Lobbying for Inclusion: Rights Politics and the Making of Immigration Policy

Lobbying for Inclusion: Rights Politics and the Making of Immigration Policy

CAROLYN WONG

Stanford University Press

00-S3453-FM 1/11/06 3:05 PM Page i

Lobbying for Inclusion

00-S3453-FM 1/11/06 3:05 PM Page ii

00-S3453-FM 1/11/06 3:05 PM Page iii

Lobbying for Inclusion Rights Politics and the Making of Immigration Policy

CAROLYN WONG

stanford university press stanford, california 2006

00-S3453-FM 1/11/06 3:05 PM Page iv

Stanford University Press Stanford, California ©2006 by the Board of Trustees of the Leland Stanford Junior University. All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or in any information storage or retrieval system without the prior written permission of Stanford University Press. Printed in the United States of America on acid-free, archival-quality paper Library of Congress Cataloging-in-Publication Data Wong, Carolyn, 1952 – Lobbying for inclusion : rights politics and the making of immigration policy / Carolyn Wong. p. cm. Includes bibliographical references and index. isbn 0-8047-5175-7 (cloth : alk. paper) 1. United States—Emigration and immigration— Government policy. 2. Pressure groups—United States. I. Title. jv6483.w66 2006 325.73 — dc22 Original Printing 2006 Last figure below indicates year of this printing: 15 14 13 12 11 10 09 08 07 06 Typeset by G&S Typesetters in 10/12.5 Palatino

2005032769

00-S3453-FM 1/11/06 3:05 PM Page v

For my mother, Rose Chu Wong, and in memory of my father, Charles Wong.

00-S3453-FM 1/11/06 3:05 PM Page vi

00-S3453-FM 1/11/06 3:05 PM Page vii

Contents

Illustrations Preface 1. 2. 3. 4. 5. 6. 7.

ix

xi

Introduction 1 Interest-Group Goals 23 Hart-Cellar Act 44 Post-Bracero Dilemmas 64 Legal and Illegal Immigration Reform 95 Revisiting Reform in a Republican Congress Conclusion 175

Notes

183

References Index

215

205

133

00-S3453-FM 1/11/06 3:05 PM Page viii

00-S3453-FM 1/11/06 3:05 PM Page ix

Illustrations

figure 1.1 Immigrants Admitted: Fiscal Years 1900 –2002

5

tables 1.1 Immigrants Admitted by Type and Selected Class of Admission, Fiscal Years 1986 –2002 7 2.1 Level of Legal Immigration: Poll Results, December 1995 2.2 Level of Legal Immigration: Poll Results, October 1996 2.3 Immigration Policies: Interest-Group Preferences

36 37

41

4.1 Unions Lobbying Independently of the AFL-CIO on Employer Sanctions, 1980s 80 4.2 Percentage of Hispanic Workers in Key Industries, 1983 5.1 Key Amendments to the Mazzoli Bill, 1984

81

110

5.2 Key Amendments to the Immigration Reform and Control Act of 1986 111 5.3 House Vote on the McCollum Amendment to Delete Amnesty, Mazzoli Bill, 1984 114 5.4 House Vote on the Roybal Amendment to Delete Employer Sanctions, Mazzoli Bill, 1984 116 5.5 House Vote on the Panetta Guest-Worker Amendment, Mazzoli Bill, 1984 117 5.6 House Vote on the Moorhead Amendment to Cap Legal Immigration, Mazzoli Bill, 1984 118

00-S3453-FM 1/11/06 3:05 PM Page x

x List of Figures and Tables 5.7 House Vote on the McCollum Amendment to Delete Amnesty, IRCA, 1986 120 5.8 House Vote on the de la Garza Amendment to Require Warrants for Searches, IRCA, 1986 122 5.9 Probability of a No Vote on the Panetta Guest-Worker Amendment, 1984 123 5.10 Key Amendments to the Immigration Act of 1990

124

5.11 House Vote on the Smith Amendment to Cap Legal Immigration, Immigration Act of 1990 126 5.12 House Vote on the Bryant Amendment to Delete All but Family Provisions, Immigration Act of 1990 128 5.13 House Vote to Pass the Immigration Act of 1990

129

5.14 Probability of a No Vote in the House on the Immigration Act of 1990 130 6.1 Amendments to the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 and Votes by Party Affiliation, 1995 –1996

148

6.2 Logistic Regression Model of a House Member’s Vote, Illegal Immigration Reform and Immigrant Responsibility Act of 1996 152 6.3 Fitted Probability of a Liberal (Proimmigration) Vote by a Hypothetical House Member, Illegal Immigration Reform and Immigrant Responsibility Act of 1996 154 6.4 Regression Model of a House Member’s Vote, Illegal Immigration Reform and Immigrant Responsibility Act of 1996 156 6.5 Logistic Regression Model of a House Member’s Vote, Illegal Immigration Reinforcement and Immigrant Responsibility Act of 1996 158 6.6 Fit Statistics for Separate and Combined Regressions

159

6.7 Pooled Regression: Chryslers-Berman-Brownback and Velazquez Amendments 160 A6.1 Vote on the Chrysler-Berman-Brownback Amendment to the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 166 A6.2 Vote on the Chabot Amendment to the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 170 A6.3 Vote on the Gallegly Amendment (Education) to the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 172 A6.4 Vote on the Pombo Amendment to the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 173 A6.5 Probability of a Yes Vote on the Chrysler-Berman-Brownback Amendment to the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 174

00-S3453-FM 1/11/06 3:05 PM Page xi

Preface

we live in an age when people can travel across oceans and continents in just a few hours. Multinational corporations frequently relocate their employees from one country to another. Manufacturers and providers of services readily import foreign workers from abroad. As the mobility of international labor has increased, national borders have become remarkably permeable. And as the nature of borders has changed, immigration policy has assumed an increasingly important role in delineating the boundaries of the nation-state. National governments have only gradually adapted their immigration policies to globalization. Relatively quick to implement policies that encourage the free trade of goods, the governments of highly industrialized democracies have been reluctant to adjust their laws to achieve an orderly and humane flow of labor into their national territories. Immigration policy is slow to change because it remains both a formal mechanism and a powerful symbol of national sovereignty. By specifying who can be admitted to—and who must be excluded from— the national political community, modern democratic states define themselves. I first became interested in how immigration policies are made because of their capacity to shape not only the makeup of the citizenry but also the character of the state. Immigration laws are enormously complex, a fabric woven of the many economic and cultural forces at work in domestic and international politics. In my analysis here, I focus on one of those forces, ethnic rights groups, organizations that have received relatively little attention in the immigration literature. In particular, I address

00-S3453-FM 1/11/06 3:05 PM Page xii

xii

Preface

a popular misperception that ethnic rights groups want more immigration simply to expand the political and social influence of “their own” ethnic groups in U.S. society. According to this view, as the influence of these groups grows, the nation’s immigration policies will continue to expand. And if ethnic rights groups can forge a coalition with employer groups— organizations that consistently demand the liberalization of immigration policies to satisfy their demand for foreign labor—some worry that immigration policies will spin out of control. From my research, I have come to a different conclusion on the role of ethnic rights groups in immigration policymaking. I find that when business interests press demands for more-expansive immigration policies, their influence is mitigated by an ethnic identity politics centered on immigrants’ rights. Writing this book would not have been possible without the help of many people. I owe the greatest intellectual debt to my teachers at UCLA, who gave invaluable guidance and constructive criticism on the design of the research project and on the earliest version of the manuscript. My principal dissertation advisers were Jeff Frieden and Miriam Golden. From Jeff, I learned the spirit and method of social science research and the tools of political-economic analysis. Miriam showed me how to carry out research with rigor and boldness of thinking. Michael Wallerstein inspired me with his creativity, incisive scholarship, and generosity. Tom Schwartz offered valuable insights about congressional politics and American political history. And Jim Denardo generously answered my many methodological questions. I want to thank a number of colleagues who offered valuable suggestions and advice. Gordon Chang, Jim Fearon, Morris Fiorna, Luis Fraga, Shanto Iyengar, and Roger Noll read and gave very helpful feedback on drafts of one or more chapters of the manuscript. Doug Rivers gave very useful advice on statistical methods. The field research in Washington, D.C., was conducted with the generous support of a fellowship from the Brookings Institution. I am grateful to Tom Mann, Kent Weaver, and Margaret Weir of the Governmental Studies Program at Brookings for their support and suggestions as I conducted the research. Joel Aberbach and John Bader also offered useful advice on the project, which was partially supported by a fellowship from the UCLA Center for American Politics and Public Policy. I also want to thank Stanford University Press. It was a pleasure to work with Amanda Moran, my editor. Judith Hibbard, the production editor, gave cheerful and valuable assistance, for which I am grateful. And I thank Carolyn Brown for her supervision of the copyediting.

00-S3453-FM 1/11/06 3:05 PM Page xiii

Preface

xiii

Warren Ko, Timmy Lu, and Linda Tran provided valuable research assistance. Finally, family members and friends have given me much encouragement. I am especially grateful to my mother, Rose Wong, and to Brian and Susan Wong, Emily Chu Wong, Alice Wong, Julia Wong, and Daniel Wong. Special thanks are owed Brant Wenegrat.

00-S3453-FM 1/11/06 3:05 PM Page xiv

00-S3453-FM 1/11/06 3:05 PM Page xv

Lobbying for Inclusion

00-S3453-FM 1/11/06 3:05 PM Page xvi

01-S3453 1/11/06 3:05 PM Page 1

chapter

Introduction

1

public debates on immigration re form in America are almost always emotionally charged because they reflect underlying tensions in two different realms of politics. In one realm—identity politics— immigration laws establish boundaries of membership in the national political community.1 By specifying who is legally admissible as a foreign national and on what terms, immigration laws speak to the nature of American identity. In the other domain—the politics of economic interests—more-tangible interests are at stake. Immigration reformers create expectations about who will “win” or “lose” as new immigrants enter the labor force and settle in local communities. Expansive policies often give rise to public concern and sometimes to overt resentment over the labor market and the fiscal effects of immigration.2 By taking a simple inventory of groups that lobby Congress on immigration issues, we can get a sense of how identity politics and the distinct politics of economic interest are intertwined. In recent decades, the most active immigration lobbyists have included Latino and Asian American rights groups, churches, humanitarian and human rights groups, population-control advocates, environmentalists, taxpayer groups, and pro–family values advocates.3 At the same time, private business associations and labor unions have undertaken lobbying campaigns to secure the economic interests of their members. Out of this disparate collection of identity and interests groups, odd and recurring coalitions of ethnic rights and business groups tend to emerge. We commonly observe these very different groups acting together in support of expansive admissions policies. Certain ethnic

01-S3453 1/11/06 3:05 PM Page 2

2

Introduction

rights advocates favor generous family-based immigration policies for their co-ethnics, while businesses that employ immigrants want to keep their lines of access to immigrant labor open. Immigration scholars emphasize that these coalitions are unusual because ethnic rights advocates and business groups more often than not occupy opposite ends of the left-right ideological spectrum (Zolberg 1990). But this cooperation is conditional, as disputes in Congress over immigration policy for temporary workers illustrate. When guest-worker policy is at issue, employer and ethnic lobbyists take up their more usual roles as competitors. The business interests favor programs that would temporarily admit large numbers of guest workers, while rightsoriented ethnic organizations generally oppose these programs unless they provide a clear path to citizenship. This book is a study of how organized economic interests and ethnic groups both cooperate and compete to influence lawmakers in Congress. Business lobbyists are mainly interested in economic outcomes; and they very often have both the voice and the funding to secure their favored outcomes. Ethnic rights groups, on the other hand, are concerned with attaining socially inclusive policies for new immigrants. Can these small organizations influence policy outcomes substantially? Or is their influence tangential to that of the more powerful economic interests? How do economic conditions and the ethnic constituencies in their districts affect legislators’ votes? I examine these questions in two ways. First, I provide an account of the congressional politics that led to the passage of three immigration bills: the Immigration Reform and Control Act of 1986 (IRCA), the Immigration Act of 1990, and the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (IIRIRA). These three pieces of omnibus legislation generated extensive debate in the 1980s and 1990s, debate that linked identity and economic policy issues in their focus on controlling illegal immigration and on regulating economic and familybased immigration.4 Second, I test a model of roll-call voting on these bills. This statistical study allows us to evaluate the ethnic and economic factors that influenced legislators’ voting choices. The legislative case studies show that lawmakers were highly responsive to American employers’ demands for access to permanent and temporary foreign workers. Associations of employers that rely on immigrant labor funded powerful lobbying machines. Given the economic resources that large corporate interests could deploy, it is no wonder that

01-S3453 1/11/06 3:05 PM Page 3

Introduction

3

certain business lobbyists often succeeded in fighting proposals to cut visa allocations for prospective workers and professional employees. More surprising was the discovery that several ethnic organizations representing Latinos and Asian Americans found ways to influence policy outcomes even though they commanded far fewer economic resources than those available to most business and union lobbyists. These ethnic organizations were most successful when they practiced an inclusive form of identity politics. Casting their demands for rights in universal terms, they formed alliances with other civil rights and humanitarian organizations. It was these coalitions that helped them sway the votes of moderate and undecided lawmakers. Ethnic minority organizations influenced policy outcomes even before the populations they represented gained significant political clout in the voting booth. Latino electoral power was emergent in some states during the 1990s (de la Garza and DeSipio 2005; DeSipio 1996; Fraga and Leal 2004), but Latino organizations were already significant players in federal policymaking on immigration issues in the early and mid1980s. As Zolberg (1999) observes: The ethnic organizations established by earlier immigrants in the United States have become legitimized as political interlocutors beyond what might be expected on the basis of their electoral weight, forming in the cultural sphere an equivalent to the corporatism that is sometimes found in the political economy sphere (but to a very low degree in the United States); this distinctly shapes the dynamics of immigration politics and also patterns the organizational strategies of more recent immigrants. (89)

The most influential organizations were nonprofit entities with professional staffs dedicated to lobbying public officials. By the 1970s, three major Latino organizations had established headquarters in Washington, D.C.: the Mexican American Legal Defense and Education Fund (MALDEF), the National Council of La Raza (NCLR), and the League of United Latin American Citizens (LULAC). The Japanese American Citizens League was founded in 1929. The Organization of Chinese Americans was established in the early 1970s. It was joined by the National Asian Pacific American Legal Consortium (NAPALC) in the 1990s. These and other ethnic organizations pressed civil rights demands on behalf of immigrants in national deliberations on immigration reform, while also acting as advocates on other issues, such as health care, education, and the protection of members of ethnic minority groups from racially motivated acts of violence.5

01-S3453 1/11/06 3:05 PM Page 4

4 Introduction In their book A Voice for Nonprofits, Berry and Arons (2003) note that nonprofit advocacy in Congress has not received much attention from interest-group scholars.6 The literature on interest groups and congressional lobbying is inclined to focus on large and wealthy interest groups rather than on the smaller nonprofits (designated in the federal tax code as 501c3s or 501c4s), which are more likely to speak for the poor and disadvantaged. One reason for this is that nonprofit organizations make up a very small segment of Washington-based interest groups.7 Nonprofit advocates did grow in number in the nation’s capital with the surge in citizens’ group activity that began in the 1960s (Walker 1991). The rise of citizens’ groups altered the nature of an interestgroup system that would otherwise have been dominated by associations formed around occupational interests.8 Recognizing the important role nonprofit groups have played in local politics since the civil rights movement, political scholars have widely studied their incorporation into municipal government (Berry and Arons 2003, 118). We know much less about the national influence of nonprofit groups that lobby on behalf of Latinos and Asian Americans. While most studies of minority representation focus on how racial and ethnic minorities can gain representation through the electoral or party system, our focus here is on how ethnic nonprofits provide informal representation for immigrants and their co-ethnics in national policymaking.9

Mixed Outcomes This study begins with passage of the Hart-Cellar Act in 1965.10 That legislation opened the nation’s door to the largest wave of immigration since the great migration to Ellis Island, which began in the 1880s and continued until the First World War. As shown in Figure 1.1, each decade after 1965 saw an increase in immigration. The unusual spike in the late 1980s was the result of a program that regularized the immigration status of a large number of undocumented aliens. The HartCellar Act also changed the ethnicity of the immigrant population: Most were no longer European but Asian and Hispanic. As we will see, this consequence of the Hart-Cellar Act was not anticipated by its framers, and it gave rise to restrictive reform movements in each decade following the law’s enactment. In the Immigration Act of 1921, the United States had adopted a system of national-origin quotas designed to limit immigration from Southern and Eastern Europe. The Chinese Exclusion Act of 1882 already had

01-S3453 1/11/06 3:05 PM Page 5

Introduction

5

2000 1800 Population in thousands

1600 1400 1200 1000 800 600 400 200 0 1900

1910

1920

1930

1940

1950

1960

1970

1980

1990

2002

Year

figure 1.1. Immigrants Admitted: Fiscal Years: 1900 –2002

barred the entry of Chinese laborers. It was the first in a series of anti-Asian immigration laws: The Immigration Act of 1917 established an Asian-barred zone; the Immigration Act of 1924 would exclude immigrants from countries across the Asian continent. Even after the McCarran-Walter Act of 1952 formally abolished the Asian-barred zone, restrictive quotas allowed only a trickle of immigrants from Asia and the Asian Pacific.11 After the Second World War, American leaders felt pressure to make the nation’s immigration laws conform to international norms. The defeat of Nazi Germany had discredited racial theories of nation building in international circles. The Geneva Convention Relating to the Status of Refugees (1951), for instance, called for contracting states to admit refugees without discrimination as to race, religion, or country of origin. Twenty years after the end of the war, the United States finally removed national-origin and racial immigration quotas with passage of the Hart-Cellar Act. Hart-Cellar allotted an equal quota of 20,000 visas per year to nations in the Eastern Hemisphere and set a 170,000-visa hemispheric limit. It also set the limit for nations in the Western Hemisphere at 120,000 visas. As Reimers (1992, 123) notes, Congress traded the termination of national-origin quotas and restrictions on immigration from Asia and the Asian Pacific for this limit on immigration in the Western

01-S3453 1/11/06 3:05 PM Page 6

6 Introduction Hemisphere. The restriction of immigration from Latin America and the Caribbean was a policy goal for a number of lawmakers. The new immigration law placed a higher priority on admitting the family members of U.S. citizens and residents than on admitting skilled workers. Its designers did not anticipate that giving a priority to family members would lead to chain migration, a trend that would dramatically change the ethnic makeup of immigrants in the last third of the twentieth century. In annual immigration intake, the main countries of origin shifted away from Europe to the developing world. In the 1950s, more than 65 percent of immigrants admitted to the United States came from Europe and Canada. By the 1980s, just 13 percent originated in Europe or Canada, while the great majority came from Asia and Western Hemispheric countries other than Canada (Borjas 1994, 1669). It is not hard to understand the underlying motivations of immigrants. Wage disparities between developing and industrialized nations create strong incentives for people to leave their homeland. Also civil strife and natural disasters in source countries have led to large-scale movements of refugees and asylum seekers (Hatton and Williamson 1994). More puzzling is the congressional response to these immigration pressures.12 When reformers initiated proposals that would have reduced legal immigration after 1965, legislators repeatedly rebuffed them. Congress kept intact family-unification policies known to lead to chain migration even through cycles of economic recession, when public pressure to restrict immigration tends to be strongest. The most expansive pressure on immigration rolls in recent decades has come from family-based, not employment-based, admissions and from the legalization programs that allowed undocumented immigrants to regularize their status under the Immigration Reform and Control Act of 1986 (Table 1.1).13 Although various national opinion polls showed that in the late 1980s, nearly half—in some studies, slightly more than half— of Americans favored cutbacks in immigration, in 1990, Congress actually decided to increase immigration limits by 40 percent.14 And in 1996, Congress left total numerical limits on legal immigration in place despite a push by congressional Republicans to restrict immigration while they held a majority in both houses. In the post-1965 period, however, immigration policies were not uniformly expansionist. Refugee and asylee policies selectively favored immigration from some countries while restricting it from others.

1995 720,461 4,267 716,194 323,458 238,122 85,336 220,360 114,664 57,712

Type and class of admission

Total, all immigrants Total, IRCA legalization1 Total, nonlegalization Preference immigrants: Family-based immigrants Employment-based immigrants2, 3 Immediate relatives of U.S. citizens4 Refugees and asylees Other immigrants

915,900 4,635 911,265 411,673 294,174 117,499 300,430 128,565 70,597

1996

601,516 — 601,516 269,328 211,809 57,519 218,575 91,840 21,773

1987

798,378 2,548 795,830 303,938 213,331 90,607 321,008 112,158 58,726

1997

643,025 — 643,344 259,499 200,772 58,727 219,340 81,719 82,786

1988

654,451 955 653,496 268,997 191,480 77,517 283,368 52,193 48,938

1998

1,090,924 478,814 620,699 274,833 217,092 57,741 217,514 84,288 44,064

1989

646,568 8 646,560 273,700 216,883 56,817 258,584 42,852 71,424

1999

1991

2001

973,977 163,342 827,888 329,321 213,123 116,198 235,484 117,037 146,046

1992

2002

904,292 24,278 891,130 373,788 226,776 147,012 255,059 127,343 134,940

1993

849,807 1,064,318 1,063,732 421 263 55 849,386 1,064,055 1,063,677 342,304 411,338 362,037 235,280 232,143 187,069 107,024 179,195 174,968 347,870 443,035 485,960 65,941 108,506 126,084 93,271 101,176 89,596

2000

1,827,167 1,123,162 720,015 275,613 216,088 59,525 237,103 139,079 68,220

fiscal year

1,536,483 880,372 669,170 272,742 214,550 58,192 231,680 97,364 67,384

1990

804,416 6,022 798,394 335,252 211,961 123,291 249,764 121,434 91,944

1994

1. The legalization programs under the Immigration Reform and Control Act of 1986 went into effect in 1989. 2. Includes spouses and children. 3. Includes immigrants issued third-preference, sixth-preference, and special-immigrant visas prior to fiscal year 1992. 4. Effective in fiscal year 1992, under the Immigration Act of 1990, children born abroad to alien residents are included with immediate relatives of U.S. citizens for calculating the annual limit of family-based immigrants.

601,708 — 601,708 269,556 212,939 56,617 223,468 104,383 4,301

1986

Total, all immigrants Total, IRCA legalization1 Total, nonlegalization Preference immigrants: Family-based immigrants Employment-based immigrants2, 3 Immediate relatives of U.S. citizens4 Refugees and asylees Other immigrants

Type and class of admission

fiscal year

ta b l e 1 . 1 Immigrants Admitted by Type and Selected Class of Admission, Fiscal Years 1986 –2002

01-S3453 1/11/06 3:05 PM Page 7

01-S3453 1/11/06 3:05 PM Page 8

8

Introduction

For example, in May 1980, President Jimmy Carter declared a policy of welcoming Cubans with “open heart and open arms,” while Haitians were being seized at sea and subsequently denied access to U.S. courts (quoted in Mitchell 1992, 45). And despite civil war in El Salvador in the 1980s, the United States denied political asylum to most Salvadorans. In the 1980s, the federal government also stepped up its efforts to control illegal immigration. The problem had intensified after 1964, when Congress terminated the bracero program, which had allowed migrant farm laborers from Mexico to enter the United States. Seeking a legal source of inexpensive labor, increasing numbers of American farmers in the West and Southwest began to rely on illegal immigrants. In turn, the federal government moved to tighten its control over unauthorized border crossings, instituting measures to apprehend undocumented migrants as they crossed the U.S.-Mexico border and sanctions against employers for hiring them. These enforcement efforts proved mostly ineffective, however, as the number of unauthorized foreigners in the United States grew to an estimated 5 million by 1995 and to 9 million by 2000.15 One question that has elicited interest among immigration scholars is why legal-admissions policy in the United States remained expansive despite pressures to restrict immigration after 1965.16 My approach is to consider the varied dimensions of immigration policy and why outcomes were mixed. The system for admitting permanent immigrants remained robustly expansionist, and policy toward highly skilled and professional foreign workers was fairly generous; but proposals to start new agricultural guest-worker programs were repeatedly defeated with one major exception: In 1986, lawmakers agreed to a brief but important trial of a new guest-worker program in agriculture, but only on the condition that foreign workers could become eligible to apply for permanent residency after working for a designated time in the temporaryworker program. Clearly there has been fluctuation in policies toward admitting different categories of immigrants over the past decades. We find similar fluctuation in integration, or social incorporation, policies, the policies that help immigrants become part of U.S. society. In the mid-1990s, for example, Congress cut social benefits for immigrants; but within a few years, it had reinstated a number of them. Why do policy outcomes vary across different dimensions? The answer has to do with the way identity politics interacts with the politics of economic regulation in different policy areas. By identi-

01-S3453 1/11/06 3:05 PM Page 9

Introduction

9

fying those interactions, I hope to shed light on the sources of stability and change in immigration policy.

Ethnic Advocacy and Regulation Ethnic advocates seek inclusive admissions and social-incorporation policies, and this has brought them into frequent conflict with both reformers who favor restrictions and proimmigration business interests. For example, while reformers were asking for limits on immigration in the 1980s, ethnic advocates were lobbying to protect family-based visa categories. Generous admissions of relatives, they argued, would facilitate the long-term social integration of immigrants already resident in the United States. In the same decade, they also worked to add special categories that would allow undocumented immigrants who had already lived in the United States for some significant length of time to apply for regular status. In both cases, ethnic advocates were adding expansionist pressure at points in the system where inclusive admissions and integration policies converged. Ethnic advocates also interjected rights demands into controversies over the economic regulation of immigration. In disputes over temporary workers, rights advocates opposed measures that would admit guest workers without granting them membership rights—for example, the right, earned after a probationary period, to permanent residency with a path to naturalization. This advocacy worked in the direction of restraining an expansionism driven by employers’ demand for unregulated access to inexpensive migrant labor. If we consider the two effects together, the course of action favored by ethnic advocates was a moderate one. By supporting inclusive policies with respect to both admissions and social incorporation, they favored a policy path that modulated new admissions according to the nation’s capacity to integrate the newest immigrants. To be sure, ethnic rights advocates were only one of many types of groups lobbying Congress, and they had to form coalitions with other forces to try to sway the votes of undecided legislators. But in general, ethnic advocacy worked as an independent influence on policy outcomes. It was not simply reinforcing the expansionist policies set in motion by economic forces. The odd coalitions that form in immigration politics are complicated by the mixed stands taken by organized labor. In the mid-1950s, the American Federation of Labor (AFL) and the Congress of Industrial Organizations (CIO) merged to form the AFL-CIO. The newly merged

01-S3453 1/11/06 3:05 PM Page 10

10

Introduction

federation abandoned the restrictionist position of organized labor, calling for an increase in immigration and an end to national-origin quotas (Tichenor 2002, 204). At the same time, the AFL-CIO continued to support policies that protected the economic interests of union members. In the 1960s, for example, it pressed Congress to pass sanctions against employers who hired illegal immigrants. But in the late 1980s, leaders of the federation realized that their advocacy of employer sanctions may have been a mistake because it led to discrimination against all foreign workers. In 2000, the AFL-CIO formally abandoned its support of employer sanctions, having questioned their effectiveness for some years, as several influential union affiliates aggressively sought to incorporate both legal and undocumented workers into their membership ranks. The AFL-CIO actively joined in coalitions with ethnic organizations to battle for immigrants’ rights (Haus 2002, 98). Because both the AFL-CIO and the ethnic advocacy organizations held liberal ideological views, this was not an odd coalition in the sense described above. After the AFL-CIO adopted a liberal stance on immigration, leadership of the anti-immigration movement was left in the hands of organizations that had originated in one strand of the environmentalist movement. Egalitarian norms constrained the anti-immigration rhetoric of these organizations. They could not overtly stir racial animus toward immigrants in the way of traditional nativists. Their arguments for reducing immigration were more limited, pointing to problems of overcrowding in American cities, competition between immigrants and natives for scarce resources, and the tendency of immigration to lower the wages of unskilled workers (Tichenor 2002, 237–238). Given these circumstances, the membership of immigration coalitions tended to shift over the course of a single policy battle, depending on which of many immigration issues was currently at stake. For instance, ethnic advocates and unions cooperated in support of family-based admissions, which employer lobbies generally supported too. In contrast, ethnic groups and unions opposed employers’ proposals to establish a traditional guest-worker program in agriculture in the 1980s and 1990s. And neither group aligned itself with the restrictive environmentalists because the environmentalists strongly opposed liberal family-based immigration policies. The tendency of immigration coalitions to shift along different issue axes is also reflected in patterns of congressional voting. Elected representatives respond to the constituencies that are proimmigration or

01-S3453 1/11/06 3:05 PM Page 11

Introduction

11

anti-immigration in predictable ways, issue by issue. I demonstrate the patterns by testing a model of floor voting in the House of Representatives on multiple immigration issues. From that examination, I draw several conclusions. As the number of foreign-born persons in a legislator’s district increases, the legislator’s vote will tend to align more closely with the policy preferences of ethnic rights advocates. The reason for this responsiveness, I suggest, is that legislators listen more closely to the advice of ethnic advocates as ethnic constituencies grow in the district. Also, ethnic factors seem to influence legislators’ votes independently of unemployment in their district or other economic variables. In some situations—namely, when growers contribute to legislators through a political action committee (PAC) to indicate their support of a traditional guest-worker program—legislators respond to the proimmigration lobbying of these business interests.

Theories of Immigration Lawmaking Political scientists have engaged in a lively discussion of the sources of inclusive and expansive immigration policies in the United States since 1965.17 One ongoing debate weighs the relative importance of societal interests and cultural-political factors in shaping policy outcomes. Freeman (1995a) has proposed an interest-driven model of immigration policymaking. The argument is built on J. Q. Wilson’s (1980) theory of client politics.18 In Freeman’s model of client politics, small, tightly organized groups that would benefit from a certain policy work with elite political actors—that is, lawmakers—to attain their policy goals in a setting largely removed from the influence of the public. Immigration policymaking fits the client-politics model because its beneficiaries are concentrated while those who pay its costs are diffuse. Specifically, employers and ethnic groups tend to benefit from liberal immigration policies: employers because they get an inexpensive source of labor, and ethnic groups because those policies tend to increase familybased immigration. The costs of the policies are borne largely by taxpayers, a widely dispersed group that pays for the social services used by immigrants; they are also borne by nonunionized native workers who compete with immigrants in labor markets. To the extent that employers and ethnic groups are intensively organized (i.e., have concentrated interests), they hold a political advantage in contests with diffuse interests: They have lower collective-action costs and can therefore mobilize

01-S3453 1/11/06 3:05 PM Page 12

12

Introduction

themselves more easily. This explains why policy outcomes are biased in the direction of expansive interests. In addition, Freeman argues, the client form of politics is supported by an antipopulist norm, which has prevented politicians from exploiting racial, ethnic, or anti-immigrant fears among the populace. Constrained discourse around immigration policy is a structural factor common to liberal democracies; it has prevented lawmakers from arguing over the ethnic makeup of migrant streams, for example. Brubaker (1995) disagrees with Freeman on this point. He argues that the adoption by the United States of a universal selection policy was not premised on democratic structure but on “changes in the shape and boundaries of discursive fields” that took place fairly recently in a specific cultural, political, and historical context (905). In a rejoinder to Brubaker, Freeman (1995b) insists that the U.S. decision to adopt universal selection criteria in 1965 should be considered irreversible in practical terms because it reflects the core values of a liberal democracy (909). Joining the debate, Joppke (2005) affirms Freeman’s view that the antipopulist norm “has come to resonate with the ethos of liberal democracy,” adding that it would be reversed only “at the price of a wholesale civilization break or regression” (19).19 The simplicity of Freeman’s client-politics model is one of its strengths, allowing it to be used in cross-national comparative studies. Especially when applied to the recent history of immigration politics in the United States, the theory provides a convincing explanation for the privileged position that business groups enjoy in the processes of immigration lawmaking. One limitation of the theory, however, is that it does not take into account the multiple dimensions of immigration policy. Although ethnic advocates have supported proimmigration policies for permanent immigrants, they also have lobbied against the expansion of traditional guest-worker programs, which do not permit migrant workers to stay in the host country and eventually apply for naturalized status. Immigration scholars have added new insights to the debate over the source of trends in immigration policy by examining the role of political institutions. Fitzgerald (1996) takes an “improvisational institutionalist” approach to the problem of explaining immigration-policy outcomes. In his view, policymakers rely on innovation to solve problems, responding to specific issues with specific fixes. He identifies three different sets of issues in immigration policy: permanent residency, refugees, and unsanctioned migration across the U.S.-Mexico border;

01-S3453 1/11/06 3:05 PM Page 13

Introduction

13

and he identifies a distinctive mode of policymaking on each dimension. In the area of permanent residency, lawmaking follows a pluralist logic. In the treatment of refugees, realism in the sphere of international relations drives decision making. And on issues of illegal workers from Mexico, “class conflict” theories best describe the decision process (59). Although Fitzgerald’s framework rightly differentiates sets of issues, it is not clear that such distinct modes of policymaking shape outcomes, especially when different policies are packaged in the same bill. Tichenor (2002) proposes an “historical-institutionalist” framework for the analysis of immigration politics, an approach that highlights the role of state actors and state structures. He describes “four interlocking processes” that help explain alternating restrictive and expansionist policies in the United States from the colonial period through the 1990s (29). The first process stems from the changing dynamics of national governing institutions, which create opportunities for political actors to try to initiate new policies; the second is the process by which changing coalitions of interests form; the third entails the influence of professional experts; and the fourth consists of international pressures that influence opportunities for policy change in the domestic arena.20 In the post-1965 period, Tichenor adds, two kinds of politics promoted expansionism. In the first, elite lawmakers were insulated from the public. In this insulated realm, several factors were significant: the “ideological convergence of liberal and conservative politicians and interest groups in favor of immigration,” the lobbying expertise of proimmigration forces, and “international pressures,” including global competition for trade (246). The second kind of politics unfolded in the public realm and consisted of the enfranchisement of Latinos. In the mid-1990s, Tichenor notes, anti-immigrant measures supported by the Republican Party brought about a surge of Latino participation in electoral politics. Latino naturalization rates increased dramatically; in 1996, for example, more than 1 million Latinos were naturalized. And between 1992 and 1996, voter registration among Latinos increased by almost 29 percent. The Republican Party responded by moderating its position on immigration to court Latino voters. It cooperated with President Bill Clinton in 2000 to pass the Legal Immigration Family Equity (LIFE) Act, for instance. Among the provisions of that act was an opportunity for certain immigrants to regularize their status who were unable to do so under a 1986 amnesty (285 –287). I found, however, that large-scale enfranchisement was not a prerequisite for ethnic groups’ attaining a voice in immigration policymaking.

01-S3453 1/11/06 3:05 PM Page 14

14

Introduction

Several Latino nonprofits had acted as immigration rights advocates in Washington, D.C., since the early 1970s. Latino and Asian American nonprofits and other ethnic and civil rights advocates played a pivotal role in legislative moves to block cuts to family-based immigration in the 1980s and 1990s. Cooperating with organized labor, they also helped obstruct California growers’ efforts to institute a traditional guest-worker program in the mid-1980s. Gimpel and Edwards (1999) attribute the long-term stability of legaladmissions policy after 1965 to a bipartisan consensus forged at that time. They argue that this consensus was possible because immigration admissions were not controversial in the mid-1960s. Immigration and unemployment levels were low, the national economy was robust, and objections to the Hart-Cellar Act on grounds that it would change the racial composition of the United States were considered racist or exaggerated (109). The consensus began to fray in the mid-1990s, as the two parties grew increasingly polarized, and as the public costs of a generous immigration policy seemed increasingly prohibitive in the eyes of taxpayers in states with large immigrant populations. One watershed event occurred in 1994, when the Republican Party in California won the gubernatorial election by exploiting immigration as a wedge issue and drawing dissatisfied taxpayers and workers away from the Democratic Party’s base.21 Immediately afterward, congressional Republicans added legal immigration cuts to their Contract with America. Gimpel and Edwards show that votes on immigration bills divided along party lines in the 1980s and 1990s. Although votes were partisan, I found that proimmigration advocates were able to block restrictive reform by winning over a handful of Republican legislators. It is true that in 1996, Congress retreated somewhat from an expansive immigration policy by cutting welfare benefits for immigrants and making it more difficult for prospective immigrants to show financial independence. However, by 2000, Congress had enacted the LIFE Act and had increased visa allocations for temporary skilled workers (Tichenor 2002, 287).

Identity Politics and Representation Ethnic advocacy organizations informally represent Latinos and Asian Americans in national policymaking circles. These organizations are not accountable to constituents in periodic elections. Although they lack formal standing as political representatives, the ethnic nonprofits

01-S3453 1/11/06 3:05 PM Page 15

Introduction

15

are able to leverage their status as spokespersons for ethnic minority interests. How is this possible? The NCLR and NAPALC serve as illustrations. These two organizations fit Gutmann’s (2003) definition of ascriptive associations.22 Ascriptive associations “organize around characteristics that are largely beyond people’s ability to choose, such as race, gender, class, physical handicap, ethnicity, sexual orientation, age, and nationality” (117). At the same time, both the NCLR and NAPALC are interest groups because they pursue the common goals of their members. And they both are what Gutmann calls “justice-friendly”: [An] ascriptive association is maximally justice-friendly if it struggles against discrimination and allies with other associations that share this aim. Ascriptive associations that are narrowly self-centered do not make such alliances, but many ascriptive associations have strategies as well as moral reasons not to be narrowly self-centered. The NAACP is a model in this regard; it explicitly aims to end discrimination for all individuals, even as it focuses its energies on African Americans. (204 –205)

In their statements of program goals, the NCLR and NAPALC did not explicitly include a commitment to end discrimination for all individuals. The NCLR formed with a mission to reduce discrimination and improve life opportunities for Latinos.23 For its part, NAPALC’s mission was to advance the human and civil rights of Asian Pacific Americans.24 Yet in practice, as Hula (1999) has pointed out, civil rights organizations tend to be “long-term coalition experts,” more inclined to form long-term coalitions than are trade associations, whose cooperation is usually of shorter duration.25 Both the NCLR and NAPALC have been active members of the Leadership Conference on Civil Rights (LCCR), for example, which specializes in coordinating national legislative campaigns to advance the basic civil rights of all people in the nation.26 In the 1990s, Latino and Asian American groups cooperated with like-minded civil rights groups; but they also formed coalitions with certain social conservatives and libertarians in an effort to preserve family-unification rights in federal immigration law.27 Family-based immigration is supported by a basic human right recognized in international conventions.28 Adopting a stance of universal rights helped these ethnic groups form broad left-right coalitions. Even among the civil rights organizations there were sometimes tensions. In the 1980s, for instance, the major Latino advocates—the NCLR, MALDEF, and LULAC—favored an amnesty for undocumented

01-S3453 1/11/06 3:05 PM Page 16

16 Introduction immigrants and opposed the imposition of sanctions against employers who hire undocumented workers. The NAACP supported the amnesty provision but also supported employer sanctions. That the NAACP would support measures to curb illegal immigration is not surprising. Researchers have found that immigration has negative effects on the employment opportunities of Americans who have not completed a high school education (Borjas, Freeman, and Katz 1992), and African Americans are disproportionately represented in the ranks of unskilled workers. After passage of the Civil Rights Act of 1964, structural shifts in the economy displaced many blacks from relatively well paying jobs in manufacturing as the flight of industry from the cities left service jobs in the lowest-paying occupations or jobs requiring technical training (W. J. Wilson 1996). Middle-class African Americans saw improvements in their life opportunities, on average. But many African Americans still lived in neighborhoods of concentrated poverty (Cohen and Dawson 1994), and remained economically disadvantaged in terms of both income and individual or family wealth (Bobo 2004). Although Latino and black leaders in the civil rights movement could not find common ground on employer sanctions, Fuchs (1990) documents the full support that black legislators gave the Congressional Hispanic Caucus in the 1980s on illegal immigration policy. In the end, amnesty for undocumented immigrants was won (though not with as generous terms as the Latino leaders wanted); but the Latino advocates’ efforts in the 1980s to remove employer-sanctions measures from legislation failed. Ethnic advocates also used outside and inside strategies to influence legislators’ decisions. Interest groups in Washington commonly combine both approaches (Gais and Walker 1991). Outside strategies include techniques like letter-writing campaigns and flying in constituents from districts to meet with elected officials in Washington, D.C. Inside strategies include the policy briefs interest groups draw up for lawmakers and executive agency officials that predict the likely effects of proposed legislation on different constituencies. In the legislative battles over immigration, ethnic advocacy groups also helped representatives write legislation and frame the relevant issues for public audiences. Ethnic advocates saw the provision of services to new immigrants expand the social networks of their supporters, creating a social base that extended beyond formal membership circles. Traditionally, political representatives of immigrant communities have provided services as a link between themselves and their constituents. In writing about

01-S3453 1/11/06 3:05 PM Page 17

Introduction

17

the nineteenth-century political machine, Boorstin (1973) describes how Irish politicians made themselves into a “personal service agency”: “They were an employment office. . . . They brought food to the hungry, and medicine to the sick. . . . They organized a benefit social for an impoverished widow, or for the family of the man crippled on his job. . . . They worked full time and year-around” (259). Ethnic nonprofits functioned much the same way, providing some of the services new immigrants needed when they settled in this country.29 Those services included help finding work, navigating systems of public health care and education, and adjusting to living in a new cultural environment. In the 1980s, community nonprofits coordinated advocacy efforts for the legalization of undocumented immigrants through national ethnic rights groups, while acting as local administrators of legalization programs. The legalization programs granted residency rights to undocumented immigrants who could show they had been residents of the United States for some specified length of time. As ascriptive associations, the NCLR and NAPALC promoted ethnic solidarity to mobilize their social base of supporters. Insofar as they worked with allies to advance universal civil and human rights, I suggest they practiced an inclusive form of identity politics.30 As Brubaker (2004) has argued, as an analytical category, identity is often fraught with problems of ambiguity, but the term identity politics refers to something more specific: Leaders try “to persuade people that they are one; that they comprise a bounded, distinctive, solidarity group; that their internal differences do not matter, at least for the purpose at hand—this is a normal and necessary part of politics, and not only of what is ordinarily characterized as ‘identity politics’” (60 – 61).

An Unresolved Issue: Guest Worker Policy In the early twenty-first century, Congress is entertaining new proposals to comprehensively reform the system of regulating economic immigration. Industrial trade lobbies have called for programs that would admit large numbers of guest workers to fill labor needs in industry. Deliberations on these proposals were temporarily suspended after the terrorist attacks of September 11, 2001. In the wake of those attacks, as the priorities of the federal immigration system shifted toward the protection of national security, the need for reform of that system became even more evident to the nation’s leaders. Enforcement mechanisms were designed to distinguish criminal or terrorist elements, who pose a threat to

01-S3453 1/11/06 3:05 PM Page 18

18

Introduction

national security, from undocumented migrants who come to the United States to fill needs in the labor force. One of the most troubling features of the system is that it creates opportunities for the abuse of millions of undocumented immigrants who are working in the United States. The threat of deportation allows unscrupulous employers to take advantage of the unprotected status of illegal immigrants. That those employers often choose to hire vulnerable undocumented workers rather than legal residents or citizens has troubling implications for American workers, especially those who work in low-skill occupations. One indication of recent change in the politics of immigration appears in the agricultural industry, where there are unprecedented efforts to cooperate across the traditional business-labor divide. Growers who employ migrants from Mexico and other low-wage countries have long been adversaries of farmworkers’ unions. But in 2000, American growers agreed for the first time to negotiate directly with the United Farm Workers over temporary-worker policy in agriculture. Eventually the growers agreed to support an earned right for temporary workers to stay in the United States in exchange for the union’s agreement to temporarily loosen regulations on wages and the housing provision in the existing foreign-farmworker (H-2A) program. Crafted in legislative language, the proposal received sixty-three cosponsorships in the Senate in the 108th Congress. This agreement between economic actors indicated a growing recognition on the part of business and labor that new systems regulating migrant workers are required in an age of mass migration and globalization. In early 2004, President George W. Bush proposed the creation of a new visa program that would allow temporary foreign workers to apply for jobs in the United States after registering in an employment database that would make the jobs available first to Americans. Although the plan did not tie workers to a single employer—as traditional guest-worker programs had done—it did require that temporary workers return home after their work eligibility ends. Those who want to become permanent residents of the United States would have to return to their home country and apply through regular channels; they would not automatically earn the right to permanent residency by virtue of their work record in this country. The Bush proposal was vague but reflected the demands of the Essential Worker Immigration Coalition (EWIC), which in 2004 represented thirty-six of the nation’s largest business and trade associations, including the U.S. Chamber of Commerce. During the spring of 2005,

01-S3453 1/11/06 3:05 PM Page 19

Introduction

19

Senators John McCain (R-AZ) and Edward Kennedy (D-MA) introduced legislation that not only met EWIC’s demands but also responded to the concerns of certain labor unions and of civil rights groups representing immigrants and ethnic minorities.31 The McCainKennedy proposal would allow workers to enter and fill low-skilled jobs after the jobs are posted in a national bank, a system that would give Americans the first opportunity to fill openings. The bill would create a three-year temporary visa renewable for up to six years. After working for four years in temporary status, workers could get in line to apply for permanent residency. Pending a decision on permanent residency, temporary status would be extendable in one-year increments. Most controversial, undocumented immigrants already living in the United States could apply for the temporary-worker visa if they paid a substantial fine as a penalty for their initial illegal entry. The McCain-Kennedy proposal faced stiff opposition from critics. Some lawmakers have opposed any measure that would allow undocumented immigrants to stay in the United States even if they pay a punitive fine; others have criticized the proposal because it does not provide an infrastructure for processing temporary-worker applications or for deporting illegal aliens. Still others argue that enforcement at the nation’s borders must be effective before a new guest-worker program is implemented. In the course of debates over immigration reform, lawmakers have been presented with proposals that have tried to address the problem of illegal immigration over the long term. Papademetriou (2002) suggests a process by which an undocumented immigrant could progress from illegal to legal to permanent-resident status. The first step would require the undocumented immigrant to register; over time, this would help bring the illegal immigrant population in the United States aboveground. After registering, the immigrant could earn “points” over a designated number of years toward the goal of legal and then permanentresident status. For example, points could be earned for demonstrating steady employment; for paying taxes; for not having a criminal record; and for “certain benchmarks of ‘civic engagement,’” signs of integration in community life (5). In Papademetriou’s argument, this process of “earned regularization” sits in sharp contrast to the concept of amnesty, which the federal government grants out of its generosity to undocumented immigrants after they have proved continuous residency. In 2002, Papademetriou envisioned the earned-regularization program as part of a larger proposal for the United States and Mexico to

01-S3453 1/11/06 3:05 PM Page 20

20

Introduction

jointly address migration policy by striking a “grand bargain.” The idea is based on the growing recognition among political scholars that in a global information economy, there is a need for forward-looking international systems for coordinating the flow of labor and capital between nation-states (see, for example, Sassen 1996). Papademetriou’s grand bargain would entail three parts. The first part is the registration of undocumented immigrants in the Unites States followed by the earnedregularization program. The second part is the establishment of a temporary-worker program for Mexican workers that treats participants with dignity. The third part is the creation of a new border security agreement between the United States and Mexico (2).32 As lawmakers address these and other issues of immigration reform, it will be useful to evaluate what political factors led to or impeded congressional reform of immigration laws in recent decades. In the 1980s, as we will see, agricultural employers did not have the political strength to push a new guest-worker program through Congress without ceding migrant workers the opportunity to apply for permanent residency. To win a congressional majority for a temporary-worker program (since expired), it was necessary to pace economic admissions to the capacity of U.S. society to incorporate those admissions.

Organization of the Chapters Chapter 2 provides an overview of where interests groups stand on immigration policy. The discussion challenges the view that immigration policymaking is captured by expansionist special interests. The chapter then turns to analyze how and why coalitions of interest groups form and shift along multiple issue axes. Chapter 3 examines the bipartisan compromise that enabled passage of the landmark Hart-Cellar Act of 1965. By removing national-origin quotas and placing a priority on family-based immigration, the HartCellar Act defused the conflict over race in U.S. visa policy. By delegating regulatory authority over permanent-labor immigration to the executive branch, the act also lessened the volatility of future interactions between employers and unions. But the failure to delegate the issue of temporary-worker admissions would lead to recurring controversy in Congress in subsequent decades. Chapter 4 analyzes the position of Hispanic rights groups and labor unions on issues affecting the working conditions and wages of migrant workers and undocumented immigrants. The first part of the chapter

01-S3453 1/11/06 3:05 PM Page 21

Introduction

21

shows how nonprofit groups representing Hispanics changed the politics of employer sanctions and traditional guest-worker programs by interjecting a rights-oriented philosophy. The second part examines the growth of immigrant sectors in union memberships, which has led some unions to ease their restrictionist efforts and others to take an openly proimmigration stance. The first part of Chapter 5 examines the legislative strategies employed by ethnic rights groups and other interest groups to obtain their favored policy in contests over legal- and illegal-immigration reform during the 1980s. Two specific legislative cases are analyzed: passage of the Immigration Reform and Control Act of 1986 and passage of the Immigration Act of 1990. The second part of the chapter tests several hypotheses for predicting a legislator’s vote on immigration issues by characteristics of the legislator’s constituents and district. Controlling for the lawmaker’s party membership and ideology, the most consistent predictor of a House member’s vote on immigration policy is the size of foreign-born or ethnic populations in his or her district. With few exceptions, the larger those populations, the more likely lawmakers are to respond to them. Chapter 6 revisits the controversy over legal and illegal immigration in the 104th Congress, which enacted the Illegal Immigration Reform and Immigrant Responsibility Act of 1996. The most important political difference from the 1980s was Republican control of both houses of Congress. But the Republican Party was hardly monolithic in demanding cuts in legal immigration, nor was the party unified in its attitude toward instituting new methods of enforcement that would increase the power of the government to collect information on individuals. A skillful legislative strategist favoring immigration expansion could frame issues to further exacerbate splits in one or both parties. This was precisely the tactic taken by a coalition of nonprofit ethnic and humanitarian groups whose purpose was to block proposals to cut family immigration and safeguard the rights of immigrants. The second part of the chapter examines the predictors of a hypothetical House member’s votes on the IIRIRA and the implications of the multidimensionality of immigration policy. The analysis supports the argument that distinct issue dimensions—legal admissions, traditional guest-worker programs, general enforcement, enforcement and privacy, and social incorporation—give rise to shifting voting coalitions in the House, some cutting across the liberal-conservative divide and others cutting along it. Chapter 6 tests a model of voting that includes

01-S3453 1/11/06 3:05 PM Page 22

22

Introduction

Hispanic and foreign-born variables as predictors; an appendix to Chapter 6 tests models using other combinations of variables, including population indicators of ethnic-specific groups (e.g., Mexican Americans, Japanese Americans, etc.), campaign donations from individual unions, and ideological measures of economic and social conservatism. Finally, Chapter 7 discusses the implications of this book’s argument in light of recent policy debates and the growing political power of ethnic minority voters. In considering the features that would define a rational and humane system for regulating admissions of low-wage workers to the United States, the chapter suggests that openness to immigration ought to be combined with grants of provisional membership rights to temporary workers and protections for fundamental human rights for all immigrants, documented or undocumented. Ethnic nonprofits representing Latinos and Asian Americans have been at the forefront of advocating this approach, and they may well find themselves in a pivotal position as negotiations on immigration reform unfold in the coming years.

02-S3453 1/11/06 3:06 PM Page 23

chapter

Interest-Group Goals

2

what policy goals do interest groups pursue in the area of immigration? This chapter examines the policy preferences of several categories of interest groups as revealed in contests over immigration policy after passage of the Hart-Cellar Act in 1965. As we will see, some groups are concerned primarily with the economic impact of immigration policies and focus their attention on maximizing the income of their members. Others are interested mainly in cultural effects— on native residents of the United States (How do immigrants change citizens’ quality of life?) or on immigrants themselves (What political and civil rights will be extended to immigrants, and will they be absorbed into the native culture?). There is a common misperception that all interest groups in the immigration-policy field pursue narrow goals. According to Briggs (1996), for example, the process of making immigration policy tends to be captured by special interests with private agendas “that simply ignore any concern for the national interest” (10); moreover, those special interests have learned to form coalitions to lobby for one another’s private ends (11). In Briggs’s account, immigration policy is both driven by politics and divorced from economic reality. Congress tries to appease special interests by legislating expansive policies, bending to the demands of business to alleviate labor shortages by recruiting immigrant workers, rather than training American citizens to fill these shortages (14). Senator Alan Simpson, the veteran immigration reformer, described the role of interest groups in a similar manner when I interviewed him in 1995.1 Interest groups had continually put up obstacles to his efforts to

02-S3453 1/11/06 3:06 PM Page 24

24

Interest-Group Goals

bring about comprehensive reform of the immigration system in the 1980s and 1990s. Each immigrant category seemed to have its own constituency—relatives of citizens, temporary or permanent workers, diversity immigrants, investors. Any measure that would have reduced numerical quotas in a specific visa category tended to generate opposition from “the groups,” Simpson remarked. And as interest groups pressed congressional representatives to add one amendment after another to immigration bills he had sponsored, the text of the legislation tended to expand enormously. Simpson described the process in colorful language: “Immigration [legislation] is like bear meat: The more you chew on it, the more it grows.” In their efforts to block immigration controls, do interest groups place their narrow goals above the national interest? To answer this question, it is important to clarify which elements of the national interest are most important. Briggs (1996) suggests that economic development ought to be placed in the forefront, and I agree that it is one important goal. But I propose using as an evaluative standard Sen’s (1999) definition of “development as freedom.” According to Sen, development is the process of expanding the range of substantive freedoms enjoyed by the people of any nation. A narrower view of development identifies it with “the growth of the gross national product, or the rise in personal incomes, or with industrialization, or with technological advance, or with social modernization” (3). But Sen refuses to limit his definition of development to these indexes alone. He writes that “freedoms depend on other determinants, such as social and economic arrangements (for example, facilities for education and health care) as well as political and civil rights (for example, the liberty to participate in public discussion and scrutiny)” (3). Sen gives us a standard by which we can evaluate development not only in third-world countries, but also in industrialized nations. Take poverty indexes, for instance. From a narrow economic perspective, we might evaluate poverty by income levels alone. But as Sen suggests, it is more useful to think of poverty as basic “capability deprivation.” By expanding the definition of poverty, we change the criteria for measuring it and the possible outcome of any evaluation. For example, as Sen writes, African Americans have lower incomes on average than white Americans, but much higher incomes than people in China; but African Americans have less chance of surviving to old age than people do in China and other developing nations (21). Early mortality rates reflect poverty in the sense that people are being deprived of basic capabilities.

02-S3453 1/11/06 3:06 PM Page 25

Interest-Group Goals

25

Drawing directly on Sen’s argument, I suggest that immigration serves the national interest to the extent to which it furthers “development as freedom.” If we focus on the economic and cultural contributions immigrants make to a society, then the extent to which political and civil rights are afforded to the newest members of a society is one baseline measure of that society’s development. In that light, the policy goals of different categories of interest groups stand in sharp contrast to one another. As I illustrate in later chapters, both unions with large numbers of immigrants on their rolls and business groups that rely on immigrant labor pursue the goal of maximizing the income of their own members; but the principal Latino and Asian American lobbyists have as one goal the promotion of a rights-based policy agenda with universal implications. Although coalitions form between business and ethnic groups on some issues—the expansion of family-based and employment-based immigration, for example—the rights orientation of ethnic lobbyists has tended to temper the demands of certain business groups for foreignworker programs that would admit large numbers of migrant laborers.

Crosscutting Cleavages Hammar (1985) distinguishes two principal categories of immigration policy: immigration control and immigrant integration. In the control category, he includes regulatory processes as well as policies that govern the “condition of immigrants,” such as length of stay and enforcement. Integration policy addresses the social incorporation of immigrants who are already residents of the country. For example, social incorporation policy would define the immigrants’ eligibility for public welfare benefits and their right to join unions. I modify Hammar’s classification scheme by distinguishing three categories of policy—admissions, incorporation (or integration), and enforcement—and numerous subcategories. The many and sometimes disparate issues represented by these subcategories form the basis of a multidimensional immigration politics. Admissions policy determines the number and origin of immigrants selected in any given year out of a pool of visa applicants; it also sets the terms of admission, including length of stay. The selection system defines the basic categories of immigrants and the number of visas allocated to each category. The possible categories are many, including permanent economic migrants (investors, professional and skilled

02-S3453 1/11/06 3:06 PM Page 26

26

Interest-Group Goals

workers, and unskilled workers), relatives of citizens and residents, temporary migrants, refugees, and political asylees.2 Incorporation policy defines the rights of immigrants and temporary migrants in terms of their social and political integration. Those rights include permission to live permanently and work in the United States, to join unions, to participate in federal or state social insurance plans, and to become citizens. Incorporation policy also addresses requirements for English-language proficiency. Enforcement policy is a separate dimension of immigration policy. Enforcement policy determines the measures used to secure the nation’s borders from illegal crossings and to detect violations of immigration laws, and sets the penalties for those violations. Enforcement policy— because it must balance the need for national security with the protection of civil rights—almost invariably engenders controversy. Of these many immigration issues, some strongly align Democrats on the left and Republicans on the right, while others foster intraparty divisions by cutting across ideological and party lines. As Gimpel and Edwards (1999) show, battles over immigration became increasingly partisan in the mid-1990s, when the acrimony that marked debate over New Deal redistributions in the 1930s resurfaced. On one side were the immigrants; on the other, taxpayers who felt burdened by the costs of transfer payments to support benefits and services to those immigrants (204). About 75 percent of all U.S. immigrants reside in just six states— California, Florida, Illinois, New Jersey, New York, and Texas. The issue of public costs is most salient in these states and in municipalities that are responsible for the delivery of public goods and services to large numbers of immigrants.3 At the federal level, economists argue, the taxes paid by most immigrants exceed the cost of federal services consumed; however, at state and local levels, the taxes paid by most immigrants are probably less than the cost of state and local services consumed.4 Although Republican ideology, which favors small government, clearly aligns with the concerns of taxpayers on immigration costs, the issue affects people who identify with both political parties. And in some states, particularly border states, public discontent with the costs of immigration is closely linked to concerns with illegal immigration. Those concerns focus not only on public spending and taxes, but also on the ability of the government to control the country’s borders and criminality.5 Taxpayers’ economic interests are also distinct from the psychological costs that some residents claim when immigration changes the

02-S3453 1/11/06 3:06 PM Page 27

Interest-Group Goals

27

cultural fabric of local communities. Although some residents value cultural diversity, others fear it (and even actively oppose it). These emotions are extremely difficult to measure, as are the cultural benefits and costs of immigration. Still, leaders of restrictionist movements may frame the issues in nativist terms or appeal to racism. And entrepreneurial politicians have used the discontent of taxpayers in states with large immigrant populations as a vehicle for election.6 That strategy becomes much less effective, of course, as immigrants are naturalized in increasing numbers and emerge as distinct voting blocs: Instead of targeting immigrants, politicians need to court their votes. Republicans and Democrats have found that their party coalitions are prone to internal fracture on the questions of which and how many immigrants should be admitted. In the Republican Party, multinational companies that employ technical personnel and the employers of lowwage immigrants are the strongest supporters of unregulated immigration. There are also widely dispersed constituents affiliated with both parties who are concerned about the costs of immigration to taxpayers and the security of workers’ jobs and wages. Divisiveness comes not only from large, well-organized constituencies; it also derives from distinct ideological wings within each party. Labor-market issues tend to unite the progrowth or free-market wing of the Republican Party in support of the unregulated movement of skilled and unskilled workers across borders. A culturally conservative wing of the party, however, uses the immigration issue to bolster arguments in a populist economic vein, claiming that free trade and free immigration are lowering American workers’ incomes. Some conservatives in the Democratic Party also champion this populist argument. On issues of regulating labor markets, the Democrats are just as sharply divided. Labor unions oppose admissions policies favored by employers when they allow workers into the country whose labor substitutes for that of union members. The Democratic Party has a large union constituency, but it, like the Republican Party, also depends on business supporters. The traditional Democratic coalition is particularly vulnerable to third-party candidates’ or conservative opponents’ strategy of driving a wedge between constituents who are racially conservative or who support populist economic arguments, and ethnic interests that favor liberal immigration policies. Furthermore, there is a fundamental tension among social liberals in the Democratic Party. These people support generous immigration and refugee policies for humanitarian

02-S3453 1/11/06 3:06 PM Page 28

28

Interest-Group Goals

reasons and, in some cases, because the policies continue a tradition of augmenting cultural diversity in American society. But social liberals are naturally ambivalent about immigration when they consider the impact of the competition between immigrants and natives in labor markets. As immigration battles unfold, the tendency of controversy to cleave both political parties has heightened the importance of negotiations among interest groups to devise policy packages that can win bipartisan support.7

Group Policy Preferences To understand the political alliances that develop around the making of immigration policy, it is important to understand the policy goals of the different players: employers, workers, civil rights advocates, religious groups, civil libertarians, and restrictionist environmentalists.

Employers The principal economic beneficiaries of a liberal policy on the admission of skilled and unskilled workers are the employers who rely on immigrant labor, or human capital, to lower production costs. By expanding the labor pool with immigrants, employers can hold wages down, even when native workers are available. And when there are actual shortages of workers to fill specific needs, immigrant workers not only increase profitability but also can fuel growth in the economy and the creation of new jobs. Employers are prone to support liberal familyunification policies when those policies enhance their ability to recruit foreign personnel. The economic benefits of freer access to immigrant labor vary with the characteristics of the industry and the labor market. Businesses that rely on immigrants naturally favor unrestricted borders. But for those companies that employ few immigrants— companies that have few economic motives for taking a stance on immigrant admissions policy— ideology may shape policy stands. One economic factor that influences a business organization’s interest in immigration is the scope of its operations. As a company’s technical and managerial operations expand globally, and with them the company’s investment in the human capital of employees living in foreign countries, it is clear that laws restricting employees’ freedom of

02-S3453 1/11/06 3:06 PM Page 29

Interest-Group Goals

29

movement across borders lessen the potential for the company to benefit from workers’ long-term or temporary relocation to work in the United States. In another example, some manufacturers face competition from foreign companies that have ready access to low-cost labor. In a global marketplace, companies in low-wage countries—although they may be disadvantaged by other economic factors— enjoy specific advantages in terms of production costs. To the extent that American manufacturers can attract low-wage workers to work in the United States, any regulation restricting the hiring of foreign or immigrant workers is likely to hinder short-term profitability and international competitiveness. A factor that drives some U.S. employers’ demands for immigrant labor is the seasonality of their production. Large numbers of foreign workers are willing to work on short-term contracts at what native workers would consider low wages. A long-standing controversy in the postwar period concerns the employment of migrant farmworkers as guest workers, recruited to harvest seasonal crops and then required to return to their home countries when the work is done. In the 1990s, companies in certain manufacturing and service industries, including software production and health care, increased their use of temporary foreign workers—skilled or semiskilled—to lower costs. U.S. employers do not generally set wages nationally, across industries, as is common among employers in other countries. They have found it cost-effective to lobby for increased immigration at the national level when a company’s own access to immigrant labor is jeopardized by new regulations or when other events induce labor shortages. The demands of employers for immigrant labor often can be met by legislation that adds tailored provisions to existing laws. Frequently, Congress creates special categories of employment visas, relaxing quotas on a case-by-case basis when occupational or industry-specific labor shortages are cited. For example, in 1989, in response to a nursing shortage, Congress allocated visas under a new category for nurses. As they do in the area of admissions policy, economic incentives have a lot to do with how employers respond to controversies over social incorporation policy. Companies that employ temporary foreign workers are more likely to oppose the liberalization of certain social rights—laws that accord foreign workers wage parity with native workers, for example. On social policies concerning immigrants’ eligibility for government-supported welfare, however, employers’ preferences may be defined more by ideological than economic considerations.

02-S3453 1/11/06 3:06 PM Page 30

30

Interest-Group Goals

Economic factors also shape employers’ stands on enforcement policy. Some of those that have come to rely on undocumented workers do not want strict enforcement of immigration laws that would impede their access to this labor source. In the 1980s, employers opposed financial penalties and other sanctions for employers that hired illegal immigrants. They also have opposed enforcement measures that would involve the government in employers’ hiring decisions. For example, both the National Association of Manufacturers and the National Federation of Independent Business took a strong stance in the 1990s against the institution of a national employment registry, which would have been used to verify the eligibility of every new worker hired. However, employer associations generally do not oppose enforcement measures that would help the government control illegal border crossings.

Workers When the flow of migrant workers into a country decreases wages for native workers and displaces native workers from jobs, both unionized and nonunionized workers can be expected to favor restrictions on immigration. The wage-displacement effects of immigration occur when immigrant workers are economic substitutes for native workers. The extent to which the migrant inflow reduces wages or takes jobs from native workers depends on the elasticity of the labor supply and the demand of jobs for which immigrant and native workers compete. Studies of the effects of immigration on wages and displacement yield contrasting results. In a study of Miami, Card (1990) concludes that the in-migration of 125,000 Cubans during the 1980 Mariel Boatlift had no appreciable negative effects on the wages and employment opportunities of local workers.8 In their study of the effects of trade and immigration on the national economy, however, Borjas, Freeman, and Katz (1992) find that “immigration flows may have contributed substantially to the poor labor market performance of the least educated American workers during the 1980s” by increasing the effective labor supply of high school dropouts (240).9 American unions vary substantially in their responses to different aspects of immigration policy. One source of that variation is the nature of the work a union’s members perform. We can expect a union to oppose liberal immigrant admissions when immigrant labor can readily replace union labor, when the work requires no special skills. The reasoning is straightforward: Unions do what is necessary to safeguard the

02-S3453 1/11/06 3:06 PM Page 31

Interest-Group Goals

31

wages and jobs of their members. But the more immigrant labor complements that of union members, the more a union is likely to lobby for liberal admissions policies. The potential for substitution or complementary effects of immigrant labor varies by skill level or occupation and across industries. For example, in the American garment industry, the entry into the labor market of immigrants employable as seamstresses can lower the wages of union seamstresses because the immigrant labor is a substitute for the native labor. But the labor of foreign tailors can complement that of seamstresses. If more foreign tailors are allowed to immigrate, this can increase the profitability of the industry and create more demand for seamstresses. If a garment workers’ union is composed mainly of seamstresses, the seamstresses would have reason to favor liberal immigration policies on the admission of foreign tailors. Two other sources of the variation we find in the policy goals of unions are incentives to control the labor supply and incentives to organize immigrant workers. Within the national union movement, the sharpest divisions on immigration policy in the 1980s and 1990s arose between craft and industrial unions. A union can be expected to favor a restrictionist policy as it becomes increasingly dependent for its power on restricting the labor supply; traditionally, craft unions are organized to do just that. If employers can gain access to immigrant or other nonunion labor with competitive skills through migrant recruitment networks, labor contractors, or other channels outside the union’s control, then those unions that especially depend on restricting the labor supply are likely to be substantially weakened by immigration.10 By contrast, industrial unions depend on industrywide organizing. When immigrants compose important sectors of the unionized labor force, union leadership faces strong pressure to support liberal family-based immigration policies in particular. Finally, a union organized along either craft or industrial lines may have strong incentives to support liberal admissions policies because immigrant or ethnic members or workers whom the union hopes to organize are advocating for those policies. The preferences of these union constituents can offset incentives working in the opposite direction, including the prospect that more immigration will lower industry wages. The organizing incentive is likely to be greater when a union is threatened by deunionization strategies or by the movement of plants overseas to low-wage countries. If policies that keep wages at lower levels are a means of saving a national union, the trade-off may be acceptable

02-S3453 1/11/06 3:06 PM Page 32

32

Interest-Group Goals

to union leaders.11 A public stance of generosity toward immigrants and immigration can be helpful in saving the union because it generates sympathy among immigrant workers, who are increasingly a source of union members in declining industries. In deciding its position on enforcement policy, the labor movement has faced a dilemma, balancing the goal of protecting union wages with that of expanding union membership among immigrants. In the 1970s, the AFL-CIO led a lobbying campaign in Congress to enact a law penalizing employers for hiring illegal immigrants. In the federation’s eyes, a stricter law would help protect native workers from employers’ attempts to lower wages by employing undocumented workers, whose status made them vulnerable to exploitation. But after employer sanctions were adopted and implemented, the unions found that some employers were using the sanctions to justify calling for immigration raids on their own workers or to fire immigrants engaged in union organizing. After expressing ambivalence on the issue for several years, in 2000, the AFL-CIO finally called for the repeal of employer sanctions. One commonality among American unions is opposition to traditional guest-worker programs, programs that as a rule do not grant foreign workers rights in the workplace or the opportunity to earn the right to apply for citizenship. The union movement as a whole objects to these programs because the foreign workers are bound to work for a single employer or else return to their native country—an arrangement that leaves them vulnerable to substandard wages and working conditions and other forms of exploitation. Moreover, in the past, employers have used large-scale temporary-worker programs to undermine unions among farmworkers (Martin 2003). More recently, temporary-worker programs have been used to recruit low-salaried professional or skilled workers in the software and health care industries. Although the union movement is not well established in certain professional sectors, the unions nevertheless see the practice as a threat to American workers. When employers look to foreign workers to fill positions, union leaders call for business to invest in training Americans. Where temporary-worker programs are already well established, unions advocate liberal policies on the social incorporation of temporary workers and champion the payment of prevailing wages and improvements in working conditions so that the temporary workers cannot be exploited and unions undermined. The policy agenda of the union movement calls for the enforcement of labor standards in industries and places of work where immigrants are concentrated.

02-S3453 1/11/06 3:06 PM Page 33

Interest-Group Goals

33

In the late 1990s, there were signs that some unions with large immigrant memberships were open to the establishment of probationary temporary-worker programs, which stand in contrast to traditional guest-worker programs. In a probationary program, foreign workers can apply for permanent residency after working for a specified time on a temporary-worker visa. In fact, growers and one farmworkers’ union have reached an agreement to advocate together for a probationary program. The interests of union members have shaped the liberal position of the labor movement on other aspects of immigration policy. Since the 1950s, the AFL-CIO has supported family-based admissions, giving them higher preference than employer-sponsored admissions. This stand not only serves the interests of union members who want family members to come to the United States; it also advances the movement’s interests in limiting the admissions of permanent laborers. In the area of social incorporation policy, unions generally favor liberal social programs as a safety net for all workers, including immigrant workers.

Civil Rights Groups Latino and Asian American rights groups make immigrant incorporation policy a high priority. Their goals include reducing discrimination, defending the human and civil rights of Hispanic and Asian immigrants, supporting social policies that help immigrants (e.g., programs to improve education and to gain fair access to housing, health benefits, and other government services). In the early to mid-1980s, the National Council of La Raza and the Mexican American Legal Defense and Education Fund made the enactment of legalization programs that would regularize the status of undocumented long-term residents their “top legislative priority.” 12 The NCLR’s goal was to maximize the number of legalization programs, to reduce the number of undocumented, and so exploitable, immigrants—a goal the organization argued was consistent with humane standards of treatment and a benefit to the community at large. The NCLR’s affiliates coordinated legalization programs in local communities, expanding their constituent base by tying newly legalized immigrants into their network. Advancing the argument that family unification is a general condition of well-being for the immigrant community, Hispanic and Asian American organizations have been at the forefront of advocating

02-S3453 1/11/06 3:06 PM Page 34

34

Interest-Group Goals

generous family-unification policies. In 1965, when Congress passed the Hart-Cellar Act, the United States made family unification the number-1 priority of its immigration policy. But policymakers underestimated the consequences of prioritizing family-based immigration at the time. One unexpected consequence was the surge in family-based immigration. Another was the limited education of new immigrants: Since 1965, immigrants entering the United States have had lower education levels on average than did previous generations of immigrants (Borjas 1994, 1676). This led immigration reformers in Congress to call for reducing the number of family-based immigrants in relation to the number of professional and skilled immigrants. The national interest in immigration policy, the argument went, was to promote economic growth and productivity through more-selective immigration policies that would favor employment-based immigrants with needed skills. Ethnic advocates responded that extended families strengthen immigrants’ contribution to economic and cultural life. They succeeded in drawing some conservative legislators to their side of the debate by casting their arguments for family-based immigration in terms of “family values.” Ethnic rights advocates do not support expanded admissions of immigrants as an end in itself. An increase in the number of immigrants in the country does not always translate into an increase in political clout for rights organizations; nor does it always advance their goals. Although expanding the population of Hispanic and Asian permanent residents may enlarge the pool of persons who later naturalize and vote, the newest immigrants do not necessarily share the political views of earlier generations of immigrants.13 Policies that increase the admission of certain categories of immigrants can actually undermine the political goals of rights groups. For example, Hispanic and Asian American rights organizations have long opposed the admission of traditional guest workers, temporary workers who enjoy neither labor nor civil rights, and who do not have the opportunity to apply for permanent residency. Ethnic rights groups have a political reason to oppose guest-worker programs for the discriminatory precedent they set: that is, the implication that denying basic rights to some members of ethnic minorities is justified by the short-term economic gains to employers, even when the migrant workers agree to those terms. By agreeing to admit a class of Hispanic immigrants, for example, without rights protections, ethnic leaders might begin to erode the national consensus that all Hispanic Americans deserve civil

02-S3453 1/11/06 3:06 PM Page 35

Interest-Group Goals

35

rights protections. Guest workers denied of rights are also likely to stand in economic competition with native workers, possibly even displacing workers of Hispanic or Asian ethnic background. Hispanic and Asian ethnic communities encompass organizations and individuals with mixed interests in labor immigration policies. Ethnic businesses that employ immigrants benefit from liberal policies on economic additions; but ethnic workers do not, particularly where they would have to compete with prospective immigrants for unskilled jobs. Competition for jobs may be one reason some Hispanic Americans oppose immigration policies that would allow immigrants to enter certain labor markets or that would regularize the status of undocumented immigrants.14 Taken as a whole, ethnic rights groups tend to take a middle position on admissions policy. They favor liberal policies on family admissions and legalization programs, but they take a restrictive position on admitting economic migrants as temporary workers without membership rights. Labor immigration is permissible only insofar as the government provides liberal rights to migrant workers. Migrant workers are admitted either permanently or temporarily with the earned right to stay permanently. A number of other civil rights organizations also have taken liberal positions on admissions and incorporation issues. In the early 1980s, the NAACP favored amnesty for undocumented workers but also supported employer sanctions, a response to its constituents who argued that illegal immigration disproportionately affects unskilled African American workers who must compete with unskilled immigrants for jobs. In 1990, the organization reversed its position on employer sanctions after the U.S. General Accounting Office found that sanctions had increased the incidence of discrimination against minorities in the hiring process. Black legislators in the 1980s also supported amnesty programs and other liberal immigration policies, and they consistently opposed employer sanctions. In his account of coalitions between black and Hispanic legislators, Fuchs (1990) documents the alliance of black legislators with Hispanic legislators on illegal immigration. Civil rights leaders often take positions on immigration policy that reflect the views of a significant sector, but not the majority, of the racial or ethnic minority group that they represent. For example, the liberal position of the NAACP—and black legislators—in the mid-1990s contrasted with the opinion of the majority of African Americans on immigration. In a national poll taken by the New York Times in December 1995,

02-S3453 1/11/06 3:06 PM Page 36

36

Interest-Group Goals ta b l e 2 . 1 Level of Legal Immigration: Poll Results, December 1995 should legal immigration into the united states be increased, decreased, or kept about the same? percentage responding

Total

Population

1,265 528 737 1,000+ 123 69 146

All Male Female White Black Hispanic Not a high school graduate High school graduate Some college College graduate Union household Nonunion household Republican Democrat Independent

445 289 379 208 1,000+ 378 409 401

Increased

Decreased

Kept at present level

Don’t know or no answer

2 2 2 1 2 8

64 64 64 67 58 45

31 31 31 29 34 45

3 3 3 2 6 2

3 1 * 5 4 2 2 3 2

64 70 68 48 60 65 67 59 68

29 26 31 44 34 30 28 35 29

4 3 2 3 3 3 2 3 1

n o t e : The poll was conducted by telephone between December 3, 1995, and December 6, 1995. The sample consisted of 1,265 adults. The responses of Asian Americans were not reported. Rows may not sum to 100 percent because of rounding. s o u r c e : New York Times Poll, Roper accession no. 0256141, question no. 021 (December 1995). Online. Lexis Nexis Academic. July 30, 2004.

at the height of debate in Congress over immigration, 64 percent of all respondents— 67 percent of whites, 58 percent of blacks, and 45 percent of Hispanics—favored a decrease in the level of legal immigration (Table 2.1). In the same poll, 29 percent of whites, 34 percent of blacks, and 45 percent of Hispanics thought legal immigration levels should remain at their current level. The opinions of Asian Americans were not reported. Another poll taken just after Congress enacted the Illegal Immigration Reform and Immigrant Responsibility Act in September 1996 also indicates differences in opinion among racial and ethnic groups. The IIRIRA placed new restrictions on illegal immigration and required legal immigrants to demonstrate greater financial resources than in the past, but it left legal-admissions levels intact. In October 1996, CBS News surveyed the nation’s adults on immigration (Table 2.2). The poll found that 50 percent of all respondents—52 percent of whites, 46 percent of

02-S3453 1/11/06 3:06 PM Page 37

Interest-Group Goals

37

ta b l e 2 . 2 Level of Legal Immigration: Poll Results, October 1996 should legal immigration into the united states be kept at its present level, increased, or decreased? percentage responding Total

Population

1,528 689 839 1,000+ 126 252 158

All Male Female White Black Hispanic Not a high school graduate High school graduate Some college College graduate Union household Nonunion household Republican Democrat Independent

537 379 444 247 1,000+ 409 585 410

Increased

Decreased

Kept at present level

Don’t know or no answer

8 9 7 6 15 18

50 50 49 52 46 35

35 35 35 35 36 40

7 5 8 7 3 7

13 7 7 10 6 9 7 8 10

47 56 51 37 49 50 50 46 55

34 30 36 46 40 34 37 39 30

5 7 7 7 5 7 6 6 5

n o t e : The poll was conducted by telephone between October 23, 1996, and October 27, 1996. The sample consisted of 1,528 adults. It included an oversample of Hispanics; a total of 291 Hispanics were interviewed. Results were weighted to be representative of a national adult population. The responses of Asian Americans were not reported. Rows may not sum to 100 percent because of rounding. s o u r c e : CBS News Poll, Roper accession no. 0277985, question no. 034 (October 1996). Online. Lexis Nexis Academic. July 30, 2004.

blacks, and 35 percent of Hispanics—wanted a decrease in the level of legal immigration. In the same poll, 35 percent of all respondents—35 percent of whites, 36 percent of blacks, and 40 percent of Hispanics— supported keeping immigration at its present level. Again, the opinions of Asian Americans were not reported. According to both polls, more blacks favored restricting legal immigration than wanted to maintain existing levels. Hispanics were evenly divided on whether to restrict immigration or retain existing levels in the 1995 survey; ten months later, they were somewhat more favorable toward keeping immigration at its current levels. Political scientists have conducted research to determine the factors that influence opinion on immigration policy. Citrin and his colleagues (1997) show the importance of beliefs about the state of the economy, anxiety about taxes,

02-S3453 1/11/06 3:06 PM Page 38

38 Interest-Group Goals and feelings about Hispanics and Asians. Burns and Gimpel (2000) demonstrate the independent role of prejudicial stereotypes of other groups among whites.15 Bobo and Hutchings (1996) offer an explanation of why blacks tend to favor restrictions on immigration. Their theory is that whites, blacks, Latinos, and Asians believe they are locked in competitive social relationships. Extending Blumer’s (1958) theory that social prejudice stems from a sense of group position, the two authors posit that members of recent (and voluntarily incorporated subordinate groups) feel less alienated than do members of long-term (and involuntarily incorporated) groups. They find corroboration in their analysis of the 1992 Los Angeles County Social Survey: The more that members of a particular racial group feel collectively oppressed and unfairly treated by society, the more likely they are to perceive members of other groups as potential threats.16 African Americans have the greatest sense of racial alienation (and they perceive greater competition with Asians than with Latinos); white Americans have the least sense of racial alienation; between these two groups are Latinos and Asians, who do not differ significantly in their sense of racial alienation. Like their Hispanic counterparts, groups that advocate for the rights of Asian Americans represent a highly diverse population. (In fact, their linguistic traditions are more diverse than those of Hispanic people.) It is impossible to generalize about the position on immigration of all Asian Americans or even of a constituent ethnic group—Korean Americans, for example, or Filipino Americans. Economic interests vary between businesses that employ immigrants and those that do not, and between immigrants in competing or complementary occupations; and immigrants with family members in prospective immigrant streams are likely to feel very differently about admissions policies than are immigrants without family seeking entry to the United States. The diverse interests of the communities they serve do not prevent Hispanic and Asian American advocates from pursuing a rights agenda that invokes the interest of the ethnic group. Flying a panethnic banner to expand the number of ethnic groups represented is a political strategy aimed at coordinating the resources of the various groups. We should be careful to distinguish, therefore, the political interests of civil rights leaders in winning specific types of immigration policies on behalf of minorities and the diverse interests in immigration policy held by the various segments of the ethnic communities that they represent.

02-S3453 1/11/06 3:06 PM Page 39

Interest-Group Goals

39

Religious Groups In the wake of Hart-Cellar, various religious organizations, representing Catholics, Protestants, and Jews, began actively lobbying for expanded immigration. The most vocal group in the 1980s and 1990s was the United States Conference of Catholic Bishops. Its goals included increasing opportunities for legal immigration, ensuring that immigrants have the right to live and work without exploitation, and protecting the right to due process of people who want to enter the United States or who are suspected of unauthorized entry.

Civil Libertarians The American Civil Liberties Union (ACLU) focuses its advocacy on issues of immigrants’ rights rather than admissions policy. On its agenda is the protection of due process rights for immigrants faced with detention or deportation proceedings. The ACLU also opposes the curtailment of immigrants’ civil rights as the federal power to monitor illegal immigrants in the population increases. In addition, the organization lobbies against proposals that would require citizens to carry national identity cards and proposals that would require employers to search a national database to confirm a job applicant’s legal status. Conservative libertarians have joined the liberal ACLU in objecting to intrusive enforcement policies.

Restrictionist Environmental Groups The groups leading the movement to restrict immigration emerged from one segment of the environmentalist movement in the 1970s. The Federation for American Immigration Reform (FAIR), an offshoot of Zero Population Growth (now Population Connection), is the most well known restrictionist organization. The goals of this membership organization in the early 1990s were (1) “to stop illegal immigration”; (2) “to reduce the amount of legal immigration”; (3) “to fix immigration quotas ‘in accordance with the demographic, natural resource and economic goals of the United States’ ”; (4) “to promote population control and economic development” in migrant-sending countries; and (5) “to promote the assimilation of immigrants into U.S. culture.” 17 Organizations like Population Connection and FAIR support reductions in immigration to help control population growth that these groups claim has outstripped

02-S3453 1/11/06 3:06 PM Page 40

40

Interest-Group Goals

environmental resources. Theirs is primarily a cultural argument: It calls on Americans to preserve their way of life by preserving scarce resources for those already living in the country. But some restrictionist environmentalists have also drawn attention to the economic interests of native workers by pointing to jobs as a resource that should be preserved for Americans.18

Coalitions and Oppositions Group preferences on multiple dimensions of immigration policy tend to give rise to coalitions and oppositions that shift in predictable ways. Table 2.3 shows how the policy preferences of several major categories of interest groups align on several dimensions of immigration policy. The table begins with the distinctions among groups on admissions policy. Employers—specifically companies that rely on immigrants for skilled and semiskilled labor—favor liberal admissions for employment-based immigrants, a position based on their economic interests. Their stand on family-based admissions is less clear because their economic interests in this instance are not clear-cut. Conversely, ethnic rights groups favor liberal admissions for family-based immigrants, but their rights agenda does not define a clear position on admitting permanent workers to the country. Therefore, there is potential for businesses that employ immigrants to trade support with ethnic rights groups on employment-based and family-based immigration. The pattern shifts when temporary-worker programs are at issue. Employer groups are the only ones that favor traditional guest-worker programs, programs that meet employers’ needs for temporary labor but that offer workers neither protections nor a path to permanent residency. Both labor unions and ethnic rights groups oppose these programs. Businesses that want access to low-wage foreign workers also have an interest in expanding probationary-worker programs. For unions and ethnic rights advocates, these programs are negotiable as long as they ensure workers’ civil and membership rights. The former pertain to the right to sue in courts and to join trade unions. Membership rights specify the time a temporary worker must work in the host country before earning the right to permanent residency. For labor unions, probationary-worker programs are negotiable as long as the temporary workers do not displace union members. Ethnic rights groups have reason to support probationary programs when those programs are an improvement on the status quo— either

Limit

Limit

Undetermined

Limit

Limit

Limit

Expand

Traditional guest workers1

Limit

Negotiable

Negotiable

Expand

Probationary workers2

temporary workers

Expand

1980s: impose employer sanctions 1990s: conditional on rights protection Conditional on rights protection

Conditional on costs

General enforcement

Expand

Limit

Limit

Limit

Enforcement and privacy

enforcement

Limit benefits

Expand benefits

Expand benefits

Undetermined

Social incorporation

1. Traditional guest workers are required to return to their home country and may not earn the right to apply for naturalization through work in the United States. 2. Probationary temporary workers may earn the right to apply for naturalization after they have worked in the United States for a designated time. 3. The entries in the table represent the position of the AFL-CIO. On some issues, a number of AFL-CIO unions took independent stands. For example, in the 1980s, the national garment workers’ union lobbied on its own for a liberal amnesty for undocumented workers; this was not a position shared by the AFL-CIO at that time.

Restrictionist environmentalists

Expand

Limit

Expand (post-1955)

Ethnic rights groups

Expand

Undetermined

Businesses employing immigrants Labor unions3

Employment-based admissions

Family-based admissions

Interest group type

legal admissions

ta b l e 2 . 3 Immigration Policies: Interest-Group Preferences

02-S3453 1/11/06 3:06 PM Page 41

02-S3453 1/11/06 3:06 PM Page 42

42

Interest-Group Goals

government-sanctioned guest-worker programs or the illegal recruitment of large numbers of undocumented workers, a practice that places those workers at great risk of exploitation. Although international interests are not depicted in Table 2.3, it should be noted that international political factors shape negotiations between governments on temporary-worker policy. The governments of source nations have a stake in protecting the rights of guest workers, and those rights are often the subject of bilateral agreements between the governments of sending and host nations. In addition, international rights groups may enter into the debate to promote standards of human rights and the rights of migrant workers. In both enforcement and social incorporation policy areas, economic interests in immigration policy define certain groups’ stances; but where economic interests are not strong, ideology is likely to play a role. Table 2.3 distinguishes two types of enforcement. General-enforcement measures—which include border control— do not engender strong controversy over privacy; the second category refers to issues that link enforcement and privacy concerns. Businesses that employ undocumented workers tend to favor lax enforcement policies; for businesses that do not employ undocumented immigrants, their position on a specific enforcement policy is conditioned on cost. Labor groups traditionally have wanted enough general enforcement to protect their members’ jobs. In the 1980s, they lobbied for employer sanctions; in the 1990s, in response to pressure from immigrant members calling for the protection of civil and labor rights in the workplace, unions began to condition their stand on enforcement on the nature of rights protections built into enforcement policy. Ethnic rights groups also oppose general-enforcement policies that violate basic civil rights or that break up families. Like unions, ethnic groups are feeling pressure from co-ethnic residents, which has the effect of making undocumented immigrants part of their ethnic identity group. Public support for strong enforcement policy diminishes when the privacy of citizens is likely to be compromised. Few would argue with strict border controls, for example; but a national databank for verifying the eligibility of all job applicants would evoke opposition because of privacy concerns. In part, it was those concerns that led to the specific terms of an employer-sanctions law passed in the 1980s, which required just minimal verification of job applicants’ identification papers. As noted earlier, the business community and ethnic rights groups have been largely opposed to enforcement that violates citizen’s privacy. Labor unions also condition their support for enforcement on protection

02-S3453 1/11/06 3:06 PM Page 43

Interest-Group Goals

43

of workers’ privacy. When privacy issues enter into enforcement policy, then, opportunities arise for cooperation among businesses, labor unions, and ethnic rights groups. Finally, labor unions and ethnic rights groups both tend to favor liberal social-incorporation policies—the provision of social benefits, for example. This is consistent with their liberal ideological bent and with their traditional support for generous social policies. Employers’ position on social-incorporation policy is undetermined because various and conflicting factors come into play. Although many business owners oppose liberal social policies because they are costly to taxpayers, some employers recognize that policies that inhibit the integration of immigrant workers into society may also impede their integration into the workforce.

Summary There is an interesting contrast between the policy goals of the economic and ethnic actors described in this chapter. Business and labor organizations are mainly concerned about maximizing the economic interests of their members. Even the steps the AFL-CIO took to embrace immigrants’ rights by changing its position on employer sanctions were motivated by economic interest: The organization wanted to expand the economic clout of the union movement through wide-scale organizing, which necessarily involved organizing immigrant workers. In contrast, ethnic groups have pursued an agenda shaped principally by their beliefs in human and civil rights. If these groups were simply interested in expanding their political base through immigration policy, they might have come down on the side of traditional guest-worker programs, focusing their efforts on legalizing temporary workers and incorporating them into their support base. Instead they have insisted on providing a minimum standard of rights for admission that would not exclude foreign workers from formal membership in the political community. The perception that special interests cooperate to wrest expansive immigration policies from Congress, then, is simplistic. Clearly there are opportunities for business and civil rights groups to cooperate on the policies that govern permanent legal admissions. But on temporary admissions, business– ethnic group cooperation tends to be constrained by the commitment of ethnic advocates to ensure some minimal standard of rights for foreign workers. In this sense, ethnic advocacy acts as a stabilizing factor in immigration politics, balancing the politics of immigrant admissions and rights.

03-S3453 1/11/06 3:06 PM Page 44

chapter

Hart-Cellar Act

3

congress passed the hart-cellar act in 1965, at a time when the civil rights movement was producing the most rapid reform of the nation’s civil rights laws since Reconstruction. As described in Chapter 1, the new law brought an end to racist policies that had underpinned immigration law for many decades. Hart-Cellar eliminated the nationalorigins quota system erected in the 1920s, a system that had discriminated against all immigrants from countries outside the northern or western regions of Europe. The new law also reversed discriminatory policies dating back to the late nineteenth century, ending special restrictions on immigration from the so-called Asia-Pacific triangle, covering most of East and Southeast Asia. In place of the national-origins quota system and restrictive anti-Asian laws, the Hart-Cellar Act gave priority to close relatives of U.S. citizens and permanent residents; to professionals, scientists, and artists of exceptional ability; and to skilled and unskilled workers in occupations for which there was an insufficient labor supply. The new law had two unexpected consequences: It expanded both legal immigration from Asia and the movement of undocumented workers across the U.S.-Mexico border. The new admission rules unleashed a pent-up demand for visas from Asian professionals and skilled workers, who, in turn, brought family members into the country under the familyunification provisions of the act. The increased activity at the border with Mexico was a response to Hart-Cellar’s imposition of limits on immigration from the Western Hemisphere. It was the first time the United States had set quotas on immigrants from this part of the world.

03-S3453 1/11/06 3:06 PM Page 45

Hart-Cellar Act

45

In addition, international developments soon would dramatically alter the size and character of immigrant and refugee flows. Prolonged economic crises would increase immigrant flows from Latin America beginning in the 1970s. And from 1975 to 1980 alone, 40,000 Indochinese refugees entered the country after the end of the Vietnam War. The expansion of Asian American and Mexican American populations following implementation of Hart-Cellar increased the influence of ethnic rights groups. These groups had emerged with the general growth of nonprofit advocacy groups in Washington, D.C., during the 1970s; and they played an increasingly important role in the making of immigration policy during the 1980s and 1990s. Ethnic advocates representing Asian American and Hispanic communities were particularly skillful in their advocacy of open-immigration policies toward their countries of origin. The partisan compromises that enabled Hart-Cellar’s passage in Congress in 1965 lent stability to its unintentionally expansive provisions into the future. When restrictionist forces later attempted to revamp the immigration system, intraparty compromises institutionalized by Hart-Cellar initially served to keep the act’s underlying political coalitions intact. Two elements of the immigration system after Hart-Cellar was enacted were forged out of compromises meant to hold together a divided Democratic Party, which had achieved unusually strong majorities in both houses of Congress while winning control of the executive branch in the 1964 elections. These two compromises also served to lessen differences among Republicans, who had successfully restricted immigration from Mexico and Canada, albeit indirectly, through a limit on immigration from the Western Hemisphere. First, the prioritization of family-based admissions under HartCellar’s seven preference categories had the effect of defusing future political conflicts over perceived imbalances in the racial and ethnic makeup of immigration streams. Family-based immigration was universal; it did not favor one racial group over another, seeking instead to reunify families divided by the migration of one or more members. Second, the Hart-Cellar compromise lessened the possibility that at some point in the future, contention between business and union interests over permanent-worker admissions would lead to contention in Congress. This was accomplished by delegating broad regulatory authority to the Department of Labor to certify that prospective employment-based immigrants would not displace American workers.

03-S3453 1/11/06 3:06 PM Page 46

46

Hart-Cellar Act

The laws guiding the process of certifying prospective skilled workers satisfied the demands of the union movement, a core constituency of the Democratic Party. By contrast, the controversial demand by western farmers for access to low-wage seasonal laborers from Mexico was not resolved. Western growers claimed that the rules for setting wage standards and other requirements under the existing agricultural guestworker program, known as the H-2A program, were too stringent. This issue would remain in contention after Hart-Cellar was passed. The twin compromise lessened the force of long-standing opposition within both parties to ending the national-origins quota system. Liberal Democratic majorities in both houses enabled a political compromise favorable to organized labor, while giving racial conservatives in the southern wing of the party a political justification for ending the national-origins quota system that they could convey to their constituencies. During the immediate postwar period, the stronghold of opposition to ending the national-origin quotas had been centered among Southern Democrats, the same group of Democrats that vigorously opposed civil rights reform during the period. The Hart-Cellar Act represented an opening of the nation’s immigration door to formerly restricted countries in Europe and an end to the laws excluding Asians, but the act was hardly liberal in its intent toward other regions of the world. To achieve a bipartisan compromise with Republicans on legislation introduced in Congress, the Lyndon Johnson administration and Democratic leaders in Congress agreed, in particular, to move federal immigration law in a decisively restrictive direction with respect to Mexican and Canadian immigration. From the perspective of managing conflict, the negotiations among lawmakers were a success: Hart-Cellar delegated broad authority to regulate employmentbased immigration, resolving debate on the issue for the foreseeable future. But the core question of regulating Mexican immigration was left a source of potential conflict, conflict that could easily reappear in the public arena of congressional lawmaking. That the Mexican immigration question was left in contention meant Congress would revisit it repeatedly. Adjustments to numerical limits and other methods of control would be legislated time and again in the years after 1965. Although amendments to the immigration law in 1976 brought per-country limits and hemispheric limits into greater concord, the problem of controlling illegal immigration went unsolved. At the same time, the symbolic value of family unification as a uni-

03-S3453 1/11/06 3:06 PM Page 47

Hart-Cellar Act

47

versal goal would blunt opposition to the family priorities of the Hart-Cellar Act for decades to come.

Racially Restrictive Laws The quotas that were established in the Immigration Acts of 1921 and 1924 were the first instances of American immigration laws that instituted numerical limits. Earlier methods of restricting immigration used qualitative criteria—making convicts and prostitutes ineligible, for example. The quota system was designed not just to limit the total number of immigrants, but also to control the racial and ethnic makeup of the immigrant population. In that respect, its rationale was consistent with the exclusion of Asians, which had legally begun in 1882, with the Chinese Exclusion Act.1 Agitation for exclusion of the Chinese based on their “unassimilablity”was centered in the west among labor unions during the mid-nineteenth century and had become a fixture of state politics in California by the 1870s (Hing 1993). Before the Civil War, immigrants from Europe settled mainly in eastern states and a few midwestern states; after the war, western states encouraged settlement. In the 1890s, the country experienced a surge of nativist sentiment. Theories of white racial superiority reinforced a fear of immigrants from southern and eastern Europe. The nation had entered an economic depression in the 1880s, and an influx of unskilled labor had lowered native wages in some industries (Hatton and Williamson 1995), prompting nativist organizations and unions to call for restriction.2 In the early 1900s, the American Federation of Labor joined forces with the Immigration Restriction League to lobby for limitations on immigration from southern and eastern Europe. With the First World War came a growing isolationism, which once again gave rise to a nativist push for limits on immigration.3 After the war, nativist and patriotic societies and the craft-based American Federation of Labor successfully pressed for restrictions on immigration. The first quota system came into being under the Immigration Act of 1921, which became known as the Temporary Quota Act. It was somewhat revised and then made permanent by the Immigration Act of 1924 (also called the National Origins Quota Act). The national-origins quota system set a numerical ceiling on admissions from the Eastern Hemisphere: a limit of 150,000 people a year. It also set a quota on immigration from each country, limiting the number

03-S3453 1/11/06 3:06 PM Page 48

48

Hart-Cellar Act

to 2 percent of the number of the people already living in the United States in 1890 (according to the census that year). Immigration in the Western Hemisphere would remain unlimited until the Hart-Cellar Act went into effect. The goal of the system, then, was to preserve the ethnic balance of the country.4 The Chinese Exclusion Act (1882) was the first of a series of restrictions on immigration set by the federal government in response to hostility toward Asians in the population. President Theodore Roosevelt signed an executive agreement in 1908 with Japan (the Gentlemen’s Agreement of 1908) to restrict the immigration of Japanese laborers. One provision of the Immigration Act of 1917 created an Asianbarred zone, which included the Far East and India.5 In 1924, at the time the National Origins Quota Act was passed, Congress found a way to exclude all Asians by making aliens who were “ineligible for citizenship” also ineligible for immigration. Two years earlier, in Takao Ozawa v. United States (260 U.S. 178, 1922), the Supreme Court had ruled that Japanese aliens were not eligible for citizenship; only free white citizens and persons of African descent could become naturalized. America’s wartime alliance with China led to the repeal of the Chinese Exclusion Act in 1943.6 Although China was given a quota of just 105 visas per year, this set a precedent for ending the ban on immigration from other Asian countries, including India and the Philippines, each of which received a quota of 100 visas per year in legislation enacted in 1946.7 Foreign-policy interests dictated these changes, but the quotas remained small to satisfy those in Congress who feared that admitting more than a few immigrants from Asia would undermine American culture and drive down living standards. The Immigration and Nationality Act (1952), known as the McCarranWalter Act, codified existing immigration law and kept the nationalorigins quota system in place. According to the act’s provisions, countries in the Asia-Pacific triangle, which encompasses most of East and Southeast Asia, would receive just 100 visas a year. And if a person of Asian ancestry came to the United States from a country in the Western Hemisphere that was not subject to quotas, his or her immigration slot would count toward the country of Asian ancestry. In addition, the McCarran-Walter Act added a four-level selection system to immigration law, which gave preference to applicants who possessed needed skills and then to the relatives of American citizens. The act also explicitly excluded communists from admission.

03-S3453 1/11/06 3:06 PM Page 49

Hart-Cellar Act

49

After the end of the Second World War, political support gradually grew in Congress for proposals to replace the national-origins quota system. The many inadequacies of the system were obvious. For one thing, it was too rigid to accommodate new sources of immigration. By the 1950s, more immigrants were coming to the United States as nonquota immigrants under special refugee or other categories than were entering the country as quota immigrants.8 Many of the immigrants who waited the longest to migrate were from the countries in Europe with the lowest quotas—Italy, Greece, Portugal, and Poland. And Britain, Northern Ireland, the Republic of Ireland, and Germany were not using their full quota allocations. In addition, the idea of a nation’s selecting immigrants on the basis of ethnic or racial criteria had become increasingly unpopular in the international community after the experience of Nazi Germany. In the postwar period, as the United States sought to assert its world leadership and forge alliances with countries in Europe, Asia, and Africa, the discriminatory principles of the American immigration system stood as an obstacle to the nation’s foreign-policy goals. Concerns about the relationship of immigration law and foreign policy motivated the executive branch to take the lead in calling for reform. Although there was widespread recognition of the antiquated and unfair basis of the old quota system, the conservatism born of the cold war lengthened its demise.9 In 1952, over President Harry Truman’s veto, Congress passed the McCarran-Walter Act, affirming the quota system and adding new measures to exclude people professing a belief in communism.10 The other factor delaying the end of the quota system was a strong conservative coalition in Congress between Southern Democrats and Republicans who had opposed social welfare and civil rights legislation from the time of the New Deal into the 1960s (Stern 1975, 55). The liberal championing of immigration reform had roots in the more general program of civil rights reform, including the need to erase formal justifications for ethnic and racial discrimination in the nation’s laws. Presidents Truman, Eisenhower, and Kennedy all favored ending the national-origins quota system. In addition, most economists agreed that a growth-oriented economy would benefit from a nondiscriminatory immigration law that could enable highly skilled professionals and workers to enter the United States from any country. By the mid-1960s, landmark civil rights reforms, including a voting rights act, had passed. And in 1965 an unusually large majority of liberal Democrats in both houses created favorable conditions for immigration reform. But conservatives among Democrats and Republicans objected to ending

03-S3453 1/11/06 3:06 PM Page 50

50

Hart-Cellar Act

national-origin and race quotas if the new selection system was likely to lead to a large influx of low-wage immigrants from Asia and Latin America. Racial conservatives in the southern wing of the Democratic Party were strongly opposed. In the end, many conservative Republican legislators and Southern Democrats agreed to replace the old quota system with a new selection system that prioritized family unification because they accepted arguments predicting that the admission of relatives of U.S. citizens would preserve the ethnic stock. And that is the explanation they gave to neutralize opposition among their constituents at home. None of the policy experts expected the new system to lead to large increases in total immigration.

Bipartisan Compromise The bill that would pass Congress as the Immigration and Nationality Act Amendments of 1965 had originally been proposed by President John F. Kennedy in a special message sent to Congress on July 23, 1963. After Kennedy’s assassination, President Johnson expressed support for the immigration reforms proposed by his predecessor; and after his landslide victory in the election of 1964, Johnson included immigration reform in a package of liberal reforms he sent to Congress. The bill proposed by Johnson, as revised and passed by Congress, was known as the Hart-Cellar Bill for its congressional sponsors, Senator Philip Hart (D-MI) and Representative Emmanuel Cellar (D-NY). The administration proposal was to phase out the national-origins quota system over a period of five years.11 The bill also would end specific restrictions on people of Asian ancestry. Congressional Democrats were hardly unified around the proposal; notably, Senator James Eastland (D-MS), chair of the Senate Judiciary Committee, strongly supported the national-origins quota system.12 In a provision that eventually was dropped because of congressional opposition, the president would have been given broad discretionary authority to reserve up to 50 percent of the quota pool to issue visas for national security reasons and another 20 percent for refugees. The original Kennedy and Johnson proposals called for admitting an increased number of highly skilled immigrants, who would have come under the “first preference” in the new system, filling the initial 50 percent of all quotas unless the president used the allocations for national security reasons or for refugees.13 But for reasons discussed below, Congress changed the category ranking, making family unification the most important selection criterion.

03-S3453 1/11/06 3:06 PM Page 51

Hart-Cellar Act

51

In the final terms of the Hart-Cellar Act, the Eastern Hemisphere was allowed a maximum of 170,000 immigrants a year, and no single country was allowed more than 20,000 visas. In line with then-current policy, initially no limit was placed on immigration from the Western Hemisphere; but the final version of the bill passed by Congress imposed a limit of 120,000 visas a year. Several special refugee programs were also proposed. On the path to becoming law, the immigration bill encountered numerous obstacles and underwent major revision. Although Congressman Cellar, the chair of the House Judiciary Committee, supported the administration bill—he was its main sponsor in the House—the chair of the immigration subcommittee, Congressman Michael Feighan (D-OH), emerged as the key player in negotiations with administration representatives. Feighan initially did not support the end of national-origin quotas: He had introduced his own immigration bill in 1964 that would have left national-origin quotas in place. But the AFL-CIO and ethnic groups had pressured him to oppose the quotas. Feighan also wanted an annual worldwide ceiling on immigration; but according to Wagner (1986), members of the executive branch stood firm in opposing a Western Hemispheric limit in negotiations with Feighan. The administration tried to appease the congressman by agreeing instead to a bill that placed a higher priority on relatives than on skilled immigrants (430). Feighan’s own constituency consisted of one thousand families that would benefit from the family-unification provisions of the bill.14 With the concessions to Feighan, a bill acceptable to the administration was reported out of the House immigration subcommittee to the House Judiciary Committee, which did not include a provision for a Western Hemispheric limit in its version of the bill. When the bill was brought to the House floor, Representative Clark MacGregor (R-MN) introduced an amendment to impose a Western Hemispheric limit. The House Republican Policy Committee issued a statement that it supported HR 2580 on August 25, 1965, but added that it believed an amendment placing the Western Hemisphere under a “reasonable numerical limitation” was essential.15 In addition, seven of the eleven Republican members on the House Judiciary Committee had signed the minority report supporting the MacGregor Amendment, but the amendment failed in a vote of 218 to 189. The opposition consisted of 205 Democrats and 13 Republicans; voting in favor were 68 Democrats and 121 Republicans. After the House turned down the proposal for a Western Hemispheric limit, the administration became more flexible in its negotiations on the issue. Senator Sam Ervin (D-NC) and Senate Minority Leader

03-S3453 1/11/06 3:06 PM Page 52

52

Hart-Cellar Act

Everett Dirksen (R-IL) struck a deal with Attorney General Robert Kennedy to include a Western Hemispheric limit of 120,000 visas in the Senate bill (Wagner 1986, 442). This agreement eventually allowed passage of the bill; when the House and Senate bills were reconciled in conference, the administration did not press for removal of the Western Hemispheric limit. The strongest opposition to the bill was centered among Southern Democrats, who articulated a conservative populist view toward immigration in floor debates in the House and Senate. A number of conservative groups, including the American Legion and the American Coalition of Patriotic Societies, also were staunch opponents.16 A coalition of liberal Democrats and moderate Republicans supported the bill, as did lobbyists for Italian American and Jewish groups, religious and humanitarian organizations, and the AFL-CIO. In the 1964 presidential election, immigration was an issue of contention between Lyndon Johnson and Barry Goldwater, his Republican challenger. The Democratic Party had included an immigration plank in its platform calling for repeal of the national-origins quota system; the Republican platform, however, “rejected proposed moderate amendments and did not allow for any major changes in immigration law.” 17 In one well-publicized speech on Labor Day in South Bend, Indiana, the Republican vice presidential candidate, William Miller, criticized both the tariff program and the immigration proposals supported by Johnson: In legislation, which Lyndon Johnson has designated as top priority, he now proposes that we completely abolish our selective system of immigration and instead open the floodgates for virtually any and all who would wish to come and find work in this country. We estimate that if the President gets his way, and the current immigration laws are repealed, the number of immigrants next year will increase threefold and in subsequent years will increase even more. These people will need jobs, but where will they find them? (Loftus 1964, 14)

Thousands in South Bend had recently lost jobs when a Studebaker plant closed. In a reply to Miller on October 8, President Johnson noted that twothirds of the total allocation of immigration quotas under the McCarranWalter Act were left unused: “We want to abolish these discriminatory quotas gradually over a five-year period and raise the overall limit by 9,000 — or 1/80th, or 1 percent of our work force.” 18 Johnson’s reelection and the large margin of Democratic seats in the 89th Congress—at least two to one—were both important factors in the passage of immigration reform that year.19

03-S3453 1/11/06 3:06 PM Page 53

Hart-Cellar Act

53

Still, Feighan and congressional Republicans did succeed in winning the two major changes in the administration’s bill; the introduction of a numerical limit (120,000 people) on immigration to the United States from the Western Hemisphere, and a change in the relative priority of family and skilled-worker categories in the preference system (the family category was placed first). In addition, the administration dropped its proposals to form an immigration board, with representatives from both the executive and legislative branches, and to give the president the power to change certain quota allocations. For its part, organized labor achieved substantial new protections for workers in the labor certification process. It failed, however, to win a guarantee that future temporary-worker programs would be outlawed, although in 1964 it had succeeded in ending the bracero program. Under the new selection process, 290,000 visas would be issued to applicants in seven categories each year, in the order of preference and the proportions listed here:20 Unmarried adult sons and daughters of U.S. citizens, to a maximum of 20 percent (58,000 visas) Spouses and unmarried sons and daughters of permanent residents, to a maximum of 20 percent (58,000 visas) Members of the professions, and scientists and artists of exceptional ability, to a maximum of 10 percent (29,000 visas) Married adult sons and daughters of U.S. citizens, to a maximum of 10 percent (29,000 visas) Adult brothers and sisters of U.S. citizens, to a maximum of 24 percent (69,600 visas) Skilled and unskilled workers in occupations for which there is insufficient labor supply, to a maximum of 10 percent (29,000 visas) Refugees (conditional entry or adjustment), to a maximum of 6 percent (17,400 visas)

The list shows a clear bias toward family-based immigration: Familyunification categories received 74 percent of the total allocation of visas, compared to 20 percent for employment-based immigrants and 6 percent for refugees. The strength of that bias was also evident in the preference given to adult brothers and sisters of citizens (fifth place in the list) versus needed skilled and unskilled workers (sixth place). Moreover, the brothers-and-sisters category was allocated the largest quota of any category. In fact, after it became clear that the category would be one of the most heavily used, some began calling the legislation the “brothers and sisters act” (Reimers 1992, 81). As before, spouses and

03-S3453 1/11/06 3:06 PM Page 54

54

Hart-Cellar Act

minor children were not restricted; that is, they could enter the United States outside the preference system.

Family Unification Liberal reformers entered legislative deliberations with the goal of ending the national-origins quota system. Conservative forces wanted the national-origins quota system retained and numerical limits on immigration tightened. In the end, a compromise granted liberal reformers the end to national-origin quotas, but appeased conservatives by introducing a numerical quota on immigration in the Western Hemisphere. The key to the compromise was family unification. The decision to make family the priority in the bill’s proposed selection system played a crucial role in the relative rankings of visa categories. And both that decision and the decision to extend family preference to the adult siblings of citizens had a political motive. Lawmakers on both sides of the immigration issue were feeling pressure. For supporters of expanding immigration, that pressure came from several directions: ethnic advocates, civil rights groups, even the international community, all urging an end to quotas based on national origin. For restrictionists, it came primarily from voters who feared that easing limits on immigration would affect their jobs and their way of life. Family unification offered both sides protection from electoral retribution. Putting family first has general appeal. By casting the immigration issue in universal rather than ethnic terms, the bill’s framers hoped to appease ethnic voters and ethnic interest groups, and to reduce competition among voters and groups for visas. Family unification also allowed these lawmakers to justify the Hart-Cellar Act to other constituents: An immigration policy that benefits families by extension benefits the social and cultural fabric of the nation. Members of the conservative opposition used the same argument with their political bases. How could voters in their districts object to a reform based on the principle of family unification?21 Over time, congressional leaders of both political parties had come to agree that the national-origins quota system needed replacing. Blocking reform was a minority in each party that wanted to preserve the old system because it feared that proposed changes would bring about an influx of immigrants from Asia and Latin America. Making family unification, rather than ethnicity, the primary basis of the selection process, along with setting equal visa limits for each country in the

03-S3453 1/11/06 3:06 PM Page 55

Hart-Cellar Act

55

Eastern Hemisphere, was a solution that was both politically feasible and in the long run least likely to generate ethnic competition for visas. The debates in committees and on the floor of both houses reflected the expectations of the Johnson administration and key proponents of the bill that the reforms would not bring about large-scale changes in immigration flows. The administration and congressional sponsors of the 1965 reforms argued that their proposals were simply a natural extension to the immigration system of the nondiscriminatory principles embodied in newly passed civil rights legislation. On March 4, 1965, in testimony before the Subcommittee on Immigration and Naturalization of the Senate Judiciary Committee, Attorney General Kennedy predicted that “immigration from any single country would be limited to 10 percent of the total . . . with the possible exception of two countries now sending more than that number: Great Britain and Germany. But this extreme case should set to rest any fears that this bill will change the ethnic, political or economic makeup of the United States.” 22 On the floor of the House, lawmakers from the South expressed concern about the prospect of swelling numbers of people from Africa and Asia who would likely enter the United States under the provisions of Hart-Cellar. For example, Representative O. C. Fisher (D-TX) opened the argument against HR 2580 this way: My chief objection to this bill is that it very substantially increases the number of immigrants who will be admitted each year, and it shifts the mainstream of immigration from western and northern Europe—the principal source of our present population—to Africa, Asia, and the Orient. . . . The elimination of the national origins quota, as proposed here, would do more than change the cultural pattern of our immigration—serious as that may be. It would have a direct effect on the numbers of people who would be pouring in from non-quota countries.23

To counter this argument, Representative Cellar had given more than one reason not to expect large growth in immigration from Asia and Africa. His reasoning, however, seems inconsistent. Although Cellar claimed that relatives would use up most of the visas available to Asian and African countries, he also asserted that the people of Asia and Africa had very few relatives: Mr. Chairman, claim has been made that the bill would bring in hordes of Africans and Asians. This is the answer to that false charge: Persons from African and Asian countries would continue to come in as heretofore, but would be treated like everyone else. With the end of discrimination due to place of birth, there will be shifts to countries other than those of northern and western

03-S3453 1/11/06 3:06 PM Page 56

56

Hart-Cellar Act

Europe. Immigrants from Asia and Africa will have to compete and qualify in order to get in, quantitatively and qualitatively, which itself will hold the numbers down. There will not be, comparatively, many Asians or Africans entering the country. Mr. Chairman, there are many factors that would limit immigration from these sources. Many countries in Africa do not use their present quota of 100. Under this bill those who have preference would come in first; that is, those coming to fathers, mothers, husbands, wives, sons, daughters, brothers, sisters, and so forth. These preferences would practically use up most of the numbers authorized from those countries. Mr. Chairman, since the people of Africa and Asia have very few relatives, comparatively few could immigrate from those countries, because they have no family ties in the United States.24

Regulations Governing Immigrant Laborers Since 1955, labor unions had supported the expansion of family-based immigration. Their concerns over the Hart-Cellar Act focused instead on issues related to employment-based immigration—tighter regulation of admissions to protect American wages and jobs, general enforcement of labor standards, and guarantees of prevailing (union) wages for immigrants. Traditionally, organized labor is an important component of the Democratic Party’s base. Republican majorities in Congress would not have been constrained overmuch by labor’s interests, and any immigration reform they enacted would likely have retained looser regulations and/or greater latitude toward employers’ hiring temporary workers without meeting labor standards or union wage levels. However, with Democrats holding large majorities in both houses in 1965, organized labor could make a strong case for its interests. This left Democrats with a problem: Labor’s interests and those of other members of the Democratic coalition—in particular, employers who relied on immigrant labor— diverged sharply on the issue of regulation. What could they do to preserve their coalition? The solution the Democrats adopted is not uncommon in situations where a coalition is faced with actual or potential conflict among its members.25 What the Congress did in 1964 was to reserve to unions and businesses a role in the case-by-case process of administrative decision making. In a number of other countries that actively encourage skilled workers to immigrate, the national legislature uses projections of labor shortages and other factors to set annual immigration targets; as a result, worker and employer organizations have regular input into the

03-S3453 1/11/06 3:06 PM Page 57

Hart-Cellar Act

57

setting of employment-based quotas.26 In the United States, Congress sets quotas on skilled and unskilled immigrants, but changes in the quotas occur only when interested parties can place the issue on the congressional agenda. In the interim, the Department of Labor holds authority for regulating immigrant laborers. In a system established after the Second World War and strengthened in 1965, immigrants admitted for the purpose of filling job vacancies must be sponsored by specific employers. The Labor Department carries out a labor-market test so that it can certify that no American worker is available and willing to fill any job in question. In addition, immigrants must be paid the prevailing wage according to Labor Department estimates. Unions have the opportunity to contest both specific claims about the availability of American workers and the department’s wage estimates. This kind of system tends to remove divisive regulatory issues from the arena of congressional politics. In 1965, when the Hart-Cellar Act was passed, the AFL-CIO’s political influence was strong. The United Steelworkers of America and other industrial unions were also in a position to wield considerable power in industrywide wage bargaining. With a Democratic administration and a liberal Democratic majority in Congress, national unions had greater influence over national economic policy than they would enjoy a decade and a half later, when immigration appeared on the congressional agenda again. In the first few decades of the twentieth century, organized labor tended to favor restrictions on immigration. By midcentury, the leaders of the national labor federation had begun to moderate their traditional views. In 1946, before their merger, for example, both the AFL and the CIO opposed a bill that would have cut immigration quotas significantly (Starr 1955, 300).27 By 1965, the AFL-CIO was ready to accept the original version of the Hart-Cellar Act, which would have given first preference to skilled workers rather than family members (Stern 1975, 185 –192). One factor in the AFL-CIO’s acceptance of that proposal was a certification system it had won in the 1950s, a system that required the secretary of labor to certify, case by case, that immigrants entering the United States to work would not adversely affect U.S. labor conditions. Under Hart-Cellar, the secretary was required to certify a much wider range of labor-market entrants from foreign countries: professionals, scientists, and artists (the third preference category); needed skilled and unskilled workers (the sixth preference category); nonquota immigrants; and all entrants from the Western Hemisphere except immediate relatives.

03-S3453 1/11/06 3:06 PM Page 58

58

Hart-Cellar Act

The response of unions to the economic and cultural effects of immigration varied greatly. Unions with large immigrant constituencies, like the Steelworkers, were among the most active and influential of the interest groups that lobbied not only for ending the national-origins quota system in the early and mid-1960s but also for moderate increases in the number of immigrants legally admitted to the United States each year. The proimmigration interests of several of the largest industrial unions in the AFL-CIO stemmed from their immigrant membership. These unions had originally been formed and led by recent immigrants; by 1965, there were many recent immigrants among their members who favored policies that would allow their relatives to immigrate to the United States. But members of all unions, to varying degrees by industry, also were threatened by prospective (or actual) job displacement and lower wages because of an increase in the number of immigrants entering the labor force. Craft unions, whose members generally competed for work with immigrant laborers, tended to favor more-restrictive immigration policies than did industrial unions, which tended to have a large proportion of immigrants among their members. Moreover, although the mid-1960s was a period of generally tight labor markets, unions usually sought protection for their members only in periods when unemployment was expected to go up or was actually rising. The AFL-CIO leadership achieved two gains from Hart-Cellar. The first was the incremental strengthening of the labor certification requirements and their extension to any person entering the United States to work. The second was the indirect gain achieved by lowering the position of skilled-worker visas in the preference scheme. Although various politicians would try to use populist trade and immigration protectionism for electoral gain after the Immigration and Nationality Act Amendments were passed, none succeeded in mobilizing the AFL-CIO to their side. By insulating itself politically from the admissions issue, the AFL-CIO was able to pursue a long-term alliance with minority civil rights organizations. That alliance still forms part of the core of the national Democratic coalition, which has not put immigration at the forefront of its policy agenda since 1965.

Unresolved Issues The AFL-CIO lost two main points of contention in the legislative debate over the Hart-Cellar Act. One year earlier, in 1964, the federation and an alliance of civil rights groups had succeeded in ending the

03-S3453 1/11/06 3:06 PM Page 59

Hart-Cellar Act

59

bracero program. Now the AFL-CIO wanted a ban on all programs to import foreign agricultural laborers, but it was not strong enough to have one written into the legislation. In addition, the federation failed to achieve its goal of giving the secretary of labor (instead of the attorney general) the power to certify a labor shortage as the precondition for admitting temporary foreign workers. Regulating temporary workers turned out to be the major unresolved issue of concern for labor. This and closely related issues would emerge and reemerge in congressional debates after the Hart-Cellar Act was passed.

A Durable Solution As the analysis here indicates, Hart-Cellar’s selection criteria and their ranking resolved a dilemma for the Democratic coalition. That is, they defused or institutionally removed from center stage conflicts that could deeply fracture the party in the future. Publicizing the bill as a “family immigration act” appeased racially conservative Southern Democrats and weakened labor’s opposition because employment-based admissions received a lower preference than family categories. Paradoxically, this effort to defuse the conflict over race lent resiliency to the familybased preference system passed in 1965, even though the consequences of the policy were very different from what the bill’s framers had predicted. Later, proimmigration ethnic advocates would artfully frame the rationale for a family-based preference system in terms of family values, a strategy that would attract the support of conservatives. The reordering of visa preference categories was also the result of a political compromise that was intended to sway votes for passage. One sector of the Democratic Party consisted of liberals from nonsouthern states, many of whom represented organized labor’s concerns about protecting American workers from job displacement by and wage competition with immigrants. In his original proposal to Congress in 1963, President Kennedy had spoken for these interests in calling for formation of an immigration wage board, which would have consisted of representatives from Congress and the executive branch. Instead of setting fixed quotas through an openly contentious process in Congress, the board would have been authorized to alter how many and what type of immigrant workers would be admitted as economic conditions changed. Republicans opposed this proposal, however, arguing that the cross-branch wage board would violate the constitutional separation of powers. With an unusually sizable representation in the 88th Congress,

03-S3453 1/11/06 3:06 PM Page 60

60

Hart-Cellar Act

liberal Democrats did succeed in granting the Department of Labor enhanced powers to certify wages and employment decisions. Racial conservatives in the southern wing of the party feared that opening the immigration door to qualified workers without regard for their race or ethnicity would lead to a sudden and large influx of immigrants from Asia and Latin America. Party leaders worked out a compromise with the expectation that it would preserve the dominance of European immigrants in later streams. The new mechanism reversed the standing priority given to skilled laborers over family members. Now, not only spouses and children but also brothers, sisters, and parents of citizens would move toward the front of the immigration queue. Because American citizens were largely of European descent, it was believed that most of the immigrant relatives of citizens would be European and that ethnic homogeneity could be preserved. But soon after Hart-Cellar was passed, large numbers of Asian immigrants began entering the country under the employment categories. This reflected the accumulated demand for visas during many decades of exclusion. Later, relatives joined the first waves of Asian immigrants. In the end, the 1965 racial compromise had precisely the opposite effect of what was intended: By the 1980s, people from Asia and Latin America made up the largest proportion of immigrants, a fact of immigration even today. This unexpected consequence has fueled recent debates between those who want to retain the status quo and those who want to increase the proportion of highly skilled immigrants and perhaps entirely eliminate unskilled-worker categories. While acceding to the demand of reformers to end the nationalorigins quota system, Republicans had insisted on limiting immigration from Western Hemispheric countries. There had been no limit on Western Hemispheric immigration before this to stem mass immigration, particularly from Latin America and Mexico. The hemispheric limit was a decidedly restrictive component of the new immigration bill that is sometimes overlooked. One reason that the limit often goes unnoticed is that the Republicans originally had proposed to limit each country in the Western Hemisphere to a maximum of 20,000 visas a year, a restriction that was not included in the final legislation. By not imposing a percountry limit on nations in the Western Hemisphere, the final agreement made it appear that the Latin American, Canadian, and European countries to which the hemispheric limit applied held a preferential position relative to the nations of the Eastern Hemisphere. Referring to the decision not to place Western Hemispheric countries under a per-country limit, Mr. Moore remarked that the nonquota status

03-S3453 1/11/06 3:06 PM Page 61

Hart-Cellar Act

61

for immigrants from the Western Hemisphere is “an important symbol of hemispheric solidarity.” 28 Before 1965, Canada and Mexico, because they were neighboring economies, enjoyed special status: The United States issued them visas on an unlimited basis.29 Later, in 1976, in response to public concern about the growth in Hispanic immigration, Congress for the first time placed both Canada and Mexico under a 20,000-visa limit, like countries in the Eastern Hemisphere. In 1978, the Western and Eastern Hemispheric limits were combined in a 290,000visa worldwide cap, and the family-preference system was applied to all origin nations. The final amended version of the Hart-Cellar Act would pass overwhelmingly, by a vote of 318 to 95 in the House on August 25 and by a vote of 76 to 18 in the Senate on September 22. It had the support of the leadership of both parties; the only opposition came from Democrats in the South. Northern Democrats supported the House bill 179 to 8, while Southern Democrats opposed it 62 to 30. In the Senate, Northern Democrats favored the bill 43 to 2, and Southern Democrats opposed it 13 to 9.30 The conference report was agreed to on September 30 by a vote of 32 to 69 in the House and by a voice vote in the Senate. As noted by Congressman Cellar, among the many interest groups that supported passage were the Lutheran Conference, the Society of Friends, the Hebrew Immigrant Aid Society, the National Catholic Welfare Conference, the Tolstoy Foundation, the Church World Service, International Social Services, B’nai B’rith, and the American Hellenic Educational Progressive Association. Within just a few years of Hart-Cellar’s passage, members of Congress were surprised by its impact. Wagner (1986) reports that of the ten countries that most commonly sent immigrants in 1965, only two— Germany and Italy—were on a March 1968 list of nations the State Department was predicting would send the most immigrants to the United States in 1969.31 According to a report by John Corry in the New York Times on March 18, 1968, however, there was no move in Congress to bring back the national-origins quota system. As one of Corry’s Senate sources remarked, “Congressmen don’t want to look like racists” (cited in Wagner, 464 – 465).

Summary The bipartisan compromise underpinning the 1965 Hart-Cellar Act would prove politically durable. Racial conflict over immigration issues

03-S3453 1/11/06 3:06 PM Page 62

62

Hart-Cellar Act

was not erased by removal of the racist immigration quotas, but it was defused in part by an agreement to give family-based visas, including visas for adult siblings of citizens, the highest priority over and above employment-based visas. Family unification is presumably a race-neutral goal: If one ethnic group benefited more from family-based immigration, a charge of racial preference or prejudice would be hard to substantiate. The Hart-Cellar Act also had positive political effects on issues of labor regulation. Its terms reduced the conflict between organized labor and business interests by delegating regulatory authority over employment-based immigration—particularly the approval of permanent-employment visas—to the Department of Labor. But the legislation left considerable authority to regulate temporary workers to congressional policymakers, which meant ongoing conflict in this area in the decades following Hart-Cellar’s passage. In the mid-1980s, influential reformers in Congress, among them Senator Alan Simpson (R-WY), would argue that Hart-Cellar’s priority system needed fundamental revision—specifically, elimination of the preference category for the adult brothers and sisters of U.S. citizens. In 1990, Congress mandated that a select commission examine and make recommendations about U.S. immigration policy. In its 1995 report, the U.S. Commission on Immigration Reform suggested that it was in the nation’s interest to attract more highly skilled immigrants and to eliminate the admission of unskilled immigrants, who had come to constitute the bulk of the family streams. It would prove hard, however, for restrictive reformers to eliminate any of the family-preference categories because the family-unification policy had created a growing constituency in new immigrant communities. Moreover, large national organizations, including the Catholic Church, were finding that immigration was a source of new members. Paradoxically, the family-preference system enabled the change that its proponents had argued would not occur: a large increase in immigration from Asia and Latin America. But the goal of U.S. immigration policy—reuniting families—was one that restrictionists would find extremely difficult to assail.32 Recognized in policy circles as representatives of important constituencies, Jewish and Italian American lobbies were active and influential in pressing for an end to national-origin quotas. Ngai (2004) notes that the Western Hemispheric limit was passed in the absence of an empowered Mexican American constituency that could have resisted it.

03-S3453 1/11/06 3:06 PM Page 63

Hart-Cellar Act

63

Some Asian American and Mexican American organizations testified in congressional hearings on Hart-Cellar, but the civil rights revolution had yet to give rise to many of the advocacy organizations that would represent these ethnic groups in Washington, D.C.33 The family-unification provisions of the Hart-Cellar Act not only helped enlarge the constituency base for rights advocacy on behalf of Asian Americans and Mexican Americans; their universality also expanded advocacy for the inclusion of all racial groups in the political community.

04-S3453 1/11/06 3:06 PM Page 64

chapter

Post-Bracero Dilemmas

4

in 1964, one year be fore the hart-cellar act was passed, Congress terminated the bracero program, a program that allowed laborers from Mexico to enter the United States temporarily to meet the seasonal demand for farmworkers. The program had started in 1942, during the Second World War, in response to claims of labor shortages by growers in Texas and California. The program admitted about 4.6 million workers between 1942 and 1964. Although the program continued for almost two decades after the war ended, opposition to importing Mexican laborers was strong among unions and humanitarian organizations.1 After Congress refused to extend the bracero program in 1964, growers in the west turned to illegal methods of procuring migrant labor from Mexico. In the early 1970s, the AFL-CIO began pressing members of the House for legislation that would stem the rising tide of illegal workers by establishing employer sanctions. According to Tichenor (2002), a number of other interest groups offered support for employer sanctions at hearings held by the House Judiciary Committee’s Subcommittee on Immigration. Among those groups were the NAACP, the United States Conference of Catholic Bishops, and the League of United Latin American Citizens. In Tichenor’s account, Leonard Carter, the NAACP’s western regional director, claimed that as a result of illegal immigrants, “the poor have been deprived of 100,000 jobs for which they might qualify” (226 –227). By 1977, the AFL-CIO had gained widespread support from House Democrats, and the Carter administration initiated an employer-sanctions bill. The bill would have imposed civil fines on

04-S3453 1/11/06 3:06 PM Page 65

Post-Bracero Dilemmas

65

employers who demonstrated a “pattern or practice” of hiring undocumented aliens and, after repeated violations, criminal penalties. The proposal also called for a legalization program that would make all undocumented aliens living in the United States before 1970 eligible to apply for permanent residency. In response, agricultural lobbyists on Capitol Hill successfully mobilized to prevent action on any employer-sanctions bill until Congress could guarantee some alternative supply of migrant labor. The H-2 program, a guest-worker program, had been established by the federal government in 1952. It allowed agricultural and nonagricultural employers to hire contract workers from foreign countries. H-2 was used by sugarcane growers in Florida and apple growers along the eastern seaboard to hire workers from Caribbean and other countries. But western growers found it too costly to use the H-2 program. The requirement that employers provide adequate housing for workers was too onerous, they argued. And the hiring process was slowed by the need to “clear” guest workers by certifying that no American worker was available to fill any job that an H-2 worker would perform (Martin 1987). Represented by the Farm Labor Alliance in the 1980s, these growers demanded a new and expansive guest-worker program. A key element was an expedited process for calling up guest workers, who would be required to work in agriculture and then return to the sending country. In the argument of the Farm Labor Alliance, access to an ample supply of migrant labor on short notice was vital to maintaining the profitability of fruit and vegetable farming, one of the most lucrative industries in California. This chapter examines the role of Hispanic rights groups and labor unions in the debates on foreign migrant workers and illegal immigrants after the bracero program ended, debates that prepared the way for passage of the Immigration Reform and Control Act in 1986. The issues Congress attempted to confront in the 1980s remain largely unsolved.

Numbers or Rights? The controversies over illegal immigration and guest-worker programs posed difficult questions for civil rights groups and for labor unions with large immigrant memberships. One strategic problem concerned the relationship between the number of migrants given legal status in the United States and the rights they would be afforded. Specifically, in the 1980s and 1990s, the rights groups and unions could have endorsed proposals that would admit large numbers of foreign workers on temporary

04-S3453 1/11/06 3:06 PM Page 66

66

Post-Bracero Dilemmas

status, without full naturalization rights. Clearly, liberal admissions would offer opportunities to individual laborers. It also could expand the constituency base of both rights groups and unions. Although ethnic groups had an interest in increasing immigrant admissions to the United States, they were unwilling to do so at the risk of creating an underclass of migrants. For American unions, expanding the labor pool by admitting temporary workers had the potential to create competition for jobs or to place downward pressure on wages . . . unless those workers were given the labor rights available to legal residents of the United States. Forced to choose between alternatives, civil rights groups and unions could not endorse numbers in the absence of rights. The impact of the numbers-versus-rights question extended beyond wages and working conditions to the norms of racial equality as they applied to members of ethnic minority groups.2 As Mendelberg (2001) notes, the “most effective way to combat an old norm and establish a new one is to pass landmark legislation, to issue momentous judicial rulings, and to engage in other highly salient signals of commitment to the new norm” (17). In her analysis, in the 1930s, cultural leaders began to communicate the idea that “racial inequality was an immoral principle” (18); and their message got increasingly louder in the 1950s and 1960s. In 1975, Congress had amended the Voting Rights Act to extend antidiscrimination laws and voting protections to language minorities. But norms on the fair treatment of aliens had not been established by the courts in the 1980s, when illegal immigration and guest-worker policy became the focus of federal immigration legislation. The Supreme Court had ruled in Espinoza v. Farah Manufacturing Co. (414 U.S. 86, 1978) that civil rights laws did not include discrimination based on alienage. Subsequently, one of the most important issues debated under the Immigration Reform and Control Act of 1986 was the creation of a class of persons protected against discrimination based on citizenship status. Hispanic and Asian American rights groups were concerned about a proposed law that would punish employers for hiring undocumented workers. They predicted that if employers felt threatened by sanctions, they would discriminate against all foreign-looking persons in their hiring decisions. Hispanic organizations played an especially influential role in policy debates over what rights should be afforded migrant farmworkers from Mexico, whose labor supported the lucrative fruit and vegetable industry in western states. It was not clear what position Hispanic rights groups should take on illegal immigration: Opinion within the Hispanic community was

04-S3453 1/11/06 3:06 PM Page 67

Post-Bracero Dilemmas

67

divided. Recent immigrants favored a more-generous policy on undocumented workers than did immigrants who had been in the United States for some time. Education was also a factor: Those with less education tended to be more lenient than those with higher levels of education. All the major Hispanic rights groups eventually came to agree that employer sanctions would be ineffective in preventing illegal immigration; moreover, these groups predicted that sanctions would compel employers to subject all foreign-looking workers to scrutiny, increasing the incidence of racial discrimination in the workplace. The labor movement was divided on the issue of undocumented workers. Although most unions supported employer sanctions, some opposed them. These unions were motivated in large part by numbers: the need to expand their organizing base. This was particularly true of labor unions with shrinking membership. Recent immigrants—legal and illegal—represented a large pool of potential members. Unions that hoped to organize widely among recent immigrants had to stand with them. If recent immigrants opposed sanctions, then so did these unions.3 In combination with the lobbying of agricultural employers, Hispanic opposition stalled passage of an employer-sanctions bill for nine years after one was first proposed by the Carter administration in 1977. When the Immigration Reform and Control Act was finally passed in 1986, it contained provisions that were supposed to reduce racial discrimination in the workplace; but because they depended on the voluntary compliance of employers, the provisions were difficult to enforce. On the issue of temporary-farmworker programs, Hispanic rights groups sided with organized labor. Their cooperation with the AFL-CIO succeeded in blocking the traditional guest-worker program that western growers wanted to include in the IRCA, and was a significant force in liberalizing the temporary-worker program that did become part of the law. The crucial element of that program was “earned-legalization rights”: Undocumented farmworkers already working in the United States retroactively earned the right to legal temporary residency, and eventually permanent residency, once they had worked a specified number of days harvesting seasonal and perishable crops as special agricultural workers (SAWs). The legalization of SAWs created an immigrant category that would eventually be adopted by about 1 million agricultural workers. As described later in this chapter, the political trade-off between numbers and rights was embodied in the details of the legislation that established the SAW category, which in the end gave both documented and

04-S3453 1/11/06 3:06 PM Page 68

68

Post-Bracero Dilemmas

undocumented migrant farmworkers liberal earned-stay rights (Martin 2004). In the bargaining that led to the creation of the SAW category and the legalization of undocumented immigrants, Hispanic rights groups engaged in a distinctive form of identity politics in which new social identities (categories) were the outcome of legislative bartering. The numbers-versus-rights trade-off was of central concern because it affected the future political power and rights of the Hispanic population and of the labor movement.

Civil Rights Advocates A handful of Hispanic advocacy groups cooperated closely in lobbying for immigration reform during the 1980s. The most prominent were the League of United Latin American Citizens, the Mexican American Legal Defense and Education Fund, and the National Council of La Raza. All three groups opened offices in Washington, D.C., in the 1970s, establishing, with a handful of other Hispanic organizations, a national base for Hispanic rights advocacy in the nation’s capital. Founded in 1929 in Corpus Christi, Texas, LULAC was the oldest of the three groups. LULAC was established as a membership organization, structured along the lines of U.S. civic organizations like the Order of Knights of America. At the organization’s founding, members were required to be native-born or naturalized citizens of Latin extraction. Later, in the 1970s, all Hispanics as well as Anglos residing in the United States were admitted. In 1968, with a grant from the Ford Foundation, LULAC’s leaders founded MALDEF in San Antonio, Texas. MALDEF began as a small litigation firm specializing in civil rights.4 It evolved into a civil rights advocacy group that has argued and won a number of landmark civil rights cases. The first was Serna v. Portales (499 F.2d 1147, 1972), which established the right of Spanish-speaking children to bilingual education. The ruling in White v. Regester (412 U.S. 755, 1973) required counties to institute single-member instead of at-large election districts where the latter diluted minority voting strength. The first case MALDEF argued successfully in the Supreme Court on behalf of undocumented immigrants was Plyler v. Doe (457 U.S. 202, 1982). In Plyler the Court found that a Texas statute that withheld state funds for educating children who were not “legally admitted” to the United States and that authorized local school districts to refuse to enroll these children violated the equal protection clause of the Constitution.

04-S3453 1/11/06 3:06 PM Page 69

Post-Bracero Dilemmas

69

The National Council of La Raza was founded in 1968 as a regional organization, the Southwest Council of La Raza. In 1973, the group changed its scope and its name, and moved its headquarters from Phoenix to Washington, D.C. There the organization began working to build the financial and organizational capacity of local affiliates and also began advocating for the interests of Mexican Americans at the national level. In the late 1970s, the council formally expanded its representation to all Hispanics. Key to the NCLR’s advocacy function are its relationships with White House staff, members of executive agencies, and members of Congress. As Sierra (1993) recounts, the council gained visibility in Washington policy circles during the Ford and Carter administrations through its political relationships with members of executive branch departments and its testimony at congressional hearings on Hispanic concerns. For example, NCLR “took the lead” in organizing a coalition of Mexican American organizations to consult with the U.S. Immigration and Naturalization Service (INS).5 The coalition successfully pushed for “increased hiring of Hispanics in the Justice Department and the Border Patrol; the appointment of a special assistant for Hispanic affairs; and the institutionalization of the coalition into a permanent advisory committee” (241–242). When the advisory board was formed, seven of its twenty-one members were board members of the National Council of La Raza. In addition, President Jimmy Carter chose NCLR members to fill administration posts. A prime example: NCLR board member Leonel Castillo was appointed head of the INS. Through an office set up in 1975 to coordinate policy research and advocacy, the NCLR provides detailed policy analysis from the council’s perspective on legislative and administrative proposals concerning Hispanics, produces press releases and issue briefs, drafts legislation, and testifies before lawmakers at hearings on proposed legislation. In his study of policy analysts at the NCLR, Cortes (1992) describes the particular value of congressional testimony: When representatives of the council appear on Capitol Hill to give testimony, they help “assure their Latino constituents and allies that their interests have been recognized.” And briefings to the executive branch allow NCLR staff members to claim that “the views of Latinos [are taken] into account during the formative stages of policymaking” (349). According to Charles Kamasaki, NCLR’s director of policy analysis, LULAC, MALDEF, and NCLR cooperated with one another in 1984 to defeat a proposal that would have instituted employer sanctions. Each

04-S3453 1/11/06 3:06 PM Page 70

70

Post-Bracero Dilemmas

group played a distinct role: LULAC’s Washington lobbyist was “in the paper every day”; MALDEF conducted legal analysis and lobbying; and the council carried out extensive research so its lobbyists would know what “would and would not sell in terms of political compromises on the Hill.” Kamasaki added: “Under our working relationship, all of us could be consistent and loyal to our institutional interest, but also loyal to each other. That’s how we succeeded in killing the bill in ’84” (quoted in Cortes 1992, 350 – 351).

Employer Sanctions and Legalization Hispanic rights groups in the 1970s came to their positions on issues of illegal immigration while opinion in the Hispanic community remained divided. That divergence of opinion was evident in numerous studies of Mexican Americans, Cuban Americans, and Puerto Ricans. Polinard, Wrinkle, and de la Garza (1984), for example, surveyed Mexican American households in two different counties of Texas when national debate over the Simpson-Mazzoli Act (the IRCA) was under way. One county was in central Texas; the other, close to the U.S.-Mexico border. Among the respondents, 49 percent supported employer sanctions, while 29 percent opposed them; 17 percent were neutral (791).6 The authors conclude that when Hispanic rights groups eventually took a strong stand in opposition to sanctions, their position ran counter to opinion in these two Mexican American communities (796). They did find greater agreement between Hispanic rights groups and the two communities on two other issues addressed by Simpson-Mazzoli: The national organizations supported an amnesty for undocumented workers, and so did 47 percent of the people surveyed; 29 percent opposed it, and 17 percent were undecided (791).7 And both organizations and a majority of survey respondents opposed the institution of a national identity card to help control illegal immigration: The authors report that 56 percent of respondents disagreed or disagreed strongly with a national ID card, and that 32 percent agreed or agreed strongly with this method of control (790). In another study, Binder, Polinard, and Wrinkle (1997) identify a tension between the economic fears aroused by immigration, and the cultural and social bonds that link Mexican Americans, especially those living along or near the border, with immigrants from Mexico (325). In an analysis of data from 756 Mexican Americans in two border counties, the authors find that age, generation, and income influence respondents’ thoughts on immigration policy. Older Mexican Americans, those who

04-S3453 1/11/06 3:06 PM Page 71

Post-Bracero Dilemmas

71

have been in the United States longer, and those with higher incomes tend to support more-restrictive immigration policies than do younger, more-recent immigrants and lower-income respondents.8 In contrast to public opinion among Hispanics on illegal immigration, the views of the Hispanic rights organizations tended to converge in the 1970s, although the major advocacy organizations took different paths to their decision on some of the issues. From the start, MALDEF was strongly opposed to any form of employer sanctions on the grounds that sanctions would endanger the rights of minorities and others. In a memo to members of the board in January 1981, MALDEF president and general counsel Vilma Martinez summarized the position on employer sanctions that the organization had taken since the 1970s, writing that they would be “certain to increase discriminatory employment practices but incapable of fair and effective implementation” because wellmeaning employers would shy away from hiring workers who look “foreign,” and biased employers would use the sanctions as an excuse to avoid hiring qualified minority workers. Employer sanctions “will not significantly open up jobs for unemployed U.S. workers” because those workers do not want most of the jobs held by undocumented immigrants. In addition, Martinez noted that MALDEF opposed creation of a national identity card system, predicting that it would endanger people’s privacy rights. For example, it would leave people who either did not have a card or simply misplaced their card vulnerable to broad searches and seizures. It also could lead to exposure of confidential financial, medical, and other records, which could be “keyed to” the identity card.9 Both LULAC and the NCLR saw their stands on undocumented immigration evolve. At its founding, LULAC’s goal had been assimilation into the mainstream culture; and illegal immigration, in LULAC’s view, stood in the way of that assimilation. In 1954, LULAC supported Operation Wetback, a federal campaign to drive undocumented workers who had crossed the southern border of the United States back to Mexico. In the 1970s, however, new leadership brought about a change in LULAC’s perspective on illegal immigration, which now emphasized immigrant rights. Although NCLR lobbied against an employer-sanctions bill initiated by President Carter in 1977, some members of NCLR’s leadership thought the organization was spending too much time on the issue. After studying the issue, Charles Kamasaki, director of the council’s Office of Research, Advocacy, and Legislation, changed his own position on sanctions and encouraged the organization to become “more activist” in

04-S3453 1/11/06 3:06 PM Page 72

72

Post-Bracero Dilemmas

its opposition to sanctions (quoted in Cortes 1992, 357). Kamasaki had concluded that employer sanctions not only were ineffective in deterring illegal immigration, but would lead to discrimination against all Hispanics. If sanctions went into force, employers would tend to lump all Hispanics together by their physical appearance.10 Both LULAC and the NCLR became convinced that employer sanctions would not work. In addition, supporting undocumented immigrants promised the empowerment of Hispanic populations. By bringing undocumented laborers aboveboard, legalization programs could give Hispanics living on the margins of society a direct voice in economic and rights-based bargaining. Both groups were proponents of broad amnesty programs that would grant temporary legal status to undocumented immigrants who had established continuous residency in the United States, status that would qualify them to apply for permanent residency in the future. The NAACP, long a proponent of immigration reform, had worked to revoke the racist national-origins quota system in the 1960s. It agreed with the Hispanic rights groups on certain legal immigration policies— eliminating statutory distinctions between native-born and naturalized citizens, and expanding immigration opportunities for refugees on an emergency basis, for example. But the groups parted company during the 1970s when the debate turned to illegal immigrants and enforcement (Tichenor 2002, 205 –206). The NAACP aligned with the AFL-CIO in support of employer sanctions. Joined by the Congressional Black Caucus, the NAACP concluded that undocumented immigration increased unemployment among native workers, with particularly detrimental effects on employment opportunities for black youth (Tichenor 2002, 234). The NAACP and Hispanic rights groups did find agreement on two issues surrounding undocumented workers: Both supported the right of undocumented aliens who had lived in the United States for a defined period of time to be eligible for naturalization and eventually to apply for citizenship, and both opposed increasing border patrols, acknowledging the potential for abuse. There were various efforts to resolve the differences between Hispanic and African American leaders. One was the Working Committee on the Concerns of Hispanics and Blacks, which, as Tichenor reports, included the NAACP, the National Urban Coalition, the National Urban League, and the NCLR. The committee was formed “to encourage cooperation and join action by the two largest U.S. minority groups” (233n58). Cortes (1992) reports his personal participation in another Washington-based

04-S3453 1/11/06 3:06 PM Page 73

Post-Bracero Dilemmas

73

“black-brown coalition” of minority advocacy groups that were members of the Leadership Conference on Civil Rights. At meetings held in 1978 and 1979, the rights organizations agreed to oppose employer sanctions. However, the AFL-CIO, a member of the umbrella group, “blocked” the LCCR from adopting that position. Although the NAACP initially agreed with the ethnic rights groups, it later returned to supporting employer sanctions (137n39). Legalization programs had the support not only of organizations representing Latinos but also of the Organization of Chinese Americans and the Japanese American Citizens League; of unions like the International Ladies Garment Workers Union; the American Immigration Lawyers Association; the city governments of New York, Los Angeles, and Chicago; the county of Los Angeles; various immigrant and refugee aid agencies; the United States Conference of Catholic Bishops, the National Council of Churches, and the American Jewish Committee; and various other groups (Cortes 1992, 138).

Temporary Farmworkers Guest-worker policy mixes economic and rights issues. Classic economic theory offers one useful way to think of foreign-worker programs. From this viewpoint, a foreign-worker program functions much like an industry subsidy: It makes an exception to federal restrictions on hiring foreign workers for one industry that is claiming economic hardship, an exception that translates into extra income for employers in that industry.11 International trade theory draws a direct parallel between regulating labor flows across borders and regulating trade and capital flows. In this view, workers constitute a factor of production. Much the same way that governments regulate trade with other nations, they write laws governing the importation of foreign labor. And the laws governing foreign labor, like those governing trade, benefit or impose costs on certain economic actors— employer associations or unions, for example. Economic theory also helps explain the mechanisms underlying the regulatory process. The decision by Congress to approve a foreign-worker program that benefits only certain constituents is a political prize, and is subject to the usual political trading that goes on among legislators from different districts. So a legislator from a steel-producing district might agree to support a foreign-worker program proposed by a colleague from an agricultural district in exchange for support for a steel subsidy.

04-S3453 1/11/06 3:06 PM Page 74

74

Post-Bracero Dilemmas

The benefits and costs that accrue to political and economic actors are only part of the story, however. The rules that govern foreign-worker programs define more than the number and type of workers an employer can import; they also define the rights of those workers as individuals—for example, the right to sue their employer in an American court of law for violations of fair-employment practices, the right to participate in collective bargaining, or the right to negotiate terms of employment with other employers when their initial contract has expired. From a simple economic perspective, supply and demand drive the movement of labor across borders. Here the decision to implement a guest-worker program is relatively straightforward, a matter of weighing economic costs and benefits. The decision-making process becomes more complex when the political implications of guest-worker programs are made part of the calculation. Consider the case of unions. Unions have an interest in protecting their members’ jobs and wage levels. For the AFL-CIO, that meant longtime opposition to traditional guest-worker programs. But for unions in industries that rely heavily on immigrant workers, the decision is complicated by the perspective of their members, many of whom have relatives who are undocumented or are themselves undocumented. These unions often support regularization and amnesty because their members do. The competing interests for and against guest-worker programs have made these programs a commodity in the political trading that goes on in Congress. What distinguishes negotiations over foreign workers from negotiations over other legislative items is the currency. When legislators barter over temporary-worker programs, they are trading in the social category, the identity, of the people who will enter the United States under those programs. That does not prevent the political trade in guest-worker programs, but it does raise the stakes for ethnic rights groups, for proimmigration unions, and for other organizations that support those programs. In the 1980s, growers used the numbers-versus-rights issue in their negotiations with ethnic rights groups. They believed that ethnic advocates would be willing to trade increases in admissions of unskilled foreign workers—which would increase the groups’ constituent base and so their political strength—for restrictions on the rights of those workers for a specified time. After a period of restriction, negotiated as part of the legislative deal, the foreign workers would have the option to become free agents in the labor market and to apply for permanent residency.12

04-S3453 1/11/06 3:06 PM Page 75

Post-Bracero Dilemmas

75

The National Hispanic Task Force on Immigration Policy, a “grouping of virtually all major national Hispanic organizations concerned with immigration issues,” advanced a rights-based argument for opposing all temporary-worker programs. Although it found no evidence that temporary farmworker programs would lessen the flow of undocumented workers, the task force predicted that the programs would “create a subclass of laborers without adequate enforcement mechanisms for the protection of civil and constitutional rights.” 13 Moreover, under the H-2 program that certain employers were requesting, there would be no mechanisms to protect workers’ civil rights or to ensure them adequate housing, health care, working conditions, and wages. Representing MALDEF in testimony before lawmakers, Vilma Martinez expressed concern that even aliens who are lawful permanent residents “are not adequately protected by federal civil rights statutes.” 14 She cited as an example the U.S. government’s failure to prohibit conspiracies by private citizens to deprive aliens of protected rights. Although the Fourteenth Amendment extends its protection to any “person” in the United States, Congress, by implementing 18 U.S.C. 241, limited its protection to citizens, allowing groups like the Ku Klux Klan to “terrorize aliens without being subjected to federal criminal prosecution for civil rights violations.” At the same hearings, other MALDEF representatives pointed to the Supreme Court’s 1978 ruling in Espinoza, that federal employment law, which prohibits discrimination on the basis of national origin, does not apply to discrimination based on alienage. This put into question “the protection afforded even lawful resident aliens by federal civil rights legislation in the areas of housing, education, municipal services, and public accommodations.” MALDEF spokespersons also addressed a recommendation by the Reagan administration for a two-year guestworker program that would bring 50,000 Mexican nationals a year into the United States on an experimental basis. The temporary workers would be allowed to work for up to twelve months and would not be tied to the agricultural sector, but they could not work in states that certified that they had an adequate supply of American workers. Although entitled to standard wages and working conditions, these temporary workers would not be eligible for unemployment insurance or public assistance programs and would not be permitted to bring their spouse or children to the United States. MALDEF criticized the proposal for not giving guest workers the right to organize or to access the

04-S3453 1/11/06 3:06 PM Page 76

76

Post-Bracero Dilemmas

courts for redress and for failing to prohibit employers from using temporary farmworkers as strikebreakers: “Workers, who over a period of time, staff our industries, pay taxes, and contribute to our economic welfare should receive the social and economic benefits accorded U.S. residents. They should be given an opportunity to remain here and to become legal permanent residents and, eventually citizens.” The compromise that finally was enacted under the IRCA in 1986 did restrict temporary workers’ rights for a period of three years; but aliens who could prove they had lived in the United States for at least three years and had worked at least ninety days in agriculture in each of those years became eligible for temporary-resident status and then, one year later, for permanent-resident status.15 The NCLR did not support the IRCA but stated that it was “probably the least damaging legislation possible given the current political climate” (Kamasaki and Briceño 1986, 3). For ethnic rights groups, generous legalization terms presented not only an alternative to creating a vulnerable underclass of undocumented workers but also an opportunity to expand both the political power of ethnic voters and the groups’ own social base. Advocacy for modified guest-worker programs, then, was in part a form of rights-oriented identity politics, in which legalized guest workers would be included as part of the ethnic identity group in the future. International rights groups have been working to establish minimum standards of fairness and decency in employment conditions for migrant foreign workers since the 1920s. The rationale has been that migrants’ participation in the labor force contributes to national development in the host country; by contributing, the foreign workers become members of the economic community and so deserve the protection of labor and civil rights laws. In the mid-1970s, the International Labour Organization (ILO), an agency of the United Nations, called for states to respect the human rights of all migrants, including those unlawfully residing or working in a nation’s territory. In 1990, the General Assembly of the United Nations adopted the United Nations International Convention on the Protection of the Rights of All Migrant Workers and Members of Their Families; the convention went into force in 2005.16

Illegal Immigration and the Union Movement During the 1980s and 1990s, the labor movement experienced internal conflict over illegal immigration. Early on, a number of unions with large immigrant memberships voiced strong opposition to employer

04-S3453 1/11/06 3:06 PM Page 77

Post-Bracero Dilemmas

77

sanctions. Their policy contrasted sharply with the position of the AFLCIO and other unions that strongly supported sanctions. The AFL-CIO would change its position on illegal immigration dramatically in 2000, following a decade of intense debate within the federation, forming a proimmigrant rights coalition with Hispanic advocates. Organized labor influenced liberals in Congress to lead the battle against illegal immigration in the 1970s. From the start of the bracero program in 1942, labor had been its strongest opponent. As Tichenor (2002) notes, in 1950 the AFL had called for legislation to “bar illegal entries of aliens” and “criminal penalties for employers who hire such (illegal) labor” (224 –225). In 1952, the labor movement succeeded in persuading Senator Paul Douglas (D-IL) to propose an employer-sanctions amendment to the McCarran-Walter Act. Although sanction provisions were included in the law, Senator James Eastland’s (D-MS) Texas Proviso nullified their effectiveness by sheltering employers of undocumented immigrants from punishment (225).17 The CIO supported an end to national-origin quotas and policies after the Second World War. In contrast, the AFL remained committed to the quota system—a restrictive reform that it had helped steer through Congress in the 1920s. In 1955, the AFL and the CIO merged; and the combined organization adopted the more-generous orientation of the CIO on immigration policy (Tichenor 2002, 204). The national labor federation wanted liberal family-unification policies to be the cornerstone of the Hart-Cellar Act; but it also wanted regulatory mechanisms that would protect union members from wage competition when legal immigrants were admitted into the workforce. The position of organized labor was highly restrictive on illegal immigration. In the 1970s, the AFL-CIO called on Congress to enact an employer-sanctions law as a deterrent to illegal immigration. At the same time, several unions were pressing the AFL-CIO from within to adopt a liberal position on amnesty for illegal immigrants. These unions also took their policy statements to lawmakers in testimony before congressional committees. Over time, the AFL-CIO would adopt their thinking. In the United States, employer associations in industries that rely on immigrant workers generally have lobbied lawmakers for more liberal laws governing the admission of temporary and permanent immigrant workers. In turn, organized labor has strongly resisted liberal immigration policies, especially in times of high unemployment. In the 1970s, the AFL-CIO actively pressed for sanctions against employers that hire

04-S3453 1/11/06 3:06 PM Page 78

78

Post-Bracero Dilemmas

undocumented workers and for a national identity card that could not easily be counterfeited. As the general pool of undocumented migrants grew in the United States after the end of the bracero program, the AFLCIO’s opposition to immigration grew stronger. That campaign culminated in passage of the Immigration Reform and Control Act of 1986. The AFL-CIO’s opposition to liberal immigration policies was rooted in its fears that immigrant workers would displace native workers and lower wage levels. Yet other unions—in particular the ILGWU and the Amalgamated Clothing and Textile Workers Union (ACTWU)—had been pushing for a more liberal admissions policy for European and non-European immigrants since the 1920s (Parmet 1981, chap. 12).18 The unions’ motivation was clear: The bulk of workers in the clothing and textile industries are immigrants. In the 1970s and early 1980s, both unions pressed the AFL-CIO to adopt a more liberal admissions policy. The ILGWU, which initially opposed employer sanctions, was even willing to support sanctions in exchange for a liberal amnesty for undocumented workers; and it lobbied Congress independently for generous amnesty throughout the 1980s. A third international union, the Hotel Employees and Restaurant Employees and Bartenders International Union (HEREBIU) took a similar position, calling for generous amnesty, but did not lobby Congress on its own.19 After 1990, four unions publicly broke from the AFL-CIO’s position on economic sanctions: the ILGWU, the ACTWU, the HEREBIU, and the Service Employees International Union (SEIU), which organized janitors and public service workers. The immediate impetus was a March 1990 report issued by the U.S. General Accounting Office (GAO) on the impact of employer sanctions. In 1986, the ILGWU, the ACTWU, and a number of Hispanic rights organizations had accepted the Immigration Reform and Control Act’s complex package of reforms, including employer sanctions, on the condition that the GAO would submit a report on the impact of the sanctions after they had been in place for three years. At that time, the groups stated their intention to call for repeal if increases in discrimination were found as a result of the sanctions. When the GAO reported in 1990 that discrimination against Hispanics had increased substantially since the act was implemented, the unions made their opposition to employer sanctions—and to the AFLCIO’s support of those sanctions—known.20 The first to call for repealing sanctions in 1990 was the ILGWU; the ACTWU and the HEREBIU followed shortly after. In contrast to these three unions, which had advocated a generous amnesty for

04-S3453 1/11/06 3:06 PM Page 79

Post-Bracero Dilemmas

79

undocumented aliens during the previous decade, the fourth union, the SEIU, was breaking with its previous restrictionist position. It was calling for repeal of employer sanctions because of changes that took place in the structure and labor markets of the cleaning industry in Southern California during and after the mid-1980s. On the other end of the policy spectrum were the craft unions in the Building and Construction Trades Department of the AFL-CIO, which led the push for strong sanctions and border controls. Among the largest internationals in this department were the United Brotherhood of Carpenters and Joiners of America, the International Union of Operating Engineers, the International Brotherhood of Electrical Workers, and the United Association of Journeymen and Apprentices of the Plumbing and Pipefitting Industry of the United States and Canada. One of these craft unions, the Carpenters and Joiners, lobbied independently from but in line with AFL-CIO policy. In addition, the AFL-CIO Professional Department lobbied Congress for strict enforcement of immigration laws; in particular, the department argued the restrictionist case of the International Alliance of Theatrical Stage Employees, Moving Picture Technicians, Artists and Allied Crafts of the United States, Its Territories and Canada. Table 4.1 lists three categories of unions that lobbied Congress independently of the AFL-CIO on the employer-sanctions law. Many were federation affiliates; in addition there were independent internationals and an AFL-CIO–affiliated local. Not all of these groups agreed on sanctions: Some wanted sanctions repealed, while others were lobbying for stricter enforcement. One independent international, the United Electrical, Radio and Machine Workers of America, consistently pressed for liberal immigration policies. Two independent longshoremen’s unions, on the other hand, in testimony before Congress in 1989, called for amendments to the IRCA that would apply employer sanctions specifically to their industry.21 One Southern California local of the International Union of Electronic, Electrical, Salaried, Machine and Furniture Workers called for repeal of employer sanctions; this local’s members work in industries that are highly sensitive to imports. The union’s efforts were a response to the increasing number of companies that were moving production from Southern California to maquiladoras, assembly or manufacturing plants along the northern border of Mexico. Table 4.2 shows the percentage of Hispanic workers in key industries in 1983.22 A comparison of Tables 4.1 and 4.2 shows that at least two of the unions that invested resources to lobby independently in opposition to employer sanctions—the ACTWU and the ILGWU—represented

04-S3453 1/11/06 3:06 PM Page 80

80

Post-Bracero Dilemmas ta b l e 4 . 1 Unions Lobbying Independently of the AFL-CIO on Employer Sanctions, 1980s

Unions AFL-CIO affiliates: ACTWU Carpenters and Joiners Farmworkers Food and Commercial Workers HEREBIU ILGWU Laundry and Dry Cleaners SEIU Theatrical Stage Employees and Moving Picture Technicians Independent internationals: Electrical, Radio, and Machine Workers Longshoremen (two unions) AFL-CIO local: Electrical and Furniture Workers, Southern California

Position on employer sanctions Liberal Restrictionist Restrictionist until mid-1980s, then liberal Restrictionist Liberal Liberal Liberal Restrictionist until mid-1980s, then liberal Restrictionist Liberal Restrictionist

Liberal

s o u r c e s : AFL-CIO (1983, 1985, 1987); House Committee on the Judiciary, Subcommittee on Immigration, Refugees, and International Law, Application of Employer Sanctions to Longshore Work: Hearing on HR 2138, 101st Cong., 1st sess., June 14, 1989; and prepared statement of Sol Chaikin before the House Committee on Education and Labor, Hearing on Employment Discrimination and Immigration Reform, 98th Cong., 1st sess., May 19, 1983.

workers in an industry with a large concentration (14.4 percent) of Hispanic workers. In contrast, Hispanics made up only 5.8 percent of the workforce in the construction industry, a percentage that translated into minimal influence within the Carpenters and Joiners union, which took a restrictionist position on illegal immigration policy.23 Still, the percentage of Hispanics in an industry’s labor force is not necessarily an indicator of a union’s policy on illegal immigration. Organizing in an industry in which Hispanics comprised 8.5 percent of hotel and motel workers and 6.3 percent of restaurant workers, the Hotel and Restaurant Employees union took a liberal policy position. Yet the Food and Commercial Workers union, which represents cannery workers in an industry with a large percentage of Hispanic laborers (15.7 percent), takes a restrictive position. The sheer percentage of Hispanics in an industry’s workforce does not determine a union’s position on illegal immigration policy. In the 1980s, unions had to make strategic decisions about how to increase their membership. If a union believed its power depended on organizing immigrant workers, it was likely to adopt a liberal immigration

04-S3453 1/11/06 3:06 PM Page 81

Post-Bracero Dilemmas

81

ta b l e 4 . 2 Percentage of Hispanic Workers in Key Industries, 1983 Industry Agriculture Horticultural services Agricultural services, except horticultural Agricultural production, livestock Nondurable goods Canned and preserved fruits and vegetables Miscellaneous and not specified food preparations and related products Personal services: Barber shops Laundry, cleaning, and garment services Hotels and motels Restaurants and bars Manufacturing Toys, amusements, and sporting goods Miscellaneous and not specified manufacturing industries Mining, metal Apparel and other finished textile products Transportation, warehousing and storage Services, dwellings and other buildings Leather and leather products Construction

Total workers (1,000s)

Hispanic workers (percent)

368 278 1,494

11.7 10.7 10.6

214

15.7

203

11.4

93 418 892 4,875

11.7 10.6 8.5 6.3

141

11.2

367 70 1,154 88 471 243 6,149

11.2 16.4 14.4 12.9 11.4 10.5 5.8

s o u r c e : U.S. Department of Labor (1988), table B-17.

position. What follows is a description of the role of immigrants in certain labor markets and the policy incentives of those unions for which immigration was a salient issue.24

Carpenters and Joiners, Longshoremen, and Theatrical Stage Employees and Moving Picture Technicians Hiring procedures—provisions for union hiring halls in their contracts with employers—linked the Carpenters and Joiners and the two longshoremen’s unions that lobbied independently for passage and strict enforcement of employer sanctions.25 In 1983, the Bureau of National Affairs reported that the incidence of hiring procedures was more than 50 percent in the construction and maritime industries, compared to 23 percent in the general sample.26 In the case of the Carpenters and Joiners, union control of labor supply was threatened by the employment practices of nonunion construction contractors. Migrant day laborers

04-S3453 1/11/06 3:06 PM Page 82

82

Post-Bracero Dilemmas

were being hired at less than half the union rate and were selected through informal migrant recruitment networks and day-laborer pools visible on the streets of major U.S. cities (and even supported, in part, by city governments in Los Angeles and elsewhere). In 1989 hearings on the application of employer sanctions, the International Longshoremen’s Association and the International Longshoremen’s and Warehousemen’s Union both called on Congress to clarify how employer sanctions in the bill that would become the Immigration Act of 1990 would apply to foreign vessels in U.S. ports. The unions wanted the legislation to expressly state that foreign crewmen and noncrewmen working in maritime operations would be subject to the same employment-verification procedures and employer sanctions as were native workers in the industry.27 The International Alliance of Theatrical Stage Employees and Moving Picture Technicians also called on the AFL-CIO to press for stricter visa requirements for foreign technical and craft workers who often entered the United States on film crews with foreign directors (AFL-CIO 1987, 469 – 470). The alliance restricts the labor supply in the film industry by charging high membership fees and usually negotiating closed-shop agreements with filmmakers.

International Ladies Garment Workers Union The ILGWU had consistently advocated for liberal immigration policies since the 1920s. The union had called on the AFL-CIO to oppose legislation imposing employer sanctions in 1979 (AFL-CIO 1987, 416) and again attempted, unsuccessfully, in 1983 to secure the federation’s opposition to the employer sanctions that would become a provision of the Immigration Reform and Control Act of 1986 (AFL-CIO 1983, 335).28 Sol Chaikin, then the president of the ILGWU, stated at IRCA hearings in 1983 that for employer sanctions to be an effective deterrent, fines would have to be set high—he suggested $20,000 (in 1983 dollars). Anything less would be treated as “a standard cost of doing business.” 29 Chaikin argued that criminal penalties and tamper-proof identification cards probably would work, but that criminal penalties were extreme and not a realistic option. In 1985, Jay Mazur, who would succeed Chaikin as president of the union the next year, shifted the emphasis of the union’s lobbying efforts. Now the ILGWU took the lead among unions in pressing for generous amnesty provisions for undocumented workers.

04-S3453 1/11/06 3:06 PM Page 83

Post-Bracero Dilemmas

83

Defending the union’s liberal position on amnesty at hearings in 1990, Muzaffar Chisti, director of the ILGWU’s Immigration Project, stated that the flow of undocumented immigrants did not lower the wages of ILGWU members and that unemployment was not a problem in the industry. In response to a questioner who claimed that reducing the supply of low-wage immigrants could lose the whole industry to imports, Chisti responded: “If you . . . look at what has happened to our industry since 1965, you would have to say that foreign trade policy has had 80 percent of the impact on that, compared to any other single issue. And therefore, for us to talk about immigration in isolation from other factors is fallacious.” 30 At a later hearing, after assessing the impact of employer sanctions on the industry, an ILGWU representative explained that the ILGWU had changed its position on sanctions. In the union’s view, sanctions had failed to achieve the wanted public policy effect of reducing illegal immigration: Instead of improving workplace conditions, sanctions had contributed to a resurgence of illegal sweatshops in the underground economy. The sanctions had “segregated the labor market” because now legal workers could not work in the illegal shops and undocumented workers could not work in the legal shops. Before the implementation of the IRCA’s employer sanctions, legitimate and nonlegitimate sectors competed for the same workers, and the union had some ability to set wages for all workers, documented and undocumented. Post-IRCA, in the “absence of competition” between the two sectors, employers who hired illegal workers were now free to establish their own workplace conditions and to set wages for undocumented workers. Employer sanctions, moreover, had led to a rise in illegal homework.31 As Chisti had stated in testimony at the 1990 hearing, the overwhelming source of pressure on the union was imports. By U.S. Department of Labor measures, several of the various industries under consideration in this study are sensitive to imports: apparel and textile manufacturing, electronics, and furniture manufacturing. Kahn (1986) has used time-series data from the 1950s to the 1980s to show that a significant portion of fluctuations in ILGWU membership can be explained by imports: That is, as competition from imports has reduced jobs in the industry, ILGWU membership has fallen (33 – 35). But studies of the garment industry in New York (Waldinger 1985, 1986) and in Los Angeles (Maram and King 1983) provide survey evidence that immigrants or temporary migrants have helped certain subsectors survive and even expand at a time when the domestic share of the U.S. garment

04-S3453 1/11/06 3:06 PM Page 84

84

Post-Bracero Dilemmas

market was dropping sharply. The garment industry is made up of three types of businesses: manufacturers who design garments; contractors who specialize in specific production tasks; and jobbers, wholesalers who act as intermediaries between producers and retailers. In both New York and Los Angeles, the highly competitive contractors make up the majority—74 percent in New York and 69 percent in California. The Waldinger and Maram-King surveys report that contractors believe that low-cost labor is critical to their survival. Summarizing these studies and others, the GAO (1988) concluded that the employment of low-cost undocumented workers in the contracting firms “has created employment for skilled native or legal workers in the preserved or expanded subsectors, although it may have kept wages low for substitute native or legal workers seeking jobs as factory operatives in those firms that would survive even without the availability of international migrant labor” (30). Waldinger (1986) has shown that the New York apparel industry is not likely to disappear with foreign competition because the spot markets that respond to fashion changes have created niches in women’s outerwear (in which production has increased relative to women’s underwear). In underwear manufacturing, mass-production runs are more efficient, and more companies in this category have moved production to the maquiladoras, the plants along the northern border of Mexico that offer U.S. business inexpensive labor and other savings. In New York, long an international center for fashion designers, companies are able to respond most flexibly to rapid changes in women’s fashion through agglomeration, by contracting out design, pattern cutting, even marketing to a number of smaller companies. Sewing is done by small contractors in Chinatown and other areas near the central garment district, their workforce in large part Chinese and Dominican sewing machine operators. New York specializes in fashion-sensitive outerwear, particularly dresses and sportswear (Waldinger 1986, 89 –122). The Los Angeles niche includes slacks, swimwear, and other casual wear, assembled by predominantly Hispanic sewing machine operators in small contracting shops throughout the Los Angeles basin. In Los Angeles, too, there is a central garment district where designers are employed by manufacturers. In this kind of industry structure, the different companies are complementary; and within each type of business, there is a hierarchy by skill level of workers. Laslett and Tyler (1989) note that employment in the clothing industry in California increased between 1973 and 1982, in contrast to its decline nationwide (112).

04-S3453 1/11/06 3:06 PM Page 85

Post-Bracero Dilemmas

85

The structure of the ILGWU was semi-industrial; locals were organized along craft lines, but the union organized all workers in the industry (Seidman 1942, 226). Union leadership in the main came from the skilled trades. After the Second World War, membership in locals of highly skilled tailors fell because of a shift in American fashion taste toward casual wear. Still, in the New York and Los Angeles regional boards and in the New York–based international union, relatively skilled workers were more senior and exerted stronger influence on union policy than the more marginally employed immigrants in small immigrantowned contracting companies (Laslett and Tyler 1989).

Amalgamated Clothing and Textile Workers Union The other union in the import-sensitive clothing industry was the ACTWU, which in the 1980s was the ILGWU’s counterpart in men’s and children’s apparel manufacturing. It also organized textile mills and allied products. The ACTWU did not lobby independently of the AFLCIO on immigration issues during the IRCA debate of the 1980s, but it supported the AFL-CIO’s position on sanctions in exchange for the federation’s taking a liberal position on amnesty.32 In 1991, the ACTWU began to play a prominent role in political negotiations, taking up the American labor movement’s banner and advocating for the inclusion of provisions to protect labor standards in the North American Free Trade Agreement (NAFTA). In numerous statements on NAFTA, the ACTWU argued for more-liberal immigration policies that would guarantee all immigrants protection under U.S. labor standards. Notice the focus on protecting the rights of immigrants in the United States, not on issues of border control (United Nations Association 1993). Like the ILGWU, the ACTWU did call for repeal of employer sanctions on the basis that they had increased discrimination toward Hispanics and expanded the informal economy in the apparel industry (UCLA 1992).

Hotel Employees and Restaurant Employees and Bartenders International Union In its 1988 report, the GAO concluded that “international migrants in the restaurant industry create jobs for waiters, bartenders and other complementary native workers. At the same time, native or legal workers seeking unskilled kitchen jobs have suffered from the competition of low-wage international migrants” (35). Migrants in the restaurant

04-S3453 1/11/06 3:06 PM Page 86

86

Post-Bracero Dilemmas

industry, like those in the garment industry, do not displace native workers; instead they fill vacancies in the growing service sectors of local economies. The leadership and stable membership of HEREBIU traditionally came from waiters and waitresses, bartenders, and cooks, although significant numbers of its members included hotel cleaners, a large proportion of them undocumented. Two case studies provide useful evidence. Bailey (1985) in New York and Morales (1985) in San Diego found that the restaurant industry is made up of three sectors: fast-food restaurants, intermediate or coffee house establishments, and full-service restaurants. In the fast-food sector, employers prefer young native workers to international migrants because of the English-language skills necessary for customer service. The intermediate sector, however, hires large numbers of undocumented migrants for back-of-the-house (kitchen) jobs; and the fullservice sector primarily employs undocumented migrants in its kitchen jobs (Bailey). Both Bailey and Morales based their studies on surveys of employers, and both assessed the probable effect of restricting the supply of international migrants. In their view, the increased cost of kitchen labor would increase the price of food, shifting consumer demand toward the fast-food sector, which does not rely on migrants. The intermediate and full-service sectors would shrink but would survive; consequently, some of the more skilled native and legal workers would lose their jobs. These case studies provide a plausible reason—the complementarity of labor—for HEREBIU’s support of liberal immigration policies. Because restaurants and hotels are in a nontrade sector, a sector that is not subject to competition from imports, the industry’s survival is not threatened as immediately by cutting off immigrant labor as the garment industry is. This may explain why HEREBIU was not as active in lobbying for liberal immigration policies as the ILGWU was.

Service Employees International Union The SEIU broke in 1991 from a position favoring immigration restriction. The union’s membership traditionally came from janitors working in the high-rise districts of Chicago, Los Angeles, New York, and other major cities. About half of its members in the 1970s were African Americans, and they would win wages and benefits valued at more than $12 an hour at the union’s peak, in 1983. These union members worked mainly for large unionized contractors. In the mid-1980s, a group of midsized

04-S3453 1/11/06 3:06 PM Page 87

Post-Bracero Dilemmas

87

companies employing a largely Hispanic migrant workforce at about $4 an hour underbid the larger companies for downtown contracts. The union was practically displaced in the local industry. According to a survey by the Bureau of Labor Statistics, average janitorial wages dropped 14 percent between 1983 and 1985.33 The SEIU responded with a strategy of areawide (rather than site-specific) organizing of the mostly undocumented Hispanic janitors, using confrontational tactics and building alliances with community organizations. As of July 1989, the union had achieved 55 percent unionization of downtown Los Angeles and had increased union wages in a master contract to between $5.00 and $6.35 per hour, compared to $4.00 to $4.65 per hour for nonunion workers (Loucky, Hamilton, and Chinchilla 1999). The SEIU’s strategy worked, but it raised a question: In this case, immigrant workers clearly had been used as substitutes for higherwage native black workers. Why, then, did the union stop supporting restrictions on immigration? Holter and Wong’s (1993) study of the impact in Southern California of the amnesty and employer-sanctions provisions of the IRCA suggests an explanation. The study found that the labor-market structure of the cleaning industry changed once the unionized black workforce was replaced by immigrants. The low wages attracted a ready supply of transient laborers from Mexico and Central America, many of whom left janitorial work as soon as they could secure a better job. With the displacement of the union and the black janitors, the janitorial jobs became part of a high-turnover and low-wage labor market. The SEIU also found that amnesty provisions for undocumented workers had opened up a pool of workers receptive to organizing: Amnesty removed the intimidation that typically had made illegal immigrants reluctant to associate themselves with the union in the years before the IRCA was passed. Increasingly during the 1980s, the background of migrants from Mexico diversified to include more workers from the cities with experience in unions, and many of them were receptive to organizing efforts. But employers were using the IRCA requirement that all workers show papers certifying their legal status to intimidate prospective union members. The SEIU generally does not maintain hiring halls, except one in San Francisco for janitors. It relies on collective bargaining rather than on restricting the labor supply to secure wage increases. When faced with an already transformed workforce and little prospect of native workers’ wanting to work in the

04-S3453 1/11/06 3:06 PM Page 88

88

Post-Bracero Dilemmas

industry, the SEIU turned to aggressive organizing campaigns among immigrant workers, a strategy that was being hampered by employer sanctions.

Unions in Agriculture and Food Processing The United Food and Commercial Workers (UFCW) and the United Farm Workers (UFW) both actively lobbied against agricultural employers’ proposals to renew guest-worker programs under the IRCA. The UFCW represents meat cutters, meatpackers, processors, and retail clerks. Traditionally, meat cutters (and later retail clerks, who have made up most of the union’s membership since its merger) argue that union wages in the food-processing industry are closely tied to union farmworkers’ wages.34 Because agricultural employers historically have attempted to break the farmworkers’ union by hiring guest workers, the UFCW lobbied on the side of the farmworkers, while generally supporting the AFL-CIO’s positions on employer sanctions. The issue of employer sanctions posed a dilemma for the UFW, as it did for the ILGWU. According to one mid-1980s survey of California farmworkers, more than three-fourths of the farmworkers in the state had permanentresident green cards, but more than half of the male respondents who were married but not accompanied by their wives were undocumented (Mines and Martin 1986). The UFW has organized documented and undocumented workers alike. The union opposed employer sanctions early in the IRCA debate—not after the IRCA’s implementation— arguing that sanctions would be used by employers to discriminate against all Hispanics; it also supported unconditional amnesty programs. The UFW’s opposition to sanctions is explained in part by its mission to protect its undocumented members. In addition, the UFW had changed its strategy, moving from organizing fieldworkers to encouraging consumer boycotts. Its membership and number of contracts dropped dramatically in the 1980s. It is possible, then, that community alliances with Hispanic rights groups— critical allies in the UFW’s consumer-boycott campaigns—were a large influence on its position on sanctions. The economic reasons for the UFW’s opposition to guest workers are straightforward.35 Agricultural employers in southwestern and western states traditionally have used guest-worker programs, particularly the bracero program, to displace unions and to lower farm wages. Between 1945 and 1965 —the bracero program was in operation from 1942 to

04-S3453 1/11/06 3:06 PM Page 89

Post-Bracero Dilemmas

89

1964 —the ratio of real farm wages to nonfarm wages dropped by about 40 percent (Runsten and Le Veen 1981, 24). Employers also overtly used Mexican laborers to break strikes. After the termination of the bracero program in 1964, efforts to organize farmworkers met with considerable success in California, largely because braceros were no longer available to break strikes. On average, the ratio of farm wages to nonfarm wages rose 50 percent from 1965 to 1980. Although a small percentage of U.S. growers had unionized workforces, they paid higher wages, on average. By the late 1960s, employers were using both “green-card” Mexican commuters, many of whom were former braceros, and illegal immigrants to break union strikes. Also growers increasingly were relying on farm-labor contractors to recruit and supervise Mexican workers. Martin (1993) reports that after 1977, when the INS had apprehended more than 1 million undocumented immigrants, the UFW called for employer sanctions and stronger enforcement of laws to control illegal immigration (66). The increase in illegal immigration, coupled with the union’s internal problems, drastically reduced the UFW’s strength in the 1980s. The number of contracts it won fell from 120 in 1980 to 20 in 1990; and the number of members the union claimed fell from 60,000 to 6,000 (67). The UFW’s leaders began to oscillate in their position on undocumented immigration in the early 1980s. UFW president Cesar Chavez continued to cite illegal immigration as a major factor in the joblessness and the low wages of settled farmworkers. UFW vice president Dolores Huerta, however, emphasized that the source of the UFW’s problems was not illegal aliens but the government’s failure to enforce labor laws (Martin 1993, 67). Both Chavez and Huerta acknowledged a reality: that illegal immigration would be difficult to control as long as U.S. employers continued to offer work to undocumented laborers. Increasing numbers of UFW members themselves were undocumented, as were many of their relatives and friends.36 During the IRCA debate, the UFW reversed its position and began actively opposing employer sanctions; it also advocated liberal amnesty provisions for illegal aliens. Like Hispanic rights organizations, the UFW argued that employer sanctions would increase workplace discrimination against all Hispanics, not just illegal workers. Employers could disrupt organizing drives by intimidating undocumented workers or even organizers. Growers could use the employer-sanctions provisions of the IRCA to require all workers to show work papers or else be subject to firing. With a substantial part of its membership base consisting of undocumented workers,

04-S3453 1/11/06 3:06 PM Page 90

90

Post-Bracero Dilemmas

it was in the union’s interest to support those workers’ right to join the union and participate in strikes. One indicator of the UFW’s base among illegal workers was its decision to become one of the California statecertified agencies funded to help undocumented workers learn English, apply for amnesty, and eventually be naturalized.

Constructing Migrant Categories by Trades In the 1980s, growers, organized labor, and Hispanic rights groups entered into negotiations with lawmakers over a temporary-farmworker program. In the final stages of the political bartering over the IRCA, the biggest obstacle was reaching a consensus on a guest-worker program. Western growers proposed the new program to compensate for the depletion of their supply of illegal workers should the IRCA pass. The House and Senate agriculture committees heard proposals and recommended various programs that were admitted for floor consideration in the form of amendments to the judiciary committees’ versions of the bill. When policy controversy heats up over foreign-worker programs in agriculture, American labor and ethnic rights groups invariably warn against starting again with anything like the bracero program. That program, which was established in response to growers’ claims of labor shortages during the Second World War, had been administered by both the U.S. and Mexican governments. The program recruited migrant farmworkers from Mexico to work in America on guest contracts at wages that were tightly regulated in comparison to post-bracero practice. By the 1960s, however, support for the program had waned as large growers increasingly mechanized their production, and as the opposition of unions and civil rights groups gained strength with the civil rights movement. After Congress refused to reauthorize the bracero program in 1964, growers who relied on Mexican farm labor resumed the large-scale recruitment and employment of illegal workers. In the 1980s and 1990s, the cross-border market in illegal immigrants and migrant laborers was the source of the most hotly debated immigration issues: the public costs of illegal immigration, the legal rights of undocumented persons, and the conflict between federal and state interests in regulating illegal immigration. Twice—in 1986 and again in 1996 — Congress responded to public pressure against illegal immigration by tightening borders and implementing internal controls. And both times, growers of

04-S3453 1/11/06 3:06 PM Page 91

Post-Bracero Dilemmas

91

perishable crops in California renewed their campaign for a guestworker program that would guarantee them a supply of migrant labor. Congress deliberated for fifteen years over various versions of the Immigration Reform and Control Act of 1986 before the act was passed. Agreement seemed closer when a deal was reached to exchange a firsttime sanction against employers that hire undocumented immigrants for an amnesty for undocumented immigrants already living in the United States. But western growers threw up a last obstacle by insisting that the federal government replace their access to illegal migrant workers; the growers called for a new guest-worker program. Senator Pete Wilson (RCA) introduced an amendment to the immigration enforcement bill, calling for a program that would allow growers to recruit and employ migrant farmworkers for up to nine months each year. Wilson’s program would have required migrants to move from farm to farm to earn a living and would have prevented them from moving out of the agricultural labor force. In the House, Leon Panetta, a Democrat representing a farm district in California’s Imperial Valley, proposed a similar program, but one that in some respects imposed fewer regulations on agribusiness than Wilson’s proposal did. Under Panetta’s program, growers needed to apply for workers only seventy-two hours in advance, and the migrant workers could stay in the United States for up to eleven months and could move between agricultural employers in a specified region. In the end, a compromise negotiated by Representative Howard Berman (D-CA) and Senator Charles Schumer (D-NY) was incorporated into the Immigration Reform and Control Act of 1986. Berman, an advocate for farmworkers’ rights, played a crucial role in securing some protection for undocumented workers. The agreement included creation of a special agricultural workers (SAWs) program, which would meet growers’ needs for migrant labor through the legalization of undocumented farmworkers. The details are noteworthy for the earnedresidency provision offered to undocumented farmworkers.37 The agreement identified two categories of agricultural employers. The first was producers of “fruits and vegetables of every kind and other perishable commodities,” with other perishable commodities defined by the secretary of agriculture. The Department of Agriculture ruled that the production of perishable commodities requires “uncertain and unpredictable” demand for seasonal farm labor; hence, producers of these crops were given more flexible access to labor, and workers— documented or undocumented—who harvested perishable crops were eligible for a special earned-residency program. Producers of non-

04-S3453 1/11/06 3:06 PM Page 92

92 Post-Bracero Dilemmas perishable crops (e.g., livestock and dairy products, cotton, grains, and sugarcane) were required to hire their workers—native or foreign, permanent or temporary—through a contract guest-worker program, the H-2A program, which was created when IRCA split the existing H-2 program into two parts. Henceforth, H-2A visas would go to seasonal or temporary agricultural workers, and H-2B visas would be issued to nonagricultural workers who were needed in seasonal or intermittent jobs, or in periods of high workload. H-2B workers typically would be employed in landscaping, in seasonal construction, in seasonal hotel work, or in certain manufacturing and food-processing jobs. Martin (1993) reports that unions and other farmworkers’ advocates lobbied for a broad definition of perishable commodities “in order to legalize as many illegal alien farmworkers as possible” (93). The legalization program allowed farmworkers who had carried out at least ninety days of seasonal agricultural work in a perishable commodity between May 1, 1985, and May 1, 1986, to become temporary resident aliens (TRAs), and then permanent resident aliens (PRAs) in December 1990 (one year earlier if the workers had worked for ninety days in 1984, 1985, and 1986). The SAW legalization program was more lenient than the amnesty provision offered under the IRCA to undocumented migrants in residence since January 1, 1982: After entering the United States without documentation, undocumented workers did not have to remain continuously in the United States to become eligible for SAW status; only one hour of seasonal farmwork was required to establish proof of a day’s work; and workers did not have to take an English test to receive a permanent-resident card (93). To satisfy growers’ concerns that their labor force would be depleted over time as the SAWs became permanent residents, lawmakers included in the IRCA a replenishment agricultural worker (RAW) program that could go into effect in 1989, three years after the IRCA was enacted, but was never used. Nonetheless, the law was unprecedented. It provided that if the secretaries of labor and agriculture projected a labor shortage, replenishment workers would be admitted for seasonal agricultural work in perishable commodities. Growers would not be required to certify that U.S. workers were not available. The replenishment agricultural workers would be free agents rather than bound by a contract; this meant they would be able to move from one agricultural employer to another. They could become PRAs after ninety days of work in perishable commodities for each of three years. To become U.S. citizens, the replenishment workers would have to work five 9-day periods in each of five years.

04-S3453 1/11/06 3:06 PM Page 93

Post-Bracero Dilemmas

93

Support for guest-worker programs is largely partisan: Republicans tend to favor them more than Democrats do. But concentrated agricultural interests cut across party lines, a fact of political life that was evident in the mid-1990s, in the debate over a traditional guest-worker program. The SAW program accepted applications during 1987 and 1988; the RAW program expired in 1992. After the two programs ended, California’s growers began lobbying for a new contract-labor program. In the 106th Congress (1999 –2000), most Republicans and some Democrats in agricultural states favored the growers’ proposals in light of long-term projections of a shortage of low-wage laborers. For almost a decade, growers had been asking Hispanic leaders to accept temporaryworker programs with the argument that some form of legalization alternative for these migrant workers would expand the Hispanic voting population in the future.38 In the face of a threatened veto by President Clinton, the growers specifically sought support from the Southwest Voter Registration Education Project, a group that focuses on citizenship and voter registration drives among Latinos. Spokespersons for both MALDEF and the NCLR opposed any contract-labor program that would lower prevailing wage and housing standards for U.S. workers. They were open to negotiating the terms of a labor program, however, if prevailing wage standards were retained and a phased-in amnesty was granted to undocumented immigrants already in the country and to contract laborers at some point in the future.39 Although growers refused to negotiate directly with the UFW and Hispanic rights groups in the years leading up to passage of the IRCA, union and ethnic-group lobbyists exerted indirect influence to a significant degree through Howard Berman’s representation of their interests. That was how they won amnesties for undocumented immigrants who could prove extended residency in the United States, for instance. The SAW compromise differed from conventional amnesties because legalization was based on a right to stay in the United States that was earned through work. In both cases, Hispanic leaders insisted on fairly lenient terms for granting undocumented residents legal status.

Summary In the trade-off between numbers and rights, winning the right of legalized workers to be eligible for legal permanent residency, which could lead to naturalized-citizen status, proved critical in the 1980s. Hispanic rights groups maintained a consistent position in the years

04-S3453 1/11/06 3:06 PM Page 94

94

Post-Bracero Dilemmas

following IRCA’s passage: It was not enough to simply expand the numbers of legalized migrants without giving them the opportunity to become citizens. Within the ranks of organized labor, debate continued on employer sanctions and amnesty throughout the 1990s. Even the restrictionist United Brotherhood of Carpenters and Joiners eventually changed its position on employer sanctions as immigrant workers joined the unions in increasing numbers and argued for a labor strategy centered on widescale organizing. Union leaders recognized that employers had used the sanctions to justify firings of immigrant organizers. After 2000, when the AFL-CIO formally decided to end its support of sanctions, the Hispanic rights groups and the labor movement became allies in efforts to expand immigrants’ rights.

05-S3453 1/11/06 3:06 PM Page 95

chapter

Legal and Illegal Immigration Reform

5

after 1965, legislative battles in congress repeatedly tested the durability of the immigration system established by the Hart-Cellar Act. Three times during the 1980s and 1990s, reformers mounted largescale efforts to restrict immigration. Those efforts resulted in passage of the Immigration Reform and Control Act of 1986, the Immigration Act of 1990, and the Illegal Immigration Reform and Immigrant Responsibility Act of 1996. This chapter begins with an account of the legislative politics leading to passage of the IRCA and the 1990 act. In each legislative battle, there were many influential players in Congress—the sponsors of the legislation, congressional party leaders, party caucuses, congressional committees, and the Congressional Hispanic Caucus. Employing both “inside” and “outside” strategies, ethnic rights and economic interest groups lobbied to influence the decisions of these political actors. The policy preferences of the different groups led to the shifting coalitions described in Chapter 2 (see Table 2.3). The second part of the chapter reports a statistical analysis of House votes on a variety of immigration issues addressed by the 1986 and 1990 legislation. As the results demonstrate, both ethnic and economic factors at the district level influenced members’ votes, but the nature and extent of their influence varied considerably according to the particular issue at stake.

Immigration Reform and Control Act of 1986 Reform of American immigration law proceeded in two stages from the early 1980s to 1990. In its final report, the Select Commission on U.S.

05-S3453 1/11/06 3:06 PM Page 96

96

Legal and Illegal Immigration Reform

Immigration and Refugee Policy (1981), a bipartisan commission, had recommended that the nation close the “back door” to illegal immigration to keep the “front door” open.1 In the initial phases of the reform effort, proposals to revise legal and illegal immigration laws were joined in the same bill; after extensive debate, the issues were separated into two pieces of legislation. Once Congress dealt with the problem of illegal immigration (the back door), it acted to open the front door somewhat to accommodate new streams of legal immigrants. In the first stage, Congress produced the Immigration Reform and Control Act of 1986. The centerpiece of the law was the institution of sanctions against employers who hire illegal immigrants. The IRCA also explicitly made it illegal for employers to knowingly hire undocumented immigrants. A law passed in the 1950s, known as the Texas Proviso, had exempted employers from prosecution for hiring illegal aliens (see Chapter 4). By the time employer sanctions were finally enacted into law in 1986, the subject had been under congressional deliberation for a full fifteen years. Opposition to employer sanctions came primarily from legislators who represented two major categories of organized interests: Mexican American constituencies and employers in agricultural districts in the West, who openly admitted their reliance on illegal migrant labor from Mexico to harvest perishable crops. The testimony of Mexican American organizations on employer sanctions and other issues debated under the IRCA is documented in Chapter 4. Several other ethnic rights groups joined Mexican American lobbyists in arguing that sanctions would increase discrimination against foreign-looking persons in the workplace. Although the Organization of Chinese Americans proclaimed its support for stronger controls over illegal immigration, the group predicted that employers would either find sanctions a convenient excuse for discriminating against Chinese Americans or be overly cautious in hiring them. The New York chapter of the Caribbean Action Lobby also cited job discrimination as a reason it opposed sanctions, adding that sanctions would impose heavy financial burdens on small and marginal businesses as well as on the victims of discrimination. Similarly, a consortium of eighteen Puerto Rican health and human services agencies in New York pointed out that employers would not be able to distinguish Puerto Ricans—who are American citizens—from undocumented Hispanics.2 (No proof of citizenship is required of U.S. citizens.) Opponents of employer sanctions wanted sanction provisions deleted from the IRCA, but they failed repeatedly. As the controversy

05-S3453 1/11/06 3:06 PM Page 97

Legal and Illegal Immigration Reform

97

went on, members of the Congressional Hispanic Caucus were faced with deciding which terms of a political trade they would accept between an employer-sanctions provision and a proposed amnesty for illegal immigrants already living in the United States. Following the recommendations of the select commission, both the Carter and Reagan administrations had supported granting some sort of amnesty to prevent fostering a large “shadow” population of illegal aliens. However, there was disagreement in the early 1980s between the Reagan administration and congressional Democrats about how many years of continuous residence in the country an illegal immigrant would have to have to receive amnesty. Conservatives in both parties in Congress opposed various compromise versions, claiming they were too generous in their amnesty provisions; on the other side, members of the Hispanic caucus were reluctant to trade an unacceptable version of an employersanctions law for amnesty. In the end, though, both sides negotiated.3

Blocking Reform of Legal Immigration The Simpson-Mazzoli Bill of 1982, which was reintroduced in 1983, was the predecessor to the IRCA. It addressed issues pertaining to both legal immigration and illegal immigration. The most far-reaching changes in the legal-immigration system were passed in the Senate, where Republicans held a majority for most of the Reagan years, but the measures did not pass in the House. Among other changes, the Senate version of the bill, which was jointly sponsored by Senators Edward Kennedy (D-MA) and Alan Simpson (R-WY), called for raising the priority of economic goals. A new category of 55,000 “independent” immigrants would be admitted under a point system: Applicants would receive points for characteristics likely to make them assimilable into labor markets. For example, points would have been awarded for education, skills in short supply, age, working experience, and English-speaking ability. The Kennedy-Simpson proposal also would have eliminated the brothersand-sisters visa category, and it would have imposed a numerical cap on legal immigration each year. The Hart-Cellar Act had not set a cap on immigration, and none has been imposed in legislation to date.4 In devising their lobbying strategy to block restrictions on legal immigration and more stringent enforcement to prevent illegal immigration, proimmigration interest groups focused their efforts on the Democrat-controlled House. Their strategy entailed forging a broad coalition across party lines at the committee and subcommittee levels.

05-S3453 1/11/06 3:06 PM Page 98

98 Legal and Illegal Immigration Reform The actions of the House Speaker and Democrat-controlled Rules Committee would also prove decisive in the outcome. The primary strategy of proimmigration advocates was to block reform proposals in Congress.5 They believed that any delay would work in their favor: Because extended controversy over immigration tended to bring out and exacerbate divisions within both parties; party leaders shied away from it. In 1983, then, proimmigration forces convinced Speaker Tip O’Neill (D-MA) to pull the immigration bill from the House agenda. The interest groups claimed that President Reagan would veto Simpson-Mazzoli if it passed through Congress. The claim was never confirmed by the administration; but as national and congressional elections neared, O’Neill was worried about alienating Hispanic voters and western growers.6 The next year, O’Neill agreed to place the Mazzoli Bill back on the House agenda, but the House Democratic Caucus and its leadership had other, more positive means to shape the substance of the final bill. The most important was their power to influence which version of the bill would serve as the basis for floor debate through their majority on the House Rules Committee and the Speaker’s influence over the chair of that committee. This was the strategy they had used two years earlier, when the immigration reform bill first reached the House floor. By issuing an open rule, which allowed unlimited amendments, and scheduling debate in the last days of the lameduck session of the 97th Congress, the Rules Committee had effectively killed the bill. The second time the Mazzoli Bill came to the floor of the House, in 1984, it had been referred to the Judiciary, Agriculture, Education and Labor, and Energy and Commerce committees, each of which produced a version of the legislation. In addition to the amendments proposed by the four committees, the Hispanic caucus had worked with ethnic rights and church groups to have hundreds of amendments ready. Just before Rules met to make its decision on how to accommodate the many amendments, members of the Hispanic caucus successfully pressured the Democratic caucus for a meeting. It was attended by 140 members. Edward Roybal (D-CA), chair of the Hispanic caucus, wanted the Democrats to instruct Rules not to issue a rule for floor debate at all, but he backed down. The Hispanic caucus did press Rules to admit key amendments it proposed; and the committee approved sixty-nine amendments the next day. The Judiciary Committee version was made the basis of the upcoming floor debate, while key portions of the other committee bills would be introduced as amendments.

05-S3453 1/11/06 3:06 PM Page 99

Legal and Illegal Immigration Reform

99

Although the Hispanic caucus’s amendments received more attention in the press because there were so many of them, the amendments offered by the Education and Labor Committee in negotiations were also important. Organized labor, through its representation on the committee, held a strong position at the informal bargaining table with other key players, including members of the Hispanic caucus.7 The version of the bill reported out of the committee called for increased labor protections in a temporary-labor program (H-2) used by eastern growers. The committee version of the bill did not support the new guestworker program that western growers were hoping for because they found the H-2 regulations too stringent.8 With labor afforded significant representation through the Education and Labor Committee, three well-organized interests had to be accommodated in the final compromise—western growers, the AFL-CIO, and Hispanic interests. In this situation, ethnic rights and church groups and the Hispanic caucus, in particular, enjoyed some advantage because they could form strategic, and shifting, alliances on two dimensions of the bill. They could stand with employers in opposing employer sanctions and calling for a loosening of certain enforcement provisions; and they could also stand with labor in opposing any guest-worker program that created an underclass of foreign workers. Among Republicans, the most significant opposition to the original proposal to reduce legal immigration (the Simpson-Mazzoli Bill) came from the Reagan administration, not from Congress. Ronald Reagan had publicly voiced his views on the economic and cultural benefits of legal immigration; in his last speech as president, for example, he praised America for welcoming immigrants to a land of freedom, to a “shining city upon a hill.” The main Republican spokesperson for the administration’s position within the House Judiciary Committee was Hamilton Fish, a Republican from New York. Fish joined with liberal Democrats to support removal of all portions of the original Simpson-Mazzoli Bill pertaining to legal immigration, which the proimmigration interest groups then claimed as a victory.

Amnesty and the Rights of Temporary Workers Democrats held a large majority in the House of Representatives in 1984: 269 to 166. The floor votes on the Mazzoli Bill, which I analyze in detail later in this chapter, divided along ideological lines. According to Congressional Quarterly’s tally of annual votes, the so-called conservative

05-S3453 1/11/06 3:06 PM Page 100

100

Legal and Illegal Immigration Reform

coalition, an informal alliance of Republicans and Southern Democrats against Northern Democrats, won nine floor votes on immigration reform in 1984 (it won a total of just fifty-nine House votes that year).9 Two of the most controversial issues in the Mazzoli Bill were amnesty and a guest-worker program: The conservative coalition lost on the former but won on the latter (although the terms of the guest-worker program were subsequently revised). A number of votes divided liberals from conservatives because they dealt with the social-incorporation dimension of illegal immigration. The primary issue was whether to grant an amnesty to illegal immigrants already living in the United States. No politician was advocating the mass deportation of illegal immigrants who had been in the country for a number of years; at issue were the terms of the amnesty. The Republican Senate and the Reagan administration wanted to limit the scope of the amnesty; Democrats wanted to expand it. The proposed guest-worker program also generated liberal-conservative divisions on the issue of migrants’ rights. Traditional guest-worker programs offer migrant workers no protections: Once their work is done, they have to return to their home country. Liberals, who were responsive to the demands of organized labor, were willing to accept a guest-worker program but only if that program guaranteed foreign workers basic rights and the earned right to stay in the country and to become citizens. The final compromise on traditional guest workers was negotiated by three Democratic legislators: Senator Charles Schumer of New York, and Representatives Leon Panetta and Howard Berman, both of California. Panetta represented agricultural districts, while Berman represented labor’s interests. There were no hearings; instead the representatives of growers, labor, and Hispanics engaged in a round of political trading. When the IRCA finally passed in 1986, it established a probationary guest-worker program, granting earned-stay rights to special agricultural workers, and allowing SAWs to take jobs outside the agricultural sector if they could show they had worked in agriculture for at least ninety days from May 1985 to May 1986. The Congressional Budget Office estimated that by the time the program ended in 1988, 1.3 million workers had claimed SAW status (Reimers 1992, 248). In addition, the act provided for replenishment agricultural workers should there be a documented shortage of seasonal farmworkers. The RAW provision was a response to growers’ concern that they would need an additional source of labor should large numbers of SAWs move out of agriculture. The provision, however, was never used. By the standards

05-S3453 1/11/06 3:06 PM Page 101

Legal and Illegal Immigration Reform

101

of all countries that import agricultural workers, the rights measure enacted under the RAW program was unprecedented in its liberalism: Foreign migrant workers essentially became free agents in domestic labor markets after a minimal time spent in agriculture. In the final vote on the IRCA in the House, HR 3810 passed by a very narrow margin, 216 to 211. Within liberal and conservative coalitions, there were divisions at the extreme ends. Some of the most conservative members thought the amnesty provision too generous, and liberal Democrats, including the Hispanic caucus, opposed employer sanctions because, they argued, the measures would result in discriminatory treatment of all Hispanics.10

Immigration Act of 1990 The question of legal immigration was finally revisited in 1988. This began the second stage of immigration reform. By the time Congress returned to the issue of legal immigration, public concern over immigration had largely receded. The national economy had sustained an upturn since the mid-1980s. It was not until 1990 that the economy began to fall into a recession; but by that time, passage of the Immigration Act of 1990 was near. The basic terms of the debate were set by the final report of the Select Commission on U.S. Immigration and Refugee Policy (1981). While affirming the priority of family-unification goals, the report called for increasing the admissions of highly trained immigrants, “human capital” that could contribute to American industry. Various policy analysts were pointing to a skill gap in the workforce.11 The projections of a skilledlabor shortage prompted high-technology companies in the computer and electronics industries, including Microsoft Corporation and smaller companies in California’s Silicon Valley, to mount a concerted effort to keep access open to trained personnel from foreign countries. Although they supported an increase in visas for skilled and professional workers, most Republicans opposed the demands of ethnic groups to increase family visas. The Democrats were more responsive to specific ethnic constituencies. For example, liberal Democrats from the Northeast, particularly from New York and Massachusetts, supported the Irish American lobby’s demand for more Irish visas. And Asian and Hispanic organizations found sympathetic Democrats to support their demand that the preference for adult siblings of citizens be retained.12

05-S3453 1/11/06 3:06 PM Page 102

102

Legal and Illegal Immigration Reform

The Proimmigration Coalition Two factors helped proimmigration forces stave off restrictive reform: a proimmigration coalition of liberals and conservatives, and logrolling to meet the diverse demands of Democratic constituencies. Like the Reagan administration, the George H. Bush administration came to support a liberal position on legal immigration.13 Although Republicans were split on the issues, the administration’s stance influenced a block of congressional Republicans to vote in the end for the expansion of legal immigration called for by the Immigration Act of 1990. Two proimmigration coalitions formed. The first was a business coalition coordinated by the U.S. Chamber of Commerce. Its goal was to obtain legal changes that would make it easier to recruit temporary employees.14 The second—informally known as the “family coalition”— brought together the same ethnic rights and church groups that had cooperated before passage of the IRCA. According to Cecilia Muñoz, a lobbyist for the National Council of La Raza, in the negotiations that led up to passage of the 1990 act, the two coalitions met formally only once or twice; but they had a tacit agreement not to oppose each other’s demands.15 Furthermore, the American Immigration Lawyers Association, a member of both coalitions, informally coordinated the flow of information about legislative goals and strategies between the two groups.16 In 1988 and 1989, the Senate had passed two very similar immigration bills, but the House did not act until 1990.17 There were significant differences between the House and Senate bills. To meet the original goal of the policy reform—to increase the number of immigrants with special skills and training—the Kennedy-Simpson Bill in the Senate allowed for a new category of independent immigrants, people who would not be admitted for employment. The bills would have employed a point system somewhat like Canada’s, which gave credit to applicants for their level of education or for work and professional experience that might fill the nation’s human resource needs. In full committee, a provision to give points for proficiency in English was removed after Asian and Hispanic groups strongly objected that it would give preference to Europeans.18 Unlike the legislation passed by the Senate, the House version of the bill did not contain a point system. A coalition of Asian American legal organizations had opposed several other provisions in the 1989 Senate bill (S 358): a proposed cap

05-S3453 1/11/06 3:06 PM Page 103

Legal and Illegal Immigration Reform

103

on family-based immigration and the deletion of the second and fifth preference categories (unmarried children of citizens and residents, and adult siblings of citizens and residents).19 The coalition’s argument was that the net effect of these measures would be to scale back opportunities for Asian immigrants to reunite with families at a time when the impact of racist anti-Asian laws, which were finally changed in 1965, was still being felt.20 The Asian American legal groups carefully positioned themselves not to block Irish Americans’ efforts to add new immigrant categories. Speaking to the specific issue of awarding points to applicants who know English, the groups made clear that they did not oppose creating new categories to allow increased immigration, particularly for those who lack the family ties necessary for immigration; moreover, they pointed out that Irish American groups did not believe a cap on legal immigration was necessary or that there was a need for reduction in family preferences. The Senate set a lower cap on total immigration than the House did: 630,000 visas versus 800,000. However, in a victory for the ethnic groups, the cap could be pierced. In 1989, an amendment offered by Senators Orrin Hatch (R-UT) and Dennis DeConcini (D-AZ) set a floor on family visas (the number of family-based visas could not fall below 216,000); as long as that floor was maintained, the cap on total visas could be raised. The amendment passed by a vote of 62 to 36. The solution devised as a response to Irish American constituencies was a novel one. It affirmed the basic principle of family unification while creating a special compensatory mechanism for countries inherently disadvantaged by the family-based system. That is, it acknowledged that family-based immigration works against older immigrant groups, whose long-term patterns of immigration leave few immediate relatives in the sending country. Because the allocation of visas to family members of citizens or residents did not, on its face, imply any ethnic favoritism, the principle stood; but to it was added a form of affirmative action for disadvantaged countries. One of the longtime members of the Senate immigration subcommittee was Edward Kennedy, who represented a large Irish constituency, not a small factor in the outcome. In its final form, the diversity-visa program allocated 55,000 visas a year (starting in 1995) to countries that received small numbers of visas under the existing system. Also, Northern Ireland was counted as a separate foreign state from the United Kingdom for the purpose of allocating country quotas.

05-S3453 1/11/06 3:06 PM Page 104

104 Legal and Illegal Immigration Reform Like the coalition of Asian American legal organizations, the family coalition responded to the Irish Americans’ demands strategically. Realizing the possibility of an increase for their specific ethnic interests if they supported one another’s demands, the Hispanic and Asian American groups in the coalition committed themselves to bartering for different visa types. The family coalition would support more visas for Irish immigrants as long as the overall package preserved the brothersand-sisters category.21 In a victory for the family coalition, the final House version of the bill increased by 300,000 the number of visas for nonimmediate relatives, including the adult siblings of citizens.22 A new nonimmigrant category was created for artists and entertainers, capped at 25,000 visas a year. And another was created for workers in religious organizations.

Accommodating Labor The House bill also included tighter protections for native workers when employers recruited foreigners to fill labor shortages. The AFLCIO for the first time had pressed for numerical restrictions on foreign workers in “specialty occupations,” temporary professional workers recruited by employers under the H-1B program. Among those workers are computer programmers, certain teachers, and other highly skilled or professional workers. In the House version of the 1990 bill, organized labor succeeded in winning a 65,000-visa limit on admissions of H-1B temporary workers. However, in some respects H-1B regulations were loosened. The requirement of a residence abroad was removed; furthermore, seeking permanent residence in the U.S. would not disqualify the H-1B applicant. Because many H-1B visa holders seek permanent residency, this made it easier for employers to recruit them. In addition, the AFL-CIO worked closely with the new chair of the House immigration subcommittee, Bruce Morrison (D-CT), to devise a set of regulatory proposals. These included a head tax on employers who hire foreign workers; tighter rules governing employers’ documentation that they have tried to recruit American workers, and a thirty-day waiting period; and a system of linking labor immigration to state and federal certification of labor shortages. These proposals naturally generated strong opposition from business lobbies. American employers had never paid a head tax on foreign workers.23 Nor had the federal government ever tied the total level of immigrant admissions to periodic measures of labor shortages, to encourage mass immigration,

05-S3453 1/11/06 3:06 PM Page 105

Legal and Illegal Immigration Reform

105

as Canada and other countries do. Although a modified head tax provision passed out of Judiciary, the House Ways and Means Committee later forced its deletion from the bill when it was under consideration by Rules.24 The argument was that the head tax was a revenue measure, which fell under the jurisdiction of Ways and Means. When Morrison attempted to raise the issue of new regulations on hiring foreign workers in the form of a floor amendment, it was voted down 194 to 229.25 The cap on the admission of temporary foreign workers under the H-1B program was retained in the final bill as a concession to organized labor.26 In 1995, the next time that regulations for hiring temporary foreign workers were revisited, Republicans held majorities in both houses of Congress, and Democrats and their labor constituencies would be cut out of the final conference negotiations. Again, regulating the hiring and employment of temporary foreign workers remains a highly partisan issue.27 The members of the conference committee split the difference between the limit on total immigration approved by the Senate (630,000 visas) and that approved by the House (800,000 visas), settling on 700,000 visas a year for three years and then a reduction to 675,000. In the end, the Immigration Act of 1990 increased legal immigration levels by more than 30 percent, and it did that with bipartisan support. Considering the goals that Senators Kennedy and Simpson had set out to accomplish in their bill, the final version of the 1990 act was unexpectedly expansive. Legal-admissions quotas were increased substantially, not limited. Moreover, quotas were increased by more than categories of skilled and professional workers; the law also anticipated increases in family-based immigration and the new diversity program.

Interest-Group Politics In congressional battles over the IRCA and the Immigration Act of 1990, various interest groups cooperated in coalitions that shifted in membership according to predictable patterns described in Chapter 2. In the debate leading up to passage of the IRCA in 1986, Hispanic and Asian American groups found common ground with farm lobbies in opposing employer sanctions; and their joint lobbying helped stall the passage of an employer-sanctions bill for many years. In contrast, on the issue of guest workers, Hispanic and Asian American groups cooperated

05-S3453 1/11/06 3:06 PM Page 106

106 Legal and Illegal Immigration Reform with organized labor. Subsequently, in the contest over the Immigration Act of 1990, the Hispanic and Asian American groups cooperated in tacit alliances with lobbyists for employers in high-technology industries, supporting liberal policies on both family-based and employment-based admissions. Ethnic rights groups formed loose alliances with employer groups on occasion; however, there were long-established working relationships between ethnic groups and their liberal church allies in the family coalition, which formed to defend family-unification visas in the battle over the Immigration Act of 1990. The political science literature on interest groups sheds light on the nature and timing of issue-specific coalitions. Through economies of scale, member groups in coalitions save resources and so are able to increase the number of legislators they can contact (Hojnacki and Kimball 1998). According to Hojnacki (1997, 61), groups must weigh the benefits of increasing their chance for success by working together against the cost of submerging their distinct identity in a coalition. The benefits of cooperation are greater with the involvement of one or more organizations that are pivotal to success. Furthermore, individual members of coalitions tend to put more work into coalitions when one or more groups coordinate the joint effort (Hojnacki 1988, 454). In the family coalition, the NCLR and other Hispanic groups played a leadership role. Coalition building is part of an inside strategy interest groups employ to influence legislation. In general, inside strategies consist of lobbying directed toward policymakers and opinion leaders in Washington, D.C. For example, inside strategies can include efforts to sway legislators’ votes by providing them with research on the likely effects of legislation, drafting legislation, even meeting with them or their staff to devise ways to approach undecided legislators. In these respects, lobbyists operate as “unpaid staff,” supplementing lawmakers paid staff members, who are often in short supply (Birnbaum 1993, 6). In contrast, outside strategies include attempts to apply public pressure on legislators by mobilizing constituents to visit, call, or send letters; to hold press conferences; or to write opinion pieces or letters to the editor (Gais and Walker 1991). Hojnacki and Kimball (1998) also found that interest groups are more likely to approach a member of Congress when they have a strong base of support in the member’s district (784). The success of such lobbying depends on the targeted legislator’s perception of the strength of the

05-S3453 1/11/06 3:06 PM Page 107

Legal and Illegal Immigration Reform

107

support.28 In my cross-sectional analyses of legislators’ voting behavior on immigration issues, I assume that lawmakers use the relative size of an ethnic population in his or her district to estimate the strength or at least the potential strength of a bloc of ethnic voters. Economic interest groups— employers’ lobbies and organized labor— contribute to political action committees that allocate campaign contributions to candidates for political office. The level of a group’s PAC contributions to individual candidates’ campaign funds should reflect the group’s interest in an issue. Unemployment rates are an indicator of workers’ economic interests in restricting immigration. In the next section are the findings from a study of how these and other variables influence legislators’ votes on immigration.

A Model of Voting: The Immigration Reform and Control Act of 1986 What characteristics of a congressional district influence a representative’s votes on immigration issues? This section investigates a model of roll-call voting in the House of Representatives on issues addressed in the Immigration Reform and Control Act of 1986 (99th Congress) and in its predecessor, the Mazzoli Bill, which was voted on in 1984 (98th Congress). We are interested in measuring the extent to which ethnic and economic factors within congressional districts predict individual representatives’ votes, controlling for the members’ party affiliation. We expect that House members are increasingly responsive to proimmigration lobbyists as the ethnic population of their district grows. However, it is not clear whether the shape of this function increases at all levels of ethnic concentration or is curvilinear if we consider research on the relationship between districting and minority representation. In their study of race and representation, Cameron, Epstein, and O’Halloran (1996) discuss several hypotheses on the relationship between the size of the black population in a district and the representation of black interests. The starting point of their investigation is V. O. Key’s (1949) race-polarization hypothesis. Key found that in 1928, elected representatives of counties with relatively large black populations had more conservative voting records than did others. Key’s thinking was that representatives in these districts felt pressured to show white conservative constituents that they did not cater to black voters. His hypothesis, then, was that black representation in a district decreases as the size of

05-S3453 1/11/06 3:06 PM Page 108

108

Legal and Illegal Immigration Reform

the black population in the district increases. In a variation on Key’s reasoning, Keech (1968) posits that a U-shaped relationship may exist between the size of the black population in a district and the liberalness of its representative’s voting record. Only if the black population is large enough to constitute a threat to nonblacks (e.g., 20 percent) does polarization begin and conservative voting increase, at which point black representation decreases. Alternatively, there may be an increasing relationship between black population size and black representation, or there may be a threshold point at which black representation increases after decreasing (because of race polarization) or staying flat (Cameron, Epstein, and O’Halloran). The relationship between the immigrant population in a district and the representative’s votes on immigration issues may follow a similar logic. For example, polarization may develop between native and foreign-born populations. After foreign-born populations grow to a certain size, however, an elected representative may feel that reelection depends on responsiveness to the foreign-born population’s interests. Thus, I investigate whether there is a curvilinear relationship between the size of the foreign-born population in a district and the representation of foreign-born persons’ interests by testing the fit of a quadratic function. In addition, I investigate the effect of specific ethnic populations, such as Mexican American, Cuban American, and Japanese American, because their interests are likely to differ. Economic interests are also tested as predictors of a representative’s vote. In a district, we expect that a concentration of employers’ interests in importing immigrant workers will increase the probability of the representative’s voting proimmigration. As unemployment levels increase, the probability of the representative’s casting a restrictionist vote should increase. More details on both variables and expectations are described below. Gimpel and Edwards (1999, 205 –208) provide a similar analysis of votes on the IRCA and the Immigration Act of 1990, as well as other bills voted on from 1965 to 1996. My results generally support their findings, but there are some differences between the two studies in the formulation of vote models. Gimpel and Edwards test the effect of a representative’s membership on the Judiciary Committee (which had jurisdiction over immigration matters) as well as the representative’s ethnicity. In general, they find that members of racial or ethnic minorities tend to vote in a liberal direction. The present study does not include these characteristics of members as independent variables.

05-S3453 1/11/06 3:06 PM Page 109

Legal and Illegal Immigration Reform

109

I include a number of ethnic-specific variables as predictors, rather than the more encompassing panethnic or racial categories “Hispanic” and “Asian,” which Gimpel and Edwards use. Instead of using the Hispanic category as an independent variable, I test for the effect of the percentage of people of Mexican or Cuban ancestry residing in the district. Instead of using the Asian/other category, I test for the effect of the percentage of people of Chinese or Japanese ancestry. I also test the influence of population concentrations of two older ethnic-immigrant backgrounds, Irish and Italian. And I test for the influence of nativity, or the percentage of foreign-born persons in a district. For the ethnic and foreign-born variables, I test the fit of a U-shaped curvilinear model, as described above. In addition, in several cases I found that the estimate of the percentage unemployed in a state does not influence votes in a restrictionist direction, as expected; consequently, I examine whether the change in unemployment over the past year is a factor. For the vote on the 1990 bill, I am interested in whether PAC contributions from employers’ groups and labor unions are a factor in members’ votes. Gimpel and Edwards (1999) test for the effect of concentrations of persons employed in agriculture on all amendments and final passage of both the IRCA and the 1990 bill, but some votes are not directly related to agricultural interests.

Votes Analyzed IRCA’s immediate predecessor was the Simpson-Mazzoli Bill (the Senate version was S 529, or the Simpson Bill; the House version was HR 1510, or the Mazzoli Bill). The debate on HR 1510 saw votes on no fewer than fifty-seven floor amendments. I analyze votes on several amendments representing issues that stood in the way of the bill’s passing: the McCollum Amendment, which proposed to regularize the status (through amnesty) of undocumented immigrants; the Roybal Amendment, which proposed imposing a fine or other sanctions on employers found to be employing illegal immigrants; the Panetta Amendment, which proposed a guest-worker program that would allow growers of perishable crops to obtain foreign workers on seventy-two hours’ notice; and the Moorhead Amendment, which would have restricted the number of legal immigrants (Table 5.1).29 In the 99th Congress, the House considered HR 3810, which in its final form would be known as the Immigration Reform and Control Act

05-S3453 1/11/06 3:06 PM Page 110

110

Legal and Illegal Immigration Reform ta b l e 5 . 1 Key Amendments to the Mazzoli Bill, 1984

Amendment

Description

McCollum Amendment

Proposed to strike the sections of the bill that provide for legalization (amnesty) and for legalization assistance. The amendment failed in the Committee of the Whole by a recorded vote of 195 to 233. (CQ vote 896) Proposed to delete sections of the bill that provide for employer sanctions and to substitute language providing for stricter enforcement of labor laws. The amendment failed in the Committee of the Whole by a recorded vote of 120 to 304. (CQ vote 857) Proposed (1) to permit growers of perishable crops to apply to the attorney general for guest workers and to expect action within 72 hours of the request if unusual circumstances hold; (2) to limit guest workers to work within a region established by the attorney general; (3) to require that employers actively recruit U.S. workers and that they provide housing and compensation for temporary foreign workers as set forth in the bill; and (4) to prohibit the use of guest workers as strikebreakers. The amendment was adopted in the Committee of the Whole by a recorded vote of 228 to 172. (CQ vote 882) Proposed to add a new section to the bill limiting nonrefugee immigration admissions to 450,000 per fiscal year. The amendment failed in the Committee of the Whole by a recorded vote of 168 to 231. (CQ vote 873)

Roybal Amendment

Panetta Amendment

Moorhead Amendment

of 1986. There were votes on twelve amendments to HR 3810; I analyze votes on the two controversial amendments described in Table 5.2. One vote was on the McCollum Amendment, which proposed to delete the amnesty provision; the other, on the de la Garza Amendment, which limited the ability of INS officials searching for undocumented aliens to enter a farm without a search warrant or the owner’s permission.

The Model of Individual Floor Votes The analysis that follows evaluates the direction and strength of party and constituency variables on an individual member’s vote. A hypothetical House member’s vote is modeled with a logistic regression

05-S3453 1/11/06 3:06 PM Page 111

Legal and Illegal Immigration Reform

111

ta b l e 5 . 2 Key Amendments to the Immigration Reform and Control Act of 1986 Amendment

Description

McCollum Amendment

Proposed to strike an amnesty provision in the bill that would have allowed aliens who entered this country illegally before 1982, and who are otherwise eligible for admission, to apply for temporary-resident status. The amendment failed in the Committee of the Whole by a recorded vote of 192 to 199. (CQ vote 1291) Proposed (1) to prohibit INS officials from entering a farm or agricultural operation without a search warrant or permission from the owner for the purpose of questioning a person believed to be an illegal alien, and (2) to permit INS officials to enter open fields without a warrant within 25 miles of U.S. borders. The amendment was adopted in the Committee of the Whole by a recorded vote of 221 to 170. (CQ vote 1289)

de la Garza Amendment

equation in which the dependent variable is dichotomous (a yes or no vote). The logistic regression allows us to calculate the size of effects of statistically significant variables on the probability of a liberal (or restrictive) vote. The study focuses on the House of Representatives, not the Senate, because of the larger number of districts represented by the House. The variation in demographic and economic characteristics over 435 districts (versus 50 states) allows testing of various hypotheses with more precision.

Independent Variables and Expectations In Tables 5.3 to 5.8, the FBORN variable represents the percentage of a district’s population that is foreign-born; the ethnic variables (e.g., MEXICAN, CHINESE, IRISH) represent persons with a specific ethnic ancestry. A log transformation of FBORN is taken to correct for a skewed distribution of the data. AGEMP is the share of total employment in agriculture. Agricultural employers organized the most resources to lobby against employer sanctions and for a guaranteed supply of migrant farm labor. I expected that as the percentage of the district population employed in agriculture

05-S3453 1/11/06 3:06 PM Page 112

112

Legal and Illegal Immigration Reform

increases, so would the likelihood of a representative’s casting a vote in favor of starting a temporary-farmworker program. The limitations of using aggregate measures of district characteristics—FBORN and AGEMP, for example—to explain constituency influences on a legislator’s votes should be noted here. These data give summary statistics for a geographic constituency; however, it is known that a member’s geographic constituency is not identical to his or her reelection constituency. Furthermore, using aggregate data can mask important differences between districts that are only observable at the level of individuals. If there are no better alternatives, however, it is reasonable to make the best possible estimate of a member’s preference function by using available measures of variables that, on theoretical grounds, can be expected to predict a member’s vote. Although the objections noted here are likely to cause some bias in the estimates, that possibility should be taken into account in interpreting coefficients. In addition, the reader should note limitations in using district population data for small ethnic groups dispersed in only a handful of states. The small numbers make it difficult to draw firm conclusions about a particular ethnic group; this limitation would apply to ethnic-specific groups of Asian ancestry, whose sample sizes are generally very small in the districts. PARTY is a dummy variable representing the party affiliation of the House member; it is coded 1 for Republican, 0 for Democrat. According to our expectations, a Democratic member of Congress is more likely than a Republican member to cast a proimmigration vote on legaladmissions and social-incorporation policies, and to oppose any guestworker program that would undermine labor and the civil rights of immigrants and minorities. Two variables measure constituencies that bear certain diffuse costs of immigration. The first is UNEMPCH, the unemployment level in the state. UNEMPCH84 – 83 is state-level unemployment in 1984 minus the level in 1983; thus, it is the recent increase in unemployment in the state.30 Similarly, UNEMPCH86 – 85 represents the change in state unemployment rates between 1985 and 1986. We generally expect a positive coefficient (sign is positive)—that is, a larger increase in unemployment over time should increase the probability of a restrictionist vote. However, unemployment is an economic variable related to workers’ interests, which are represented by an internally divided AFL-CIO. The second measure here is BRDSW, a dummy variable coded 1 for border or southwestern states (and 0 otherwise) that have large concentrations of immigrants; hence, taxpayers in these states bear the costs of

05-S3453 1/11/06 3:06 PM Page 113

Legal and Illegal Immigration Reform

113

incorporating undocumented immigrants into their public service system. The border and southwestern states include Arizona, California, Florida, New Mexico, and Texas. I’ve also included Florida in this group. Although Florida is neither a border state nor a southwestern state, taxpayer costs enter immigration politics in the state: Its geographic location has made it a destination for immigrants from Cuba and Central America. A dummy variable, SOUTH, is included in Model 1 for states in the Deep South. These states are coded 1, except Florida, which is counted as a BRDSW state. The purpose of the SOUTH variable is to evaluate whether the traditional conservatism of southern representatives is a significant influence on votes. In addition, controlling for southern states may alter the estimates of the effect of concentrations of blacks in a district (BLACK), a variable included in all three models. In the analysis of the various votes, Models 1 and 2 test the influence of certain specific concentrations of ethnic populations within a district. The principal ethnic advocacy groups that lobbied for amnesty and against employer sanctions were Mexican American and Asian American. Finally, Model 3 replaces the ethnic-specific variables (MEXICAN, CUBAN, CHINESE, JAPANESE, IRISH, and ITALIAN) with a single variable for the percentage of foreign-born persons (FBORN) in a district. (Again, I perform a log transformation on FBORN to correct for a skewed distribution in the data.) Goldin’s (1994) study of roll-call votes in an earlier period demonstrates the influence of foreign-born rather than ethnic-specific populations on congressional voting.31

Results of the Analysis The findings are reported in Tables 5.3 to 5.8. Notice the headnote in each table that describes the dependent variable and shows the total number of votes for and against, and the total vote broken down by party and by liberal (northern) and conservative (southern) segments of the Democratic Party. The votes by party suggest that Republicans are more likely than Democrats to vote for restrictive policies. For example, a Republican member of the House is likely to support employer sanctions and to oppose amnesty. The reader should note that in coding the votes legislators cast (i.e., the dependent variable), in most cases a liberal vote is coded 0 and a restrictionist vote is coded 1, regardless of whether the vote was a yes or no on the specific amendment. For some issues—

05-S3453 1/11/06 3:06 PM Page 114

114

Legal and Illegal Immigration Reform

ta b l e 5 . 3 House Vote on the McCollum Amendment to Delete Amnesty, Mazzoli Bill, 1984 Dependent variable: Delete amnesty: yes  1 195 –233 Republicans: 114 –50, Democrats: 81–183 Northern Democrats: 26 –148, Southern Democrats: 55 –35 Independent variable CONSTANT ln(FBORN) AGEMP PARTY UNEMPCH84 – 83 BRDSW states SOUTH states BLACK MEXICAN CUBAN CHINESE JAPANESE IRISH ITALIAN N 2 log likelihood Correctly predicted

Model 1

Model 2

.9031 (.9724) —

.4623 (.8205) —

.00565 (.0330) 1.7488*** (.2571) .1435 (.1968) 1.7283*** (.4419) 1.9977*** (.4504) .6848 (.9654) .0624* (.0271) .1985 (.1691) .8760 (.7896) 1.4750 (.7944) .0626 (.0456) .1493* (.0712)

.00397 (.0317) 1.6354*** (.2488) .2479 (.1906) 1.5724*** (.4474) —

428 430.826 74.5

Model 3 1.1063 (.5808) 1.2612*** (.1832) .0414 (.0317) 1.7695*** (.2469) .2531 (.1919) 1.2828*** (.3351) —

.2549 (.5506) .0763* (.0277) .2508 (.1839) 1.3480 (.8121) 1.2251 (.7721) .0201 (.0402) .2091* (.0693)

.1030 (.5547) —

428 454.889 69.6

428 460.155 71.6

— — — — —

* p  .05; ** p  .01; *** p  .001. n o t e s : Standard errors are in parentheses. Paired yes votes are counted as yes; paired no votes, as no. Absentee votes and abstentions are omitted from the regression. Cases with missing values are omitted. ln(FBORN) is ln(percentage of the district’s total population that is foreign-born). BLACK is the percentage of the district’s population that is black. MEXICAN, CUBAN, CHINESE, JAPANESE, IRISH, and ITALIAN indicate the percentage of the district’s population of these specific ethnic ancestries. (s o u r c e s : Census tape stf3d, 1980 census; and U.S. Census Bureau, Congressional Dis-

(continued)

05-S3453 1/11/06 3:06 PM Page 115

Legal and Illegal Immigration Reform

115

guest-worker programs and certain enforcement policies—however, classification of a position as liberal or restrictionist is not entirely clear. As shown in Table 5.5, for example, I use the 0 code (the liberal indicator) for a vote in favor of Representative Panetta’s guest-worker program, and the 1 code for a vote opposing it. (Liberals generally oppose traditional guest-worker programs like the one proposed by Panetta.) In general, in modeling the effect of foreign-born populations, the quadratic function did not exhibit a good fit in tests on any of the votes. Thus, there is no evidence that the function specifying the relationship between foreign-born population size and a member’s vote is U-shaped. In Model 3, as expected, FBORN is a strong predictor of a liberal vote on amnesty (Tables 5.3 and 5.7), employer sanctions (Table 5.4), capping legal immigration (Table 5.6), and requiring warrants for searches (Table 5.8). The MEXICAN variable has a similar effect on the same issues; the only exception is that it is not statistically significant, although the effect is in the expected direction in the vote on warrants for searches (Table 5.8, Models 1 and 2). AGEMP is statistically significant in all models in just two cases: the votes on the Panetta Amendment, to start a traditional guest-worker program (Table 5.5), and on the amendment that would require warrants for searches (Table 5.8). This tends to support the proposition that the concentration of industry in a district affects lawmakers’ votes: That is, lawmakers may find it difficult to vote against employers when the employers’ interests are directly affected. In general, legislators representing border or southwestern states

NOTES TO TABLE 5.3 (continued ) trict Data Books, 98th Congress, for districts in ten states for which boundaries changed between the 98th and 99th Congresses.) AGEMP is the percentage of the district’s total population employed in agriculture. (s o u r c e s : Census tape stf3d, 1980 census; and Congressional District Data Books, 98th Congress.) For PARTY, Republican  1, Democrat  0. UNEMPCH84 – 83 is the unemployment rate in the state in 1984 minus the comparable rate in 1983; UNEMPCH86 – 85 has an equivalent meaning. Because a point estimate was only available by congressional district in 1979 and 1989 (in summary tapes for the decennial census), I tried estimating the unemployment rate in other years (e.g., 1983) by multiplying the total number of unemployed people in the state in that year by the ratio (census count unemployed in district in 1979)/(census count unemployed in state in 1979). The results were not appreciably different from those achieved by using the state-level data reported here. (s o u r c e : The state-level estimates at middecade are taken from Bureau of Labor Statistics, Geographical Profiles of Employment and Unemployment, various years.) BRDSW states include Arizona, California, Florida, New Mexico, and Texas. SOUTH states include Alabama, Arkansas, Georgia, Louisiana, Mississippi, North Carolina, and South Carolina (states in the Deep South except Florida).

05-S3453 1/11/06 3:06 PM Page 116

116

Legal and Illegal Immigration Reform ta b l e 5 . 4 House Vote on the Roybal Amendment to Delete Employer Sanctions, Mazzoli Bill, 1984 Dependent variable: Delete employer sanctions: yes  o 120 –304 Republicans: 20 –140, Democrats: 100 –164 Northern Democrats: 86 – 89, Southern Democrats: 14 –75 Independent variable CONSTANT ln(FBORN) AGEMP PARTY UNEMPCH84 – 83 BRDSW states SOUTH states BLACK MEXICAN CUBAN CHINESE JAPANESE IRISH ITALIAN N –2 log likelihood Correctly predicted

See notes to Table 5.3.

Model 1

Model 2

Model 3

1.5001 (1.0748) —

2.0644* (.9748) —

.0233 (.0383) 1.2501*** (.3013) .1013 (.2196) .2091 (.4222) 2.6445*** (.7857) 2.9121* (1.2155) .l3l4*** (.0325) .1620 (.1304) .1758 (.1555) .00214 (.0722) –.0472 (.0504) .0517 (.0749)

.0360 (.0384) 1.2295*** (.3001) .0516 (.2188) .0835 (.4328) —

1.9362** (.6263) –.5896*** (.1672) –.0375 (.0352) 1.2789*** (.286) .0186 (.2098) .7436* (.3095) —

1.9928* (.8683) 0.1455*** (.0334) .1036 (.1091) .2505 (.1668) .00153 (.0750) 0.0588 (.0478) .0156 (.0714)

.7698 (.4633) —

424 375.449 78.5

424 395.035 78.3

424 424.033 75.2

— — — — —

05-S3453 1/11/06 3:06 PM Page 117

Legal and Illegal Immigration Reform ta b l e 5 . 5 House Vote on the Panetta Guest-Worker Amendment, Mazzoli Bill, 1984 Dependent variable: Start guest-worker program: yes  0 228 –172 Republicans: 138 –15, Democrats: 90 –157 Northern Democrats: 36 –124, Southern Democrats: 54 –33 Independent variable CONSTANT ln(FBORN) AGEMP PARTY UNEMPCH84 – 83 BRDSW states SOUTH states BLACK MEXICAN CUBAN CHINESE JAPANESE IRISH ITALIAN N –2 log likelihood Correctly predicted See notes to Table 5.3.

Model 1

Model 2

Model 3

1.3467 (1.1617) —

8518 (.9529) —

.1752*** (.0453) 3.3766*** (.3790) .5750* (.2451) 1.4893** (.4582) 2.6580*** (.5688) .1779 (.9671) .0496* (.0236) .3729 (.4071) .3246 (.2355) .1391 (.0873) .0956 (.0577) .1988* (.0860)

.1705** (.0450) 3.2277*** (.3735) .6789** (.2384) 1.1828** (.4571) —

1.747* (.6950) 1.1677*** (.2083) 1174** (.0435) 3.2456*** (.3537) .8495*** (.2443) 1.4162*** (.4006) —

.1347 (.5926) .0690** (.0243) .5886 (.4712) .4525 (.2403) .1178 (.0914) .00917* (.0486) .2800*** (.0830)

–.2574 (.5725) —

408 305.057 83.3

408 333.405 74.9

408 334.239 81.1

— — — — —

117

05-S3453 1/11/06 3:06 PM Page 118

118

Legal and Illegal Immigration Reform ta b l e 5 . 6 House Vote on the Moorhead Amendment to Cap Legal Immigration, Mazzoli Bill, 1984 Dependent variable: Cap legal immigration at 450,000: yes  1 168 –231 Republicans: 111– 46, Democrats: 57–185 Northern Democrats: 17–140, Southern Democrats: 40 – 45 Independent variable CONSTANT ln(FBORN) AGEMP PARTY UNEMPCH84 – 83 BRDSW states SOUTH states BLACK MEXICAN CUBAN CHINESE JAPANESE IRISH ITALIAN N –2 log likelihood Correctly predicted

See notes to Table 5.3.

Model 1

Model 2

.2965 (1.0944) —

.3847 (.9822) —

.0585 (.0372) 2.4619*** (.3002) .2026 (.2177) 2.5710*** (.4948) .9547 (.4335) .8433 (1.0294) .0632** (.0235) .5366 (.4738) .4701 (.3849) .2882 (.4651) .0313 (.0515) .2392** (.0839)

.0541 (.0367) 2.3930*** (.2922) .2558 (.2152) 2.4634*** (.4935) —

399 366.186 77.4

Model 3 .2499 (.6396) 1.0767*** (.2011) .0365 (.0364) 2.4835*** (.2886) .2735 (.2137) 2.4162*** (.4049) —

.5367 (.7166) .0693** (.0238) .6170 (.4923) .5378 (.3948) .2960 (.4531) .0517 (.0477) .2721** (.0829)

.3068 (.6238) —

399 371.304 78.2

399 380.729 76.2

— — — — —

05-S3453 1/11/06 3:06 PM Page 119

Legal and Illegal Immigration Reform

119

are more restrictionist than are legislators from other states; and the estimates are statistically significant in all tests. The effect of an increase in unemployment in a state over the past year (UNEMPCH) is statistically significant in two cases. On the Panetta Amendment, an increase in the rate of unemployment over the year preceding the vote (UNEMPCH84 – 83) tends to increase a legislator’s proclivity to vote for the guest-worker program (Table 5.5). And the unemployment increase between 1985 and 1986 (UNEMPCH86 – 85) has a restrictionist effect on the vote to delete the amnesty provision in the IRCA (Table 5.7).32 Ideological factors related to regional characteristics are captured indirectly in the control variable for states in the Deep South; the SOUTH variable is statistically significant on the votes to delete the amnesty provision from the Mazzoli Bill (Table 5.3) and to implement Panetta’s guest-worker proposal (Table 5.5). This finding indicates a conservative tendency in lawmakers’ voting on these issues in southern states. On the issue of employer sanctions, the statistically significant estimate on BLACK and MEXICAN (in Models 1 and 2) indicates the basis for a coalition that formed between representatives of largely Mexican American and black districts (Table 5.4, Models 2 and 3). Black civil rights organizations initially were ambivalent on the enforcement issues debated in the IRCA: Many were concerned that illegal immigrants were displacing black workers in unskilled labor markets. However, the Congressional Black Caucus decided to join the Congressional Hispanic Caucus in opposing the sanctions on the grounds that sanctions would open the door for violations of the civil rights of all Hispanics, not just illegal immigrants. Members of the Congressional Black Caucus generally represent districts with significant black populations, and often hold safe seats in congressional elections. This allows them to regularly cast liberal votes that do not necessarily mirror the views of the majority of their constituents (Tate 2003). Apparently, in the votes of representatives of predominantly black and Hispanic constituents, the liberal alliance of blacks and Hispanics is a factor that overrides the economic competition between blacks and Hispanics in unskilled labor markets. On the votes concerning issues other than sanctions, the variable BLACK is not statistically significant. In most cases, the estimates of specific ethnic constituency variables, such as MEXICAN or CHINESE, are not statistically significant because they represent relatively small minorities. There are some notable

05-S3453 1/11/06 3:06 PM Page 120

120

Legal and Illegal Immigration Reform ta b l e 5 . 7 House Vote on the McCollum Amendment to Delete Amnesty, IRCA, 1986 Dependent variable: Delete amnesty: yes  1 192 –199 Republicans: 124 – 40, Democrats: 68 –159 Northern Democrats: 21–131, Southern Democrats: 47–28 Independent variable CONSTANT ln(FBORN) AGEMP PARTY UNEMPCH86 – 85 BRDSW states SOUTH states BLACK MEXICAN CUBAN CHINESE JAPANESE IRISH ITALIAN N –2 log likelihood Correctly predicted

See notes to Table 5.3.

Model 1

Model 2

Model 3

.5319 (.8697) —

.3727 (.7326) —

.0231 (.0390) 2.3461*** (.2952) .4480* (.1910) 1.5939* (.5150) 1.6842** (.5188) .9443 (1.088) .0678* (.0270) .3026 (.2833) .8548 (.8656) 1.9766* (.8762) .0432 (.0505) .1727* (.0789)

.0321 (.0378) 2.2795*** (.2913) 0.6211*** (.1818) 1.3821** (.5235) —

1.0348** (.3693) 1.4851*** (.2114) .0872* (.0383) 2.4975*** (.2950) .5432** (.1713) 1.0758** (.3950) —

395 357.607 78.5

.4824 (.6577) .0831** (.0275) .3573 (.3158) 1.2064 (.8942) 1.6765 (.8709) .00999 (.0458) .2073** (.0772)

.0895 (.6815) —

395 369.887 76.7

395 367.784 77.2

— — — — —

05-S3453 1/11/06 3:06 PM Page 121

Legal and Illegal Immigration Reform

121

exceptions. On three votes, statistically significant estimates indicate that representatives of largely Mexican American districts opposed employer sanctions (Table 5.4), Panetta’s guest-worker program (Table 5.5), and a numerical cap on legal immigration (Table 5.6). On two issues related specifically to agricultural interests (traditional guest workers and search warrants), the variable JAPANESE is statistically significant, probably because Japanese Americans live in relatively large concentrations in the farming districts of the Central Valley in California. The Nissei Farmers’ League in that region was one of the most active groups lobbying for a guest-worker program in the 1980s. Representatives from the largely Japanese American districts also favored granting amnesty to illegal immigrants (Tables 5.3 and 5.7). There is no clear evidence that representatives from largely Irish American districts favored deleting the employer sanctions provision of the Mazzoli Bill (Table 5.4, Model 2). Irish American lobbies, in fact, admitted that many illegal immigrants in the Northeast come from Ireland. Of particular interest is the potential for shifting coalitions to form between the elected representatives of agricultural employers on the one hand and representatives of Mexican American constituencies on the other. Growers favored the traditional guest-worker program; ethnic nonprofits and labor unions opposed it on the grounds that farmworkers would not enjoy civil and labor rights. Representatives from districts with greater employer interests in migrant farm labor are more likely to vote for a traditional guest-worker program, as indicated by the negative sign on the AGEMP coefficient. Those with large Mexican American populations are more likely to vote no (notice the positive number for the MEXICAN coefficient). When the FBORN variable replaces the ethnic-specific variables, there is evidence that a coalition between representatives of districts with concentrated agricultural interests and representatives of districts with large foreign-born populations could form: On the issue of whether warrants should be required for INS searches in open fields, FBORN and AGEMP constituencies influence legislators’ votes in the same direction. A negative sign on the coefficients for FBORN and AGEMP, which are statistically significant in Model 3 in Table 5.8, indicates that legislators representing large constituencies of those types generally favored the warrant requirement. In addition, districts with large concentrations of Japanese favored the warrant requirement, for reasons noted above. On the amnesty issue (Tables 5.3 and 5.7), there is no evidence of the

05-S3453 1/11/06 3:06 PM Page 122

122

Legal and Illegal Immigration Reform ta b l e 5 . 8 House Vote on the de la Garza Amendment to Require Warrants for Searches, IRCA, 1986 Dependent variable: Require warrants for field searches: yes  0 221–170 Republicans: 85 –79, Democrats: 136 –91 Northern Democrats: 98 –58, Southern Democrats: 38-33 Independent variable CONSTANT ln(FBORN) AGEMP PARTY UNEMPCH86 – 85 BRDSW states SOUTH states BLACK MEXICAN CUBAN CHINESE JAPANESE IRISH ITALIAN N –2 log likelihood Correctly predicted

See notes to Table 5.3.

Model 1

Model 2

.2427 (.6808) —

.2704 (.6175) —

.0777* (.0347) .4200 (.2424) .0313 (.1667) .3189 (.4256) .7510 (.3981) .2649 (.4313) .0126 (.0225) .0292 (.0618) .7230 (.4158) 4.0104*** (.9674) .0202 (.0410) .1410* (.0684)

.0848* (.0340) .3994 (.2399) .0713 (.1560) .4753 (.4167) — .3192 (.4182) .0196 (.0225) .0351 (.0620) .6283 (.3989) 3.9888*** (.9675) .00307 (.0394) .1170 (.06640)

393 449.206 66.9

393 452.832 64.6

Model 3 .7652* (.3080) .3310* (.1519) .1325** (.0361) .4364 (.2258) .0318 (.1481) 1.1445*** (.3463) — .0529 (.3350) — — — — — — 393 488.869 62.1

05-S3453 1/11/06 3:06 PM Page 123

Legal and Illegal Immigration Reform

123

ta b l e 5 . 9 Probability of a No Vote on the Panetta Guest-Worker Amendment, 1984

Democrat in BRDSW states Democrat not in BRDSW states Republican in BRDSW states Republican not in BRDSW states

Mean1

10% FBORN 2

20% FBORN 3

10% AGEMP 4

.41 .74 .03 .10

.68 .89 .08 .26

.83 .95 .16 .44

.26 .60 .01 .05

n o t e : Computations are based on logit estimates in Model 3 of Table 5.5. 1. Computed with variables at means: ln(FBORN): 1.34; AGEMP: 4.15; UNEMPCH84 – 83: –2.09; BLACK: .17. 2. Ten percent of the total population in the district is foreign-born; all other variables are at their mean. 3. Twenty percent of the total population in the district is foreign-born; all other variables are at their mean. 4. Ten percent of the total employment in the district is in agriculture; all other variables are at their mean.

potential for a coalition to form between representatives of districts with a large foreign-born population and representatives of agricultural districts. The former tended to vote to retain an amnesty provision, while there is no appreciable effect of agricultural interests on the outcome of the votes on amnesty. Table 5.9 interprets the estimated logit coefficients in Model 3 tested on the Panetta Amendment to the Mazzoli Bill. The table shows the probability of a Republican or Democratic legislator’s casting a negative vote on the traditional guest-worker program as a hypothetical concentration of a foreign-born population in the member’s district increases in size. As the proportion of the foreign-born population rises from 10 percent to 20 percent, the probability of a negative vote increases for both Democrats and Republicans—a relationship that holds in all states.

A Model of Voting: The Immigration Act of 1990 The House vote on the IRCA— on controlling illegal immigration—was very close.33 By contrast, the final vote on the Immigration Act of 1990 — 231 in favor, 192 opposed—reflected a clear choice to expand legal immigration. The range of issues addressed by the 1990 bill (HR 4300) allows us to test the influence of PAC contributions from a set of employers in the high-technology sector; it also offers further exploration of the effect of ethnic constituencies, PAC contributions from labor, and other variables.

Votes Analyzed Among the votes analyzed here is the roll call on passage of the House version of the Immigration Act of 1990. In addition, I include two

05-S3453 1/11/06 3:06 PM Page 124

124

Legal and Illegal Immigration Reform ta b l e 5 . 1 0 Key Amendments to the Immigration Act of 1990

Amendment

Description

Smith Amendment

Proposed cap on annual immigration at 630,000 permanent immigrants: 430,000 family-based immigrants, 150,000 employment-based immigrants, and 50,000 diversity immigrants. The amendment failed by a recorded vote of 143 to 266. (CQ vote 790) A substitute amendment that proposed (1) to delete all except the family-unification provisions of HR 4300; and (2) to retain the current level of immigration at 530,000 persons per year, with 115,000 visas designated for the spouses and children of resident aliens, and 10,000 visas designated for the unmarried adult children of legal immigrants. The amendment failed by a recorded vote of 165 to 257. (CQ vote 810)

Bryant Amendment

important votes on restrictive amendments (Table 5.10) that would have reversed the expansionist thrust of the act. The first is the Smith Amendment, which would have capped legal immigration at 630,000; the second is the Bryant Amendment, which would have deleted all provisions of the law except the generous family-based preference categories. Both were soundly defeated. In this analysis, we test the same hypotheses described in the analyses of the House votes on the Mazzoli Bill (1984) and the Immigration Reform and Control Act of 1986. In addition, because data are available on PAC contributions in 1990 (they were not available for the mid-1980s), we can test the influence of PAC contributions from industry groups and labor unions to specific representatives. Ethnic nonprofit groups do not make PAC contributions to political candidates. In order to measure the influence of ethnic and foreign-born constituencies, I continue to use the FBORN and ethnic variables (e.g., MEXICAN, CHINESE, JAPANESE, IRISH). The hypothesis here is that the greater the PAC contributions from companies in an industry to a legislator, the more likely the legislator is to vote to increase access to foreign or immigrant labor for that industry. In the same way, when the issue is protection for workers, we expect that PAC contributions from unions increase the probability of a representative’s voting in support of that protection. In Model 3, the independent variables again include the log of FBORN, PARTY, UNEMPCH90 – 89, BRDSW states, and BLACK. I expect the direction of their influence to be much as it was in the contest over the IRCA in the 1980s.

05-S3453 1/11/06 3:06 PM Page 125

Legal and Illegal Immigration Reform

125

The measure of concentrated employer interest is industry-specific. In this case the relevant industries are the communications and electronics industries. To get a proxy measure of their political influence, I used PAC contributions, taking a log transformation to correct for a skewed distribution in the data. Since 1988, data on PAC contributions from industries have been available, coded with a three-digit standard industrial classification. I obtained data on PAC contributions from two categories of unions: building trades unions (BLDTRUNPAC) and industrial unions (INDUNPAC). Again, comparable data were not available for the early and mid-1980s. In general, the building trades unions are made up of craft unions. The positioning of the PARTY, SOUTH, and ethnic variables in Model 1 parallels that described for the IRCA votes.

Results of the Analysis Tables 5.11, 5.12, and 5.13 report the results of the vote analysis. Again, the signs on the foreign-born and ethnic variables are in the expected direction. The weight and statistical significance of the logged FBORN variable are both strong. In the final vote on the Immigration Act of 1990, the effect of PAC contributions from the high-technology communications industry is not statistically significant. It is not clear why higher PAC contributions from the high-technology industry influence the vote on the Bryant Amendment in a favorable direction. The amendment would have cut all of the legislation’s provisions except certain family-unification provisions. The bill was numerically expansive did not contain certain protections for workers. The effect of a rise in unemployment in the state over the year preceding the 1990 votes (UNEMPCH90 – 89) does not have an appreciable effect. In the case of the Smith Amendment (Table 5.11), the effect is actually in the wrong direction and statistically significant. One explanation for this result is that by assigning the same change in unemployment level to every district in a state, we probably are picking up unspecified but fixed effects of variations across states, as though we had included a dummy variable for each state.34 The anomalous result may be driven by the high value of unemployment change in several New England states; among these, the largest is Massachusetts, where unemployment increased by 2 points between 1989 and 1990. The

05-S3453 1/11/06 3:06 PM Page 126

126

Legal and Illegal Immigration Reform ta b l e 5 . 1 1 House Vote on the Smith Amendment to Cap Legal Immigration, Immigration Act of 1990 Dependent variable: Cap legal immigration at 630,000: yes  1 143–266 Republicans: 104 –56, Democrats: 39 –210 Northern Democrats: 8 –160, Southern Democrats: 31–50 Independent variable CONSTANT ln(FBORN) ln(HITECHPAC) ln(BLDTRUNPAC) ln(INDUNPAC) PARTY UNEMPCH90 – 89 BRDSW states SOUTH states BLACK MEXICAN CUBAN CHINESE JAPANESE IRISH ITALIAN N –2 log likelihood Correctly predicted

Model 1 1.1338 (.9768) —

Model 2 .4767 (.8657) —

Model 3 4.3428*** (.9812) .9797*** (.1984) .0129 (.0565) .0408 (.0592) 0837 (.0608) 2.0334*** (.4047) .5202* (.2401) .7860* (.3830) —

.0263 (.0575) .00483 (.0633) .1041 (.0629) 2.1013*** (.4227) .5021* (.2518) .9951* (.4786) 1.4518** (.4982) .1895 (1.2918) .0408 (.0296) .1686 (.2323) .8175 (.9413) .8782 (.8611) .0645 (.0555) .1729 (.0974)

.00257 (.0554) .0453 (.0615) .0869 (.0619) 1.8470*** (.4001) .6159* (.2430) .7379 (.4639) — .5620 (.8136) .0482 (.0295) .2403 (.3289) .9358 (.9154) .7369 (.8296) .0348 (.0500) .1860* (.0943)

.4320 (.8123) —

407 336.896 80.3

407 346.229 78.9

407 347.705 78.6

— — — — —

* p  .05; ** p  .01; *** p  .001.

(continued)

05-S3453 1/11/06 3:06 PM Page 127

Legal and Illegal Immigration Reform

127

House delegation from Massachusetts nonetheless voted quite liberally and is heavily responsive to Irish constituents, who played an active part in obtaining pro-Irish immigration provisions in the bill. In general representatives from states in the Deep South (SOUTH) voted in a restrictionist direction, likewise, representatives from the border or southwestern states tended to vote for immigration limits. PAC contributions from building-trade unions influenced votes on the final bill in a favorable, or liberal, direction (Table 5.13). Although several building-trade unions lobbied in the 1980s for restrictive policies on illegal immigration, it is not surprising that PAC contributions from these unions influenced votes in a liberal direction on the question of legal immigration. In negotiations before the floor vote, the AFL-CIO supported the liberal increases in family-based immigration provided for under the Immigration Act; the federation also accepted increases in employment-based immigration on the condition that limitations on temporary workers would be imposed. Considering the substantial increase in total immigration levels called for, it is notable that the labor unions did not oppose the legislation.

NOTES TO TABLE 5.11: Standard errors are in parentheses. Paired yes votes counted as yes; paired no votes, as no. Absentee votes and abstentions are omitted from the regression. Cases with missing values are omitted. ln(FBORN) is ln(percentage of the district’s total population that is foreign-born). BLACK is the percentage of the district’s population that is black. MEXICAN, CUBAN, CHINESE, JAPANESE, IRISH, and ITALIAN indicate the percentage of the district’s population of these specific ethnic ancestries. (s o u r c e : Census tape stf3d, 1990 census.) HITECHPAC is communications-industry PAC contributions to the representative in the preceding election cycle in units of $10,000. The 3-digit industry codes are those for telephone utilities, telecommunications services and equipment, electronics manufacturing and services, and computer equipment and services. The logarithmic transformation is started by 1 unit; experiments with the start constant were carried out to achieve normality in the distribution of the data. (s o u r c e : Data file, Center for Responsible Politics, Washington, D.C.) BLDTRUNPAC is PAC contributions from AFL-CIO building trade unions in the preceding election cycle in units of $1 (notice the different units from HITECHPAC). The logarithmic transformation is started by 1 unit. (s o u r c e : Data file, Center for Responsible Politics, Washington, D.C.) INDUNPAC is PAC contributions from AFL-CIO industrial unions in the preceding election cycle in units of $1. The logarithmic transformation is started by 1 unit. (Source: Data file, Center for Responsible Politics, Washington, D.C.) For PARTY, Republican  1, Democrat  0. UNEMPCH90 – 89 is the unemployment rate in the state in 1990 minus the comparable rate in 1989. See the note to Table 5.3 on UNEMPCH for other methods of estimating unemployment that were tried. (s o u r c e : The state-level estimates at mid-decade are taken from Bureau of Labor Statistics, Geographical Profiles of Employment and Unemployment, various years.) BRDSW states include Arizona, California, Florida, New Mexico, and Texas. SOUTH states include Alabama, Arkansas, Georgia, Louisiana, Mississippi, North Carolina, and South Carolina (states in the Deep South except Florida).

05-S3453 1/11/06 3:06 PM Page 128

128

Legal and Illegal Immigration Reform ta b l e 5 . 1 2 House Vote on the Bryant Amendment to Delete All but Family Provisions, Immigration Act of 1990 Dependent variable: Delete all provisions except family provisions: yes  1 165 –257 Republicans: 113–58, Democrats: 52 –199 Northern Democrats: 11–160, Southern Democrats: 41–39 Independent variable CONSTANT ln(FBORN) ln(HITECHPAC) ln(BLDTRUNPAC) ln(INDUNPAC) PARTY UNEMPCH90 – 89 BRDSW states SOUTH states BLACK MEXICAN CUBAN CHINESE JAPANESE IRISH ITALIAN N –2 log likelihood Correctly predicted

See notes to Table 5.11.

Model 1

Model 2

2.0432* (.9506) —

.6926 (.8234) —

.0855 (.0562) .00506 (.0642) .1594* (.0651) 1.5769*** (.4293) .1566 (.2451) 2.0791*** (.5574) 2.1544*** (.4931) .6922 (.7590) .0861* (.0394) .5858 (.5735) 2.0206 (1.1737) .5985 (.9187) .1029 (.0537) .1664 (.0968)

.1231* (.0541) .0696 (.0596) .1188 (.0613) 1.1379** (.4000) .3439 (.2294) 1.8096** (.5555) —

418 337.908 82.8

Model 3 5.8051*** (.9933) 1.2479*** (.1991) .1105* (.0549) .0499* (.0589) .1214 (.0619) 1.4328*** (.3984) .1837 (.2290) 1.2797*** (.3820) —

.8307 (.5648) .1018* (.0397) .7260 (.5720) 2.2715 (1.1601) .3791 (.8769) .0402 (.0476) .1670 (.0921)

.8778 (.5248) —

418 359.463 79.7

418 365.301 80.6

— — — — —

05-S3453 1/11/06 3:06 PM Page 129

Legal and Illegal Immigration Reform ta b l e 5 . 1 3 House Vote to Pass the Immigration Act of 1990 Dependent variable: Pass Immigration Act: yes  0 231–192 Republicans: 45 –127, Democrats: 186 – 65 Northern Democrats: 159 –13, Southern Democrats: 27–52 Independent variable CONSTANT ln(FBORN) ln(HITECHPAC) ln(BLDTRUNPAC) ln(INDUNPAC) PARTY UNEMPCH90 – 89 BRDSW states SOUTH states BLACK MEXICAN CUBAN CHINESE JAPANESE IRISH ITALIAN N –2 log likelihood Correctly predicted See notes to Table 5.11.

Model 1

Model 2

Model 3

.8501 (.9433) —

1.4894 (.8801) —

.0214 (.0616) .1651* (.0693) .0684 (.0706) 1.1160* (.4543) .3270 (.2615) 3.1397*** (.7245) 1.3124** (.5074) .0547 (.8360) .1933*** (.0511) .8252 (.5654) 1.5745 (1.1123) .8146 (.9308) .0299 (.0536) .1672 (.0979)

—

5.3557*** (.9975) 1.5088*** (.2125) .0302 (.0604) .1878** (.0673) .0393 (.0711) 1.3142** (.4328) .1135 (.2405) 1.5958*** (.4133) —

419 314.686 81.6

.2102** (.0675) .0467 (.0703) .9029* (.4435) .4244 (.2513) 3.1222*** (.7417) — .2730 (.6439) .2103*** (.0518) .9392 (.5635) 1.8240 (1.1412) .6847 (.9266) .000806 (.0501) .1750* (.0958) 419 321.842 80.7

.6537 (.6831) — — — — — — 419 327.541 82.8

129

05-S3453 1/11/06 3:06 PM Page 130

130

Legal and Illegal Immigration Reform ta b l e 5 . 1 4 Probability of a No Vote in the House on the Immigration Act of 1990

Democrat in BRDSW states Democrat not in BRDSW states Republican in BRDSW states Republican not in BRDSW states

Mean1

10% FBORN2

.47 .14 .81 .45

.17 .05 .58 .10

n o t e : Computations are based on logit estimates in Model 3 of Table 5.12. 1. Computed with variables at mean – ln(FBORN): 1.35; UNEMPCH90 – 89: .23; ln(HITECHPAC): 2.06; ln(BLDTRUNPAC): 5.17; ln(INDUNPAC): 5.36; BLACK: .175. 2. Ten percent of the total population in the district is foreign-born; all other variables are at their mean.

Table 5.14 interprets the parameter estimates in the Immigration Act of 1990 for a hypothetical Democrat and a hypothetical Republican representing a district in a border or southwestern state and not representing a district in a border or southwestern state. For example, for a Republican in a BRDSW state, increasing FBORN from the mean (6.2 percent before the log transformation) to 10 percent decreases the probability of a negative vote on the bill by about 23 percent, holding other variables at their mean.

Summary The two case studies of immigration policymaking in 1984 and 1986, and in 1990 suggest that interest-group lobbying contributed to the durability of the preference system established under the Hart-Cellar Act. As they did in earlier periods, business interests played a major role in keeping America’s immigration door open. And intensive lobbying by a proimmigration coalition of ethnic rights and church organizations helped preserve a liberal family-based immigration policy. In the battle over the IRCA, Hispanic organizations in effect took a moderate position on the question of admitting migrant laborers. In contrast to restrictionists, Hispanic groups pressed for liberal familyunification policies and protection of the civil rights of immigrants and their relatives. In contrast to free-marketers, they opposed the unregulated (or loosely regulated) admission of temporary foreign workers. From the study of the roll-call votes, we can conclude that the economic interests of agricultural employers influenced votes in favor of

05-S3453 1/11/06 3:06 PM Page 131

Legal and Illegal Immigration Reform

131

traditional guest-worker programs. We do not have clear evidence that PAC contributions from high-technology employers influenced members’ votes in a proadmission direction, although high-technology companies lobbied for expansive policies on admitting workers. The analysis of PAC contributions from unions suggests that lobbying by building trade unions influenced members’ voting choices in a liberal direction on legal immigration, while lobbying by industrial unions did not have an appreciable effect. In the post-1965 period, unions did not mount lobbying campaigns against legal immigration as the AFL did at the turn of the twentieth century and again in the 1920s. Although we do know that the AFL-CIO has lobbied for restrictions on illegal immigration in recent decades, we unfortunately do not have data on the political contributions the restrictionist building-trade unions made in the early and mid1980s. We found that the past year’s change in unemployment levels in a state is not a significant predictor of a representative’s vote. However, the consistent effect of the dummy variable included for border or southwestern states, whose representatives tended to vote in a restrictionist direction, may reflect the interests of taxpayers in those states. A better measure of taxpayer interests would help confirm this. Clearly, concentrations of Mexican Americans or foreign-born persons who are affected by contested immigration policies exert a strong influence on the policy votes of their legislators, controlling for the party affiliation of those legislators. As concentrations of Mexican American constituents grow, they generally influence votes in a liberal direction. For example, their representatives in Congress are more likely to oppose guest-worker programs in the absence of protections and earned-stay rights for temporary foreign workers. But ethnic interests are by no means monolithic. Although Asian American rights organizations opposed guest-worker programs, a concentration of Japanese Americans tended to influence votes in the opposite direction on the issue of employing migrant farmworkers; traditionally, the Japanese have been a significant ethnic sector among farmers in California who rely on migrant workers. Although Polinard, Wrinkle, and de la Garza (1984) provide evidence that more Mexican Americans supported sanctions than opposed them (see Chapter 4), we find that concentrations of Mexican American constituents tend to influence members to vote against employer sanctions. The aggressive lobbying campaign of Hispanic rights groups

05-S3453 1/11/06 3:06 PM Page 132

132

Legal and Illegal Immigration Reform

against the sanctions conveyed the message to legislators that sanctions went against the interests of Mexican Americans. A large constituency of Mexican Americans may well have made a member more receptive to this argument. It was in immigration battles that Hispanic rights groups emerged as significant players in national policymaking in the 1980s. Their advocacy efforts would continue to play a large role in congressional immigration politics in the 1990s.

06-S3453

1/11/06 3:06 PM Page 133

chapter

Revisiting Reform in a Republican Congress

6

after passage of the immigration act of 1990, Congress did not deliberate again on immigration reform until 1995, after the 1994 midterm election in President Clinton’s first term. The case study of the 1995 –1996 legislative contest reported in this chapter presents an important counterpoint to the two cases analyzed in Chapter 5 because of two changes in the political environment. In the midterm election, the Republican Party won control of both houses of Congress for the first time in forty years. The election produced seventy-three House Republican freshmen who were resolved to bring about conservative change through their numbers. And the most populous state of the Union, California, was witnessing a groundswell of public opinion against illegal immigration. Proposition 187, an initiative to deny public services to illegal immigrants, had won the support of about 59 percent of California’s voters in 1994. The grassroots campaign for its passage brought renewed attention to the ongoing problem of illegal border crossings from Mexico, as did several highly publicized attempts to land boats filled with illegal immigrants from China. With the state facing a serious budget crisis, California’s Republican governor, Pete Wilson, seized on the illegal immigration issue as a central component of his successful campaign for reelection in 1994. His Democratic challenger, Kathleen Brown, opposed Proposition 187. In the Republican victories nationwide and in California in 1994, the principal sponsors of the 1996 immigration bill saw a political opportunity to introduce sweeping immigration reform. Senator Alan Simpson

06-S3453

1/11/06 3:06 PM Page 134

134

Revisiting Reform in a Republican Congress

(R-WY) and Representative Lamar Smith (R-TX) introduced similar bills in the Senate and House calling for major cuts in legal immigration and for stricter enforcement measures. The policy proposals were very similar to those recently recommended by the bipartisan U.S. Commission on Immigration Reform, headed by Barbara Jordan. The Simpson and Smith bills included proposals to eliminate the preference category for adult brothers and sisters of citizens, to reduce allocations of other family visas and the total number of family visas allocated each year, to require employers to pay foreign workers the prevailing wage plus at least a fixed percentage of that wage, and to impose tighter restrictions on hiring H-1B temporary skilled workers. Because the Clinton administration had endorsed the commission’s report, it appeared to many observers that Congress might be on the verge of the first major scaling back of legal immigration in a hundred years. There were earlier signs, however, that the path to reforming legal-immigration policies would not be smooth. All mention of reforming admissions policy had been removed from the Republicans’ Contract with America during the 1994 congressional elections because of divisions on the issue in the national party. The most influential Republicans who opposed cuts in legal immigration were Jack Kemp (a former congressman from New York and secretary of housing and human services under President Ronald Reagan) and William Bennett (a former education secretary under Reagan), who also had opposed Proposition 187. The two men published their opinions on immigration through a think tank they had cofounded, Empower America, sounding themes very close to the proimmigration editorial views of the Wall Street Journal. Simpson and Smith used a familiar strategy to circumvent proimmigration conservatives and liberals: They tacked their proposals to cut legal immigration onto an illegal-immigration bill.1 Their assessment— that there was a strong consensus in favor of passing a bill to strengthen the laws against illegal immigration—would prove correct. In part, that consensus was built out of the policy failures of the employer-sanction provisions of the Immigration Reform and Control Act of 1986. The U.S. Immigration and Naturalization Service and almost every advocacy group on both sides of the immigration issue agreed that sanctions were not working, that they were not deterring illegal immigration. Restrictionists blamed the failure of sanctions on Congress’s refusal in 1986 to institute some form of national identification system to verify the

06-S3453

1/11/06 3:06 PM Page 135

Revisiting Reform in a Republican Congress

135

legal status of workers and prospective hires. Under what many of the original IRCA reformers considered its weakened provisions, employers only had to check that a work applicant held one of a list of identification documents, but there was no reliable or effective way to check a document’s authenticity. As a result, a new underground industry was thriving, producing fraudulent documents for sale at street prices. In the first of two reports, the U.S. Commission on Immigration Reform (1995) had recommended a phone employment-verification system, coordinated by the Social Security Administration and the INS, that would require employers to verify by a phone call to a federal agency the legal eligibility of each new hire based on the employee’s social security number. Although initially the commission had stated that a recommendation on legal-immigration reform would require more study, it decided to formulate a recommendation once Simpson and Smith made clear their intention to introduce omnibus immigration bills in the 104th Congress.

Illegal Immigration Reform and Immigrant Responsibility Act of 1996 Substantial changes in the law ultimately were passed under the IIRIRA to increase the federal government’s authority to control the nation’s borders. In its final form, the act added new bars to admission: An alien unlawfully present in the United States for 180 days but less than a year would be barred from admission in any legal status for three years. An alien unlawfully present for more than a year would be barred from admission in any legal status for ten years.2 The new law also overhauled the procedures for excluding and removing aliens, eliminating several avenues for judicial review of INS decisions. An INS official, for example, would be able to unilaterally rule that an alien is inadmissible at a port of entry unless the alien expresses the intent to claim asylum. The IIRIRA also would add one thousand U.S. border police each year for five years, to reach ten thousand by the year 2000. The act included a provision to start a pilot phone-verification system. New financialresponsibility rules limited the admission of aliens likely to become public charges. And the sponsors of immediate relatives would now have to prove an income of 25 percent above the poverty line; before enactment of the IIRIRA, the level was set at the poverty line.3 At the

06-S3453

1/11/06 3:06 PM Page 136

136

Revisiting Reform in a Republican Congress

same time that these restrictive measures were enacted, the entire quota system, including the preference for the adult siblings of citizens, was left intact.

Left-Right Coalition As it had in the political maneuvering that led to passage of the Immigration Act of 1990, the family coalition of Hispanic and Asian American rights groups and church groups focused on building a left-right coalition to defeat the Simpson-Smith Bill. The strategy was to frame the defense of family immigration in a way that would appeal to conservatives in the Republican Party. Meanwhile, a separate coalition of high-technology companies formed to defend the status quo in labor immigration policy. To build ties with conservatives, leaders of the family coalition worked closely with a network of progrowth libertarians and economic conservatives in the Washington, D.C., advocacy community. The most important members of this network were associated with the libertarian think tank, the Cato Institute, whose positions on immigration were influenced by proimmigration economist Julian Simon.4 Meeting for a while as “the left-right coalition” (the informal name its members gave the coalition), the diverse groups chose a legislative strategy of calling on Congress to split the immigration bill into two bills— one dealing with illegal immigration and the other with legal immigration. In the coalition’s argument, legal and illegal immigration represent two very different sets of policy problems. It contended that Congress should not imply that legal immigrants—people who contribute positively to the nation’s economic and cultural development—are being punished for the crimes of illegal immigrants. In its demand to split the bill, the leftright coalition was joined by a business coalition made up largely of high-technology companies and lobbyists from the National Association of Manufacturers (NAM). The relationship between the business coalition and the left-right coalition would later become strained as pressure mounted on the business groups to make a deal with Lamar Smith to preserve business interests at the expense of family visas. It was widely expected at the outset of the 104th Congress that the ideological fervor of the House Republican freshmen would give a large boost to the restrictionist cause, to scaling back legal immigration. In a powerful rebuff to the Clinton administration and the Democrats in Congress, these seventy-three Republicans had been elected on what they

06-S3453

1/11/06 3:06 PM Page 137

Revisiting Reform in a Republican Congress

137

proclaimed was a revolutionary conservative platform, the Republican Contract with America.5 Early in the first session, Elton Gallegly (R-CA), who had just been elected to his fifth term in the House, headed a task force on illegal immigration; the task force solidified a consensus calling for tough measures to combat illegal immigration. Threatened with the loss of relatively free access to temporary workers and with cutbacks in certain permanent-employment visas under the Smith Bill, business lobbies mounted several high-profile fly-ins to Washington, D.C. The largest corporate interests that spoke out against the bill were from the computer industry—Microsoft, Intel, and Texas Instruments among them. The National Association of Manufacturers coordinated cross-industry lobbying efforts on two fly-in days, during which more than fifty corporate executives came to Washington. The companies participating argued that their international operations made the employment of immigrants vital to their profitability.6 Faced with this flurry of lobbying by the business community, Representative Smith offered to lessen the burden the new bill would impose on employers of foreign workers in exchange for the business coalition’s support of his effort to keep the legal and illegal parts of the bill joined. The business coalition, misreading the strength of anti-immigration forces, decided to cut its losses. Its leaders accepted Smith’s deal, freeing a number of Republicans on the House and Senate Judiciary Committees to vote against splitting the bill at markup (the process of revising a bill). Once its business allies withdrew, the family coalition was alone in calling for the House to split the bill, and it lost. The bill was reported out of the House Judiciary Committee on October 24, 1995, by a vote of 23 to 10. As one of NAM’s key lobbyists reflected afterward, the political trade that the business coalition made with Smith proved unnecessary.7 This became evident when momentum turned in favor of splitting the bill in the Senate Judiciary Committee. The forces calling to split the bill knew that time was on their side. If all the titles on legal immigration were deleted and, as a consequence, debate on a legal-immigration bill was forced to await passage of an illegal immigration bill, the current Congress would be over. The longer consideration of legal-immigration reforms could be delayed, the more likely it was that the reforms would not be enacted. In addition, voter pressure on Congress to reduce legal immigration did not seem a significant factor; after all, Proposition 187 in California had been about illegal immigration. Clearly it was better for the business coalition not to endorse any changes in the status quo on employment than to endorse the less-restrictive measure Smith was

06-S3453

1/11/06 3:06 PM Page 138

138

Revisiting Reform in a Republican Congress

offering. By the time the House floor vote occurred on March 21, 1996, there were indications that the prospects for splitting the bill in both the House and Senate were becoming favorable. In the Senate Judiciary Committee, Spencer Abraham (R-MI), a freshman, indicated his willingness to oppose Simpson; he worked with Edward Kennedy (D-MA) to build a bipartisan coalition on the committee to split the bill. Realizing that momentum was shifting in its favor, the business coalition changed its position and called for splitting the bill before the House vote.8 After the House Judiciary Committee had marked up the bill, as the floor fight in the House neared, two conservative freshmen, Representatives Dick Chrysler (R-MI) and Sam Brownback (R-KS) agreed to cosponsor a floor amendment with Howard Berman (D-CA) to split the immigration bill. Their aides, together with a handful of aides representing liberal Democrats, acted as informal whips on the amendment, coordinating the effort to win its passage. In the Senate, Abraham held a joint press conference with Paul Simon (D-IL) and a spectrum of interest-group representatives to announce their intention to organize a broad left-right coalition to split the bill. The interest groups included NAM, the National Council of La Raza, the National Asian Pacific American Legal Consortium, and the newly formed Union of Needletrades, Industrial and Textile Employees (UNITE).9 Through a series of “Dear Colleague” letters, the lead sponsors of the amendment to split the bill promoted their proimmigration views in both chambers. Identifying themselves with the progrowth wing of their party, the Republicans linked immigration with free trade and free markets. Abraham also made clear to his colleagues in the Senate that he was not opposed to reforming the laws regulating legal immigration and would cooperate with them later in passing a bill for that purpose; but at this time he firmly called for procedurally separating the legal- and illegal-immigration proposals.10 The family coalition, implementing an important outside strategy, encouraged the aides of Representatives Chrysler, Berman, Brownback, and Xavier Becerra (D-CA) to function as whips on the amendment. Two active participants in the family coalition regularly attended the aides’ meetings, helping them craft arguments and amendments.11 For about a month in the spring of 1996, leaders of the Hispanic, Asian American, and church groups in the family coalition met with representatives of the Cato Institute and of Empower America and other conservative organizations to discuss legislative strategy. The participants called themselves a left-right coalition; but in the period leading up to the first floor debate in the House, the family coalition and the conservative

06-S3453

1/11/06 3:06 PM Page 139

Revisiting Reform in a Republican Congress

139

groups decided to meet separately while continuing to coordinate their efforts through individual representatives. Two other groups— Grover Norquist’s Americans for Tax Reform and the Alexis de Tocqueville Institution—also joined the collection of conservative interest groups. And just before the House vote on the amendment to split the immigration bill, the conservatives were able to win the Christian Coalition to their side.12

Labor’s Position Significantly, the push for immigration reform did not come from organized labor, as it had in the 1980s. In fact, by 1994, the AFL-CIO executive council had announced that it was reconsidering its earlier support of employer sanctions in response to the U.S. General Accounting Office report in 1990 that claimed that IRCA sanctions had increased discrimination against all Hispanic workers. The AFL-CIO had strong incentives to soften its position on sanctions. Since the early 1990s, the industrial unions in the AFL-CIO had adopted an aggressive organizing drive to boost their falling membership, and immigrant workers were an important target. In 1995, the AFL-CIO elected as its new president John Sweeney, a former president of the Service Employees International Union. SEIU had spearheaded the wide-area organizing drive to bring immigrant janitors in large urban centers of the country into the union. In 1990, the union had called for repeal of employer sanctions, as did the International Ladies Garment Workers Union, the Amalgamated Clothing and Textile Workers Union, and the Hotel Employees and Restaurant Employees and Bartenders International Union (see Chapter 4). In 1995, the AFL-CIO was facing a dilemma. On the one hand, it supported family-based immigration, an argument for splitting the legal-immigration and illegal-immigration portions of Simpson-Smith. On the other hand, the federation—in particular, the Professional Department—wanted increased protection for workers through tighter regulation of the H-1B program, which admits temporary employees from foreign countries in certain professional occupations.13 Sweeney resolved the problem by speaking out in favor of the amendment before the House floor debate began. With a Republican majority controlling both chambers of Congress, unions lacked the strong representation they had had when Democratic majorities ruled the committees with jurisdiction over labor issues. Robert Reich, secretary of labor in the first Clinton administration, took the

06-S3453

1/11/06 3:06 PM Page 140

140

Revisiting Reform in a Republican Congress

lead in promoting tighter regulations for H-1B workers. Reich worked with Edward Kennedy, the most outspoken senator on the issue, and testified on numerous occasions on employer abuses of the H-1B program. In the end, however, their efforts were to no avail: The Clinton administration had come to support splitting the bill, and that meant deleting all provisions regarding the H-1B program. And Kennedy received little response to his attempts to add protections for American workers to the illegal-immigration bill.

Political Parties The Republican leadership did not play a public role in the intraparty disputes over legal immigration: Party leaders did not come out in favor of or against specific cuts; nor did they take a public position on splitting Simpson-Smith. Significantly, though, key Republican leaders, through their influence on the House Rules Committee, did allow a floor vote on the Chrysler-Berman-Brownback Amendment to split the bill, while denying more-specific amendments, like one favored by Asian American interests to save the brothers-and-sisters preference category. With one organized group among House Republicans espousing a progrowth view of immigration, and the veteran chair of the immigration subcommittee and many conservatives favoring cuts, the majority party was sharply divided. To complicate matters, the majority leader, Dick Armey (R-TX), was openly associated with the progrowth wing of the party and had issued recent public statements opposed to large-scale cuts in immigration.14 And House Speaker Newt Gingrich (R-GA) also had expressed proimmigration views at various times early in the debate on Simpson-Smith. But the popular support for restriction in California created strong counterpressure on the leadership, whose goal was to reelect a Republican majority. Representatives from states with high levels of immigration also were divided. As their votes would later indicate, members of the California delegation, in meetings of the House Republican Conference held to discuss the immigration bill, were uniformly in favor of cutting legal immigration as well as illegal immigration.15 The Florida delegation was divided: Representatives of large Cuban American constituencies tended to favor proimmigration policies. Also divided were the delegations from Texas and Arizona. Although immigration was an emotional issue in states where many immigrants live, most congressional Republicans were not directly pressured by constituents on immigration issues . . . nor were they very knowledgeable about those issues.

06-S3453

1/11/06 3:06 PM Page 141

Revisiting Reform in a Republican Congress

141

Working with Representatives Chrysler and Brownback was Steve Chabot of Ohio, one of the conservative Republicans elected in 1994. Chabot sent his aide to help the other aides working on splitting the immigration bill. In the House Judiciary Committee, Chabot had led the opposition to the proposed verification system. Unable to win the support of Howard Berman and Barney Frank (D-MA), who represented labor’s interest in ensuring an effective verification system in the workplace, Chabot lost. But to win a majority in the House vote, Lamar Smith was forced to modify the original verification proposal, making it experimental in some states and voluntary on the part of employers. Although Democratic leaders in the House, like their Republican counterparts, removed themselves from the early stages of debate on legal immigration, in the end the party leaders—and most of the Democrats—supported the amendment to split the bill. The crucial change came from the Clinton administration. It vacillated in the months before the House vote, first openly endorsing the Commission on Immigration Reform’s report, which was largely in accord with the Simpson and Smith bills’ proposed cuts in legal immigration.16 Subsequently, the administration sent conflicting messages. Then, the day before the bill came to the floor, the administration announced its support of the Chrysler-Berman-Brownback Amendment. With the House Democratic leadership in favor, and with approximately one-third of House Republicans voting to split the bill, the amendment passed by a vote of 238 to 183. Any cuts in legal immigration were deferred for the time being.

The Bill’s Provisions Once the amendment to split the Smith Bill succeeded, it was not difficult for most of the important provisions related to enforcement— or illegal immigration—to win support from a majority of House members. One of those provisions doubled the spending on border control, which Clinton already had increased in the face of Proposition 187’s popularity in California. Another key provision tightened the financial requirements for the sponsors of family immigrants, requiring them to sign legally binding affidavits of support. A sponsor’s income would be counted as part of the alien’s income in tests qualifying the alien for public assistance. The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 had already been passed by the 104th Congress and signed into law by President Clinton. Under this reform, welfare recipients generally were

06-S3453

1/11/06 3:06 PM Page 142

142

Revisiting Reform in a Republican Congress

required to work after two years of public assistance. Families could not receive assistance for more than five cumulative years; and states could tighten that time limit. The act restricted the eligibility of noncitizens for Temporary Assistance for Needy Families (TANF), Medicaid, Supplemental Security Income, and the food stamp program. No immigrants arriving in the United States after the reform was enacted could receive TANF or Medicaid during their first five years of residency. Moreover, the states could deny TANF and Medicaid to noncitizen legal immigrants who had come before the law was passed and to new immigrants even after they had lived in the country for five years. One issue in the Smith Bill became the focus of partisan politics as the 1996 presidential and congressional elections neared. The Republican conference had voted to endorse a controversial amendment that would be introduced by Representative Gallegly. The amendment proposed to give states the authority to deny public schooling to children who were living illegally in the United States. Although some conservative Democrats, principally from the South, supported the Gallegly Amendment, most opposed it. Some claimed it victimized innocent children; others opposed it because law enforcement officials, including police unions, argued that putting these children out of school and onto the streets would increase crime. Speaker Gingrich took the unusual step of coming to the House floor to call on Republicans to support Gallegly’s amendment. This was widely seen as a political move to help House Republicans in California, whose reelection would be vital to retaining the Republican majority in 1996. When President Clinton threatened to veto the entire bill if the noschooling provision was included, the debate in conference stalled movement on the bill for months. No equivalent amendment was proposed for the Senate bill. Influential Republicans—including Senator Robert Dole of Kansas, the Republican candidate for president—wanted to force Clinton’s hand and then use the issue to win votes in California. Others, including Senator Simpson, believed that the Gallegly Amendment would kill illegal-immigration reform.

Deliberations in the Senate Judiciary Committee Simpson originally introduced two separate bills in the Senate, one dealing with illegal immigration and the other with legal immigration. However, to produce a bill similar to the Smith Bill in the House, Simpson agreed to join the two bills after the illegal-immigration bill had moved

06-S3453

1/11/06 3:06 PM Page 143

Revisiting Reform in a Republican Congress

143

through committee.17 The immigration subcommittee approved Simpson’s motion to join the two bills, at which point the bill came before the Judiciary Committee. A bipartisan coalition had formed in the Judiciary to split the immigration bill; it was led by the ranking Democrat, Edward Kennedy, and freshman Republican senator Spencer Abraham. By the time the committee was ready to mark up the bill, the business coalition had rejoined the family coalition in calling for two bills. The business coalition was not alone in realizing that the two-bill strategy could now win; Simpson, too, could count the votes on the Judiciary Committee that were being announced— or discussed privately among its members— in support of the Kennedy-Abraham proposal. To try to avert defeat, Simpson offered a concession to the business coalition that, if successful, would have split the business groups from the family coalition. In a meeting with business lobbyists just before the Senate markup, Simpson agreed to drop all the provisions in the joint bill that addressed employment issues.18 His offer failed, however, to sway the leaders of the business coalition, who realized that under the loosely structured rules of Senate floor debate, Simpson could later offer a floor amendment to put back the employment provisions.19 March 14, 1996, the Senate Judiciary Committee voted 12 to 6 to separate the Simpson Bill into two parts. Two weeks later, the committee voted 13 to 4 to report the bill for consideration by the Senate. The potential for another split in the proimmigration coalition emerged over provisions that would protect native workers from the threat of displacement and lower wages because of temporary foreign workers admitted to the country under the H-1B program. Those provisions were lost when the legal-immigration component of Simpson’s bill was split off from the illegal-immigration component. Kennedy wanted to insert several amendments that would protect American workers. Although Simpson tried to use the issue to form an alliance with Kennedy, the amendments failed to win majority support in the Judiciary Committee. The family coalition did not press for the amendments. It feared that reintroducing the conflict over H-1B workers would have forced business interests and Spencer Abraham, in particular, out of the proimmigration coalition.20

Senate Floor Action Once the Senate Judiciary Committee voted to leave the legalimmigration titles out of the bill, Alan Simpson brought the issue to the

06-S3453

1/11/06 3:06 PM Page 144

144

Revisiting Reform in a Republican Congress

floor of the Senate. His amendment was to reattach the legal and illegal bills, a motion that lost overwhelmingly, 20 to 80. Two factors accounted for the large margin of loss. First, the vote was on a procedural issue—to reattach the deleted titles of the bill related to legal immigration—rather than on the merits of reducing legal admissions. As a result, some Republican senators who would have voted to cut back legal immigration were able to oppose the procedural motion, to insist that the two issues were distinct. Second, the family coalition and business groups held firmly together. This proimmigration coalition was able to swing to its side a voting bloc of free market– oriented Republicans who, although they were also social conservatives, tended to put their commitment to economic conservatism first.21

Conference and Negotiations with the Clinton Administration A significant difference between the Senate and House bills was that the Senate bill did not include the Gallegly Amendment, the proposal to allow states to deny a public education to children who were living in the country illegally. As the family coalition mounted a new lobbying campaign to oppose the Simpson Bill, proponents of the bill feared that including the Gallegly provision would kill it. Forty-five senators had already signed a letter threatening a filibuster if the provision were included. Despite Dole’s pressure on the Republicans to include the provision so that Clinton would have to veto the bill as he had promised (which would cost the president politically in California) or backtrack (which at least would embarrass Clinton), the Republicans held firm; and the provision was not included in the final version of the bill. The thirteen Democratic members of the conference committee were excluded from negotiations to resolve the differences between the Senate and House bills. As a result, the labor protections championed by Kennedy were omitted. In the end, the Clinton administration succeeded in having some of the most extreme measures removed. One example was a measure that would have allowed the deportation of legal immigrants who relied on Medicaid or other federal means-tested programs for more than twelve months during their first seven years in the United States. We can draw three comparisons between the efforts to restrict immigration in 1990 and in 1996. Both campaigns started out with the goal of reducing the number of nonimmediate relatives eligible for

06-S3453

1/11/06 3:06 PM Page 145

Revisiting Reform in a Republican Congress

145

admission—specifically the adult siblings of citizens—and reducing total intake. In these two policy goals, both efforts failed. First, in 1990 and again in 1996, the family-unification priority generated broad bipartisan support and served as the basis of a proimmigration coalition that drew from liberal Hispanic, Asian American, and church groups, on the one hand, and conservative groups on the other. Moreover, there was no inherent conflict of interest between business and labor when the issue was family-based immigration.22 The demand of Irish Americans for an increase in the number of visas allocated to Ireland is an interesting case because it was a demand for special legal treatment for specific visa applicants from a specific country. Clearly, the proliferation of such demands could generate damaging competition among ethnic groups in America. The solution reduced the threat of that competition by redressing an imbalance—giving applicants a chance to participate in a random lottery if their home countries were receiving relatively few visas through the family-preference system. At the same time, it preserved the underlying goal of family unification. Second, Simpson and Smith’s efforts to reduce legal admissions failed in 1996 in part because Republican leaders chose to respond instead to public pressure to contain the costs of immigration. House Republicans and Senate Majority Leader Dole, who would run in 1996 for president, calculated that they could win votes at the polls by politicizing controversies over social-incorporation policy rather than legal-admissions policy. Similarly, the House Speaker’s attempt to play the immigration card by calling on Republicans to vote for the Gallegly Amendment was aimed at helping to reelect California Republicans, whose constituencies were angry about the growing number of illegal immigrants in the state.

A Model of Voting: The Illegal Immigration Reform and Immigrant Responsibility Act of 1996 We now turn to an analysis of votes in the House on the IIRIRA. Here, as in Chapter 5, I test a logistic regression model of a House member’s vote with ethnic, economic, and party variables as predictors. To explore more deeply the significance of economic conditions in congressional districts, I evaluate the effect of the average wage change in the district over the previous four years (since implementation of the Immigration Act of 1990). Data are available on average wage change at the

06-S3453

1/11/06 3:06 PM Page 146

146

Revisiting Reform in a Republican Congress

county level.23 Because counties do not align geographically with congressional districts, I estimated the average wage change per job in the congressional district by taking a weighted sum of wage rates, where the weights were the approximate proportion of the population in each geographic segment formed by overlapping the county and congressional district territories. The first step of the analysis is to evaluate the direction and strength of party and constituency variables on an individual member’s vote. The logistic (logit) equation is reduced to a simpler form than the equation tested in Chapter 5. Two ethnic-specific variables—the percentage of a district’s population that is of Mexican or Cuban descent—are replaced by a single Hispanic indicator (HISP), the percentage of a district’s population that is Hispanic. The variables for black, Chinese, Japanese, Irish, and Italian are not included. In the appendix to this chapter I report the results of testing some alternative models, which include variables for specific ethnic groups and labor unions. Simplifying the model facilitates the second step of the analysis, which is to evaluate whether the issues voted on are sufficiently dissimilar that they represent distinct dimensions in a policy space.24 Whereas the IRCA (1986) focused on illegal immigration, and the Immigration Act of 1990 focused on legal immigration, the IIRIRA included key amendments related to both illegal and legal immigration. The multiplicity of issues allows us to formally test for their dissimilarity in votes on a single bill. In one part of the analysis, for example, I test the validity of my classification of amendments to IIRIRA according to district issue categories, as shown in Table 6.1; in another, I test whether the issues of legal admissions and social incorporation—as represented by two different amendments—are dissimilar. The implications of multidimensionality are discussed in theoretical terms by William Riker (1986, 1988, 1996). If the policy space is multidimensional and outcomes are decided by majority voting, this implies that a political actor may be able to alter the outcome of a policy battle by introducing a new dimension of conflict—an amendment to a bill, for example. According to this line of reasoning, when a new issue is defined and introduced, coalitions tend to realign. Studying interest groups specifically, Baumgartner and his colleagues (2001) suggest that interest groups may redefine issues for this purpose, when these groups have reason to work with members of Congress to strategically redefine issues if by redefining issues they can shift the balance of votes in their preferred direction.25

06-S3453

1/11/06 3:06 PM Page 147

Revisiting Reform in a Republican Congress

147

The legislative history of the IIRIRA presents an example. Restrictionists initially attempted to lay the basis for cutting legal immigration by joining the issues of the public costs of legal immigration and the public costs of illegal immigration. By separating the legal-immigration titles from the illegal-immigration titles, proimmigration forces tried to change the political focus from public costs to the economic and social benefits of legal admission. This strategy divided certain key Republican representatives who were not eager to restrict either family-based or employment-based immigration, from the restrictionist coalition. Similarly, when restrictionists attempted to shape conflict around the issue of a national database to verify employment eligibility, some conservatives found common ground with civil libertarians. Table 6.1 shows the amendments analyzed in this section and their issue categories. Legal admissions relates to the permanent admission of family-based and employment-based immigrants; the ChryslerBerman-Brownback Amendment effectively preserved existing law governing both types of immigrants. The other categories are social incorporation; traditional guest workers, general enforcement; and enforcement and privacy (i.e., enforcement policy that invokes privacy issues as well). The table also shows the total vote on each amendment, and a breakdown of that vote by party.26

Variables and Expectations I model the vote of a hypothetical legislator with a logistic regression equation in which the binary dependent variable is coded 0 or 1. I first examine the factors influencing votes on the issues of legal admissions and temporary farmworkers because these issues are expected to align voting coalitions differently. A vote in favor of a liberal legaladmissions policy is coded 1. The proadmission position on traditional guest workers is coded 1. A description of the variables and expectations follows. As the proportion of the district population that is of Hispanic ancestry (HISP) increases, the probability of a member’s voting in a proimmigration direction on a legal-admissions vote should increase, all else being equal.27 Compared with the other constituency interests, members of ethnic minorities with many relatives living abroad have the most direct interest in family-unification policies; on a vote concerning the admission of traditional guest workers, however, the Hispanic variable should have a negative effect.

General enforcement General enforcement Social incorporation Enforcement and privacy Legal admissions General enforcement Enforcement and privacy Enforcement and privacy Traditional guest workers Social incorporation

Beilenson Amendment1 Bryant Amendment2 Canady Amendment3 Chabot Amendment4 Chrysler-Berman-Brownback Amendment5 Gallegly Amendment (education)6 Gallegly Amendment (verification)7 McCollum Amendment8 Pombo Amendment9 Velazquez Amendment10 Failed 120 –291 Failed 170 –250 Adopted 210 –207 Failed 159 –260 Adopted 238 –183 Adopted 257–163 Failed 86 –331 Failed 191–221 Failed 180 –242 Failed 151–269

Total yes-no vote 4 161 182 79 75 213 43 100 157 21

Rep. 116 9 28 79 162 44 43 91 23 129

Dem.

223 71 50 152 158 20 187 129 76 211

Rep.

67 178 156 108 25 142 143 91 165 58

Dem.

no votes

n o t e s : Not included in this table (or the analysis) is the Burr Amendment, which proposed a 6-month extension of a program that would issue temporary visas to foreign nurses. This table does not report the vote of one Independent. Tables A6.1–A6.4 indicate this Independent member’s vote on four amendments. 1. Proposed (1) to strike a $12 million authorization for the construction of a 14-mile triple fence along the San Diego border with Mexico and (2) to authorize instead $110 million for increased barriers and roads (CQ vote 71). 2. Proposed to require public medical facilities to provide the INS with identifying information on illegal aliens treated or else lose federal reimbursement (CQ vote 73). 3. Proposed to require immigrants issued visas under the diversity program or employment-based classifications to pass a standardized English test administered by the Department of Education (CQ vote 78). 4. Proposed to strike a provision for a voluntary program allowing employers to verify the immigrant status of employees by calling a toll-free number (CQ vote 76). 5. Proposed to strike provisions to limit numbers of legal immigrants and to restrict the preference system (CQ vote 84). 6. Proposed to allow states to deny public education to children in the country illegally (CQ vote 75). 7. Proposed to establish a mandatory program in 5 states to require employers to verify the immigrant status of employees by calling a toll-free number (CQ vote 77). 8. Proposed to improve the design of and material used in social security cards to secure against fraud (CQ vote 72). 9. Proposed a 3-year pilot program in agriculture that would allow employers to hire temporary and seasonal workers for no more than 10 months (CQ vote 85). 10. Proposed to strike provisions to disqualify undocumented aliens from applying for benefits (e.g., food stamps, Medicaid) for their U.S.-born children (CQ vote 74).

Issue category

Amendment

yes votes

ta b l e 6 . 1 Amendments to the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 and Votes by Party Affiliation, 1995 –1996

06-S3453 1/11/06 3:06 PM Page 148

06-S3453

1/11/06 3:06 PM Page 149

Revisiting Reform in a Republican Congress

149

The independent variable SAL is a control for the mean income from salary or wages in each district. This variable likely captures more than one effect. Salary levels may influence the attitudes of taxpayers on the costs of supporting public benefits for illegal immigrants or lowwage legal immigrants. Low-salaried districts, then, might be more likely than high-salaried districts to contain low-salaried immigrants (who use public safety nets). The high-salaried districts bear the greater tax burden for the costs of the services. Another possibility is that salary is associated with varying levels of labor-market competition. For example, higher-salaried workers may compete with higher-salaried immigrants. Still another possibility is that salary is a proxy for education level in the district, which may be correlated with more or less liberal attitudes on whether it is appropriate to deny services to illegal immigrants.28 Although on theoretical grounds we do not have a clear prediction of the effects of median salary, I enter it as a control variable because any or all of these effects may be at work. Two labor-market variables are interacted with a dummy variable for districts in states bordering or close to low-wage countries that send a large proportion of immigrants to the United States. The first, WGCHBRD, is the magnitude of the change (drop) in average wage levels in the district from 1991 to 1995; the second, UNEMPBRD, is the level of unemployment in the district in 1995. Both are expected to have a negative effect on the probability of a member’s voting favorably on legal admissions or traditional guest workers.29 For the index of a member’s ideological proclivity (IDEOL), I use the first dimension D-nominate score described by Poole and Rosenthal (1997).30 This score places a legislator’s views along a liberal-conservative continuum in which the relevant issue is presumed to be the role of the government in the economy. Movement along the continuum in a positive direction represents a shift toward a more conservative (limited government) position; movement in a negative direction represents a shift toward a more liberal (activist government) stance. The expectation is that a conservative (positively signed) D-nominate score will tend to predict a proimmigrant or pro–foreign worker vote on issues that activate the debate on the government’s regulation of the use of temporary or permanent foreign workers. The D-nominate score should have less influence on votes that evoke strong intraparty cleavages, such as privacy issues related to enforcement (the privacy-and-enforcement category).31

06-S3453

1/11/06 3:06 PM Page 150

150

Revisiting Reform in a Republican Congress

Also included are variables for the level of PAC contributions from groups representing four categories of economic interests; hightechnology industries (HITECHPAC), fruit and vegetables growers (FVPAC), building-trade unions (BLDTRUNPAC), and industrial unions (INDUNPAC). We would expect PAC contributions from a group of employers that has a direct interest in importing labor from abroad to be positively related to a member’s vote on admitting permanent or temporary foreign workers. In the amendments to the IIRIRA, two categories of employers lobbied aggressively for open access to foreign workers. One consisted of high-technology employers, who were concerned about access to immigrant and foreign workers in the computer industry. I include the dollar amount of PAC contributions from employers in this group as the independent variable HITECHPAC. The other consisted of agricultural growers who tend to rely on low-wage migrant labor in California; in the debate over the IIRIRA, these growers favored starting a guest-worker program. The other two variables here represent PAC contributions from unions. AFL-CIO unions have an interest in restricting immigration that could threaten union jobs and could lower wage levels. But increasingly in the 1990s, the labor movement refrained from supporting policies that could be construed as violating immigrant rights; instead, it was aggressively organizing immigrant workers to mitigate steep declines in union membership. These conflicting pressures led the AFL-CIO to differentiate its lobbying efforts by issue area. On issues of enforcement, the AFLCIO did not actively lobby for stricter sanctions against employers or even border controls in the IIRIRA deliberations, as it did in the 1970s and early 1980s. Accordingly, I do not include a variable representing union interests in regressions for the enforcement-and-privacy issues.32 On legal admissions, the AFL-CIO gave its support to the ChryslerBerman-Brownback Amendment primarily because the amendment reflected the federation’s support for the current family-based preference system. Although professional unions especially favored the restrictions on temporary-worker visas that were eliminated by Chrysler-Berman-Brownback, the diversity of union interests inside the federation led it to support the status quo. Model 1 includes the metric score of the ideological proclivities of individual members; Model 2 replaces the metric ideology score with a dummy variable for the member’s party (1 if Republican, 0 if Democrat); and Model 3 removes both ideology and party variables, leaving the constituency variables alone.

06-S3453

1/11/06 3:06 PM Page 151

Revisiting Reform in a Republican Congress

151

Results of the Analysis Table 6.2 shows the results of testing the three models on two different issues: legal admissions and traditional guest workers. The strongest and most consistent influence on a member’s vote is ideological score or party. The percentage of Hispanics in a district is a strong predictor of a member’s liberal vote on legal admissions. On traditional guest workers, the dollar amount of PAC contributions from fruit and vegetable growers is statistically significant in all three models. Table 6.3 interprets the estimated coefficients in terms of fitted probabilities. It shows the change in probability of a hypothetical member’s casting a liberal vote when the value of a single independent variable of interest increases by some fixed amount from the mean, holding others constant. I interpret coefficients for which p is less than .15 in Table 6.3, using Model 1 except in the case where I am evaluating the influence of PARTY, which is represented in Model 2. If we compare the columns for legal admissions and traditional guest workers in Table 6.2, the signs on the coefficients of key independent variables are in the expected direction. A concentrated Hispanic constituency in a district has an appreciable influence on a member’s vote on legal-admissions policy. As shown in Table 6.3, a 4 percent increase from the mean in the Hispanic proportion of a district population (HISP) increases the probability of a liberal vote by 10 percent on legal admissions. On the issue of traditional guest workers, the standard error of the coefficient for the Hispanic variable is large, so I do not interpret this coefficient in the table. The effect of the salary variable (SAL) is statistically significant but almost negligible in magnitude. The variables for average wage change in border districts (WGCHGBRD) and unemployment in border states (UNEMPBRD) do not have an appreciable effect on votes on either legal admissions or traditional guest workers.33 However, UNEMPBRD has some effect on other issues, as we will see in Table 6.5. The table shows the strong effect of party affiliation (PARTY). For example, the probability of a Republican member’s voting liberally on legal admissions is 10 percent when other variables are held at their means; the probability of a Democrat’s voting liberally on that issue is 45 percent. With variables other than PARTY at their mean values, the probability of a Republican’s voting in favor of starting a guestworker program is 86 percent, while the probability for a Democrat’s is 42 percent.

06-S3453

1/11/06 3:06 PM Page 152

152

Revisiting Reform in a Republican Congress ta b l e 6 . 2 Logistic Regression Model of a House Member’s Vote, Illegal Immigration Reform and Immigrant Responsibility Act of 1996 legal admissions 1 dependent variable: vote to admit  1

Independent variable

Model 1

Model 2

Model 3

traditional guest workers 2 dependent variable: vote to admit  1 Model 1

Model 2

Model 3

.2709 .2911 1.4947** .5854 .6059 1.7309*** (.5878) (.5748) (.5092) (.6171) (.5808) (.5192) ln (HISP) .3173** .3745** .4070** .1013 .2053 .2748* (.1471) (.1385) (.1308) (.1525) (.1365) (.1297) SAL .000019 .000026 .000012 .00005* .00005** .00003 (.000015) (.000015) (.000014) (.000016) (.000015) (.000014) WGCHBRD 2.4390 4.0306 2.7384 4.3622 1.7358 1.8646 (5.5891) (5.0743) (4.7792) (5.7071) (5.2564) (4.9289) UNEMPBRD .1106 .1122 .1087 .0797 .0523 .0655 (.1096) (.1000) (.0935) (.1250) (.1124) (.1042) IDEOL 2.6103** — — 3.4262** — — (.4062) (.4855) PARTY — 1.9664*** — — 2.1217*** — (.3854) (.4088) ln(HITECHPAC) .6667 .6054 .3878 — — — (.3637) (.3530) (.3402) — ln(FVPAC) — — — .1929* .1825*** .2040*** (.0600) (.0550) (.0528) ln(BLDTRUNPAC) .1022 .1470** .1642*** .0594 .0181 .0543 (.0530) (.0506) (.0478) (.0576) (.0515) (.0480) ln(INDUNPAC) .0280 .00314 .1611* .0530 .1044 .2601*** (.0665) (.0657) (.0542) (.0696) (.0653) (.0549)

CONSTANT

N  2 ln likelihood Correctly predicted

421 372.011

420 397.963

421 426.291

422 354.415

422 399.106

422 429.510

80.3

77.6

76.2

79.1

76.2

75.4

* p  .05; ** p  .01; *** p  .001 n o t e s : Standard errors are in parentheses. Paired-yes votes are counted as yes; paired-no votes, as no. Absentee votes and abstentions are omitted from the regression. Cases with missing values are omitted. HISP is the percentage of the congressional district’s total population that was Hispanic in 1989. (Source: Census file stf3d, 1990 census.) SAL is the mean level of 1989 household income from salary and wages in the congressional district. (Source: Census file stf3d, 1990 census.) WGCHBRD interacts a dummy variable coded 1 for border states (0 otherwise) with the percentage increase in average wage per job in the congressional district between 1991 and 1995, estimated for each congressional district from yearly economic data reported at the county level by the Bureau of Economic Affairs. Border states include Arizona, California, Florida, and Texas. To obtain the district-level estimate of average wage increase, I first calculated the size of the population of each part of the congressional district geographically overlapping a county using a geographic information system (by summing the population of wholly contained census tracts).

(continued)

06-S3453

1/11/06 3:06 PM Page 153

Revisiting Reform in a Republican Congress

153

The sign on the variable representing the dollar amount of PAC contributions from employers in the computer and communications industries is in the expected direction (liberal) but is not statistically significant at the standard p equals .05 level. If we nonetheless interpret it in Table 6.3, an increase in dollar contributions of $1,426 increases the probability of a legislator’s casting a liberal vote on legal admissions by 3 percent. PAC contributions also have a statistically significant effect on a legislator’s vote on the traditional guest-worker issue: That is, a $15,821 increase in contributions from fruit and vegetable growers increases the probability of a favorable vote on the issue by 6 percent.34 If the dollar amount of PAC contributions from labor unions influences a member’s vote in a restrictionist direction, then the legislator should respond to the unions’ protectionist concerns. In Model 3, PAC contributions from industrial unions seems to exert a protectionist influence in voting on traditional guest workers, but that effect is eliminated when we control for ideology (Model 1) or party (Model 2). On the issue of legal admissions, however, we do have some evidence of union influence, but it is in a liberal direction. Although the AFL-CIO generally supports more-restrictive laws to regulate hiring and the wages paid to temporary professional workers, that position was

NOTES TO TABLE 6.2 (continued ) The estimate of the district’s average wage per job is a weighted average of the county-level average wage per job, where the weights are the proportion of the district’s population falling within the overlapping county. UNEMPBRD interacts the dummy variable for border states with unemployment (i.e., the percentage of the civilian labor force unemployed in 1995), estimated for each congressional district from county data in the Local Area Employment Series from the Bureau of Labor Statistics. The district-level estimate is calculated in the same way as the increase in the average wage per job noted above. IDEOL reflects Poole and Rosenthal (1997) D-nominate scores on dimension 1. For PARTY, Republican  1, Democrat  0. HITECHPAC is PAC contributions to the representative in the 1995 –1996 election cycle from the communications industry in units of $10,000. The logarithmic transformation is started by 1 unit. (Source: Data file, Center for Responsible Politics, Washington, D.C.) FVPAC is PAC contributions to the representative in the 1995 –1996 election cycle from fruit and vegetable growers in units of $1,000. The logarithmic transformation is started by 1 unit. (Source: Data file, Center for Responsible Politics, Washington, D.C.) BLDTRUNPAC is PAC contributions to the representative in the 1995 –1996 election cycle from AFL-CIO building trade unions in units of $1. The logarithmic transformation is started by 1 unit. (Source: Data file, Center for Responsible Politics, Washington, D.C.) INDUNPAC is PAC contributions to the representative in the 1995 –1996 election cycle from AFL-CIO industrial unions in units of $1. The logarithmic transformation is started by 1 unit. (Source: Data file, Center for Responsible Politics, Washington, D.C.) 1. Chrysler-Berman-Brownback Amendment to delete all provisions of the bill that would change legaladmissions policy; effectively preserving the liberal status quo. 2. Pombo Amendment to establish a traditional guest-worker program.

06-S3453

1/11/06 3:06 PM Page 154

154 Revisiting Reform in a Republican Congress ta b l e 6 . 3 Fitted Probability of a Liberal (Proimmigration) Vote by a Hypothetical House Member, Illegal Immigration Reform and Immigrant Responsibility Act of 1996 Independent variable All predictors at means (except PARTY) HISP: Increase from mean of 4% (not logged) SAL: Increase from mean of $5,000 WGCHBRD: Increase from mean of .1% UNEMPBRD: Increase from mean of .5% IDEOL: Increase from mean of .2 PARTY: Member is Republican PARTY: Member is Democrat ln(HITECHPAC): Increase from mean of $1,426 ln(FVPAC): Increase from mean of $15,821 ln(BLDTRUNPAC): Increase from mean of $3,000

Legal admissions

Traditional guest workers

.32 .42 — — — .22 .10 .45 .35 — .35

.60 — .60 — — .75 .86 .42 — .66 —

n o t e s : See the notes to Table 6.2 for an explanation of the variables. Table entries give probabilities based on logistic regression coefficients for which p value .15. For the variables representing Hispanic population and all PAC contributions, the increase from the mean is about 1 standard deviation. Except for the PARTY rows, cell entries interpret coefficients for Model 1, Table 6.2. The PARTY rows interpret the coefficient for party affiliation of a member in Model 2, Table 6.2.

muted in the IIRIRA debate by the federation’s support for the status quo in family-based admissions. An increase in PAC contributions from building-trade unions by about 1 standard deviation, or $3,000, from the mean increases the probability of a vote in favor of legal admissions (i.e., for the Chrysler-Berman-Brownback Amendment) by about 3 percent.35 Because we do not have strong evidence of the influence of PAC contributions from industrial unions, we do not interpret the coefficient on that variable. The results of testing for a U-shaped curvilinear relationship between the Hispanic variable (HISP) and the dependent-variable increases were generally negative. There was a good fit of a quadratic equation in only one case: the vote to institute a mandatory verification system (one of Gallegly’s amendments). The mandatory verification system was unpopular except among a relatively small number of representatives from districts in California with moderate but growing numbers of Hispanic constituents. Because their districts stand out as restrictionist and contained moderately large numbers of Hispanics, those numbers appear to be a threshold. In contrast, the restrictionist votes are more evenly

06-S3453

1/11/06 3:06 PM Page 155

Revisiting Reform in a Republican Congress

155

spread across districts with varying concentrations of Hispanics, except in the heavily Hispanic districts, which are decidedly liberal.36

Issue Dissimilarity The next step of this analysis is to investigate whether votes on a particular set of amendments that seem similar actually can be grouped together in a single issue dimension in a policy space; alternatively, each of the amendments represents a distinct issue. If the latter is true, then the policy space will be characterized by high dimensionality. The results of the regression thus far suggest that legal admissions and traditional guest workers are two very disparate issues, considering the opposite direction of signs on the Hispanic, ideology, and party variables across the two issue categories in Table 6.2. But it is possible that sets of other votes can be classified together. I consider three groupings: general enforcement, enforcement and privacy, and social incorporation.37 As noted earlier, Table 6.1 shows how votes can be grouped within issue categories. We next examine the validity of this categorization. For the dependent variables in models shown in Table 6.4, I code a liberal vote (a vote for laxer enforcement) on general enforcement or enforcement and privacy as 1; a vote for tighter enforcement is coded 0. On social incorporation, voting for fewer restrictions on immigrants in terms of their eligibility for public benefits and less-stringent requirements for proficiency in English are considered liberal and coded 1; tighter restrictions and more-stringent requirements are coded 0. For the dissimilarity test, I used Model 1. In the pooled data, the dummy variables (D1, D2, etc.) distinguish the votes on the individual amendments. Table 6.4 reports the results of carrying out the pooled regressions, while Table 6.5 reports the results of carrying out the regressions separately. The null hypothesis is that the fit of the separate regression models is not significantly better than that of the pooled model; this would indicate that the issues are not different. The alternative hypothesis is that the separate regressions provide a significantly better fit; this would suggest that the amendments represent different issue dimensions. We compare the 2 log likelihood statistics (deviance) for the pooled and unpooled regressions. The deviance is roughly chi-square distributed. The sum LB of the 2 log likelihood values for the regressions per-

D1  1 if vote is to strengthen border fence D2  1 if vote is to require medical facilities to report illegal immigrants D3  1 if vote is to start database to identify undocumented workers

PARTY

IDEOL

UNEMPBRD

WGCHBRD

SAL

ln (HISP)

CONSTANT

Independent variable

Model 2

1.7091*** (.2192) —

2.2201** (.2634) —

.0640 .6480 (.4134) (.3569) .3967** .4868*** (.1011) (.0853) .00002 .000004 (.000011) (.000001) 5.4612 1.6972 (4.0488) (3.2270) .1161 .1291* (.0789) (.0640) 4.6308** — (.2612) — 3.7234*** (.2044) 1.2179** .8170*** (.2551) (.2080)

Model 1

—

.9031*** (.1467)

.4680** (.1534)

—

.1545 (.2738) .5304*** (.0639) .00003*** (.000007) .9312 (2.3521) .1418* (.0471) —

Model 3

general enforcement1

—

1.9854** (.1658)

1.9788*** (.1655)

.2214 (.1312) —

3.3899** (.3121) .2284** (.0655) .00006** (.000008) 2.0807 (2.4717) .2078 (.0503) —

Model 2

—

—

3.4378** (.3113) .2120* (.0660) .00005*** (.0000076) 2.0026 (2.4694) .2028*** (.0502) .0610 (.1131) —

Model 1

1.9785 (.1655)

—

—

—

3.4342** (.3111) .2038* (.0642) .00005** (.000008) 2.0511 (2.4673) .2008** (.0500) —

Model 3

enforcement and privacy2

—

—

—

.4587 (.4647) .6563** (.1234) .000001 (.000012) 8.9458 (4.9960) .2779* (.0998) 4.0769** (.2628) —

Model 1

—

—

3.1077*** (.2069) —

.8261* (.3977) .6655*** (.1022) .000011 (.000011) 3.5532 (3.8320) .2325** (.0767) —

Model 2

—

—

—

—

.2679 (.3223) .6494*** (.0794) .00002* (.000008) 1.8178 (2.8506) .2032*** (.0575) —

Model 3

social incorporation3

ta b l e 6 . 4 Regression Model of a House Member’s Vote, Illegal Immigration Reform and Immigrant Responsibility Act of 1996 Dependent variable: liberal vote  1, restrictionist vote  0

06-S3453 1/11/06 3:06 PM Page 156

— —

1,251 964.962 82.5

— —

1,251 747.418 86.8

1,251 1537.645 64.9

—

—

1,248 1468.361 67.1

—

1.2898*** (.1631)

1,248 1462.428 66.3

—

1.2943** (.1633)

1,248 1468.652 67.4

—

1.2895 (.1631)

837 545.895 85.1

1.3998** (.2398)

—

837 724.263 80.8

.9988*** (.1945)

—

837 1044.502 64.0

.6141*** (.1490)

—

* p  .05; ** p  .01; *** p  .001. n o t e s : Standard errors are in parentheses. Paired-yes votes are counted as yes; paired-no votes, as no. Absentee votes and abstentions are omitted from the regression. Cases with missing values are omitted. HISP is the percentage of the congressional district’s total population that was Hispanic in 1989. (Source: Census file stf3d, 1990 census.) SAL is the mean level of 1989 household income from salary and wages in the congressional district. (Source: Census file stf3d, 1990 census.) WGCHBRD interacts a dummy variable coded 1 for border states (0 otherwise) with the percentage increase in average wage per job in the congressional district between 1991 and 1995, estimated for each congressional district from yearly economic data reported at the county level by the Bureau of Economic Affairs. Border states include Arizona, California, Florida, and Texas. To obtain the district-level estimate of average wage increase, I first calculated the size of the population of each part of the congressional district geographically overlapping a county using a geographic information system (by summing the population of wholly contained census tracts). The estimate of the district’s average wage per job is a weighted average of the county-level average wage per job, where the weights are the proportion of the district’s population falling within the overlapping county. UNEMPBRD interacts the dummy variable for border states with unemployment (i.e., the percentage of the civilian labor force unemployed in 1995), estimated for each congressional district from county data in the Local Area Employment Series from the Bureau of Labor Statistics. The district-level estimate is calculated in the same way as the increase in the average wage per job noted above. IDEOL reflects Poole and Rosenthal (1997) D-nominate scores on dimension 1. For PARTY, Republican  1, Democrat  0. 1. Pools votes on the Beilenson (build triple fence on border), Bryant (require medical facilities to report illegal immigrants), and Gallegly (allow public schools to eject illegal immigrant children) amendments with dummy variables as shown. 2. Pools votes on the McCollum (make social security cards tamper-proof), Chabot (start database to identify illegal immigrants, voluntary participation), and Gallegly (start database to identify illegal immigrants, mandatory participation) amendments with dummy variables as shown. 3. Pools votes on the Velazquez (delete provisions to deny public benefits to illegal immigrants) and Canady (require more stringent English proficiency test for legal immigrants) amendments with dummy variables as shown.

D4  1 if vote is to make social security cards tamper-proof D5  1 if vote is to deny public benefits to undocumented immigrants N –2 log likelihood Correctly predicted

06-S3453 1/11/06 3:06 PM Page 157

420 316.856 82.1

2.1035*** (.6252) .2373 (.1605) .00001 (.000016) 7.5150 (6.2180) .1864 (.1251) 4.2458*** (.4342)

1.0876 (.8087) .4279* (.1873) .00003 (.000023) 4.8678 (7.1829) .1216 (.1553) 5.5388*** (.6414) 411 188.589 89.8

Bryant

Beilenson

* p  .05; ** p  .01; *** p  .001. See notes to Table 6.4.

N –2 log likelihood Correctly predicted

IDEOL

UNEMPBRD

WGCHBRD

SAL

ln (HISP)

CONSTANT

Independent variable

420 230.636 88.1

.1734 (.7360) .5266** (.1892) .00001 (.000021) 12.2399 (8.0265) .2057 (.1576) 4.6068*** (.4131)

Gallegly (education)

412 538.201 56.6

2.0069*** (.4508) .3502** (.1096) .00006*** (.000013) .2637 (4.0475) .1907* (.0819) .3441 (.1848)

McCollum

419 527.117 62.3

1.4733** (.4776) .1773 (.1049) .00005*** (.000014) 1.0920 (4.3004) .0866 (.0855) .4382* (.1846)

Chabot

amendment

417 378.242 81.3

3.8086*** (.5424) .0357 (.1390) .00006*** (.000014) 5.6648 (4.3911) .2982*** (.0884) .4001 (.2329

Gallegly (verification)

420 255.908 86.4

1.2302 (.6962) .5766** (.1753) .000005 (.000019) 7.3505 (7.3173) .1413 (.1437) 4.0816** (.3726)

Velazquez

417 286.392 83.2

.6757 (.6024) .7238*** (.1746) .000005 (.000016) 10.9393 (7.0585) .4074** (.1460) 4.1284** (.3794)

Canady

ta b l e 6 . 5 Logistic Regression Model of a House Member’s Vote, Illegal Immigration Reinforcement and Immigrant Responsibility Act of 1996 Dependent variable: liberal vote  1; restrictionist vote  0

06-S3453 1/11/06 3:06 PM Page 158

06-S3453

1/11/06 3:06 PM Page 159

Revisiting Reform in a Republican Congress

159

ta b l e 6 . 6 Fit Statistics for Separate and Combined Regressions

LB  2  log Li LA LA  LB  2 p

General enforcement

Enforcement and privacy

Social incorporation

736.081 18 df 747.418 8 df 11.337 10 df .25

1443.56 18 df 1468.361 8 df 24.801 10 df .01

542.3 12 df 545.895 7 df 3.595 5 df .25

formed separately on each amendment is reported in the first row of Table 6.5. In the second row, LA is the 2 log likelihood statistic for the pooled regression. The third row shows the change in deviance, or LA  LB. For enforcement and privacy, the change in deviance has a chisquare value of 24.801 with 10 degrees of freedom (df); the probability value (p) is less than .01. Thus, we can reject the null hypothesis with statistical reliability. The issues hypothesized as similar under the enforcement and privacy category (in Table 6.1) are actually dissimilar. For the general enforcement category, the change in deviance is 11.337 with 10 df; now p is greater than .25; for social incorporation, it is 3.595 with 5 df, and p is again greater than .25. We cannot reject the null hypothesis that the issues classified under the general enforcement category are similar. And we cannot disconfirm the hypothesis that the issues classified under social incorporation are similar. Using the same method, we can test dissimilarity between two specific amendments. The Chrysler-Berman-Brownback amendment represents the legal admissions category. The Velazquez amendment represents the social incorporation category. Table 6.6 shows a test of the influence of the same covariates on the two votes. (For the Velazquez vote, the results are also reported in Table 6.4.) If we add the 2 log likelihood statistics for the two separate regressions, the sum of 380.216 and 255.908 is 636.124. Adding the number of parameters, the combined number is 12. We use the chi-square value of 636.124, with 12 degrees of freedom, to measure the aggregate fit of the models run separately. In the pooled regression, the votes on the two amendments are combined in one set of data. We use a dummy variable to distinguish the votes on the Velazquez amendment. The 2 likelihood value is 651.754,

06-S3453

1/11/06 3:06 PM Page 160

160

Revisiting Reform in a Republican Congress

ta b l e 6 . 7 Pooled Regression/Chrysler-Berman-Brownback and Velazquez Amendments: Dependent variable: Proimmigration vote  1, restrictionist vote  0

Independent variable CONSTANT ln (HISP) SAL WHCHBRD UNEMPBRD IDEOL D1 = 1 if vote is on Velazquez Amendment, i.e., to deny public benefits to undocumented immigrants N 2 log likelihood Correctly predicted

ChryslerBermanBrownback Amendment (legal admissions)

Velazquez Amendment (social incorporation)

Pooled regression

.1122 (.5329) .2998* (.1422) .000030* (.000015) .1515 (5.4916) .1574 (.1101) 3.0112*** (.2909)

1.2302 (.6962) .5766 (.1753) .000005 (.000019) 7.3505 (7.3173) .1413 (.1437) 4.0816 (.3726)

.3167 (.4195) .4247*** (.1086) .000018 (.000011) 3.0849 (4.2597) .1537 (.0846) 3.4720*** (.2279)

—

—

1.6952*** (.2223) 841 651.754 79.2

421 380.216 78.1

420 255.908 86.4

* p  .05; ** p  .01; *** p  .001. n o t e s : Standard errors are in parentheses. Paired-yes votes are counted as yes; paired-no votes, as no. Absentee votes and abstentions are omitted from the regression. Cases with missing values are omitted. HISP is the percentage of the congressional district’s total population that was Hispanic in 1989. (Source: Census file stf3d, 1990 census.) SAL is the mean level of 1989 household income from salary and wages in the congressional district. (Source: Census file stf3d, 1990 census.) WGCHBRD interacts a dummy variable coded 1 for border states (0 otherwise) with the percentage increase in average wage per job in the congressional district between 1991 and 1995, estimated for each congressional district from yearly economic data reported at the county level by the Bureau of Economic Affairs. Border states include Arizona, California, Florida, and Texas. To obtain the district-level estimate of average wage increase, I first calculated the size of the population of each part of the congressional district geographically overlapping a county using a geographic information system (by summing the population of wholly contained census tracts). The estimate of the district’s average wage per job is a weighted average of the county-level average wage per job, where the weights are the proportion of the district’s population falling within the overlapping county. UNEMPBRD interacts the dummy variable for border states with unemployment (i.e., the percentage of the civilian labor force unemployed in 1995), estimated for each congressional district from county data in the Local Area Employment Series from the Bureau of Labor Statistics. The districtlevel estimate is calculated in the same way as the increase in the average wage per job noted above. IDEOL reflects Poole and Rosenthal (1997) D-nominate scores on dimension 1.

06-S3453

1/11/06 3:06 PM Page 161

Revisiting Reform in a Republican Congress

161

with 7 df. The change in deviance between separate and pooled regressions is 15.630. The difference in degrees of freedom is 12  7, or 5. For a chi-squared value of 15.630 with 5 df, p is less than .01. Thus, we can conclude with statistical reliability that the two amendments are dissimilar. The policy space thus appears to have high dimensionality. The preference functions of a legislator differ greatly between votes on legal admissions and traditional guest workers. When we test for dissimilarity of issues in the enforcement-and-privacy category—the proposal to establish a temporary or mandatory phone verification system or tamperproof social security cards—we have strong confidence that the issues are dissimilar.

Further Analysis of Separate Votes Having found evidence of high dimensionality, we proceed to look at the results of individual regressions on separate amendments using Model 1. We already have discussed the results of testing the three models on the Chrysler-Berman-Brownback (legal admissions) and Pombo (traditional guest workers) amendments. Table 6.5 shows results for testing Model 1 on eight other amendments related to general enforcement, enforcement and privacy, and social incorporation. The votes on the Chabot, McCollum, and Gallegly (verification) amendments are distinctive. Each amendment addressed privacy concerns as part of enforcement policy. The coefficient on the ideological variable (IDEOL) was statistically significant in the case of the Chabot Amendment, which proposed to strike a provision that would begin a voluntary program enabling employers to verify the status of an employee through a telephone call to a national database. But on the McCollum Amendment to improve the design of social security cards (making them more tamper-proof) and on the Gallegly Amendment to make a telephone verification system mandatory in five states, the ideology score is not statistically significant. The appendix to this chapter differentiates the influence of ideological and social conservatism, for the Chabot vote in particular. As the percentage of unemployed persons in border states (UNEMPBRD) increases, the probability of a restrictionist vote increases on certain issues: creating a tamper-proof social security card; the proposal for a mandatory verification system; and the requirement that immigrants entering under the diversity program pass a standardized

06-S3453

1/11/06 3:06 PM Page 162

162

Revisiting Reform in a Republican Congress

English test. These results probably reflect the diffuse costs of illegal immigration on unemployed workers living near the border. The Hispanic variable influences a vote in a liberal direction (favoring less enforcement when privacy concerns are involved) on five of the eight amendments represented in Table 6.5: the Beilenson Amendment (to strike authorization for building a 14-mile triple fence at the border); the Gallegly Amendment (to give states the option to deny schooling to children who are illegal immigrants); the McCollum Amendment (to create tamper-proof social security cards); the Velazquez Amendment (to retain provisions for public benefits for immigrants); and the Canady Amendment (to test the English proficiency of new diversity immigrants). Thus a concentration of Hispanic constituents in a congressional district influences the votes of the legislator in a liberal direction, not only on legal admissions but also on matters of general enforcement (Beilenson), enforcement with privacy (McCollum), and social incorporation (Velazquez and Canady).

Discussion of Findings A mix of political, ethnic, and economic factors influenced legislators’ votes on the IIRIRA amendments, which themselves covered a wide range of issues. Although party and ideology variables are the strongest predictors of a member’s vote, some key votes—in particular, two votes on enforcement and privacy— do not divide neatly along the conservative-ideology dimension. The influence of PAC contributions is strong when agricultural interests are lobbying for traditional guest workers; the influence of PAC contributions from companies in the computer and communications industries is in the expected direction but not significant at standard levels. There are some interesting results concerning the influence of organized labor. For the building-trade unions, there is some evidence that PAC contributions influenced legislators’ votes in a proimmigration direction on issues of legal admissions. This finding is consistent with the results we found in our test of the Immigration Act of 1990 in Chapter 5. It confirms that the union movement is no longer a restrictionist force in the making of immigration policy. Apart from party and ideology, the percentage of Hispanics in a district predicts a liberal vote on six of ten amendments at standard statistical levels in models that controlled for either ideology or party. For the

06-S3453

1/11/06 3:06 PM Page 163

Revisiting Reform in a Republican Congress

163

other four amendments, the sign on the coefficient is in the expected direction—that is, in a liberal direction— except where a proadmission stand on guest workers could not be classified in traditional liberal versus conservative (restrictionist) terms.

Summary The first part of this chapter provides an account of the congressional battle over passage of the Illegal Immigration Reform and Immigrant Responsibility Act of 1996. Latino and Asian American groups played a leading role in the proimmigrant family coalition, which formed mainly to preserve yearly allocations of family-unification visas in a climate of backlash against immigration. Their strategy was to cast the liberal demand for the preservation of family visa categories in a universal light. All Americans benefit from family solidity, they argued. From this position, the ethnic groups were able to construct a left-right alliance with economically conservative groups to lobby jointly for a liberal policy on legal admissions—including employment-based visas. It was also possible to form a left-right alliance on issues of enforcement when enforcement measures threatened to violate privacy rights. In contrast, the position of Latino and Asian American groups diverged sharply from that of conservatives on traditional guest-worker policy, where the rights of foreign workers were at stake. In the second part of the chapter, the analysis of House votes on the IIRIRA and its amendments builds on the findings in Chapter 5 concerning the influence of ethnic and economic interests in immigration voting. As we did in the votes on the IRCA and the Immigration Act of 1990, we find that increasing the percentage of foreign-born persons in a congressional district increases the likelihood of a representative’s voting in a liberal direction on several policy issues. Increasing contributions from fruit and vegetable growers tends to influence members’ votes on traditional guest workers in a liberal direction. But PAC contributions from labor unions do not influence legislators’ votes in a restrictive direction—a result also found in the vote study of the Immigration Act of 1990. And contributions from unions actually push legislators in a liberal direction on legal admissions when we replace the conservative ideology index with party. The multidimensionality of the policy space makes it theoretically possible for a political actor to introduce amendments that will realign

06-S3453

1/11/06 3:06 PM Page 164

164

Revisiting Reform in a Republican Congress

voting coalitions. In the votes analyzed in this chapter, the immigration policy space appears to be highly multidimensional. Voting on amendments on legal admissions, guest workers, and social incorporation policy clearly took place along different issue dimensions. In the mid1990s, ethnic advocates took advantage of multidimensionality as they helped legislators implement amendment strategies that effectively blocked cuts to legal admissions and the establishment of a guestworker program.

06-S3453

1/11/06 3:06 PM Page 165

Appendix: Alternate Models

this appendix provides further analysis of House floor votes on the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 for the reader who would like to compare the results with those discussed in Chapter 5. Four amendments are included in this analysis. The first is the Chrysler-Berman-Brownback Amendment (Table A6.1), adopted by a vote of 238 to 183, which called for deleting the titles in the immigration bill that related to legal immigration. In effect, the amendment split the bill into two separate pieces of legislation and deferred consideration of the legal-immigration issues to a later (unspecified) time. The proposal had earlier been defeated in the House Judiciary Committee; it was revisited and adopted on the House floor. It is important to note three differences between the logistic regression models shown in the tables here and those shown in the tables in Chapter 5. First, Model 1 includes a variable for PAC contributions from building-trade and industrial unions, as in Chapter 5; but in Model 2, the data on PAC contributions are broken down by specific unions. Second, I use National Journal’s ideological rankings of House members on separate scales of economic and social conservatism for every member. Third, in Model 3, I evaluate a set of logistic regression models that includes the ethnic-specific variables (e.g., MEXICAN, CUBAN, CHINESE, etc.) used in Chapter 5 but not in Chapter 6. The National Association of Manufacturers did not support the Chrysler-Berman-Brownback proposal in committee, but the organization did lend its support to the amendment on the floor. The employers

06-S3453

1/11/06 3:06 PM Page 166

166

Appendix: Alternate Models

ta b l e a 6 . 1 Vote on the Chrysler-Berman-Brownback Amendment to the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 Dependent variable: Delete titles restricting legal immigration: yes  1 238 –183 Republicans: 75 –158, Democrats: 162 –25, Independent: 1– 0 Northern Democrats: 121–9, Southern Democrats: 41–16 Independent variable CONSTANT ln(FBORN) ln(HITECHPAC) ln(FVPAC) ln(BLDCONPAC) ln(BLDTRUNPAC) ln(INDUNPAC) ln(UNITEPAC)

Model 1 1.1463 (.7661) .6269** (.1817) .7311* (.3715) .0137 (.0551) .0457 (.0427) .1116* (.0566) .00583 (.0701) —

2.2647* (.8857) .2371 (.2606) .8470 (.3730) .0145 (.0565) —

—

—

.0119 (.016) .0148 (.0105) .01549 (.01334)

.0218 (.0156) .0149 (.0105) .01322 (.01327)

ln(HEREPAC)

—

ln(CARPENTPAC)

—

ln(LONGSHOREPAC)

—

ln(COMMWRKPAC)

—

BRDSW states

1.988* (.7064) .6259** (.1855) .7570* (.3719) .0124 (.0555) .0345 (.0425) —

1.0212 (.6016) .3749 (.2150) 1.3145** (.4547)

—

UNEMPCH95 –94

Model 3

.0205 (.0889) .0508 (.0955) .00321 (.0711) .1010 (.0543) .0582 (.1701) .0774 (.0717) 1.1799* (.5761) .4030 (.2170) 1.3858** (.4620)

ln(SEIUPAC)

PARTY

Model 2

SOUTH states ECONIDEOL SOCIDEOL BLACK

—

.0304 (.0909) .0641 (.1012) .00608 (.0727) .1085 (.0574) .0738 (.1765) .0928 (.0751) 1.3619* (.6076) .4745* (.2383) 1.6112** (.5349) 1.0522 (.5823) .015 (.0162) .0163 (.0108) .0318 (.0179) (continued)

06-S3453

1/11/06 3:06 PM Page 167

Appendix: Alternate Models

167

t a b l e a 6 . 1 ( Continued) Independent variable

Model 1

Model 2

MEXICAN

.0215 (.0194) .0771 (.0756) .1060 (.2210) .1189 (.3773) .1040 (.0833) .0559 (.0551)

CUBAN CHINESE JAPANESE IRISH ITALIAN N –2 log likelihood Correctly predicted

Model 3

418 374.337 79.2

418 377.027 78.2

418 356.396 78.7

* p < .05; ** p < .01; *** p < .001. n o t e s : Standard errors are in parentheses. Paired yes votes are counted as yes; paired no votes, as no. Absentee votes and abstentions are omitted from the regression. Cases with missing values are omitted. ln(FBORN) is ln(percentage of the district’s total population that is foreign-born). BLACK is the percentage of the district’s population that is black. MEXICAN, CUBAN, CHINESE, JAPANESE, IRISH, and ITALIAN indicate the percentage of the district’s population of these specific ethnic ancestries. (Source: Census tape stf3d, 1990 census.) HITECHPAC represents contributions from electronics and communications industries PACs in the preceding election cycle in units of $10,000. The logarithmic transformation is started by 1 unit. The 3-digit industry codes are those for telephone utilities, telecommunications services and equipment, electronics manufacturing and services, and computer equipment and services. (Source: Data file, Center for Responsible Politics, Washington, D.C.) FVPAC is contributions from fruit and vegetable growers PACs in the preceding election cycle in units of $1. The logarithmic transformation is started by 1 unit. (Source: Data file, Center for Responsible Politics, Washington, D.C.) BLDCONPAC is contributions from builders’ associations, and construction and public works PACs in the preceding election cycle in units of $1. The logarithmic transformation is started by 1 unit. (Source: Data file, Center for Responsible Politics, Washington, D.C.) BLDTRUNPAC is PAC contributions from AFL-CIO building trade unions in the preceding election cycle in units of $1. The logarithmic transformation is started by 1 unit. (Source: Data file, Center for Responsible Politics, Washington, D.C.) INDUNPAC is PAC contributions from AFL-CIO industrial unions in the preceding election cycle in units of $1. The logarithmic transformation is started by 1 unit. (Source: Data file, Center for Responsible Politics, Washington, D.C.) UNITEPAC represents the sum of contributions from the International Ladies Garmet Workers Union and Amalgamated Clothing and Textile Workers Union PACs in the preceding election cycle in units of $1. The two unions merged to form UNITE in 1995. The logarithmic transformation is started by 1 unit. (Source: Data file, Center for Responsible Politics, Washington, D.C.) SEIUPAC represents contributions from the Service Employees International Union PAC in the preceding election cycle in units of $1. The logarithmic transformation is started by 1 unit. (Source: Data file, Center for Responsible Politics, Washington, D.C.) HEREPAC represents contributions from the Hotel Employees and Restaurant Employees and Bartenders International Union in the preceding election cycle in units of $1. The logarithmic transformation is started by 1 unit. (Source: Data file, Center for Responsible Politics, Washington, D.C.)

(continued)

06-S3453

1/11/06 3:06 PM Page 168

168

Appendix: Alternate Models

primarily interested in the terms of the amendment were hightechnology companies in the computer and communications industries. PAC contributions from high-technology employers are in the expected direction and are statistically significant. As in Chapter 5, the influence of the log of FBORN is strong. We also find that legislators from districts in border states with many immigrants (represented by BRDSW) voted in the expected direction on the IIRIRA, as they did on the IRCA and the Immigration Act of 1990. Also, in the vote on the Chrysler-Berman-Brownback Amendment, PAC contributions from building-trade unions influence a legislator’s vote in a liberal direction—much as they did in the vote to pass the Immigration Act in 1990. The statistical significance of the BLDTRUNPAC variable does not indicate a particularly active lobbying effort on the part of the building-trade unions, however. In fact, both the Industrial Union Council and the Building and Construction Trades Department of the AFL-CIO wanted family visas protected. The AFL-CIO Department of Professional Employees had reservations about legislative strategies to split off the legal immigration bill because it contained restrictions on the employment of temporary skilled and professional workers. The Chrysler-Berman-Brownback Amendment won the support of 75 Republicans in the House (about one-third of the total); 25 Democrats voted against it. The two freshman Republican sponsors, Chrysler and Brownback, identified themselves as economic conservatives “first” and social conservatives “second.”1 NOTES TO TABLE A6.1 (continued) CARPENTPAC represents contributions from the Carpenters and Joiners in the preceding election cycle in units of $1. The logarithmic transformation is started by 1 unit. (Source: Data file, Center for Responsible Politics, Washington, D.C.) LONGSHOREPAC represents contributions from merchant marine and longshoremen unions PACs in the preceding election cycle in units of $1. The logarithmic transformation is started by 1 unit. (Source: Data file, Center for Responsible Politics, Washington, D.C.) COMMWRKPAC represents contributions from the Communications Workers of America PACs in the preceding election cycle in units of $1. The logarithmic transformation is started by 1 unit. (Source: Data file, Center for Responsible Politics, Washington, D.C.) For PARTY, Republican  1, Democrat  0. UNEMPCH95 –94 is the unemployment rate in the state in 1995 minus the comparable rate in 1994. See the note to Table 5.3 on UNEMPCH for other methods of estimating unemployment that were tried. (Source: The state-level estimates at middecade are taken from Bureau of Labor Statistics, Geographical Profiles of Employment and Unemployment, 1994 and 1995.) BRDSW states include Arizona, California, Florida, New Mexico, and Texas. SOUTH states include Alabama, Arkansas, Georgia, Louisiana, Mississippi, North Carolina, and South Carolina (states in the Deep South except Florida). ECONIDEOL and SOCIDEOL are the member’s ranking on economic conservatism and social conservatism in the National Journal.

06-S3453

1/11/06 3:06 PM Page 169

Appendix: Alternate Models

169

The second vote studied is the Chabot Amendment (Table A6.2), which was defeated 159 to 260. The amendment was intended to delete the provisions of the Smith Bill that would begin test programs in some states of an employee-verification system. Before bringing his bill to the floor, Congressman Smith revised the proposal that had passed in committee; the floor version made participation by employers voluntary rather than mandatory, which enabled the defeat of Chabot’s amendment. On what was essentially an enforcement issue that generated opposition from civil libertarians on the left and right, it is not surprising that the National Journal ratings of economic and social conservative ideology would work in opposite directions. According to the signs on the ideological parameter estimates, the more economically conservative a member (ECONIDEOL), the more likely the member is to vote yes on deleting the employee-verification system. This would represent a conservative libertarian position. The variable for social conservatism (SOCIDEOL) is not statistically significant here. The vote on the Chabot Amendment is the only instance for which we have evidence that representatives from largely black districts voted for stronger enforcement (a no vote) and in the opposite direction from representatives of districts with a large concentration of immigrants. The estimate on BLACK is statistically significant whether a control is added for SOUTH or not. It is also of interest that PAC contributions from the Service Employees International Union influenced yes votes on the Chabot Amendment. The SEIU is one of the fastest-growing industrial unions in the AFL-CIO because of its organizing campaigns among janitors, most of whom are immigrants. SEIU has openly opposed employer sanctions since the early 1990s. The third vote analyzed here is the vote on the Gallegly Amendment to give states the option of denying a public education to children who are illegal immigrants (Table A6.3). The Republican House leadership whipped support for this amendment, which was passed 257 to 163, largely along party lines. The ideological measures are so strong in this case that they drive the significance of party and other variables to lower-than-standard levels. The fourth vote studied is the vote on the Pombo Amendment to establish a new guest-worker program (Table A6.4); it was defeated 180 to 242. The coefficient on FVPAC, the variable representing agricultural interests in a district, is highly significant and strong, as is the coefficient on FBORN.2 There is a difference, however, in the influence of economic

06-S3453

1/11/06 3:06 PM Page 170

170

Appendix: Alternate Models

ta b l e A 6 . 2 Vote on the Chabot Amendment to the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 Dependent variable: Delete employee-verification system: yes  0 159 –260 Republicans: 79 –152, Democrats: 79 –108, Independent: 1– 0 Northern Democrats: 54 –75, Southern Democrats: 25 –33 Independent variable CONSTANT ln(FBORN) ln(HITECHPAC) ln(FVPAC) ln(BLDCONPAC) ln(BLDTRUNPAC) ln(INDUNPAC) ln(UNITEPAC)

Model 1 .5676 (.5867) .1082 (.1318) .1371 (.3098) .0173 (.0458) .0497 (.0346) .0273 (.0489) .00788 (.0565) —

1.1128 (.6804) .2506 (.2002) .1960 (.3281) .00544 (.0488) —

—

—

1.2939* (.5085) .2481 (.1835) .0215 (.3464) —

.0465 (.0601) .1336* (.0577) .0137 (.0559) .0597 (.0411) .185 (.0991) .0114 (.0500) 1.1497* (.4787) .2458 (.1847) .2398 (.3596) —

.0587*** (.013) .0251* (.00915) .01606* (.00802)

.0598*** (.0131) .0274** (.00941) .01727* (.00832)

.0377 (.0630) .1368 (.0614) .0388 (.0584) .0410 (.0452) .1158 (.1095) .0062 (.0528) 1.258* (.5108) .3152 (.1941) .9238* (.4378) 1.0535* (.4540) .0614*** (.0139) .0263 (.00966) .0247* (.0106) (continued)

—

ln(HEREPAC)

—

ln(CARPENTPAC)

—

ln(LONGSHOREPAC)

—

ln(COMMWRKPAC)

—

UNEMPCH95 –94 BRDSW states SOUTH states ECONIDEOL SOCIDEOL BLACK

Model 3

.6147 (.5237) .0462 (.1371) .1772 (.3151) .00381 (.0468) .0537 (.0354) —

ln(SEIUPAC)

PARTY

Model 2

—

06-S3453

1/11/06 3:06 PM Page 171

Appendix: Alternate Models

171

t a b l e a 6 . 2 ( Continued) Independent Variable

Model 1

Model 2

MEXICAN

—

—

CUBAN

—

—

CHINESE

—

—

JAPANESE

—

—

IRISH

—

—

ITALIAN

—

—

416 509.572 509.72

416 496.664 63.2

N –2 log likelihood Correctly predicted

Model 3 .0201 (.0151) .4625 (.2431) .1020 (.1031) .0752 (.1154) .0585 (.0717) .0944 (.0521) 416 470.948 67.1

See notes to Table A6.1.

and social conservatism. To the extent that the economic-conservatism index captures a member’s preference for unfettered access to foreign labor (as opposed to conservative views on social policy), ECONIDEOL is a statistically significant predictor of a vote in favor of the guestworker program.3 At just under the standard probability level (at about p  .06), the influence of a largely black district in Model 2 works in the same direction as the influence of a large concentration of foreign-born persons in a district. Hispanic rights groups have long opposed traditional guestworker programs in the absence of civil rights protections for temporary farmworkers. That was the gist of their argument against the Pombo Amendment, and it well may have forged a black-Hispanic alliance similar to the one that developed in the 1980s over potential civil rights violations related to employer sanctions. By the calculations shown in Table A6.5, an increase in HITECHPAC contributions of $1,500 above the mean in the sample (about $340) increased the likelihood by about 6 percent that a Democratic or Republican representative outside the southwestern border states would vote in favor of the Chrysler-Berman-Brownback Amendment. The first conclusion we can draw from this study of roll-call votes is that apart from the influence of party membership, a concentration of foreign-born constituents in a congressional district exerts the most

06-S3453

1/11/06 3:06 PM Page 172

172

Appendix: Alternate Models

ta b l e a 6 . 3 Vote on the Gallegly Amendment (Education) to the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 Dependent variable: Allow states to deny public education to illegal immigrant children: yes  1 257–163 Republicans: 213–20, Democrats: 44 –142, Independent: 0 –1 Northern Democrats: 25 –104, Southern Democrats: 19 –38 Independent variable CONSTANT ln(FBORN) ln(HITECHPAC) ln(FVPAC) ln(BLDTRUNPAC) ln(INDUNPAC) PARTY UNEMPCH95 –94 BRDSW states SOUTH states BLACK ECONIDEOL SOCIDEOL N 2 log likelihood Correctly predicted

Model 1

Model 2

Model 3

4.9252*** (1.1428) .2112 (.2189) .3069 (.5080) .00513 (.0784) .0822 (.0753) .0456 (.0879) .2055 (.7495) .4685 (.3182) .7069 (.6232) —

1.2458* (.5023) .9588*** (.1869) .5841 (.4012) .00577 (.0639) .00401 (.0665) .0736 (.0761) 3.1865*** (.4714) .58121* (.2827) .7979 (.4911) —

.00049 (.01668) .0781** (.0229) .0587*** (.0147)

4.939*** (1.1453) .2226 (.2258) .3082 (.5068) .00501 (.0785) .0825 (.0752) .0463 (.0878) .2192 (.7524) .4563 (.3232) .7162 (.6246) .1471 (.7235) .001235 (.01866) .0786*** (.0231) .0589*** (.0147)

416 214.580 87.5

416 214.539 87.5

418 281.731 85.9

.0350** (.0120) — —

See notes to Table A6.1.

consistent and strongest influence on policy votes (except on the Chabot Amendment) compared with other constituency variables.4 There is some evidence that PAC contributions from employer lobbies influence House members’ votes on immigration when access to temporary or permanent foreign workers is directly threatened. Contributions from two categories of employers matter most: fruit and vegetable

06-S3453

1/11/06 3:06 PM Page 173

Appendix: Alternate Models

173

ta b l e a 6 . 4 Vote on the Pombo Amendment to the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 Dependent variable: Start traditional guest-worker program: yes  1 180 –242 Republicans: 157–76, Democrats: 23–165, Independent: 0 –1 Northern Democrats: 5 –125, Southern Democrats: 18 – 40 Independent variable CONSTANT ln(FBORN) ln(HITECHPAC) ln(FVPAC) ln(BLDTRUNPAC) ln(INDUNPAC) PARTY UNEMPCH95 –94 BRDSW states SOUTH states BLACK ECONIDEOL SOCIDEOL N 2 log likelihood Correctly predicted

Model 1

Model 2

3.6207*** (.8921) .6249** (.1864) .00201 (.3747) .1900* (.0581) .1002 (.0586) .116 (.0690) .2247 (.6073) .0108 (.1902) .1924 (.4519) —

3.3334*** (.9158) .4851* (.1942) .0448 (.3937) .1875* (.0585) .1175 (.0601) .1354 (.0739) .4599 (.6382) .1160 (.2024) .0329 (.4666) 3.0108*** (.7448) .03630 (.0188) .0639*** (.0180) .00305 (.0109)

.01274 (.01273) .0677*** (.0177) .00319 (.0105) 419 361.295 76.8

419 337.895 78.8

Model 3 –.4648 (.4802) .847*** (.1660) .1616 (.3416) .1807** (.0550) .00615 (.0526) .1041 (.0672) 2.1591*** (.4370) .05244 (.1931) .3727 (.4149) — .004734 (.01075) — — 420 358.370 77.1

See notes to Table A6.1.

growers, and high-technology companies that employ foreign technical personnel. Where data on PAC contributions from unions are available, the study shows several instances, all involving legal-immigration issues, where building-trade unions influenced members’ votes in a liberal (rather than restrictionist) direction. The SEIU also exerted influence on the vote on the Chabot Amendment, which would have eliminated an

06-S3453

1/11/06 3:06 PM Page 174

174

Appendix: Alternate Models

ta b l e a 6 . 5 Probability of a Yes Vote on the Chrysler-Berman-Brownback Amendment to the Illegal Immigration Reform and Immigrant Responsibility Act of 1996

Democrat in BRDSW states Democrat not in BRDSW states Republican in BRDSW states Republican not in BRDSW states

Mean1

10% FBORN2

HITECHPAC + $1,5003

.13 .71 .08 .31

.43 .82 .14 .45

.41 .77 .11 .37

n o t e : Computations are based on logit estimates in Table A6.1. 1. Computed with variables held at means: ln(FBORN): 1.46; ln(HITECHPAC): .2024; UNEMPCH95 – 94: –.442; BLACK: 12.097; ln(BLDTRUNPAC): 4.017; ln(INDUNPAC): 3.32; ln(BLDCONPAC): 4. 97; ln(FVPAC): 1.31. 2. Ten percent of the total population in the district is foreign-born; all other variables are at their mean. 3. HITECHPAC increases $1,500 above the mean (approximately 2 standard deviations); all other variables are at their mean.

employee-verification system. Labor unions have not mounted restrictionist lobbying campaigns around immigration, as AFL unions did at the turn of the twentieth century and again in the 1920s. Although we know that the AFL-CIO lobbied for restrictions on illegal immigration in the early 1980s, we unfortunately do not have data on the contributions of restrictionist building-trade unions for that period. Recent change in unemployment levels in a state is not a significant predictor of a district representative’s vote. However, the consistent effect of the dummy variable included for southwestern border states, whose representatives tended to vote in a restrictionist direction, probably reflects the interests of taxpayers in those states. A better measure of taxpayer interests would help confirm this. There is a change over time in the influence of a large black population in a district where enforcement issues are concerned. In the controversy that surrounded the Immigration Reform and Control Act in the 1980s, a higher concentration of blacks in a district led the district’s representative to vote in a liberal direction—that is, to vote against measures to tighten enforcement of laws against illegal immigration. By the mid-1990s, when the IIRIRA was under consideration, the alliance between black and Hispanic legislators on enforcement issues had weakened, at least in regard to the Chabot Amendment.

07-S3453-CON 1/11/06 3:06 PM Page 175

chapter

Conclusion

7

during the early years of the twenty-first century, Congress began a new series of deliberations on how to institute a rational and humane system for managing the inflow of foreign migrant labor. As it was in the past, the most pressing problem for lawmakers was controlling illegal immigration. Each year, hundreds of undocumented workers risk their lives illegally crossing the southern border of the United States to seek job opportunities and a better life for their families. According to estimates offered by one well-known demographer, by the year 2000, the number of undocumented immigrants living in the United States had reached 9 million (Martin 2004, 61), and most were Mexican migrants. Underlying the debate over temporary foreign workers are ethical questions concerning our obligations as a society to grant at least partial membership status to all people who live and work in the United States over the long term. Whether migrants have entered the country legally or illegally, creating a class of excluded persons without rights to assert their interests weakens the practice of democracy (Carens 1991; Walzer 1981). From a practical standpoint, many immigration policy experts have concluded that illegal-immigration flows from developing countries to labor-importing states are impossible to control entirely (see, for example, Cornelius and Tsuda 2004). When demand for foreign labor is strong and persists over time, and when huge wage disparities exist between sending and receiving countries, the most-resourceful migrants find ways to cross national borders illegally, circumventing border controls. This process is facilitated by migrant networks as well as by labor

07-S3453-CON 1/11/06 3:06 PM Page 176

176

Conclusion

brokers who help foreign workers make travel arrangements and settle in the new country. Under these circumstances, undocumented workers have become integral to the labor force in certain American industries, including garment manufacturing, construction, and hotels and restaurants, where they are part of an informal economy. Reformers argue that the growing informal economy in the United States is unsatisfactory from a number of perspectives.1 Yes, it provides jobs for illegal immigrants; but because they are working in an underground economy, they are vulnerable to exploitation by unscrupulous employers. Employers also pay a price for hiring undocumented workers: They are subject to penalties for violating federal immigration laws.2 Moreover, although their contributions to the economy are valuable, the presence of undocumented migrants in large numbers undermines the credibility of the nation’s immigration laws. The terrorist attacks of September 11, 2001, added a new dimension to the illegal-immigration problem. Federal authorities threw their energies into trying to identify potential terrorists among people attempting to enter the country as immigrants and among immigrants already living in the United States. The attacks led to a restructuring of the American immigration system and a reordering of its policy priorities. In 2003, the U.S. Citizenship and Immigration Services, a bureau of the newly formed Department of Homeland Security, assumed responsibility for immigration services (e.g., issuing visas and naturalization) and for the enforcement functions that had formerly been housed in the U.S. Immigration and Naturalization Service. A challenge still facing the new system is distinguishing between unauthorized immigrants who enter the country for economic reasons and those who come with the intention to do harm. One of the goals of this book is to shed light on the politics of immigration reform. Organized groups representing a variety of economic and cultural interests play a critical role in the making of immigration policy in Congress. And their influence is enhanced by the conflict over immigration issues within the Democratic and Republican parties. Guest-worker legislation in the late twentieth century provides a good example of the growing importance of interest groups in shaping and reshaping immigration policy. From the 1970s through the 1990s, lobbyists for western growers repeatedly suggested proposals to replace the existing agricultural guest-worker program (the H-2A program) with less costly alternatives. They objected to the H-2A program for several reasons. It did not allow expedited access to migrant workers; it

07-S3453-CON 1/11/06 3:06 PM Page 177

Conclusion

177

required them to pay workers’ housing expenses; and it gave the Department of Labor the authority to set wage rates, potentially increasing their costs of labor. But labor unions and Hispanic rights groups strongly opposed traditional guest-worker programs, and none were passed by Congress. In 2000, the United Farm Workers and growers’ representatives did enter into negotiations on a new program that would allow undocumented farmworkers to legalize their status in exchange for concessions to growers’ demands for changes in H-2A wage requirements and labor regulations. By the end of the year, the two sides had agreed on a concept for compromise legislation. But the terrorist attacks of September 11, 2001, took the subject of migrant farm laborers off the congressional agenda for a time. In 2003, the UFW and the growers made another attempt to reach an agreement on issues that had been outstanding in their previous talks. The growers agreed to the union’s demand for earned-stay rights, that undocumented farmworkers have the right to legalize their status after working continuously in agriculture for a specified period. They also agreed that H-2A workers would have the right to sue in federal court if the terms and conditions of their employment contract were violated.3 In return, the union agreed to a three-year wage freeze (at 2002 levels) for H-2A workers. If after three years Congress did not set new wage levels, wages for the next year (and subsequent years if necessary) would be set according to a formula that allowed for inflation.4 The first test of support in the Senate for AgJobs (S 359) took place in April 2005. Rather than directly addressing the bill, the Senate took a vote on cloture to end a threatened filibuster, which would have prevented consideration of AgJobs as an amendment to a supplemental appropriations bill. The vote was 53 to 45 in favor of cloture, 7 votes shy of the 60 votes needed to invoke cloture. As a result, AgJobs was not considered by the Senate as a whole. Still, one newspaper called the agreement underlying the legislation a “grand compromise” that would be revisited in Congress again in the near future (Fresno Bee 2005). The bill appeased rural constituencies and Republicans by easing farmers’ access to temporary foreign workers; and it satisfied Democrats, organized labor, and ethnic rights groups by giving an estimated 500,000 undocumented immigrants the opportunity for legalization.5 The year 2005 also saw important progress in negotiations among interest groups on temporary-worker policy for all low-skilled immigrant workers, agricultural and nonagricultural. In May, a coalition of em-

07-S3453-CON 1/11/06 3:06 PM Page 178

178

Conclusion

ployers’ groups, unions, and rights organizations announced support for the Secure America and Orderly Immigration Act (S 1033, HR 2330). The bill’s Senate sponsors included John McCain (R-AZ) and Edward Kennedy (D-MA); its House sponsors, Jim Kolbe (R-AZ), Jeff Flake (R-AZ), and Luis Gutierrez (D-IL). Known as the McCainKennedy immigration bill, this proposal would increase border protections while creating a portable visa for unskilled and low-skilled workers.6 Applicants would have to show they have a job offer before being eligible for the visa. Portability would give workers more freedom than they have under the existing H programs, including the freedom to move from one employer to another.7 Employers would not be required to pay set wages and meet terms of work established by government agencies, as they do under the existing H-2 program (and in the AgJobs proposal); instead they would be required to pay the same wages and provide the same work conditions as U.S. workers similarly employed. Workers could file complaints with the Department of Labor but could not file suit in federal court. And workers who are unemployed for more than forty-five days would have to return to their home country until they find another job.8 Since 1999, the Essential Worker Immigration Coalition, an alliance of industry and trade associations, has worked to revise the H-2B guestworker program, to facilitate the process of importing temporary workers in nonprofessional and nonagricultural occupations. The program proposed under McCain-Kennedy would expand the number of workers admitted each year. Under the existing H-2B program, only 66,000 workers are allowed in per fiscal year; McCain-Kennedy would make 400,000 visas available the first year, and the number could be increased if the cap were reached during the year.9 The bill also would allow temporary workers holding the new H-5A visa to apply for naturalization after four continuous years as an H-5A worker. The visa itself would run for three years and could be renewed for up to six years. Moreover, pending a decision on a naturalization application, workers could extend their H-5A visa one year at a time until the decision is made, without returning to their home country. Undocumented workers present in the United States before the bill’s enactment would have an opportunity to legalize their status by first paying a $1,000 fine and application fee, working for six years as an H-5A worker, and then paying another $1,000 fine and meeting other eligibility requirements, including a knowledge of civics. Finally, temporary work-

07-S3453-CON 1/11/06 3:06 PM Page 179

Conclusion

179

ers would be eligible for the same labor law protections as their American counterparts. Rights organizations welcomed these provisions of the McCainKennedy Bill and announced they would work for its passage.10 The AFL-CIO, however, did not endorse the bill when it was introduced. Several individual unions that did endorse the proposal would decide to leave the AFL-CIO a few months later, a decision based on organizing policy, not immigration policy.11 Among the unions endorsing McCainKennedy were the Laborers’ International Union of North America, the Northwest Treeplanters and Farmworkers United (Pineros y Campesinos Unidos del Noroeste), the Service Employees International Union, the UFW, and UNITE HERE. In light of the split in labor’s ranks, it is useful to revisit a question asked in Chapter 2: Has the process of immigration policymaking been captured by expansionist special interests? This may be of greater concern now because labor, which has played a role in protecting workers’ interests, is potentially weakened. According to one view, interest groups lobby for the admission of immigrants without regard to the economic effects of immigrant workers on the national welfare (Briggs 1996, 10). In particular, businesses that rely on immigrant labor lobby for liberal immigration policies that expand the labor supply in their own industries but not necessarily in other industries where new sources of labor are needed. And according to the capture theory, ethnic groups want the number of immigrants in the country to go up to strengthen the groups’ political clout. When ethnic groups advocate expansionist policies, then, these policies tend to be politically selfperpetuating: that is, immigration expands proimmigration constituencies. To secure liberal policies for their separate ends, moreover, business and ethnic groups have learned to cooperate in lobbying campaigns. If we extend the logic of the argument, there may be serious cause for concern that expansionist immigration policies will spin out of control. It is true that employers for the most part favor liberal admissions policies for both temporary and permanent immigrants, but ethnic rights groups do not always advocate for larger numbers of immigrants. The policy choice is made issue by issue in a multidimensional policy space. Hispanic rights groups have been in the forefront of lobbying campaigns to preserve the generous family-admissions policies enacted in 1965. Compared with other preference categories, the family-based

07-S3453-CON 1/11/06 3:06 PM Page 180

180

Conclusion

categories are the source of the largest immigration intakes. But these rights groups often have lobbied in opposition to employers’ requests to augment their labor supply with traditional guest workers. The multidimensionality of immigration policy also opens up opportunities for trading issue by issue. For example, the number of immigrants admitted in any single visa category can generally be differentiated from the rights those people will enjoy after they have entered and settled. In legislative negotiations on a single piece of legislation, then, we are likely to see political trading over numbers and rights. Another of my purposes in this book is to illustrate the nature of ethnic identity politics centered on immigrants’ rights. Immigration in the late twentieth century brought a new cast of ethnic actors onto the policymaking stage. As the number of Latinos and Asian Americans elected to Congress gradually rose, a sector of nonprofit organizations established national lobbying arms in Washington, D.C., to advocate for the civil rights and general welfare of Latino and Asian panethnic identity groups. The influence of these organizations on members of Congress has depended on their ability to refine a distinctly modern form of ethnic advocacy. Their lobbying strategy relies at base on networks of ethnic volunteer associations and service agencies in local areas. After 1965, the U.S. immigration system no longer excluded persons of designated races. Thus modern-day ethnic advocates have focused their lobbying efforts on defending nondiscriminatory admissions policies, protecting immigrants’ rights, and providing the opportunity for undocumented long-term residents in the United States to regularize their status. Immigrant rights advocacy today is in several respects reminiscent of civil rights advocacy in the 1950s and 1960s. In both cases, identity politics has been the vehicle for a progressive-rights agenda that has widened the scope of democracy. Ethnic groups in America assert their ethnic particularity in response to actions that would exclude them from equal participation and integration—the aim of the minority civil rights movement since the 1960s. In other words, the goal is integrationist, not separatist. At the same time, ethnic identity politics relies on symbols and tools that assert differences, and these sometimes can be used to separate minority ethnic groups from prospective and actual allies. Divisiveness can be mitigated by the need to defend against an antiimmigrant backlash: Here, an assertive identity politics are necessary to coordinate collective action in opposition to the backlash and to preserve minority civil rights. There are many instances in which ethnic

07-S3453-CON 1/11/06 3:06 PM Page 181

Conclusion

181

identity politics have a narrow or exclusivist orientation; ethnic identity is an avenue for mobilizing a response to discrimination and racism based on ethnic difference. Still, the goal of that response—fair treatment for all—is universal. I have suggested that the broader the benefits of a proposed policy, the greater the ability of an ethnic rights group to influence lawmakers across traditional liberal-conservative divides. In each of the case studies in this book, Hispanic lobbyists demonstrated considerable skill in casting issues in universal terms, which allowed them to forge left-right coalitions and to target conservative legislators in both parties with their demands. Party affiliation is the strongest factor in predicting how an individual legislator is going to vote, but intraparty conflicts and multiple issue dimensions make it theoretically possible to divide either a liberal or a conservative voting coalition. In the 1990s, the United States demonstrated that it was more willing to open its southern border to the free flow of trade than of people.12 Congress imposed increasingly harsh measures to control the unauthorized entry of migrant workers. In 1993, though, it approved the North American Free Trade Agreement, landmark legislation that lowered barriers to free trade in North America. The free-trade agreement increased benefits for Canadian and Mexican businesses and for professional immigrants, but it did not address the rights of low-wage foreign workers. Compared with other interest groups, the ethnic nonprofits that represent Latinos and Asian Americans are favorably positioned to advocate a universal-rights agenda that transcends divisions of race and nationality. Why? Because one segment of their constituent base is made up of migrants living in the United States who remain citizens of other countries. We might otherwise expect ethnic lobbyists to take an especially narrow perspective on immigration policy based on the particularism of ethnic-group interests. But rather than favoring simple increases in the number of visas available to immigrants or migrant workers, ethnic lobbyists have argued from a universal-rights perspective against any large-scale guest-worker program that would not permit the step by step inclusion of the migrant workers into the polity. Their broad perspective may also be a factor in the political effectiveness of ethnic nonprofits in terms of making immigration policy. That process entails bilateral agreements between sending and receiving countries: Policy is no longer the product of domestic politics alone; and its provisions can no longer be evaluated by a proprietary system of measures. The context of foreign relations tends to increase the

07-S3453-CON 1/11/06 3:06 PM Page 182

182

Conclusion

weight of evolving norms on human and migrant rights. At the same time, human and migrant rights conventions once adopted can only be enforced effectively by nation-states, and the ethnic nonprofits play a political role in domestic policy circles as advocates of national laws protecting these rights. More immediately, the influence of the ethnic nonprofits is likely to increase over time in congressional policymaking. The policies they advocate strike a middle ground between the proposals of forces who advocate restriction of all immigration, on the one hand, and those who favor the unregulated admission of workers for economic reasons, on the other. Consequently, ethnic rights advocates have the potential to appeal to legislators with moderate views who can swing votes in the legislative setting. In contemporary America, the voice of nonprofits may seem small and appear to go unnoticed; but ethnic rights groups may prove to be one of the most vital forces in molding an immigration policy that will serve America’s economic and democratic development.

08-S3453-END 1/11/06 3:06 PM Page 183

Notes

chapter 1 1. Political scholars have contributed significant literature on the relationship between immigration and citizenship. Regarding the debate on the rights of temporary foreign workers and their access to citizenship, see Rogers Smith (1997); on the force of international consensus on norms concerning immigrant and migrant rights, see Soysal (1996). 2. This viewpoint aligns with Zolberg’s (1999) thinking, that the politics of immigration unfolds in distinct cultural and economic spheres. In the cultural domain, immigrants, including foreign workers, are a “political and cultural presence” with a putative impact on the host country’s “‘way of life,’ ‘cohesiveness,’ or in current discourse ‘identity’” (84). In the economic domain, immigrants are primarily considered as workers; and the putative effects of immigration are felt primarily in the labor market, where the public perception is that immigrants lower wages and, in the worst-case scenario, take jobs away from natives. 3. I use the terms Latino and Asian American as ethnic categories, but it should be noted that they are also supracategories. The Latino supracategory includes persons with ancestral origins in Latin America: Mexicans, Puerto Ricans, Cubans, Salvadorans, and many others. Similarly, the Asian American category includes Koreans, Chinese, Hmong, Filipinos, and other ethnic groups whose members have ancestral origins in Asia. One organization, the National Asian Pacific American Legal Consortium, serves Pacific Islanders as well as Asian Americans. I also use the terms Latino and Hispanic interchangeably. 4. Because my primary interest here is the processes in which ethnic and economic politics intersect, I do not analyze the politics of admitting refugees or asylum seekers, whose admission is based on foreign policy grounds and humanitarian considerations. In the period under study, the most important reform of refugee law occurred in 1980.

08-S3453-END 1/11/06 3:06 PM Page 184

184

Notes to Chapter 1

There are several general classes of permanent immigrants. The principal categories are employer-sponsored, family-sponsored, refugees, and political asylees. In addition, investors who will create jobs are given preference. Temporary migrants, such as tourists, students, and guest workers, are not formally classified as immigrants in American law. 5. The contemporary influence of ethnic lobbyists has historical precedent. Reimers (1992) notes the role played by ethnic lobbyists in the repeal of the Chinese Exclusion Act in the 1940s; by the 1960s, ethnic and religious organizations and refugee aid groups “became the major nongovernmental groups influencing American immigration policy” (12). Ngai (2004) has shown that EuroAmerican immigrants played a significant role in passage of the Hart-Cellar Act (1965); in contrast, Mexican American and Asian American communities had “virtually no agency” in the reform movement (263). 6. Yet research on interest groups has taken very fruitful directions since the mid-1990s. Baumgartner and Leech (1998), in an evaluation of interest-group research, suggest the importance of large-scale surveys in this subfield of study. They also argue that “literature on interest groups will have a great impact only if it can combine a focus on important questions of democratic representation with the evidence needed to answer those questions” (188). 7. Berry and Arons (28 – 39) cite two studies that demonstrate how nonprofits are only a small part of the interest-group community in Washington, D.C. One is by Schlozman and Tierney (1986, 77); the other is by Berry (1999, 20). 8. Walker notes that “citizens groups built around a compelling moral cause or single issue” were always fewer in number than “associations founded upon occupational communities” (34). But he also reports that their number is growing and that half of the groups he surveyed in 1985 did not form until 1965 (63). 9. For a review of the literature on minority representation, see Tate (2003, 6 –20). 10. The Hart-Cellar Act was codified as the Immigration and Nationality Act Amendments of 1965. 11. The McCarran-Walter Act of 1952 was codified as the Immigration and Nationality Act. 12. See Cornelius and Rosenblum (2005) for a useful review of models of migratory behavior and migration policy. 13. Carter and Sutch (1998) trace the history of American immigration policy, showing the lack of correspondence between business cycles and changes in immigration policy. 14. In a 1990 study titled “American Attitudes Toward Immigration” by the Roper Organization, researchers found that 48 percent of Americans felt that the United States accepts too many immigrants each year (cited in House Committee on the Judiciary, Subcommittee on Immigration, Refugees, and International Law, IRCA’s Anti-Discrimination Amendments Act of 1990 Hearing, 101st Cong., 2d sess., June 27, 1990, 322). An earlier study, conducted in 1985 by Associated Press/Media General, found that 54 percent of adults in the United States believed new laws should be passed to make immigration more difficult

08-S3453-END 1/11/06 3:06 PM Page 185

Notes to Chapter 1

185

(cited at 318). For an analysis of opinion in the mid-1990s, see Gimpel and Edwards (1999, 27– 45). The authors document responses to a CBS News/ New York Times monthly poll taken in September 1994 in which just over 59 percent of respondents favored a decrease in legal immigration. Among Republicans, the total was almost 65 percent; among Democrats, 60 percent. 15. Martin (2004, 61) cites a personal communication from demographer Jeffrey Passel as the source of these estimates. 16. Tichenor (2002) examines this question, which he calls the puzzle of “expansionism.” Freeman (1995a) also addresses the issue in his theory of client politics. 17. There is a growing literature on the effect of economic interests on immigration policymaking. Kessler (1998) extends the Heckscher-Ohlin theory of international trade to study the distributional effects of labor mobility on the interests of socioeconomic actors. As Hollifield (1992) has argued, liberal rights traditions and laws have made it increasingly difficult to limit the civil rights of noncitizens. The liberalization of American immigration law in 1965 was an indirect consequence of reforms brought about by the rights revolution of that era. Cornelius, Martin, and Hollifield (1992, 9 –12) have pointed out that when national governments want to control “unwanted” or illegal immigration, rights-based liberalism ties their hands. Consequently, it is harder for governments to carry out such measures as surveillance to identify illegal immigrants or deportations without due process. Peter Shuck (1984) suggests that communitarian legal traditions explain much of the expansiveness of recent admissions policy decisions. In this view, communitarian ideology encompasses notions of group rights, which are extended to immigrant-ethnic minorities (49 –50) See also Shuck (1992), 37–91. 18. Wilson describes four types of regulatory politics: (1) interest-group politics, in which both the costs and benefits of a policy are narrowly concentrated; (2) majoritarian politics, in which both costs and benefits are dispersed; (3) entrepreneurial politics, in which costs are concentrated, but benefits are dispersed; and (4) client politics, in which costs are dispersed, but benefits are concentrated. 19. Modifying Freeman’s model, Joppke considers the role of the courts as a “separate source of expansiveness and inclusiveness toward immigrants.” In Joppke’s view, “the legal system is the true province of the ‘anti-populist norm’ right identified by Freeman as backing up client politics” (18). 20. For an examination of the complex interaction between international and domestic politics of migration, see Rosenblum (2004). 21. For an analysis of the politics of Proposition 187, see Mac Donald and Cain (1998). 22. I refer to these ascriptive associations as ethnic groups and ethnic rights groups, using the terms interchangeably in this study. The context should make clear whether the term ethnic group or ethnic rights group refers to an organization or to a category of people who identify with an ethnic group, to one degree or another. Some ethnic identity organizations unify members in pursuit of common interests; in this case, the ethnic identity organizations can also be con-

08-S3453-END 1/11/06 3:06 PM Page 186

186

Notes to Chapter 1

sidered interest groups. See Gutmann for a discussion of the distinction between identity groups and interest groups (13 –15). 23. Cortes (1992) reports that in 1978, the National Council of La Raza, after establishing its headquarters in Washington, D.C., set out its goals as follows: (1) training and technical assistance to a growing number of affiliated, Latino nonprofit community self-help and service organizations throughout the nation; (2) research and advocacy addressing national public issues of special concern to Latino communities in the United States; (3) increased public awareness of Latino community needs and programs; (4) creation of new national, regional, and local organizations for various Latino community economic and political empowerment projects; and (5) strengthening the council’s own competence and endurance as a national institution for social change (307– 308). 24. In 1995, NAPALC described its mission as follows: “to advance the legal and civil rights of the nation’s 7.3 million Asian Pacific Americans” (12). 25. Hula distinguishes the Leadership Conference on Civil Rights, which he describes as “long-term coalition experts,” from coalitions in the transportation domain, which favor short-term coalitions (13). 26. The LCCR was founded in 1950 by A. Philip Randolph, founder of the Brotherhood of Sleeping Car Porters; Roy Wilkins, executive secretary of the NAACP; and Arnold Aronson, leader of the National Jewish Community Relations Advisory Council. For an analysis of when interest groups pursue coalition strategies, see Hojnacki (1997). 27. In forming these left-right coalitions, the ethnic rights groups skillfully practiced a strategy of expanding the scope of battle, which, as Schattschneider (1975) describes, is a critical ingredient of politics. 28. According to the Universal Declaration of Human Rights (Art. 16, para. 3), the “family is the natural and fundamental group unit of society and is entitled to protection by society and the State” (Green 1956, 176). 29. Camarillo (1991) describes a misconception that local and national ethnic organizations emerged in the 1960s (15); his account of their history begins in 1848. 30. For a contrasting view, see Skerry (1993), who criticizes the approach of the Hispanic civil rights establishment. According to Skerry, Hispanic rights groups claim the status of a minority and demand reparations for past discrimination. He argues that the majority of Mexican Americans care about making their way into the political and social mainstream, not about restitution for past discrimination. Political empowerment, he suggests, will result from urban electoral strategies similar to those taken earlier by other ethnic immigrant groups. 31. The McCain-Kennedy Bill was titled the Secure America and Orderly Immigration Act (S 1033). Representatives Jeff Flake (R-AZ), Luis Gutierrez (D-IL), and Jim Kolbe (R-AZ) introduced the House version (HR 2330) of the bill. A more-detailed account of the legislation appears in Chapter 7. 32. Registration of undocumented immigrants would help identify and establish the accountability of the large undocumented population in the United

08-S3453-END 1/11/06 3:06 PM Page 187

Notes to Chapter 2

187

States. Papademetriou argues that such accountability is necessary to establish a stable, orderly immigration system. Establishing a temporary-worker program would benefit the Mexican economy, which has an oversupply of unskilled and low-wage workers; it also would benefit employers’ interests in the United States, where there is a demand for low-cost Mexican labor. Another benefit for the United States would be Mexico’s cooperation in strengthening border security. chapter 2 1. Simpson, a Republican from Wyoming, served in the Senate for eighteen years; he did not run for reelection in 1996. The interview took place in Washington, D.C., on June 12, 1995. 2. Refugee and asylee policy raises a set of separate foreign policy and humanitarian concerns, which I do not analyze in this study. However, it should be noted that refugee issues are sometimes considered alongside family-based and employment-based admissions issues, which can further expand the dimensionality of immigration bills. 3. A National Research Council study (cited in Martin 2004) found that in 1996, the average immigrant and his or her descendants paid $80,000 more in taxes than the value of tax-supported services they consumed that year. One of the study’s assumptions was that the children and grandchildren of immigrants have both an average education and an average level of income compared with other Americans (Martin, 73). 4. There is no agreement among economists about the size of payments to immigrants through the welfare system as a whole. For a discussion of why economists come to very different conclusions about whether immigrants pay their way in taxes, see Borjas (1994). He explains that the calculations are accounting exercises, in which the conclusions are highly sensitive to assumptions. Martin (2004, 73) discusses the varied public finance effects of immigration in different states. For a general framework on assessing the fiscal impact of immigration, see McCurdy, Nechyba, and Bhattacharya (1998). 5. In the mid-1990s, when illegal immigration was an especially hot issue, a Los Angeles Times poll (conducted and released in August 1996) revealed that 74 percent of Republicans and 71 percent of Democrats felt that illegal immigration was a “major problem.” 6. Politicians in California’s statehouse catapulted anti-Chinese immigration restrictionism to national prominence in the last quarter of the nineteenth century. An early avenue was to sue the federal government for the costs of incorporating immigrants. Similarly, the governors of California and Florida filed suits against the federal government in the early 1990s for reimbursement of the costs of illegal immigration. 7. Zolberg (1990) observes that immigration controversy historically has led to “odd” partnerships. The alliances are odd because they cut across liberalconservative divides. For example, on the side of expanding immigration are

08-S3453-END 1/11/06 3:06 PM Page 188

188

Notes to Chapter 2

business groups and ethnic advocates; on the restrictionist side, unions and nativists join together. 8. One explanation Card offers for this finding is that immigrants may have generated new economic activity, which created jobs and offset the effect of the increased labor supply on local workers. Borjas, Freeman, and Katz (1992) offer another explanation, that workers are mobile. Workers who competed with the new immigrants may have moved to other cities, in which case the wage effects of the labor inflow would have been diffused across different cities (242). 9. Borjas, Freeman, and Katz note the impact of trade here as well, through the implicit use of unskilled labor in the manufacturing of imported goods. For an example of arguments that emphasize the economic benefits of immigration to receiving countries, see Simon (1989). For evidence that native wages were lowered by immigration in the period before the Immigration Act of 1924 was passed and a system of national-origin quotas was established, see Hatton and Williamson (1995). 10. This argument builds from a framework analyzing union incentives in Rees (1979, 30 – 45). 11. The motivation for accepting liberal immigration policies is not likely to be pressure from employers to lower wages; in that case, the union could simply agree to lower union wages. 12. Charles Kamasaki testifying before the Senate Committee on the Judiciary, Subcommittee on Immigration and Refugee Policy, Hearing on the Immigration Reform and Control Act of 1986 (Public Law 99-603), 100th Cong., 1st sess, April 10, 1987. At the time, Kamasaki was NCLR’s director of policy analysis; today he is the organization’s senior vice president. 13. For an analysis of the emerging power of the Latino vote, see de la Garza and DeSipio (1999). 14. For example, using 1992 National Election Survey data, Citrin and his colleagues (1995) found that Hispanics did not have distinctly proimmigration views. 15. Citrin et al. and Burns and Gimpel specify different models for testing the influence of attitudes. Their findings on the relative strength of black, white, and Hispanic opinion on restricting immigration reflect differences in their models. 16. The Los Angeles County Social Survey examines a large multiracial sample of the general population. It has measured the beliefs, perceptions and attitudes that form individual’s attitudes toward race, immigration, and affirmative action. 17. The statement of FAIR’s goals is taken from Cortes (1992, 214). 18. Although FAIR’s views are sometimes portrayed as classic nativism, FAIR’s founder, John Tanton, cast his beliefs about national unity in terms of a melting pot: “I’m a great believer in trying to emphasize the similarities among people” (quoted in Cooper 1986, 1210). In years when Congress was considering passage of legislation that would control both illegal and legal immigration, FAIR reported that its membership grew substantially—from 5,000 in 1982 to 45,000 in 1988 (Cortes 1992, 228).

08-S3453-END 1/11/06 3:06 PM Page 189

Notes to Chapter 3

189

chapter 3 1. Before the mid-1870s, most regulation of immigration was left to the states. In 1864, Congress enacted the first law concerning contract labor, prohibiting contracts with immigrant laborers for more than one year and conditions of slavery or voluntary servitude. But it was in the Act of March 3, 1875, that Congress for the first time assigned direct regulation of immigration to the federal government, prohibiting the entry of criminals and prostitutes. That same year, the Supreme Court ruled in Henderson v. Mayor of New York (92 U.S. 249, 1875) that state regulation of immigration was unconstitutional because it usurped the exclusive power of Congress to regulate foreign commerce. Under Henderson, states could no longer require shipmasters to pay bonds and head taxes for passengers. In 1891, the Immigration Service was established to regulate the growing influx of immigrants. 2. One strong grass-roots organization that espoused nativism in calling for restrictions on immigration and more-stringent requirements for naturalization was the American Protective Association (Tichenor 2002, 71). 3. Higham (1992) gives a penetrating account of the relationship between nativism and nationalism. 4. Ngai (1999) examines how “national origins” in the Immigration Act of 1924 were invented: They “comprised a constellation of reconstructed racial categories. . . . At one level, the new immigration law differentiated Europeans according to nationality and ranked them in a hierarchy of desirability. At another level, the law constructed a white American race, in which persons of European descent shared a common whiteness that made them distinct from those deemed to be not white” (67). 5. The act also imposed a literacy requirement on all adult immigrants and increased the head tax on new immigrants. 6. The Act to Repeal the Chinese Exclusion Acts, to Establish Quotas, and for Other Purposes (57 Stat. 600-1) was signed into law in December 1943. 7. The Act of July 2, 1946, amended the Immigration Act of 1917; it was also known as the Luce-Cellar Act. 8. Immigrants from the Western Hemisphere did not fall under individualcountry quotas. Congress also admitted some disabled persons by “mortgaging” future admissions quotas. For details, see U.S. House Committee on the Judiciary (1995, 586). 9. For a full account of the politics leading to the end of the national-origins quotas, see Wagner (1986). 10. In his veto message, Truman stated: “There are only a few examples of the absurdity, the cruelty of carrying over into this year of 1952 the isolationist limitations of our 1924 law. In no other realm of our national life are we so hampered and stultified by the dead hand of the past, as we are in this field of immigration” (1961, 443 – 444). 11. The unused visas would be placed in a reserve pool during the five years; later they could be allocated to quota immigrants through a new system

08-S3453-END 1/11/06 3:06 PM Page 190

190

Notes to Chapter 3

based on preferences according to an applicant’s skill, family relationship to citizens and permanent residents, and labor shortages (Stern 1975, 29). 12. Eastland was also responsible for blocking an employer-sanctions bill in the 1970s, when it was considered on the floor of the Senate. Although he was known as a social conservative, Eastland had been a grower and had relied on Mexican labor, as did growers among his constituent base in Mississippi. 13. Attorney General Robert Kennedy argued that the national-origins quota system was antiquated: “A maid or an unskilled laborer from a northern European country can enter this country within a matter of weeks, while scientists or doctors or other highly skilled persons from less favored countries wait for months and years” (testimony before the Senate Committee on the Judiciary, Subcommittee on Immigration and Naturalization, Hearings on S 500, 89th Cong., 1st sess., March 4, 1965, S. Rep. 216). 14. Wagner cites Congressional Quarterly Weekly Report (October 8, 1965, 2042), in which Feighan told a reporter that the provision would benefit one thousand families in his district. 15. Statement of the House Republican Policy Committee on HR 2580, 89th Cong., 1st sess., Congressional Record 111 (August 25, 1965): H 21789. 16. Congressional Quarterly Almanac 21 (1965): 462. 17. This account of the impact of the 1964 presidential elections relies largely on Stern (1975, 51–52). Also see Congressional Quarterly Almanac 20 (1964): 324. 18. Congressional Quarterly Almanac 20 (1964): 324. 19. In the House, the breakdown was 295 Democrats and 140 Republicans; in the Senate, there were 68 Democrats and 32 Republicans. 20. The details of the preference system were modified in 1990, but the basic principle of giving preference to close family members over immigrants with needed skills did not change. 21. Moreover, conservative lawmakers could point out that the calculus used to allocate visas to each preference category would preserve the existing ethnic makeup of the nation. 22. Hearings on S 500, 217. 23. HR 2580, 89th Cong., 1st sess., Congressional Record 111 (September 20, 1965): H 24482. 24. HR 2580, H 21757–21758. 25. An example: In 1934, Congress passed the Reciprocal Trade Agreements Act (RTAA), which delegated authority to set tariffs for separate countries to the executive branch. Before this law was passed, Congress regularly passed legislation setting tariffs on individual goods—a process that repeatedly led to battles among organized interests and impeded the liberalization of trade. A Democratic initiative, the RTAA protected the party from conflict between labor and business constituencies. The RTAA was initially opposed by Republicans; but by the late 1940s, the GOP had come to agree with the RTAA’s general approach. There is a large literature on why congressional coalitions delegate lawmak-

08-S3453-END 1/11/06 3:06 PM Page 191

Notes to Chapter 4

191

ing authority to executive agencies and the consequences of that delegation. For example, see McCubbins, Noll, and Weingast (1987, 243 –277). 26. In Canada, for example, the 1976 Immigration Act gave parliament authority to set numerical targets each year; formerly, the numbers were established by administrative regulation. In New Zealand, the national government also sets annual immigration targets. Both countries have a policy of actively encouraging workers with needed skills to immigrate. 27. Mark Starr was the longtime (1935 –1960) educational director of the International Ladies Garment Workers Union. He notes that in the spring of 1946, representatives from both the AFL and the CIO testified in congressional hearings on the Gossett Bill (HR 3663) that the organizations opposed restrictions in the bill that would have cut immigration quotas in half. 28. HR 2580, H 21789. 29. Even so, there were conflicts between the U.S. and Mexican governments over the issue of guest workers. Guest workers did not come to the United States on permanent visas—which were issued without limitation—but as workers under contract for fixed periods of time. Those who overstayed or who crossed the border without authorization were subject to deportation. Periodically the U.S. government carried out widely publicized efforts to detain and deport illegal immigrants, as it did in the 1950s campaign known as Operation Wetback. 30. Congressional Quarterly Almanac 21 (1965): 478. 31. That list was made up of China, Germany, Greece, India, Italy, Korea, the Philippines, Portugal, Poland, and Yugoslavia. 32. By contrast, the governments of Canada and Australia had decided in the 1960s that the principal goal of their immigration policy would be to attract needed human capital. 33. The Japanese American Citizens League, for example, testified in favor of removing racial restrictions against Asians. See Hearings on S 500, S. Rep. 619 – 630. chapter 4 1. Craig (1971) analyzes the nature of the opposition to the bracero program and the politics of its demise. Calavita (1992) examines how institutional fragmentation in the federal bureaucracy and other bureaucratic dynamics shaped the program. 2. Throughout this discussion, I use Mendelberg s (2001) definition of norm: “an informal standard of social behavior accepted by most members of the culture that guides and constrains behavior” (17). 3. For civil rights groups, incorporating undocumented immigrants into their constituent base was a natural step once they recognized that undocumented immigrants were woven into the communities and families of members of grassroots groups or unions. 4. MALDEF was modeled after the NAACP Legal Defense and Educational Fund, an offshoot of the NAACP that was founded to offer legal services to poor

08-S3453-END 1/11/06 3:06 PM Page 192

192

Notes to Chapter 4

blacks. Among the cases it argued was Brown v. Board of Education (347 U.S. 483, 1954), in which the Supreme Court found that separate is inherently unequal— a ruling that changed the lives of all minority people in the United States. 5. In 2003, the INS was made part of the Department of Homeland Security and was renamed U.S. Citizenship and Immigration Services (USCIS). 6. The percentages described throughout this discussion do not sum to 100 percent. Polinard, Wrinkle, and de la Garza do not report a no-answer category. 7. The authors also found that first-generation immigrants were more likely to support amnesty than were their children or grandchildren. 8. In an earlier study, de la Garza and his colleagues (1993) examine the views on legal immigration in a sample of 1,746 Mexican, 682 Cuban, and 589 Puerto Rican individuals. A majority of Mexicans supported giving immigration preference to immigrants from Latin America, but Cubans and Puerto Ricans did not. Mexicans and Puerto Ricans were more likely than Anglos to agree that too many immigrants were entering the country. 9. Vilma S. Martinez, memorandum to members of the board of directors of MALDEF on immigration policy positions, January 14, 1981 (MALDEF Records, 1968 –1983, RG7, Box 35, Stanford University Library Special Collections, Stanford, Calif.). 10. The rationale for NCLR’s later position against sanctions is summarized in Kamasaki and Escutia (1984). 11. Here I use the term foreign-worker program interchangeably with guestworker program. Both terms refer to programs that import foreign employees on temporary contracts; when those contracts expire, the workers are required to return to their homeland unless specific provisions permit them to stay. 12. Rick Swartz (former head of the National Immigration Forum), interview with the author, September 15, 1994, Washington, D.C.; and Charles Kamasaki, interview with the author, September 7, 2000, Washington, D.C. 13. The description of the National Hispanic Task Force and its prediction about farmworker programs are taken from the Martinez memorandum to the board of directors of MALDEF, January 14, 1981. 14. Testimony before the Senate Committee on the Judiciary, Subcommittee on Immigration and Refugee Policy, and the House Committee on the Judiciary, Subcommittee on Immigration, Refugees, and International Law, Joint Hearings on the Final Report of the Select Commission on Immigration and Refugee Policy, 97th Cong., 1st sess., May 6, 1981. 15. Eligibility for permanent-resident status did not go into effect until 1989. The law made an exception to the three-year residency rule for foreign laborers who worked at least ninety days in agriculture between May 1985 and May 1986. These workers became eligible for temporary-resident status immediately and for permanent-resident status two years after receiving their temporary status or three years after the bill’s enactment, whichever was later. 16. By October 2005, just twenty nations—the minimum required—had ratified the convention. Most were migrant-sending countries; no Western migrant-receiving country has ratified it.

08-S3453-END 1/11/06 3:06 PM Page 193

Notes to Chapter 4

193

17. The Texas Proviso stated that hiring undocumented aliens did not constitute “harboring” persons unlawfully in the United States; its inclusion in the bill was considered a victory for growers who used illegal immigrant labor. See Reimers (1992, 51–52). 18. The ACTWU was formed from a merger of the Amalgamated Clothing Workers of America (ACWA) and the Textile Workers Union of America (TWUA) in 1976. The ACWA was the active proponent of proimmigration policies in the 1920s. 19. The ILGWU and the ACTWU merged in 1995 to form UNITE; UNITE and the Hotel and Restaurant Employees merged to form UNITE HERE in 2004. 20. In 2001, the United Brotherhood of Carpenters and Joiners of America withdrew from the AFL-CIO because of disagreements over how the federation was prioritizing its use of funds. In 2005, four unions that had led the labor movement’s campaigns for immigrants’ rights—UNITE HERE, the SEIU, the International Brotherhood of Teamsters, and the United Food and Commercial Workers International Union—broke with the AFL-CIO because they believed it was not devoting enough resources to wide-scale organizing. All are part of the Change to Win coalition, which formed in 2004 to press for reforms in the AFL-CIO. At the time of this writing, the coalition also included the United Brotherhood of Carpenters and Joiners of America, the Laborers International Union of North America, and the United Farmworkers. 21. House Committee on the Judiciary, Subcommittee on Immigration, Refugees, and International Law, Application of Employer Sanctions to Longshore Work: Hearing on HR 2138, 101st Cong., 1st sess., June 14, 1989. 22. Labor force statistics published by the Department of Labor do not distinguish between foreign-born and native workers. 23. As Hispanic immigrant workers increased their numbers in the construction industry during the 1990s, their influence in unions of construction workers grew. In 1992, about 4,000 Hispanic drywall workers went on a wildcat strike in Southern California. Two-thirds of these workers would later join the Carpenters and Joiners (Milkman and Wong 2000), which would eventually reverse its stand in support of employer sanctions. 24. Information on union positions was collected from testimony at congressional committee hearings on immigration from 1970 through 1992; AFL-CIO convention proceedings during those years; union journals; a series of openended interviews with union officials in Los Angeles and, by telephone, in New York during 1993; and the transcript of a hearing in which unions testified on the employer-sanctions law to the Los Angeles County Federation of Labor in 1991. 25. Hiring procedures refers to hiring halls or a union role in furnishing candidates for employment (Bureau of National Affairs 1983, 7). 26. The bureau also identified the printing, retail, and service industries as industries that regularly make hiring procedures a component of their contracts with employers. 27. Application of Employer Sanctions to Longshore Work. 28. At the AFL-CIO’s convention in 1983, as an alternative to the ILGWU’s

08-S3453-END 1/11/06 3:06 PM Page 194

194

Notes to Chapter 4

resolution, opposing sanctions, the SEIU introduced a resolution supporting sanctions “only if” they effectively deter unscrupulous employers. 29. Prepared statement of Sol Chaikin before the House Committee on Education and Labor, Hearing on Employment Discrimination and Immigration Reform, 98th Cong., 1st sess., May 19, 1983, 243. 30. Testimony before the House Committee on the Judiciary, Subcommittee on Immigration, Refugees, and International Law, and the House Committee on Education and Labor, Immigration Task Force, Immigration Act of 1989: Joint Hearings, part 3, 101st Cong., 2d sess., February 21, 1990, and March 1, 7, 13, and 14, 1990, 190. (This hearing addressed what was later called the Immigration Act of 1990.) 31. House Committee on the Judiciary, Subcommittee on Immigration, Refugees, and International Law, IRCA Anti-Discrimination Amendments Act of 1990: Hearing, 101st Cong., 2d sess., June 27, 1990, 432 – 444. 32. Ron Blackwell (assistant to the president of ACTWU), phone interview with the author, August 25, 1993. 33. This information on the SEIU is from the GAO (1988). 34. In 1979, the Retail Clerks International Union and the Amalgamated Meatcutters Union merged to form the UFCW. 35. The information here and to the end of this section relies on Martin (1993). The UFW’s position on guest workers can be found in House Committee on Agriculture, Hearing on HR 1510, 98th Cong., 1st sess., June 15, 1983, 65 – 81. 36. Although it is not possible to get accurate numbers of illegal members, most scholars agree that the UFW membership includes a substantial proportion. See, for example, Martin, Vaupel, and Egan (1988). 37. The provisions of the program can be found in Title III of the IRCA. 38. Swartz, interview. 39. Marisa Demeo (legislative lobbyist for MALDEF), interview with the author, September 11, 2000, Washington, D.C.; and Charles Kamasaki, interview with the author, September 16, 1994, Washington, D.C. chapter 5 1. The commission was chaired by Father Theodore Hesburgh, a staunch supporter of civil rights, who served as president of Notre Dame from 1952 to 1987. 2. House Committee on Education and Labor, Hearing on Employment Discrimination and Immigration Reform, 98th Cong., 1st sess., May 19, 1983, 237–242. 3. According to Rick Swartz, former head of the National Immigration Forum, Hispanic legislators and lawmakers representing affected districts wanted to delete the employer-sanctions provision altogether. In turn, some conservatives wanted to strike the amnesty provision. Legislative strategists on both sides of the IRCA controversy measured the strength of their coalition by the votes on these two amendments. Either group could have killed the delicate compromise between sanctions and amnesty that was keeping the IRCA afloat (interview with the author, September 15, 1994, Washington, D.C.).

08-S3453-END 1/11/06 3:06 PM Page 195

Notes to Chapter 5

195

4. In a 1988 version of the Kennedy-Simpson Bill, the cap was 590,000, not counting refugees; in 1989, the Senate passed another version of the bill that set a flexible ceiling of 630,000 immigrants not counting refugees. 5. Tichenor (2002) described their “guerilla tactics” (260). 6. O’Neill’s decision was reported in Congressional Quarterly Weekly Report 42 (1984), 1456 –1462. The story was also recounted to the author in an interview with Harris Miller (an aide to Representative Romano Mazzoli during the IRCA deliberations), on June 17, 1996, in Washington, D.C. 7. The same situation would not recur under a Republican Congress until a decade later, when similar proposals to control illegal immigration and start a guest-worker program were debated. 8. The H-2 program imported workers to harvest sugarcane in Florida and various other crops in eastern states. Western growers did not want to adhere to H-2 regulations, which required employers to provide housing for seasonal foreign workers. 9. By Congressional Quarterly’s definition, the conservative coalition is not organized or distinguished by ideological ratings; rather it is a voting alliance that manifests itself when a majority of Republicans and Southern Democrats vote against a majority of Northern Democrats. 10. Such divisions were clearest in the southwestern border states of Arizona, California, New Mexico, and Texas; in these states members voted 15 for and 64 against the bill. The Hispanic caucus members’ strength was also centered in the border states. 11. Workforce 2000 (1987) was highly influential. In it, Johnston and Packer predicted that the nation’s workforce in 2000 would be knowledge-based, that workers in service industries would need higher skill levels than in the past. 12. By the time the Immigration Act of 1990 was passed, other special categories of visas had been added to its provisions. For example, international events led the George H. Bush administration to press for special treatment for Chinese students who might be persecuted if they returned to their homeland after the Chinese government’s crackdown on prodemocracy forces in 1989. 13. Rick Swartz (in the September 15, 1994, interview with the author) reported that Bush was persuaded by two kinds of arguments. One was the potential of immigration to aid in economic growth; the other was the opportunity to woo Hispanic and Asian voters. 14. Congressional Quarterly Almanac 45 (1989): 267. 15. Cecilia Muñoz, interview with the author, September 16, 1994, Washington, D.C. 16. Charles Kamasaki (senior vice president of NCLR), interview with the author, September 16, 1994, Washington, D.C. 17. With a bill yet to be decided, Congress approved a special measure in 1988 that allowed more Irish immigrants to enter the country and also extended the visas of foreign nurses. President Reagan signed HR 5115 (PL 100-658) on November 15, 1988. The extension for foreign nurses was a response to hospitals’ claims of a widespread shortage of American nurses.

08-S3453-END 1/11/06 3:06 PM Page 196

196

Notes to Chapter 5

18. Ethnic rights and church lobbies worked closely with Representative Berman, who at the committee stage of deliberations had offered up a separate bill with Senator Paul Simon (D-IL) as an alternative to the Kennedy-Simpson Bill. The Simon-Berman bill did not include points for English proficiency. 19. The coalition included the Asian American Legal Defense and Education Fund, the Asian Law Alliance, the Asian Law Caucus, the Asian Pacific American Legal Center of Southern California, Na Loio No Na Kanaka—Lawyers for the People of Hawaii, and Nihomachi Legal Outreach. 20. House Committee on the Judiciary, Subcommittee on Immigration, Refugees, and International Law, Immigration Act of 1989 (part 1), Hearings on S 358, HR 2448, and HR 2646, 101st Cong., 1st sess., September 27, 1989, 589 – 620. 21. Muñoz, interview. 22. The 1990 immigration reform would establish a numerical cap that was “pierceable” by unrestricted admission of certain immediate relatives of citizens and other categories. 23. The head tax provision would have required large employers (200 or more workers) to pay a tax of $1,000 for every foreign worker, and smaller employers (50 to 199 workers) to pay a tax of $500 for every foreign worker. The proceeds were to be used for a training fund for U.S. workers. 24. Morrison’s original proposal in subcommittee would have required employers to pay newly hired immigrants 105 percent of the prevailing wage; in addition, they would have been required to pay a tax amounting to 15 percent of each foreign worker’s wages into a job-training fund for U.S. workers. These proposals would be resurrected again in the recommendations of the U.S. Commission on Immigration Reform, headed by former Congresswoman Barbara Jordan (D-TX), which issued a report to Congress in 1995. Morrison was a commissioner. 25. The floor amendment would not have imposed a tax on employers; it would have required employers who hire foreign workers to certify that they provided skills training or tutoring in schools. After it was defeated, a provision was passed by voice vote on the method by which employers must certify a labor shortage before bringing in foreign workers. They could follow the existing labor certification process, administered by the Department of Labor, or they could attest that they had tried to recruit U.S. workers and then give domestic workers thirty days to lodge a protest. 26. Other provisions tightening the laws on hiring foreign workers were included in the final version of the bill. For example, the bill prohibited domestic airlines and shipping companies from bringing in temporary foreign workers to perform services on a U.S. aircraft or vessel during labor disputes. 27. Another concession was granted to Joe Moakley (D-MA), chair of the House Rules Committee, who used his influential position to obtain a stay of deportation and special “temporary protected status” for Salvadorans. In addition, persons fleeing countries that were engaged in armed conflict or that had undergone a natural disaster would be allowed to apply for a stay of deportation. 28. According to Michael Hill, a lobbyist for the United States Conference of Catholic Bishops, one simple indicator that ethnic rights and church lobbyists

08-S3453-END 1/11/06 3:06 PM Page 197

Notes to Chapter 6

197

use to predict the openness of a legislator to proimmigration arguments is the percentage of foreign-born or Hispanic constituents in his or her district. Lobbyists also assess the lawmaker’s voting record. Hill spoke with the author on June 6, 1996, in Washington, D.C. 29. This proposal to cap legal immigration occurred before the portions of the Simpson-Mazzoli Bill pertaining to legal immigration were removed from the bill. 30. Information on unemployment levels in congressional districts is available only every ten years (it is collected as part of the decennial census). To estimate unemployment at middecade points at the congressional-district level, I used Bureau of Labor Statistics estimates of annual or sometimes monthly unemployment data at the state or metropolitan-area level (see notes for Table 5.3). 31. In her study of immigration politics in Congress from 1890 to 1921, Goldin finds that as the percentage of foreign-born persons in a district increases, so does the probability of a legislator’s voting liberally on immigration admissions. 32. As I note in the summary of the results of testing the unemployment variable in 1990, the effect of the unemployment variable must be interpreted with caution. Assigning the same state-level value to all districts in the state is problematic because statistically significant results may reflect fixed effects equivalent to those that we would achieve by adding a dummy variable for each state. 33. The IRCA’s final form mixed expansionary and restrictive measures. 34. As with the IRCA votes, when the interval was extended to four or five years, the findings were similarly negative on the effects of unemployment change. Substituting point estimates of unemployment rates in the state rather than using a measure of unemployment change yielded no significantly different results. chapter 6 1. Simpson, who led the Senate effort to pass the restrictive legislation, initially thought that legal- and illegal-immigration issues should be separated and, accordingly, introduced two bills. But he agreed to join the two pieces of legislation so that a legislative strategy could be coordinated in the House and Senate. 2. Battered women and children are exempt. In addition, certain periods are exempt from the time count, including time spent in the United States before age 18 and the time during which an asylum application is pending. 3. The new level of earnings is required for both the sponsor’s family plus the alien’s family. 4. According to Steve Moore, a former senior fellow at the Cato Institute, the institute published policy analyses—see, for example, Simon (1995)—that argued that immigration was not a “zero-sum game,” in which each new immigrant took away the job of an American, but, on balance, a source of new employment and growth. Moore, who helped draft the Kemp-Bennett proposal for a proimmigration plank in the Republican platform in 1994, spoke with the author in an interview on September 17, 1994, in Washington, D.C.

08-S3453-END 1/11/06 3:06 PM Page 198

198

Notes to Chapter 6

5. Here, too, there had been indications that the Republicans were not united on the issue of cutbacks in legal immigration. When the Contract with America was being drafted before the 1992 elections, the immigration issue was removed. 6. Phyllis Eisen, a coordinator of the NAM-led coalition, reported that the companies with international concerns included computer companies in the main but also several pharmaceuticals, a large automaker, and others. She spoke in an interview with the author on September 30, 1996, in Washington, D.C. 7. Eisen, interview. 8. NAM issued an alert to signal its support for splitting the bill (a copy is in the author’s possession). In the September 30, 1996, interview, Eisen recounted the shift in the business coalition’s and NAM’s assessment between the markup in the House Judiciary Committee and the House floor vote. 9. UNITE was formed in 1995, the product of the merger of the International Ladies Garment Workers Union and the Amalgamated Clothing and Textile Workers Union. 10. Ray Kehtledge (counsel to Spencer Abraham), interview with the author, July 26, 1996, Washington, D.C.; Paul Ryan (aide to Sam Brownback), interview with the author, July 22, 1996, Washington, D.C.; and Fred Nelson (aide to Representative Steve Chabot, R-OH), interview with the author, June 21, 1996, Washington, D.C. 11. The representatives were Michael Hill from the United States Conference of Catholic Bishops and Rick Swartz, a former head of the National Immigration Forum and now an independent consultant (Michael Hill, interview with the author, June 6, 1996, Washington, D.C.). 12. Rick Swartz, interview with the author, July 1, 1996, Washington, D.C. 13. Under the H-1B program, employers can hire foreign workers to work in the United States on a temporary (nonimmigrant) basis in a specialty occupation. A specialty occupation requires the application of a body of specialized knowledge and a bachelor’s degree or the equivalent in the specific specialty (e.g., science, education, medicine and health care, or business). The AFL-CIO has two principal objections to expansion of the H-1B program. It argues that the program, instead of filling spot shortages, creates a surplus of workers. In addition, it claims that the Department of Labor uses inaccurate methods to ensure that H-1B workers do not have an adverse effect on prevailing wages. 14. In his book The Freedom Revolution (1995), in a chapter on free trade and immigration, Armey writes: “Holding to the view that legal immigrants are a benefit to our economy and society in general, we should under no circumstances reduce the current quotas for legal immigration. We can, however, reform the way visas are distributed. Some should be given to people fleeing tyranny abroad or coming from newly liberated areas such as the former Communist nations. Others can be handed out by lottery or given to family members of American citizens” (267). 15. The description of the Republican conference discussions on immigration is based on the author’s interview with Paul Ryan. 16. The Clinton administration responded to the political pressure in California for tightened federal enforcement of immigration laws with in-

08-S3453-END 1/11/06 3:06 PM Page 199

Notes to Chapter 6

199

creased funding for border patrols. However, President Clinton decided not to introduce a bill on illegal immigration in the 103d Congress. According to Roberto Suro, a reporter for the Washington Post, this was a decision with which the Democratic leadership concurred. The leadership’s thinking was that it could not control the outcome of an immigration bill, that the members would be “all over the map” (Roberto Suro, interview with the author, July 23, 1996, Washington, D.C.). 17. Alan Simpson, interview with the author, June 12, 1995, Washington, D.C. 18. The business groups had scheduled a national lobbying day in Washington one day before the Senate markup session was scheduled to open. According to Simpson’s remarks on March 28, 1996, when the Judiciary Committee began its markup deliberations, he had met with approximately ninety lobbyists the day before (author’s notes). 19. On the morning of the Judiciary Committee markup, the business coalition sent a letter to Spencer Abraham indicating that it would continue to support the Kennedy-Abraham Amendment to split the bill, in spite of Simpson’s concessions. Abraham’s counsel reported to the author that the business interests realized Simpson could easily reintroduce the employment provisions on the floor of the Senate (Kehtledge, interview). 20. Kehtledge, interview. 21. According to Kehtledge (interview), Phil Gramm’s final speech before the floor vote helped turn this bloc of votes. Gramm, from Texas, was representative of Republicans who were both socially conservative and oriented toward a free-market view of immigration. 22. Actually, organized labor did not resist employment-based immigration reform as a matter of course. Its primary concern in debates on the IIRIRA was the effect of temporary foreign workers on domestic workers. 23. Data for congressional districts are not readily available. Data for counties are available for the mid-1990s (published in annual series), but not for the mid-1980s. 24. Money (1999) discusses the politics of immigration policymaking in a two-dimensional policy space, where immigration control and integration are the two principal dimensions. See also Money (1995). 25. The concept of issue definition is described in Baumgartner and Jones (1993). An important theoretical question raised by this theory of multidimensionality is why endless realignments of voting coalitions do not occur. A clear constraint on such cycling of alternatives are rules limiting the number or type of amendments that can be introduced. In the case of the IIRIRA, the Republican leadership on the Rules Committee did not issue an open rule for amendments. By not issuing an open rule, the committee was prohibiting endless cycling. 26. I derived the main issue content of each House amendment from numbered summaries in Congressional Quarterly Weekly Report (March 23, 1996, 834 – 840). I did not include in the analysis (or in Table 6.1) the Burr Amendment, which proposed a six-month extension of a program granting temporary visas to foreign nurses. 27. The results show the influence of a concentration of Hispanic people in a

08-S3453-END 1/11/06 3:06 PM Page 200

200

Notes to Chapter 6

congressional district. The results are similar when the variable for the percentage of Hispanics in the district is replaced with the percentage of foreign-born. 28. In a study of county-by-county voting in California on Proposition 187, Mac Donald and Cain (1998) found that higher-income counties voted more liberally than did lower-income counties; that is, they tended to vote against Proposition 187. In addition, an increase in the number of college-educated persons in a county increased the vote against Proposition 187. The authors offer two possible explanations for this effect of education: Those with higher levels of education may feel “less materially threatened” by an influx of immigrants; or higher levels of education may produce “more tolerant and less nativist racial and cultural attitudes (or perhaps both)” (294). Their study did not allow them to distinguish between these two possible causes. 29. Important variations in labor-market conditions are not represented in the available data for congressional districts. For example, the labor of highsalaried workers may complement that of low-salaried immigrants; or the labor of low-salaried native and immigrant workers may substitute for each other. Furthermore, labor-market incorporation of immigrants varies by locality, but the boundaries of congressional districts do not necessarily correspond to those of local markets. 30. It should be pointed out that there is a problem of autocorrelation that arises from using ideology scores derived from legislators’ voting records. The estimated coefficients may be biased upward. Furthermore, the interpretation of a unit change in the ideology index is usually not straightforward. Thus, I replace the ideology (D-nominate) score with a party variable (PARTY) in Model 2. The robustness of estimated coefficients may be compared across columns in Table 6.2, which reports results for the alternative models. 31. Poole and Rosenthal argue that the ideological index (the dimension-1 score), can statistically explain broad patterns of congressional voting over a 200-year span. Although they identify a much less prominent dimension that divides parties internally, scores measuring dimension 2 since the 1980s do not have a clear interpretation and probably represent statistical noise. Before that time, in the immediate postwar and civil rights period, the dimension-2 scores represent divisions inside the Democratic Party over race. 32. In separate results, when I include variables for PAC contributions from building-trade unions and from industrial unions in the pooled regressions for the general-enforcement and enforcement-and-privacy categories, the PAC variables do not produce statistically significant estimates. 33. The control for average income in a district from wages or salary gives statistically significant results, but the magnitude of the coefficient is negligible. Tests of the effects of other variables are available from the author. For example, the proportion of blacks residing in a district is not an important predictor of a legislator’s vote, even though concentrated black communities have a higher than average share of unskilled workers. Although immigrant competition in labor markets probably affects low-skilled workers more than high-skilled workers (Borjas, Freeman, and Katz 1992, 212 –244), many representatives of largely black districts take ideologically liberal stands on immigration (Fuchs

08-S3453-END 1/11/06 3:06 PM Page 201

Notes to Chapter 7

201

1990, 293 – 314). Asian concentrations in a district are generally not large enough to have a statistically significant effect except in a few agricultural districts in the San Joaquin Valley in California, where Japanese American farmers have lobbied for liberal guest-worker policies. 34. These fitted probabilities are estimated using the logit regression coefficients reported in Table 6.2. The $1,426 and $15,821 increases are approximately the size of 1 standard deviation, after reversing the log transformation of the data. 35. Building-trade and industrial unions have sometimes taken opposing stands on immigration (see Chapter 4). Because building-trade unions are organized as craft unions, they rely on restricting access to labor markets; hence, they have been more restrictionist at times than industrial unions, which rely on industrywide organizing. 36. Results of tests for curvilinear relationships are available from the author. 37. I thank Doug Rivers for suggesting the method used in this section for testing the dissimilarity of issues. Any errors in the analysis are my own. appendix: alternate models 1. Paul Ryan (aide to Sam Brownback), interview with the author, July 22, 1996, Washington, D.C. 2. This amendment was offered by Representative Richard Pombo (R-CA) for the House Agriculture Committee. 3. The position of the libertarian Cato Institute, a major think tank that influenced the immigration debate in the mid-1990s, suggests that economic conservatives can support guest-worker programs. 4. Except in the case where six variables were added to the model representing specific ethnic groups that overlap with the foreign-born variable, the proportion of a district’s population that is foreign-born remained statistically significant. chapter 7 1. The arguments of the Select Commission on U.S. Immigration and Refugee Policy (1981) are still pertinent to contemporary debate. 2. Even if businesses that presently employ undocumented immigrants wanted to alter their employment practices to accord with federal immigration laws, there would be no reliable way to distinguish authorized workers from others. 3. The UFW did not win its demand that H-2A workers have full coverage under the federal Migrant and Seasonal Agricultural Worker Protection Act (AWPA), which was passed in 1983. The AWPA excludes H-2A workers. 4. In addition, employers could provide a housing allowance to workers instead of providing housing facilities. The wage freeze accommodated the demands of eastern growers, while the option of providing a housing allowance rather than physical housing met the demands of western growers. Because Cal-

08-S3453-END 1/11/06 3:06 PM Page 202

202

Notes to Chapter 7

ifornia, Oregon, and Washington set relatively high minimum wages (higher than the federal rate), western growers already would have to pay wages under those constraints; however, constructing housing would be costly. Many eastern growers had already constructed housing for H-2A workers; their concerns about costs focused on wages. 5. The legislation would have given temporary work permits to undocumented workers who had worked at least 100 days in agriculture in the eighteen months prior to December 31, 2004. After working in agriculture for at least 360 days over six years, they could apply for permanent residency in the United States. 6. Several categories of nonimmigrant aliens would explicitly not be covered by the new visa program: aliens entering under the H-2A (agricultural) and H-1B (professional specialty occupations), L (executive, managerial), O (athletes, artists, and entertainers), P (aliens with extraordinary abilities in science, arts, education, business, or athletics), and R (religious workers) programs. 7. Portability is not a feature of the AgJobs proposal. AgJobs workers obtain their visa from their employer and must stay with that employer. However, AgJobs would give workers the right to enforce their labor contracts in federal court. 8. Under the bill proposed by McCain and Kennedy, H-5A workers would be paid the same wage as provided by the U.S. employer to workers similarly employed the same occupation at the same place of employment. This proposed requirement differs somewhat from the definition of the prevailing wage used by the Department of Labor. The prevailing wage is the average wage paid to similarly employed workers in the occupation for which the foreign worker is requested and in the area of intended employment. Prevailing wage rates differ from adverse-effect wage rates (AEWRs), which are the minimum wage rates that the Department of Labor has determined must be offered and paid to U.S. and foreign workers (H-2A visa holders) so as not to adversely affect U.S. workers. Under current law, employers of H-2A workers must pay the higher of the AEWR, the applicable prevailing wage, or the statutory minimum wage. H-2B workers must be paid the prevailing wage. H1-B workers must be paid the higher of the prevailing wages or the actual wage paid by the firm to workers with similar skills and qualifications. 9. If the cap was reached in the first three months of the fiscal year, an additional 80,000 visas would be made available, and the cap for the next year would increase to 480,000; if the cap was hit in the second quarter, an additional 60,000 visas would be made available, and the cap for the next year would be 460,000. If the cap was hit in the third quarter, 40,000 more visas would be available and the cap increased to 440,000 in the next year. If the cap was reached in the last quarter, 440,000 visas would be available the next year. Other provisions allow for adjustments up and down according to labor-market conditions. 10. This was not the first bill of its kind, but rights groups supported it more strongly than they had previous bills. For example, the National Council of La Raza (2005) issued a press release “applauding” the McCain-Kennedy Bill. The

08-S3453-END 1/11/06 3:06 PM Page 203

Notes to Chapter 7

203

previous year, the group had issued a much more reserved statement on S 2010, a bill introduced by Senators Chuck Hagel (R-NE) and Tom Daschle (D-SD). According to the NCLR, the Hagel-Daschle Bill was a “constructive step forward” in providing permanent legal status but needed to “expand its labor protections considerably.” Rights groups opposed a competing bill introduced in July 2005 by Senators John Cornyn (R-TX) and Jon Kyl (R-AZ), the Comprehensive Enforcement and Immigration Reform Act of 2005 (S 1438). The Cornyn-Kyl Bill proposed a new portable W visa that would allow newly enrolled temporary workers to work for two years but would require them to return home for one year before becoming eligible again for another two years. For undocumented immigrants already in the United States, the bill would allow undocumented immigrants to work under the W visa for five years; after the fifth year, they would be required to leave. 11. For more details about this decision, see note 19 in Chapter 4. 12. Robinson (2003) notes a general trend among industrial states—an openness to trade but not people.

08-S3453-END 1/11/06 3:06 PM Page 204

09-S3453-BIB 1/11/06 3:07 PM Page 205

References

AFL-CIO. 1983. Proceedings of the Fifteenth AFL-CIO Constitutional Convention. Washington, D.C.: AFL-CIO. ———. 1985. Proceedings of the Sixteenth AFL-CIO Constitutional Convention. Washington, D.C.: AFL-CIO. ———. 1987. Proceedings of the Seventeenth AFL-CIO Constitutional Convention. Washington, D.C.: AFL-CIO. Armey, Dick. 1995. The Freedom Revolution. Washington, D.C.: Regnery Press. Bailey, Thomas. 1985. “A Case Study of Immigrants in the Restaurant Industry.” Industrial Relations 24, no. 2 (spring): 205 –221. Baumgartner, Frank R., Jeffrey M. Berry, Marie Hojnacki, David Kimball, and Beth L. Leech. 2001. “Lobbying and Issue Definition.” Grant proposal, Grant SBR-0111224. National Science Foundation, Arlington, Va. Baumgartner, Frank R., and Bryan D. Jones. 1993. Agendas and Instability in American Politics. Chicago: University of Chicago Press. Baumgartner, Frank R., and Beth L. Leech. 1998. Basic Interests: The Importance of Groups in Politics and Political Science. Princeton, N.J.: Princeton University Press. Berry, Jeffrey M. 1999. The New Liberalism. Washington, D.C.: Brookings Institution Press. Berry, Jeffrey M., and David F. Arons. 2003. A Voice for Nonprofits. Washington, D.C.: Brookings Institution Press. Binder, Norman, J. L. Polinard, and Robert Wrinkle. 1997. “Mexican American and Anglo Attitudes Toward Immigration Reform: A View from the Border.” Social Science Quarterly 78 (2): 325. Birnbaum, Jeffrey H. 1993. Lobbyists: How Influence Peddlers Work Their Way in Washington. New York: Times Books. Blumer, Herbert. 1958. “Race Prejudice as a Sense of Group Position.” Pacific Sociological Review 1: 3 –7.

09-S3453-BIB 1/11/06 3:07 PM Page 206

206

References

Bobo, Lawrence. 2004. “Inequalities That Endure? Racial Ideology, American Politics, and the Peculiar Role of Social Science.” In Changing Terrain of Race and Ethnicity, edited by M. Krysan, A. Lewis, and T. Forman, 13 – 42. New York: Russell Sage Foundation. Bobo, Lawrence, and V. Hutchings. 1996. “Perceptions of Racial Group Competition: Extending Blumer’s Theory of Group Position to a Multiracial Social Context.” American Sociological Review 61, no. 6 (December): 951–972. Boorstin, Daniel. 1973. The Americans: The Democratic Experience. New York: Random House. Borjas, George J. 1994. “The Economics of Immigration.” Journal of Economic Literature 32 (December): 1667–1717. Borjas, George J., Richard Freeman, and Lawrence E. Katz. 1992. “On the Labor Market Effects of Immigration and Trade.” In Immigration and the Workforce, edited by George Borjas and Richard B. Freeman, 213 –244. Chicago: University of Chicago Press. Briggs, Vernon M., Jr. 1996. Mass Immigration and the National Interest. 2d ed. New York: M. E. Sharpe. Brubaker, Rogers. 1995. “Comments on ‘Modes of Immigration Politics in Liberal Democratic States.’” International Migration Review 29, no. 4 (winter): 903 –908. ———. 2004. Ethnicity Without Groups. Cambridge, MA.: Harvard University Press. Bureau of National Affairs. 1983. Basic Patterns in Union Contracts. Washington, D.C.: Bureau of National Affairs. Burns, Peter, and James G. Gimpel. 2000. “Economic Insecurity, Prejudicial Stereotypes, and Public Opinion on Immigration Policy.” Political Science Quarterly 115, no. 2 (summer): 201–225. Calavita, Kitty. 1992. Inside the State: The Bracero Program, Immigration, and the I.N.S. New York: Routledge. Camarillo, Albert. 1991. “Mexican-American and Nonprofit Organizations: An Historical Overview.” In Hispanics and the Nonprofit Sector, edited by Herman E. Gallegos and Michael O’Neill, 15 – 32. New York: Foundation Center. Cameron, Charles, David Epstein, and Sharyn O’Halloran. 1996. “Do MajorityMinority Districts Maximize Substantive Black Representation in Congress?” American Political Science Review 90, no. 4 (December): 794 – 812. Card, David. 1990. “The Impact of the Mariel Boatlift on the Miami Labor Market.” Industrial and Labor Relations Review 43 (January): 245 –257. Carens, Joseph H. 1991. “Migration and Morality: A Liberal Egalitarian Perspective.” In Free Movement: Ethical Issues in the Transnational Migration of People and of Money, edited by Brian Barry and Robert E. Goodin, 25 – 47. University Park: Pennsylvania State University Press. Carter, Susan B., and Richard Sutch. 1998. “Historical Background to Current Immigration Issues.” In The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration, edited by James P. Smith and Barry Edmonston, 289 – 366. Washington, D.C.: National Academy Press. Citrin, Jack, Donald P. Green, Christopher Muste, and Cara Wong. 1997. “Pub-

09-S3453-BIB 1/11/06 3:07 PM Page 207

References

207

lic Opinion Toward Immigration Reform: The Role of Economic Motivations.” Journal of Politics 59 (3): 858 – 881. Cohen, Cathy J., and Michael C. Dawson. 1994. “Neighborhood Poverty and African American Politics.” American Political Science Review 87: 286 – 302. Cooper, Ann. 1986. “John Tanton: Immigration Reformer Stirs the Melting Pot.” National Journal, May 17, 1210. Cornelius, Wayne A., Philip L. Martin, and James F. Hollifield. 1992. “Introduction: The Ambivalent Quest for Immigration Control.” In Controlling Immigration: A Global Perspective, edited by Wayne A. Cornelius, Philip L. Martin, and James F. Hollifield, 3 – 41. Stanford, Calif.: Stanford University Press. Cornelius, Wayne A., and Marc R. Rosenblum. 2005. “Immigration and Politics.” Annual Review of Political Science 8: 99 –119. Cornelius, Wayne A., and Takeyuki Tsuda. 2004. “Controlling Immigration: The Limits of Government Intervention.” In Controlling Immigration: A Global Perspective, 2d ed., edited by Wayne A. Cornelius, Takeyuki Tsuda, Philip L. Martin, and James F. Hollifield, 3 – 48. Stanford, Calif.: Stanford University Press. Cortes, Michael E. 1992. “Policy Analysis and Interest Groups: The Case of Immigration Reform.” PhD diss., University of California, Berkeley. Craig, Richard. 1971. The Bracero Program: Interest Groups and Foreign Policy. Austin: University of Texas Press. de la Garza, Rodolfo O., and Louis DeSipio. 1999. “Awash in the Mainstream: Latinos and the 1996 Elections.” In Awash in the Mainstream: Latino Politics in the 1996 Election, edited by Rodolfo O. de la Garza and Louis DeSipio, 3 – 46. Boulder, Colo.: Westview Press. ———. 2005. “Between Symbolism and Influence.” In Muted Voices: Latinos and the 2000 Elections, edited by Rodolfo O. de la Garza and Louis DeSipio, 13 – 60. Lanham, Md.: Rowman & Littlefield. de la Garza, Rodolfo O., A. Falcon, F. C. Garcia, and J. Garcia. 1993. “Attitudes Toward U.S. Immigration Policy.” Migration World 21: 13 –16. DeSipio, Louis. 1996. Counting on the Latino Vote: Latinos as a New Electorate. Charlottesville: University of Virginia Press. Diamond, Jeff. 1998. “African American Attitudes Toward United States Immigration Policy.” International Migration Review 32 (2): 451– 470. Fitzgerald, Keith. 1996. The Face of the Nation: Immigration, the State, and the National Identity. Stanford, Calif.: Stanford University Press. Fraga, Luis R., and David L. Leal. 2004. “Playing the ‘Latino Card’: Race, Ethnicity, and National Party Politics.” Du Bois Review: Social Science Research on Race 1: 297– 317. Freeman, Gary. 1995a. “Modes of Incorporation Politics in Liberal Democratic States.” International Migration Review 29 (4): 891–902. ———. 1995b. “Rejoinder.” International Migration Review 29 (4): 909 –913. Fresno Bee. 2005. “AgJobs Bill Falls Short in Senate.” April 20. Fuchs, Lawrence H. 1990. “The Reactions of Blacks to Immigration.” In Immigration Reconsidered, edited by Virginia Yan-McLaughlin, 293 – 314. Oxford, Eng.: Oxford University Press.

09-S3453-BIB 1/11/06 3:07 PM Page 208

208

References

Gais, Thomas L., and Jack L. Walker Jr. 1991. “Pathways to Influence in American Politics.” In Mobilizing Interest Groups in America: Patrons, Professions, and Social Movements, edited by Jack L. Walker Jr., 103 –121. Ann Arbor: University of Michigan Press. Gimpel, James G., and James R. Edwards Jr. 1999. The Congressional Politics of Immigration Reform. Boston: Allyn & Bacon. Goldin, Claudia. 1994. “The Political Economy of Immigration Restriction in the United States, 1890 to 1921.” In The Regulated Economy: A Historical Approach to Political Economy, edited by Claudia Goldin and Gary D. Libecap, 223 –257. Chicago: University of Chicago Press. Green, James Frederick. 1956. The United Nations and Human Rights. Washington, D.C.: Brookings Institution Press. Gutmann, Amy. 2003. Identity in Democracy. Princeton, N.J.: Princeton University Press. Hammar, Thomas. 1985. European Immigration Policy: A Comparative Study. Cambridge, Eng.: Cambridge University Press. Hatton, Timothy J., and Jeffrey G. Williamson. 1994. “International Migration 1850 –1939: An Economic Survey.” In Migration and the International Labor Market 1850 –1939, edited by Timothy J. Hatton and Jeffrey G. Williamson, 3 – 32. London: Routledge. ———. 1995. “The Impact of Immigration on American Labor Markets Prior to the Quotas.” Working Paper 5185, National Bureau of Economic Research, Washington, D.C. Haus, L. 2002. Unions, Immigration, and Internationalization: New Challenges and Changing Coalitions in the United States and France. New York: Palgrave Macmillan. Higham, John. 1992. Strangers in the Land: Patterns of American Nativism, 1860 – 1925. 2d ed. New Brunswick, N.J.: Rutgers University Press. Hing, Bill Ong. 1993. Making and Remaking Asian America Through Immigration Policy, 1850 –1990. Stanford, Calif: Stanford University Press. Hojnacki, Marie. 1988. “Organized Interests’ Advocacy Behavior in Alliance.” Political Research Quarterly 51: 437– 459. ———. 1997. “Interest Groups’ Decisions to Join Alliances or Work Alone.” American Journal of Political Science 41: 61– 87. Hojnacki, Marie, and David C. Kimball. 1998. “Organized Interests and the Decision of Whom to Lobby in Congress.” American Political Science Review 92, no. 4 (December): 775 –790. Hollifield, James. 1992. Immigrants, Markets, and States: The Political Economy of Postwar Europe. Cambridge, Mass.: Harvard University Press. Holter, Darryl, and Kent Wong. 1993. “Immigrant Worker and Labor Law Reform: The Organizer’s Perspective.” Typescript, UCLA Institute of Industrial Relations, Los Angeles. Hula, Kevin W. 1999. Lobbying Together: Interest Group Coalitions in Legislative Politics. Washington, D.C.: Georgetown University Press. Johnston, William B., and Arnold E. Packer, with contributions from Mat-

09-S3453-BIB 1/11/06 3:07 PM Page 209

References

209

thew P. Jaffe. 1987. Workforce 2000: Worker and Workers for the 21st Century. Indianapolis: Hudson Institute; Washington, D.C.: U.S. Department of Labor. Joppke, Christian. 2005. Selecting by Origin: Ethnic Migration in the Liberal State. Cambridge, Mass.: Harvard University Press. Kahn, Shulamit. 1986. “Union Membership Trends: A Study of the Garment Workers.” Monthly Labor Review 109 (6): 33 – 35. Kamasaki, Charles, and Rose Briceño. 1986. “An Overview of the Immigration Reform Act.” Perspectivas Publicas, November 25. NCLR Office of Research, Advocacy, and Legislation. Kamasaki, Charles M., and Martha M. Escutia. 1984. “Second Thoughts on Employer Sanctions: A Brief Analysis.” Perspectivas Publicas, March, 1– 8. NCLR Office of Research, Advocacy, and Legislation. Keech, William R. 1968. The Impact of Negro Voting. Chicago: Rand McNally. Kessler, Alan. 1998. “Distributional Coalitions, Trade, and the Politics of Postwar Immigration.” Paper presented at the 1998 annual meeting of the American Political Science Association, Boston. Key, V. O. 1949. Southern Politics in State and Nation. Knoxville: University of Tennessee Press. Laslett, John, and Mary Tyler. 1989. The ILGWU in Los Angeles, 1907–1988. Inglewood, Calif.: Ten Star Press. Loftus, Joseph A. 1964. “Miller Attacks Tariff Programs: Says Johnson Would Flood U.S. with Foreign Goods.” New York Times. September 8, 14. Loucky, James, Nora Hamilton, and Norma Chinchilla. 1989. “The Effects of the Immigration Reform and Control Act on the Garment, Building Maintenance, and Hospitality Industries in Los Angeles.” Final report to the Division of Immigration and Policy Research, U.S. Department of Labor, Washington, D.C. Mac Donald, Karin, and Bruce E. Cain. 1998. “Nativism, Partisanship, and Immigration: An Analysis of Proposition 187.” In Racial and Ethnic Politics in California, vol. 2, edited by M. B. Preston, B. Cain, and S. Bass, 277– 304. Berkeley, Calif.: Institute of Governmental Studies Press. Maram, Sheldon L., and J. King. 1983. “The Labor Market Impact of Hispanic Undocumented Immigrants: An Analysis of the Garment and Restaurant Industry in Los Angeles.” Report for the Rockefeller Foundation, New York. Martin, Philip L. 1987. “California’s Farm Labor Market.” UC AID Issues Paper 87-1, Agricultural Issues Center, University of California, Davis. ———. 1993. Trade and Migration: NAFTA and Agriculture. Washington, D.C.: Institute for International Economics. ———. 2003. Promise Unfulfilled: Unions, Immigration, and the Farm Workers. Ithaca, N.Y.: Cornell University Press. ———. 2004. “The United States: The Continuing Immigration Debate.” In Controlling Immigration: A Global Perspective, 2d ed., edited by Wayne A. Cornelius, Takeyuki Tsuda, Philip L. Martin, and James F. Hollifield, 51– 85. Stanford, Calif.: Stanford University Press.

09-S3453-BIB 1/11/06 3:07 PM Page 210

210

References

Martin, Philip L., Suzanne Vaupel, and Daniel L. Egan. 1988. Unfulfilled Promise: Collective Bargaining in California Agriculture. Boulder, Colo.: Westview Press. McCubbins, Mathew D., Roger C. Noll, and Barry R. Weingast. 1987. “Administrative Procedures as Instruments of Political Control.” Journal of Law, Economics and Organization 3, no. 2 (fall): 243 –277. McCurdy, Thomas, Thomas Nechyba, and Jay Bhattacharya. 1998. “An Economic Framework for Assessing the Fiscal Impacts of Immigration.” In The Immigration Debate: Studies on Economic, Demographic, and Fiscal Effects of Immigration, edited by James P. Smith and Barry Edmonston, 13 – 65. Washington, D.C.: National Academy Press. Mendelberg, Tali. 2001. The Race Card: Campaign Strategy, Implicit Messages, and the Norm of Equality. Princeton, N.J.: Princeton University Press. Mexican American Legal Defense Fund (MALDEF). 1968 –2002. Records. Stanford, Calif.: University Library Special Collections, Stanford University. Milkman, Ruth, and Kent Wong. 2000. “Organizing the Wicked City: The 1992 Southern California Drywall Strike.” In Organizing Immigrants: The Challenge for Unions in Contemporary California, edited by Ruth Milkman, 169 –198. Ithaca, N.Y.: Cornell University Press. Mines, Richard, and Philip L. Martin. 1986. A Profile of California Farmworkers. Berkeley, Calif.: Giannini Foundation. Mitchell, Christopher. 1992. “Introduction: Immigration and U.S. Foreign Policy Toward the Caribbean, Central America, and Mexico.” In Western Hemisphere Immigration and United States Foreign Policy, edited by Christopher Mitchell, 2 – 30. University Park: Pennsylvania State University Press. Money, Jeannette. 1995. “Two-Dimensional Aliens.” Paper presented at the 91st annual meeting of the American Political Science Association, Chicago. ———. 1999. Fences and Neighbors: The Political Geography of Immigration Control. Ithaca, N.Y.: Cornell University Press. Morales, Richard. 1985. “Under One Roof: Mexican Immigrants and NativeBorn Workers in the Non-Union Restaurant Industry of San Diego County.” PhD diss., University of California, Berkeley. National Asian Pacific American Legal Consortium (NAPALC). 1995. NAPALC Review 12, no. 1 (spring). National Council of La Raza (NCLR). 2004. “NCLR Statement on Hagel/ Daschle Immigration Proposal.” News release, January 22. Available at <www.nclr.org/content/news/detail/25047/> (September 30, 2005). ———. 2005. “NCLR Applauds Introduction of Bipartisan Comprehensive Immigration Bill.” News release, May 12. Available at <www.nclr.org/content/ policy/detail/1058/> (September 30, 2005) Ngai, Mae M. 1999. “The Architecture of Race in American Immigration Law: A Re-examination of the Immigration Act of 1924.” Journal of American History 86 (June): 67–92. ———. 2004. Impossible Subjects: Illegal Aliens and the Making of Modern America. Princeton, N.J.: Princeton University Press. Papademetriou, Demetrios G. 2002. “A Grand Bargain: Balancing the National Security, Economic, and Immigration Interests of the U.S. and Mexico.” Wash-

09-S3453-BIB 1/11/06 3:07 PM Page 211

References

211

ington, D.C.: Migration Policy Institute. Available at <www.migrationpolicy. org/files/bargain.pdf> (September 25, 2005) Parmet, Robert D. 1981. Labor and Immigration in Industrial America. Boston: Twayne. Polinard, Jerry, Robert Wrinkle, and Rodolfo de la Garza. 1984. “Attitudes of Mexican Americans Toward Irregular Mexican Immigration.” International Migration Review 18 (3): 782 – 899. Poole, Keith T., and Howard Rosenthal. 1997. Congress: A Political-Economic History of Roll Call Voting. New York: Oxford University Press. Rees, Albert. 1979. The Economics of Trade Unions. Chicago: University of Chicago Press. Reimers, David M. 1992. Still the Golden Door: The Third World Comes to America. 2d ed. New York: Columbia University Press. Riker, William. 1986. The Art of Political Manipulation. New Haven: Yale University Press. ———. 1988. Liberalism Against Populism. Prospect Heights, Ill.: Waveland Press. ———. 1996. The Strategy of Rhetoric. New Haven: Yale University Press. Robinson, Mary. 2003. “Human Rights and Ethical Globalization.” Tanner Lecture, Stanford University, Stanford, Calif., February 12. Rosenblum, Marc R. 2004. “The Transnational Politics of U.S. Immigration Policy.” Monograph 3. University of California, San Diego Center for Comparative Immigration Studies, La Jolla. Runsten, David, and Phillip Le Veen. 1981. “Mechanization and Mexican Labor in California Agriculture,” Monograph 6. Program in U.S. Mexican Studies, University of California, San Diego. Sassen, Saskia. 1996. Losing Control: Sovereignty in an Age of Globalization. New York: Columbia University Press. Schattschneider, Elmer Eric. 1975. The Semisovereign People: A Realist’s View of Democracy in America. Fort Worth: Harcourt Brace Jovanovich. Schlozman, Kay Lehman, and John T. Tierney. 1986. Organized Interests and American Democracy. New York: Harper & Row. Seidman, Joel. 1942. The Needle Trades. New York: Farrar and Reinhart. Select Commission on U.S. Immigration and Refugee Policy. 1981. U.S. Immigration Policy and the National Interest. Final report. Washington, D.C.: U.S. Government Printing Office. Sen, Amartya. 1999. Development as Freedom. New York: Random House. Shuck, Peter. 1984. “The Transformation of Immigration Law.” Columbia Law Review 84 (January): 49 –50. ———. 1992. “The Politics of Rapid Legal Change.” Studies in American Political Development 6, no. 1 (spring): 37–91. Sierra, Christine Marie. 1993. “The Political Transformation of a Minority Organization: The Council of La Raza, 1965 –1980.” PhD diss., Stanford University, Stanford, Calif. Simon, Julian L. 1989. “Views of Economists and Other Social Scientists Towards Immigration.” In The Economic Consequences of Immigration, edited by Julian Simon, 350 – 357. Oxford, Eng.: Basil Blackwell.

09-S3453-BIB 1/11/06 3:07 PM Page 212

212

References

———. 1995. Immigration: 1995. The Demographic and Economic Facts. Washington, D.C.: Cato Institute/National Immigration Forum. Skerry, Peter. 1993. Mexican-Americans: The Ambivalent Minority. Cambridge, Mass.: Harvard University Press. Smith, Rogers. 1997. Civic Ideals: Conflicting Visions of Citizenship in U.S. History. New Haven: Yale University Press. Soysal, Yasemin N. 1996. “Changing Citizenship in Europe: Remarks on Postnational Membership and the National State.” In Citizenship, Nationality and Migration in Europe, edited by David Cesarani and Mary Fulbrook, 17–24. London: Routledge. Starr, Mark. 1955. “Labor and Immigration.” Current History 29, no. 171 (November): 300. Stern, William Samuel. 1975. “H.R. 2580, the Immigration and Nationality Amendments of 1965 —A Case Study.” PhD diss., New York University, New York. Tate, Katherine. 2003. Black Faces in the Mirror: African Americans and Their Representatives in the U.S. Congress. Princeton, N.J.: Princeton University Press. Tichenor, Daniel J. 2002. Dividing Lines: The Politics of Immigration Control in America. Princeton, N.J.: Princeton University Press. Truman, Harry S. 1961. Public Papers of the Presidents of the United States. Washington, D.C.: U.S. Government Printing Office. UCLA Institute of Industrial Relations. 1992. “Beyond the Free Trade Debate: Labor’s Future in California and Mexico.” Proceedings of a conference at the UCLA Institute of Industrial Relations, Los Angeles, April 23 –24. United Nations Association. 1993. “The Social Implications of a North American Free Trade Agreement.” Report of the Economic Policy Council of the United Nations Association, New York. U.S. Commission on Immigration Reform. 1995. Legal Immigration: Setting Priorities. Washington, D.C.: U.S. Government Printing Office. U.S. Department of Labor, Bureau of Labor Statistics. 1988. “Employed Civilians by Detailed Industry, Sex, Race, and Hispanic Origin, 1983.” Labor Force Statistics Derived from the Current Population Survey, 1948 – 87, Bulletin 2307 (August 1988). U.S. General Accounting Office (GAO). 1988. Legal and Illegal Workers’ Wages and Working Conditions. Washington, D.C.: U.S. Government Printing Office. ———. 1990. Immigration Reform: Employer Sanctions and the Question of Discrimination. Washington, D.C.: U.S. Government Printing Office. U.S. House Committee on the Judiciary. 1995. “Brief History of U.S. Immigration Policy.” In Immigration and Nationality Act (Reflecting Laws Enacted as of May 1, 1995) with Notes and Related Laws, 10th ed., 578 –592. Washington, D.C.: U.S. Government Printing Office. Wagner, Stephen Thomas. 1986. “The Lingering Death of the National Origins Quota System: A Political History of United States Immigration Policy, 1952 –1965.” PhD diss., Harvard University, Cambridge, Mass. Waldinger, Roger. 1985. “Immigration and Industrial Change in the New York

09-S3453-BIB 1/11/06 3:07 PM Page 213

References

213

City Apparel Industry.” In Hispanics in the U.S. Economy, edited by George Borjas and M. Tienda, 323 – 349. New York: Academic Press. ———. 1986. Through the Eye of the Needle. New York: Columbia University Press. Walker, Jack L., Jr., ed. 1991. Mobilizing Interest Groups in America: Patrons, Professions, and Social Movements. Ann Arbor: University of Michigan Press. Walzer, Michael. 1981. “The Distribution of Membership.” In Boundaries: National Autonomy and Its Limits, edited by Peter G. Brown and Henry Shue, 1– 35. Totowa, N.Y.: Rowman & Littlefield. Wilson, James Q. 1980. “The Politics of Regulation.” In The Politics of Regulation, edited by James Q. Wilson, 357– 394. New York: Basic Books. Wilson, William Julius. 1996. When Work Disappears: The World of the New Urban Poor. New York: Knopf. Zolberg, Aristide. 1990. “Reforming the Back Door: The Immigration Reform and Control Act of 1986 in Historical Perspective.” In Immigration Reconsidered: History, Sociology and Politics, edited by Virginia Yan-McLaughlin, 315 – 339. New York: Oxford University Press. ———. 1999. “Matters of State: Theorizing Immigration Policy.” In The Handbook of International Migration: The American Experience, edited by Charles Hirschman, Philip Kasinitz, and Josh DeWind, 71–93. New York: Russell Sage Foundation.

09-S3453-BIB 1/11/06 3:07 PM Page 214

10-S3453-IX 1/11/06 3:07 PM Page 215

Index

Italic page numbers indicate material in tables or figures. Page numbers followed by “n” indicate notes. Abraham, Spencer, 138, 143 ACLU (American Civil Liberties Union), 39 Act to Repeal the Chinese Exclusion Acts, to Establish Quotas, and for Other Purposes (1943), 189n6 ACTWU. See Amalgamated Clothing and Textile Workers Union (ACTWU) admissions policy: as category of immigration policy, 25 –26; employers’ goals, 28 –29, 40, 41; Hispanics’ goals, 35; industrial union goals, 31, 58; interestgroup goals, 40, 41; workers’ goals, 30 – 32 adverse-effect wage rates (AEWRs), 202n8 AFL. See American Federation of LaborCongress of Industrial Organizations (AFL-CIO) African Americans. See blacks AgJobs, 178, 179, 202n5, 202n7 Alexis de Tocqueville Institution, 139 Amalgamated Clothing and Textile Workers Union (ACTWU), 78, 79 – 80, 85, 139. See also Union of Needletrades, Industrial and Textile Employees (UNITE) Amalgamated Clothing Workers of America, 193n18

American Civil Liberties Union (ACLU), 39 American Federation of Labor (AFL), 47, 57, 77, 191n27. See also American Federation of Labor-Congress of Industrial Organizations (AFL-CIO) American Federation of Labor-Congress of Industrial Organizations (AFL-CIO): Chrysler-Berman-Brownback Amendment, 150, 168; employer sanctions, 10, 32, 43, 64, 72, 77– 80, 139; family-based admissions, 33; guest-worker programs, 74; H-1B program, 139 –140, 198n13; Hart-Cellar Act, 52, 57–59; illegal immigration, 77; Illegal Immigration Reform and Immigrant Responsibility Act, 139; Immigration Act of 1990, 104 – 105, 127; immigration policy change, 9 –10; national-origin quotas, 51, 77; Secure America and Orderly Immigration Act, 180; temporary-farmworker programs, 67; union breaks with, 193n20. See also American Federation of Labor (AFL); Congress of Industrial Organizations (CIO) American Immigration Lawyers Association, 73, 102 American Protective Association, 189n2 amnesty, 87– 88, 97, 99 –101, 121, 123, 194n3 anti-Asian immigration laws, 4 –5 2

10-S3453-IX 1/11/06 3:07 PM Page 216

216

Index

antipopulist norm, 12, 185n19 Armey, Dick, 140, 198n14 ascriptive associations, 15 Asian American Legal Defense and Education Fund, 196n19 Asian American legal organizations coalition, 102 –103, 196n19 Asian American rights groups, 38, 45, 66 Asian Americans, 63, 183n2, 184n5 Asian Law Alliance, 196n19 Asian Law Caucus, 196n19 Asian Pacific American Legal Center of Southern California, 196n19 asylees, 6, 8 Australia, 191n32 AWPA (Migrant and Seasonal Agricultural Worker Protection Act) (1983), 201n3 Beilenson Amendment (Illegal Immigration Reform and Immigrant Responsibility Act), 162 Bennett, William, 134 Berman, Howard, 91, 93, 100, 138, 141, 196n18 blacks: attitudes toward immigration, 35 – 38, 36, 37; as constituents, 107–108, 169, 171, 174 –175; effect of immigration on, 16 border control funding, 141, 198n16 bracero program, 64, 77, 90. See also farmworkers, temporary Brownback, Sam, 138, 169. See also Chrysler-Berman-Brownback Amendment (Illegal Immigration Reform and Immigrant Responsibility Act) Bryant Amendment (Immigration Act of 1990), 124, 124, 125, 128 Burr Amendment (Illegal Immigration Reform and Immigrant Responsibility Act), 199n26 Bush, George H., 102, 195nn12 –13 Bush, George W., 18 business coalition: and ethnic rights groups, 1–2; Illegal Immigration Reform and Immigrant Responsibility Act, 137–138, 143, 144, 199n18, 199n19; power of, 2 – 3 California: anti-Chinese immigration restrictionism, 187n6; contract-labor pro6

gram, 93; illegal immigration costs, 187n6; Illegal Immigration Reform and Immigrant Responsibility Act, 140; Proposition 187, 133, 137, 200n28; public opinion on illegal immigration, 133; Republican Party, 14 Canada: annual immigration targets, 191n26; Hart-Cellar Act, 46; immigration policy goal, 191n32; visa limits, 61 Canady Amendment (Illegal Immigration Reform and Immigrant Responsibility Act), 162 capture theory, 180 Caribbean Action Lobby, 96 Carter, Jimmy, 8, 64 – 65, 69, 97 Carter, Leonard, 64 Cato Institute, 136, 138, 197n4, 201n3 Cellar, Emmanuel, 50, 51, 55 –56. See also Hart-Cellar Act (1965); Luce-Cellar Act (1946) Chabot, Steve, 141 Chabot Amendment (Illegal Immigration Reform and Immigrant Responsibility Act), 161, 169, 170 –171, 174, 175 Chaikin, Sol, 82 chain migration, 6 Change to Win coalition, 193n20 Chavez, Cesar, 89 China, 48, 187n6, 195n12 Chinese Exclusion Act (1882), 4 –5, 47, 48, 184n5, 189n6 Chisti, Muzaffar, 83 Chrysler, Dick, 138, 169 Chrysler-Berman-Brownback Amendment (Illegal Immigration Reform and Immigrant Responsibility Act): AFL-CIO, 150, 168; left-right coalition, 138; political parties, 140, 141; Republican Party, 169; voting models with ethnic-specific variables, 165, 166 –168, 168 –169, 171– 172, 174 CIO. See Congress of Industrial Organizations (CIO) citizens’ groups, rise of, 4, 184n8 civil libertarians, 39 civil rights groups. See ethnic rights groups cleaning industry, 86 – 87 client-politics model, 11–12, 185n16 Clinton, Bill: Gallegly Amendment, 142, 144; illegal immigration, 141, 198n16;

10-S3453-IX 1/11/06 3:07 PM Page 217

Index Illegal Immigration Reform and Immigrant Responsibility Act, 139 –140, 141; Legal Immigration Family Equity Act, 13; temporary-worker programs, 93; U.S. Commission on Immigration Reform report, 134 communitarian ideology, 187n17 Comprehensive Enforcement and Immigration Reform Act (2005), 203n10 Congressional Black Caucus, 72, 119 Congressional Hispanic Caucus, 96 –97, 98 –99, 119 Congress of Industrial Organizations (CIO), 57, 77, 191n27. See also American Federation of Labor-Congress of Industrial Organizations (AFL-CIO) conservative coalition, 99 –100, 195n9 contract-labor programs, 93 Contract with America, 14, 134, 136 –137, 198n5. See also Republican Party Convention Relating to the Status of Refugees (1951, Geneva), 5 Cornyn, John, 203n10 Cornyn-Kyl Bill (2005), 203n10 Cubans, 8, 30, 192n8 cultural diversity, attitudes toward, 26 –27 Daschle, Tom, 202n10 DeConcini, Dennis, 103 de la Garza Amendment (Immigration Reform and Control Act), 110, 111, 115, 122 Democratic Party: 1964 election, 52, 190n19; amnesty, 97; ethnic constituencies, 101; Gallegly Amendment, 142; guest-worker programs, 93; Hart-Cellar Act, 45 – 46, 52, 56, 59 – 60, 61; illegal immigration, 187n5; Illegal Immigration Reform and Immigrant Responsibility Act, 141; labor market regulation, 27; national-origins quota system, 49 –50; Reciprocal Trade Agreements Act, 190n25; Simpson-Mazzoli Bill, 100; social liberals, 27–28 Department of Labor, 45, 57, 59 – 60, 177, 179, 198n13 Dirksen, Everett, 51–52 D-nominate scores, 149, 200nn30 – 31 Dole, Robert, 142, 144, 145 Douglas, Paul, 77

217

Eastern Hemisphere immigration quotas, 47, 51 Eastland, James, 50, 77, 190n12 economic interests, politics of, 1, 183n2 economic theory, 73 –74 Edwards, James R., Jr., 14, 108 –109 El Salvador refugees, 8, 196n27 employers: admissions policy goals, 28 – 29, 40, 41; enforcement policy goals, 30; Hart-Cellar Act, 45 – 46; incorporation policy goals, 29; policy goals, 28 – 30, 40 – 43, 41 employer sanctions: AFL-CIO, 10, 32, 43, 64, 72, 77– 80, 139; Amalgamated Clothing and Textile Workers Union, 78, 85, 139; Asian American rights groups, 66; Carter administration, 64 – 65; Congressional Black Caucus, 119; Congressional Hispanic Caucus, 119; ethnic rights groups, 15 –16, 70 –73, 96; Hispanic rights groups, 66 – 67, 69 –70, 71–72, 131–132; Immigration Act of 1990, 82; Immigration Reform and Control Act, 82, 96 –97, 134 –135, 194n3; interest group activities in 1970s, 64 – 65; International Alliance of Theatrical Stage Employees, Moving Picture Technicians, 82; League of United Latin American Citizens, 64, 69 –70, 71, 72; Mexican American Legal Defense and Education Fund (MALDEF), 69 –70, 71; NAACP, 35, 64, 72 –73; National Council of La Raza, 69 –70, 71–72; Service Employees International Union, 78 –79, 86 – 88, 139, 193n28; unions, 67, 76 – 81, 80; United Farm Workers (UFW), 88 –90; United Food and Commercial Workers International Union, 88 employment-based admissions, 40, 41, 45 – 46 employment visas, 29 Empower America, 134, 138 enforcement policy: black constituents, 174 –175; as category of immigration policy, 26; employers’ policy goals, 30; interest-group goals, 41, 42 – 43; and privacy, 41, 42 – 43; workers’ policy goals, 30 – 32 entrepreneurial politics, 185n18 environmentalists, 10, 39 – 40, 41 Ervin, Sam, 51–52 2

10-S3453-IX 1/11/06 3:07 PM Page 218

218

Index

Espinoza v. Farah Manufacturing Co. (1978), 66, 75 Essential Worker Immigration Coalition, 18 –19, 179 ethnic rights groups: coalition with business groups, 1–2; Congress, 10 –11; employer sanctions, 15 –16, 70 –73, 96; family-based immigration, 33 – 34, 40, 41; guest-worker programs, 34 – 35; Hart-Cellar Act, 62 – 63; interest-group goals, 33 – 38; legalization, 73; nature of, 181–183; policy goals, 33 – 38, 40 – 43, 41; post-bracero dilemmas, 68 –76; power of, 3, 184n5; and regulation, 9 –11; temporary farmworkers, 73 –76 FAIR (Federation for American Immigration Reform), 39 – 40, 188n17 family-based immigration: AFL-CIO, 33; as basic human right, 15, 186n28; civil rights groups’ policy goals, 33 – 34, 40, 41; Hart-Cellar Act, 6, 45, 53 –56; interest-group goals, 40, 41; Republican Party, 101; and social incorporation, 9; support of, 10 family coalition: Illegal Immigration Reform and Immigrant Responsibility Act, 136, 138, 143, 144, 198n11; Immigration Act of 1990, 102, 104 Farm Labor Alliance, 65 farmworkers, 29, 46, 67– 68, 91–93; effect on voting, 111–112, 115, 121; temporary, 18, 67– 68, 73 –76, 100 –101. See also bracero program federal regulation of immigration, 189n1 Federation for American Immigration Reform (FAIR), 39 – 40, 188n18 Feighan, Michael, 51, 53 Fish, Hamilton, 99 Fisher, O. C., 55 Fitzgerald, Keith, 12 –13 Flake, Jeff, 179, 186n31 Florida, 140, 187n6 Fourteenth Amendment, 75 Freeman, Gary, 11–12 Gallegly, Elton, 137, 142 Gallegly Amendment (Illegal Immigration Reform and Immigrant Responsibility Act), 142, 144, 161, 162, 169, 172 garment industry, 31, 83 – 84. See also 8

Amalgamated Clothing and Textile Workers Union (ACTWU); International Ladies Garment Workers Union (ILGWU) General Accounting Office (GAO), 78, 139 Gentlemen’s Agreement (1908), 48 Gimpel, James G., 14, 108 –109 Gingrich, Newt, 140, 142 Goldwater, Barry, 52 Gossett, Bill, 191n27 Gramm, Phil, 199n21 guest-worker programs, 177–180; AFLCIO, 74; civil rights groups, 34 – 35; Democratic Party, 93; economic theory, 73 –74; Immigration Reform and Control Act, 91–93; interest-group goals, 40, 41, 42; international trade theory, 73; Mexican workers, 191n29; Republican Party, 93; unions, 32 – 33, 74; United Farm Workers, 178, 201n3. See also H-2A program; H-2B program; probationaryworker programs; temporary workers Gutierrez, Luis, 179, 186n31 H-1B program, 104, 105, 139 –140, 143, 198n13, 202n6 H-2 program, 65, 75, 92, 99, 195n8 H-2A program, 46, 92, 177–178, 201n3, 202n8 H-2B program, 92, 179, 202n8 H-5A visas, 179 –180 Hagel, Chuck, 202n10 Hagel-Daschle Bill, 202n10 Haitians, 8 Hart, Philip, 50 Hart-Cellar Act (1965), 44 – 63; AFL-CIO, 52, 57–59; Asia, 44; Asian American organizations, 45; Asian Americans, 63; bipartisan compromises, 45 – 46, 50 –59; Canada, 46; chain migration, 6; codification, 184n11; consequences, 4, 5 – 6, 44, 60, 61– 63; Democratic Party, 45 – 46, 52, 56, 59 – 60, 61; as durable solution, 59 – 61; Eastern Hemisphere, 51; employers, 45 – 46; as end to racist policies, 44; ethnic rights groups, 62 – 63; EuroAmerican immigrants’ role in passage, 184n5; family-based immigration, 6, 45, 53 –56; farmworkers, 46; H-2A program, 46; immigrant laborer regulations, 56 –

10-S3453-IX 1/11/06 3:07 PM Page 219

Index 59; Latino organizations, 45; Mexican Americans, 62 – 63; Mexico, 44, 46; national-origins quota system, 44, 46, 50, 189n9; Republican Party, 45 – 46, 52, 53, 59, 60, 61; unions, 45 – 46, 56 –59; unresolved issues, 58 –59; visas, 5 – 6, 53, 59 – 60, 190n21; Western Hemisphere immigration limits, 44, 45, 51–52, 53, 60 – 61, 62 Hatch, Orrin, 103 head tax proposal, 104 –105, 196nn23 –24 Henderson v. Mayor of New York (1875), 189n1 HEREBIU (Hotel Employees and Restaurant Employees and Bartenders International Union), 78, 80, 85 – 86, 139 Hesburgh, Theodore, 194n1 Hispanic rights groups: criticism of, 186n30; employer sanctions, 66 – 67, 69 – 70, 71–72, 131–132; Hart-Cellar Act, 45; Immigration Reform and Control Act, 130; power of, 3; stance in 1980s, 15 – 16; temporary-farmworker programs, 67– 68 Hispanics: admissions policy goals, 35; attitudes toward immigration, 35 – 38, 36, 37; Illegal Immigration Reform and Immigrant Responsibility Act voting model, 147, 151, 162, 163; Immigration Reform and Control Act, 78; participation in electoral politics, 13; versus Puerto Ricans, 96; and Republican Party, 13; subcategories, 183n3; workers in key industries, 79 – 81, 81, 193n23. See also Mexican Americans historical-institutionalist framework, 13 Hotel Employees and Restaurant Employees and Bartenders International Union (HEREBIU), 78, 80, 85 – 86, 139 House Education and Labor Committee, 98, 99 House Judiciary Committee, 98, 99, 105, 137, 138, 141 House Rules Committee, 98, 140 House Ways and Means Committee, 105 Huerta, Dolores, 89 human rights, 15, 186n28 identity politics: and representation, 14 – 17; underlying tensions in, 1, 183n2 IIRIRA. See Illegal Immigration Reform

219

and Immigrant Responsibility Act of 1996 (IIRIRA) ILGWU. See International Ladies Garment Workers Union (ILGWU) illegal immigration, 76 –90, 80, 81; AFLCIO, 77; Amalgamated Clothing and Textile Workers Union, 85; Clinton administration, 141, 198n16; Democratic and Republican concerns, 187n5; Hotel Employees and Restaurant Employees and Bartenders International Union, 85 – 86; International Ladies Garment Workers Union, 82 – 85; number of illegal immigrants, 8; Service Employees International Union, 86 – 88; unions in agriculture and food processing, 88 –90 Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (IIRIRA), 135 –145; AFL-CIO, 139; and Alan Simpson, 142 –144, 199n18; Beilenson Amendment, 162; Burr Amendment, 199n26; business coalition, 137–138, 143, 144, 199n19; California, 140; Canady Amendment, 162; Chabot Amendment, 161, 169, 170 –171, 174, 175; Clinton administration, 139 – 140, 141; conference and negotiations with Clinton administration, 144 –145; deliberations in Senate Judiciary Committee, 142 –143; Democratic Party, 141; and Edward Kennedy, 138, 143, 144; family coalition, 136, 138, 143, 144, 198n11; Florida, 140; Gallegly Amendment, 142, 144, 161, 162, 169, 172; H-1B program, 143; labor’s position, 139 –140; left-right coalition, 136 –139, 144; McCollum Amendment, 161, 162; National Asian Pacific American Legal Consortium, 138; National Council of La Raza, 138; PAC contributions, 150, 151, 153 – 154, 162 –163, 168, 173 –174, 200n32; political parties, 140 –141; Pombo Amendment, 171, 173, 201n2; provisions, 141–142; Republican Party, 140 – 141, 145; Senate floor action, 143 –144; Union of Needletrades, Industrial and Textile Employees, 138; Velazquez Amendment, 162. See also ChryslerBerman-Brownback Amendment (Illegal Immigration Reform and Immigrant Responsibility Act) 2

10-S3453-IX 1/11/06 3:07 PM Page 220

220

Index

Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (IIRIRA) voting model, 145 –163, 148, 165 –175; amendments, 148; discussion of findings, 162 –163; with ethnicspecific variables, 165 –175, 166 –168, 170 –171, 172, 173, 174; further analysis of separate votes, 161–162; Hispanic constituents, 147, 151, 162, 163; issue dissimilarity, 155 –161, 156 –159, 160 – 161; results of analysis, 151–154, 152 – 153, 154; variables and expectations, 147, 149 –150 ILO (International Labour Organization), 76 immigrants, number admitted, 4, 5, 7 Immigration Act of 1917, 5, 48, 189n7 Immigration Act of 1921, 4, 47 Immigration Act of 1924, 5, 47, 48, 189n4 Immigration Act of 1990, 101–105; AFLCIO, 104 –105, 127; Bryant Amendment, 124, 124, 125, 128; and Edward Kennedy, 103; employer sanctions, 82; family coalition, 102, 104; H-1B program, 104, 105; Irish immigrants, 103 – 104, 145; and Kennedy-Simpson Bill, 102; PAC contributions, 123, 124, 125, 127, 131; proimmigration coalition, 102 –104; and Simon-Berman Bill, 196n18; Smith Amendment, 124, 124, 125, 126 –127; unions, 104 –105 Immigration Act of 1990 voting model, 123 –130; results of analysis, 125 –130, 126 –127, 128 –130; votes analyzed, 123 – 125, 124 Immigration and Nationality Act (1952). See McCarran-Walter Act (1952) Immigration and Nationality Act Amendments (1965). See Hart-Cellar Act (1965) Immigration and Naturalization Service. See U.S. Immigration and Naturalization Service (INS) immigration cap, 97, 195n4 immigration lawmaking theories, 11–14 immigration policy categories, 25 –26 Immigration Reform and Control Act of 1986 (IRCA), 95 –101; amnesty, 97, 99 – 101, 121, 123, 194n3; blocking reform of legal immigration, 97–99; citizenship status, 66; conservative coalition, 99 – 0

100, 195n9; de la Garza Amendment, 110, 111, 115, 122; Democratic Party, 100; discrimination against Hispanics, 78; and Edward Kennedy, 97; and Edward Roybal, 98; employer sanctions, 82, 96 –97, 134 –135, 194n3; farmworkers, 91–93; guest-worker program, 91– 93; H-2 program, 92, 99, 195n8; Hispanic rights groups, 130; Irish Americans, 121; Japanese Americans, 121; Kennedy-Simpson Bill, 97, 195n4; legalization program, 6; McCollum Amendment, 109, 110, 110, 111, 114 –115, 120; Mexican Americans, 121; Moorhead Amendment, 109, 110, 118, 197n29; National Council of La Raza (NCLR), 76; numbers admitted under legalization program, 7; Panetta Amendment, 109, 110, 115, 117, 119, 123; racial discrimination in the workplace, 67; Reagan administration, 98, 99, 100; Republican Party, 99, 100; Roybal Amendment, 109, 110, 116; temporary farmworkers, 100 – 101; temporary workers, 76, 99 –101, 192n15 Immigration Reform and Control Act of 1986 (IRCA) voting model, 107–123; independent variables and expectations, 111–113; model of individual floor votes, 110 –111; results of the analysis, 113 –123, 114 –115, 116 –118, 120, 122 – 123; votes analyzed, 109 –110, 110, 111 Immigration Restriction League, 47 improvisational institutionalist approach, 12 –13 incorporation policy: as category of immigration policy, 26; employers’ policy goals, 29; and family-based immigration, 9; fluctuation in, 8; interest-group goals, 41, 43; workers’ policy goals, 32 – 33 independent immigrants, 97, 102 INS. See U.S. Immigration and Naturalization Service (INS) inside strategies, 16, 106 integration policy. See incorporation policy interest-group goals, 23 – 43; admissions policy, 40, 41; civil libertarians, 39; civil rights groups, 33 – 38; coalitions and op-

10-S3453-IX 1/11/06 3:07 PM Page 221

Index positions, 40 – 43, 41; employers, 28 – 30; employment-based admissions, 40, 41; enforcement policy, 41, 42 – 43; familybased immigration, 40, 41; guestworker programs, 40, 41, 42; incorporation policy, 41, 43; probationaryworker programs, 40, 41, 42; religious groups, 39; restrictionist environmental groups, 39 – 40; scope, 23 –24; temporary workers, 40, 41, 42; workers, 30 – 33 interest-group politics, 105 –107, 185n18 interest group research, 4, 184n6 International Alliance of Theatrical Stage Employees, Moving Picture Technicians, 79, 82 International Labour Organization (ILO), 76 International Ladies Garment Workers Union (ILGWU), 73, 78, 79 – 80, 82 – 83, 85, 139. See also Union of Needletrades, Industrial and Textile Employees (UNITE) International Longshoremen’s and Warehousemen’s Union, 82 International Longshoremen’s Association, 82 international trade theory, 73 International Union of Electronic, Electrical, Salaried, Machine and Furniture Workers, 79 IRCA. See Immigration Reform and Control Act of 1986 (IRCA) Irish Americans, 103 –104, 121, 145, 195n17 Japan, 48 Japanese American Citizens League (JACL), 3, 73, 191n33 Japanese Americans, 121, 131 Johnson, Lyndon, 50, 52, 55 Jordan, Barbara, 134, 196n24 Kamasaki, Charles, 69 –70, 71–72 Kemp, Jack, 134 Kennedy, Edward: H-1B program, 140; Illegal Immigration Reform and Immigrant Responsibility Act, 138, 143, 144; Immigration Act of 1990, 103; Immigration Reform and Control Act, 97; Secure America and Orderly Immigration Act, 179; temporary-worker policy, 18

221

Kennedy, John F., 50, 59 Kennedy, Robert, 51–52, 55, 190n13 Kennedy-Abraham Amendment (Illegal Immigration Reform and Immigrant Responsibility Act), 143, 199n19 Kennedy-Simpson Bill: and Immigration Act of 1990, 102; and Immigration Reform and Control Act, 97, 195n4 Key, V. O., 107–108 Kolbe, Jim, 179, 186n31 Kyl, Jon, 203n10 Labor Department, 45, 57, 59 – 60, 177, 179, 198n13 Laborers’ International Union of North America, 180, 193n20 labor unions. See unions Latino rights groups. See Hispanic rights groups Latinos. See Hispanics Leadership Conference on Civil Rights (LCCR), 15, 72 –73, 186nn25 –26 League of United Latin American Citizens (LULAC): employer sanctions, 64, 69 – 70, 71, 72; founding, 68; headquarters, 3; stance in 1980s, 15 –16 left-right coalitions, 15, 136 –139, 144, 186n27 legal immigration, 102, 133 –134, 184n14, 188n18, 192n8, 197n29, 198n14 Legal Immigration Family Equity (LIFE) Act (2000), 13 legalization programs, 6, 67– 68, 73 liberalism, rights-based, 185n17 liberals, social, 27–28 LIFE Act (2000), 13 Los Angeles County Social Survey, 38, 188n16 Luce-Cellar Act (1946), 189n7 LULAC. See League of United Latin American Citizens (LULAC) MacGregor, Clark, 51 majoritarian politics, 185n18 MALDEF. See Mexican American Legal Defense and Education Fund (MALDEF) maquiladoras, 79, 84 Martinez, Vilma, 71, 75 McCain, John, 18, 179 2

10-S3453-IX 1/11/06 3:07 PM Page 222

222

Index

McCain-Kennedy Bill. See Secure America and Orderly Immigration Act McCarran-Walter Act (1952), 5, 48, 49, 77, 184n11 McCollum Amendment: to Illegal Immigration Reform and Immigrant Responsibility Act, 161, 162; to Immigration Reform and Control Act, 109, 110, 110, 111, 114 –115, 120 Mexican American Legal Defense and Education Fund (MALDEF): contractlabor programs, 93; employer sanctions, 69 –70, 71; founding, 68, 191n4; headquarters, 3; policy goals, 33; stance in 1980s, 15 –16; temporary-worker programs, 75 –76 Mexican Americans: as constituents, 121, 131–132; effect of, 115; Hart-Cellar Act, 62 – 63; Immigration Reform and Control Act, 121; lack of agency in immigration reform, 188n4; national identity card system, 70, 71; public opinion, 70 – 71, 192n8. See also Hispanics Mexico: bracero program, 64, 77, 90; grand bargain with, 19 –20, 187n32; guest workers, 191n29; Hart-Cellar Act, 44, 46; improvisational institutionalist approach to, 13; migrant workers from, 8, 18; visa limits, 61 Migrant and Seasonal Agricultural Worker Protection Act (AWPA) (1983), 201n3 migrants, temporary, 183n4 Miller, William, 52 Moakley, Joe, 196n27 Moorhead Amendment (Immigration Reform and Control Act), 109, 110, 118, 197n29 Morrison, Bruce, 104 –105, 196nn24 –25 multidimensionality of immigration policy, 146 –147, 181, 199n25 NAACP: as ascriptive association, 15; employer sanctions, 35, 64, 72 –73; and Leadership Conference on Civil Rights, 15, 186n26; stance on immigration, 16; and Working Committee on the Concerns of Hispanics and Blacks, 72 NAACP Legal Defense and Educational Fund (LDF), 191n4 2

NAFTA (North American Free Trade Agreement), 85, 182 Na Loio No Na Kanaka—Lawyers for the People of Hawaii, 196n19 NAM. See National Association of Manufacturers (NAM) National Asian Pacific American Legal Consortium (NAPALC), 3, 15, 17, 138, 183n3, 186n24 National Association of Manufacturers (NAM), 30, 136, 137, 138, 165, 198n8 National Council of Churches, 73 National Council of La Raza (NCLR): as ascriptive association, 15; contract-labor programs, 93; employer sanctions, 69 – 70, 71–72; founding, 69; Hagel-Daschle Bill, 203n10; headquarters, 3; identity politics, 17; Illegal Immigration Reform and Immigrant Responsibility Act, 138; Immigration Reform and Control Act, 76; and Jimmy Carter, 69; mission, 15, 186n23; policy goals, 33; Secure America and Orderly Immigration Act, 202n10; stance in 1980s, 15 –16; and Working Committee on the Concerns of Hispanics and Blacks, 72 National Federation of Independent Business, 30 National Hispanic Task Force on Immigration Policy, 75 national identity card system, 70, 71 National Origins Quota Act (1924). See Immigration Act of 1924 national-origins quota system, 44, 46, 47– 50, 51, 77, 189n11 nativist organizations, 47, 189n2 NCLR. See National Council of La Raza (NCLR) New Zealand, 191n26 Nihomachi Legal Outreach, 196n19 Nissei Farmers’ League, 121 nonprofit organizations, 4 norm, defined, 191n2 North American Free Trade Agreement (NAFTA), 85, 182 numbers-versus-rights question, 65 – 68 nurses, foreign, 195n17, 199n26 O’Neill, Tip, 98 Operation Wetback, 71, 191n29

10-S3453-IX 1/11/06 3:07 PM Page 223

Index

223

Organization of Chinese Americans (OCA), 3, 73, 96 organized labor. See unions outside strategies, 16, 106

racial/ethnic group differences, 35 – 38, 36, 37; and salary levels in district, 149, 200n28 Puerto Ricans, 96

PAC contributions: building trade unions, 127, 131, 154, 162 –163, 168; Illegal Immigration Reform and Immigrant Responsibility Act, 150, 151, 153 –154, 162 – 163, 168, 173 –174, 200n32; Immigration Act of 1990, 123, 124, 125, 127, 131; unions, 127, 131, 150, 153, 162 –163, 174 Panetta, Leon, 91, 100 Panetta Amendment (Immigration Reform and Control Act), 109, 110, 115, 117, 119, 123 Papademetriou, Demetrios G., 19 –20, 186n32 perishable commodities, 91–92 Personal Responsibility and Work Opportunity Reconciliation Act (1996), 141–142 Plyler v. Doe (1982), 68 policy goals: Asian American organizations, 38; civil libertarians, 39; employers, 28 – 30, 40 – 43, 41; ethnic rights groups, 33 – 38, 40 – 43, 41; Mexican American Legal Defense and Education Fund, 33; National Council of La Raza, 33; religious groups, 39; restrictionist environmental groups, 39 – 40, 41; unions, 30 – 33, 40 – 43, 41; workers, 30 – 33 Pombo, Richard, 201n2 Pombo Amendment (Illegal Immigration Reform and Immigrant Responsibility Act), 171, 173, 201n2 Population Connection, 39 – 40 portable visas, 179, 206n6 poverty as capability deprivation, 24 prevailing wage, 206n8 privacy and enforcement policy, 41, 42 – 43 probationary-worker programs, 33, 40, 41, 42. See also guest-worker programs Proposition 187 (California), 133, 137, 200n28 public assistance for immigrants, 141–142 public opinion: illegal immigration, 133; immigration cutbacks, 6, 184n14; Mexican Americans, 70 –71, 192n8;

racial discrimination in workplace, 67 racially restrictive laws, 44, 46, 47–50, 51, 77, 189n11 RAWs. See replenishment agricultural workers (RAWs) Reagan, Ronald, 75 –76, 97, 98, 99, 100, 195n17 Reciprocal Trade Agreements Act (1934), 190n25 refugees, 5, 6, 8, 13, 196n27 registration of undocumented immigrants, 19 –20, 186n32 regulation, and ethnic advocates, 9 –11 Reich, Robert, 139 –140 religious group goals, 39 replenishment agricultural workers (RAWs), 92 –93, 100 –101 representation and identity politics, 14 –17 Republican Party: California, 14; ChryslerBerman-Brownback Amendment, 169; family-based admissions, 101; Gallegly Amendment, 142, 169; guest-worker programs, 93; Hart-Cellar Act, 45 – 46, 52, 53, 59, 60, 61; illegal immigration, 187n5; Illegal Immigration Reform and Immigrant Responsibility Act, 140 –141, 145; immigration costs, 26; labor market regulation, 27; and Latinos, 13; Legal Immigration Family Equity Act, 13; likelihood of voting for restrictive policy, 113, 115; national-origins quota system, 49 –50; Reciprocal Trade Agreements Act, 190n25; Simpson-Mazzoli Bill, 99, 100. See also Contract with America restaurant industry, 86 Roosevelt, Theodore, 48 Roybal, Edward, 98 Roybal Amendment (Immigration Reform and Control Act), 109, 110, 116 salary levels, and public opinion, 149, 200n28 Salvadoran refugees, 8, 196n27 SAWs. See special agricultural workers 2

10-S3453-IX 1/11/06 3:07 PM Page 224

224

Index

schooling, public, for illegal immigrant children, 142 Schumer, Charles, 91, 100 Secure America and Orderly Immigration Act, 18, 179 –180, 186n31, 202n5, 202n8, 202n10 Select Commission on U.S. Immigration and Refugee Policy, 95 –96, 97, 101, 201n1 Sen, Amartya, 24 Senate floor action, on Illegal Immigration Reform and Immigrant Responsibility Act, 143 –144 Senate Judiciary Committee, 137, 138, 142 –143 Serna v. Portales (1972), 68 Service Employees International Union (SEIU): amnesty, 87– 88; break with AFL-CIO, 193n20; Chabot Amendment, 169, 174; employer sanctions, 78 –79, 86 – 88, 139, 193n28; illegal immigration, 86 – 88; Secure America and Orderly Immigration Act, 180 Simon, Paul, 138, 196n18 Simon-Berman Bill, 196n18 Simpson, Alan: cuts in legal immigration, 133 –134, 197n1; Gallegly Amendment, 142; Hart-Cellar Act, 62; Illegal Immigration Reform and Immigrant Responsibility Act, 142 –144, 199n18; role of interest groups, 23 –24; Simpson-Mazzoli Bill, 97 Simpson-Mazzoli Bill. See Immigration Reform and Control Act of 1986 (IRCA) Simpson-Smith Bill. See Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (IIRIRA) Smith, Lamar, 133 –134, 136, 137, 141 Smith Amendment (Immigration Act of 1990), 124, 124, 125, 126 –127 social incorporation policy. See incorporation policy social liberals, 27–28 Southwest Voter Registration Education Project, 93 special agricultural workers (SAWs), 67– 68, 91–93, 100 specialty occupations, 104, 198n13 state regulation of immigration, 189n1 Sweeney, John, 139 4

Takao Ozawa v. United States (1922), 48 Tanton, John, 188n18 taxes, and services consumed by immigrants, 26, 187nn3 – 4 temporary migrants, 183n4 Temporary Quota Act (1921). See Immigration Act of 1921 temporary workers: and Bill Clinton, 93; farmworkers, 18, 67– 68, 73 –76, 100 – 101; Immigration Reform and Control Act, 76, 99 –101, 192n15; interest-group goals, 40, 41, 42; and membership rights, 9; Mexican American Legal Defense and Education Fund, 75 –76; Reagan administration, 75 –76; Southwest Voter Registration Education Project, 93. See also guest-worker programs; probationary-worker programs terrorism, 177 Texas Proviso, 77, 96, 193n17 Tichenor, Daniel J., 13 trades, migrant categories, 90 –93 Truman, Harry, 49, 189n10 UFCW (United Food and Commercial Workers), 80, 88, 193n19 UFW (United Farm Workers), 18, 88 –90, 178, 180, 196n19, 201n3 unemployment increase, effect on voting, 112, 119, 125, 127, 197n32, 197n34 Union of Needletrades, Industrial and Textile Employees (UNITE), 138, 193n19, 198n9. See also Amalgamated Clothing and Textile Workers Union (ACTWU); International Ladies Garment Workers Union (ILGWU) unions: admissions policy goals, 30 – 32; bracero program, 77; building trade unions, 127, 131, 154, 162 –163, 168, 201n35; craft unions, 31, 58, 79; employer sanctions, 67, 76 – 81, 80; enforcement policy goals, 30 – 32; guest-worker programs, 32 – 33, 74; Hart-Cellar Act, 45 – 46, 56 –59; illegal immigration, 88 – 90; Illegal Immigration Reform and Immigrant Responsibility Act, 139 –140; Immigration Act of 1990, 104 –105; incorporation policy goals, 32 – 33; industrial unions, 31, 58, 131, 153, 201n35; mixed stands of, 9 –10; PAC contribu-

10-S3453-IX 1/11/06 3:07 PM Page 225

Index tions, 127, 131, 150, 162 –163, 174; policy goals, 30 – 33, 40 – 43, 41; probationaryworker programs, 33; temporaryfarmworker programs, 67; temporaryworker policy, 18. See also specific unions UNITE. See Union of Needletrades, Industrial and Textile Employees (UNITE) United Association of Journeymen and Apprentices of the Plumbing and Pipefitting Industry of the United States and Canada, 79 United Brotherhood of Carpenters and Joiners of America, 79, 80, 81– 82, 193n20, 197n23 United Farm Workers (UFW), 18, 88 –90, 178, 180, 193n20, 201n3 United Food and Commercial Workers (UFCW), 80, 88, 193n20 United Nations International Convention on the Protection of the Rights of All Migrant Workers and Members of Their Families, 76, 192n16 United States Conference of Catholic Bishops (USCCB), 39, 64, 73, 198n11 United Steelworkers of America, 57, 58 UNITE HERE, 180, 193nn19 –20 Universal Declaration of Human Rights, 186n28 U.S. Chamber of Commerce, 102 U.S. Citizenship and Immigration Services (USCIS), 177, 192n5. See also U.S. Immigration and Naturalization Service (INS) U.S. Commission on Immigration Reform, 62, 134, 135, 141, 196n24 U.S. General Accounting Office (GAO), 78, 139 U.S. Immigration and Naturalization Ser-

225

vice (INS), 69, 134, 177, 192n5. See also U.S. Citizenship and Immigration Services (USCIS) USCCB (United States Conference of Catholic Bishops), 39, 64, 73, 198n11 Velazquez Amendment (Illegal Immigration Reform and Immigrant Responsibility Act), 162 visas: Canada, 61; China, 48; employment visas, 29; H-5A visas, 179 –180; HartCellar Act, 5 – 6, 53, 59 – 60, 190n20; Mexico, 61; portable visas, 179, 202n7; W visas, 202n10 voting models. See Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (IIRIRA) voting model; Immigration Act of 1990 voting model; Immigration Reform and Control Act of 1986 (IRCA) voting model Voting Rights Act, 1975 amendment, 66 wage rates, 202n8 Western Hemisphere immigration limits, 44, 45, 51–52, 60 – 61, 189n8 whites, attitudes toward immigration, 35 – 38, 36, 37 White v. Regester (1973), 68 Wilson, J. Q., 11, 185n18 Wilson, Pete, 91, 133 workers’ policy goals, 30 – 33. See also unions Workforce 2000 (Johnston and Packer), 195n11 Working Committee on the Concerns of Hispanics and Blacks, 72 W visas, 202n10

2

10-S3453-IX 1/11/06 3:07 PM Page 226

6