999 could be classified in four basic categories: “proreform,” 2
For example, Moscow Mayor Yurii Luzhkov, whose OVR alliance was excluded from the majority coalition, denounced Gennadii Seleznev’s election as chairman of the Duma “one of the first signs of the upcoming Bolshevik dictatorship” (RFE/RL Newsline, January 19, 2000).
Coalition Politics in the New Duma
235
“Communist,” “nationalist,” and “party of power.” Sometimes the latter tendency is characterized as “centrist,” but this is misleading; generally politicians adhering to the current “party of power” are a residual category composed of pragmatists who prefer the benefits of office to any policy commitments. Compared with past elections, in this election, the reformist elements were more successful at uniting under a common banner; the Communist elements were more fragmented; nationalists were both more fragmented and electorally weaker; and the “party of power” divided into two rival organizations: the Primakov-Luzhkov alliance (Fatherland-All Russia, or OVR), and the Unity organization that gained popularity at an astonishing rate thanks to Putin’s soaring public reputation.3 As a result, the distribution of seats in the new Duma by political tendency differed substantially from the two previous Dumas. In this Duma, the left opposition was much smaller, and the progovernment forces much larger. However, neither the Left nor the Right commanded a majority, as Chapter 1 shows. A second similarity with the previous Duma was that, once elected, both party-list and SMD deputies immediately sought to join factions and registered groups. Although during their campaigns many candidates claimed that they were independent of party influence, once elected, nearly every deputy affiliated with one of the Duma factions or registered deputy groups. Only nineteen deputies chose to stay outside of any faction or registered group in the new Duma. Factions and registered groups are given crucially important organizational resources, which politicians value both for their benefit in Duma legislative politics and for their usefulness in electoral campaigns. Among these are the automatic right to office space and funding for staff. Another crucial right is that the leader of each faction and registered group is represented as a voting member of the steering body for the Duma, the Council of the Duma. The Council of the Duma is made up of the heads of all factions and groups on a parity basis. At least two of the factions in the 1996–9 Duma – Yabloko and the Communists – effectively turned their faction offices into election headquarters. Moreover, both in the first and second Dumas, deputies who were members of nonparty groups, such as the New Regional Policy group in the first Duma and Russia’s Regions in the second, worked actively to develop electoral organizations of their own. While the deputies elected from party lists have the automatic right to form Duma factions, start-up groups formed by other deputies must qualify to register with the secretariat in order to obtain the same rights. To register, a group must have a minimum of thirty-five members, a threshold rule that was established in the first days of the Duma which convened in 1994 and has been consistently reaffirmed since then. It is notable that almost no group that falls short of the thirty-fivemember floor survives, but that group size sometimes hovers tellingly just 3
On the decline of the OVR alliance and the rise of Unity, see Chapters 7 and 10.
236
table 11.1 Faction Affiliation Changes between January 2000 and September 2001 Factions as of January 31, 2000 CPRF Factions as of September 2001 CPRF Agro-Industrial Independent LDPR Russia’s Regions People’s Deputy Yabloko Unity OVR SPS total
81 4
AgroIndustrial 1 38 1
Independent
LDPR
Russia’s Regions
People’s Deputy
Yabloko
Unity
OVR
SPS
Total
31
84 42 12 12 41 53 18 80 43 36
31
421
2 5
3
3
12 1 2
1
38
1 50
2
1
1 18
1
86
2 2 40
14
15
38
55
72 2
3
20
78
2 41 44
Coalition Politics in the New Duma
237
table 11.2 Factional Affiliation as of September 5, 2001
CPRF Agro-Industrial Independent LDPR Russia’s Regions People’s Deputy Yabloko Unity OVR SPS total
Number
Percent
86 43 15 12 45 59 18 83 44 38 443
19.4 9.7 3.4 2.7 10.2 13.3 4.1 18.7 9.9 8.6 100
over the thirty-five-member threshold needed to maintain the organizational benefits of faction status (Remington 2001). Moreover, factional affiliations in the Duma are becoming more stable. Over the period from January 2000 to September 2001, only 8 percent of the members switched their faction membership.4 (See Tables 11.1 and 11.2.)
ideological, partisan, and electoral interests of deputies Results of a survey of candidates running for the Duma in 1999 shed light on the motivations that affect the new deputies as they form groups and coalitions, and made their initial decisions on governance in the newly convened Duma.5 The purpose of the survey was to gather evidence bearing on the relationship between candidates’ policy-based, partisan, and electoral motivations and their attitudes about the institutional structure of the new Duma. The findings of the survey are useful in helping us explain how the 4
5
This applies to the 421 deputies who were members throughout the January 2000 to September 2001 period. Table 11.1 indicates the direction and magnitude of faction switching for these deputies. It does not show the affiliations of members who entered after January 2000 or of those who died or departed for other reasons. The columns reflect the distribution of deputies by their January 2000 factional affiliation. Rows reflect the distribution of membership of members as of September 2001. Thus the table shows that the Communists suffered a slight net loss (four members went to the agrarians and one to Unity, losses that were not quite offset by gains from among the agrarians [one member] and independents [two members]). Unity, on the other hand, was a net importer, having gained more members from the ranks of the Communists, independents, LDPR, PD, and OVR than it lost to the independents and SPS. SPS was also a net importer. Table 11.2 indicates the factional affiliation among all members listed on the Duma Web site as of the beginning of September 2001. See appendix for details of the survey.
238
Thomas F. Remington
orientations of the newly elected members of the Duma toward their parties, their voters, and their policy agenda are related to the formation of crossfactional coalitions. In seeking understanding of deputies’ behavior, we can draw upon recent developments in positive theories of legislative institutions. A number of recent studies have proposed simple but powerful models of U.S. congressional behavior in which members seek to maximize reelection chances or influence over public policy through particular institutional arrangements. Given simple assumptions about members’ motivations, scholars have demonstrated that institutions such as procedural rules, committee powers, party agenda control, bicameralism, and separation of powers provide solutions to particular types of collective dilemmas for members and shape the outcomes of decision making in particular directions (for surveys of this literature, see Shepsle and Weingast 1994; Smith 2000). However, many scholars question how well the full range of politicians’ behavior can be captured by models that reduce their motivations into a single master objective function. Much as Fenno (1995 [1973]) concluded that members of Congress could best be understood as pursuing combinations of three goals in particular, policy making, reelection seeking, and influence within the chamber, recent studies of the coalition behavior of European parties have emphasized that theories based on the assumption that parties are primarily concerned with maximizing their influence over public policy do not adequately account for key choices made over joining, supporting, or opposing a governing coalition (Strom 1990; Huber 1996 [a], 1996 [b]). As Strom observes, “Some parties are more office motivated, others more concerned about the pursuit of votes or policy” (1990, 242). As a result, parties differ in their assessments of the benefits and costs of entering a government. Elsewhere, Smith and Remington (2001) have analyzed the series of decisions that deputies elected to the first Duma in 1994 made over their governing arrangements, including the faction system, the election of chamber and committee leaders, the choice of an electoral law, and cohesiveness in floor voting. We found that depending on the salience of a particular institutional issue for deputies’ goals and the degree to which deputies could make knowledgeable inferences about the consequences of alternatives for their interests, much of the variation in deputies’ behavior could be explained by knowing three attributes: their preferences over policy issues, the type of electoral mandate they held, and their factional affiliation. We found that factional influence on deputies’ voting behavior could not be entirely decomposed to its policy-based and electoral components, but appeared also to reflect members’ common stake in a partisan collective good. To be sure, factions differed according to the nature and uniformity of members’ concerns. Not surprisingly, deputy groups (that is, those groups that met the thirty-five-member threshold and were registered) that tended to be made
Coalition Politics in the New Duma
239
up of single-mandate district deputies tended to be less homogeneous in their views on public policy, but their members nonetheless shared common views on how the Duma should be governed (Smith and Remington 2000).6 In the 1999 survey of candidates, as in previous surveys of candidates (Remington and Smith 1998), there was a dominant issue dimension along which deputies were aligned. A principal components analysis of responses to a battery of questions about public-policy issues extracted a factor explaining about 21 percent of the variance in responses.7 This factor loads especially highly on questions about whether there should be private ownership of land, whether privatization of state enterprises should be accelerated, the importance of social assistance measures, and the desirability of protectionism.8 Candidates were scaled along this issue dimension, which we can regard as a “left-right” axis where the left pole represents the pro-Communist position and the right a liberal, market-oriented position.9 A candidate’s scale score is a summary measure of his or her preferences over basic policy choices concerning the orientation of Russia’s economic and political system. In a political system where voters were perfectly informed about elected representatives’ activities and voters rewarded or punished representatives accordingly at the polls, we might expect that maximizing policy influence would also maximize politicians’ chances of being reelected. But Russia is far from being such a system. There is little indication that voters either in SMD or in party-list races reward or punish incumbent deputies based on their voting records – indeed, it is virtually impossible for constituents to know their representatives’ voting records. Therefore we should expect that politicians’ electoral interests might generate different political objectives than their policy-based goals – not necessarily contradictory goals, but separate. This would be particularly marked in an electoral system such as that of Russia, where half of the deputies hold SMD mandates and half party-list mandates. We might therefore expect that on matters directly affecting their chances for reelection, deputies would be aligned differently than on issues more closely related to their policy concerns. For example, on a vote about changing the election law, deputies elected from SMDs, regardless of political 6
7 8
9
Note that a deputy elected on a party list is permitted to join a different faction or deputy group, and there is a good deal of movement in and out of factions and groups over the life of a Duma. The second strongest factor accounted for only about 11 percent of the variance. Factor loadings are .69 for the item on whether land should be subject to free purchase and sale; .73 for whether the privatization of state enterprises should be accelerated; .69 for whether the state should support all citizens or only those who cannot help themselves; and .63 for whether protectionism toward Russian industry is needed or not. Positive scores are the leftist (Communist) side of the spectrum. Scores ranged from −2.6 to +1.9 (negative scores are associated with the promarket, liberal end of the spectrum; positive with the Communist end).
Thomas F. Remington
240
table 11.3 Mean Scores on Left-Right Dimension by Party List
Party Communist Party of the Russian Federation Other Party List Zhirinovsky’s Bloc Fatherland-All Russia Unity Our Home is Russia Yabloko Union of Right Forces Single-Member Districts Only
Mean Score
Standard Deviation
Number of Cases
1.021
.479
30
.266 .242 .055 −.049 −.223 −.567 −1.153 −.010
1.043 .693 .956 .988 .949 .967 .815 .904
46 18 24 18 33 28 20 135
Source: Survey of candidates, 1999.
ideology, might line up on the opposite side from party-list deputies. But on issues concerning Western investment in Russia, deputies on the left might line up against deputies on the right, regardless of electoral category. In this case, a candidate’s electoral mandate type would be unrelated to his or her policy score. The survey evidence is consistent with this argument. We find that among our respondents there were no statistically significant differences in the mean score on the left-right dimension among candidates by electoral mandate type. However, as we would expect, there were marked and statistically significant differences among some parties on this dimension. Party-list candidates running on the Communist, Yabloko, and SPS’s lists differed significantly from the mean for all candidates – the Communist Party of the Russian Federation (CPRF) anchoring the left end of the spectrum, SPS and Yabloko the right (see Table 11.3). On the other hand, the mean scores of other parties’ list candidates were not statistically distinguishable from the mean for the sample, while the standard deviations tended to be high. Thus, some parties had distinctive left-right positions, while others did not. This is consistent with the expectation that some parties adopted more programmatic electoral strategies, campaigning on policy-based appeals, while others preferred a vaguer appeal free of specific policy objectives. In particular, the three “parties of power,” Unity, Our Home is Russia (NDR), and OVR, have medians at the mean point for all deputies and their members are dispersed across much of the left-right spectrum. Therefore “centrist” is a misleading label to apply to such parties: they are vehicles of convenience for the political interests of officeholders with little concern for a coherent policy stance. Figure 11.1 presents the same data graphically in the form of a boxplot showing the median scores
Coalition Politics in the New Duma
241
REGR Factor Score 1 for Analysis 1
3
2
286
1
0
-1
-2
-3 N=
101 989
135
30
none
24
18
OVR CPRF
33
18
Zhirinovksy
28
20
Unity
46
other
Yabloko
NDR
SPS
Candidate's Party List figure 11.1 Mean Factor 1 Score by Party List.
and twenty-fifth and seventy-fifth quartile scores for each group on this dimension. Note that the CPRF and SPS are both the most distinctive and cohesive ideologically of the parties. If the electoral interests of newly elected deputies differ in a systematic way between those who were elected through a party list and those who were elected from an SMD, and between members of more programmatic parties and those belonging to pragmatic “parties of power,” we should expect to see evidence of these differences showing up in candidates’ campaign strategies.10 There should be distinct differences among candidates in the way they invoke the past record of the Duma and make promises about a future legislative agenda. We would expect that, on balance, candidates representing a particular SMD would emphasize a more personal and less ideological relationship with their voters, while candidates running on party lists would emphasize the image and stance of their party, with more programmatic parties emphasizing their legislative agenda more than the broad, pragmatic parties of power. Candidates running both in district and party-list races 10
On the nature of the contrast between programmatic and patronage-oriented parties in postCommunist Europe generally, see Kitschelt et al. (1999). Regina Smyth (2001) develops this distinction in the context of Russian electoral parties.
242
Thomas F. Remington
might exhibit behavior somewhere between the two pure types. Generally speaking, this is what we found. The record of the last six years of Duma activity gave all candidates a target for praise and blame, and the survey showed that most candidates did make reference to the activities of the previous Duma in their campaigns.11 Overall, 77 percent of the candidates said that they referred to the Duma’s legislative achievements and failures in their campaigns, and 84 percent said that they cited specific legislation that they would work to get passed when elected. But the party-list candidates were more likely to do so than were candidates running only in SMDs. Eighty-three percent of list candidates as opposed to 73 percent of SMD candidates said that they called attention to the parliament’s past legislative decisions in their campaigns; and 89 percent of list candidates as opposed to 78 percent of SMD candidates said that they promised to try to pass specific bills if elected.12 Not surprisingly, party-list candidates were far likelier to point with pride to specific laws their party or faction had pushed through the Duma than were SMD candidates (78 percent to 50 percent).13 Likewise, party-list candidates were more likely to promise to work to change the constitution if elected than were SMD candidates (67 percent to 59 percent) and to promise that they would try to change the way parliament was run (46 percent to 36 percent).14 We get the impression that party-list candidates on the whole were more concerned with taking policy stances, an impression confirmed by the fact that the policy-oriented parties (Communists, SPS, and Yabloko) were far more likely than other party-list candidates to campaign by promising to work for passage of specific laws (over 90 percent), to refer to previous legislation passed by parliament (over 90 percent), to call for revising the constitution (80 percent), and to cite their parties’ legislative accomplishments (89 percent). (Among candidates running on the OVR, NDR, and Liberal-Democratic Party of Russia [LDPR] lists, by comparison, only 67 percent said they were calling for constitutional reform; 76 percent said they called attention to past legislation, 88 percent promised to pass legislation, and 68 percent referred to their parties’ legislative accomplishments.) In short, party-list candidates were generally more likely to campaign by referring to specific legislative matters past and future than were district candidates, and among party-list candidates, those running on the lists of the more programmatic parties were likelier to do so than were candidates 11
12 13 14
Most of our respondents were not current deputies. Of those who were (12.5 percent of the sample), there was much less inclination to criticize the Duma for its performance. The difference in the mean ratings given by the incumbents and the nonincumbents was strongly statistically significant (