Economics Of Biotechnology

This page intentionally left blank Copyright © 2007, New Age International (P) Ltd., Publishers Published by New Age ...

Author: T V S Ramamohan Rao

33 downloads 759 Views 2MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

This page intentionally left blank

Copyright © 2007, New Age International (P) Ltd., Publishers Published by New Age International (P) Ltd., Publishers All rights reserved. No part of this ebook may be reproduced in any form, by photostat, microfilm, xerography, or any other means, or incorporated into any information retrieval system, electronic or mechanical, without the written permission of the publisher. All inquiries should be emailed to [email protected]

ISBN (13) : 978-81-224-2638-0

PUBLISHING FOR ONE WORLD

NEW AGE INTERNATIONAL (P) LIMITED, PUBLISHERS 4835/24, Ansari Road, Daryaganj, New Delhi - 110002 Visit us at www.newagepublishers.com

To

The Inquisitive Reader


PREFACE Major technological innovations of the 20th century consisted of •

mechanical devices including the steam engine, the automobile, and mechanical contraptions involving automation of the assembly lines of manufacturing processes • electronic devices and the advent of information technology Advances in physics had a major role in many of these developments. This made some observers suggest that the 20th century belonged to physics. The other major driver of the 20th century economic development was the chemical technology. Of particular significance were the chemical technologies related to crude oil distillation, agricultural inputs, and pharmaceuticals. The major developments in the context of the pharmaceutical sector consisted of the discovery of new chemical entities of therapeutic value by combining chemicals whose properties are well established and protein synthesis using chemical reactions. However, there have been doubts about the sustainability of these technologies in the future. The two major reasons are •

the fast depletion of crude oil resources has been eroding the input structure on which many of these technologies depend, and • significant, though perhaps unexpected, externalities in the form of air and water pollution and the associated greenhouse gas effects like global warming In addition, the use of chemicals, exemplified by the use of fertilizers and pesticides in the era of the green revolution, had the effect of destroying plant nutrients in the soil and gave rise to pest infestation largely more immune to the use of pesticides. These pesticides also entered foodgrains, fruits and vegetables, and water that we consume regularly. Many of these ill effects were recognized only belatedly. There is a more fundamental level at which conventional technologies have been found to be insufficient. It is obvious that every country wants to be self-sufficient with respect to food and healthcare. The technologies that are currently in use have been developed to ensure this. However, the green revolution in agriculture, which is vastly dependent on

186

Economics of Biotechnology

(viii)

mechanization and chemical technologies, appears to have run its course and shows signs of a decline in productivity. There is a possibility of shortages in the future if supply of traditional varieties slows down and/or the demand outstrips present supply potential. In other words, it has been recognized that most of the major technological inventions related to these conventional technologies are already in place. New developments can only be marginal. As a result, the technological developments in mature industries will be a result of learning-by-doing and cumulative use of technologies. Therefore, the large corporations will themselves undertake much of the R&D in-house. Limits on the potential of the technologies as well as limitations of competence and organizational culture of the large firms restrain further progress. It was also pointed out that much of the technological development in the present day technologies is a result of random search through various possibilities. There is no systematic logical basis for progress in the R&D activities. This is especially true of traditional plant breeding and the search for new chemical entities that have therapeutic value for the pharmaceutical industry. The usefulness of biochemical reactions, albeit at a rudimentary level, has been acknowledged for many years. For instance, it is well known that bacteria are the essential change agents to convert milk into yogurt. Similarly, yeast is used in brewing beer and wine. Such discoveries were also a result of hit and miss processes somewhat akin to the discoveries in chemical technology. A major revolution, in the form of life sciences and biotechnology, emerged before the turn of the 21st century. Modern biotechnology consists of a set of techniques that involve manipulation or change of the genetic inheritance of living organisms including plants. This technology utilizes some known biological processes to alter others found in nature. More pertinently, modern biotechnology, operating at the more fundamental genetic level, renders the process of change more specific to a given task and is more scientifically deterministic. To be somewhat more specific, modern biotechnology is made possible by the recognition that •

proteins are the workhorses of living cells

•

genes are recipes for proteins

•

specific genes are linked to specific cells

In other words, biochemical processes observed in living organisms can replace conventional chemical synthesis to produce proteins and other chemicals of biological value. The techniques of modern biotechnology can be broadly described as •

recombinant DNA (rDNA)

•

monoclonal antibodies (MAbs)

•

bioprocessing techniques (cell fusion, cloning)

Against this backdrop, the value chains, leading to commercialization, consist of the following links.

(ix)

Introduction

187

• Identifying and cloning the gene of interest (molecular biology) • Bioprocessing (scaling up of laboratory technology) • Clinical testing (of pharmaceutical products and field trials of crops) • Production and marketing of final products It was suggested that biotechnology •

has the prospect of reversing the ill effects of the earlier technology (usually designated as bioremediation), • will probably ensure a greater sustainability of new patterns of economic development However, progress in biotechnology has been somewhat slow probably because the limits of traditional technologies have not reached menacing proportions as yet. Of late, biotechnology and associated aspects of commercialization are approaching a relatively stable pattern though further developments and changes are expected. A study of the economics of biotechnology therefore appears feasible at this juncture. The most distinctive characteristic of modern biotechnology is its knowledge intensity both in the development of the basic science and bioprocess engineering. This has two dimensions. Firstly, the R&D, that is necessary to develop scientific knowledge, cannot be taken up by the large firms. For, they do not have the organizational culture and structures to undertake basic research. Perforce, the R&D developments originate in universities and research laboratories. Secondly, consider the process of such knowledge transfer to private firms. In the context of conventional mechanical and chemical technologies the following paradigm appears to operate. The university scientists that develop technologies train their students in their use. In turn, private firms employ these students. They assist in the transfer of technology and its implementation. Over the years, there is an adequate pool of expertise (both in learning new knowledge as well as developing tools to implement them) within the private firm. Hence, it would appear that transfer of formal knowledge is adequate only in a mature state. Since biotechnology is in its early stages private firms do not have knowledge workers and they also do not have any experience in constructing and implementing bioprocessing tools (fermentation processes may be somewhat of an exception). Hence, the scientists, who developed the technical knowledge, must work with the companies that undertake higher levels of processing so that informal knowledge (intricacies in its actual use) can also be transferred for its efficient adaptation. Such necessity for intricate relationships persists even at the level of clinical tests (of medicines) and field trials (of crops). These features make it necessary to visualize and implement novel organizational forms. They include • • •

networking of major manufacturers with small biotechnology firms (that predominantly deal with knowledge intensive phases of the value chain) strategic alliances with multinational corporations full vertical integration (through mergers and acquisitions) and so on. On a few occasions, like the Human Genome Project, even open source organizational structures have been considered efficient. Economic analysis concerns itself with the

188


(x)

relative efficiency of one of these arrangements over the others in terms of costs and the private (as well as social) benefits. Consider the network technologies, viz., transportation, telecommunications, and information technology. Some of these are considered as public goods. To appreciate this, a good road network is necessary to have an efficient transportation system. Similarly, a telephone network is efficient only when a large number of connections can be handled. The internet also has similar features. Consequently, during some phases of the development of such technologies it was felt that the large and lumpy investments cannot be recovered by private firms in a reasonable time and/or the products (like the internet) were expected to have largely defense or public use. Public investment flagged off such industries and eventually, once the base is set, private investments became viable. No such defense needs or public good properties seem to apply in the context of biotechnology. Much of the R&D and investment have been in the private sector though on occasions public expenditure has been discernible. Large investment requirements are one feature of biotechnology. Further, since the developments are in their early phases, there are pronounced risks of failure. These features add to the necessity for organizational networking to share specialized knowledge and the associated financial risks. These organizational arrangements determine the patterns of R&D and investment financing in biotechnology. One of the primary concerns of economic analysis relates to the degree of monopoly generated in the process and the associated implications for biodiversity (especially in the context of agricultural biotechnology and animal population). Other significant features of R&D in biotechnology have been pointed out. First, scientific discoveries are slow and consist of discoveries of the structure of a few cell lines, fragments of protein structures and so on. Development of a final product (of value to consumers) therefore takes time and may depend on the expertise of several scientists. Second, imitation is simple, once the scientific knowledge is available, though the initial discoveries are not. These features made it necessary to protect property rights of scientific knowledge. Secrecy, as a mechanism to protect such knowledge, inhibits swift and efficient progress of technology since no one individual or laboratory can have all the requisite competence. Patents and IPRs, at the level of novelty of the discovery of scientific knowledge (as opposed to a demonstration of the utility to a consumer), was thought of as an efficient way of protecting property rights while ensuring a faster diffusion of useful knowledge. These features of biotechnology exclusively motivated the change in the patent regime. It was also felt that patenting knowledge (not merely products of use to consumers) is feasible because biotechnology developments are not cumulative. In particular, it was noted that the discovery of one protein does not provide the firm any specific advantage in discovering another. Note, however, that once patent rights are granted, the exclusivity clause meant that other firms and/or scientists can obtain information only on a contractual basis. Monopoly elements in these transactions slow the process of knowledge diffusion. The ensuing market structure and contractual arrangements condition the pace of progress in biotechnology. They also determine the sharing of risks and benefits among the stakeholders. Clearly, the economics of biotechnology must come to grips with these issues.

(xi)

Introduction

189

There is fragmentary evidence that biotechnology will be more productive and cost efficient, less expensive to consumers, or very profitable for producers except when they charge exorbitant prices. Developed countries push for patents and IPRs only to ensure that their industries and firms are established before any real crisis emerges. Others resist it because they do not want this dependence on any large scale. As noted above, no new technology can be said to bestow benefits without creating any diseconomies. There is a fear that the biotechnology revolution, which operates at the more basic genetic level, may give rise to mutations that may have far more devastating effects on human, animal, and plant welfare. Private benefits of conglomerates, especially of the short run variety, may overshadow these consequences as biotechnology gathers momentum. Similarly, trade across national boundaries becomes especially susceptible to abuse if the labeling and testing procedures are inadequate, time consuming, and/or expensive. Economic analysis is concerned with these tradeoffs as well. Devising suitable regulatory mechanisms and examining their economic implications becomes mandatory. This is also an essential aspect of the economics of biotechnology. As in much of the industrial organization literature there are two perspectives from which a study of biotechnology can be approached. • The firm (micro perspective) • The developmental (macro) perspective This book adopts the first route. The second approach is reflected in some books available on the market. The reader may wish to supplement that information as the need arises. The present book endeavors to cover many of the aspects of biotechnology alluded to above. It is at an elementary level so that even an undergraduate student, familiar with some basic microeconomic theory, can appreciate the issues and their resolution. Some basic fundamentals have been reviewed in appendices 2 and 3 to assist the reader. The material covered in this book was presented in an undergraduate class. I benefited from the interaction and comments. Similarly, some of my colleagues and friends read the manuscript and offered useful advice. They were helpful in improving the readability of the manuscript. I am thankful to all of them. I would consider my effort worth while if the reader develops interest in this emerging area of economics.

Kanpur November - 2006

T.V.S. Ramamohan Rao


190


CONTENTS (vii)

PREFACE

Chapter 1 INTRODUCTION 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8

1-16

CHANGING TECHNOLOGY ...1 MODERN BIOTECHNOLOGY ...2 ORGANIZATIONAL ISSUES ...5 PATENTS AND IPRS ...6 MARKET STRUCTURE ...9 ETHICAL AND ENVIRONMENTAL CONCERNS ...12 GOVERNMENT POLICY ...13 LOOKING AHEAD ...15

Chapter 2 ORGANIZATIONAL STRUCTURE

17-34

Chapter 3 IPRS AND PATENTS

35-58

2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8

3.1 3.2

THE SCOPE ...17 KNOWLEDGE INTENSITY ...18 BIOTECHNOLOGY KNOWLEDGE ...20 NETWORK ORGANIZATION ...22 NATURE OF CONTRACTS ...25 SHARING FIXED COSTS ...30 ECONOMIC CONSEQUENCES ...32 OTHER ASPECTS ...33

WHY PROTECTION? ...35 PATENTS AS PROTECTION ...38

(xiv) 3.3 3.4 3.5 3.6 3.7 3.8

79-96

THE BACKGROUND ...79 PATTERNS OF DEMAND ...81 PRODUCTIVITY ...85 VARIABLE COSTS ...89 WELFARE EFFECTS ...93 SUMMING UP ...96

Chapter 6 MARKET STRUCTURE AND PRICING 6.1 6.2 6.3 6.4 6.5 6.6 6.7

59-78

THE ISSUES ...59 ROLE OF PUBLIC INVESTMENT ...61 R&D IN SCIENTIFIC KNOWLEDGE ...64 RISKS OF R&D ...66 COMPLEMENTARY R&D ...69 AGRICULTURAL EXTENSION SERVICES ...70 BIOPROCESSING ...72 PHYSICAL CAPITAL ...73 FINANCING CONSTRAINTS ...76 FURTHER CONSIDERATIONS ...78

Chapter 5 DEMAND, COST, AND PRODUCTIVITY 5.1 5.2 5.3 5.4 5.5 5.6

191

BIOTECHNOLOGY PATENTS ...41 IPR AGREEMENTS ...44 TRIPS AGREEMENT ...46 CONSEQUENCES OF PROTECTION ...50 ISSUES OF CONCERN ...53 MODIFICATIONS TO PATENT REGIME ...54

Chapter 4 INVESTMENT AND FINANCING 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 4.10

Introduction

NATURE OF MARKETS ...97 DEFINING MARKET CONCENTRATION ...99 SOURCES OF CONCENTRATION ...101 MONOPOLY POWER AND PRICING ...102 DIFFERENTIAL PRICING ...106 DYNAMIC PRICING ...110 IN RETROSPECT ...110

97-112

192


(xv)

Chapter 7 ETHICS AND ENVIRONMENT 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8

ISSUES AT STAKE ...113 ETHICAL ISSUES ...114 ENVIRONMENTAL ISSUES ...115 ETHICAL ISSUES IN AGRICULTURAL BIOTECHNOLOGY ...115 ETHICAL ISSUES IN DRUG DEVELOPMENT ...118 ENVIRONMENTAL ISSUES ...119 INTERNATIONAL CONVENTIONS ...121 LESSONS AND CONTROL ...124

Chapter 8 GOVERNMENT POLICY 8.1 8.2 8.3

125-135

AN OVERVIEW ...125 SCIENTIFIC R&D ...127 SCIENTIST VS. NBF CONTRACT ...132

Chapter 9 CONCLUSION 9.1 9.2 9.3 9.4 9.5

113-124

137-141

THE TECHNOLOGY ...137 ORGANIZATIONAL ASPECTS ...138 PRODUCT PROFILES AND MARKETS ...139 NEGATIVE EFFECTS ...140 STEADY STATE ...141

APPENDICES APPENDIX I TECHNICAL TERMS APPENDIX II ECONOMIC CONCEPTS APPENDIX III MATHEMATICAL BACKGROUND REFERENCES INDEX

143-164 145 149 157 165 179


Chapter 1

INTRODUCTION

1.1 CHANGING TECHNOLOGY Traditional agriculture was based on using nature’s bounties for crop production. In particular, conventional farming • • • •

uses seeds preserved from the previous crop depends on natural sources of water utilizes natural manures (fertilizers) like cow dung, and obtains assistance from earthworms and wooden ploughs to improve the quality of land The productivity of land, that such farm practices sustain, turned out to be inadequate for the food security necessary for the growing population. Chemical technology improved the yield of crops by providing • fertilizers (nitrogenous, phosphatic, and sodium) to enhance the growth of plants • pesticides to control worms and insects that destroy crops, and • herbicides to remove the growth of weeds Nitrogenous fertilizers are based on the naphtha feedstock obtained from the distillation of crude oil. Many chemical products (like Vaseline and aromatic compounds) also have crude oil distillation as their base. Some of them involve fairly complex chemical technologies. The acknowledgement that oil reserves are depleting very fast created an awareness to explore alternatives. The usefulness of biochemical reactions, albeit at a rudimentary level, has long been acknowledged. For instance, bacteria are the essential change agents that convert milk into yogurt. Similarly, yeast is used in brewing beer and wine. Though the existence of such

2


biochemical processes is well known (once they are established through a trial and error approach) the reasons for their occurrence is not well established. The emergence of modern biotechnology is changing these production processes significantly. This has the potential to operate in such areas as • genetically modified seeds • biofertilizers • biopesticides and herbicides • specialty chemicals Consider the context of producing fruits, vegetables, and flowers. Plant breeding conventionally meant identifying robust plant varieties, cross breeding to obtain more and better productive varieties and so on. These methods and their success depend on random trials. On the other hand, modern biotechnology goes a step further by genetically modifying the plants. This process is more scientific, systematic, and predictable. Hence, the fluctuations in productivity associated with conventional plant breeding can be largely eliminated. (Of course, it is not always easy to establish the superiority of modern biotechnology in terms of the growth in average productivity per se.) The animal world is also expected to undergo a major shift in reproductive biology. For, cloning with the help of modern biotechnology has the prospect of creating more and better varieties. As of today the entire pharmaceutical sector (diagnostics and therapeutics) depends on chemical technology. This has been a hit and miss process as in the case of agriculture. Modern biotechnology, operating at the more fundamental genetic level, renders the diagnosis as well as medication more scientifically deterministic. Modern biotechnology also has the prospect of providing biofuels to replace the rapidly dwindling fossil fuels as well as biochips for semi conductor applications. However, such applications will take more time. Technological advances, propelled by the steam engine, the automobile and so on as well as several chemical technologies brought about environmental degradation and pollution in their wake. Modern biotechnology (and the entire area known as bioremediation) is expected to be of valuable help in solving many of these ecological problems. Stated very broadly, modern biotechnology, though not a total replacement for conventional technologies, is changing the basic structure of technology in many areas rapidly.

1.2 MODERN BIOTECHNOLOGY Modern biotechnology consists of a set of techniques that involve manipulation or change of the genetic inheritance of living organisms. This technology utilizes some known biological processes to alter others found in nature. Such a transformation leads to a technology of greater productivity compared to what the natural processes themselves can offer.

Introduction

3

The techniques of modern biotechnology can be broadly described as • recombinant DNA (rDNA) • monoclonal antibodies (MAbs) • bioprocessing techniques (cell fusion, cloning) All the pertinent technical terms have been described in some detail in Appendix 1. Such a transformation is made possible by the recognition that • proteins are the workhorses of living cells • genes are recipes for proteins • specific genes are linked to specific proteins Consider the above mentioned techniques sequentially. The rDNA technology allows direct manipulation of the genetic material of individual cells. It can be used in a wide range of industrial sectors to develop micro-organisms that produce • • •

new products existing products more efficiently otherwise scarce products (which occur in small quantities in nature and are very difficult to isolate in pure form) • microorganisms that degrade toxic wastes • new varieties of agriculturally-important plants and so on. Cell fusion is the artificial joining of cells to form a new cell by combining the desirable characteristics of two or more cells. This technique produces large quantities of monoclonal antibodies (MAbs). They can be used in the purification of proteins and the diagnosis of diseases. As mentioned briefly earlier, a clone refers to a cell or collection of cells containing identical genetic material. Clones are produced from a single parent cell. Cloning therefore consists of using clones to produce plants or animals starting from such cells. Cloning may reproduce what is already in existence or create novel plant varieties that would not be possible using traditional breeding methods. Some of this activity may read like science fiction and involves deep ethical considerations. Biotechnology, described in the form of the above techniques, is not an end in itself. Instead, it is an input in the value chains with various kinds of final (i.e., marketable) products. The relevant industrial sectors are agriculture, specialty chemicals and food additives, animal products, pharmaceuticals, environmental control products, and bioelectronics. These can be stated in somewhat greater detail as follows. Agriculture

–

hybrid seeds biopesticides biofertilizers plant breeding

4

Economics of Biotechnology Healthcare

–

medicines vaccines therapeutics and diagnostics gene therapy

Industry

–

industrial enzymes polymers biofuels fermentation products

Environment

–

effluent and waste water management bioremediation biosensors creation of germplasms

Clearly, biotechnology is an input in the value chains of these final outputs. At a more fundamental level, the value chain of any of these products consists of the following links. • Identifying and cloning the gene of interest (molecular biology) • Creating the hybrid cells • Legal filing for approval and protection • Clinical tests (for drugs) and field trials (for crops) • Marketing the final product A somewhat related area of biotechnology is also receiving attention for its commercial importance. Note that the total amount of information, available in the form of techniques of biotechnology, is vast and growing. The human genome project, that ambitiously attempts to map the entire genetic structure of humans, adds substantial information database. The area of bioinformatics therefore attempts to utilize the now available information technology to systematize information retrieval in this context. Two aspects of biotechnology should be kept in perspective before proceeding further. •

•

Each of the techniques in the broad area of biotechnology is an essential building block. A variety of products can be developed on the basis of the same techniques. Hence, the economics underlying the development of these techniques is crucial. Most of the issues regarding the commercialization of biotechnology are similar to those of other industrial processes. However, some issues are specific to biotechnology. For instance, the ethics of using biotechnology based testing for cancer or other severe diseases. Similarly, patenting processes was deemed inadequate because market competition is at the product level. Product patents came into force from January 2005 under the WTO (World Trade Organization) regime. Even the idea of patenting fragments of knowledge (not related to a final product of value to a consumer) has been accepted in the area of biotechnology.

Introduction

5

1.3 ORGANIZATIONAL ISSUES A systematic approach to the development of the various links of the value chain is a prerequisite for any commercial success. A large variety of organizational arrangements emerged. Each of these alternatives attempts to balance organizational (core) competence and economic viability. At a very broad level the organizational designs are based on their final objective. The two distinct patterns are • technology base • product base A technology based organization uses a technique of the biotechnology sequence as an asset. It endeavors to provide mainly services. A finer division of these activities consists of • sequencing and mapping • gene target identification • gene target validation • protein target validation • identification and optimization of the bioprocess • clinical testing of therapeutics and pharmaceuticals The basic stages of scientific development and bioprocess engineering are knowledge intensive. Hence, basic scientific research is undertaken in • university departments • government sponsored research laboratories • some large chemical companies that have expertise in protein synthesis Clearly, the problem at this stage is financing because this phase of work is far removed from a marketable product. Bioprocess engineering is also knowledge intensive. However, it requires more than the laboratory tools. For, the scale of operations is different by an order of magnitude. Further, in the context of biotechnology the knowledge interface requirement is still rather high. As a result the following arrangements emerged. • Start a small biotechnology firm close to a university community • The university community accepts contract research for a large chemical firm • In-house R&D of the large chemical firm Usually the financing is by • • • •

the entrepreneur of the small firm venture capital finances from the chemical firm, or the government

6


The contracts generally stipulate payments for reaching well-defined guideposts along the way to the eventual scaling up of scientific knowledge for commercial use. It is not surprising if, after a certain amount of success is recorded, the small firm is taken over by the more resilient chemical manufacturer. For, it reduces their exposure to risk though there may be a danger that the technology leaks out before the firm derives the expected gains. The alternative is the more integrated product based organization. In this organizational form the activities are sequenced with the development and marketing of a specific product in perspective. For instance, a pharmaceutical firm may target • diagnostics (for cancer, AIDS and so on) • gene therapy • therapeutics in general • vaccines • blood products In the context of agriculture the final products may be •

seeds, e.g., Bt cotton produced by inserting the insecticide Bt gene from the bacterium Bacillus thuringiense (thereby allowing the plant to produce a toxin that destroys the gut of the invading insects) or the Roundup Ready soybeans (Roundup is a chemical herbicide from Monsanto that may destroy the crops along with the weeds when it is sprayed on the fields containing the crop). • pesticides and herbicides to be used for spraying plants (outside the seed) Clearly, the organizational structure requirements are more inclusive. In general, they may consist of • in house development in large chemical firms • obtaining technology by licensing • creating a subsidiary • forming a joint venture and so on. Investments in product based firms are more risky but they create more profits. The technology provider creates the value of the product and will endeavor to control the value and share it with other market participants based on their additional costs, risks assumed, and bargaining position. In sum, it must be acknowledged that the developments in biotechnology are contingent on substantial changes in the organization of existing firms or the introduction of new firms at various stages in the value chain. The choice of the organizational form depends on the stages of production that the firm undertakes, the tasks that it chooses to perform, the expected revenues and the specific contract clauses it chooses.

1.4. PATENTS AND IPRs The classical description of property used to relate to land, buildings, and physical things. Industrialization and ownership based on limited liability added a dimension to it. For, the

Introduction

7

ownership of stocks and shares of a firm had to be acknowledged as property. The information age and knowledge intensity of recent technology made it necessary to treat even scientific knowledge as property. This is the most crucial dimension of patents and IPRs in the context of biotechnology. Until about the late 1960s experimental biologists, unlike their counterparts in high energy physics and other sciences, were reluctant to reveal their results freely outside their laboratories. The basic reason for this was the fact that a great deal of this fundamental research could not be associated with any product from which consumers derive value. Since the patent laws at the time applied the utility doctrine, i.e., the usefulness of information to consumers of goods and services, secrecy was the only protection they had for their proprietary knowledge. Knowledge diffusion, that is very essential to develop further results and at an efficient pace, was difficult due to the secrecy. Consider the case of an individual or a firm engaged in R&D activities related to biotechnology. Their inventions may be at three levels: the basic science, bioprocess engineering, or products that can be marketed to consumers. Examples may be genetically modified cell lines that produce monoclonal antibodies for diagnostics and therapy, or genetic databases that combine sequence data with protein structure and possible function. Inventions at each stage, that constitute intellectual property, involve high fixed costs. They can be recovered, through commercial exploitation, only over a certain length of time. There is thus a necessity to protect inventions over the required time horizon. One alternative available to the individual firms is to maintain secrecy of their invention until they can themselves exploit the market. At the most they may license the use of their inventions to a selected few who can be trusted to maintain secrecy of the crucial biological material or information. Where secrecy is used to protect intellectual property, access to materials and information relies on the negotiation of private contracts between the parties. It may then be possible to recover the market value of the invention. This approach protects the original inventor from piracy or imitation by competitors. However, this has two deleterious effects. First, it forecloses the possibility of others developing competitive technologies to produce the same product. Second, once the invention is known, it may be possible to develop related products that are also of commercial value and value enhancing to society. In particular, the pace of the genomics revolution (and the mapping of human genome sequences) and the diffusion of those research results (including the algorithms to analyze them) depends on open access to genetic data. The original inventor may not work on the development of derivative products if the costs are too high and/or there is a fear that they will erode profits from the original discovery. This will happen even if there is an expected increase in social welfare. The alternative is to provide the inventor monopoly rights for its use and/or sale for a limited time period. However, it is necessary to ensure that the veil of secrecy is lifted. Hence, the basic condition for granting a patent is that the inventor must disclose the innovation in a written description in such a way that any knowledgeable person can put it to practical use (move forward to develop alternative bioprocesses and/or derivative products).

8


There were objections to patenting biotechnology initially. The United States Patent and Trademark Office (USPTO) argued against patenting of genes on the grounds that they are • discoveries (identifying something that already exists) and not inventions • products of nature and not new • the basic core of humanity should not be owned by anyone as property However, a 1980 U.S. Supreme Court judgment changed all that. The Diamond vs. Chakraborty case was about the patentability of a genetically modified bacterium. The court held that such genetic material is patentable because there is novelty. Subsequent gene or DNA patents have claims that they cover nucleotide (DNA or RNA) sequences that encode genes or fragments of genes. As a general rule, of course, patents cannot cover a substance in situ (inside and resident) in the human body. But they can if they are isolated from their natural source. The principal requirements for patentability are that the invention is new (is not already documented), involves an innovative step (not otherwise obvious), and has industrial or other useful capability. In biotechnology applications the common forms of claims involve • an apparatus or a device • a process or a product that can be manufactured • a method of treatment or testing (diagnostic tool) In general, the inventors may include in the patent claims a wide variety of related activities that can be developed in the future on the basis of the basic invention. The scope of claims (or, alternatively, the breadth of claims) will constitute a legal part of the patent. Two issues have therefore become pertinent. First, bioprocessing of basic discoveries towards a marketable product may take several forms. In the IPR act of 1970 the Indian government sought to facilitate cheap technology acquisition and to enhance technological self-reliance. Unlike the Paris Convention it restricted the range only to process patents and not product patents (including imported products with patented technology). This enabled the Indian pharmaceutical industry to use its reverse engineering competence to develop generic drugs for the Indian market. Under these conditions a process patent loses its value before the fixed costs are recovered. Hence, under the current WTO dispensation the patents are for products irrespective of the basic science and bioprocessing concepts involved. In a similar fashion the Union for Protection of Varieties (UPOV) 1991 amendment for crops and plants • •

disallows farmers the privilege of retaining or re-using seeds for self-cultivation requires a plant breeder to buy genetic dependency rights before he can market a “cosmetically bred” variety. The general justification is that the yield vigor of genetically modified crops decreases in subsequent growing seasons. There are several concerns about the ill effects of possible genetic mutations as well. Second, the broader the patent the more difficult it is for others to develop derivative products. As a consequence, the incentive for the original inventor to invest further may also be dampened. Hence, the patent regimes have shown a concern that granting a broad patent too early may inhibit development of related products that assure biodiversity. This concern is quite evident in the context of plant and animal kingdom.

Introduction

9

The 1970 Act provided patent protection of 7 years for food, chemicals, and pharmaceuticals. It was 14 years for the other products. Under the Paris Convention it was 20 years for all. However, the current WTO TRIPS agreements impose a uniform 20 years for all products. There is a different kind of concern with patenting diagnostic kits for cancer and other dreaded diseases. The point is that a doctor’s use of these on a patient may not be ethical if the patient has a chance of longer survival if he/she does not know that he/she is suffering from and/or susceptible to such diseases. On the whole patents • may inhibit technological progress and increase monopoly power, or • accelerate progress and competition by sharing information Judgments, about one or the other being dominant, are contested.

1.5 MARKET STRUCTURE Developments in biotechnology, both in agricultural applications and pharmaceuticals, proceed in four distinct stages. In the context of agricultural applications they can be described as follows. •

DNA sequences from various organisms (plant and bacteria) are inserted into the genome of chosen plant variety. The new gene constructs have economic value if they have the desired characteristics (herbicide tolerance or insect resistance). • The innovator, or a new biotechnology firm, field tests the variety following government regulation requirements. • The firm then seeks deregulation from the government. Once it is granted the firm may cross this variety with others and pass on its genetics without the necessity for further approvals. • Industrial production and marketing of the transgenic variety is generally taken over by the large chemical companies already in the market. The pharmaceutical sector also develops through four similar phases. • The university or research laboratory identifies new genes or proteins as useful in diagnostics or therapeutics. • A new biotechnology firm (NBF) sets up a process for producing the drug. • A NBF or a large pharmaceutical company then processes it through the five mandatory phases of regulatory approvals. They can be designated as follows. 1. Filing an initial application with the regulator. 2.

Phase 1 trials about the safety of the drug in healthy subjects.

3.

Phase 2 trials about the effectiveness of the drug in a small sample of patients with the target disease.

4.

Phase 3 trials about the efficacy of the drug in a large sample of patients.

5.

A new drug application for regulatory review and approval.

10


• A large pharmaceutical firm then produces and markets the drug. The firms in both these lines, if they can be so designated, are of two broad types. •

Research intensive companies that specialize in discovery and development of new biotechnologies. • Firms that specialize in commercialization and marketing activities. The markets therefore consist of the linkages between agencies which produce basic science, bioprocess engineering, regulatory approval, and production and marketing. The usual contracts between them are for the offer of inputs, sharing costs, and sharing revenues received (royalties). The necessity to protect secrets of intellectual property makes these alliances take the form of networks. Their relationship and the choice of contract terms resemble a bilateral monopoly. On occasions the contracts also provide the large firms the first option to license new discoveries by NBFs. More specifically consider the pharmaceutical sector again. One view of the evolving market structure is that small companies will specialize in discovery and the large firms (which produce chemical based drugs and are already familiar with clinical testing procedures, the regulatory processes, and have the physical assets and marketing expertise) will concentrate on the production and marketing of final products to the consumers. The contracts between the firms at these successive stages therefore depend on the • ability to raise finances from the government and private sources • share of costs accepted by the parties • risks involved in the transactions, and • prior experience of the parties in negotiating deals. Consider the market interface between the pharmaceutical firms and the consumers. A firm, in this context, is not merely the final producer of the new drug. Instead, it consists of the network of firms (institutions) that develop the drug. In practical terms this means that there is a high degree of concentration for a specific therapeutic category even if it is over the limited time of the patent. This is a consequence of the large fixed costs and possible economies of scale. However, the drug market exhibits two other specific features. •

The drug discovery process is not cumulative. In particular, innovations in one therapeutic category do not guarantee a higher probability of success in another. Hence, no one firm can command excessive monopoly power. • The market demand for firms is fragmented. That is, the market leadership in any one drug does not assure the firm that they will gain in related product markets. However, the growth of the firm will depend on the number of drugs that it discovered and marketed (horizontal diversification). In general, though, no single firm can hope to command a large market share in the pharmaceutical industry. An exception may arise if a large pharmaceutical firm is marketing substitutable drugs based on chemical technology and biotechnology. For, there is a high degree of substitutability on the demand side and there are economies of scope on the supply side. The economies of

Introduction

11

scope are obviously a result of common fixed costs incurred in clearing regulatory requirements and the marketing and distribution expenditures. (Some details of the concepts and sources of economies of scale and scope have been outlined in Appendix 2.) In certain therapeutic categories the new drug produced from biotechnology may directly compete with the one based on chemical technology and produced by another firm. The markets and the prices for the drug will then depend on consumer preferences. New entrants into final product markets, when they are competing with a large pharmaceutical firm, are likely to face competition because they lack marketing experience. Return to the market for agricultural biotechnology. There are some obvious differences in comparison to the pharmaceutical markets. •

•

•

•

•

Suppose a new gene discovery can be inserted into a number of crops. Examples are the familiar Bt cotton and Bt corn or the Roundup Ready soybeans and cotton. It so happens that there are many more varieties of Roundup Ready soybeans than there are of cotton. The soybean market is more competitive. However, notice that all varieties are not equally efficient in all soil climatic conditions. This limits their competitiveness. New discoveries in agricultural biotechnology are cumulative. A variety of corn that produces more oil is initially discovered. In subsequent iterations the crops are made pesticide free. Therefore, there are greater barriers to entry and the degree of competitiveness is lower. Seeds, that contain the genetic coding, are both complementary and substitutable to conventional chemicals and herbicides. For example, Roundup Ready soybean seeds are complementary products to the glyphosate in Roundup. They are, however, substitutable to the herbicides traditionally utilized to control weeds in soybean crop. Strong demand complementarities suggest that a single firm producing both these products will be more profitable. For, this firm can price its products so that the use of complementary products (tying sales) can be encouraged. Seed companies have sufficient monopoly power. For example, Monsanto makes the farmer dependent on its chemical herbicide Roundup if Roundup Ready seeds are used in the cultivation of soybeans, cotton and so on. Hence, they are in a position to charge a technology fees in addition to a price for the seed. In part this may be necessitated by the high fixed cost of producing biotechnology embodied seeds. There are fairly stringent standards to ensure food safety. For instance, in the U.S.A. the tolerance level of Bt toxins is 5 percent. In the European Union it is even more stringent at 1 percent. Even so, the consumer preferences are still toward non-GM (genetically modified) foods in contrast to GM foods. This sensitivity has led the firms to mandatory labeling of foodstuffs. The costs of transportation, storage, and marketing of non-GM foods increased as a result. This aspect of the market inflates the prices of non-GM foods.

12


The economic analysis of the markets for biotechnology must reflect these features explicitly in its modeling efforts.

1.6 ETHICAL AND ENVIRONMENTAL CONCERNS From the time of the green revolution farmers have been utilizing large quantities of fertilizers, other nutrients, pesticides, and so on. One of the major problems has been the reduction of soil fertility and the ability of the soils to regenerate their productivity. Secondly, a substantial portion of these chemicals are drained out into the surrounding fields. This too has the potential to adversely affect farm productivity and reduce the capacity of the ecosystem to return to normalcy following the initial disturbance. Thirdly, it was observed that crops produced on such farmlands contain pesticide residues. In sum, ecological and health related issues have been raised in the context of the use of conventional chemicals. Genetic modification of crops raises a series of related issues. First, whenever plants contain traits of herbicide tolerance and insect resistance there is a risk of such species surviving the crop cycle and leading to the creation of more herbicide resistant weeds and insects. That is, the planting of GM crops may induce genetic alterations of wild plants and genetic pools of major insects. Second, when agricultural inputs flow into surrounding farm lands the possibility of aggressive insect populations finding their way into those farm lands increases. In addition, wind blown pollen from Bt crops may affect natural surroundings. In particular, it was reported that such pollen has killed insects like the monarch butterfly. Third, there has been a concern that Bt crops may damage biochemical cycles. Toxic wastes from such GM plants may enter the soil through the roots. Similarly, such toxins may also reach the soil when the plants are decomposed on the farm after the harvest. In turn, they may affect a whole range of interacting species like bacteria, viruses, and insects. At a different level, there have been concerns about the safety of GM foods with respect to human and animal consumption. It has been suggested that the new proteins formed in the food products based on biotechnology may • •

themselves act as allergens or toxins alter the metabolism of the food producing plant causing it to produce new allergens or toxins Equally important is the concern that the genetically modified foods reduce their nutritional value. For instance, it was pointed out that herbicide resistant soybeans contain smaller amounts of isoflavons, which are important phytoestrogens, believed to prevent many cancers in women. Historically, farmers relied on the seeds produced on their farms for crop production in subsequent years. This resulted in natural adaptation of seed varieties to local agro-climatic conditions. Such rich biodiversity had an essential role in preserving the productivity of farm lands. The widespread use of homogenous genetically modified varieties will unavoidably lead to erosion of such biodiversity. The problem gets compounded by the fact that the earlier non-GM varieties deteriorate over time if they are not cultivated over long stretches of time.

Introduction

13

The risks of biotechnology may then be summarized as follows. • • • • •

Invasive weeds and insect species may be created Wild pools of species may be contaminated Monoculture may lead to the neglect of traditional varieties There will be a reduction in biodiversity There will be negative effects on humans and animals when they consume GM products. In the final analysis the policy maker must examine the trade-off between the positive and negative effects of both conventional chemical use and the new biotechnology adaptations. The Cartagena protocol on biosafety 2003 is the main international agreement that provides guidelines to national governments in issues pertaining to biodiversity. The approach is precautionary. It attempts to prevent harm by banning uncertain products. Similarly, the Indian Environmental Act 1986 attempted to ensure biosafety by stipulating that while planting Bt varieties the farmers should maintain a buffer zone around their farm areas. In the final analysis, disseminating information and labeling products may shift the choices to the consumer. Two major concerns have been expressed with regard to the therapeutic applications of biotechnology. • • •

There is a possibility of creating antibiotic resistant genes The new drugs may have potential allergic reactions While diagnostic kits for cancer, AIDS and so on may be an advantage to those that voluntarily use them there are ethical issues about the physicians using them on patients without their consent. The pharmaceutical market is far more organized compared to the agricultural sector. Hence, GM products of the pharmaceutical companies, when they are released on the market, may be more reliable.

1.7 GOVERNMENT POLICY There is widespread recognition that basic scientific research • emerges only as a result of uncertain search however organized it may be, and • takes a long time before anything of commercial value emerges. Private firms do finance some scientific research. However, history is replete with examples where the basic research funding is initially provided by the government. One of the more recent examples is that of the internet. It was initially funded and developed for defense purposes in the U.S.A. The search for new materials, especially those with potential for replacement of fossil fuels or solving major health problems, have been receiving attention from about 1970. Basic biotechnology research is of this genre. Successful research is then transferred to the private sector for commercial exploitation.

14


In general, governments refrained from directly undertaking or financing applied research. But it was found that government intervention is essential especially when quick developments and practical use are indicated. The classic example is that of penicillin. Briefly recall the following well known facts. In 1928 Alexander Fleming identified that penicillin mould had antibiotic properties. Oxford scientists then spent almost a decade isolating the essential agent, producing it in laboratories in small quantities, and testing its clinical efficiency. The arrival of World War II necessitated the manufacture of penicillin on a large industrial scale. The engineers, at the U.S. department of agriculture, working on fermentation technologies, utilized government finances to produce output of penicillin in industrial scale fermentors. After the war private firms used this technology in the production of a variety of antibiotics. In the context of chemical as well as biotechnologies there is an acknowledgement that different process engineering methods can lead to the same final product. In the absence of product patents a firm engaged in bioprocess engineering may not be in a position to recover its sunk costs if rivals undercut it. This feature is the basic reason for government financing of some applied research as well. In the context of biotechnology it has also been recognized that there will be lengthy approval processes even after an industrial scale bioprocess is identified. This feature contributes to further risks of private funding of such research. In some cases government funding of pharmaceutical products, until they are cleared through phase 3 clinical trials, has been sought. Alternatively, it was argued that 20 year product patents should apply only after a drug receives approval to market it. There have been debates about the efficient sharing of investment between the government, private venture capital, and the private sector. The alternative of tax concessions to the private sector for such applied research has often been considered. Some areas of pharmaceutical biotechnology are under strict government supervision both because of their cost implications and ethical concerns. They include • laboratory production of stem cells • regeneration of human tissue • possible generation of insulin producing cells in the pancreas of diabetics. The transfer of a successful technology to the corporate sector has also been a matter of government policy. The following issues are pertinent. •

To what extent should the scientists, who are originally responsible for the inventions, or the institutions they represent be compensated? • Will a technology licensing and royalty share contract be efficient? • Will a profit tax imposed on a private firm be adequate for the government to recover the initial investment they make toward technology development? The government is also seized with issues of protecting intellectual property from piracy at least until the innovating firm recovers its sunk costs. The pros and cons of providing such monopoly power have been extensively investigated.

Introduction

15

Consider the context of agricultural biotechnology. The farming community is fragmented, accustomed to certain practices (such as replanting seed saved from the farm), and find it difficult to acquire the new knowledge and practice it efficiently. The need for extension education is far more important now than it was earlier. The costs, to the private firms producing GM seeds etc., is excessive. For, the fine tuning of agronomic processes is highly location specific. Government organizations are perforce the change agents even in such a context. The products of biotechnology, in the areas of agriculture as well as pharmaceuticals, require a close scrutiny and on a recurrent basis simply because they have profound implications for the safety of human health. Setting standards, testing and certification, and monitoring performance have also become essential functions of the government. In general, providing appropriate information coupled with monitoring has become an important function of the government agencies. Will these government policies be sufficient to ensure that • •

the farmer and the consumer get fair prices the price of pharmaceuticals and the quantities supplied adhere to welfare maximizing norms? We cannot be sure of these once a monopoly is granted to some firms. Once again it is imperative that the government restructure antitrust policies, price controls (in the form of differential pricing) and so on. Drug delivery is often conditioned by national health service policies. For, the services rendered through this channel are as significant as the health care through private hospitals. In other words, government policy has a profound impact on the demand for health services in general and diagnostic and therapeutic services in particular. It was already noted that concerns of biodiversity and environmental impacts have resurfaced at a different and perhaps more difficult level. This is so despite the fact that certain discoveries of germplasms are helpful in solving some of the environmental problems. Government policy is also directed towards such issues. Both agricultural biotechnology and pharmaceutical developments are such that they tend to displace conventional chemical technologies and other related methods. They displace workforce and render them obsolete unless their skills can be upgraded. The private sector firms will not pay attention to this rehabilitation requirement. Government intervention is again called for.

1.8 LOOKING AHEAD Recall that both agricultural and pharmaceutical sectors have been dependent on chemical technologies for well over fifty years. The most prominent technological developments were based on crude oil and its distilled components as the basis. The erosion of this basis and the ill effects of the use of petroleum products have been the driving forces behind the recent technological developments. In particular, the biotechnology revolution is one such response.

16


The basic purpose of economic analysis of biotechnology is to examine the following issues. • • •

What are the inputs necessary to make the technology develop? What are the investments and costs of developing the technology? What organizational arrangements will efficiently move the technology from the laboratory to the market and the consumer? • Will such arrangements necessarily involve government policy? What are the consequences of patents, government financing of R&D and so on? • What will be the effect of new technologies on prices of products and consumer welfare? In particular, what are the economic consequences of biotechnology displacing conventional technologies? • What are the regulatory measures necessary to ensure safety in terms of human health? What price and welfare effects will such regulatory practices have? The fundamental analytical tools are those of micro-economic theory and industrial organization. Networking is one of the basic principles that need attention. The possibility that product patents of biotechnology will displace conventional technologies and thereby endanger biodiversity is of real concern. The economic implications of this are fairly widespread and need urgent attention. The conflict between private and social benefit at almost all levels of the organization are evident. For example, in its seed contracts, for Roundup Ready soybeans and cotton, Monsanto stipulates that the farmers should not preserve seeds even for their own reuse. This is important to maximize their profit from the sale of seeds. Is it possible to argue that this maximizes social welfare? If not, can contract law and IPR regimes be so structured as to eliminate the ill effects? The economics of biotechnology must address even such issues explicitly.

Market Structure and Pricing

17

Chapter 2

ORGANIZATIONAL STRUCTURE

2.1 THE SCOPE Every firm, irrespective of the industry in which it is operating, endeavors to • •

monitor opportunities for its growth identify its existing competencies and develop new competencies to take advantage of its opportunities • implement its choices by developing an appropriate flow of information and materials within the firm. In such a milieu the external environment of the firm will consist of the • technological knowledge • market for new products, and • sources of finance It may contain signals with respect to • opportunities and profit making avenues, or • threats and problems that inhibit development The internal environment, especially its organizational structure and the core competence of the management, may have to undergo substantial changes if the firm wishes to realize the expected benefits. In particular, the organization of the firm should have • • • •

sensors to monitor and analyze the external environment develop suitable strategies in reaction to these changes capabilities to make the requisite adjustments, and resources to implement efficient adaptations

18


In this context the organizational structure defines the • scope of the organization (e.g., in terms of the extent of vertical integration) • division of labor between the subunits, and • flow of work within and across subunits Individuals, associated with the firm and/or working in distinct divisions of it, • offer their expertise • learn from each other, and • internalize existing values and/or redefine organizational goals Such organizational learning, and the synergies derived from it, are essential ingredients for the efficiency of the firm. Traditionally, the organizational units within the firm • maintained their decision making autonomy, and • bargained with the general manager for a common pool of assets This is the general nature of a M-Form organization. In sum, the basic interest of the economic analysis of organizational structures is to understand • • • •

how certain economic forces govern the choice of organizational structure how organizational structures affect productivity and growth of firms how explicit contractual arrangements between the subunits of the firm emerge, and the implications of organizational arrangements for market structure and consumer welfare

2.2 KNOWLEDGE INTENSITY One feature of traditional industries, viz., the mechanical and chemical based activities, should be recorded. Most of the technological developments in these industries were a result of learning-by-doing and cumulative use of technologies. Much of the R&D was, therefore, a result of the efforts of the workers within the firm. It can therefore be said that the knowledge necessary to develop products and new technologies was easily available in most firms. The more recent industries, such as information technology and biotechnology, are knowledge intensive. •

The knowledge available within the firm is inadequate to develop technologies and/or products. • Scientific and/or specialized knowledge must be obtained from outside sources • Assimilation of knowledge requires constant interaction with experts. To a large extent this is a result of the fact that these technological developments are relatively new whereas the traditional technologies attained a steady state much earlier.

Organizational Structure

19

However, it must be noted that the discovery of new chemical entities for pharmaceutical applications and other technological developments in traditional industries have been proceeding on the basis of random trials and a hit and fail method of analysis. On the contrary, developments in the emerging technologies appear to be far more logically organized and calculated. To that extent their discoveries are knowledge intensive. Similarly, to the extent that traditional technologies are in a steady state, the basic scientific knowledge is well understood, documented, and transmitted to the workers and engineers employed by firms. In this sense the workers in those industries already have the scientific knowledge and practical experience in their use to facilitate their participation in the requisite technological development. It can also be claimed that most of the innovations in these traditional technologies have been minor and the scope for any major breakthroughs is small. The new technologies did not reach such a state as yet. Hence, there is a shortage of trained scientists that can be employed in the industry and entrusted with the development and discovery of technology. The newness of these technologies also indicates that every new development is a major innovation that cannot be built in house based on learning-by-doing of the existing scientists and engineers. Dependence on outside scientific knowledge is a result of this state of the technological developments. Knowledge intensity should be interpreted in this specific sense. Therefore, the emerging knowledge intensive technologies do not strictly confirm to the organizational structure logic alluded to above. Instead, they require a different type of organizational structure for their success. In the context of information technology the requisite scientific knowledge is often generated within a firm and protected mostly by copyright. However, it cannot be presumed that all requisite knowledge can be developed and provided in house. It is necessary, at least on some occasions, to interface with other firms (especially smaller firms that specialize in developing fragments of a larger product). There is a necessity for product specific networking in either case. Such networks may be created from knowledge and talents within the firm, obtained through open source strategies (hacking and value enhancing contributions, not destructive activities, to source code is an example), or through business process outsourcing. To elucidate this further, consider the following. Assume that the firm receives a contract to develop a new source code for a specific task (usually called a product). To complete the task it may be necessary to pool together the core competencies of two or more individuals (or divisions) within the firm. A team (or network) is formed for a short time to complete the task. It will be dismantled once the product is delivered. A different network is assembled as and when a new task is to be achieved. Note that the network may elicit help from individuals in another firm (e.g., a software engineer from Tata Consultancy Services in India may be drafted by Erickson of Sweden). The team will be under the control and management of one firm until it delivers the product. Outsourcing is possible when the job can be divided into independent modules so that close collaboration is not needed. The team members remain with their respective organizations and get the job done. (It may be acknowledged that individuals drafted into a team need not

20


belong to any specific organization. They may be independent experts providing their services on a contractual basis.) Outsourcing is possible even when there is some cumulativeness in the development of the product. Of course, this involves greater coordination between the constituents of the network. Open sourcing is a peculiar network when no specific product is targeted (it evolves over time without any premeditation) and there is no clear a priori knowledge of which set of hackers will be in a position to add value. Note that contracting knowledge from outside eliminates the costs of developing and maintaining a large variety of talents in house. The disadvantages of this approach relate to maintaining an information base and the costs and risks associated with short term contracts (e.g., frictions and delays in learning to work as a team as well as explicit payment mechanisms). The specific choice of network relations depends on these relative costs. In general, when they depend on knowledge intensity, firms learn to identify appropriate network members and they in turn learn to work in teams. This process results in organizational learning, spillovers, and concomitant improvements in technology and productivity. Network organizational structures cannot survive if this is not the case.

2.3 BIOTECHNOLOGY KNOWLEDGE In the context of biotechnology, the development of a commercially viable product entails • • •

basic research bioprocess engineering and the scaling up of laboratory technology clinical testing (of medicines) and field trials (of crops) and getting government and/or regulatory approvals, and • production and marketing of the product No single firm has all the necessary expertise. Each successive stage of product development requires the expertise of the previous level. This must be obtained through networking as a result of the knowledge intensity. However, such networks are almost entirely with agents outside the firm. Some open sourcing policies have been recorded. See, for instance, Rai (2002, 2005). There are several levels at which transfer of knowledge through network structures can be conceptualized. •

•

Note that, unlike in any other area of science and technology, biotechnology activities like the discovery of a few cell lines, partial information about protein structures, and so on can be patented. The cumulativeness of knowledge towards the development of products of utility to the ultimate consumer necessitates networking in the first instance. After the initial discovery there is a necessity to scale up laboratory technology for industrial use. This phase of development is also generally knowledge intensive.


21

•

Once a firm develops a Bt or RR variety it will explore the possibility of introducing both the traits in a crop. Even such horizontal integration necessitates networking.

•

If a Bt or RR variety of one crop is successfully developed, the firm may entertain the idea of using the technology in some other crop. It was noted that this would generally require new knowledge that may not be available with the firm. This horizontal expansion of the firm also dictates networking.

However, it is obvious that a firm targeting biotechnology development in agriculture does not have any advantage in developing pharmaceutical products. At a more micro level, it was noted that the synthesis of one protein does not provide the individual, or the firm, any specific advantage in discovering another. See, for example, Giescke (2000) and Malerba and Orsenigo (2002). Networks have different constituents and tend to be event specific and of a very short run nature. In essence, different combinations of networks emerged in the context of different technologies and product applications. The three major components of these networks are •

scientists contributing basic research

•

new biotechnology firms (NBFs) involved in bioprocessing (this is the most distinct aspect of the industry)

•

large pharmaceutical and seed companies at the downstream stages

Note that the NBFs emerged as the essential link between the scientific community and the large seed companies and chemical firms engaged in marketing the final products. The knowledge interface between the scientists and the NBFs that scale up technology is significant. Further, the relationships are event specific and somewhat of a short run nature. Network organizations will perhaps be the only efficient organizational choice. Clearly, information and scientific knowledge is one pivot on which biotechnology networks hinge. The other aspects of biotechnology knowledge that need to be addressed are •

large capital investments at each stage

•

significant uncertainties associated with these technology developments

So far the emphasis was on the fact that one level of knowledge specialists do not know much about the knowledge required at the next level. It is equally true that the scientists and managers of NBFs do not always have the expert knowledge in generating the finances necessary for each level of activity. In a similar vein, it may be argued that financial arrangements also have a significant role in the development and marketing of a product of biotechnology. The large seed company or the pharmaceutical firm may have the resources but may not be convinced of the commercial viability of a new idea particularly in its early stages. A network relationship between a scientifically oriented NBF and a specialist financial company may complement each other. However, organizational learning in networks at this stage may not be as strong as it would be at the technological stages.

22

Economics of Biotechnology The sources of finance are generally from

• the government (i.e., public funding) • the venture capitalist • the capital market (including debt), and/or • the internal sources available with large firms It is necessary to visualize an organizational mechanism to obtain an efficient mix of finances if the product’s potential to maximize social welfare is to be realized in practice. In much of the literature dealing with financial economics and financial management the optimal mix is defined only with respect to the cost of financing activity. But, as Jensen and Meckling (1976) and Williamson (1988) emphasized, each of these sources of finance has idiosyncratic implications with respect to governance and organizational control. The organizational consequences should, therefore, be kept in perspective. Hence, in the context of biotechnology the organizational arrangements consist of defining an appropriate combination along both these dimensions. Different combinations of networks of firms emerged in the context of different technologies and product applications. This is one of the distinctive features of biotechnology. Powell et al. (1999) and Niosi (2003) argued that the network arrangements extend to all aspects of the value chain generally. They also pointed out that the strength of activity in each link, and not just their number, matters. This will be examined in detail in chapter 4.

2.4 NETWORK ORGANIZATION The concept of a network organization and the reasons for its emergence in practice can be examined more explicitly. Consider the scaling up of a bioprocess once the basic science is known. To begin with note that neither the conventional university organization nor that of a corporate structure are conducive to the required development if they act in isolation. In particular, university research is often • •

directed to discovery and professional publication rewarding the scientists in the form of promotions, status, honors, and research funding • supportive of free dissemination of information By way of contrast, commercial research is • supportive of only that research that has a commercial promise • secretive until the full market value is realized • owned by firms and the reward structure is limited These organizational differences are such that a formal organizational alliance between them is impractical. In particular,


23

•

information, especially knowledge regarding the use of scientific knowledge, cannot be purchased on the market for a price. Similarly, it may not be available with any subunit of a formal organization for it to be transmitted through a hierarchy • in a competitive environment, with rapidly changing technology, it is necessary to utilize opportunities as they arise and over a short time horizon. There will be significant sunk costs if the formal organization sets up a subunit to internalize every such opportunity. It may not be possible to sustain the unit over a sufficiently long time to enable the firm to recover such sunk costs A far more important problem is generally highlighted. Much of the scientific knowledge concerning biotechnology is such that the mere transfer of formal knowledge is inadequate. Instead, there is a necessity for more direct involvement of the scientist in explaining the use of scientific knowledge and making adaptations to it as the need arises. This is necessitated by the fact that scientific knowledge and technology cannot be formally codified and communicated in writing. Further, the corporate firm may not have the competence to absorb and implement such tacit knowledge even if it is offered. The appropriate form of the organization should be in a position to generate absorptive capacity in the sense of accumulated knowledge, skills, and organizational routines necessary to identify and utilize externally generated knowledge. These issues have been highlighted in Liebeskind et al. (1996) and Argyres and Liebeskind (2002). The organization of bioprocess technology therefore entails • access to complementary knowledge from sources external to the firm • learning to coordinate different knowledge skills, competencies, and assets, and • aligning incentives to foster greater harmony The network organization is a solution to the difficulties of coordinating activities of different institutions where different but complementary activities are undertaken with the commercial exploitation of scientific research in perspective. In a broad sense it can be defined as the relationship between different groups of individuals to exchange knowledge, services, and goods with the express understanding that there will be mutual support. These groups generally come together to execute specific projects over a short time horizon. In essence, therefore, a network organization involves exchange between the firm and other organizations (like the university scientists) external to the firm. It is generally understood that they operate in an implicit contract based on shared norms and values. However, somewhat more realistically, desired efficient behavior may be elicited through value sharing royalty contracts, monitoring performance, and commitment inducing share of fixed costs and investment financing. The following advantages of network structures are evident. • •

They enable NBFs to access more, qualitatively superior, and reliable scientific knowledge and assimilate it more efficiently Transactions in network organizations, since they are basically short term in nature, provide flexibility to switch sources at short notice

24

Economics of Biotechnology • • • •

• •

See, for

The association of university scientists with a NBF signals the quality of the firm’s research to both the capital and resource markets Usually R&D of the scientists is financed by public agencies. To that extent the sunk costs to the NBF, of sourcing knowledge, is reduced A NBF, that has a large number of connections with scientists, can prevent competitive firms from having access to certain types of critical expertise The more scientists are working on a crop, the more likely they are to find new research methods, new genes, new germplasms, or new knowledge about the crop that will reduce a NBF’s cost of developing a new variety In addition to providing access to knowledge for immediate projects, information from the external linkages may evolve into important sources of new product ideas A large number of links means that the NBF learns to adapt to diverse management styles of the scientists or the universities that they represent. In other words, over time there is organizational learning, though not about technology, as network connections increase example, Liebeskind et al. (1996) and Audretsch and Stephan (1996).

However, it is possible to argue that NBFs do not have any additional advantages by having many links with university scientists (i.e., over and above what they gain from one such contact). •

Scientists may prefer secrecy with respect to their more promising discoveries. They may agree to license only the least promising compounds. Conversely, NBFs may not take up even promising compounds if they are already handling a large number of diverse activities • There may be organizational dissonance as the number of links increases. For, the organizational goals may clash rather than synchronize with the goals of the NBF It is therefore obvious that there will be diminishing returns to network size and the most efficient size of the network depends upon the • • •

development of a specific product extent to which the participation of the knowledge specialist is a requirement cost reductions that can be achieved at the downstream level due to an increase in the size of the network • the extent to which scientific, technological, or market risks can be reduced by pooling complementary assets and competencies It should be acknowledged that the NBF may make the scientists share in some of the costs of bioprocessing stage. This may be one mechanism to obtain greater compliance and commitment. This approach may enable the NBF to take up a larger number of activities and also improve the efficiency of the organization. To this point in the analysis only the number of links in the network have been considered to be important for the performance of a firm. However, it should be clear that the qualitative and quantitative nature of the links matters. Hence, in addition to the number of links several authors suggest other measures.


25

•

Number of field trials or clinical trials (input measure). See, for instance, Oehmke and Wolf (2003) and Schimmelpfenig (2004) • The number of approvals achieved (an output measure) • R&D grants obtained • Number of patents filed • The nature (e.g., equity participation) and quantity of financing • The nature and extent of risks absorbed Three observations are in order. First, a network organization need not necessarily imply links with organizations or individuals outside the firm. Modern technologies, especially biotechnology and information technology, are such that a task can be efficiently concluded only by drawing on the expertise of diverse groups of talent. Hence, it has been observed that whenever a new task needs to be tackled the firm puts together a small team from within the firm so that they can network efficiently and deliver the required product. Clearly, utilizing some outside expertise is one of the options in network formation. The team gets dissolved as soon as the task is completed. Second, a particular scientific idea may not translate into a marketable product. Hence, the network organization fails if it is entered into too early. On the other hand, too late an entry means the scientist has already invested a lot of money, is in a loss, and has very low bargaining power. Timing becomes the essence of success. The nature of the network organization, even for similar products, may turn out to be quite different. Third, a network organization fails in its function, especially in the context of private industry, if commercial secrets cannot be reined in within the firm. Such a danger may be particularly significant while dealing with outside experts. The purely short term contractual relations exacerbate problems of this nature. Thus, network organizations are gaining importance and have their efficiency generating properties. However, it should be understood that they are unsuitable in some contexts due to adverse selection as well as moral hazard. This is one of the reasons for some firms preferring full vertical integration by merger or takeover.

2.5 NATURE OF CONTRACTS Recall that the networking between university scientists and the NBFs is meant to • • •

obtain and implement scientific knowledge reduce uncertainty in market valuation of the assets of the NBF provide a greater assurance of quality (on the basis of the reliance on patents of the scientists etc.) • reduce costs of producing a given output, and • share costs to elicit commitment Each organizational relation with an external link in a network is a specific event. It may not be repeated over time. However, it is necessary to have some formal agreement (contract) for sharing gains or payments for achieving specific tasks. For, otherwise disagreements

26


may arise ex post if it is left open. Stated somewhat differently, it is necessary for the NBFs to exhibit equitable behavior in each of these transactions because trust and reciprocity are essential for the efficiency of network relationships. For, in the absence of such trust it may be difficult to establish such links in the future even if they are desirable and necessary. Stated more explicitly, a specific contract between a star scientist and a NBF will be to •

cooperate and create the synergy necessary for the success of the project on hand, and • share in the value generated through bioprocessing In most cases, there is an agreement, before the start of the network collaboration • on the distribution of intellectual property • on the property rights of the resulting patents, and • the terms of license to other parties Assume that the scientist offers his expertise and effort in scaling up technology. In its turn, the NBF provides its skills in bioprocessing and incurs the necessary capital and other expenditure. The NBF may expect a market value of its activity to be m = expected market value of the commercially usable new technological development Note that there is no tangible marketable output at this stage of transformation. Hence, this is an imputation by the financial markets and/or the larger chemical firms who eventually utilize this to achieve marketable production. (Appendix 2 contains an outline of the CAPM concepts used in this context. This is one efficient method for evaluating uncertain projects.) The value is achieved through the cooperative effort of both the parties. However, the expectations may not be fulfilled. First, there may be an adverse selection of the network partners. Further, after the agreement is reached either of the parties may decide to free ride on the effort of the other. This reduces the value of the outcome randomly. Second, the pharmaceutical sector (in particular, production of human health care products, including human diagnostic and therapeutical products, and associated treatment delivery systems) is characterized by severe competition (from the established chemical firms). This uncertainty is compounded by appropriation problems (since not all aspects of scientific knowledge can be protected by patents), high degree of uncertainty of returns to R&D, and so on. Hence, NBFs cannot determine in advance that any particular R&D activity will lead to a valuable discovery. The realized market value may be (m + u) where u = a random variable with E(u) = expected value of u = 0 and V(u) = variance of u = σ2 (a constant) Assume that the opportunity cost of the scientist’s effort can be represented by m2/2δ where δ represents the following. •

The degree to which the scientist has to participate at the bioprocessing stage to guide the NBF. A low value of δ indicates that more informal knowledge transfer and involvement of the scientist is necessary.


27

•

It may also indicate the skills of the scientist in his interface with the NBF. From this perspective a larger δ indicates a greater skill and a possible cost reduction. • Greater organizational learning, if it materializes, necessitates lower involvement of the scientist in the operations of the NBF. That is, a large value of δ may also signal greater organizational learning. The above specification posits diminishing returns to the effort of the scientist. Similarly, postulate that km2 = investment of the NBF in the organization Some technologies are more expensive than others. The magnitude of k reflects this degree of difficulty. The only way to compensate the scientist for his effort is to offer a fraction α of the realized market value. Clearly, a larger value of acts as an incentive for efficient performance. The net return to the scientist can be represented by πs = α(m+u) – m2/2δ Note that the scientist may be risk averse. That is, he will find it difficult to accept a large value of σ2 given his opportunity cost. Hence, the net value of the contract to the scientist will be Vs = αm – m2/2δ – λα2σ2 where λ > 0 measures his degree of risk aversion. On the other hand, the gain to the NBF is πn = (1–α) (m+u) – km2 By way of contrast, the NBF may be risk neutral. For, he may be operating in several ventures and can therefore diversify the risk. (However, note that his risk aversion, if it exists, will not affect the qualitative nature of the results that follow.) Hence, the net value for the NBF can be written as Vn = (1 – α)m – km2 The scientist can be expected to choose m targeted for a given α. Maximizing Vs results in m = αδ This is often designated as the incentive constraint of the scientist. Clearly, the scientist targets a higher m depending on • his level of skill, and • the share of value he can recover One characterization of the behavior of the NBF is to examine the choice of α that maximizes the total net value given the incentive constraint of the scientist. For, in the ultimate analysis, the purpose of creating the NBF is to generate the maximum benefit for all the parties. Observe that the net value is N = Vs + Vn = m – m2/2δ – km2 – λα2σ2

28


The value of α that maximizes N is α = δ/(δ + 2kδ2 + 2λσ2) The net value of this network organization turns out to be N1 =

δ2/2(δ + 2kδ2 + 2λσ2)

This is an increasing function of δ. Observe that as δ → ∞ the scientist need not incur any cost in his association with the NBF. That is, the knowledge transfer does not require the expertise of the scientist at any informal level. Instead, knowledge transfer is purely formal. Under these conditions scientific knowledge can be procured by making a fixed technology payment. There will be no necessity to offer royalties proportional to the value generated. This is the implication of α → 0 as δ → ∞. Further, organizational learning in the network relationship can be represented by an increase in δ and/ or a reduction in k. For, ultimately, one or both the parties become more efficient if they learn about the knowledge requirements of the other through their association in the network. A reduction in k, given the costs of the scientist, implies that the scientist is accepting a greater share of the cost. Hence, he must receive a greater share α of the value generated. It can also be verified that the total output obtained from the network relationship is T = δ2/(δ + 2kδ2 + 2λσ2) and the total cost of producing this output is C = δ3(1 + 2kδ)/2(δ + 2kδ2 + 2λσ2)2 It can therefore be inferred that the productivity of the network organization is P1 = 2 +4λσ2/δ(1 + 2kδ) The advantages of networking will be lower, if the value of the informal interaction of the scientist goes down. Similarly, an increase in organizational learning increases productivity. The following model reflects some of the economic implications of an increase in the size of the network organization. The profits of each firm can be represented by πs = α(m + u) – m2/2δ so that m = αδ as before. The NBF, or the large firm as the case may be, produces m* = nαδ if it has n links with upstream firms. Given the organizational dissonance alluded to earlier the investment and spending by the larger firm will be kn2m2. Then, it follows that πn = n(1–α) (m + u) – km2n2 and the net value of the network of alliances is N = nαδ – nα2δ/2 – kn2α2δ2 – λnα2σ2


29

The optimal choice of α will therefore be α = δ/(δ +2λσ2 + 2kνδ2) and the total output is T = nδ2/(δ +2λσ2 + 2knδ2) The total cost of producing this output is C = nδ3(1 + 2knδ)/2(δ + 2λσ2 + 2knδ2)2 Consequently, the productivity of the network becomes P2 = 2 + 4λσ2/δ(1 + 2knδ) This expression decreases as n increases. That is, productivity decreases as the network size increases. Danzon et al. (2005, p.319) noted the occurrence of such decreases in productivity. It can also be shown that the optimal n, that maximizes N, is given by n = [δ(2 – α) – 2λσ2]/4kαδ2 Hence, n increases as δ rises initially but highly skilled downstream firms are unlikely to find networking with larger firms profitable. Instead, as Danzon et al. (2005, p.321) pointed out, “small firms take advantage of asymmetric information to out-license least promising compounds (while) retaining their more promising compounds to be developed internally.” It should also be noted that the large firms tend to prefer fewer network relations, each of which yields higher sales. The following observations are also pertinent. •

• •

•

The fundamental basis for networking will be lost if the scientist need not interact with the NBF in the process of transferring the informal knowledge associated with scientific knowledge. This is reflected in n → 0 as δ → ∞. That is, under these conditions formal transfer of knowledge and a payment of technology fees to each of the scientists will be adequate and efficient. Though P2 is a decreasing function of n, it must be noted that T increases with n. This is the only advantage of networking. It is of course possible that organizational synergies can be achieved at some stage as n increases. The productivity of the network organization will be higher during that phase. The maximum size of the network occurs when δ = 4λσ2/(2 – α)

That is, the NBF accepts a lower informal interaction with the scientists the greater the royalty share it must pay. Stated differently, a greater α will induce the NBF to choose projects where they require only a lower continuous involvement of the scientists. Similarly, when the project is less risky the NBF would be more willing to undertake it and it would be willing to involve the scientist on a more continuous basis. It cannot, however, be claimed that the NBF will undertake only those products that require this value of δ.

30


Note that one of the advantages of the network organization is the expectation of greater commitment. For, as Teece (1980, p. 232) pointed out, “internal trading changes the incentives of the parties (i.e., aligns them more closely with the goals of the organization) and enables … (them to attenuate) costly haggling … and other non-cooperative (disruptive) behavior.” In particular, the association of a university scientist with a NBF will improve his competence to move towards a marketable product. There will be a reduction in the scientist’s perception of risk in the project. Further, the association of a star scientist may improve the image of the NBF on the capital market thereby reducing the uncertainty in its valuation. It is therefore possible that the variance of u becomes V(u) = σ2/δ The corresponding optimal value of α is α = δ2/(δ2 + 2kδ3 + 2λσ2) and the net value will be N3 = δ3/(δ2 + 2kδ3 + 2λσ2) Clearly, N3 > N1 if δ > 1. That is, the network organization is at an advantage whenever the variance can be reduced through such collaboration. The following observation is important in the context of agricultural biotechnology. Usually the government agencies offer extension services to the farmer to help him improve productivity. The model will be similar to the above. The farmer’s role is the same as that of the scientist and that of the NBF identical to the extension service. Interpret m as the gross welfare generated by the output produced by the farmer. The fraction of welfare not accruing to the farmer goes to the rest of the community and not necessarily the agency providing the extension services. The net value concept remains the same.

2.6 SHARING FIXED COSTS It is by now clear that the question about sharing fixed costs at every stage of the network organization becomes important. First, such sharing of risk may reduce the liability of both the parties. Second, the NBF sharing in the fixed costs of clinical trials, drug approvals, and so on will induce the NBF to offer the requisite intangible knowledge in addition to the formal knowledge. It will then provide a commitment to the large pharmaceutical company in so far as it transmits all the scientific information that it has truthfully. Part of the reason is of course its own interest in recovering its investment. In the process of sharing fixed costs the NBF also gains some control on the operations of the large firm. As a result the NBF can scrutinize the decisions of the large firm and ensure alignment with its viewpoint. Let σ be the share of the NBF (denoted as F hereafter) in the cost km2. It will be assumed that this has the effect of reducing the variance to σ2/s. More pertinent to the argument will be the changes in the cost to the large pharmaceutical firm (identified by P in the sequel). The following observations are relevant.


31

•

P has to pay F an extra share of output by way of royalties due to the increase in his share of capital and the increase in the bargaining power. • P has to make greater effort, and perhaps incur greater costs, to convince F to provide the commitment abinitio • The costs of negotiation, management, and conflict resolution increase with s. This may arise purely due to differences in expectations, management and organizational values. The costs of making collective decisions can be quite substantial, especially when the firms have diverse preferences. It will be postulated that αskm2 represents the total costs with the understanding that αs > 1. The share of F will however be skm2. The profit for F will now be πf = α(m+u) – m2/2δ – skm2 Given his risk aversion the value he attaches to πf is Vf = αm – m2/2δ – λα2σ2/αs Observe that F will not accept a large s because he incurs additional cost. Further, there is a possibility that F experiences liquidity constraints while raising the resources required for capital investments. Both these considerations suggest that F will prevail on the choice of s. Hence, he can be expected to choose s and m to maximize Vf. This results in s = (λ/k)1/2 (σ/δ) m = αδ* where δ* = βδ and β = 1 – 2 (λk)1/2σ The net value of the contract is N = m – m2/2δ – αskm2 – λα2σ2 = αδβ – α2β2δ/2 – 2β(λk)1/2α2δσ P will therefore choose α = 1/[1+2(λk)1/2σ] The efficient choice of s so obtained has the following properties. F will accept a higher s to counteract the effect of a low δ. That is, he will signal greater commitment to P. • There is a direct relationship between s and λσ2. A larger λ and/or σ2 necessitates F indicating a greater commitment to P. • As m increases P may seek greater security and commitment. F will therefore agree to a larger s. It must however be acknowledged that an increase in s beyond a point may be a disadvantage to P because F gains control. However, the choice of an efficient s, whose effect is through the variance, generally neutralizes the effect of a low δ. •

32


2.7 ECONOMIC CONSEQUENCES It was generally noted that • each NBF has associations with many scientists • every large firms has links with many NBFs • there are far fewer large firms in comparison to smaller firms The economic reasons are as follows. •

Zucker and her associates argued that a large firm, which is already networked to several smaller firms, has greater popularity as being reliable. This allows it to increase its upstream connections. • Under the existing IPR regime it is not possible for any one upstream firm to link with many downstream firms. This is one reason for fewer downstream firms. There are, however, deeper reasons for this phenomenon. • Riccaboni and Pammolli (2002) argued that larger firms have certain types of expertise like scaling up of plants, providing expertise in clinical trials and taking a product through the regulatory process. However, these skills are difficult to acquire. Hence, the growth in the number of downstream firms is not commensurate with the growth of basic knowledge about biotechnology. • Rothaermal and Deeds (2004) noted that by networking with a large number of NBFs a large firm can prevent competitive firms from having access to certain types of critical expertise. This reduces the number of downstream firms. • Niosi (2003) noted that large firms have deep pockets (many sources of, and large amount of, finances). They can use the finances to take up downstream activities of a large number of NBFs. The converse is definitely not valid. However, it is possible to argue that large firms do not have any additional advantages by having many links with smaller firms (i.e., over and above what they gain from one such contact). •

•

•

Schimmelpfennig et al. (2004) maintain that the more scientists are working on a crop, the more likely they are to find new research methods, new genes, new germplasms, or new knowledge about the crop that will reduce a private firm’s cost of developing a new variety. Decarolis (1999) pointed out that in addition to providing access to knowledge for immediate projects, information from these external linkages may evolve into important sources of new product ideas. Powell et al. (1999) argues that a large number of links means the large firms learn to adapt to diverse management styles of the NBFs. This may help them develop their network over time as also consolidate their advantage. In other words, over time there is organizational learning, though not about technology, as network connection increases.


33

•

Rothaermal (2001) and Danzon et al. (2005) noted that large firms have cumulative advantages of having been there already. For example, when they deal with a large number of smaller firms, larger firms learn more about the differences in their organizational culture, and develop greater capacity to adapt to diverse organizational arrangements. This organizational learning enables them to expand their network even wider. • Powell et al. (1999) also argued that as the large firms develop deeper networks and also cover a large number of functions (scientific information, organizational knowledge, and finances) they become central (or more important) relative to competitors. This enables them to obtain larger research grants, non-operating incomes, larger sales revenue and so on. • Rothaermal (2001) and Rothaermal and Deeds (2004) also observed that incumbents that adapt to the new technology via interfirm cooperation with new entrants can eliminate competition while gaining advantage. There have been concerns about negative effects and diminishing returns. •

•

Danzon et al. (2005) noted that there can be adverse selection. For example, the NBFs may agree to license only their least promising compounds and develop others in house. On the other hand, large firms may not take up even promising compounds if they are already handling a large number of diverse activities or they feel that the new compound will immediately cannibalize an existing drug that they are manufacturing. Powell et al. (1999), and Decarolis and Deeds (1999), Bottazzi et al. (2001), and Danzon et al. (2005) pointed out that as the number of links increases there may be organizational dissonance rather than synergy because the organizational structures of different NBFs may clash rather than synchronize with the goals of the large firm.

2.8 OTHER ASPECTS The basic result of this chapter is that network organizations, i.e., contracts with one or more outside agents, are essential when there is significant interdependence and internalizing all activities is uneconomical. Such an organizational arrangement may improve productivity if there are synergies and significant organizational learning. However, this cannot be taken for granted due to the emergence of organizational dissonance as the size of the network increases. In general, it may not be possible to turn an unsuccessful firm around by merely adopting a network organizational structure. In practice, there may be tendencies to choose n > nopt (to prevent competition from using the assistance of the scientists) or n < nopt (if financial constraints are stringent). Some inefficiency is bound to persist even dynamically. In general, it is difficult to assert that information synergies dominate the network choice. As noted above, they may also be motivated by the desire to monopolize, financial constraints, and so on. To that extent, productivity increases from network organizations will not be a foregone conclusion.

34


The NBF may invest a small part of the total requirement initially to ascertain the probability of success. It may undertake the rest of the investment only with this probability. The implications of such choices for the efficiency of network organizations have been examined in Filson and Morales (2005). When networking turns out to be inefficient some firms prefer full vertical integration by resorting to mergers and takeovers. In general, the existence of organizational learning and spillover effects in network organizations cannot guarantee productivity increases. The exact conditions under which this can be surmised with a fairly large probability remain an empirical question. Given the present state of analysis any generalizations may be hazardous. However, having a network is an advantage over not having it so long as it is utilized judiciously. Throughout this chapter diffusion of scientific information through a network organization was among a limited number of clearly identified individuals and firms. This approach is one way of keeping secrets among the members of the network. As noted earlier open source policies may be superior if they can materialize. However, it is generally considered more desirable to disseminate knowledge more widely. IPRs and patents provide a mechanism to achieve this. The next chapter will deal with the pertinent details.

IPRS and Patents

35

Chapter 3

IPRS AND PATENTS

3.1 WHY PROTECTION? A couple of decades ago, the legalities of obtaining samples of microbes from plants and animals were straightforward. In many instances one could simply arrive at the site, collect samples, and take them without bothering about legal issues. Such samples could be freely transferred anywhere in the world. This is not possible in the commercial world of today. Hence, it is necessary to examine • the alternatives • their evolution over time • merits and disadvantages of each choice At the other extreme, until about the late 1960s, experimental biologists, unlike their counterparts in high energy physics and other sciences, were reluctant to reveal their results freely outside their own laboratories. The basic reason for this was the fact that a great deal of this fundamental research could not be associated with any product from which the consumers derived value. Secrecy was the only protection for their proprietary knowledge because the patent laws at that time applied the utility doctrine, i.e., usefulness of information to consumers of goods and services as a precondition for the grant of a patent. However, any one group of scientists has limited capabilities and resources to develop knowledge efficiently. As a result, the developments in biotechnology were proceeding in a fragmented way. Only a few cell structures have been discovered with difficulty. Further progress, towards a protein structure or a marketable product was difficult and inefficient due to secrecy. Secrecy, as an organizational mechanism, proved inadequate for efficient knowledge diffusion. It was necessary to conceptualize an organizational mechanism that allows fast and efficient knowledge transfer.

36


Contract research was one possible alternative. That is, a scientist or a firm may provide the knowledge that he developed, laboratory tools etc. to another scientist or firm who maintains secrecy and develops them further. This organizational mechanism is subject to three limitations. • • • A better

There is no guarantee that a transfer of formal knowledge is enough There is no assurance that secrecy will be maintained except in close knit groups Such small groups may not have the required competence organizational structure was sought.

In the previous chapter it was noted that firms enter into network relations with suitable parties to •

develop the scientific information that they discover

•

limit the availability of crucial biological material to competitors, and

•

internalize the informal knowledge that they develop in its utilization

Examples may be •

genetically modified cell lines that produce MAbs for diagnosis of diseases

•

genome data bases that combine sequence data with protein structure.

Such networking is generally conducive to transfer of informal knowledge as well. However, due to their short term and transient nature, such networks are vulnerable to knowledge leaks. In other words, the synergies due to networking may offer an inadequate protection in the pursuit of secrecy. It was also noted in the previous chapter that network organizations may fail for a variety of reasons. A further complicating aspect of knowledge generation and transfer has been pointed out in Fink (2000), Bottazzi et al. (2001), and Grabowski (2002). As they pointed out, pharmaceutical inventions are such that •

imitative product development (reverse engineering, generic drugs) is not very difficult • imitation costs are extremely low in comparison to inventor’s costs (scientific discoveries in the pharmaceutical sector are expensive to discover, develop, and obtain regulatory approvals) • knowledge erosion is relatively fast since new ideas are generated all the time The above features, regarding knowledge diffusion, therefore make these advances weakly appropriable from the viewpoint of the innovating firm. That is, the firm will find it difficult to recover the fixed costs of drug development if left entirely to competitive market forces. Since every stage of biotechnology development is expensive any mechanism that rewards only the final product discovery (as with the market mode) does not compensate the early discoverers who are essential to achieve the latter stage developments. This is the crucial aspect of cost recovery in the context of biotechnology firms.

IPRS and Patents

37

Agricultural biotechnology is also subject to similar problems. In particular, Monsanto and other firms invested substantial amounts to develop Bt and RR varieties of seeds. They also invested in obtaining regulatory approvals and setting up marketing networks. However, traditionally the farmers are accustomed to preserving seed for reuse. That is, the practices of the farming community are inimical to the recovery of such costs. Hence, the plant breeder, who made substantial investments in genetically modifying crops, cannot recover his costs. Note, however, that R&D can act as a barrier to entry. For, the speed of innovation raises the costs and risks of innovation. This may result in a natural monopoly position. In such a case there is no further necessity for protection. However, this appears to be an inadequate mechanism in practice. Hence, from the viewpoint of the innovator and the firm, there is a necessity to protect intellectual property to • provide security from piracy and imitation, and • recover the large fixed costs of R&D The necessity to protect intellectual property can be articulated from the vantage point of social welfare as well. This too has the above two dimensions. First, it can be argued that providing open access to genetic data is necessary to encourage the diffusion of research results. For, removal of research findings (e.g., nucleotide sequences) from the public domain will restrict development of derivative knowledge necessary to make the genetic information practically usable. Consider the following example. Myriad initially developed BRAC1 and BRAC2 breast cancer tests. By keeping the information exclusive and excluding other clinical testing services from its use they prevented other firms from developing more and better tests. Knowledge secrecy is therefore a disadvantage from the viewpoint of social welfare. Second, under competitive market conditions, the firm may not be in a position to convert social welfare into profits for itself. This would make it difficult for the firm to recover its fixed costs while maintaining the maximum possible social welfare. This will be elaborated further in the next section. In sum, it must be noted that providing some protection may • release knowledge to be used for common good, but • creates monopoly power thereby reducing the extent of social welfare achieved. Mechanisms design must keep an appropriate balance between these two aspects in perspective since it may not be possible to achieve both of them simultaneously. Some changes were brought about in the early 1970s, viz., the introduction of • • • •

plant breeder rights protection of plant varieties material transfer agreements copyrights and trademarks

38


In addition, in some countries, it was made mandatory for all publicly funded scientific research, including research tools, to be made freely available to all interested scientists and industry within six months after publication. However, enforcing copyrights after publication will be generally difficult for the simple reason that detection of violations is costly. A more sophisticated institutional mechanism had to be conceptualized. The only well known organizational mechanism was patents. There was some precedence that pointed towards this alternative. First, in 1911 Learned Judge Hand upheld a patent on purified human adrenaline made via a new process. The patent was not simply on the process, but also on the purified natural substance. Second, in 1975 Kohler and Milstein, discovered that individual immune system cells, that generate antibodies to a specific antigen, could be fused with immortal cancer cells to create a small factory for producing antibodies. They did not patent it. Hybritech was the first to use monoclonal antibodies in diagnostic kits sold to doctors and hospitals to identify the presence of diseases (e.g., AIDS) or heightened hormonal levels (e.g., pregnancy tests). It received a patent covering the whole family of diagnostic kits. Patents generally provide a 20 year exclusive market protection if the following conditions are satisfied. • •

Novelty, i.e., it was not known earlier Non-obviousness; in particular, it is not something already occurring in nature and not discovered earlier • Full and complete disclosure so that any one knowledgeable about the trade can reproduce the production process In practice, patent claims should also specify their scope. That is, claims should define what the inventor considers to be the technological territory that he claims to be under his control by suing for infringement if necessary.

3.2 PATENTS AS PROTECTION For all practical purposes it is obvious that there is a necessity to keep information and access to genetic data open. For, this is the only mechanism to encourage diffusion of research results and new innovation. It is also clear that there is a commercial need to protect inventions in order to create revenues from investments in R&D. Patents are one answer to these problems. Article 7 of the TRIPS agreement states the objective as “the protection and enforcement of intellectual property rights (with the objective of contributing) to the promotion of technological innovation and to transfer and (disseminate) technology, to the mutual advantage of producers and users of knowledge and in a manner conducive to social and economic welfare, and to balance the rights and obligations.” They consist of granting a 20 years exclusive market protection (i.e., monopoly use and marketing rights) in return for full and complete disclosure of information so that anyone knowledgeable about the trade can reproduce the production process. There is a possibility that some discoveries, e.g., a block buster drug, may be in a position to recover its costs much sooner. Similarly, the costs of minor

IPRS and Patents

39

innovations may be low. This will also enable the firm to recover costs quickly. In such cases it will be difficult to defend a uniform patent. However, this is the current consensus. Its economic rationale is not properly documented. Such a patenting arrangement, by making the knowledge available to everybody as soon as possible, is expected to enable others in the industry to invent around the basic concept and create competing products. This process of generating product variety under different patents for each of the variants is helpful in breaking the monopoly granted to the patent holder. Static micro-economic theory explains the role of patents in recovering costs in the following manner. Consider Figure 1. In this figure D represents the demand for the patented product. MR is its corresponding marginal revenue curve. MC represents the marginal cost of producing Y units of output. It is well known that in a competitive market the firm will offer output Yw on the market at a price pw. This maximizes social welfare and Yw is said to be efficient. Innovative new products, e.g., crops based on genetically modified varieties, may have a low demand at the outset. The producer surplus, represented by the area pwCE, may not be sufficient to cover the fixed costs (not shown in the figure). This is the basic appropriability problem. Now suppose that the firm is granted monopoly rights through a patent. It will now choose a profit maximizing output Ym and price pm. This results in Ym < Yw; i.e., the monopoly restricts output. p,MC G A

Pm Pw E O

MC CD

B F Ym

MR Yw

Y

Fig. 1 It charges a price pm > pw and the producer surplus changes to pmAFE. This increase in profits (if any) may be adequate to recoup the fixed costs. This is the basic justification for the grant of a patent. The emergence of monopoly reduces social welfare by the area ACF. This is the conventional concept of deadweight loss. The agency granting patents should consider ways of recovering the loss. Shavell and Ypersele (2001), for instance, argued that if the government grants a subsidy pwCG the monopoly firm will indeed offer Yw. However, this is unrealistic. Alternative solutions will be considered in the sequel. In general, it may be argued that the patent system can, and does, make an attempt to put in place adequate instruments to curb monopoly power of the patent holders in the exercise of the rights granted by law. The other viewpoint, which argues that monopoly power can in fact be reduced, refers to the possible cost reducing effects of the new innovation and knowledge disclosure. For, this is one of the reasons why the new product is deemed superior. To simplify the exposition

40


assume that the marginal cost MC = c (a constant) for all values of Y. Similarly, let the demand curve be p = a–Y Hence, the output Ym chosen by the monopoly is such that c = pm(1 – 1/η) where η is the elasticity of demand. Consequently, pm = cη/(η–1) On the other hand, the welfare maximizing Yw and pw are Yw = a – c, and pw = c However, note that Ym = (a – c)/2, and pm = (a +c)/2 Hence, it follows that (a – c)/2 = c/(η – 1) Consequently, the deadweight loss is given by D = (pm – pw) (Yw – Ym)/2 where pm – pw = [cη/(η–1)] – c = c/(η–1) and, similarly, it can be verified that Yw – Ym = (a – c)/2 = c/(η–1) Hence, it follows that D = c2/(η–1)2 Clearly, D reduces as c falls and increases whenever η is reduced. A very general result is available in Rao (2004, p.329). This result suggests that the advantages of cost reduction may reduce the losses due to monopoly power and the consequent reduction in η. Therefore, it may be argued that monopoly markets, to the extent they give rise to dynamic efficiencies emanating from the ability of large firms in the concerned industries to conduct socially beneficial R&D, lower costs of production (process innovation) and offer a wide range of product varieties (for instance, in a given therapeutic class and product innovation). These will be conducive to enhancing social welfare. An alternative approach to the welfare issue is due to Spence. Spence addressed the question about the extent to which a firm can convert consumer welfare into revenue for itself.

IPRS and Patents

41

The standard Spence formula, explicitly derived in appendix 2, informs us that Y

Yp(Y) = (1 – 1/η) ∫ p(y) dy 0 Hence, as η reduces a smaller portion of the consumer welfare can be converted into revenue for the firm. That is, even if patents provide monopoly for the firm, it will prefer products with a high elasticity of demand. If this is indeed the case the deadweight loss of strong IPR protection will not be very much. Both the approaches suggest that a high elasticity of demand will be less damaging. Spence’s argument suggests that this will be the case more often. However, there is no empirical evidence for this in practice. Notice that the Spence approach is not designed to incorporate the effect of cost reduction on social welfare. Deadweight loss calculations appear to be superior. Of course, it is quite another thing to argue that policy considerations under the TRIPS agreement do not pay attention to these calculations. Observers, like Watal and Mathai (1995), are only saying that commercial interests of the larger corporate firms, rather than larger social interests, dominated the TRIPS negotiations. The issue here is, should they? Or, is there a need for a balance?

3.3 BIOTECHNOLOGY PATENTS There were objections to patenting biotechnology initially. The U.S. Patents and Trademarks Office (USPTO) argued against patenting genes on the ground •

that they are discoveries (identifying something that already exists) and not inventions,

• products of nature are not new, • the basic core of humanity should not be owned by anyone as property However, two 1980 decisions of the U.S. Supreme Court changed all that. The Diamond vs. Chakraborty case was about patentability of a genetically modified bacterium. The court held that such material is patentable because there is novelty. Subsequent gene or DNA patents have claims that they cover nucleotide sequences that encode genes or fragments of genes. As a general rule, of course, patents cannot cover a substance in situ (inside the body or resident) in the human body. But they can if they are isolated from their natural source. Secondly, the U.S. Supreme Court ruled that genetically altered life forms require patenting. A decision by the court allowing an oil company to patent an oil eating microorganism (bioremediation) set precedence and opened up massive possibilities, including that of the exploitation of genetic engineering for commercial purposes. Lakhsmikumaran and Pillai (2005) pointed out that the Calcutta high court decision in the Dimminaco A.G. vs. controller of patents and designs had a similar basis. The more recent third amendment to the Indian Patents Act 1970, and its ramifications for biotechnology innovations, has been examined in Abrol (2005) and Rangnekar (2006).

42

Economics of Biotechnology In general, patentability and challenges to patent grants revolve around three aspects. •

To be patentable an invention must meet the requirements of novelty, non-obviousness, and utility (note that utility is not the major emphasis any longer) • The disclosure must be sufficient to enable everyone skilled in the art to make and use all the documents of the invention claimed in the patents • Claims should define what the inventor considers to be the scope of his invention, the technological territory he thinks to be his to control by suing infringements Biotechnology patents are peculiar in so far as they are not directed to any specific and marketable product. Instead, biotechnology inventions cover genetic materials. The gene or DNA patents generally relate to nucleotide (DNA or RNA) sequences that encode genes or fragments thereof. They consist of a combination of definitions of new processes, methods, and compositions. Genetic patents may also be directed to devices for use in testing and diagnostic kits. In other words, some genomic discoveries have been granted patents based solely on the new composition or sequences of a random piece of genetic material without knowing its function but only in the hope that it will constitute an important part of a gene. In general, patent applications may pertain to • • • • • • •

genes or partial DNA sequences such as cDNAs, promoters, and enhancers proteins encoded by these genes and their functions in organisms vectors used for the transfer of genes from one organism to another genetically modified cells, plants and micro-organisms processes used for the manufacture of a genetically modified product genetic tests for diseases that utilize genetic sequences or proteins drugs developed on the basis of the knowledge of proteins and their biological activity See, for example, Abrol (2005). Stated mildly, there has been a greedy rush to apply for and grant patents though the nature of the invention and its utility are at best nebulous. As Correa (2001) put it, “thousands of patents are granted every year in the United States for minor, purely trivial developments or for substances (including genes) that already exists in nature and which have been merely discovered but not invented by their would be owner.” It is therefore necessary to recognize that the most important task is to understand what a patent covers, i.e., the extent of protection it provides the owner. The scope (breadth) of biotechnology patents creates a host of new problems. First, consider the possibility that a patented biotechnology material requires further improvement and processing before any final product of commercial use emerges. The most obvious example is the Cohen-Boyer patent on rDNA. If another firm wishes to pursue this activity it needs a license from the patent holder. The patent holder may license the use of its patent to others for an appropriate payment. However, a patent holder may hold rivals hostage if they need licenses for a large number of nucleic acids. For instance, the development of a medicine may depend on genomic technologies, receptors, assays, and high throughput

IPRS and Patents

43

technologies. Similarly, enriched vitamin A rice (popularly called the golden rice) is based on technology that spans 70 patents held by 31 different organizations. This phenomenon is usually designated as patent thickets. They tend to increase the transaction costs of reaching agreements with various components needed to proceed with a product development. Since each of these patent holders claims a royalty (a share of revenues generated in the product market) the costs may increase prohibitively due to royalty stacking. Second, some patents claim a very broad scope. For example, Human Genomic Sciences (HGS) of the U.S. claimed a patent for a gene though its function was not known. It was only asserted that it will be a research reagent or material for diagnostics. Subsequently, it was discovered that it was the docking receptor CCR5 used by the HIV virus to infect a cell. Thus, through access to a broadly defined patent HGS gained an undue advantage to block further research. The patent offices have not been able to moderate such claims. Diffusion of knowledge, the very purpose for which patents are granted, has been in jeopardy. Third, the data exclusivity issue, though related to these, must be highlighted. Many testing procedures for genetic discoveries and clinical tests of bio-pharmaceutical products have been granted exclusive rights (often called data exclusivity). Usually the larger chemical companies provide data relating to clinical tests on drugs and field trials of GM seeds to the regulator while requesting marketing approval. Since these tests are also expensive (accounting for as much as 50 percent of the costs of drug development) the patent holder desires that the data be kept a secret unless they license their use to other parties. The law allows this for a period of five years if the patent has not expired already. This issue appears relevant for a country like India because it has the technical capabilities to conduct such tests and the pharmaceutical majors expect MNCs to outsource such business to them. See, for example, Maria and Ramani (2004). In the context of agricultural biotechnology it should be kept in mind that plants and their varieties cannot be patented through GM seeds alone. It is therefore recommended that plant breeders and others be provided sui generis (literally second to none or a system of its own) modes of protection. This will be considered in the next section. In general, it can be claimed that companies may under invest or under develop biotechnology products from the viewpoint of social welfare if IPR protection is defined too narrowly. On the other hand, broadly defined protection may reduce competition and lead to excessive monopoly power and high product prices. It is therefore more reasonable to provide patent protection to functional genomic discoveries that identify the role of specific genes and those that have implications for the design of new products. However, it should be clear that the uncertainty about the scope of patents can have considerable negative effects. For, firms may delay investments if they expect another firm to enter the market under patent protection. But the patent holder himself may also delay the exploitation of a patent if the extent of legal protection against imitators is not clear. Balancing the many aspects alluded to above has been a difficult task while defining patent provisions and the law.

44


3.4 IPR AGREEMENTS It is generally argued that • •

diagnostic, therapeutic, and surgical methods of treating humans and animals, and plants, animals, and essential biological processes for their production are not patentable. This is the spirit of article 27(3) of the TRIPS agreement. In general, IPR protections expect countries to put a sui generis system of protection of intellectual property in place in all such contexts. The IPR agreements relevant to the biotechnology area are • • • • • Consider

plant breeder rights traditional knowledge and geographic indicators trademarks trade secrets bioinformatics and databases each of these in turn.

The traditional concern of the private sector plant breeding initiatives was on the development of hybrid varieties. But, with the advent of biotechnology, new varieties displace traditional varieties on a large scale. However, the habit of the farmer has been to reuse seed from their crops for planting it in the next season. This makes it difficult for the breeders to recover their investment. The Plant Variety Protection (PVP) certificates provide one form of protection. Essentially, they protect the genetic makeup of a specific plant variety even if the varieties have similar characteristics. The basic requirement is of novelty and distinctiveness. However, there is a necessity for non-obviousness (inventive step) or usefulness (industrial appropriability) as in the case of patents. The PVP system of protection also acknowledges that products of biotechnology are essentially delivered through seeds. This may signal preventing the reuse of seed if the genetic resources are conserved and maintained for possible future use through “gene banks.” However, in practice, the more efficient mechanism of delivery is “on farm” seed preservation. When this is acknowledged the PVPs make a balance by allowing farmer’s rights consisting of • reuse of seeds from crops • use of protected varieties for further breeding • cross licensing for development of plant varieties However, in more recent times, biotechnology firms came up with “terminator seeds” that are sterile and do not produce any crop if replanted. There is also a recognition that the productivity of GM seeds depend crucially on the agro climatic conditions that vary considerably. It is therefore possible to adapt a GM variety to suit local conditions. See, for example, the evidence in Morse et al. (2005) on Bt cotton. It is difficult to cover such a spectrum through a single PVP.

IPRS and Patents

45

Traditional knowledge about plants is specific to a region. However, its date of origin is uncertain. So is its ownership if that concept is meaningful at all. In particular, the knowledge is in unwritten form. It is, however, acknowledged that the variety of crops suitable to a specific agroclimatic condition has evolved over time. Such traditional knowledge about plant breeding is an important source of new varieties. It also provides the background stock for new crops produced by genetic engineering. See, for example, Schaal (2004). Hence, it should be protected. This is sought to be achieved through copyrights. The practical difficulty here is in creating institutions and awareness about the need to obtain such copyrights. Trademarks apply to all goods and services in a similar fashion. They also provide legal protection for an unlimited amount of time. They are meant to distinguish goods or services of one firm from those of another. Consequently, even colors and symbols constitute important parts of a trademark. Products of biotechnology undergo stringent tests, regarding human health and safety, before regulatory approval to market them. The regulator usually demands such information from the inventor. This information is also expensive and it was observed that it may account for as much as 50 percent of the development of a drug. Hence, the inventors are interested in the regulator maintaining secrecy (trade secrets) with respect to such data. This is granted for upto five years after regulatory approval. During this period the right to license the use of data on clinical tests and field trials rests with the original seed or pharmaceutical firm. In general, it can be argued that the IPR system is driven by commercial considerations of product differentiation and planned obsolescence rather than genuine improvements in product characteristics. Agricultural biotechnology is also posing issues of biodiversity and environmental concerns. In this context, biodiversity refers to the variability among living organisms from all sources. Such biodiversity, as observed in practice today, is a result of centuries of adaptation of plants and other species to the environment they live in. Such biodiversity, as opposed to commercial uniformity that biotechnology firms seek to impose through patents and IPRs, constitutes the very essence of maintaining the ecological balance. Long term viability of biotechnology necessitates addressing the tradeoff between commercial interests of a few and the ecological balance necessary for the survival of many. The Convention on Biodiversity (CBD) and Cartagena Protocol (2003) are the most pertinent. The goals of CBD are the • conservation of biodiversity • sustainable use of its components • fair and equitable sharing of benefits • including appropriate access • transfer of relevant technologies and products To pursue these objectives, the CBD •

recognizes sovereignty of countries and their genetic resources

46

Economics of Biotechnology •

focuses on in situ (within the body) conservation of genetic resources (not in gene banks) • recommends protection of technical knowledge The Cartagena Convention deals mostly with international trade. It emphasizes the need for •

an adequate level of protection in the field of safer transfer, handling, and the use of genetic materials • minimizing risks for human health, and • monitoring transboundary movements The following approaches have been suggested. •

Prior informed consent (based on scientific knowledge and tests); the exporting country should inform the importing state of the nature and hazards of shipping GM products and obtain written consent • Refuse such consignments if not satisfied about safety or destroy the lot if illegally shipped However, one of the concerns is that such rejection will be a non-tariff barrier under the TRIPS agreement. A resolution is not as yet in sight.

3.5 TRIPS AGREEMENT Historically, each country specified its own regimes of IPR protection to suit their development imperatives. The Swiss watch industry and German glass are examples. The Indian pharmaceutical industry successfully developed the market for generic drugs as a result of the Patents Act 1970 that allowed only process patents. A concept of a universal patent does not exist. Firms must apply for independent patent protection in every country where they intend to exploit their commercial interest. Trading across international boundaries is assuming significance in recent years. The Biotechnology Revolution and Information Technology are at the cutting edge of such developments. This calls for some uniformity of laws in different countries to facilitate the spread of knowledge and diffusion of innovative technologies. The TRIPS agreement under the WTO is an attempt at such uniformity. The three broad components of the IPR regime are • degree of protection; rigid and excessive protection • use of patents and enforcement; abuse of IPRS • dispute settlement; especially restrictions on technology diffusion It is expected that TRIPS agreement will alter the patterns of • • • •

technology transfer and imitation foreign direct investment domestic innovation international incentives for trade

IPRS and Patents

47

Assume that an individual or a firm in country 1 developed a new product and applied for a patent in his country. The knowledge necessary to manufacture the product is no longer a secret even across international boundaries. Hence, two possibilities exist. •

Another individual in country 2 may develop a somewhat different process of producing that product, undertake manufacture, and sell it in his country. • Having produced the output, perhaps at a lower cost, the firm in country 2 may export the product to country 1 and undercut it. The TRIPS agreement attempts to prevent these by insisting that •

patents must be for products and not just processes; what this means is that a product, irrespective of the process of arriving at it, will be considered equivalent and hence a violation of patent rights of the firm in country 1 • there must be uniformity over patent life over which the innovator will have a monopoly right of refusal for others to use the production process; TRIPS defines it as 20 years for all products and/or technologies • the original patent holder should have exclusive marketing rights (EMRs) The firm in country 1 applies for a patent in country 2 as well. Within the rules of the TRIPS agreement country 2 can refuse the patent only if • the products and technologies are morally or ethically indefensible • they harm human health • they are inimical to national safety Three aspects of EMRs must be noted. •

Suppose the firm in country 1 applies for a patent in country 2. However, the grant of a patent takes some time. Further, in the case of agrochemicals and products the right to sell is commercially more critical. Under TRIPS agreement EMRs must be granted to the patent applicant for five years or until the patent application is decided. Of course, such EMRs apply only if the firm has a EMR in some other country • EMRs can prohibit imitators from selling the product in country 2 even if they can produce it. • The grant of an EMR in country 1 also means that even the patent holder cannot import the product from country 2. This is a protection against the possible cost and price differentials in the two countries. It was necessary for TRIPS to address another aspect of patent grants. Assume that country 2 granted patent rights to the firm in country 1. Clearly, they have done so, expecting • •

the possible spread of knowledge and technology in country 2 the availability of the innovative products in country 2

On occasions the firm from country 1 may try to use its monopoly power and refuse to invest in country 2, produce the output in country 2, and/or market it there. This may be

48


deemed an abuse of the patent. TRIPS provides a mechanism by which country 2 can force the firm to issue a compulsory license to another firm in country 2. More specifically, a firm in country 2 can apply for a compulsory license provided •

the original innovator does not start production in country 2 even after three years of the grant of the patent in that country

•

production is not started even after four years of the patent filing in country 2

However, TRIPS stipulates that the compulsory license need not be granted if the patent holder can offer valid reasons. Suppose a compulsory license is not granted. Does this mean that the patent holder forfeits his rights? TRIPS agreement allows the patent holder two years time to rectify the situation. The rights will be deemed to have been forfeited after that. There is another dimension of compulsory licensing under the TRIPS agreement. Host countries may require the patent holder to provide a license to a local producer at a reasonable cost in case of a national emergency. It is difficult to define what constitutes a “national emergency” and what a “reasonable cost” will be. This may lead to violation of patents somewhat arbitrarily. Under PVP a provision for compulsory licensing is dictated by the strong commitment for public interest. Under this provision a holder of plant breeder’s rights cannot •

refuse any applicant

•

impose unreasonable terms of license

The material transfer agreements for genetic materials are also of this nature. Note that this is important simply because most of the genetic inventions are not final products. Instead, they have to be developed further. Hence, the spirit of the patent will be violated if knowledge diffusion is blocked. The TRIPS agreement considered one other aspect. Suppose the firm in country 1, i.e., the patent holder, sold some units of the commodity to an individual X in country 2. X may, in his turn, sell it to others in country 2. Will this be a violation of patent rights? The TRIPS agreement gives the rights of decision to country 2. For, the patent holder is deemed to have exhausted his rights after the sale to X. Assume that X sells the product to someone in country 1. Will this be a violation? As per TRIPS agreement it will not be so provided country 1 offered a most favored nation treatment to country 2. In the context of biotechnology the firm has to go through a regulatory approval process after applying for a patent. In the context of agricultural biotechnology this involves extensive field trials. The pharmaceutical sector has to go through clinical tests. In both these contexts the patent holder must submit all the test data to the regulator. It was therefore pertinent to find out if this data, like the patent filing details, become public property that everyone can access freely. The developed countries insist that this expensive data must be protected. Under the TRIPS agreement such test data remains an exclusive property of the patent holder for five years from the point of regulatory approval or the end of the patent life

IPRS and Patents

49

whichever is earlier. Anyone else that desires to use such data must get a license from the patent holder. In general, the attitude of the developed countries is that they should have monopoly rights • over the technology since they developed it • over production and monopoly pricing for goods that they alone can sell • to make it difficult, if not impossible, for others to develop the technology The developing countries doubt the sincerity of the developed countries with respect to developing their technical capabilities and/or free trade. The commercial interests of the developed countries dominate the TRIPS agreements. Geographical indicators is another contested area of the TRIPS agreement. Article 22.1 defines them as “indications which identify a good as originating in the territory of a member, or a region or locality in that territory, where a given quality, reputation or (some) other characteristic of the good is (typical) to its geographic origin.” To be more specific, consider the following. Neem trees can grow everywhere. One country calling it neem cannot prevent others from using the same product (the only thing is that it may have to be called something else and not neem; that is the trademark or copyright for the label alone can be protected legally). In that case nobody can have any exclusive right. There is a different viewpoint. Basmati rice was produced in some parts of India for centuries. Some other country may now take this variety, produce it, and call it basmati as well. Does this give them a right to exclusivity just because it has not been patented in India earlier? Two issues have arisen. • •

Can neem, basmati, and so on be considered as geographical indicators? Should patents and formal written documents, that are of much recent origin than traditional knowledge transmitted orally through generations, be a requirement to acknowledge intellectual property? Can developing countries really afford the costs of doing this? Will it really add to their social welfare? There is no easy resolution. However, TRIPS acknowledges that they can be protected through copyrights (again a formal registration). This prevents the use of a specific word like basmati. However, it cannot prevent others from its production, use, or sales under a different label so long as the consumer is not deceived into believing that the product is original basmati. Bioinformatics will also be covered under copyrights. Such rights prohibit copying or reproduction of protected work. However, TRIPS introduces rental rights, i.e., the right to authorize others to use it for a rent. It is difficult to monitor the use beyond the legitimate first renter. The advantages to developing countries, if any, as a result of the introduction of these agreements, are not at all obvious. Some investigations in this direction will be outlined in the next section.

50


3.6 CONSEQUENCES OF PROTECTION There is a fairly general acknowledgement that IPR protections can have far reaching consequences for the economic functioning of markets. Prominent among them are the following. • • • • •

Industrial concentration and monopoly power Prices of products Changes in social welfare Incentives for local firms to develop competitive research and product variety Discourage developmental work since generics cannot be introduced until the patent expires • Foreign direct investment • Technology transfer However, very limited information is available about many of these aspects. In particular, •

much of the literature deals with chemical based pharmaceutical industry and some of it deals explicitly with the experiences of developing countries. Consequently, it is difficult to assert that these results carry over to biotechnology patents as well • the studies relating to agricultural biotechnology mostly relate to the U.S.A. A few aspects are also pertinent in the context of less developed countries This limited information will be presented in this section to provide a general direction to this analysis. Consider each of these aspects in turn. Assume that a MNC succeeded in generating a new innovation. Suppose a patent is granted. The initial impact of this is a monopoly and an increase in concentration. However, it is expected that knowledge is freely available. Consequently, there is an expectation that it stimulates others to work around it, create new methods and varieties, and patent them in turn. The more the variety and patents (in a given therapeutic class or for a specific disease) the greater the competition in the long run. This may materialize even before the expiry of the original patent. But this may not happen. For, •

•

•

it may not be easy to invent around the patent (difficulty with scientific knowledge itself). Schimmelpfennig et al. (2004) noted that it has been difficult to build up varieties in cotton and corn (Bt varieties). However, varieties of soybean (RR varieties) have been relatively easy a developing country may not have the scientific base and absorptive capacity to create new knowledge even if it is possible. (see, for instance, Maria and Ramani (2004).) successive introduction of varieties may have smaller market shares to make them commercially viable; i.e., even the MNC will not find proliferation of varieties profitable

IPRS and Patents

51

Lanjouw (1997) also observed that once a patent is granted in the country where the MNC is located, all others, including potential international competitors, have access to it. As such a patent granted for the same product is neither necessary nor sufficient for knowledge diffusion and variety enhancing patents in less developed countries. In general, it is difficult to determine the optimal product variety. Hence, it is not possible to clearly say that a specific variety that materializes as a result of IPR protection is inefficient. Studies, such as Subramanian (1995) and Watal (1995), considered the expected price increases due to the monopoly power inherent in the grant of IPR protection. Based on reasonable estimates of the elasticity of demand and marginal costs they found that the expected price increase in drugs can be anywhere upto 75 percent of the pre-1995 prices. However, as Watal and Mathai (1995) noted, it is difficult to attribute the entire low elasticity of demand to IPR protection alone. For, the large chemical companies derive significant market advantages from their marketing strategies, trademarks, and other promotional campaigns as well. It is widely believed that patent holders cannot charge monopoly prices as expected by earlier studies. For, there exist several substitutable drug varieties in every therapeutic class. Some of them may be older off patent drugs. Others may be produced under competitive patents. One of the examples cited is that of quinolenes (antibiotics). This suggests that an innovator has two types of competitors; producers of different rival products and firms dealing in generic drugs. IPRs may give rise to competitive variety that restricts the price increases that any one patent holder may be in a position to exploit. This is the general position of the research based pharmaceutical MNCs. Three forces are at work in determining the prices of such variety. First, suppose a patented product is sold at a high price. Then, the demand curve for substitutable products shifts to the right. The rival firms then choose prices to maximize their profits. They will be generally higher. This is the general conjecture in Chamberlin’s monopolistic competition. Hence, larger variety may result in cascading price increases rather than reduction. Second, in addition to the number of substitutable products the elasticity of substitution between them also matters. In general, a lower elasticity of substitution and therefore a lower elasticity of demand, has the effect of increasing prices. Third, across international boundaries, where transfer pricing regulations apply, it is not possible to maintain significantly different prices. Fink (2000) found the variety effect of hypotensives to be price reducing. A similar result was reported in the context of quinolenes by Chaudhuri et al. (2003). Litchenberg and Philipson (2002) offered a general theoretical discussion of these substitution effects. However, it is generally pointed out that the estimated price effects are quite sensitive to the elasticity of substitution among varieties within a therapeutic class. An important policy implication of these price changes was suggested in Chaudhuri et al. (2003). When an IPR protection results in price increases the host country may impose price controls. There is a limit to this because of the transfer pricing regulations. Suppose there is a variety (or access) effect. Then, a compulsory licensing arrangement, to the extent it makes variety possible, may be a superior policy from the viewpoint of the less developed country.

52


Recall that a shift of some part of the consumer surplus to producers is the other expected effect of IPR protection. In general, the results suggest that •

profits for the MNCs from developing countries may be a fairly significant share of consumer welfare though the quantum is too small for the MNCs (relative to their own home country) and also the costs of seeking and maintaining a patent in the developing country • the increase in product variety generally reduces the profit accruing to the MNC. The more important point from Watal and Mathai (1995) is that these profits accrue to the MNC and often repatriated to their parent country. Hence, an increased IPR protection, that a less developed country offers, is unlikely to result in any increased R&D in that country. The results with respect to welfare losses, measured by the deadweight loss (or the income compensation necessary to maintain the same level of utility) exhibit similar results. Fink (2000) also pointed out that “welfare losses are lower the more price elastic is overall demand in a therapeutic group and higher the degree of substitutability among chemical entities. The latter effect is relatively more pronounced … because the presence of a larger off-patent market segment makes therapeutic competition more effective.” See also Chaudhuri et al. (2003). Innovations pertaining to agricultural biotechnology are generally delivered through seeds. Hence, IPR protection to seed companies may allow them to increase the prices of seed. This may not be fully compensated by the reduction in other costs like spraying pesticides. The increase in the equilibrium prices of crops and the corresponding welfare losses depend on the shift in the supply curve. Moschini (2001) reported some attempts to disentangle the various effects of IPRs. It is often suggested that pharmaceutical firms in developing countries, like India and elsewhere, have the technical capabilities to generate R&D. For, after all, in the absence of patent protection, they could successfully create a market for generic drugs. Hence, it can be expected that providing a strong patent protection will spur them into relevant R&D. As Lanjouw (1997) pointed out, the large pharmaceutical firm may be compelled to do this if the generics route is eliminated. However, there is a possibility that they will find it more profitable to obtain technology from MNCs on the basis of license instead of their own R&D. Similarly, the firms in the less developed countries may find it more advantageous to do clinical testing work for the MNCs on an outsourcing basis. The risks involved in the large amounts of investment as well as the difficulties of making suitable financial arrangements may deter them from doing R&D work that will result in patents for them. Local development of competitive drugs or development of drugs for locally specific diseases like malaria and leprosy appear to be a remote prospect. An obvious negative effect of strong IPR protection is more likely. For, even the R&D efforts they undertook for the development of generic drugs would be lost. On the whole, the gains to domestic pharmaceutical firms in less developed countries may turn out to be negative. Observe that when a MNC is given patent protection in a less developed country the profits accrue to the former. The claim is that this will spur MNCs to offer technology transfer

IPRS and Patents

53

to less developed countries through foreign direct investment. Similarly, it was suggested that they may invest on R&D for local diseases. However, the MNCs generally feel that •

the profits are not large enough to be attractive (drug delivery at personal cost to the patient rather than health insurance or national health schemes is one of the deterrents) • the presumed cost advantages in less developed countries (of cheap skilled labor availability) is offset by other administrative bottlenecks and tariffs on import of precision machinery etc. • there are problems of infrastructure, information acquisition, commitment to contracts (Maria and Ramani (2004)), and price controls On balance, there is a general feeling that the less developed countries will lose, and definitely not gain anything, by offering strong IPR protection to MNCs. If there is some enterprising activity to begin with then IPRs may act as catalysts but they cannot, by themselves, spur local firms to become more enterprising in the short run.

3.7 ISSUES OF CONCERN There have been serious ethical and environmental concerns about the use of biotechnology products especially with respect to food products. This resulted in a labeling and testing regime for agricultural products traded across international boundaries. The precautionary approach, viz., to reject a lot if an importing country is not satisfied about its safety, has been put in place. However, this does not receive any specific mention in the TRIPS agreement. The mandatory labeling of GMOs and GM derived products may turn out to be a powerful incentive to develop handling and processing systems characterized by market segmentation and/or identity preservation. Implementing an identity preservation system is costly because to supply non-GM goods it is necessary to keep traditional crops segregated from the GM varieties throughout the production and marketing system—through production, storage, transportation, processing, and distribution. This may lead to a reduction in social welfare. The problem is that this is also a non-tariff barrier under WTO and is illegal. But it cannot be neglected. The other issue is about differential pricing. To understand the simplest mechanics of its operation reconsider Fig.1. In this diagram the monopoly is offering a quantity Ym at price pm. However, from a social welfare viewpoint the most efficient output is Yw at price pw. To keep the profit incentives in tact while attempting maximization of social welfare it is possible to visualize market segmentation. For example, the regulator may insist that the patent holder should supply the output YwYm at price pw. This restores welfare maximum. If the operation of the market is within a country such segmentation may have to be achieved through the physician network and suitable labeling and prescription. If the lower price segment is the market in a less developed country the issues involved are different. First, under the transfer pricing regulations a MNC cannot charge a price p1 in country 1 that is far below the lowest price pI it charges in any other country. Hence, this arrangement may not be practical. In fact,

54


it is possible that the MNC sets prices for all its segments keeping the global market demand in perspective. This fails to maximize social welfare. Second, if the MNC is unwilling to adopt differential pricing, a less developed country may force it to accept compulsory licensing to a domestic firm that can deliver the output at a lower price. Third, even if such differential pricing is possible, there is a necessity for institutional and legal mechanisms to ensure that the product is not privately traded across the market segments. As noted earlier TRIPS agreement does not specify any mechanism to achieve this. In general, as the IPR commission acknowledged, it is expensive for less developed countries to establish and operate elaborate IPR protections under the TRIPS agreement. It may not be socially desirable to divert resources, which are already inadequate to take care of education and health, toward administering a costly TRIPS agreement. Hence, several dilemmas remain to be addressed.

3.8 MODIFICATIONS TO PATENT REGIME Property rights generally consist of • a right to own • a right to use or rent it to others • a right to modify Implicitly such property rights allow the owner to exclude others from the use of such property • if he so chooses • if others are not willing to pay the rent specified The present day patent regime confers these property rights. However, note that such property rights applied historically to final consumers. The question now is: should any or all these rights apply to knowledge development as well? The existing patent regime appears to accept this without any reservations. However, Article 27.3 of the TRIPS agreement allows governments of sovereign countries to exclude certain types of inventions from patents if national interests are at stake. That is, one extreme form of reaction is to deny such patent rights altogether. This results in secrecy and hinders knowledge diffusion. In some cases, like the National Institute of Health in the U.S., the agency stipulates that they will not allow certain types of discoveries of fundamental knowledge to be patented. In the absence of an objective way of classifying different types of knowledge development and specifying entitlements this remains subjective. The question about financing such R&D is also pertinent. This may not be a serious issue so long as adequate public funding is available for all such fundamental research. However, over the years, there has been a reduction in the government involvement in biotechnology R&D primarily because it did not contribute to defense needs in any major way. There is a

IPRS and Patents

55

possibility that R&D financed by such institutions would be inadequate. A further objection may be raised even if adequate finances are provided. In particular, the contribution of any one fundamental R&D to the value addition obtained from a final product of utility may far exceed the cost of generating it. Therefore, any knowledge that contributes to such private value addition in the ultimate analysis should be compensated adequately. Public funding institutions may not be in a position to ascertain such value additions apriori let alone compensate the innovators adequately. In other words, denying patents per se also hinders knowledge generation and diffusion. Rai (2005a) argues that there is a need for improving access by requiring publicly funded scientists and research institutions to put data and certain types of research into the public domain, or, at a minimum, to license them widely and non-exclusively for a reasonable fees. Non-exclusivity reduces transaction costs and improves the range and quality of resulting products. However, the question of deciding what constitutes reasonable fees cannot be resolved objectively. Further, under the current patent regime, there is no way of compelling private firms to accept non-exclusivity. This solution is also not adequate since private R&D constitutes a major portion of biotechnology research. Some individuals, that patented discoveries, voluntarily agreed to offer non-exclusive access to their knowledge to everyone that needs to use it to move the knowledge forward. The Cohen and Boyer patent for rDNA is one such example. There are two problems with this approach. First, the problem of cost recovery must be resolved. One argument is that in the context of biotechnology mere transfer of formal knowledge will not be sufficient to use it. The scientist, that allows exclusive access to the knowledge he developed, may still charge a consultancy fees for providing the informal knowledge. This may, in itself, be sufficient especially when the use of knowledge spreads widely. Second, there is a possibility that only the manufacturer of a final product sold on the market will usurp all the benefits of the chain of discoveries. Hence, most innovators will be reluctant to use this approach. Another alternative is to persuade the patent holder to offer his patented knowledge on a collaborative basis or an open source mode. Rai (2005b) is depending on voluntary action after granting exclusivity. The problem with this argument is as follows. Suppose a scientist is granted a patent with the exclusivity clause in place and with a broad scope. Why should he agree to share it for free or at low cost? In particular, a scientist, who is aware that the final product developer is capturing the entire value added, will not accept this arrangement. Rai (2005b) then falls back on the argument that markets in developing countries add very low value to patents. Hence, she feels that innovators can easily be persuaded to provide patented knowledge to them on a non-exclusive basis. But this creates a variety of new problems if firms in developing countries, in turn, sell their products in industrial countries. Hence, even this approach is not practical. If none of the above solutions appear practically feasible the only option is to leave the decision, to enforce exclusivity or leave the knowledge as a public good, to the scientists themselves and/or the institutions that they belong to. However, the current patent regime already made them feel that they can derive benefits by exploiting the monopoly power

56


granted by knowledge patents. Consequently, it would be unrealistic to expect them to relent. Rai and Eisenberg (2004) alluded to such attitudes of WARF (the Wisconsin Alumni Research Foundation). It would indeed be paradoxical to grant patent rights, with all its implications of property rights, and then expect the patent holders to be persuaded that they should allow nonexclusive use of their patent in the larger interest of social welfare. For all practical purposes it is by now acknowledged that discoveries of knowledge need protection by applying the utility doctrine with somewhat greater flexibility (not insisting on utility to a consumer of final products). This appears to be a necessity in the context of biotechnology. For, private individuals and firms, that finance such activities, cannot recover the high costs of R&D otherwise. Exclusivity enables patent holders to claim royalties from those that use their results of R&D. The exclusivity clause has become a hindrance to knowledge diffusion though patent filing releases information in the public domain. Several observers argued a case for narrowing patent rights to restore parity. See, for example, Abrol (2005), Correa (2005), and Rai (2005a). However, as noted in the previous section, only Rai (2005a,b) has some analysis of alternatives. Two situations are conceivable. First, there may be a cumulative chain of ‘n’ inventions before a final product emerges. Second, a new R&D effort, perhaps at the downstream level, may require knowledge embodied in ‘n’ earlier patented innovations that are not interrelated in the above sense. The requirements of knowledge diffusion in these two cases are somewhat different. In the first case each stage of invention can be considered as a different marketable product. The patent holder may then be allowed to negotiate a license to a user at the downstream level and extract rents based on the perceptions of the two parties regarding the value added in that particular use. Similarly, he may grant licenses to many firms pursuing different types of applications and developments. However, in line with the conventional patent regime, the patent holder’s rights should be deemed to have been exhausted after the first stage license. This reduces the burden of negotiating a license with all the early down stream innovators. The main advantage of this approach is in allowing faster diffusion of knowledge. It also reduces the burden of transaction costs on the final stage innovator. Clearly, the patent holder at each stage may negotiate the license based on his perception of value that his knowledge contributes. The only disadvantage is that early stage innovators may not be in a position to assess the ultimate market value of their innovation. Will the patent holder not grant licenses to competitors pursuing the development of the same marketable product? There may be some short run difficulties. However, long term reputation will be at stake if such moral hazard persists. Hence, it is unlikely in the long run. A second approach is more practical. Begin with the observation that the innovator at the final product stage can recover costs when a marketable final product is available. The licensing contract for upstream knowledge is similar to subcontracts for parts and franchise bidding in any other industrial context. Confronted with risks in the ultimate product market

IPRS and Patents

57

the upstream firm has two choices. First, choose the rents ex ante to resolve the risk. This is somewhat similar to the model alluded to above. Second, wait for the risk to be resolved ex post and claim returns accordingly. More often than not, the second choice is more efficient and it can be implemented if knowledge about the market is not difficult to verify. This suggests the following modification to the patent regime. Suppose that the system of patenting knowledge is continued with two conditions. First, each of the early stage innovators will be under obligation to provide the knowledge on a non-exclusionary basis. Second, the entire chain of related innovators must be compensated if and when later stage R&D results in a marketable product. This accelerates knowledge diffusion while preserving the appropriability of intermediate discoveries of knowledge. The practical problem of apportioning the eventual benefits among them can be resolved by making payments proportional to the costs incurred at each stage of R&D development. This generally results in recovering more than the costs involved and closer to the contribution of any one aspect of knowledge to the ultimate value. Hence, the objections raised in the context of public funding of R&D will not arise. The patent application must normally specify the territory that the applicant considers his own and thus exclude others from it. Hence, the following subtle points can be introduced. First, he may be required to spell out the developments for which the use of knowledge will be allowed on a non-exclusionary basis. Two examples can be offered. (a) The knowledge under consideration is not known to result in any marketable product either within the scope of the patent or outside of it. (b) In some cases there may not be any market for early knowledge developments because utility is not obvious. Second, there may be some parts of scientific knowledge that he considers far removed from a final product. In such a case he may be asked to specify the parts of knowledge that he will exclude unless a payment is made. Note that the second model proposed here is quite general. It is applicable even in the case where the ‘n’ upstream innovations are not cumulatively related. Further, as noted above, this model may turn out to be superior even in the context of cumulative knowledge developments. However, in general, it must be noted that the new model will be usable only if the patent application contains information about all the prior patented knowledge utilized in the downstream development. It is also necessary for every scientist, at the intermediate stages of development, to declare the costs of their R&D. From an operational viewpoint there may be a necessity for some institutional mechanism to ensure that payments are properly made and redress grievances if they arise. One objection to the new scheme may still arise. Note that under the present patent regime a scientist can claim payments as early as possible. What then is the incentive for him to wait until some final product of utility is marketed? Two points may be recorded as possible answers to this question. First, it is well known, from the economic theory of incomplete contracts, that such ex ante resolution is inefficient under conditions of risk. Second, under the present patent regime there will be fewer users and/or uses of knowledge that the patent holder developed. This is primarily due to the costs that must be paid before the value is realized and the extensive transaction costs. When the new model is in operation, there will

58


be widespread knowledge diffusion and a better chance of value enhancement. Hence, the losses due to the time lags may be more than compensated. However, it must be acknowledged that this is an empirical matter so that any apriori judgement about the superiority of one over the other may not be warranted. On the whole, as in the established practice, accepting full property rights for products of utility can be justified. However, they should not apply to at all stages of knowledge development. Otherwise the pious hope that patents will result in extensive knowledge diffusion will not materialize.

Investment and Financing

59

Chapter 4

INVESTMENT AND FINANCING

4.1 THE ISSUES In the context of the biotechnology industry R&D activities are of several types. Prominent R&D activities are in areas like • basic scientific discoveries of cell structure, protein synthesis • the process of transforming laboratory research into marketable products • clinical tests of drugs or field trials of crops • the manufacturing and market release of final products The first two phases of biotechnology industry are such that many new NBFs do not have any developed and readily marketable product. It was noted in chapter 2 that the primary consideration, of a NBF considering an alliance with a university scientist, is the number of patents he has and how active he is in current scientific developments. This will determine the usefulness and the likely success of the NBF. It also defines its ability to convince would be investors about its potential to generate the requisite finances. Similarly, a NBF with many patents on industrial processes will be an advantage to a large firm forging links with it. In other words, when considering the R&D decision at the earlier stages, where there is no tangible marketable output, the main consideration is the market value of its assets. This may be purely a valuation by the accountants or the capital market. The products of the NBF are essentially scientific knowledge with little or no established reputation and experience high rates of obsolescence. They also lack tangible assets. As a result their growth depends primarily on personal savings (and may be angel capital from friends). The government recognizes the social value of such discoveries and comes to their rescue by

60


providing finances, facilities, tax subsidies and so on. Later stage firms, or mature NBFs, derive their finances from conventional sources comprising of • venture capital sources • commercial banks and financial institutions • capital markets • private equity placements • investment from foreign MNCs At higher levels of the biotechnology value chain, involving manufacture and distribution of marketable products targeted to specific consumer needs, the innovative knowledge must be embedded in physical capital. Investments in R&D and physical capital will be complementary assets at this stage of development. Private biotechnology firms determine both of them to maximize the discounted cashflows that they generate by utilizing the market to conduct the sale of their products. One important factor in generating the cashflows is the nature of market demand. Consider agricultural biotechnology. The seed industry, especially the one dealing with GM seeds, is highly concentrated. They, however, have competition from conventional varieties. It should also be noted that the conventional varieties, over time, have been adapted to local agro climatic conditions whereas the currently available GM varieties are not equally adaptable. These features make the demand for GM seeds somewhat elastic despite the monopoly achieved through IPR protection. Another important feature of the seed industry is the extension services. In general, the private companies, that sell GM seeds, offer assistance regarding the application of pesticides, herbicides, and so on. But the farmer has to pay for these. Further, the assistance is generally not site specific. On the contrary, the government funded agricultural extension services work closely with the farmer and offer specific assistance to each farmer. The demand for GM seeds depends on these arrangements as well. Several factors determine the demand for GM crops and foods produced from them. The most important among them is the consumer resistance to GM crops. Many still prefer nonGM varieties. The demand for these varieties is, however, depending on the cost of labeling GM products. To a large extent this problem is affecting the demand for GM crops across national boundaries as well. Price controls and support prices have always been a determinant of demand. The differences in the marketing and distribution channels also contribute to this. Pharmaceutical products are subject to a different set of problems. • • •

• •

The patent protection for drugs is far more stringent compared to GM foods There is much greater monopoly power However, there are conventional chemical technology based drugs in many therapeutic classes. This variety, along with some alternatives in the biotechnology area, offer competition. This is especially valid in countries like the U.S.A. where even alternative formulations have been granted independent patents. See, for instance, the examples in Correa (2001). In the less developed countries the drug market experiences significant price controls. There is no meaningful national health scheme or insurance cover against illness in several countries. The patients must pay for the expensive drugs from their


61

incomes. This reduces the demand considerably. The situation is better wherever national health schemes cover at least some expenses of health care. So it is with insurance. The basic point is that patterns of drug delivery have a fundamental bearing on the demand for pharmaceutical products. The cost implication of R&D and the production of the resulting high quality products are equally important in determining the growth of biotechnology firms. For, fundamentally, a discovery is useful only if it results in the reduction of cost of delivering a unit of product of given quality. The high costs and risks associated with the development of biotechnology necessitated public policies and spending to ensure that socially desirable innovations materialize. As noted earlier, R&D in biotechnology products and the actual production of specific output depend on the financial arrangements. In addition to the sources mentioned above, the following have also been significant. •

Intercorporate and private placement of equity holdings tend to cushion some of the risks of these investments • Venture capital has also been active at this higher level of the value chain However, it should be noted that public funding of activities at this stage of production are much less compared to the first two phases of the development of biotechnology. In general, even public institutions have been encouraged to raise finances by licensing their patented technology to private firms. The investments in basic R&D and later stages of biotechnology will now be examined with these aspects in perspective. As Cohen and Levin (1989) noted, there is no justification to exclusively emphasize the effect of market structure as the dominant determinant of R&D and capital investments in the biotechnology sector.

4.2 ROLE OF PUBLIC INVESTMENT Most technologies, including the internet development, have four distinct characteristics. •

The initial developments have been sponsored and financed by the government because they have some implications for defense • The published formal knowledge (in some cases through patent filing) is adequate for others to reproduce it and create materials to use it. In particular, there was no specific necessity for the original scientist to participate in the transfer of knowledge. That is, the requirement of informal knowledge is minimal or non-existent • There is no necessity for transfer of any physical materials or laboratory tools along with the knowledge • Most of them are directly related to products of utility to consumers Against this backdrop the universities, where knowledge development is initiated, trained young scientists in the new discoveries and knowledge. The private firms then used this pool of

62


talent to implement the technological breakthrough. In sum, the development of technological knowledge has been considered as a public good and transferred to the private firms through the educational process. In a few cases, like the agriculture extension services, even public research organizations participated in the development and transfer of the requisite R&D. Biotechnology development, at least in the initial phases, has been different in all these respects. In addition, much of the R&D remains in the hands of private firms due to the absence of any utility for defense purposes. However, there has been a significant involvement of the public sector in basic R&D in countries like Germany, Japan, and India due to several institutional and cultural considerations. In particular, the public sector had a role in • setting up research laboratories and sponsoring research • transfer of knowledge to the private firms • finances and other incentives to reduce risks See, for instance, Giesecke (2000) and Lehrer and Asakawa (2004). One of the important questions pertain to the relative proportions of the private and public R&D. In addition to this quantitative dimension the major issue is about the areas of application that would be pertinent to each of these sectors of activity. The following framework may therefore be utilized to examine some of the issues involved in the efficient transfer of public scientific knowledge to private firms. Initially a private firm identifies promising R&D activities. However, it does not have the scientific knowledge, considers the risks too high, and/or appropriability too low. The private firm expects the government to resolve the risk before they undertake further development (bioprocessing for instance) and production. Hence, it is possible to view the government as a partner in the development of R&D after the private firm spots some worthwhile activity. Clearly, the possibility that the public sector initiates the necessary action for the private firms to follow it up at an appropriate time is also a real possibility. Precedence may not be the essence of creating socially beneficial R&D. See, for example, Khush (2004). Another important feature of the development of biotechnology is the emergence of the new biotechnology firms (NBFs). They have been the essential link between specialized research institutions (either public or private) and the large pharmaceutical and seed companies that manufacture and market the final products. The basic reasons are as follows. No private firm has adequate scientific manpower or the laboratory tools to undertake the more fundamental activities in the early phases of their development. Perforce they must be developed and incubated in specialized research institutions (either public or private) and transferred to private firms through appropriate organizational mechanisms. These technologies are also such that a formal transfer of blueprints (or the disclosure at the patent filing stage) is generally inadequate to make efficient use of the results of R&D. The transfer of informal knowledge, made possible only through close interaction with the scientists, has become a compulsion. Irrespective of the composition chosen, the government and its institutions had to play a crucial role in providing arrangements for technology transfer between various actors in the network.


63

Hence, it should be acknowledged that developing network organizations, for the transfer of informal knowledge in their functioning, will be critical. Realistically, when private and/ or public firms depend on the transfer of informal knowledge, they learn to identify appropriate network members and they in turn learn to work in teams. This process results in organizational learning, spillovers, and concomitant improvements in technology and productivity. There have been two other reasons for the emergence of the NBFs. First, most of the developments of biotechnology are capital intensive. The institutions undertaking the basic R&D generally do not have access to requisite finances. The NBFs, with their more significant capabilities in scaling up technology for commercialization, have better access to venture capital and other sources of finance. Second, there have been significant risks in R&D at the present stage of biotechnology development. In particular, many research findings may fail at the bioprocessing or the regulatory stage so that cost recovery is not commercially possible. The NBFs act as a buffer between R&D institutions and the large firms and absorb the risks that are intrinsic to progress in biotechnology. For many years, knowledge developed in R&D laboratories (both private and public) was considered a public good and was accessible to anyone that wished to utilize it and move it to the market. However, after the 1980s, the patent regime allowed patenting of such knowledge and discoveries from R&D. Further, public organizations in many countries have been encouraged to recover the costs of R&D through contracts and licensing. Creating an atmosphere for R&D and its efficient diffusion through patenting and other public policy measures has also been an important feature of the developments in biotechnology. In sum, public policy with respect to the biotechnology sector consists of the following. •

The government may create and/or support public institutions that undertake such research • Public R&D may precede private R&D in order to create the atmosphere for the later to flourish • Public institutions may offer agricultural extension services to reduce costs • The government may finance such private R&D • Public policy may nurture venture capital and foreign direct investment • Public financing of national health schemes may augment the demand for medicines and their appropriability • The government may create a suitable patent and IPR regime Ramani (2002) has a perceptive observation. As she pointed out, the above approaches to public policy focused on two ends of commercialization, viz., public research organizations and final product manufacturers. The effort was to retain the decision making autonomy of each of the institutions involved, to the extent possible, and hope that formal interaction between them will develop to benefit society at large. This was also noted in Raina (2003). However, as Ramani pointed out, “the indispensable intermediate link to (translate) scientific knowledge into technological competence was largely skipped”.

64


4.3 R&D IN SCIENTIFIC KNOWLEDGE The simplest formal model of R&D expenditures in the early stages of biotechnology development proceeds as follows. Let E be the expenditure on R&D at any point of time t. To begin with assume that a private firm is financing the entire expenditure. This is usually designated as an input measure of R&D. It would be equally plausible to conceptualize the number and strength of patents obtained by the firm as a measure of R&D. This will be the output measure of R&D. The fundamental nature of the analysis will not change materially irrespective of which measure is employed. Hence, the rest of the analysis will be based on the input measure. The scientific knowledge so discovered accumulates over time. Such R&D may be useful in accelerating the nature of discoveries as well because each successive discovery is built on past knowledge and experiences. This common observation is reiterated in Cohen and Levin (1989). Denote by R the stock of this accumulated R&D. It is generally acknowledged that new discoveries make some existing knowledge obsolete. This is the spirit of Schumpeterian creative destruction. Bottezzi et al. (2001, p.1164) summarized this process in the following way. When a new innovation enters the market it is rapidly diffused throughout the market. Invariably this is achieved through a displacement of older varieties where they exist. It also spurs competitive R&D in generating competing varieties, e.g., equivalent drugs in the same therapeutic class. It is important to recognize that such changes tend to occur before the patent on the original discovery expires. However, the rate of decay of R depends on many factors. Assume that εR; ε > 0 represents the reduction in the stock of R at any point of time. The trajectory of R over time can therefore be represented by dR/dt = E – εR The firm can be expected to choose E at each point of time so as to maximize the present discounted value of the valuation that accountants and/or other interested parties assign to the stock of R&D of the firm. The valuation of R is subject to diminishing returns for the following reasons. •

•

•

•

An increase in R generally indicates that the firm has already appropriated most of the profitable investment opportunities. The incremental value expected from further R&D will consist of activities that yield lower returns. As the level of R&D increases, and the NBF becomes larger, it requires extensive network relationships. Due to organizational differences between the constituents and associated coordination problems it must be expected that there will be diminishing returns to increases in R. A large NBF may develop greater monopoly power. However, in line with the Spence formula, it becomes difficult for it to extract all the value that the potential users of the technology associate with it. An increase in R also suggests that the firm has used up most of the financial resources available to invest in R&D. Hence, further R&D can be taken up only by


65

using more expensive sources of finance. This reduces the additional value generating potential of R&D. • Much of R&D, in particular, activities related to biotechnology, is very specific to the firm in the initial stages when a patent is in force. If there is any failure it becomes very difficult for the firm to recover its finances due to the low liquidation value of these specific assets. See, for instance, Correa (2001) and Sporleder and Moss (2004). Hence, the gross value addition from a stock R of R&D can be written as V = Rα; 0 ≤ α ≤ 1 This has been achieved at a cost E at time t. Therefore, the net value addition is V = Rα – E The choice problem for the firm is therefore to define E as a function of time in such a way as to maximize ∞

∫ e–rt (Rα – E) dt 0

subject to dR/dt = E – εR This problem in calculus of variations can be solved by Pontryagin’s maximum principle. It results in a choice of E such that αRα–1 = r + ε Note that a net unit increase in R necessitates an increase in E equal to (1 + ε/r) where 0

– ∫ e–rtε dt = ε/r

–∞

represents the depreciation of all earlier R&D stock. Hence, the interest cost (or, cost per unit increase in R) will be r+ε. This is usually designated as the user cost of R&D stock. Similarly, αRα–1 is the increase in the value of the R&D stock per unit increase in it. Hence, the choice of R, and consequently E, is optimal if αRα–1 = r + ε Following Jorgenson (1963) this can be written as R/V = α/(r + ε) This way of expressing the solution suggests that the optimal R • •

increases with the gross value V that the firm desires to achieve increases if the productivity (or the value generating potential) a of R is higher

66


decreases with an increase in the user cost. In particular, if the firm experiences faster decay of its R&D stock due to competitive pressures, it will tend to reduce R

•

decreases with an increase in the discount rate (this is a reflection of the willingness on the part of the entrepreneur to wait for a longer time to recover the value)

Assume that the government provides IPR protection to all such investments. One of the consequences of this policy is that ε will fall because competitive imitation is no longer possible. The only effective competition is by creating a new technique that serves the same function and get it patented. Secondly, it may be easier to extract the possible gross value relatively easily. It may then be expected that α will decrease. Thirdly, the firm may no longer be impatient with respect to recovering value as soon as possible. For, it has the life of the patent available to it to get the money back. That is, r itself may be lower. These aspects of the effects of IPR protection were noted in Alfranca and Huffman (2003) and Sporleder and Moss (2004). All these forces tend to increase the R&D expenditures that a firm undertakes.

4.4 RISKS OF R&D It is generally observed that the R&D investments in biotechnology are highly risky. Coupled with high investment this risk inhibits firms from undertaking investment. This can be elucidated in the following manner. Suppose the private firm spends E on R&D. Let p be the probability of success. Clearly, the expenditure adds to the stock of R&D only if it is a success. Hence, the expected change in R can be represented by dR/dt = pE – εR Suppose the firm chooses E so as to ∞

Max

∫ e–rt [Ra – E]dt 0

Subject to dR/dt = pE – εR The optimal value of R will then be R = αpV/(r + ε) p < 1 indicates that the cumulative expenditure on R&D will decrease. It is therefore possible to view equity investments by venture capital and capital markets in general as a sort of seed money that enables the firm to assess the probability of success. The private sector may then make the rest of the investment with the probability p of success that it identifies. This role of the equity capital can be incorporated into the above model in the following way. Suppose the capital market invests (1–f)E on R&D that enables the private firm to find out the probability of success of the R&D effort it is trying out. See, for instance,


67

Allen and Gale (1999). The private firm makes an investment with a probability p that it discovers. The expected value of the private investment is therefore pfE. The total expenditure on R&D is therefore [1 – f(1–p)]E. The private firm chooses E so as to ∞

Max

∫ e–rt [Rα – pfE] dt 0

subject to dR/dt = [1 – f(1–p)]E – εR The optimal choice of R is therefore given by R = αV [1 – f(1 – p)] / (r + ε)pf It may now be verified that the expression e = [1 – f(1 – p)] / pf is a monotonic decreasing function in f. Therefore, it can be concluded that private R&D increases with equity spending. The above analysis does not take into account the possibility that • •

too much (1 – f) may discourage the private firm from making any R&D investment there will be a reduction in the proportion of benefits that the private firm can appropriate because, having made some investment, the capital market will not allow the private firm to appropriate all the benefit. However, despite these limitations, this analysis indicates that there is another channel through which equity placements can affect private spending on R&D. One further observation is in order. As noted in an earlier chapter, biotechnology innovations are not always cumulative. In particular, the discoverer of one protein does not have any specific advantage in discovering another. The net value of R&D expenditure may only be V = AEα – E The optimal R&D expenditure will then be E = (Aα)1/(1–α) An increase in α is still conducive to an increase in E. Note that the fraction f of investment may be government spending. The rest of the analysis is still valid. Some disadvantages of such government policy may now be recorded. •

The government may put limits on the type of R&D undertaken as well as control the prices of the resulting products sold on the market. This was noted in Kalaitzandonakes (1999).

•

The policy makers experience information asymmetry with respect to the probability of success of a venture and the expected market value of the innovation. This

68

Economics of Biotechnology may make them reluctant to finance some otherwise worthwhile investments. See, for example, Ferreira and Brooks (2000) and Lerner et al. (2003). •

Recent policies, of the government, allow public institutions, through whom financing is channeled, to claim royalties from the revenue generated. This limits the motivation to increase R. See, Lerner et al. (2003).

In the context of biotechnology the failure may be at later stages like bioprocessing and regulatory approvals. In other words, the firm may spend E and create R but recover Rα with only a probability p. Consider the problem of ∞

Max

∫ e–rt (pRα – E) dt 0

subject to dR/dt = E – εR This results in an optimal R given by R/V = pα/(r + ε) With this in perspective the government may agree to share (1–f)E of expenditure to encourage R&D. Such a public policy results in the choice of R/V = α/f(r + ε) Clearly, public policies of this nature also assist in improving the R&D culture of the firm. Note that the institutional arrangement, like patents and providing monopoly power to the firm, may adequately compensate the firm. This alternative is perhaps superior in practice. Another dimension of risk is specific to agricultural biotechnology. GM products are new to the consumers. There is some resistance especially with respect to genetically modified food. That is, there is a problem with the appropriability of GM products in the eventual product market. In such a case p represents the degree of appropriability. Hence, the government may spend (1–f)E to augment the market demand for the products of the private firm. A case in point, as noted in Just and Heuth (1993), may be the public financing of national health schemes meant to augment the demand for medicines. It is obvious that the above analytical argument will hold even in this context. Further, the availability of and preference for conventional non-GM products persists. Competition on the market for GM products is from such non-GM varieties as well. As noted in the above context, a part of the expenditure E may then be on marketing of biotechnology. However, granting a patent and monopoly rights would be a more efficient alternative. For, it can limit competition from other GM products and imitative generic products. The resulting increase in appropriability mitigates the difficulties associated with the risk of a low value of p.


69

4. 5 COMPLEMENTARY R&D In the early stages of the development of biotechnology, especially due to the high costs of R&D, any one scientist or a private research institution is developing only a part of the knowledge necessary to move towards a final product. In most cases, the knowledge about one of these aspects is necessary to create the next stage of the technology. Knowledge diffusion and progress in R&D slows down if such fragments of knowledge remain a secret. However, many scientists preferred such secrecy and developing complementary R&D on their own. For, at least until the 1980s, patenting knowledge, that does not directly result in utility to a final consumer, was not possible. The effect of this approach and alternatives to it can be analyzed in the following manner. Let Ej (j = 1, 2) represent the expenditures on these two activities and let Rj (j=1,2 ) represent the stocks of R&D built up in these two directions. As before, it can be expected that dR1/dt = E1 – ε1R1 dR2/dt = λE2 – ε2R2 This formulation presumes that λ < 1 is the efficiency with which the first firm can undertake R&D related to the complementary knowledge. Similarly, the expected market value of the innovations will be V = θR1αR2β Note that the private firm decided to pursue R&D regarding R2 while maintaining secrecy with respect to R1. It is therefore vulnerable to competitive pressure. For, someone else may discover R1 in the meantime and patent it. Hence, it may be in a position to retrieve the expected value addition only with some probability θ. It is also possible that the developments will fail at the regulatory stage. This risk can also be included in the specification of θ. The private firm therefore chooses E1 and E2 so as to ∞

Max ∫ e–rt (θR1αR2β – E1 – E2) dt 0

subject to the above two differential equations. Clearly, the optimal choices will be R1/V = θα/(ε1 + r) R2/V = θβλ/(ε + r) One alternative available to the firm, when it recognizes complementarily, is to enter a private network arrangement and choose an optimal contract for sharing gains. An agreement of this nature is feasible if the commercial secrets can be reined in within the partners of the network. Assume that one private firm develops R1, and contracts with a second firm to develop R2. Let s be the share of the accruing value paid to the first firm. The optimal choice of the first firm

70


will be R1/V = sα/(ε + r) Similarly, the choice of the second firm, which is the most efficient in developing R2 and hence λ = 1, becomes R2/V = (1 – s)β/(ε2 + r) This contractual agreement results in a higher level of R1 if s > θ. Similarly, R2 will be higher if s< 1 – θλ. Hence, specialization and networking will be an efficient organizational arrangement whenever θ < s < 1 – θλ. This mechanism is superior whenever θ and/or λ is small. However, there may be a holdup and inefficiency if the share s demanded by the first firm is outside these limits. Private network arrangements may then be replaced by the government research laboratory working on R1 and providing it to the private firm that develops R2. The advantage of this agreement is that even low values of s may be acceptable. The recent experience of most of the countries is that the government is encouraging such contractual agreements. Public policy is then limited to the specification of the limits on s with social welfare in perspective. However, note that government policy, amounting to merely allowing patent rights for fragments of knowledge may be adequate to induce private firms to enter into network arrangements that generate efficient R&D of complementary knowledge. The existence of risk at various levels, as noted in the previous section, necessitates government policies to encourage bearing such risks. The analytical details will be very similar. The above analysis assumed that the transfer of formal knowledge regarding R1 is sufficient to proceed with the development of R2. However, it is well known that transfer of informal knowledge has a critical role in biotechnology development. This will be considered in detail in the sequel.

4.6 AGRICULTURAL EXTENSION SERVICES One of the most important features of the network relationships in biotechnology is the transfer of informal knowledge. In the context of agricultural biotechnology this takes the form of extension services. Of late several private seed companies have been providing such services to the farmers. However, traditionally the public sector institutions were entrusted with this role. From an operational viewpoint it consists of the accumulation, over time, of the expertise in the use of seeds that embody new technologies with the hope that the farmer will eventually have enough knowledge in the use of such technologies. The degree of risk of failure is not as severe in the context of agricultural extension services. The basic issue related to R&D at this stage of bioprocessing is the accumulation of the stock of informal knowledge in the use of a new technology that has already been established through an earlier stage of R&D. Let R be the stock of R&D available for transfer. The farmer generally spends an amount E at time t to use the seeds that embody such technologies. Further,


71

when he gets the knowledge about extension services, he has to pay an additional amount P. The accumulated knowledge S can be expressed as dS/dt = γP – ηS and the value of output generated at time t will be V = EαSβ Hence, the farmer can be expected to choose E and P to maximize the present discounted value of net returns ∞

∫ e–rt (EαSβ – E – P) dt 0

subject to the above differential equation. The optimal choices will therefore be E = αV βV/S = (r + η)/γ Hence, it follows that S/V = βγ/(r + η) Note that a constraint S ≤ S can be imposed on the maximization problem. In that case, P will be equal to S* once it is attained. *

The question that should be addressed is: will government provision of extension services improve the rate of accumulation of S? Consider the choice problem of the farmer. He now maximizes F = EαSβ – E by choosing E such that αEα–1Sβ = 1 Hence, E = αV and substitution yields V = αα/(1–α)Sβ/(1– α) Hence, the problem for the government is to choose P in such a way as to ∞

Max

∫ e–rt [αα/(1–α) Sβ/(1–α) – P] dt 0

subject to dS/dt = γP – ηS The optimal choice satisfies the equation S/V = βγ/(1 – α)(r + η) Clearly, since 0 < α < 1, this quantity is larger than the farmer’s choice when he is paying a private firm for extension services. Hence, public provision of extension services helps accumulate S faster.

72


4.7 BIOPROCESSING One of the distinctive features of biotechnology is the necessity to provide informal knowledge in the process of transferring scientific R&D to a higher stage of development. The new biotechnology firms (NBFs) undertake bioprocessing based on basic patented scientific knowledge obtained from university scientists and other research laboratories. There is no assurance of success at bioprocessing and regulatory stages despite the accumulation of informal knowledge. Assume, to begin with, that the scientist or the private firm expects the bioprocessing firm to pay an amount P to offer such informal knowledge. However, the efficiency of the bioprocessing firm in assimilating it cannot be instantaneous. It is more realistic to expect such a firm to accumulate the stock of requisite knowledge over time and with repeated use of knowledge transferred to it. Let γ represent the efficiency with which the firm can assimilate such informal knowledge and assume that the stock of knowledge decays at a rate η. Then, the net addition to S at time t will be determined by dS/dt = γP – ηS There will be a specific amount E of expenditure on conducting the biotechnology activity itself. However, the activities are subject to the risk of failure. Hence, only a fraction pE of the expenditure contributes to value addition. The other aspect that needs attention is the problem of acquisition of patented knowledge. As the number of such fragments, that need to be acquired, increases the license fees, either upfront or royalty payments, increases. This reduces the actual amount of E available to biprocessing itself. The fraction p captures this aspect as well. The gross value addition may therefore be written as Vg = (pE)αSβ The problem for the firm is then to ∞

∫ e–rt [(pE)αSβ – E – P] dt

Max

0

subject to dS/dt = γP – ηS The solution to this problem is E = αPαV, and S/V = βpαγ/(r+η) where V = EαSβ Clearly, some government intervention, to reduce costs and improve recovery rate, is in order. It may be argued that patented knowledge, at least some fundamental aspects of it that have widespread application, should be made available for non-exclusive license at no cost or for a predefined fixed payment. In fact, the Cohen and Boyer rDNA patent is made available on this basis. Such an adjustment, apart from decreasing costs, allows faster diffusion of knowledge.


73

Suppose, on the other hand, that the government offers to spend (1–f)E so that the firm can explore the probability p of success before investing fE. The solution to the problem will now be E/V = α [1 – (1 – p)f]α/pf, and S/V = βγ [1 – (1 – p)f]α/(r + η) It is readily apparent that such government interventions improve the biprocessing activity and hence knowledge diffusion. For the sake of analytical generality it must be acknowledged that some amount of R&D and discovery of new knowledge is implied even at the bioprocessing stage. In other words, the expenditure E contributes to the development of R&D. The stock of R&D then contributes to the value of the firm. This modification does not change the results in any fundamental way.

4.8 PHYSICAL CAPITAL Consider the latter stages of the biotechnology value chain where the firm is producing output for a final consumer market. The choice of the optimal stock of capital is somewhat analogous to the above. Let K = stock of physical capital at any time t It will be assumed that this deteriorates at some rate δ. Suppose I = gross investment in physical capital at time t Then, the net addition to the stock of capital is given by dK/dt = I – δK The output from the use of this capital can be represented as Y = Kβ ; 0 ≤ β ≤ 1 Let the market price per unit of Y be unity for purposes of analytical simplicity. Similarly, let q = price per unit of I The firm therefore chooses K to maximize the present discounted value of cashflows. That is, ∞

Max ∫ e–rt [Kβ – qI] dt 0

subject to dK/dt = I – δK The optimal value of K is given by K = βY/ q(r + δ) if Y is the output target of the firm. As before the user cost of capital is q(r+δ).

74


One pertinent question is the following. Suppose the firm has a monopoly power in the market for Y. This may be, for instance, a result of patent protection. What will be the effect on the optimal choice of K? To answer this question let the demand curve for Y be written as Y = p–η, where η = elasticity of demand Then, the revenue accruing to the firm is Revenue = Kβ(1–1/η) = Kθ (say) where θ = β(1–1/η) < β because η > 1 in the operationally relevant range. The optimal K then becomes Km = θY1–1/η/ q(r + δ) This quantity is definitely less than the optimal K obtained earlier under the assumption of a competitive market. It can also be shown that this quantity is less than the welfare maximizing value of K. For, recall, from the Spence formula, that Kθ = revenue = (1 – 1/η) welfare Consider the welfare maximizing choice of K. It is such as to ∞

Max

∫ e–rt [ ηKθ/(η–1) – qI] dt 0

subject to dK/dt = I – δK The optimal value of K then satisfies the equation ηKθ–1/(η–1) = q(r + δ)/θ Therefore, it follows that Kw = ηθY1–1/η/(η – 1) q(r + δ) > Km for all relevant values of η. The usual claim made in favor of patent protection is that the firm will use the monopoly profits to increase the stock of capital. However, the above formulation does not justify such a claim. Surely, the relevant monopoly power, if any, is not reflected in the elasticity of demand per se. The conventional result can, however, be rescued if an increase in K allows the firm to appropriate the market by shifting the demand curve to the right. Replace the demand curve by Y = p–ηKφ, where φ = degree of appropriability


75

The revenue will now be Revenue = Kθ + φ/η = Kψ (say), where ψ = θ + φ/η The optimal choice of K is therefore Km = ψY1–1/η+φ/βη/q(r + δ) > βY/q(r + δ) if φ > β. In other words, there will be at least some values of φ such that the monopoly increases its investment in physical capital. The problem of appropriability remains if φ < β. The necessity for capital market intervention to restore welfare maximum is indicated. Suppose the firm accepts a fraction f of qI. Then, the firm will choose Kg = θY/fq(r + δ) This will be equal to Kw if f = (1 – 1/η) Y–1/η = (1 – 1/η)p Clearly, as η increases, the firm can be made to bear a greater share of the cost without reducing welfare substantially. Lower values of η and difficulties in appropriability necessitate greater equity financing of capital investments of the firm. These results may now be generalized by including the firm’s investment in R&D. The only essential modification is in the production function. For, it now depends on both types of capital. Let Y = RαKβ ; 0 ≤ α,β ≤ 1 The problem for the firm is now to choose K and R so as to Max

∞

∫ e–rt [RαKβ – E – qI] dt 0

subject to dR/dt = E – εR dK/dt = I – δK It can be readily verified that the optimal choices are R = αY/(r + ε), and K = βY/(r + δ) That is, the results developed earlier carry over. It should now be obvious that • •

R chosen by a monopoly will be less than the welfare maximizing ideal if the elasticity of demand is low it can be restored partially if an increase in R shifts the demand curve to the right and enables the firm to appropriate the profit potential inherent in the market.

76


capital market intervention, in the form of partially financing R&D expenditures may have the effect of restoring welfare maximum. One further aspect of the demand for biotechnology products should be kept in perspective. In the context of pharmaceutical products, in particular, the demand depends not only on the private individual’s ability to pay for them but also on the nature of the national health schemes and health insurance policies that they have. Similarly, if the physician does not prescribe a specific brand drug the pharmacist may substitute a cheaper chemical-based product. Agricultural biotechnology is also subject to similar changes in demand due to consumer preferences for traditional varieties over GM products. The qualitative nature of the above analysis is still valid though the quantitative choices of R&D expenditures change as a result of these policies.

4.9 FINANCING CONSTRAINTS Early stages of biotechnology development involve large investments with no marketable output for a considerable time. Even at later stages, when the products enter regulatory approval phase the uncertainties are rather large. The scientist discoverers and/or the NBFs do not have their own resources for the necessary investments. Consequently, there are financing problems associated with the • • •

sources of finance cost of capital organizational and motivational problems associated with the method of finance chosen by the entrepreneur Consider the following simple examples to appreciate the role of uncertainty on the cost of capital. Let a venture capital firm provide capital v to a NBF. Suppose the risk free market rate of interest is r. Then, the venture capitalist expects to recover v(1 + r). However, when the amount v is invested in a biotechnology firm the probability of success is p; 0 ≤ p ≤ 1. If he charges an interest rate rv the expected recovery is pv(1+rv). Hence, the likely choice of rv is such that p(1 + rv) = 1 + r That is, if r = 0.05, and p = 0.75, then rv = 0.2. In other words, the interest rate increases to 20 percent as opposed to a normal 5 percent. Private equity placements have a similar effect. The analysis of the earlier sections therefore suggests that both R&D investments as well as those on physical capital will be reduced ceteris paribus. Note that, as Aghion and Tirole (1994) and several others noted, the uncertainty associated with biotechnology projects also affects the market value of the assets of the firm and its revenue generating ability. The positive effect of sourcing finances from outside are • •

limited liability of the NBF increased commitment of the NBF to make the venture succeed


77

•

the venture capitalist and/or the large firm may also contribute its expertise and control capabilities to ensure efficient performance (see, for example, Wolf and Zilberman (1999)). • Gerpacio (2003) pointed out that in the context of agricultural biotechnology external financing from government sources is an essential complement of private research on seeds. For, extension services, in the form of distribution systems, grain harvest, post harvest facilities, and marketing are essential to achieve success. Similarly, drug delivery through national health policies, and insurance schemes define the success of pharmaceutical research. • In addition, association with a large outside firm may signal higher value and ability to attract additional finances as necessary. See, for instance, Lensink et al. (2003). On the downside it must be recognized that •

there may be a reduction in the level of motivation of the NBF and a reduction in the returns from its effort • the government, if it is the source of finance, may put limits on the type of activities undertaken as well as control the prices of the products sold on the market. This was noted in Kalaitzandonakes (1999). • the venture capitalist and the large firms may emphasize marketability rather than social welfare thereby restricting some applications that may have the potential to augment social wellbeing • outside investors experience information asymmetry with respect to the probability of success of a venture and the expected market value of the innovation. This may make them reluctant to finance some otherwise worthwhile investments. See, for example, Ferreira and Brooks (2000) and Lerner et al. (2003) • the outside financing agent will also generally claim property rights and royalties from the revenue generated. Due to the limited bargaining power of the NBF the larger firm or private financing agent may usurp disproportionate control rights. See, Lerner et al. (2003). The negative effect may be somewhat reduced in the case where large business houses provide the finances. In particular, unlike short term financing by outside agents they provide stable long term relationships. See, for example, Lensink et al. (2003). In general, private placements can be arranged quickly. They also allow greater monitoring of the managers of the NBF. However, all such finances may be available only at the margin and after public sector institutions made substantial investments. Lerner et al. (2003) reported that when there is a reduction in government financial support NBFs are more likely to seek alliances with large firms rather than depend on the capital market. One of the reasons may well be that their experiences reduce problems of information asymmetry. The other reason may be the possibility of obtaining assistance in getting regulatory approvals, marketing products, and so on. In sum, there are various factors that affect both the market value of R&D and its costs. As of now it is difficult to model each of these effects in detail. However, the following approach

78


illustrates their effects. Fundamentally, the returns to the NBF from cumulative R&D represented by R can be written as sRα. There are forces that tend to reduce s as well as those that contribute to an increase in s. Similarly, let i represent the generalized cost incurred by the NBF in effectively integrating an expenditure E on R&D. The problem of the optimal choice of E can therefore be represented as ∞

Max

∫ e–rt [sRα – iE] dt 0

subject to dR/dt = E – εR The optimal R can therefore be written as R = αsV/i(r + ε) Clearly, the NBFs seek finances from outside sources to the extent that they can augment the market value of their assets in excess of the costs of obtaining finances and loss of control. However, as of now, it is difficult to precisely specify the extent to which external financing would be beneficial to a NBF.

4.10 FURTHER CONSIDERATIONS It was noted, in chapter 2, that network organizational structures are prominent in the biotechnology industry. When a large chemical or seed firm finances a part of the R&D expenditure of a NBF it will expect to receive a share of the value generated. Intuitively, the share of the large firm will contain a risk premium in addition to the share of finances that it provides. However, it has been difficult to extend the models of section 4.2 to accommodate this aspect. Refer to Section 4.8. The value of the firm, given K and R, depends on both. However, the formulation of the model, of the optimal choices of K and R, is that the user cost of capital of either of these investments depends only on δ or ε. It would be more plausible to expect both δ and ε to affect the cost of capital of R and K. Further modifications of the framework are indicated. Financing constraints, coupled with the risks of biotechnology R&D investments, can be generally expected to influence the expenditure on R&D. However, it has been rather difficult to articulate formal models. Models, dealing with capital market imperfections on the choice of K alone, do not provide adequate guidelines. Recall, from chapter 2, that the principal–agent model provides a useful approach to network relationships. It can therefore be argued that the choice of R and K must be articulated from a similar perspective. However, such dynamic generalizations are not available so far. On the whole, several questions have yet to be addressed though some useful results about R&D investments are available.

Demand, Cost, and Productivity

79

Chapter 5

DEMAND, COST, AND PRODUCTIVITY

5.1 THE BACKGROUND The foregoing chapter considered the environment in which R&D and capital investment activity takes place in the context of biotechnology. Irrespective of the locus of these investments, viz., private firms or public sector institutions, the ultimate purpose of biotechnology is to improve the wellbeing of the consumers and the firms that manufacture the products. To that extent it is necessary to understand something about the demand for such products. Consider the case of pharmaceutical products. In any given therapeutic class, directed towards a specific disease, there are substitutable products. Some of them may be based on biotechnology, others on chemical based technologies, and so on. In a few cases, where the consumer directly pays for them, he may reveal his preferences. However, in most of the situations encountered in practice, it is the physician or the pharmacist who make the choice on the basis of their assessment of the patients’ ability to pay, their experience with the drug, and so on. It is also possible to argue that the ability to pay may be conditioned by the insurance policy of the patient and/or national policies of public health. Demand theory has not been able to come to grips with all these complexities. Instead, most of the literature considers the problem in a conventional manner. The main emphasis is on the effect of substitutable products within a given therapeutic class. The context of the demand for diagnostic kits is much less clear. For, it is essentially the judgement of the physician that determines their use in the case of any one patient. A battery of tests and drugs may go together while determining efficient treatment. Given this scenario there is no significant resistance to the use of biotechnology drugs. Some ethical issues regarding stem cell transplants and so on have been raised but they do not probably have any significant effect on the demand for pharmaceutical products. There are three types of issues with respect to agricultural biotechnology. Firstly, instances can be found where genetically modified foods and non-GM foods are accepted by consumers

80


without any distinction. Soybean usage in the U.S.A. seems to be of this kind. Secondly, in the context of milk products the animals may be treated with rBST (recombinant bovine somatotropin) but the milk itself is free from any genetically modified material. Even so, consumers have been demanding that these products be appropriately labeled and their preference for non-genetically modified varieties honored. The sensitivity is much greater in the context of food crops like soybeans (especially in the European Union (EU)), rice, wheat, and others. Thirdly, there is only a derived demand for genetically modified seeds like Bt cotton. The economic factors, that determine such derived demand, are different from those alluded to in conventional demand theory. In sum, understanding the demand for biotechnology products requires different sets of tools depending on the context. It is not possible to detail all the idiosyncracies. A few major aspects will be examined. The success of biotechnology depends on a variety of other factors as well. In the context of agricultural biotechnology the most important aspects are the • • • •

complementary input use, such as pesticides, fertilizer etc. price of genetically modified seed increase in output achieved by pesticide resistance and herbicide tolerance increases in prices achieved (may be purely due to the monopoly power of the firm rather than superior quality and consequent consumer willingness to pay) In particular, the profit of a firm can be represented by π = pyY – TC where π = total profit py = price per unit of Y Y = total output, and TC = total cost of producing and marketing Y In turn, TC can be written as TC = pS S + pPP where ps = price per unit of S S = quantity of seeds used pp = price per unit of pesticides P = volume of pesticide use Bt varieties of seeds reduce the use of pesticide. However, there will be significant changes in the use of complementary inputs like irrigation, fertilizer, and so on. Cost reduction is not assured unless something is known about the use of other factors of production. It is therefore


81

necessary to investigate the • use of inputs on a given farm • the price of Bt seeds • total cost per hectare of cultivation • changes in output achieved through the use of GM varieties These factors determine total cost. Improvement in the production conditions and/or an increase in productivity may still not result in profitability for the firm adopting biotechnology. It depends on the consumer acceptance of biotechnology products and their willingness to pay and/or the monopoly power of the firm. With these dimensions of the use of biotechnology in perspective this chapter analyzes • demand for the products of the firm • productivity of biotechnology • costs of production • profitability of new products • distribution of gains between consumers and firms However, the effects of biotechnology on the structure of markets and price determination will be the emphasis of the next chapter. For, they present conceptually different issues for economic analysis.

5.2 PATTERNS OF DEMAND As noted above, one of the important links is the evaluation of the consumer’s willingness to pay. This is generally stated in the form of the demand curve. Consider the demand for pharmaceutical products. Assume that the consumer is choosing a non-GM (based on traditional chemical technology) versus a GM product. Both are available on the market and the products are indistinguishable so long as they are not specifically labeled. Under these conditions the total amount of output bought, viz., X = Xn + Xg where Xn = amount of non-GM drugs bought, and Xg = amount of the GM drugs bought is the only variable entering the utility (gross value) function of the consumer. The utility function can be stated as U = aXε + Y ; a > 0 where Y = any other commodity in the consumer budget or simply the expenditure on all other goods (assuming py = 1).

82


The consumer chooses X and Y so as to maximize U subject to the income constraint pxX + Y = I where p = price per unit of X Note that p is the same for both Xn and Xg because they are indistinguishable. The demand for X will then be X = Ap–η where η = 1/(1–ε), and A = (aε)η It is generally noted that even within the category of GM drugs there are substitutable varieties in the same therapeutic class. Added to this is the option of traditional chemical technology based drugs. Though they are substitutable in general, their prices are kept distinct in the market. This is surely one dimension along which competition manifests itself. When Xn and Xg are indistinguishable they can be considered as perfectly substitutable. Hence, the utility function can be represented by U = XgαXnβ It can be readily verified that the demand for GM drugs will now be Xg = α(I – Y)/(α + β)pg so that the elasticity of demand is unity. However, in practice, this is unrealistic. Another way of representing the substitutability, following Fink (2000), is to write X = (wnXnρ + wgXgρ)1/ρ ; wn + wg = 1 where w’s are the distribution parameters that represent the relative importance of the two types of preparations and ρ is a parameter reflecting the extent of substitution. It is well known that the elasticity of substitution between them is σ = 1/ (1 – ρ) That is, Xn and Xg are not perfectly substitutable. The gross value now becomes U = a (wnXnρ + wgXgρ)ε/ρ + Y The consumer maximizes U subject to the budget constraint pnXn + pgXg + Y = I where pn = price per unit of Xn, and pg = price per unit of Xg Explicit optimization results in Xg = E wn–σ (pg/p)-σ


83

where the composite price index p is given by p = [wn–ρpg1–ρ + wg–ρpn1–ρ]–1/ρ, and E = I–Y In some recent literature there is a suggestion that non-GM varieties may be preferred so long as their supply lasts. In other words, GM varieties can be sold only when the non-GM supplies are exhausted. This subtlety can be incorporated in the above framework by placing suitable limits on the value of the parameter a. However, pharmaceutical products are generally not labeled and the consumer (or his physician who writes the prescription) does not discriminate between them except for the utility they offer and the purchasing power of the patient. Imperfect substitutability will primarily be a result of cumulative experiences with a specific drug. However, their prices may remain distinct thereby suggesting that the above analysis is more meaningful in practice. The above formulation is generally valid even in the context of agricultural biotechnology so long as the GM and non-GM varieties coexist. However, the analytical requirements are somewhat different if the products are labeled. One of the major modifications, as suggested by Giannakis and Fulton (2000) and Fulton and Giannakis (2004), concerns the fact that the firms do not apriori know the relative preferences of a consumer for non-GM versus GM varieties. Further, the specification must be consistent with the stylized fact; viz., that consumers prefer non-GM products when both are offered at the same price. Hence, the following assumptions are pertinent. • •

Each consumer purchases 1 unit of the commodity in whatever form they choose. c represents the distinguishing characteristic of a consumer relating to the choice between non-GM and GM varieties. • αc represents the consumer’s aversion to GM products. That is, consumers with large c are more likely to prefer the non-GM products. • Assume that c varies randomly over (0, C) and postulate a uniform distribution. A typical consumer has three options on which he can spend his income; the non-GM variety, the GM variety, or some other commodity (viz., the Y in the above model). To simplify comparisons measure the quantities of these purchases in such a way that they yield the same gross value U per unit of consumption. The net values are then represented by Un = U – pn for the non-GM variety Ug = U – pg – αc for the GM variety, and Us = U –1 for the substitutable variety The price per unit of the substitutable product is parameterized as unity as in the previous modeling exercise. A consumer is indifferent between a GM and non-GM variety provided Ug = Un. That is, U – pn = U – pg – αc

−

84


Hence, the corresponding value of c is c = (pn – pg)/α All the consumers, with a value of c less than or equal to this, will prefer the GM product. Hence, Xg = quantity of the GM product demanded = (pn – pg)/α Note that in this formulation pg ≤ pn invariably. The empirical validity of this implication can be contested if the unit cost of producing the GM variety is higher. Consider the possibility that the consumer prefers the substitutable product over the GM variety. This will occur whenever U – 1 ≥ U – pg – αc or, equivalently, c ≥ (1 – pg)/α Assume, for purposes of argument, that pn > 1. Then, it is obvious that Y = (1 – pg)/ α and, consequently, X n = C – Xg – Y The above argument assumed that the non-GM and GM foods are labeled and sold at different prices. Consider the situation where there is no such labeling. Then, pg = pn since they are not distinguishable. Let θ represent the probability of an actual purchase being a GM variety. Assuming that the consumer can recognize it ex post (i.e., when he consumes it) the expected utility will be Ue = θ(U – p – αc) + (1 – θ) (U – p) = U – p – θαc X will be preferred over Y if and only if U – p – θαc > U – 1 That is, 1 – p > θαc Hence, the demand for X becomes X = (1 – p)/θα The above formulation assumed that the consumer derives the same gross value from the consumption of a unit of GM and non-GM products. Differences in valuation can be built into the framework in the following manner. Rewrite the net value as Ug = θg – pg U n = θ n – pn with the possibility that θg > θn. The consumer will then buy GM products if and only if


85

θg > pg + (θn – pn) The demand for GM products will then be Xg = C – (pg + θn – pn) It may be expected that pg > pn though θg/pg > θn/pn. To an extent this scaling was implied in the earlier formulation. Neither of these approaches contains any reasons for the preference of non-GM over GM products. They only allow a derivation of the demand curves if such a distinction exists. One further observation is in order. Most biotechnology based products are produced in networked organizations. Similarly, most of these products are supported by strong patent rights. In either case, the firms derive a degree of monopoly power that has not been taken into account in obtaining the demand curves so far. There is no definite analytical argument to fall back upon. The following approach is indicative of the general direction. Assume that there are Xg number of consumers of GM products with each one purchasing one unit of output. As Xg increases the marginal valuation of the commodity decreases. Let pg = α – aXg where α = maximum price consumers are willing to pay, and Xg = number of units of output bought It can be expected that only Xg/ψ individuals will pay if property rights are not strict. The rest will use pirated versions. That is, when property rights are not perfect and costless to enforce the patent holder may not be in a position to exclude all pirating or he may find it costly to do so. Such weakly enforced property rights would result in payments from a fraction ψ of customers only. The marginal valuation will therefore become pg = α – aXg/ψ This is one approach suggested in the literature. A second version suggests that pirating will occur only when pg is high. Consequently, in the presence of weak property rights, there is a fraction φ of value such that any price above that leads to piracy. There is a limit to monopoly exploitation of IPRs. Hence, the effective demand curve can be represented by pg = φ(α – aXg) ; φ < 1 It may be concluded that some issues related to the demand for GM versus non-GM foods have not been resolved.

5.3 PRODUCTIVITY When products of biotechnology and conventionally produced output coexist in the market the market penetration of biotechnology products depends on advantages either in terms of the quality of products, lower prices, and/or advantages in the cost of production. Monopoly prices of biotechnology products, that are a result of patents and IPR protection, are unlikely

86


to be lower. In the case of corn the higher oil content has been value enhancing. In general, however, there is no overwhelming quality improvement in the use of products of biotechnology. This makes the study of productivity and costs and their contribution to profitability important. The simplest analytical argument would be that the productivity of biotechnology seeds (Bt cotton or roundup ready soybeans) and other such inputs like rBST save the farmers from utilization of certain resources and thereby increase productivity. In other words, partial productivity measures like Y/S = output/quantity of seed used = average productivity of seeds, and ∂Y/∂S = marginal productivity of seeds convey adequate information about productivity improvements. One difficulty with this approach is that the prices of biotechnology seeds and inputs like rBST are more expensive relative to the conventional factors of production. That is, the advantages due to increases in partial productivity measures may be more than neutralized by such price increases. It is necessary to conceptualize a productivity measure that accounts for this. Suppose pn is the price per unit of non-GM seed whereas pg is the price of GM seeds. Then, pgSg/pn can be viewed as the equivalent non-GM seed use when Sg units of GM seeds are utilized. The partial productivity measure may be modified as pnY/pgSg = average productivity of conventional seed equivalent. Clearly, whenever pg > pn, the expected productivity increases will have to be discounted. A more serious difficulty with partial productivity measures is that they do not adequately account for cost reductions of substitutable inputs. For instance, by their nature Bt and RR varieties of seeds economize on the use of pesticides or herbicides as the case may be. However, some inputs, such as fertilizers and irrigation, may be complementary in that GM varieties need more of these resources to achieve the expected output increases. In essence, S, in itself, cannot explain all the output increase achieved by using GM seeds. Other factors contribute to the output increases significantly. Basically, therefore, a total productivity measure is necessary to appreciate the advantages derived from biotechnology. To keep the presentation simple assume that the production process utilizes only two inputs, seeds (S) and pesticides (P). Let ps and pp denote the respective prices per unit. The pesticides used in both GM and non-GM production are the same. Hence, pp does not change whatever type of crop is cultivated. However, ps may be different. It is therefore convenient to convert all types of inputs into units of P. In particular, the use of seeds is equivalent to psS/pp equivalent units of pesticide use. Hence, the total input use for either technology can be written as TFU = total factor use = psS/pp + P


87

The total factor productivity can now be defined by TFP = Y/TFU = PpY/(psS + ppP) It is important to acknowledge one other possibility. When GM technology displaces nonGM technology it is possible that the number of inputs will also change. For instance, a certain periphery of the farm may be mandated to use non-GM varieties for ecological and environmental reasons. Similarly, the nature of harvesting GM crops change as the Bt varieties yield larger outputs. The approach to total productivity suggested here can accommodate such changes as well. One implication of this approach is important. Suppose the total productivity of GM varieties is higher. This may be a result of •

a reduction in ps/pp

•

a reduction in S

•

a reduction in P

•

an increase in Y per hectare

•

a reduction in pp obtained in the short run due to the reduction in the demand for P

The following numerical example may be helpful in appreciating the issues at stake in greater detail. Consider the production of non-GM varieties first. Assume that the production process requires seeds (S), pesticides (P), fertilizer (F), and labor (L). Let one unit of output (Y) be produced and sold at a price 1 per unit. Assume that the cost of seeds

=

0.2

cost of pesticides

=

0.15

cost of fertilizer

=

0.15

cost of labor

=

0.3

Then, the productivity =

1/0.8 = 1.25

Contrast this with the production of a GM variety. Let the prices of these seeds be 1.3 (or, 30% more than non-GM). The cost of seeds

=

0.26

Let the other costs change as follows. cost of pesticide

=

0.075 (50% reduction in cost for Bt varieties)

cost of fertilizer

=

0.18 (20% increase)

cost of labor

=

0.36 (20% increase because output increases)

Y = output from the same quantity of seed = 1.2 Suppose the price per unit of Y is the same. Then the revenue is 1.2 and the cost is 0.875. Hence, the productivity is 1.43. GM varieties are then superior.

88


However, suppose for the sake of argument that seed costs double. Then the productivity is 1.2/1.075; i.e., it decreases. Assume, instead, that the costs of fertilizer and irrigation increase by 50%. The productivity decreases. In other words, the entire choice of inputs changes when moving from a non-GM to a GM variety. An appropriate mix of input may be necessary to maintain an increase in output. This may not, however, result in an increase in productivity. These aspects have been examined empirically in the context of several biotechnologies. Some representative examples will now be considered. (a)

(b)

(c) •

•

•

Consider the case of rBST. This is a genetically engineered hormone injected into cows to increase milk yield. The technology is simple, does not require heavy fixed investments, and has minimal start-up costs. However, it was reported that complementary inputs and certain management practices are necessary to obtain a higher yield of milk. In particular, the farm should invest in total mixed ration (TMR) that improves the quality of feed mix supplied to cows. This investment is large and can be justified only when the herd size is sufficiently large. It was reported that the increase in the costs of complementary inputs generally reduces total factor productivity. In other words, the increases in milk yields, if any, have not been commensurate with associated costs. The increases in the quality of milk in the form of increased shelf life have been inconsequential. It was also pointed out that increased yields may occur only with younger herds. Biotechnology cannot be a miracle to increase productivity if the cows are already barren or their productivity is low. These results were recorded in Foltz and Chang (2002) and Barham et al. (2004). The use of porcene somatotropin (PST) appears to have a different impact. PST is produced in the pituitary of pigs. It is a naturally produced protein. Supplemental PST produced through genetic modification was reported to have affected feed efficiency, average daily weight gain, and production of leaner meat. TFP increased in general. See, for instance, Lemieux and Wohlgenant (1989). Bt cotton is widely cultivated and its productivity extensively studied. The following observations are salient. A major problem with growing cotton is the crop damage due to insects and pests; especially the bollworms. The damages caused by bollworms are more important in the early phases of the crop cycle and hence have a significant impact on plant development and crop yield. See, for instance, Klotz-Ingram et al. (1999), Qaim (2003), and Bennet et al. (2004). Cotton production requires herbicides to control weeds. Two or more herbicides are necessary at planting stage. Post-emergent (i.e., after the weeds are noticed) herbicides will be needed at later stages in the production cycle. This was noted in Klotz-Ingram et al. (1999). Bt varieties of cotton contain in-built resistance to pests. It was reported in Qaim (2003) and Bennet et al. (2004) that the use of insecticide can be reduced to about


89

one third of the conventional varieties. That is, even Bt adopters had to use some bollworm insecticide. For, as Bennet et al. (2004) remarked, the resistance to pests that is characteristic of the Bt varieties, diminished with the age of the plant. The advantage of Bt varieties is not only in the quantitative reduction in pesticide use but also in the number of sprayings thereby reducing other variable costs as well. See Qaim (2003) and Traxler (2004). Koltz-Ingram et al. (1999) pointed out that Bt cotton growers may discontinue the use of BT foliar sprays and pyrethoids. • There is no difference in the use of pesticides against sucking pests. • Morse et al. (2005) reported that Bt varieties save on irrigation costs while they use a great deal more of inorganic chemical fertilizers. • Costs of harvesting Bt varieties are reported to be higher since the yield is higher. Refer to Qaim (2003). • Bt seeds costs are higher than traditional seed. Farmers also generally pay technology fees to cover the fixed costs of R&D. See, for example, Koltz-Ingram (1999) and Qaim (2003). • The increases in the yield per hectare are critically dependent on the soil, nutrient application, pest pressure and a variety of other factors. This was noted in Huang et al. (2003). When all these factors are taken into account it was generally concluded that there is no significant gain in the total factor productivity as a result of the use of biotechnology. Fulton and Keyowski (1999) reported similar results for herbicide tolerant canola.

5.4 VARIABLE COSTS The second building block to determine the welfare gains from biotechnology is the cost of production and supply behavior of firms. It has been acknowledged that the fixed costs of conducting the research, the share of this cost that the private firms need to absorb, and the intellectual property rights accorded to private investors have a decisive role. The definition of TFP utilized in the previous section suggests that the average cost of production can be written as AC = pp/TFP In the short run, to the extent that the adoption of Bt varieties reduces the use of pesticides, pp will be reduced. Hence, the reduction in AC will depend on the reduction in pp if TFP is declining simultaneously. Both Qaim (2003) and Morse et al. (2005) concluded that Bt varieties do not generally reduce the average cost primarily due to high costs of Bt seeds. However, this section will only consider the variable costs of delivering output from technologies whose operational feasibility is already established. There is a universal acknowledgement that the yield on farm lands, for the application of the same seed variety and pesticide/herbicide combination recommended by the seed company, tends to be very different. This depends on the agro climatic conditions, the initial

90


incidence of insects and weeds and so on. Analysis of variable costs should keep this in perspective. It is generally suggested that the GM varieties of crops increase the productivity of farm lands. For example, the crop season for Bt cotton can be extended thereby increasing the yield. Similar is the case of yield increases if pest elimination and herbicide tolerance can be achieved. However, note that pest control inputs differ fundamentally from other inputs. For, though potential output from given seeds cannot be altered, increases in yield are a result of pest control or herbicide tolerance. That is, only the fraction of potential output recovered depends on pesticide or herbicide usage. Refer to Lichtenberg and Zilberman (1986). This is reflected by writing the production function as Y = actual crop yield = f (X) G(Z) where X = vector of inputs like seeds and fertilizer, and Z = quantities of pesticide or herbicide In practice, the upper limit on G(Z) is unity. Hence, the logistic specification G(Z) = [1 + exp(a – bI – cBt)]–1 where I = amount of insecticide sprayed on the farm, and Bt = amount of insect resistance expected from the Bt variety of seeds was considered satisfactory. See, for example, Qaim (2003). The following alternative is sometimes conceptualized. Notice that the seed company, that sells the GM seeds, specifies fixed quantities of seed and pesticide use per hectare. This may indicate a possible reduction in the average cost of producing the crop if the use of GM seeds reduces the use of pesticides while increasing the price of seed. However, depending on the nature of the farm, the benefits may vary. Technically there may be perfect substitution between the pest control offered by the Bt seeds and the external use of pesticides. However, the efficiency with which they eliminate pests may vary. This is usually reflected by a CobbDouglas production function Y = XαZβ ; 0 ≤ α, β ≤ 1 The values of α and β in this specification are the efficiency parameters. However, note that the substitution between X and Z is not a result of the variations in the prices of X and Z. Hence, the derivation of a conventional cost function from this will be inadequate. The following approach will be more direct. The cost of producing Y is C(Y) = pgX + qZ


91

where pg = price per unit of GM seeds, and q = price per unit of the insecticide The average cost of production is AC = C(Y)/Y In general, it was noted that C(Y) associated with GM varieties exceeds those of non-GM crops. Hence, even if GM crops decrease average costs for the agro climatic conditions for which they were designed, they may not be effective if the incidence of insects is higher. The third aspect that received attention is the possibility that the output of the GM varieties is qualitatively superior. For example, genetically modified corn contains more oil. The production and cost function specifications can address this issue by an appropriate choice of output measure. In the context of the above example the amount of oil extracted is a better measure compared to the quantity of corn produced. The above approaches estimate the average cost of GM and non-GM crops separately. However, it can be expected that the actual choices of the farmer depend on profit maximization. It would be useful to investigate the comparison of average costs in such a framework. The following approach, adapted from Lamarie and Marette (2002), and Zilberman et al. (2004) is illustrative. Assume that the nature of the farm can be characterized by its pest infestation or the incidence of weeds. Suppose that, as a result of this, the farmer can get only a fraction θ of the potential output from the farm. Let θ be a random variable distributed uniformly over (0,1). Consider a non-GM variety of crop being planted on the farm. Denote by xn the quantity of seed and pesticide combination required per hectare. Suppose the potential output from the farm is yn. Then, the profit from the farm is πn = pθyn – qnxn where p = market price of a unit of yn , and qn = price per unit of xn A farmer will use these seeds if and only if πn ≥ 0. That is, θn ≥ qnxn/pyn The average cost of production on this farm will be ACn = qnxn/θyn for θ ≤ θn Hence, the maximum average cost for the non-GM crop is ACnm = p when θ = θn Now, suppose that Bt varieties or Roundup Ready varieties of seeds are available. When these crops are planted the profit becomes πg = pθyg –qgxg

92


The assumption here is that the GM and non-GM crops are indistinguishable and there is no labeling. This formulation captures the benefits of GM varieties through yg only. A suitable modification is needed if θ decreases over the long run. Clearly, a farmer, considering the GM variety in isolation, will adopt it if and only if θ ≤ qgxg/pyg The minimum AC is again ACgm = p at θ = θg Consider the possibility that θg < θn.. Clearly, this happens if the costs of using GM varieties are lower and/or the yields are higher. GM varieties will be always preferred over non-GM varieties if this occurs. However, the case where θn < θg is more appropriate in empirically observed contexts. For, as of now, both non-GM and GM varieties coexist in practice. Clearly, ACn < ACg at θ = θg. Under these conditions the farmer will prefer the GM varieties over the non-GM crops whenever pθyg – qgxg > pθyn –qnxn That is, pθδyg > δcg where δyg = yg – yn, and δcg = qgxg – qnxn Hence, the relevant range of θ is θg1 ≥ δcg/pδyg The only relevant case for analysis is where θg1 > θg. For, otherwise the farmer will not use the GM varieties. A comparison of ACg1 with ACg now suggests that ACg > ACn at θ = θg implies that ACg1 > ACn for all θg1 ≥ θg. It is always more expensive to use the GM seeds. However, in practice, several studies, such as Klotz-Ingram et al. (1999) and Qaim (2003) suggest that ACg < ACn. Two possibilities have been recorded. First, the costs may be reduced. For, production of non-herbicide tolerant crops typically requires two chemical applications: one pre-emergent and the other post-emergent. The post-emergent application controls only a limited spectrum of weeds. In the context of herbicide tolerant crops the chemical is applied only once. This improves the yield by removing competition with herbs for moisture and nutrients. It also eliminates the cost of additional machine operations over the field. Second, though the cost per hectare (seed + pesticide) increases, the yield increases substantially so as to make AC lower. This may however be observed only on farms where the initial incidence of pests is low. Huygen et al. (2004) observed that average costs are higher if labeling costs are included. In sum, it is not possible to establish the average cost reduction achieved by planting GM seeds either theoretically or empirically. It may occur only in some cases where the initial pest incidence is low.


93

The adoption of Bt varieties depend on profitability even when the average costs are rising. For, π = [pyY/TC –1]TC where π = total profit py = price per unit of Y TC = total cost of producing and marketing Y The profitability for the firm adopting biotechnology also depends on consumer acceptance of such products and their willingness to pay and/or monopoly power of the firm. Both in the context of rBST milk and Bt cotton it was generally reported that the profitability and profits did not increase. See, Barham et al. (2003), Huang et al. (2003), Bennet et al. (2004), and Caswell et al. (1998).

5.5 WELFARE EFFECTS Recall, from section 5.2, that the demand curves for GM and non-GM products can be expressed in the form Xg = (pn – pg)/α, and Xn = C – (pn – pg)/α Explicitly plotting the net utility from the two varieties, as in Fig. 1, it can be inferred that U A

Un

B

Un Ug

O

C* Fig. 1

c

the GM varieties are preferable from the consumer viewpoint so long as c < c* = (pn – pg)/α. Hence, it follows that the increase in consumer surplus as a result of the introduction of the GM crops is the area ABUn. Algebraically it is given by CS = (pn – pg)2/ 2α In the above analysis it was assumed that the two products have been labeled explicitly and sold at different prices. To consider the case where they are indistinguishable recall from

94


section 5.2 that the demand for X = Xn + Xg becomes X = (1 – p)/θα where θ is the fraction of the GM crop that is sold on the market. The net consumer benefit will then be CS = (1 – p)2/2θα Labeling will improve consumer welfare if and only if (pn – pg)2θ > (1 – p)2 That is, labeling will be beneficial whenever the products can be distinguished sharply and priced accordingly and/or there is a greater fraction of GM crop on the market. Giannakis and Fulton (2002) presented such a comparison of welfare effects. Fulton and Keyowski (1999) made an attempt to capture the benefits to the producer in the following manner. Suppose the firms differ with respect to some characteristic like the incidence of pests or the abilities of the farmer. Assume that this characteristic f is uniformly distributed over (0,F). Represent the profit from the GM and non-GM crops by πg = pgyg – qgxg + γf, and πn = pnyn – qnxn + βf The notations here are the same as in section 5.3. Since the GM seeds better resist pests it can be expected that γ > β. Notice that GM crops will be utilized on the farm whenever f ≥ ∆/(γ – β) where ∆ = (pnyn – pgyg) – (qnxn – qgxg) It can be therefore inferred that Xn = supply of non-GM output = ∆/(γ – β), and Xg = F – ∆/(γ – β) Clearly, both are increasing functions of the respective prices. The net increase in profit is given by the area CBB* in Fig. 2.

Fig. 2

96


Xdn = demand for non-GM products = C – (pn – pg)/α, and = supply of non-GM varieties = ∆/ (γ – β) Hence, the market determines pn by equilibrating supply and demand. The net welfare is the area between the demand and supply curves upto the point of equilibrium. A similar analysis holds for GM crops. Extensions, to determine qg and qn, , can also be conceptualized. In practice, the GM crops are produced under patent protection. Hence, the seed producers and/or the farmers will determine qg and pg. Such extensions of the analysis are also straight forward. Market segmentation, achieved through labeling, may enable the producers of GM varieties to extract some consumer surplus thereby making their production more profitable. Even these analytical details of Lence and Hayes (2003), Anderson et al. (2004), and Noussair et al. (2004) can be developed in an analogous manner.

5.6 SUMMING UP The present chapter highlighted some essential features of biotechnology. It also made an attempt to examine how economics can explain the emergence of these features and the consequences for pricing, investment, and economic welfare. The empirically observed patterns are rather diverse. They vary with the particular biotechnology under consideration as well as the nature of the emerging market structure for specific products. As such it is difficult to visualize one all encompassing analytical framework. The variety of conflicting inferences from different models may be desirable but more attention will be necessary to examine the sources of these differences. Frisvold et al. (2003) noted that R&D investments create dynamic gains and not just one time advantages. Hence, all the static productivity, profit analyses reported above are inadequate. Lence and Hays (2003) observed that it is possible that initially productivity and profits are nil or negative. However, dynamically the market picks up with increased acceptance of GM foods. Hence, a static analysis is inadequate. Moon and Balasubramanian (2001) pointed that the core of the controversy over biotechnology foods is the extent to which consumers perceive benefits from agricultural biotechnology relative to its risks. The role of benefit perceptions in shaping willingness-topay premium for non-GM was evaluated. Overall, risk perception exerted a stronger influence on the willingness-to-pay for non-GM foods than did benefit perceptions. In general, as Caswell et al. (1998) noted, the economic impact of biotechnology is likely to be incremental and not dramatic as claimed by its proponents.


97

Chapter 6

MARKET STRUCTURE AND PRICING

6.1 NATURE OF MARKETS Theories of industrial organization generally acknowledge markets for products that are directly sold to the consumers. For, the ultimate value of production depends on their satisfaction. What the firm can recover out of the value to the consumer determines cost recovery and profits to the firm. In the context of biotechnology the two major markets for their products are pharmaceutical and agricultural products (mainly crops but they may also include animal products). As yet, there are a few such final products in the biotechnology industry. It may, as a result, be important to consider markets for some intermediate products as well. Gamberdella et al. (2000) consider four sets of variables as relevant sources of competitiveness. •

Size and structure of biomedical education and research

•

Basic institutions governing labor markets, skilled researchers, and managers

•

Intellectual property rights and patent laws

•

Nature and intensity of competition in the final product market

It may turn out that within the pharmaceutical group the market for diagnostic kits will be different from that of prescription drugs. This happens due to two reasons. First, the technical complexity and the costs of production are far less in the context of diagnostics. Consequently, entry and exit are easier and a large number of firms may be expected in this market. Second, patients do not directly buy the diagnostic kits. Instead, it is the physicians who determine what need to be bought and used. Though they are expected to take the income of the patient into account in their choices it need not always be the case. The markets will surely change as the health insurance coverage increases. Even in a specific therapeutic

98


class market demand depends on the combination of drugs that a physician prescribes and whether or not he chooses the combination produced by the same firm. In the context of biotechnology drugs it was noted that a firm may not have any particular advantage in producing a second drug merely because it was successful in the first. That is, quite independent of the physician’s propensities, the market structure depends on the production and marketing competencies of the firm and its costs. The market for each drug may have its own dynamics if the economies of scope are not significant. Two other aspects have been highlighted. First, within each therapeutic class there may be competing products based on biotechnology. The economies of scope and the nature of patent rights may determine the number of products of each firm and the number of firms in the market. Second, many drugs are such that substitutable products are currently produced through the use of chemical technology. Products of biotechnology may not be distinct and exclusive. There is also no assurance that such drugs will be cheaper. At least in the early stages of drug development, where the scaling up of laboratory technology is still in progress, the chemical firms may have an advantage. The market for drugs of a given therapeutic class may therefore consist of both chemical and biotechnology firms. Any one firm may find it advantageous to segment the market for its products based on the mechanism of drug delivery. For instance, they may consider the bulk market through the public health system as distinct from private prescription sales. Large chemical firms are finding it increasingly advantageous for them to integrate into the biotechnology markets. Initially this may be the result of their competence in clinical tests and their ability to take a product through the regulatory process. The mergers and acquisitions may also eventually give them the scientific and technical competence that they may find expensive to develop in-house. Defining the nature of the market becomes complex if a network of firms, including NBFs and large chemical firms, produce a drug. For, even if the chemical firm has competitors for its product it has distinct dynamic advantages if it is highly connected to NBFs. However, the relationships between them are not arms length contracts. Instead, they are incomplete contracts of a long duration. In general, biotechnology firms in the pharmaceutical industry exhibit various degrees of vertical integration and contractual relationships. It is difficult to apriori claim that all links of the vertical chain belong to the same industry. Fundamentally, therefore, the degree of concentration in any therapeutic segment depends on a variety of factors. Gamberdilla et al. (2000) claim that concentration in the biotechnology industry is low because • •

the industry is composed of many therapeutic classes and a wide range of technologies the successful introduction of a new drug within a class and its advantages do not last long. A major innovation is followed by product and process innovations by competitors


99

•

an early innovator does not have any major advantage in introducing major drugs later Consider the markets for agricultural biotechnology. There are three distinct segments here: seeds, crops, and animal products. Biopesticides and biofertilizers may also be sold independently and not necessarily embodied in seeds. Consider the market for seeds. The Bt varieties of cotton and corn or the Roundup ready soybeans are distinct classes by themselves. However, some competition is possible because some varieties have been developed to suit specific agricultural climatic zones. Even so it was observed that only a few firms dominate the market. For all practical purposes, the fine division achieved in conventional hybrid technologies will not be possible in biotechnology due to the high fixed costs of adaptations. The more important problem has been the tying of sales of seed with other fertilizers etc. produced by the same firm. This effectively reduces the number of competing firms in the market. Competition is from the producers of conventional hybrid varieties though their technology is different. The difficult question to address is whether the demand side relationships and/or the technical relationships should form the basis for defining the industry boundaries. Given the preferences of the consumers, say, toward non-GM varieties of crops so long as they are available it would appear that the markets for GM and non-GM crops are distinct. However, the costs of labeling non-GM foods increases their cost and make GM crops more competitive at the margin. Taking GM and non-GM firms together to define an industry will perhaps be more reliable. Some authors seem to feel that some biotechnology firms, especially NBFs, may be dealing only with intermediate level technologies. They may sell their products, contract with seed companies, or form joint ventures. The definition of markets and products to be included in the definition are subjective. Roijakkers et al. (2005) argued that biotechnology industry is characterized by a dual market structure. On the one hand a group of large, integrated, international and established companies and on the other hand, a group of relatively small, specialized firms. In the usual industrial organization literature there will be competition within the two groups but nothing between them. In biotechnology industry the competition among large firms to force linkages with smaller firms is significant. Since learning effects are not very strong there may be no longer term alliances. Instead, there will be many short term contracts. The boundaries for defining the level at which a market should be conceptualized is, as yet, a pragmatic choice depending on the purpose of analysis. In general, either technological and/or demand relationships determine the final choice.

6.2 DEFINING MARKET CONCENTRATION Historically, the interest in market structure arises from the fact that the monopoly power of firms determines prices of products and welfare losses. The elasticity of demand, number of firms, four firm concentration ratios (based on market share of sales), and the Herfindhal index have been utilized.

100

Economics of Biotechnology It is well known that the Lerner measure of monopoly power L = (p – MC)/ p = 1/η

depends on the elasticity of demand. The market structure can be characterized as a monopoly if η is low. This measure enables the analyst to consider various degrees of monopoly power. Intuitively, the larger the number of firms the lower the concentration or the higher the degree of competition. Hence, the number of firms (or numbers equivalents as defined below) has been utilized to characterize the market structure. Either due to economies of scale, the degree of diversification, or the economies of being established in the market, some firms tend to be big even when a large number of small firms exist in the market. This is typical of the biotechnology industry. For, as observed earlier, a fairly large number of small NBFs begin to explore the potential to commercially exploit a scientific discovery. The successful firms, among these, will either merge with or are taken over by the large seed and chemical companies eventually. Studies by Ricabboni and Pamolli (2002) consider these issues fairly exhaustively. The purport of this argument is that a few large firms may have all the effective market power even when the market has a large number of firms. The market share of the four largest firms is usually called the four-firm concentration ratio (CR4). Assume that the market consists of n identical firms. The market share of each of them will be 1/n and CR4 = 4/n. Consequently, n = 4/CR4 can be considered as the numbers equivalent in any general setting. The Herfindhal index of concentration utilizes the market shares more explicitly. Let si; i = 1, 2,…, n denote the market share of the ith firm. Then, the Herfindhal index of concentration is defined as H = Σsi2 Observe that if s1 = 1 and si = 0 for i ≠ 1 this measure reduces to H = 1. On the other hand, if si = 1/n for all i, H = 1/n. In general, 1/H is also a numbers equivalent concentration measure. Pharmaceutical firms may have a market advantage if they are highly diversified. This may arise in one of two ways. First, by producing a variety of products they may experience economies of scope in demand (this will be considered in detail in the next section). In effect, they may be able to save on costs of promotion and may reach physicians more effectively if they have a diversified product range. Second, the advantages may be in the costs of production if there are economies of scope. That is, by spreading the fixed costs over a wider range of outputs the firm may be in a position to lower the prices of their products. A Herfindhal index of diversification will be useful in the first context. To take the production economies into account define s i = share of the ith product in the variable cost of the firm. Then, a similar Herfindhal index would indicate that a high value of H cannot provide much of an advantage to the firm. In the context of biotechnology many NBFs may not be producing any final product. They may still have some monopoly power while dealing with large firms. Conversely, a


101

large firm that has extensive technological and/or financial links with NBFs will have advantages in dealing with them as well as in the markets for final outputs. Roijakkers et al. (2005) characterize these as dual market structures. Basically the monopoly power of the large firm and the small firms is in their respective specialization and core competence. The number of links and the concentration of links may then characterize the market structure. The strength of the links may be measured by the number of patents or the finances offered as the case may be. Since the number of large firms in downstream markets is lower this type of concentration tends to increase while moving toward final product links. It is quite clear that large chemical firms have advantages with respect to technical expertise, regulatory links, and finances (deep pockets and ability to take risks). These features may enable them to undertake a larger number of clinical trials of drugs for eventual regulatory approval. An ex ante measure of the concentration of clinical trials may then be a good measure of ex post market power. Field trials have the same function in the context of agricultural biotechnology. A large seed firm that is currently undertaking a large number of field trials for a variety of crops has the prospect of achieving greater market advantage by offering a more diversified product range. Though a number of other finer points can be included in the construction of concentration indices this analysis captures the essential aspects of the degree of concentration in biotechnology markets.

6.3 SOURCES OF CONCENTRATION The sale of biotechnology gene components and cells are highly concentrated only because they are still under patent protection. However, more genes, owned by more NBFs, are entering the market thereby reducing concentration to some extent. With more licensing, compulsory or otherwise, the concentration in the downstream industries will reduce further. See, for instance, Schimmelpfenig (2004). In very general terms concentration may be a result of economies of scale in the production process. Biotechnology research activities have been generally rather expensive. Hence, the large seed companies are basically a result of the desire to spread the costs of biotechnology over a wide range of output and the market. A natural monopoly like situation appears to operate. In the pharmaceutical market such advantages seem to be for products in specific therapeutic categories and for a limited time. The large fixed costs are also the reason for the emergence of scope economies. These may be related to production technology or to market demand. In general, as Malerba and Orsenigo (2002) pointed out, biotechnology innovations are not cumulative in so far as the development of one gene or protein does not create any advantages in developing another. Even so firms tend to diversify into a variety of gene and product discoveries in the hope that they can spread fixed costs over a diversified range of products. The technological economies of scope that a firm can hope to achieve are rather severely limited.

102


In the agricultural biotechnology context small companies have been able to operate on lower costs and develop niche markets for the product traits that they discover. However, the large chemical companies increasingly find that biotechnology based seed production generally complements the use of their chemical fertilizers, pesticides, and so on. For example, the herbicide tolerant roundup ready soybean production requires more of the compatible herbicide. Thus, the production of both may impart certain economies of scale and scope. In the context of insect resistant Bt varieties of corn, cotton, etc. the chemical company experiences a reduction in the use of chemical insecticides. The production of Bt varieties of seeds, however, spread the fixed costs giving rise to some economies of scope. In general, as Just and Heuth (1993) and Malerba and Orsenigo (2002) remarked, technology related economies of scope do not appear to be the major reason for product diversification of the large chemical companies. A more plausible explanation is the economies of scope in demand. As Just and Heuth (1993) argued, scope economies in demand can be said to materialize if the firm can generate greater net profits by marketing two or more products together. In the pharmaceutical market certain chemical and biotechnology related products complement each other. This is especially valid in the context of diagnostic kits and drug cocktails for the treatment of AIDS. The other major source is the nature of physician prescriptions, public health schemes, and promotional activities of large firms. The chemical and seed firms in agricultural biotechnology area derive advantages from tie-in sales of biotechnology related seeds and chemical supplements. In the early stages of biotechnology innovation small firms, operating in niche segments, dominated the market. However, with subsequent developments it has become profitable for large firms to diversify so that they can take advantage of the economies of scope in demand. This tends to increase the concentration in the industry.

6.4 MONOPOLY POWER AND PRICING Recent studies of the effect of monopoly power on the pricing of products tend to consider a market in which n firms produce and market homogeneous or very closely substitutable products. Subramanian (1995) and Watal (1996) are cases in point. To illustrate their essential argument consider a market with n firms. Let Yi the output of firm i; i = 1,2,…,n. Assume that the market demand curve is p=a–Y where p = price per unit of Y Y = ΣYi = total output sold by the n firms Postulate that the cost of production of the ith firm is Ci = c i Y i


103

Assume, further, that the firms are Cournot rivals. That is, they maximize profits by choosing their output taking the outputs of all other firms as parametric. The profit for the ith firm is then πi = Yi (a – ΣYi) – ciYi = (a – ci)Yi – YiΣYi The profit maximizing choice of Yi satisfies the equation (a – ci) – Yi – ΣYi = 0 Summing over all i yields na – Σci – (n+1) ΣYi = 0 Therefore, ΣYi = an/(n + 1) – Σci/(n + 1) and, consequently, Yi = a/(n + 1) + Σci/(n + 1) – ci p = (a + Σci)/(n + 1) The Lerner measure of monopoly power of the ith firm is therefore Li = (p – c)/p = Yi/p = (Yi/Y) (Y/p) = siY/p where si = market share of firm i It should also be noted that the elasticity of demand is η = – (dY/dp)(p/Y) = p/Y Hence, it can be concluded that Li = si/η It varies directly with si and inversely with the elasticity of demand. The level of the industry monopoly power can be represented by L = ΣsiLi = H/η where H = Herfindhal index of concentration Consider the more realistic situation in which the products of the n firms are imperfect substitutes. Let the demand curve for the ith product be pi = a – bYi – Σ*Yj

104


where Σ* is the summation over j≠i = 1,2,…,n The profit of the ith firm is πi = (a – ci)Yi – bYi2 – YiΣ*Yj The profit maximizing Yi, maintaining the assumption of Cournot rivals, satisfies the equation (a – ci) – 2bYi – Σ*Yj = 0 As before, summation over i results in na – Σci – 2bΣYi – (n–1) ΣYi = 0 That is,

ΣYi = (na – Σci)/ (2b+n–1)

It can be readily verified that Li = (pi – ci)/pi = 1/ηi where ηi is the elasticity of demand for the ith product in Cournot equilibrium. As is evident it is not possible to define si any longer and it should be no surprise that Li does not depend on si of any sort. As biotechnology matures it was noted that the large chemical companies, that originally produced agricultural chemicals or hybrid seeds, also tend to integrate into biotechnology based seeds. The usual argument in economic analysis is that such diversification enables the firm to practice bundling and tying and thereby derive monopoly power and associated benefit of increased profit. This issue can be best examined through a numerical example. A general result is of course not available because the phenomenon that will be emphasized in the following analysis is not universally valid. To begin with assume that a biotechnology firm is producing seeds (Y1) and a chemical firm is offering complementary fertilizer/herbicide combinations (Y2). Let the demand curves for the two products be p1 = 10 – Y1 + Y2 p2 = 8 + Y1 – Y2 Postulate that the costs of production are C1 = 5Y1 C2 = Y22 If the two firms operate independently as Cournot rivals their profit maximizing choices will satisfy the equations 5 – 2Y1 + Y2 = 0 8 + Y1 – 4Y2 = 0 Consequently, the equilibrium output choices will be Y1 = 4, Y2 = 3 p1 = 11, p2 = 9


105

and the total profit that both the firm obtain will be π = 42 The bundling argument now suggests that the large chemical firm will find it advantageous to integrate the production of seeds based on biotechnology. This will provide it greater monopoly power and profit. The integrated firm can be expected to maximize total profits π = 5Y1 – Y12 + 2Y1Y2 + 8Y2 – 2Y22 The firm therefore chooses Y1 = 9, Y2 = 6.5 p1 = 7.5, p2 = 10.5 The bundling argument is that the firm will sell the seed and fertilizer together, rather than independently. The bundle they sell will be B = 1 unit of Y1 along with 6.5/9 units of Y2 The total profit for the integrated firm will then be π = 44.5 and the firm gains by such diversification. This tying argument takes the market information as pivotal to such organizational change. Suppose, however, that the farmer knows that one unit of Y1 must be combined with one unit of Y2 to obtain the maximum crop productivity. Then, when the firms are operating independently, only 3 units of Y1 can be sold. This reduces the combined profit to 31 and leaves an inventory of 1 unit of output with the firm producing seeds. This also implies a waste of resources both in the form of capital stock of the firm producing seeds and the variable factors utilized in production and inventory. It is of course possible that the firms learn over time and make some correction. However, such coordination is difficult to achieve and expensive. The integrated firm, that attempts to maximize profits without paying attention to the technical constraint, faces a similar problem. 2.5 units of Y1 remain in inventory and the profit reduces to 25.75. Bundling will not be an advantage if the constraint is neglected. It is reasonable to argue that the diversified firm has the ability to obtain the technical information and the managerial expertise to utilize it in its decision process. Under these assumptions the production constraint is B = 1 unit of Y1 combined with 1 unit of Y2 The corresponding price per bundle will be 18. The profit function can now be written as π = 13B – B2 so that the optimal production choice is Y1 = 6.5 = Y2 The profit for the firm is now 42.25. Tying agreements of this nature can be implemented by a diversified firm to its advantage. However, note that it reduces resource use (or optimize it) so that society also stands to gain.

106


The tendency in economic analysis is to condemn such monopoly practices. However, as the above analysis indicates, a certain amount of caution is necessary. For, the advantages, to the society, of the optimal resource use may outweigh the redistribution of the value to the firm. It is rather clear from the foregoing analysis that firms in the biotechnology industry will have some market power. They may also experience higher costs. Due to one or the other of these factors the market price will be high. However, this does not necessarily imply that there is always a loss in social welfare. Instead, the efficient welfare maximizing choices depend on the demand and cost conditions of the specific case. Sweeping generalizations can be quite misleading.

6.5 DIFFERENTIAL PRICING Studies dealing with pricing of biotechnology products, and pharmaceutical products in general, concluded that the monopoly power granted by the IPR regime generally makes the prices higher. Some of them believe that poorer consumers, especially in the developing countries, will be deprived of life saving drugs because their ability to pay is low. Differential pricing arguments have been set up against this background. The primary argument is that markets should be segmented on the basis of their ability to pay and different prices set up in the two (or more) segmented markets. Three distinct formulations are discernible. They will be considered in turn. Assume that the market can be segmented into two different elasticity zones. It is normally expected that the more vulnerable section of the consumers will have larger elasticity of demand. Suppose, now, that the firm maximizes profits by offering different prices in the two segments of the market. Let p1 = f (Y1) p2 = g (Y2) represent the two demand curves. Similarly, assume that C = c (Y1+Y2) is the cost function. Then, profit maximization requires that MR1 = MR2 = MC where MR1 = marginal revenue in market 1 = p1 (1 – 1/η1) MR2 = p2(1 – 1/η2) and MC = marginal cost Consequently, the prices are such that p1/p2 = (1–1/η1)/(1 – 1/η2)


107

Therefore, the firm charges a lower price in the market where the elasticity of demand is higher. The following exception should be noted. Assume that the market demand curves are p1 = 10 – Y1 p2 = 20 – Y2 Let the cost of production be C = 0.5 (Y1 + Y2)2 It can be readily verified that the profit maximizing choices of the firm are Y1 = 0, Y2 = 6.67, p = 13.33, and π = 66.7 when the firm charges the same price from all the consumers. If the firm does discriminate, its choices will become Y1 = 1.25, Y2 =6.25, p1 = 8.75, p2 = 13.75, and π = 58 (approximately) This does not increase profits for the firm. The firm will not cater to the segment where the willingness to pay is lower. The low ability to pay does not necessarily mean greater elasticity. Instead, it may only mean a shift to the left with the same elasticity. Does the firm charge a lower price in the market where the willingness to pay is lower? Let the demand curves in the two markets be p1 = 10Y1–1/2 p2 = 20Y2–1/2 Suppose the cost of production is C = 5 (Y1+Y2) It can be verified that the optimal choice for the firm is Y1 = 1, Y2 = 4, p1 = 10 = p2 The ability for price discrimination is essentially due to the differences in the elasticity of demand whether or not it reflects the ability to pay. One further aspect should be kept in perspective. Suppose a MNC is producing output at its home base and catering to both the home market and a foreign market. Then, price discrimination, as described above, occurs. However, note that the MNC has the option of producing in the foreign country where the costs of production may be lower. This, in itself, may enable it to offer a lower price in the foreign market. Some organizational issues should be taken into account for this possibility to materialize. These will be considered in the sequel. The other two formulations are based on welfare maximization. Consider Fig.5. Clearly, a monopoly firm will offer output at price pm. This maximizes its profit so long as it cannot

108


discriminate between consumers. However, the welfare maximizing choice of output, Yw, can be restored if the firm offers the additional output at a price pw. The firm would be willing to offer it so long as the area ABC is positive and the markets are kept segmented; i.e., output YmYw will not be sold once again to the rest of the market. As noted earlier there is a necessity for suitable organizational arrangements to achieve this. p,MC pm

MC

pw

A

B C

D MR O

Tm

Yw

Y

Fig. 5 Ganslandt et al. (2001) argued that MC based pricing may not adequately cover the fixed costs of R&D. Therefore, they suggest that an organization, like the WHO, should create a fund that will reimburse pharmaceutical firms the entire sunk costs incurred in drug development. The above analysis does not fully support this viewpoint. However, an appropriate empirical evaluation is necessary to concretely assert that differential pricing suggested above will be adequate. Ramsey pricing goes a step further. It seeks to maximize consumer utility subject to a zero profit constraint. Consider the problem Y1

Y2

0

0

Max ∫ f(y1) dy1 + ∫ g(y2) dy2 subject to Y1f(Y1) + Y2g(Y2) – C(Y1+Y2) = 0 Using the conventional Lagrange multiplier method the first order conditions for maximum yields f/g = [f( 1 – 1/η1) – c1]/ [g(1 – 1/η2) – c2] so that f and g satisfy the equation c1(1/g – 1/f) = 1/η1 – 1/η2, or f/g = η2/η1 i.e., a higher price will be charged in the market with a lower elasticity of demand. However, this is much more difficult to implement. For, unlike the previous two cases the firm is not willing to adopt this scheme voluntarily (to maximize its profit).


109

Danzon (1997, 1998) considered the application of this principle to cover sunk costs. The following salient points may be recorded. •

Ramsey pricing assumes a zero monopoly profits (a normal rate of return allowed). Hence, it is necessary to have a regulated price regime to implement such pricing. A free market operation will resist its use. • In the international context it is not possible to have a coordinated regulatory process for price fixation. In fact, some countries, that do not want to pay any part of the sunk costs, may negotiate lower prices. • From an operational viewpoint, regulated price regimes rarely produce meaningful information on the elasticities of demand. The basic foundation of Ramsey pricing may not be available to the regulator even within a given country. Watal (2000,2001) examined some of the institutional issues involved in the implementation of differential pricing. The following two are crucial. •

It may be necessary to label the products for each of the markets. The colors used for drugs may be one such marker. Size and packing have been utilized extensively to delimit the markets. On the other hand, lower priced drugs may be available through public health schemes with the definite understanding that the physicians will provide them only to the poorer sections of the population. • When trading is across national boundaries the low income country should be expected to guarantee that the low priced drugs will not be reexported. This can be covered under the WTO agreements. The issue of parallel imports has drawn considerable attention. See, for example, Maskus (2001) and Scherer and Watal (2001). A few details will be taken up in the next chapter. Suppose the MNC allows production in a low income country. Of course, technical capability is a prerequisite. In addition, the MNC needs a guarantee that its proprietary technology is safe and that low priced drugs will not be exported. Appropriate institutional arrangements will be necessary to make differential pricing successful. The other thorny aspect is the nature of differential prices. For, under the transfer pricing regulations trade across national boundaries the lower prices •

cannot be rationalized on the basis of the argument that costs of production are lower than comparable products in the foreign country • should not be such as to provide a greater rate of profit than comparable products in the foreign country • should not result in a rate of return on capital in excess of that prevalent in the foreign country Appendix 2 contains a more detailed analysis of these transfer pricing rules. The other problem that has been receiving attention is reference pricing. Suppose a large firm offers a drug at a low price in a developing country. The consumers in the patent country of the firm want justification for why they are paying much higher prices. In other words, differential pricing limits the monopoly pricing power of the firm in all its markets.

110


On the whole, it can be concluded that at least in the short run until patent protection is exhausted it would be worthwhile to put some differential pricing in operation. This practice is already in existence with respect to TB and AIDS drugs.

6.6 DYNAMIC PRICING The above two sections considered the pricing problem from a static perspective. However, two dynamic effects have been brought to light indicating that the actual prices may be lower. First, it was noted that the adoption of GM varieties involve wide ranging changes in production practices. Consider the case of herbicide resistant roundup ready soybeans. These seeds allow no-till planting of crops and a higher density per hectare. As a result the machinery used for planting as well as spraying pesticide are different. Further, once the investments in the machinery are made they cannot be used elsewhere. That is, they are irreversible and sunk. The only way to recover the cost is by repeated use over time. Consequently, the firms try to avoid the possible loss of future market that may result from static monopoly pricing. In other words, pricing practices recognize the changing market over time and the fixed costs. The monopoly power of the seed producer will also decrease as a result. See, for example, Perrin and Fulginiti (2001), Demont et al. (2004), and Weaver and Wessler (2004). Second, in the pharmaceutical sector it was generally observed that, contrary to expectations, the prices are lower when a drug is under patent but increases once the generic drugs enter the market after the expiry of the patent. Bhattacharya and Vogt (2003) explain this in the following manner. Note that the physicians determine the demand for therapeutics in general. They find it costly to learn about new drugs constantly. The pharmaceutical companies tend to set up promotional campaigns that maximize the stock of information and experience with the doctors as a priority over static monopoly pricing. For, they wish to ensure a large market share before the generics enter the market. After the expiry of the patent the branded drug loses its more elastic segment of the market to competitive generics. However, the stock of goodwill that they accumulated with the physicians allows pharmaceutical firms to charge higher prices for their branded drugs. To an extent such price increases will also be necessary to recover the costs of promotion incurred earlier.

6.7 IN RETROSPECT Most biotechnology based products are in their early stages. As yet they are sold under patent protection. Consequently, the market is highly concentrated. A few adaptations of seeds, to suit different agro-climatic zones may make monopoly severe in specific segments but reduce it overall. When the patents are off and competition develops it is difficult to predict the degree of competition that emerges. For, unless costs can be reduced significantly, smaller market segments may not sustain a large number of competitive varieties. Competition may be effective only in markets like the diagnostic kits where the investments are low.


111

Labeling non-GM products may provide an effective means of discrimination. The increase in non-GM prices may offer some advantages to the GM varieties though consumer resistance may still be important. Gambardella et al. (2000) noted that “the competitiveness of the industry cannot be assessed by looking at the individual firms, but also at the broader set of institutions, infrastructures, and policies that influence the actions of companies, and even more important, at the dynamic interactions between these levels of analysis.”


Chapter 7

ETHICS AND ENVIRONMENT

7.1 ISSUES AT STAKE Nature has its own rhythms and cycles. They are inadequate to satisfy our requirements. By their very definition modern technologies, be they mechanical, chemical, or biological, are such that they try to utilize some features of nature to counter other, albeit undesirable, properties of nature. This process tends to enhance the welfare of individuals and the society. Concepts of central heating of homes and offices, air-conditioning, and so on are representative. However, a central problem in the adoption of any new technology is the urgency, on the part of private firms, in commercializing technology with the objective of recovering the sunk costs of R&D and generating profits quickly and over a short time horizon. This may subvert welfare of the society in general. This is by now widely acknowledged. More significantly, such commercial greed did not acknowledge the ethical and environmental consequences of the adoption of such technologies. See, for instance, Dickens (2004). For, in particular, note that every technology has some negative consequences for welfare. Even so, societies and individuals • •

were compelled to accept the negative effects because there was no other choice accepted the consequences of technology because the positive aspects of their productivity were overwhelming • accepted technologies because the negative effects (like environmental degradation) were not even evident at the beginning (they were evident only after cumulative use) In a way, ethical and environmental concerns can never be articulated in any objective and comprehensive manner. Hence, it is not possible to conceptualize a once for all resolution of such issues. It is, therefore, necessary to be alert to the possibility of negative consequences

114


of biotechnology and their economic impact on individuals and the society. It is perhaps fortuitous that prior information is sought in the context of biotechnology. A brief outline will be presented keeping this in perspective.

7.2 ETHICAL ISSUES Consider the case of an individual whose life expectancy at a point of time is 65 years. Suppose he suddenly becomes seriously ill (ignore the reasons for the present). Assume that the science of medicine has only one remedy for the ailment and that the longevity reduces by 10 years after the person is cured. Would it be ethically defensible for a doctor to administer this drug? The economic consequences of either choice may dictate what is ethical. Dynamite and atomic energy can either improve human welfare or destroy it. It depends on how we use them. In general, technology itself is free from ethics. See, for example, Thompson (2000) and Dickens (2004). The ethical question is: should it be used for destructive purposes? Consider the issue of genetic feticide (bias against female babies). The current social norm is for every male to have a life partnership with only one girl. The ethics of sex selection, apart from the religious overtones, becomes an ethical issue because the one-for-one relationship cannot be maintained. Is it not possible to change the social norm? Or, will that be unethical in some absolute sense? A time may come when the balance is against the males. What happens then? Mechanical inventions and chemical technologies were at the apex of the industrial revolution of the 20th century. Modern manufacturing is mostly automated. Each individual working on the assembly line deals with only a small part of a total product. In the early stages it was pointed out that this robs the individual of the creative expression involved in manufacturing the product in its entirety. If a worker is operating only one machine, which produces a small part of the final product, all the time, he cannot even discern the value of his activity to the user of the final product. Such mechanization of production alienates him from the product itself. As Knight (1933, p.21) puts it, “specialization in itself is an evil, measured by generally accepted human ideals. It gives us more products, but in its effects on human beings as such it is certainly bad in some respects and in others questionable.” Automation also reduced the demand for labor. Even this was regarded as unethical. There is occasional resistance to the use of mechanical contraptions. However, the enormous increase in productivity, that such technologies provide, tilted the choice in their favor. Societies accepted the negative consequences as inevitable. The advent of the green revolution necessitated large land holdings for the effective use of irrigation, fertilizers, and harvest equipment. Of necessity this meant that some small and marginal farmers lost their land and may be livelihood that they are accustomed to. Some considered this displacement to be unethical. However, the productivity increases created an uneasy truce.

Ethics and Environment

115

Use of chemical technologies created a different kind of problem. For, fertilizer and pesticide residues in foodstuffs have been recorded. This was perhaps not anticipated when these technologies were introduced. They are a result of cumulative use. The ethical question is the amount of such residues that may be deemed acceptable. Similar ethical issues appear in the context of biotechnology. They continue to persist because alternatives (conventional chemical technologies) are still available. There is also no clear evidence of superior productivity of biotechnology. The religious issues and social norm considerations may appear to hold sway due to this. It is useful to consider them objectively and keep their economic consequences in perspective.

7.3 ENVIRONMENTAL ISSUES As noted above most of the ethical issues arise in the context of the impact of the use of a product or technology on specific individuals. Of course, some ethical issues may also refer to their impact on society in general. By way of contrast, environmental issues refer to external effects on individuals, other than the users themselves, of the use of a product or a technology. Environmental issues have some common features and others that are distinct. In general, some technological changes may result in unexpected and unwanted consequences; e.g., air pollution from the use of fossil fuels, effect on soils of excessive fertilizer use in the context of the green revolution (in particular, the sterility of soils and pest infestation of unexpected proportions). A few examples will clarify the issues involved. To make my life comfortable I need some wooden furniture in my house. Trees must be cut to make them. It is not just I but all others in society want such amenities. More trees are cut (during a fixed interval of time) in comparison to what nature can regenerate. This process reduces the oxygen in the air and the green cover against floods, soil erosion etc. The negative environmental effects, it is said, outweigh the positive economic benefits. In the final analysis, at any point of time, only a few technological solutions are available to overcome the pressing problems of the society. The choice may be between extinction now or extinction tomorrow depending on how technology adapts itself to ever increasing challenges. As the Nobel laureate James Meade remarked earlier, neither the selfishness of the current generation nor its altruism may be the best economic solution to environmental concerns.

7.4 ETHICAL ISSUES IN AGRICULTURAL BIOTECHNOLOGY The ethical concerns can be broadly classified into three groups. They relate to the perception of • • •

individuals and the effects on them as individuals society; some activities are commercially beneficial to the firms but affect the society adversely, others benefit the society though they are not commercially beneficial the duties of the current generation towards posterity

116


Assume that there is a shortage of food. This may leave many individuals starving or left under nourished. Whatever may be the negative effects of GM foods they may help these individuals to crawl out of this problem. Is it ethical to leave them starving because it is unethical to use GM foods (this is colorfully reflected in the resentment “there is a gene in my food”)? The point of the argument is as follows. The use of chemical fertilizers in non-GM foods left pesticide residues that are harmful. And yet we use them because the alternative of starvation is not efficient. Can it be conclusively established that the negative effects, on individuals, of the use of GM foods, are greater? In particular, there is a possibility that GM foods will leave Bt toxins and other forms of toxins in food. As Altieri and Rosset (1999) pointed out, genetic modification of seeds and plants may also alter the metabolism of the food producing plant and make it produce new allergins or toxins. Also see Mizan (2000). Will the toxins be more harmful to individuals than the pesticide residues? The fragmentary evidences available at the moment cannot provide any conclusive answer. However, private firms are mindlessly pushing products and technologies that are relatively easy to discover. The long run effects of the introduction of GM crops are not clear as yet. McGloughlin (1999) and Mizan (2000) also raised the possibility that GM products may also decrease the production of essential nutrients thus reducing the nutritive and protective value of food. In particular, it was noted that round up ready soybeans are inferior due to reduced quantities of isoflavons that are known to be anti-cancer agents. Creating such disadvantages to individuals, some argue, will in itself be unethical. However, as Robinson (1999) remarked, “where do divine responsibilities (natural cycles and dependence on them) end and man’s begin? The dividing line is not clear, and all human endeavor could be said to interfere with God’s will to some extent.” Biotechnology applications to the animal world raise some ethical issues. For instance, it is known that rBST increases production of milk in cows. But it is also claimed that the animals are at risk because several health problems arise. Since rBST is administered through an injection this is an in vivo GM technology. Is it ethical to make the animal suffer for interest of humans? See, for example, Thompson (1999). However, it was pointed out that a certain total feed management practice will eliminate the problems. It is not used simply because it is too expensive. There is perhaps a need for certain regulatory practices to eliminate such undesirable effects of biotechnology. Rejecting them as unethical may not be the most efficient solution. Similarly, as Giescke et al. (2004) remarked, “genetic improvements of staple crops like the sweet potato, and cash crops like tea and coffee remain commercially unattractive for large biotechnology companies. The benefits of poor nations must be addressed by working on these crops as well.” A related aspect was noted in Robinson (1999). “Quinova is a traditional crop of the Andes and the indigenous farmers have been breeding it for the prevailing conditions for centuries. In 1994 a patent was issued to two U.S. agronomers covering the use of CMS in Bolivian cultivar ‘Apalena’. Granting a patent on a staple food crop from a poor country to outsiders sets a dangerous and disturbing precedent and must be regarded as ethically unsound.”


117

Geographic indicators is a contested area of the TRIPS agreement. Article 22.1 defines them as “ indications which identify a good as originating in the territory of a member, or a region or locality in that territory, where a given quality, reputation or (some) other characteristic of the good is (typical) to its geographic origin.” To be more specific, consider the following. Neem trees can grow in every country. One country calling it neem cannot prevent others from using the same product (the only thing is that they may have to call it something else and not neem; that is, the trademark or copyright for the label alone can be protected legally.) Basmati rice was produced in some parts of India for centuries. Some other country may now take this variety, produce it, and call it Basmati as well. (Before the mania for patents developed collecting such samples, exporting them, and cultivating them elsewhere was common place. Nobody objected. But in the post patent regime even this is a violation.) Does this give them a right to exclusivity just because it was not patented in India earlier? TRIPS acknowledges that they can be protected through copyrights and trademarks. Conventional plant breeding methods developed efficient varieties suitable to specific agro climatic conditions. For the present biotechnology seeds do not have the same flexibility. Even so, if the GM varieties replace the conventional non-GM seeds many marginal farmers may become bankrupt. Recall that this ethical issue was raised in the context of the green revolution as well. Of greater concern is the concept of terminator seeds. Traditionally, farmers have been accustomed to preserving seed for planting the next cycle. In fact, the germplasm of many countries contain only such robust varieties. At least initially it was felt that biotech seeds will also be similar. However, some possibilities of a decrease in productivity with repeated use have been sighted. These natural processes determine the demand for seed at any point of time. The biotech seed companies feel that this demand is inadequate to maximize their commercial interest. Monsanto, for example, writes into their contract with their farmers a requirement that they buy seed every year thereby prohibiting replanting. Worse still, MNCs created terminator seeds that are sterile and cannot be used for replanting. This is clearly unethical in that it is not an inbuilt compulsion of technology but an artifact set up to maximize the profits of large seed companies. The unfavorable distribution of gains, in favor of seed companies, assumes unethical proportions. For a brief discussion of this issue the reader may refer to Robinson (1999). In the U.S. a method for producing corn syrup with high fructose content has been developed. As a result their imports of sugar from developing countries has gone down. The question being raised is that any substitution that affects farmers from less developed countries is unethical. Of course, the answer from the MNCs would be that they must yield if they cannot be competitive. The argument about other GM crops is also similar both within a country and across national boundaries. The loss of exports for farmers of developing countries is unlikely to be acknowledged as unethical. Consider the case of Round up ready soybeans. Adoption of this variety necessarily increases the demand for Round up. This gives an opportunity for Monsanto to tie the sales of seeds and herbicides and tilt the market advantage in its favor. The antitrust arguments consider this unethical because it is anti competitive.

118


One thing is by now clear. The private commercial interests of some individuals may be in conflict with the personal interests and welfare of other individuals. The following issues are pertinent in a broader social context. • Unequal access to individuals • Equity in the distribution of gains • Loss of livelihood from traditional knowledge As noted earlier, non-GM foods are as yet available in adequate quantities. Hence, there is no pressure to use GM foods. Ethical questions are debated only because the cushion of nonGM foods is available. What happens when the demand out strips their availability? Will the answers to the ethical questions change when this happens? In the final analysis, when it comes to adverse effects of biotechnology on individuals, the primary ethical concern is their right to information and decision. See, for instance, Gesche et al. (2004). No other recommendation can be justified on any objective basis.

7.5 ETHICAL ISSUES IN DRUG DEVELOPMENT Consider the context of cancer and other severe diseases. Assume that an in vitro test becomes available. A doctor knows that there will be an adverse emotional impact if the result of the test is positive and revealed to the patient. Should such tests be allowed at all? What if they are conducted after taking the patient’s consent? It can even be argued that such knowledge may have a positive impact on the individual and his welfare since he can take adequate precautions and/or rearrange his activities. Issues related to birth control, gender identification of fetus, and designer babies belong to the same category. However, as Dickens (2004) argued, parents may be able “to spare their children from natural chance or lottery of individual genetic constitution.” The other side of the argument is that such advantages subvert goals of equality of opportunity in society. This is an ethical dilemma. In vivo tests may be more harmful. The ethical concerns may then be genuine. For, as Oh (2002) remarked, if the investigator (who is doing the clinical tests) administers the drug, without prior consent of the patient, the principle of individual autonomy and rights will be broken. However, as he pointed out, taking consent may impact social welfare adversely. For, the testing process may get delayed for lack of volunteers. The efficient trade off between individual and social welfare cannot be defined in any objective manner. The question being raised is the following. Suppose, in the natural course of events, someone gets sick. It would be ethically defensible if a biotechnology drug is used to cure such a patient. However, it would be ethically wrong if similar drugs are used on a normal, healthy person to enhance performance. As Dickens (2004) puts it, it would be morally indefensible to administer memory blocking drugs that result in socially irresponsible but self-satisfied behavior of individuals. (Recall that we regularly test athletes for the use of performance enhancing drugs). It is by now well known that patients in developing countries, who are suffering from severe diseases like cancer, cannot afford the prices charged by the MNCs. Part of the reason


119

may be that they pay the costs on their own in the absence of any comprehensive national health plan or insurance coverage. How should this issue be addressed? Who should? Let the governments of the developing countries take the responsibility is the answer from the MNCs. Some issues, noted earlier in the context of agricultural biotechnology, can be discerned here as well. For example, the MNCs tend to ask firms in developing countries to do clinical testing of potentially hazardous drugs. The private firms in developing countries fall prey to this in the interest of making profits. They pay scant attention to the negative effects on human beings. The MNCs, preoccupied with maximizing profits from their R&D efforts, do not pay attention to the development of drugs for malaria, Hepatitis B etc. prevalent in developing countries. Clearly, there has been an unequal access to the benefits of biotechnology in the early stages of development so far. (Note that similar accusations regarding other technologies are possible.)

7.6 ENVIRONMENTAL ISSUES One of the positive aspects of biotechnology is the prospect of bioremediation. The following observations are pertinent. •

Microorganisms have been created to clean up environmentally obnoxious materials such as engine oil. They were useful in cleaning up oil spills. This technique was utilized at Haldia and Mathura refineries recently. • Biologically altered production methods for indigo, for instance, turned out to be environmentally safer. Common colon bacterium E.coli has been engineered to synthesize indigo. The biotech indigo is indistinguishable from synthetic chemical material in handling dye-mix preparations and performs as efficiently. • To the extent GM crops have resistance to common pests, the use of conventional pesticide is reduced and consequently the associated pollution (in fact, it was held that frequent spraying of conventional pesticides has been harmful to the health of the farmers and their families as well. This was noted in Schaal (2004)). • Reduction in the use of chemical fertilizers and irrigation water, to the extent they are possible, will assist in reducing environmental pollution • GM bacteria have been utilized as bio-sensors for detecting land mines • Genetically modified plants have the capacity to isolate and separate toxic substances. These plants are sown as a lawn and the resulting plants harvested and disposed off as toxic waste. See, for example, Schaal (2004), Mizan (2000), Khan et al. (2004). Schaal (2004) also noted that Bt produces a family of crystalline proteins that inhibit insect growth. These cry proteins are considered environmentally friendly insecticide. In fact, the bacterium is used as a natural insecticide in organic farming.

120


While there are some beneficial aspects of biotechnology a number of environmental issues have been brought to light. Most of these are in the context of agricultural biotechnology. They generally address the following issues. •

•

•

•

• • •

Consider •

•

When pests and herbs are reduced as a result of the GM varieties there is every chance that birds and animals will be affected. For, this results in less food and cover for insects and bird species higher up the food chain that feed on weeds and insects. The use of pesticides and the resulting loss of feed for birds and animals may affect biodiversity adversely Transmission through air may result in contamination of nearby farms as well. The potential transfer through the flow of genes from herbicide resistant crops to wild or semi-domesticated relatives can lead to the creation of super weeds. The use of irrigation water and the runoff into neighboring fields and sources of drinking water have negative effects on the population at large. The genetic material may be transmitted through irrigation flow. When agricultural inputs flow into the surrounding farm lands the possibility of aggressive insect populations finding their way into these farm lands increases. The potential transfer through the flow of genes from herbicide resistant crops to wild or semi-domesticated relatives can lead to the creation of super weeds. At the end of a cropping season the remaining GM plants are ploughed into the soil. This may change the genetic makeup of the farm as well as the neighboring farms. The use of Bt crops affects non-target organisms and ecological processes. Bt toxin present in crop foliage plowed under harvest can adhere to the soil colloids. This affects the invertebrate populations in the soil that breakdown organic matter and fulfils other ecological roles There is a potential for herbicide resistant varieties to become serious weeds in other crops The loss of soil nutrients and creation of more and vigorous pests There is a possibility that vector recombination generates new virulent strains of viruses, especially in transgenic plants engineered for viral resistance with viral genes. Recombination between RNA and a viral RNA inside the transgenic crop may produce a new pathogen leading to more severe disease problems. This was noted in Altieri and Rosset (1999). these issues in turn. Some species survive beyond the plantation period. Whenever these plants contain herbicide resistant genes, the risk that such species survive beyond crop cycles and pollinate with weeds around farming areas is high. Such uncontrolled cross pollination is the main reason for the rise of herbicide resistant weeds. These resistant weeds then invade natural plant communities beyond farms. See, for example, Sampath (2004). Biotechnology derived species might affect non-target organisms. For instance, wind blown pollen from crops may affect natural surroundings. In particular, it was reported that Bt pollen kills monarch butterflies.


121

•

The existence of pests on a farm helps birds and other species like butterflies to thrive. Pest control through Bt varieties cuts this off. It is possible that there will be a net increase relative to the conventional system in the amount of wildlife early in the growing season in the GM treatment (arising from the opportunity to leave spraying until later in the season) but a net relative decrease (due to the efficiency with which weeds are removed) later in the season, after the broad spectrum herbicide is applied. As yet it is difficult to determine which of these two effects predominate. Assessing concomitant effects on other organisms is, at best, hazardous. Often, the new GM varieties crossed on their own with local landraces and native species. This has a profound effect on biodiversity, by altering agricultural practices, by introducing species that displaced native species or by altering community dynamics. Sometimes whole sections of chromosomes are transferred introducing genes that may produce undesirable traits like early dropping of seed or reduced crop yield. See, for example, Schaal (2004). • The production of weedy hybrids is another concern. The worry is that when a GM crop hybridizes with a wild ancestor, the hybrid offspring will lead to the formation of a vigorous weed. Hybrids have an enhanced fitness and are resistant to attack by some lepidopterons. Bt hybrids have greater seed production thus raising the specter of gene flow altering both the gene pool and providing a new weedy taxon. With this hindsight there are environmental concerns in the use of biotechnology. As of now most of these effects are speculative. It must also be noted that not all environmental effects of biotechnology are bad. In particular, techniques of bioremediation also have the prospect of cleaning up chemical pollution. Caution may be justified. It would be worthwhile to examine the possibilities and their economic consequences further.

7.7 INTERNATIONAL CONVENTIONS Note that the ethical issues are generally of two types. They are either biodiversity related or they pertain to the unequal distribution of gains among participants in a given country or across different countries in their trading relations. Similarly, issues of biosafety may relate to the effects of biotechnology products on individuals or environmental problems as they pertain to societies in general or across international boundaries in the context of international trade. The pertinent questions are as follows. •

Since each country is sovereign in its decision making it can put some regulations in place to protect its citizens. What actions have individual countries taken in this regard? How efficiently are these regulations implemented? • International trade is governed, in recent years, by protocols like the WTO agreements and the Commission on Biodiversity. To what extent do these regulations address the issues involved? Are they adequate to address the issues of concern? To begin with notice that by now almost all countries have rigorous procedures for field trials and approval of crops and clinical trials and regulatory approval for drugs and diagnostic

122


kits. However, there are sharp differences. For example, in the U.S. the tolerance limit for genetic materials is 5 percent whereas in the European Union it is 1 percent. This lack of uniformity should be expected across different countries. It does not pose any problem for domestic transactions. However, it becomes a bottleneck if trading across nations is conceptualized. On the other hand, the WTO insists on scientific evidence as the only determining factor. According to this principle other countries cannot deny use and trade in GM products if one country has produced scientific evidence about its safety. The only redressal available is through the grievance committee of the WTO. However, the WTO arrangements are conditioned more by the profit making implications of private firms rather than the ethical and environmental concerns of member countries. It is unlikely that a consensus can be reached on any uniform standards that need to be adhered to. Can there be any other mechanism for reconciliation? The following details about the regulatory environment as it pertains to biodiversity and biosafety have already been recorded in chapter 3. They will be recalled here to place the argument in a proper perspective. The Convention on Biodiversity (CBD) and Cartagena Protocol (2003) are the most pertinent. The goals of CBD are the • conservation of biodiversity • sustainable use of its components • fair and equitable sharing of benefits • appropriate access to and transfer of relevant technologies and products To pursue these objectives CBD • •

recognizes sovereignty of countries and their genetic resources focuses on in situ (within the body) conservation of genetic resources (not in gene banks) • recommends protection of technical knowledge The Cartagena Protocol deals mostly with international trade. It emphasizes the need for •

an adequate level of protection and monitoring of transboundary movements to ensure safe transfer, handling, and the use of genetic materials • minimizing risks for human health and environment The following approaches have been suggested. •

Prior informed consent (based on scientific knowledge and tests); the exporting country should inform the importing state of the nature and hazards of shipping GM products and obtain written consent • Importing countries have a right to refuse such consignments if they are not satisfied about safety or destroy the lot if illegally shipped The WTO is a multinational trade organization focusing primarily on mechanisms for free trade across national boundaries. Environmental problems of common concern to all countries will be a subject of consideration under WTO regulations. But if one country declares a product


123

safe others cannot reject it. Similarly, under the WTO agreements all varieties of a product will be treated alike. Countries cannot discriminate between them on the argument that they are not local varieties that suit their agro climatic conditions. However, many countries feel that biotechnology related seeds offer few varieties. They are not necessarily tailored to suit the great variety of argo economic climates that we experience. The vast array of old varieties, obtained through conventional breeding will become useless if they are not utilized for a long time. Switching to biotechnology may result in both these types of losses due to reduced biodiversity. Most countries find both these aspects disturbing. However, the preoccupation with trade renders any discrimination on this basis as a non-trade barrier under the WTO configuration. To reach an agreement on high enough standards may avoid clashing with trade related issues but it is difficult to achieve it. The Cartagena Protocol allows individual country governments to discriminate on the basis of safety and environmental concerns. This clashes with WTO agreements. That is why the U.S.A. and EU did not sign the Cartegena Protocol. For all practical purposes they say that non-trade barriers are not acceptable to them. The question being raised is this. If countries are sovereign in their decision making will they not impose different constraints? Some of them may be considered legitimate in the interests of biodiversity and safety. Others may be deemed as rules to protect trade and local enterprises against invasion by MNCs. How to distinguish between these two? Whose decisions should be honored? WTO principles claim to depend on scientific principles that cannot be refuted by any one country. Others claim that this is not enough because there are still issues about the quantum of damage that is acceptable to any one country. This may have to be weighed against the loss in trade itself. Differences in judgment create an impasse. The WTO has a dispute settlement mechanism designed to deal with disagreements between members over interpretations. Further, the committee on Trade and Environment of the WTO recognizes potential conflicts between trade liberalization objectives and environmental protection objectives. However, they tend to give a greater emphasis to trade related matters. In contrast, the Convention on Biosafety is a multilateral environmental agreement without a clear mechanism to settle a dispute if an exporting country disagrees with a unilateral trade barrier imposed by the party of import. This, as well as other related issues, have been considered exhaustively in Isaac et al. (2001). The problem is to find a way of resolving the contradiction between the goals of WTO and the Cartegena Protocol. Voluntary labeling of GM products, to identify market demand for specific types, came as a result of this. What it means is that, even before the regulators (country governments) in the importing country ask for a test, the firm in the exporting country will declare the biological content. The other alternative is to set up a uniform stringent standard for everybody to follow. This may not be quite acceptable under the issue of biodiversity. From the viewpoint of economic analysis there is a necessity to clearly define the tradeoff between the needs for food and welfare as opposed to the advantages of a cleaner environment.

124


However, this is not likely to emerge so long as non-GM varieties of food are available in sufficient quantities. Note that some ethical issues, especially those related to unequal access of biotechnology to the less developed countries, have not been addressed adequately. The only mechanism currently in place is differential pricing. Perhaps the only way out is for individual countries to formulate suitable policies for themselves. The MNCs and international trade regimes are unlikely to address them. One point should be clear by now. The issues of ethics and environment raised are common to almost all technologies. To claim that biotechnology effects are devastating is an exaggeration.

7.8 LESSONS AND CONTROL There is a justifiable concern that current field biosafety tests tell us little about potential risks associated with commercial scale production of transgenic crops. A main concern is that international pressures to gain markets and profits is resulting in firms releasing GM crops too fast and without consideration for their long term impacts on people or the ecosystem. Ascertaining the associated social costs and controlling the negative effects has hardly begun. Put in a different way • There are several directions in which biotechnology developments are progressing • Only some of them can be said to enhance social value • Significant negative effects are possible in the use of certain technologies It would perhaps be more prudent to obtain reasonably complete information before granting regulatory approvals. Unfortunately, the commercial greed and economic power of a few large corporations may be forcing early introduction of varieties that may lead to undesirable effects over the long haul. The efficient response will not be despair. Nor should it be advisable to throw the baby with the bath water. Instead, the need is for greater caution, proper regulation, and policy reforms that curb or reduce the deleterious effects. In particular, it is necessary to educate farm workers about • safe ways of spraying • disposal of pesticide containers • practices of integrated pest management See, Gupta and Chandak (2004).


125

Chapter 8

GOVERNMENT POLICY

8.1 AN OVERVIEW R&D, whether it is scientific knowledge or applied science, materializes only when • •

individuals in the society are motivated to do it obtain the necessary support (including the receipt of the associated rewards) from private firms and/or the government • the social and organizational culture is favorable to it In some countries, especially based on private enterprise, there is adequate R&D culture. However, in our case, there is hardly any culture of private R&D in the industrial sector. Instead, there is excessive dependence on public institutions and government funding to carry out R&D. It is generally agreed that some R&D activities have a public good nature. For example, the road network for transportation, and defense related activities (at least until the activity reaches a mature stage, e.g., the internet). The government must generally finance such R&D. Some activities, such as • •

•

agricultural extension services public health and insurance schemes to cover the disadvantaged sections of the society require public funding on a continuing basis. Such public financing of national health schemes may augment the demand for private goods (medicines) and their appropriability. A similar argument suggests that defense based public investment has spillover effects on private investment

126


In general, public investment and/or financing may affect the supply of goods as well as the demand for them. In the initial phase of development of some industries the government has another fundamental role. Apart from financing private research it has to support public institutions that undertake research. This is basically due to the lack of appropriability of research despite its contribution to social welfare. Stated somewhat more generally, government sponsored research in publicly funded and managed institutions •

complements private research; there are many socially desirable activities that the private sector will not take up because they cannot make enough profit. Perforce the government must find ways of achieving such activities in practice. Similarly, there are certain activities, like the communication networks, where private investment becomes profitable only when a critical amount of activity materializes. That is, public R&D may precede private R&D in order to create a cost effective atmosphere for the later to flourish. • may be competitive with private research in the context of most biotechnology activities, and the human genome project in particular. There is an apprehension that knowledge developed and patented by the private sector will not be readily available for further developments. Hence, public research may have to be accelerated in competition with private research so that the overall stock of knowledge increases and can be licensed freely to all users. A very useful approach to these issues can be found in Ishibashi and Matsumara (2005). However, in areas like the semi conductor devices and biotechnology it was felt that • • •

they do not have any public good character private firms can appropriate (recover costs on the market) the results of their R&D Private sector seed companies have been offering extension services and recovering their costs. There is enough demand because food is a necessity. Hence, public spending is not necessary • Patents and IPR protection may be adequate to sustain private R&D spending • at the most the high risk of initial scientific research may need to be shared by the government Public funding of the activity is also considered to be inhibitive because • too much political intervention curtails the freedom of the scientists to pursue their curiosities and hence the nature and scope of private R&D • too much public investment crowds out private initiative and investment. Appropriability of private investment decreases with the volume of public investment. For instance, why would a farmer pay for extension services if he gets it free? There is also a feeling that public financial resources •

would be inadequate in emerging areas like biotechnology

Government Policy

127

• may be directed to areas that will not necessarily be the priority of private firms In addition, there are problems associated with transferring knowledge from the public research institutions to the private sector firms. In particular, the • •

organizational arrangements (they are very different in these two sectors) financial arrangements like licensing or taxation of returns are crucial issues that determine the efficiency of such transfer. Since financing has such an important role, it is necessary for the government to •

define its policies with respect to foreign direct investment and foreign institutional investments • create an atmosphere wherein venture capital and other financial institutions can function efficiently (some regulation is necessary even if the government does not underwrite such activities) One other aspect of government intervention was deemed to be of crucial importance. The release of some GM foods and pharmaceuticals may prove hazardous to • individual consumers • the society • the environment Biosafety and regulation of GM products becomes a crucial issue in such a context. More generally, the social objectives and definition of socially desirable projects is moving beyond appropriability and marketability. In sum, the major issues for government policy pertain to production and/or provision of some goods that are socially desirable but for which private enterprise cannot recover costs efficiently • financial arrangements that provide a catalytic effect for the development of private activity or complement private investment • regulation and control of investment and production in view of the ethical and environmental implications. These are the crucial issues for government policy if some details are set aside. The following sections consider each of these aspects in some detail. •

8.2 SCIENTIFIC R&D Scientific discoveries are generally unpredictable. For example, Monsanto engineers found some bacteria in the sludge outside their herbicide plant producing Round Up. This created the curiosity to introduce the genetic traits of these bacteria in Round Up ready soybeans. Such accidental discoveries are commonplace in scientific R&D. A variety of defense related R&D expenditures, financed by government sources, gave rise to revolutionary breakthroughs like the internet and biotechnology (especially the Human Genome Project). Public health interests and, more recently, the emphasis on containing terrorist activities have been leading

128


to increased government involvement in financing R&D. (Should the government organize and coordinate these activities is a question that can be kept distinct from public funding of such activities.) The basic economic rationale for public research is the insufficient appropriability. That is, certain activities are such that no one individual can claim ownership and/or recover the costs of selling the resulting products on the market. For instance, scientific R&D production requires finances (and other resources) and the results are conventionally available to everyone free of cost. In most cases, and biotechnology appears to be an especially severe case, the investments are large, the uncertainties are significant, and the time lags between the discovery and market returns very large. Each of these aspects deters private firms to undertake R&D on any large scale. Consider the following model to provide a brief overview of the process of R&D with public funding as an alternative. Suppose a private firm targets a volume S of scientific discoveries. Let S2/2δ represent the total investment in such R&D. In this formulation, δ is a reflection of the R&D culture, motivation of the private investor, organizational competence, and technical skill of the group. However, there is no assurance that the expectations will be fulfilled. Suppose the actual result is S + u where u is a random variable with E(u) = 0, and v(u) = σ2 (It is of course possible that the higher the S targeted the greater the variance. That is, v(u) may be S2σ2 for instance.) Assume that a unit value is associated with every unit of S. The net value to the private investor is then given by V = aS – S2/2δ – λS2σ2 where λS2 is the degree of risk aversion of the investor. It is expected that it increases with the volume of investment. (The reader is invited to work out alternative formulations.) In this formulation a represents the degree of appropriability of the returns of R&D from the viewpoint of the private investor. It depends on the public good nature of the investments. (Some scientific discoveries may be more valuable intrinsically. Then, it is possible to view a as a method of capturing both of these aspects of R&D.) Clearly, the investor will choose S to maximize V. This results in S = aδ/(1 + 2λδσ2) It can be readily verified that S increases as a and/or δ increases. However, it decreases with an increase in λ. S = 0 in the extreme cases where a = 0, δ =0, or λ and σ2 are high. Pure public goods, for which a = 0, will not be offered by the private firms. Similarly, a high degree of risk inhibits private investments in R&D. These situations signal the need for the government to finance such projects (or, at least, share some of the investment financing.) It has been assumed so far that the net value is positive. However, situations of R&D in the form of pure public goods are conceivable. In other words, the potential net value may be negative if the entire R&D expenditure must be financed by the private firm. The government may then step in and finance a fraction of the expenditure on R&D. See, for example, Lawlor (2002).

Government Policy

129

Consider the case where the government finances a fraction f of S2/2δ. Under conditions of partial appropriability the government may claim a share g of the value of the resulting R&D. The choice of g can also be looked upon as a form of taxation. The net private value of the investment can then be written as Vp = private benefit = (1–g)aS – (1–f)S2/2δ – λ(1–f)S2σ2, and Vg = benefit to the government = gS – fS2/2δ P chooses S such that S = (1–g)aδ/(1–f) (1 + 2λδσ2) and the government chooses S given by S = gδ/2f It can therefore be inferred that in equilibrium f/(1–f) = μ (some constant), and g/(1–g) = 2μa/(1+2λδσ2) Formulating the choice of optimal μ necessitates further assumptions. However, some important results can be deduced from this exercise. •

When the risk is high the government can only claim a lower share of S in comparison to the investment that they make. In particular, g = 0 only if a = 0 or S2 is infinitely large • In general, the government’s ability to recover costs will be low if a < 0.5 • Appropriability need not depend on the public good nature alone. Suppose the gestation periods of converting scientific research to marketable product is long. The time discounting, or the impatience of the private forms to recover their investments, makes a small • Some scientific discoveries may not result in any marketable product contrary to expectations • If there are fixed costs of R&D there may be problems of recovery. Public funding will then be necessary As noted in chapter 4, the process of R&D and discovery is continuous over time. Hence, a dynamic model may capture the effect of government spending more truthfully. The basic model can be rewritten in the following form. Let the government expect the private NBF to only finance a fraction f; 0 ≤ f ≤ 1 of the expenditure E. The optimization problem for the firm, returning to the notation developed in chapter 4, will be ∞

Max ∫ e–rt[Rα – fE]dt 0

130

Economics of Biotechnology subject to dR/dt = E – εR

It is straightforward algebra to show that the optimal R will now be R = αV/f(r+ε) That is, a smaller value of f surely results in larger R&D activity. Note, however, that if α is kept high enough by IPR protection and f is also chosen to be low enough to achieve the same socially optimal value of R there may be a volume of R&D far in excess of requirement. This was noted in Phillips and Stovin (1999). Similarly, Byerlee and Fischer (2001) noted that “complementarity with the private sector should be the central criterion in priority setting for public research organizations. In the early stages, public sector support is often the key to private sector entry into the market. However, once the private sector is established, the public sector is often reluctant to withdraw, and in many cases becomes a competition. This may be justified under certain conditions to maintain a competitive seed market in a situation of a potential monopoly supplier, but in many cases, such as hybrid seeds, the public sector continued to carry out such research well beyond justification.” On the other hand, a decrease in f may also imply that a large part of R&D is such that its results cannot be appropriated by private firms. That is, α may decrease. R may fall if α is sufficiently small. This was noted in Cohen and Levin (1989). However, a formulation of the optimal f has been elusive primarily because it is difficult to define the socially optimal R. In general, taxation can be viewed as an effective policy measure to restore social welfare. For, suppose the optimal R, that maximizes social welfare is Rw = wαV Suppose, now, that the government decides to tax a fraction g of value of the firm at each point of time. Then, the firm ∞

Max

∫ e–rt [(1–g)Rα – fE] dt 0

Subject to dR/dt = E – εR It can be readily verified that the optimal choice of R becomes R = α(1–g)V/f(r + ε) The government should then choose g such that f/(1–g) = w, or g = 1–f/w For illustrative purposes, let f = 0.5, and w = 1.25. Then, it is obvious that g = 0.6. The tax rate must be 60 percent if the share of public funding is 50 percent.

Government Policy

131

One qualification is in order. Suppose the government tries to balance its budget by choosing gV = (1–f)E It is then obvious that (1–g)V = Rα – (1–f)E so that (1–g)V – fE = Rα – E so that the tax policy neutralizes the effect of government funding R&D. Somewhat more general formulations are presented in Hall (2002) and Hall and VanReenen (2000). These formulations tend to view such policies as affecting the user cost of capital. In contrast, the static models indicate a more direct relationship between government policies and the optimal choice of R&D. Is government financing the only alternative? The following observations are pertinent. •

Patents and IPRs can improve appropriability even at the scientific level

National health schemes, publicly supported health insurance policies, and agricultural extension services have the effect of improving the marketability • Public intervention may reduce the cost of obtaining finances. For instance, the cost of internal finance increases infinitely if there are quantitative limits to its availability. Venture capital, foreign direct investment and so on may augment supply and thereby reduce the cost of financing. Several other aspects of public intervention have been pointed out. •

Some low value projects, that are pertinent to disadvantaged groups, may not be attractive to private firms. Public investment may then induce movement towards better social value projects and this shift may not be related to appropriability per se. The total value that can be generated itself becomes a consideration • S may depend on some public knowledge and not merely expenditure on R&D. In such a case, public knowledge, financed by the government, improves S • Public funding, say on agricultural extension services, will improve the skills of private farmers and allow them to utilize improved scientific knowledge. This enables them to generate a greater S for a given ‘a’ Wolf and Zilberman (1999) and Lawlor (2000) considered issues of this nature. •

Tax policies operate uniformly on all types of investment. On the other hand, as noted above, the public interventions are specific to different needs. As such taxation is an inferior choice. See, for example, Hall and van Reenan (2000). Lawlor (2000) also pointed out that the undifferentiated nature of the operation of tax credits is such that even much applied and developmental research, that can equally well attract sufficient private funding on its own, may also end up being subsidized. This is also inefficient from a social point of view.

132


One caveat is in order. Appropriability and/or risk is not a consideration when defense needσ are involved. Such investments in R&D cannot be avoided. Public investment is inevitable in such cases. The above arguments carry over even in the context of bioprocessing and the relationship between NBFs and large firms. The following points are noteworthy. • • •

‘a’ may increase at the downstream level (closer to marketable product) Some risk at the scientific discovery stage is already resolved. This renders σ2 low However, there are chances of GM product being rejected at field trials or clinical tests. This makes σ2 high On an average, it can be expected that it increases much more than σ2. That is, the large firm will accept a greater share of investment. Government intervention may not be important. Should the government finance R&D or involve public institutions in doing R&D that can be transferred to private firms eventually? Such open source concepts, even when initial research is funded by private organizations, turned out to be a viable business model in the context of information technology. It is also advocated in the context of some areas of biotechnology. See, for instance, Rai (2005).

8.3 SCIENTIST VS. NBF CONTRACT The relatively large sunk costs, uncertainty regarding the success of a biotechnology discovery, and the appropriability of the expected gains have been significant obstacles in its development. However, it is well known that there are substantial welfare gains from the application of biotechnology. When confronted with similar problems in other technological contexts governments intervened to make the requisite investments before passing on the manufacturing responsibilities to private firms. Financing scientific research in universities is one such activity. That is, the scientist NBF network considered in the previous section has a prior network link between the government and the university system. Almost invariably governments also found that university research has its drawbacks in terms of the speed of delivery as well as the proximity of university science to commercially exploitable ideas. In the context of agriculture the requirements of extension services are too large for university scientists alone. Hence, governments set up specialized research laboratories to accelerate commercialization. They provide an intermediate network link between university scientists and NBFs at least on some occasions. It is therefore worthwhile to examine this nexus in some analytical detail. The classic justification for government financing of scientific activity is its appropriability. Common examples are public health and bioremediation techniques used in the context of environmental pollution. To fix ideas properly, assume that a specific scientific discovery has the prospect of generating social welfare w though it is uncertain. In particular, let the actual

Government Policy

133

welfare gains be (w+u) where u is a random variable with E(u) = expected value of u = 0, and V(u) = variance of u = σ2 (a constant) Assume that a university scientist, or a private research firm as the case may be, spends some of its resources to achieve this breakthrough. Let w2/2δ represent these expenses where δ = the level of skill. This cost may be looked upon as the opportunity cost of the scientist’s time to himself and/or the organization to which he belongs. Suppose, to consider a fairly favorable case, the society allows the scientist a share α of the welfare generated. Then, the net return of the scientist is πs = α(w + u) – w2/2δ α may simply reflect the value he places on promotions, status, access to more research funds, and so on. The scientist will be generally risk averse because the effort may not result in any gains. As a result the net value of the discovery, to him, will be Vs = αw – w2/2δ – λα2σ2, where λ = his degree of risk aversion He will choose w = αδ so as to maximize Vs. The net social welfare can be represented by N = w – w2/2δ – λα2σ2 For all practical purposes, the welfare w is generated at a cost comprising of the scientist’s time and his risk aversion since the society loses as a result of both. Taking the incentive constraint of the scientist into account N becomes N = αδ – α2δ/2 – λα2σ2 Since the relationship between the two parties under consideration is a cooperative enterprise it will be expected that the choice of α will be such as to maximize N. Hence, the efficient α will be α = δ/(δ + 2λσ2) Assume that the market valuation process results in this α and the corresponding w = δ2/(δ + 2λσ2) It can be easily verified that the cost incurred by the scientist = δ3/2(δ + 2λσ2)2, and expected value of πs = δ3/2(δ + 2λσ2)2 Let F be the sunk cost for the research worker. Whenever this is sufficiently large the net expected gains may be inadequate to entice the scientist to proceed with the discovery. A welfare increasing technology will not materialize. Consider the possibility that the government agrees to pay a fraction β of the scientist’s

134


variable cost or some portion of F. In the former case Eπs changes to Eπs = δ3/ 2(1–β) [δ + 2(1–β)λσ2]2 which is higher for all values of β > 0. In other words, a suitable choice of β will ensure that the potential social welfare materializes in practice. The above argument implies that the large fixed costs of biotechnology research may be the primary reason for government financing to ensure the realization of the potential social welfare. Filson and Morales (2005) offered an alternative explanation in a different, but related, context. They consider a specific biotechnology activity, say a drug development, that is expected to cost I. The project has only a probability p of success but not known apriori. Suppose the project is successful. Then, the society is expected to derive welfare w from it. The expected welfare is then pw and the net social welfare N = pw – I There is a possibility that pw < I. In such a case, a private biotechnology firm will not undertake the project and the society does not get any welfare. The government, as stated above, has the option of financing a part of I to make the project viable. Somewhat similar to the approach of Filson and Morales (2005) consider an alternative. The government may initially invest an amount e. This may enable the project to start and assess the probability p of success. The private investor will invest (I–e) with probability p and nothing otherwise. The expected level of investment is p(I–e). Similarly, the expected welfare is pw. Consequently, the net expected benefit becomes N = p[w – (I–e)] – e = p(w–I) – e(1–p) The private firm’s gain can be represented as πp = p[w – (I–e)] = (pw – I) + I(1–p) + pe This quantity may be positive even if pw < I. To obtain an idea about the optimal e assume that the private firm puts in an effort e2/2δ to enable it to discover the actual value of p. In such a case N = p[w –(I–e)] – e – e2/2δ and the optimal e will be e = (1–p)δ This quantity increases with the expected skill level of the private firm. An alternative formulation will be to assume that good projects are easy to identify while bad ones are not. Hence, e will necessarily vary inversely with p. Let e = c/p where c is a constant.

Government Policy

135

It can be now shown that the optimal choice of e is e = (I/c)1/2 This quantity varies directly with I but will be less than that amount. Hence, the government spending, by assisting the firm in obtaining information about p, enables it to generate the welfare expected from the project. Other alternatives, like the possible reduction in variance that such expenditures entail, may be examined as well. One important point should be addressed. How does the government administer the subsidy that it is willing to offer? Clearly, this requires an organizational arrangement. The general suggestion is that some government sponsored agency may be set up as a public research laboratory where incubation and preliminary testing is carried out. Successful discoveries may then be transferred to the private sector through appropriate networking arrangements. However, such institutions experience information asymmetry and develop their own internal bureaucracies and value systems that may be inimical to the development of the biotechnology industry.


Chapter 9

CONCLUSION

9.1 THE TECHNOLOGY Fundamentally the current developments in biotechnology pertain to the identification of the biological processes through which plants and humans generate certain proteins. Once scientific knowledge develops to a point where cell structures, cell functions, and biochemical reactions are fully understood they have the prospect of replacing conventional chemical technologies that have served the same purpose over the years. There are two fundamental advantages of this replacement. •

Erosion of the conventional resource base of chemical technologies need not be a constraint on the development and production of proteins and so on. In fact, there is a good prospect of biofuels replacing rapidly depleting fossil fuels. It is also expected that there will be discoveries of many new materials. Developments in nano technologies may turn out to be catalysts in this change. • The current developments, of chemical based drugs and the fertilizer–pesticide applications on crops, act on a global scale (the entire human body, the entire farm area and so on) thereby reducing the effectiveness of the chemicals and creating undesirable side effects (orally administered drugs affect the liver function, fertilizer and pesticide residues are found in food). Biotechnology operates at a more micro level. As such the drugs and pesticides can be targeted to the exact location of the human body and plants where they are needed. This reduces the cost of application as well as minimizes the side effects. From this vantage point biotechnology may be encouraged for purely scientific interest in the short run. It should be acknowledged that biotechnology would be socially beneficial in the long run. In other words, as of today, biotechnology should be developed as a backstop technology that can be used when chemical based drugs and/or conventional plant breeding is no longer feasible (i.e., their economic viability is surpassed).

138


As of today a vast array of technologies are being developed and a large number of biotechnology products will be available on the market in the foreseeable future. One thing should, however, be clear. Biotechnology does not offer any miracles either in the form of new methods of production, increased productivity, or reduced cost of production. Traditional technologies cannot be completely replaced. In a similar vein it can be argued that biotechnology does not as yet have the prospect of replicating the flexibility and biodiversity achieved by conventional technologies. It is possible that this is purely due to the early stages of biotechnology developments.

9.2 ORGANIZATIONAL ASPECTS Four distinct aspects of commercialization of biotechnology have been recorded. • Knowledge intensity • Capital requirements • High risks • Concern about biodiversity and safety All these features have a bearing on the type of organizational structures found in the industry. Since the area of biotechnology is still new • very few experts with specific knowledge are available • private industry, even if it wants to attract them, find the supply too short • the same is true of laboratory tools to scale up production As a result some public sector involvement has been mandated. It consists of • setting up educational and training programs • setting up and sponsoring research • transfer of knowledge to private firms • tax holiday and other incentives to reduce financial risks Two observations are pertinent. •

The nature of public organizations is such that they tend to attract a lot of political patronage once they are established. As such it becomes difficult to dismantle them at a later stage. The transitory public-private network relationships should change so that public firms withdraw when they have outlived their utility • Patents and IPRs granted to private firms may offer superior incentives as also avoid the above problem The nature and extent of government intervention is, at best, a pragmatic choice. Economic theory does not offer adequate guidelines. However, as in the context of other technologies, there will be long run government intervention in • • •

regulation pricing equitable distribution of gains

Conclusion

139

In other words, the steady state in biotechnology, once it is achieved, is not likely to be very different from other industries. With respect to conventional technologies knowledge diffusion and transfer of informal knowledge is straight forward. The scientist develops the knowledge and technology, trains young students in their use, private firms employ the students, and they in turn achieve efficient transfer. This mechanism is not fully developed in the context of biotechnology. Hence, network structures emerged as a pragmatic choice. In the long run private firms would have developed adequate experience so that the networks tend to be internalized (as is the case of information technology). Note that in conventional technologies specialized jobs and stable job profiles emerged over the years. However, the experiences are different in the context of the more recent knowledge intensive technologies. For example, in the context of information technology, each new production requires expertise of various kinds. Some of them are available within the firm. Others must be acquired from outside. As such some form of networking and contract jobs dominate. It appears that biotechnology is also heading in the same direction. Financial resources available to a firm are insufficient in the initial years. As such venture capital steps in to hedge the risk. However, over time private firms • •

develop adequate resources of their own do not require the same amount of capital because investments will be marginal in a steady state That is, even financial arrangements will change towards a more conventional and predictable structure over the long haul. It can be concluded that even in the organizational context it is unlikely that there will be any fundamental differences when compared to other technologies. Consider the role of patents. The distinctive feature of biotechnology is patenting of knowledge (that does not necessarily relate to a final product valued by consumers). In the initial phases, when progress is slow and expensive, this may be necessary. There is already awareness that this slows down the process of knowledge diffusion. In about 20–25 years it can be expected that • •

local entrepreneurs would have developed generic varieties of GM seeds that are efficient in the agro climatic conditions of specific countries the TRIPS agreement would have lost much of its value since great many competing technologies would have been developed.

9.3 PRODUCT PROFILES AND MARKETS There is, by now, a general agreement that R&D and investments in biotechnology have been rather expensive. It is also generally recognized that biotechnology does not have any significant defense applications and as such public funding has been difficult to achieve. A major share of activity is in the private sector. A few multinationals dominate the market.

140


One of the consequences of this feature is that even when public funding is available private firms accept control only so long as the R&D does not yield any profit. They want the government to withdraw as soon as they can make profit. In other words, private firms want public investment to be driven by their self interest rather than social welfare. Even developing countries are coaxed into accepting government financed R&D which may result in products of value to developed countries and profits for MNCs. Perhaps the developing countries can afford to wait until they can obtain products suitable for their requirements. A second aspect of biotechnology is that MNCs tend to concentrate on biotechnology products that are relatively easy to discover. Seed production in the agricultural context is illustrative. Once a new product is discovered the MNCs want even developing countries to use it (free trade comes as a handy argument) even before the long term effects of their use can be ascertained. For, after all, this is the only sure way of expanding their markets. The marketing strategies of the MNCs are also questionable. First, they created a widespread speculation that conventional agriculture as well as the green revolution have run their course. A quick conclusion has been drawn that people in developing countries will starve if they do not accept biotechnology driven by the MNCs. However, as yet traditional varieties appear to be superior and the apprehensions are a false alarm meant to maximize the profits of private MNCs. Second, consider the following kind of “scientific evidence” offered in Morse et al. (2005). They contend that seeds, produced in India even by Monsanto, have a lower productivity compared to seeds of MNCs imported from developed countries. On the face of it such attempts appear to be deliberate subversion efforts of MNCs commercial interest. Keeping competition out to establish monopoly power is very much an accepted institutional practice. It is therefore not surprising that MNCs may be trying to destroy traditional agriculture in developing countries by sending in harmful biotechnology products. Unfortunately, the commercial greed of some individuals in developing countries may contribute to this without any regard for social welfare. It was also noted earlier that the MNCs are unwilling to set up production units in developing countries even when it is less expensive to do so. As a result they are made to pay higher prices. The monopoly power generated by patents has been misused. It is rather early to speculate about the product profiles and price policies that will emerge in a steady state.

9.4 NEGATIVE EFFECTS The main point about biodiversity is that a large number of GM seeds, adapted to a vast range of soil and climatic conditions are not available. In the absence of such variety implementing biotechnology seeds in a hurry would result in • •

lower productivity loss of traditional germplasm (which need to be stored on the field rather than in seed banks) and possible destruction of varieties developed over the years after a great deal of experimentation

Conclusion

141

The development of an equally large GM variety base is inhibited by • •

one MNC or GM firm not having all the soil, climatic, and seed samples the high costs associated with modification of each variety relative to the small localized demand It is possible to overcome the first problem with adequate effort and at the cost needed to do so. In a sense, the technological difficulty can be circumvented. The economic problem is more difficult. It is perhaps possible to reduce costs with cumulative experience. The more difficult question is that of who pays? If GM seeds are more expensive then traditional varieties (which they will almost certainly) and if profits (if any) accrue to the MNCs the question of equity will arise. However, the question will probably lose its relevance if over time it can be demonstrated that the traditional varieties lost their productivity and GM gained to such a point that only GM is economical. Similarly, there would be certain inevitability for the use of GM even if it is not economical. This may also reduce protests. Either advances in technology and/or economic compulsions may propel greater use of GM crops and products. Until then issues of biodiversity will remain active. The environmental issues are structurally the same as those raised in the context of the use of fossil fuels or other chemical technologies. They are a result of the technology. However, they can be circumvented, or minimized, either by technological modification and/or economic policy. There are social costs of environmental degradation in the use of technology. This reduces social welfare. However, both the technological and economic remedies to eliminate it are expensive. As such the prices of the products of these technologies will increase. This may result in a decrease in demand. The consequences for a reduction in social welfare are then obvious. Some balance must be sought. Similarly, purely economic sanctions, given technology, may reduce economic activity with attendant implications for social welfare. Once again it is necessary to look for a compromise. The argument about the effects on future generations may not get the attention it deserves so long as there is sufficiently rapid technological progress. Detailed work on these aspects, as they relate to biotechnology, has not yet been initiated in sufficient detail. In the final analysis efficient solutions may need to compromise on biodiversity as well as biosafety for the sake of the welfare of the present generation (for, extinction would only mean that there cannot be any future generations!) A more optimistic view would be that the past two centuries have been such that the economic problems- technology- economic policy cycles have been virtuous and resulted in overall improvements in social welfare. Biotechnology revolution may yet be another phase in this virtuous cycle.

9.5 STEADY STATE In sum, the claim that the 21st century belongs to biotechnology the same way as physics was the leader of the 20th century may well turn out to be correct. But this will fundamentally change the way industrial activity is organized. The contours of this steady state are slowly emerging.


Appendices


Appendix 1

TECHNICAL TERMS Acquired Immunity—The cellular immune system causes, through simulation, immunity towards a specific antigen. Amino Acid—The fundamental building block of a protein molecule. Antibodies—Protein molecules produced by B-cells. They are produced in response to the presence of a specific antigen. Antigen—A molecule (foreign substance) capable of activating B-cells and T-cells to induce the formation of an antibody. Assay—A method of determining the presence or quantity of a component. B-Cell—A lymphocyte in the bone marrow that matures into an antibody secreting cell. Bioremediation—A biotechnology to clean up environmental pollution created by chemical technologies and hydrocarbon use. Cell—The smallest structural unit of living organisms that is able to grow and reproduce independently. Clone—A group of cells with an identical genetic structure. DNA—Nucleic acids are macromolecules formed from repeating units called nucleotides. A nucleotide consists of a purine or pyrimidine base linked to a sugar phosphate. In deoxyribonucleic acid (DNA) a purine (adenine A, or guanine G) or thymine/cytosine is linked to deoxyribose sugar phosphate. In ribonucleic acid (RNA) a purine (A or G) or a pyrimidine (uracil U or Cytosine C) is linked to a ribose sugar phosphate. Nucleotides are polymerized by the reaction between the phosphate group of the nucleotide with the sugar of another to produce a long polymer called polynucleotide. The DNA macromolecule in living cells is normally present in the form of two nucleotide polymers closely associated with each other to form a twisted spiral (double helix). The two strands in the double helix are held together by hydrogen bonding between adjacent bases in the helix.

146


DNA Hybridization—When the DNA molecule is exposed to near boiling temperatures, it will unzip by breaking the bonds between base pairs. The DNA molecule will then change from its double stranded normal state to two single strands. When complementary strands are exposed to cooler temperatures they will bind together again. This process is called hybridization. ELISA Test—Several methods are available for determining the presence of genetic materials in foodgrains, oil seeds, processed food, vaccines, and so on. ELISA is the protein based enzyme linked immunosorbent assay technique. The ELISA test identifies a specific antibody reaction that marks the presence of the new protein produced in GM crops. Dipstick tests are now available to quantitatively detect biotech content in grain samples within two hours. Several tests may be required to eliminate the presence of GM traits since the color coding corresponding to each protein is different. However, it is more reliable and faster than PCR based tests. Enzyme—A protein that accelerates the rate of chemical reaction. Genes—Elements of germplasm that transmit a hereditary character and forms a specific part of a self-perpetuating DNA in the cell nucleus. Gene Expression Technique—A method to identify the physical manifestation of the information contained in a gene. Gene Probes—The process of locating and analyzing those single genes that, if defective, can cause a genetic disease. Diagnostic tests based on gene probes have been developed for cystic fibrosis, sickle cell anemia, and so on. Genome—The genome of a certain type of organism defines the comprehensive sequence of nucleotide pairs constituting its DNA material. DNA can be arranged in a single circular chromosome or in several linear chromosomes, each of which is segmented into informational units referred to as genes. Each gene, in turn, contains within its DNA the specific instructions to produce a protein, i.e., the actual coding of the protein sequence, and information about the conditions of this expression. During the gene expression, different cellular elements come together to produce that protein. This process occurs through an intermediate molecule called messenger RNA (mRNA) whose sole purpose is to communicate by means of translation, the information in the genetic code of the gene into the structural and functional elements that specify that protein. Genomics is the study of an organism’s genome including the location, structure, sequence, regulation, and functions of its genes. In Situ—Within the human body. In Vitro—Outside the human body. MAb—Monoclonal Antibody—Cell fusion is the artificial joining of cells to form a new cell by combining the desirable characteristics of two or more other cells. This technique produces large quantities of monoclonal antibodies (MAbs). Markers—Specific proteins to distinguish cancerous tumor cells to make the cancer a clear target for the Mab therapeutics. Once a diagnostic antibody against a distinctive marker has been made it can often be developed into a drug.

Appendix 1 Technical Terms

147

NIRS Test—Near infrared spectroscopy is the other tool that is faster and less expensive compared to the ELISA test. It is based on the principle that the pattern of absorption or reflection of NIRS light is unique for every compound. Thus the identification of the quantity of materials like oil, proteins, and starches becomes relatively easy. Polymerase Chain Reaction—PCR is a technique through which a particular DNA sequence is reproduced exponentially. It involves the use of two DNA probes, that flank the ends of the DNA of interest, adding the enzyme DNA polymerase and the four chemical bases constituting the DNA. Polypetide—A compound that yields amino acids on hydrolysis but has a lower molecular weight than a protein. Probes—Radioactive phosphorous introduced into one of the DNA strands to view the hybridization process. Promoters—Substances that can, in very small quantities, increase the activity of a catalyst. Protein—A protein is composed of hundreds or thousands of amino acids. Proteome—The study of the protein structure and activities present in the cell. rDNA—Recombinant technology uses enzymes to cut and paste fragments of DNA to make recombinant DNA molecules. Typically, a DNA sequence of interest, called the insert, is pasted together with a vector, a piece of DNA that enables the recombinant molecule to be replicated and harbored in a host organism. Recombinant molecules are constructed for the purpose of cloning the DNA, i.e., making a large number of copies of a single molecule. T-Cells—A type of lymphocyte originating in the bone marrow. Vector—The agent used to carry new DNA into a cell. Viruses or plasmids are often used as vectors.


Appendix 2

ECONOMIC CONCEPTS

A2.1 SPENCE FORMULA The availability of a product generally results in some utility or value to the consumer. The price he will pay per unit of output purchased on the market is a reflection of this value. However, there are diminishing returns to value addition as the output increases. The market is also generally such that the firm can charge only one uniform price from all its customers. Hence, the market price is necessarily a reflection of the lowest value to the consumer of the last unit of the product. Clearly, this means that the firm cannot fully convert the entire value to the consumers into revenue for itself. For analytical purposes, it is useful to derive a relationship between the value and sales revenue. This is the basic objective of the Spence formula. The theory of consumer demand generally indicates that the consumer’s optimal choice of an output of a product is such that ∂U/∂Y = λp, where U = utility or value Y = volume of output p = price per unit of Y λ = marginal utility of a unit of money For algebraic convenience it is generally postulated that the units of money are defined in such a way that λ = 1. In such a case p = ∂U/∂Y, and

150

Economics of Biotechnology Y

U = ∫ p(y) dy 0

Of course, the integration is valid if and only if the demand curves satisfy the integrability condition. However, any system of demand curves derived from a regular utility function will satisfy the condition. The purpose of analysis is to set up a relationship between U and the sales revenue R = Yp(Y) Consider the definition of the elasticity of demand, viz., η = – (dY/dp) (p/Y) From this it follows that p = – ηY (dp/dY) Therefore, it can be inferred that Y

U = ∫ p(y) dy 0 Y

= – ∫ ηY (dp/dY) dY 0

Y

= – η [Yp – ∫ p(y) dy] 0 Y

Hence,

Yp(Y)η = – (1 – η)

∫ p(y) dy 0

Therefore, the relationship being sought is Y

Yp(Y) = (1 – 1/η) ∫ p(y) dy 0

This is the Spence formula. Since η > 1 in the operationally relevant range it can be verified that the revenue is almost always less than the total value generated. They will be equal only in competitive markets.

A2.2 ECONOMIES OF SCALE AND SCOPE Following Chandler (1990) economies of scale may be defined as “those that result when increased size of a single operating unit producing or distributing a single product reduces the unit cost of production or distribution.” In practice, the size of a firm has been measured by

Appendix 2 Economic Concepts

151

•

the capacity of the firm (in terms of the potential maximum output, the stock of capital, or many other fixed or sunk costs), or

•

actual output produced

Consider the short run total cost curve of a firm. It depends on the •

stock of capital (fixed in the short run)

•

actual level of production

•

prices of various factors of production

It may be written as C = aKα + bYβ ; α,β < 1, where C = total cost of production K = potential output or stock of capital Y = actual level of production In general, if the fixed cost is written as F = aKα a part of the economies of scale arise from spreading F over a wider range of output. For, the average fixed cost F/Y decreases with an increase in Y. However, as the above formulation suggests, F may itself not increase proportionately with the capacity level of output. This is also an aspect of economies of scale. From this vantage point the fixed costs (generally called the sunk costs because such assets do not have any value outside the activity for which they are designed) are the major source of economies of scale. Drug discovery, drug development (i.e., clinical tests of drugs), and field trials of crops and such R&D activities should be considered as sources of economies of scale. See, for example, Cockburn and Henderson (2001) and Fulton and Giannakis (2001). The other component of the short run cost is the variable cost depending on the level of output. Efficient use of the variable factors of production will be at the apex of such economies of scale. Chandler (1990) expressed these differences in the following way. •

•

“In the older, labor intensive industries, increases in the output of a manufacturing firm came primarily by adding more machines and more workers to operate them. In the newer industries, expanded output came from a drastic change in the capitallabor ratios. It came by improving and rearranging inputs; by using new or greatly improved machinery, furnaces, and other equipment; by reorienting the processes of production within the plant; by placing several intermediary processes employed in making a final product within a single firm; and by increasing the application of energy.” “The potential cost advantages could not be fully realized unless a constant flow of materials through the firm was maintained to ensure effective capacity utilization. The sunk costs are much higher. Thus, the two decisive figures in determining

152


costs were rated capacity and throughput (the actual amount of output). The throughput needed to maintain minimum efficient scale requires careful coordination not only of the flow through the process of production but also of inputs from suppliers and the flow of output to intermediaries and users.” The following aspects are significant while explaining the economies of scale emanating from the fixed cost component. •

Consider the production process that consists of using a cylindrical pipe to transport some liquid. The cost of construction of the pipe (of a fixed length) depends on the radius and the surface area. On the other hand, the capacity increases with volume (or the square of the radius). Hence, the cost per unit of transporting the capacity level of output decreases with the volume. • At any given time, industrial equipment is available only in fixed sizes and capacities. There are increasing returns, for every given size of equipment, upto the capacity level of operation because this enables the firm to spread the fixed costs. • An increase in size may also lead to economies in maintenance staff. For, the law of large numbers suggests that breakdowns are more predictable. Consequently, maintenance staff need not increase proportionately with size. • Drug development (in addition to research on discovery) may experience increasing returns to scale. For, undertaking an activity on a larger scale permits the adoption of more efficient techniques. • Firms generally incur substantial expenditures in clinical trials of drugs and field trials of crops and for obtaining regulatory approval. When the firm increases its level of production these costs need not be incurred all over again. This is valid though the production of additional units of output will require additional cost. Large firms will derive economies of scale by spreading regulatory costs over more output. This aspect was noted in Fulton and Giannakis (2001) and Cockburn and Henderson (2001). Consider the economies of scale in the variable factors. The following arguments are pertinent. •

Assume that a product must be processed on two machines. Then, the production process consists of five elementary operations, viz., giving input to machine 1, processing on machine 1, transferring the inprocess material to machine 2, processing on machine 2, and storing the final output. There will be some idle time on both the machines if one person handles all the steps. Further, his skills may not be best suited to operate both the machines either. Hence, the marginal product will be small initially. However, if the number of workers increases steadily towards five each of the operations can be carried out independently (and everybody can be a specialist in a particular task). There is a gain in efficiency due to the repetition of a single task as well as a reduction in the idle time. Hence, it will be expected that the marginal product will increase steadily until five workers are employed. This suggests that the average costs of production decrease as output increases.


153

•

Large firms may have advantages in the financial markets over smaller firms. In particular, large firms may be in a position to finance their working capital requirements through bank credit obtained at lower interest rates. (For, the banks can save on transaction costs by handling a few large accounts.) In sum, economies of scale can be expected due to • economies of increased physical dimensions of plant • existence of indivisibilities • specialization and division of labor • economies (due to transaction costs) of massed resources By way of contrast, economies of scope are those resulting from the use of processes within a single operating unit while producing or distributing more than one product. More formally, suppose C(Y1,Y2) < C(Y1,0) + C(0,Y2) In other words, the cost of producing Y1, Y2 together is less than that of producing them separately. The production process is then said to exhibit economies of scope. In practice, the following kinds of economies of scope can be identified. •

When the firm has excess capacity and the market for its primary product is saturated it will try to expand into related products. If technology is malleable and accommodates the production of another product without disrupting the production of the main product then there will be an efficient use of capital assets. • Some factors of production are public (non-rival) in the sense that once they have been acquired for use in producing one product, they are costlessly available for production of another. R&D expenditures are a good example especially when inventions are cumulative. Suppose a biotechnology firm isolated a particular gene. The firm can use this to expand its activities. For example, if the firm desires to develop seeds for a new crop, it need not invest in the R&D once again. • Competence in the production of one product line may create competence to produce another related product. This is also a sunk cost explanation of economies of scope. • Some production processes exhibit cost complementarities. That is, the marginal cost of producing one product falls as the output of the other increases. For example, if one chemical is made from a byproduct of another, then increased production of the latter may reduce the marginal cost of the former. Economies of scope in demand have also been noted. As Just and Heuth (1993) argued, scope economies in demand can be said to materialize if the firm can generate greater net profits by marketing two or more products together. In the pharmaceutical market certain chemical and biotechnology related products complement each other. This is especially valid in the context of diagnostic kits and drug cocktails for the treatment of AIDS. The other major source is the nature of physician prescriptions, public health schemes, and promotional activities of large firms. The chemical and seed firms in agricultural biotechnology area derive advantages

154


from tie-in sales of biotechnology related seeds and chemical supplements. In the early stages of biotechnology innovation small firms, operating in niche segments, dominated the market. However, with subsequent developments it has become profitable for large firms to diversify so that they can take advantage of the economies of scope in demand. Network effects are also often a case of economies of scope. For, in industries like biotechnology, communications, cyberspace, and so on a larger network of relationships between firms and related products enables the firm to develop faster and at a lower cost.

A2.3 VALUATION OF RISKY ASSETS Consider the valuation of the common stock of a firm by a shareholder. Let the investor have one unit of money at his disposal. One possible investment is a bank deposit. This provides him a fixed return on his investment. Let this be written as qf. Investing in the stock market is another alternative. Suppose the investor opts for a diversified portfolio with the weightage generally assigned in the stock exchange price index. The basic motivation for this choice is of course the attempt to reduce the risk by holding a diversified portfolio. Suppose the stock price index is qm. This is random. It may also be difficult to assign any probability to different values that it takes. The third alternative is to choose the common stock of one of the companies instead of the diversified portfolio. Suppose the share price is q. Clearly, this price q is random. It is likely to be more volatile than qm and perhaps much less predictable. Suppose the investor decided to buy this specific stock. Then, conceptually, for some value of β he is considering the investment in the specific risky asset to be equivalent to putting a fraction β of his money in qm and the rest in qf. This is the spirit of the Von Neumann approach to the valuation of the common stock of a specific firm. That is, β of qm q (1 – β) of qf Or, more explicitly, write q = βqm + (1 – β)qf Hence, based on the expected value calculation, E(q) = β E(qm) + ( 1 – β) qf The prospective shareholder must reveal this choice of β. An attempt can now be made to provide an analytical approach to identify the value of β. Consider the more general portfolio p

γ

of q

(1 – γ)β

of qm

(1 – γ) (1 – β) of qf


155

That is, let the investor put a fraction γ of his saving in the specific common stock, a fraction (1 – γ)β in the diversified portfolio, and the rest in the risk free investment. Then, the return from the general portfolio can be written as p = γq + (1 – γ)βqm + (1 – γ) (1 – β)qf The expected value of the generalized portfolio is given by E(p) = γ E(q) + (1 – γ) [β E(qm)+ (1 – β) qf] = γE(q) + (1 – γ)E(q) = E(q) That is, the shareholder considers these two alternatives to be equivalent in expected value terms for all values of γ. It can then be surmised that he chooses γ to minimize the variance if he is risk averse. The variance of p is given by V(p) = γ2 V(q) + (1 – γ)2 β2 V(qm) + 2γ (1 – γ) β Cov (q, qm) Hence, the optimal value of γ can be readily computed. It should now be noted that the portfolio p is reduced to q if and only if the optimal choice of γ = 0. In other words, while valuing the common stock of the company the shareholder can be expected to reveal β = Cov (q,qm) / V(qm) From this it follows that the price he is willing to pay for a unit of common stock of this company can be written as q = β* qm + (1 – β*) qf = constant + β* qm For all practical purposes this CAPM (capital asset pricing model) valuation of the common stock is the incentive constraint of the shareholder. It articulates the preference of the shareholder in relation to the overall market price of stocks and shares.

A2.4 TRANSFER PRICING REGULATION Strategic transfer pricing may have several negative effects. A few prominent issues can be highlighted. First, a foreign company may charge a lower price (in contrast to what domestic firms can offer) mainly because the market demand is elastic. It is possible even when they do not have any cost advantage. The consumers may find this advantageous. But the domestic industry may suffer. It may then be necessary to check such pricing. Second, just the opposite may also happen. For instance, a foreign pharmaceutical firm may deny selling life saving drugs except at a high price simply because patents provide it such protection. In other words, low cost production, even when it is possible, may not be allowed. Instead, high production costs will be cited as a justification for charging higher prices. Third, the firm may take advantage of lower tax rates to derive greater than normal profits. This may apply to income taxes in the context of services, profit taxes and excise taxes, and sales taxes in the case of commodities and so on. The basic effects of these distortions may be (a) changing the distribution of gains between the contracting parties without altering efficiency (as in the case

156


of price discrimination), (b) affect efficiency (especially when motivational problems arise), or (c) impinge on third parties outside the contractual relationship. There is resistance to regulation in the first case. However, arms-length transfer pricing regulations are put in place in the other two contexts. Three basic alternatives have been proposed; comparable uncontrolled price (CUP method) or a resale price method, cost plus method (CPM), and profit split method (PSM). The rationality behind these methods is as follows. CUP method – Pricing of any product or service is based on the cost of production. Every firm should be expected to use a cost comparable to the nearest rival or the cost of producing output within the country if foreign competition is involved. Brazil argued with Cipla on these grounds with respect to the manufacture of drugs for AIDS. A similar situation arose in the context of producing statins for cardiological uses. CPM method – There can be differences in the markup used for pricing even when the costs are comparable. The stipulation is therefore the comparison with the percentage markup by the nearest rival product. PSM method – An alternative to comparing markup will be to examine profit rates and rates of return on capital. In particular, it is generally agreed that gross profit to operating cost should be comparable. Putting the conceptual and operating details aside it should be recognized that there can be misuse of transfer pricing when there is information uncertainty. Some contracts, based on transfer pricing, need more elaborate control.

Appendix 3 Mathematical Background

157

Appendix 3

MATHEMATICAL BACKGROUND

A3.1 CONSTRAINED OPTIMIZATION Consider the problem of using a scarce resource for the production of a product. Assume that x units of the resource produces an output Y = f(x) The firm, that uses resource x to produce y can obtain a value v = g(x) in this process of production and sale. v need not always be the profit generated. Suppose the firm owns the resource and R units of it are available. The problem for the firm is to choose x to maximize v subject to x ≤ R. Referring to Fig. A.1 two situations are conceptually possible; R may be Rl or Rr. That is, the free maximum of v is not possible g(x)

V

O

Ri

X m Rr Fig. A.1

X

158


if R= Rl or the resource R will not be fully utilized if R = Rr. In the first case, the last unit of the resource used in the production of y adds something positive to the value v. In the second case, some units of resource are redundant in the context of increasing the value v. Some units of the resource R are then of zero value to the firm. A general solution to the problem may now be outlined. Define the Hamiltonian H = g(x) + λ (R – x) In this formulation, H is the total value of the resource to the firm. g(x) is the value from transforming x units of the resource to output y. (R – x), the rest of the resources may also have the potential of increasing y and adding to v. Therefore, λ can be looked upon as the marginal value of a unit of the resource. λ is usually designated as the Lagrange multiplier. However, this economic interpretation is very useful in appreciating the logic behind the method of solution. Clearly, the firm wants to maximize H. Suppose, now, that R is fully utilized to produce y. Then, R = x. In fact, when R = Rl, an additional unit of the resource still has a positive value. Then, λ > 0. It is possible, on the other hand, that Rr of the resource is available. Then (R – x) > 0 and the available resource will not be fully utilized. That is, additional units of the resource, beyond xm, will not add to the value of the firm. Consequently, it can be inferred that λ = 0. A formal statement will now be that λ > 0 if R < xm = 0 if R > xm This is the well-known Kuhn Tucker condition. To obtain the efficient value of x, to solve the constrained maximization problem, consider the free maximum of g(x). Suppose x*, the free maximum, is less than the available R. Then, there is a surplus of resources and the constraint is redundant. Nothing better than x* can be chosen. Suppose, on the other hand, that x* > R. Then, x * cannot be attained. Only a constrained maximum is possible. The x* for the constrained maximization problem is equal to R.

A3.2 DYNAMIC OPTIMIZATION Consider the following problem. Suppose a firm has an initial stock of capital x0 at time t0. Assume that it estimated the likely demand for its products at t1 > t0 and decided that it will need a stock of capital x1 to cater to the expected market demand. The firm wants to know the best route through which it can augment x0 to x1. The accompanying Fig. A.2 suggests three possible approaches to investment. In general, there is infinite choice.


159

x

B

x1 A

x0

O t0

t1

t

Fig. A.2 How will the firm be affected if it chooses one path instead of the other? Suppose, the firm increases x0 to x1 at t0. It incurs a cost. However, the production capacity cannot be fully utilized until t1. That is, the investment cannot be recovered efficiently. Suppose, instead, that it postpones the entire investment until time t1. This saves costs but the firm loses revenue for t ≤ t ≤ t1 because it cannot increase production commensurate with the expected changes in demand. Hence, there is some optimal x(t) for each t between t0 and t1. Further, the optimal choice depends on the value generated over the time horizon. Let v represent the value at t. Clearly, it depends on x as well as dx/dt. For, investments in excess of or short of requirements dictated by an increase in demand do not add value. The problem for the firm is to choose x(t) so as to t1

Max ∫ v(x, dx/dt) dt t0

subject to x(t0) = x0, and x(t1) = x1 Clearly, v explicitly depends on the information about market demand and the cost of production. This is the classic two point boundary value problem in calculus of variations. A solution to this problem can be developed by using simple calculus. Assume that x(t) is the optimal trajectory and x*(t) = x(t) + dx(t) is any neighboring trajectory. Of course, this should also satisfy the boundary conditions. That is, x*(t0) = x0 = x(t0) + dx(t0) = x0 + dx(t0) Hence, dx(t0) = 0. Similarly, dx(t1) = 0. Consider v = v[x*, dx*/dt] = v[x+dx, dx/dt + d2x/dt2] Expanding the right hand side using first order Taylor’s series yields v = v(x,dx/dt) + (dv/dx) dx + [dv/d(dx/dt)] (dx/dt)

160


Therefore, it can be inferred that t1

t1

t1

t1

t0

t0=

t0

t0

∫ v* dt = ∫ v dt + ∫(∂v/∂x) dx + ∫[∂v/∂(dx/dt)] d(dx/dt) Consider

∫ [∂v/∂(dx/dt)] d(dx/dt) dt = [∂v/∂(dx/dt)] (dx/dt) – ∫ {d[∂v/∂(dx/dt)]/dt] (dx/dt) Since dx/dt = 0 at t0 as well as t1 it follows that t1

t1

t1

t0

t0

∫ v dt = ∫ v dt + ∫ [(∂v/∂x) – d{∂v/∂(dx/dt)}/dt] (dx/dt) dt *

t0

Therefore, the derivative of t1

∫ v(x, dx/dt) dt t0

with respect to x(t) will be zero if d{∂v/∂(dx/dt)}/dt = ∂v/∂x This is usually designated as the Euler condition for optimization. By way of an example consider the problem t

Max ∫ {xα – [(dx/dt) + εx]} dt 0

subject to x0 = 0, and x1 = x* For this problem v [x,dx/dt] = xα – [(dx/dt) + εx] ∂v/∂x = αxα–1 – ε ∂v/∂(dx/dt) = – 1 Therefore, the optimal x is such that αxα-1 – ε = 0, or x = (ε/α)1/(α–1) A more general problem can now be formulated. Consider t1

Max ∫ v(x,u) dt t0


161

subject to dx/dt = f(x,u) where u is the choice of the firm. It is usually designated as a control. This problem can be solved by using the Lagrange multiplier method. Define H(x, dx/dt, u) = v(x,u) + λ [f(x,u) – dx/dt] Clearly, the maximization problem is equivalent to t1

Max ∫ H(x, dx/dt, u) dt t0

The Euler condition may now be utilized to solve the problem. Note that in the modified specification both x and u can be treated as independent choices. Further, ∂H/∂(du/dt) = 0 since H is independent of du/dt. Similarly, ∂H/∂(dx/dt) = – λ Hence, the two Euler conditions for the optimization problem are dλ/dt = – ∂H/∂x, and ∂H/∂u = 0 This is the Pontryagin’s maximum principle in its essential detail. Inequality constraints can be handled by suitable Lagrange multiplier techniques and Kuhn-Tucker conditions. Consider the following example. T

Max

∫ (xα – E) dt 0

subject to dx/dt = E – εx Construct the Hamiltonian H = xα – E + λ(E – εx) Then, by Pontryagin’s maximum principle, dλ/dt = – ∂H/∂x = – αxα–1 + λε and since H contains the expression – E (1–λ) linearly, the optimal E must be such that λ = 1 since an optimal E > 0 is sought. It now follows that αxα–1 = ε, or x = (ε/α)1/(α–1) and the corresponding E will constitute the optimal solution.

162


Note, incidentally, that this example is the same as the previous one. But Pontryagin’s maximum principle is far more general.

A3.3 NASH EQUILIBRIUM Almost all economic problems deal with exchange between two or more individuals. Invariably, each party makes an attempt to pursue maximization of its own goal. Economic theory posits that the market, as an organizational mechanism, arbitrates between these divergent interests and brings about an equilibrium in which maximum gains to a maximum number of individuals can be achieved. In practice, the bargaining process is not impersonal because only a finite number of individuals are involved in the exchange. Economic theory needs to grapple with a way of describing their strategic interaction and characterizing an efficient equilibrium. Efficient terms of contract or bargain must be specified to proceed with economic analysis. Nash equilibrium is one such concept. Consider the following. Suppose two individuals, say X and Y, are involved in the bargain. Let x and y be their respective choices. By the nature of the problem both of them recognize that though they wish to maximize their personal gains, what they can achieve depends on the choice of the other. This can be characterized by writing their respective gains as π1 (x,y) and π2 (x,y). Clearly, while choosing x the individual X must either know y or assume it to be parametric to begin with. That is, conceptualize X maximizing π1 for a given y in his choice of x. This results in x = f(y) Economic theory characterizes this as the reaction function of X or as his participation constraint. In a similar fashion, the choice of Y can be written as y = g(x) where g is not necessarily f even if it exists. Clearly, the choice (x*,y*) that satisfies both these participatory constraints is such that neither of them can do any better for the optimal choice of the other. For, exchange cannot take place unless both of them consent. Such a (x*,y*) is called a Nash equilibrium if it exists. –1

Consider the following simple numerical example to illustrate the usefulness of this concept and its implications. Let X produce an input x that Y requires to produce an output y. Assume that the production function is X = y2 X incurs a cost C = x2 to produce the input. Postulate, for simplicity, that Y sells output in the market at a price p per unit. The bargain is essentially for a price q at which X offers a unit of x to Y. The profit function of X can be represented by πx = qx – x2


163

Hence, he will demand a price q such that q = 2x This is the incentive constraint of X. Similarly, the profit for Y is πy = py – qy2 Hence, Y would be willing to offer a q such that p = 2qy This represents the incentive constraint of Y. The Nash equilibrium value of q is such that q = 2x = 2y2 = p/2y Hence,

p = 4y3; or, y = (p/4)1/3

Consequently, the choice of q is q = 2 (p/4)2/3 Observe that the total profit, or value, generated by the transaction is the sum of the profits of both X and Y. That is, N = net value of the contract = py – x2 = py – y4 The value of y that maximizes N is such that p = 4y3, or y = (p/4)1/3 This is the Nash equilibrium value. Stated briefly, whenever a unique Nash equilibrium exists, it maximizes the net value of the exchange.

A3.4 PRINCIPAL AGENT PROBLEM Economic theory acknowledges that due to strategic reasons X and Y may not reveal their reactions truthfully. It would also be expensive for the other party to obtain this information accurately if they try to do so on their own. As a result, there is always a certain degree of uncertainty in approaching these equilibrium concepts. Fundamentally, uncertainty may be external to the parties in exchange or it may be intrinsic due to their patterns of behavior. Therefore, a modification of the Nash equilibrium concept is sought. The most popular is that of Kawasaki and McMillan (1987). Assume that a principal Y commissions the production of x through an agent X. Let the agreement be that X will produce x at a cost x2/2δ where δ is a characterization of his efficiency. However, in the execution of the contract, X may find it impossible to deliver x at a cost x2/2δ due to reasons beyond his control

164


(environmental uncertainty) or, he may not be motivated to deliver all of x (usually this is designated as reneging or moral hazard). Let the output delivered be x+u where u is a random variable with E(u) = expected value of u = 0 V(u) = variance of u = σ2 Normally, in such contracts the principal agrees to pay the agent a fraction f of the revenue from the sale of x. Then, the profit for X is πx = f(x+u) – x2/2δ assuming that the market price of x is unity. The KM formulation postulates that X will be risk averse. The value of πx to him will then be Vx = fx – x2/2δ – λf2σ2 where λ > 0 represents his degree of risk aversion. Given a f, he agrees to offer x such that x = fδ KM now postulates that Y chooses f so as to maximize the net value of the contract. Note that the gain to Y can be written as πy = (1–f) (x+u) However, in general, the principal is risk neutral since he has many other avenues of diversifying his risk. Hence, he maximizes the net value of the contract N = x – x2/2δ – λf2σ2 subject to the participatory constraint of X, namely, x = fδ Write N in the form N = fδ – f2δ/2 – λf2σ2 The f that maximizes N is the efficient solution in the KM model. That is, f = δ/(δ+2λσ2) It is an increasing function of δ and a decreasing function of λ. Note that the Nash equilibrium characterization does not adequately take risk aversion into account. However, from a practical viewpoint, it is difficult to assert a priori that one of these concepts is superior to the other. Empirical evidence has not been conclusive either.

References

165

REFERENCES Abrol, D. (2005), The Question of Patentability of Biotechnological Subject Matter: Issues and Options Before the Policy Makers, Asian Biotechnology and Development Review, 7, 53-73. Alexander, M.T., O. Furrer, and D. Sudharshan (2003), A Hierarchical Framework of New Products Development: An Example from Biotechnology, European Journal of Innovation Management, 6, 48–63. Alfranca, O., and W.E. Huffman (2003), Aggregate Private R&D Investments in Agriculture: The Role of Incentives, Public Policies, and Institutions, Economic Development and Cultural Change, 51, 1–21. Allen, F., and D. Gale (1999), Diversity of Opinion and Financing of New Technologies, Journal of Financial Intermediation, 8, 68-89. Alston, J.M., and R.J. Venner (2002), The Effects of the US Plant Variety Protection Act on Wheat Generic Improvement, Research Policy, 31, 527-542. Alteiri, M.A., and P. Rosset (1999), Ten Reasons Why Biotechnology Will Not Ensure Food Security, Protect the Environment and Reduce Poverty in the Developing World, AgBioForum, 2, 155-162. Anderson, K., L.A. Jackson, and C.P. Nielsen (2004), Genetically Modified Rice Adoption: Implications for Welfare and Poverty Alleviation, World Bank Policy Research Working Paper, WPS 3380. Argyres, N.S., and J.P. Leibeskind (2000), Governance Inseparability and the Evolution of the U.S. Biotechnology Industry, Presented at the 2001 ISNIE Conference, Berkeley. Arora, A. (1994), Evaluating Technological Information and Utilizing It, Journal of Economic Behavior and Organization, 24, 91-114. Arora, A. (1996), Contracting for Tacit Knowledge: The Provision of Technical Services in Technology Licensing Contracts, Journal of Development Studies, 50, 223-256.

166


Arora, A., and A. Fosfuri (2003), Licensing the Markets for Technology, Journal of Economic Behavior and Organization, 52, 277-295. Arora, A., and A. Gambardella (1990), Complementarity and External Linkages: The Strategies of the Large Firms in Biotechnology, Journal of Industrial Economics, 38, 361-379. Audretsch, D.B., and P.E. Stephan (1996), Company-Scientist Locational Links: The Case of Biotechnology, American Economic Review, 86, 641-652. Austin, D.H. (1993), An Event Study Approach to Measuring Innovative Output: The Case of Biotechnology, American Economic Review, 83, 253-258. Austin, D.H. (2000), Patents, Spillovers, and Competition in Biotechnology, Resources for the Future. Avramovic, M. (1996), An Affordable Development? Biotechnology, Economics, and the Implications for the Third World (London: Zed Books). Balasubramanian, D., C.E.A. Bryce, K. Dharmalingam, J. Green, and K. Jayaraman (2004), Concepts in Biotechnology (Hyderabad: Universities Press). Barham, B.L., D. Jackson-Smith, and S. Moon (2000), The Adoption of rBST on Wisconsin Dairy Farms, AgBioForum, 3, 181-187. Barham, B.L., J.D. Foltz, D. Jackson-Smith, and S. Moon (2004), The Dynamics of Agricultural Biotechnology Adoption: Lessons From rBST Use in Wisconsin 1994-2001, American Journal of Agricultural Economics, 86, 61-72. Benett, R.M., Y. Ismael, U. Kambhampati, and S. Morse (2004), Economic Impact of Genetically Modified Cotton in India, AgBioForum, 7, 96-100. Berglund, D., and M. Clarke (2000), Using Research and Development to Grow State Economies, available at www.nga.org/eda/files/2000RESEARCH.pdf Betemps, S., and G.Venuprasad (2004), Contributing to Improved Health Care Through Promotion of Regional Trade Expansion, India Trade Center. Bhattacharya, J., and W.B. Vogt (2003), A Simple model of Pharmaceutical Price Dynamics, Journal of Law and Economics, 46, 599-626. Bhattacharya, S., and S. Guriev (2004), Knowledge Disclosure, Patents, and Optimal Organization of Research and Development, Discussion Paper No. TE/04/478, London School of Economics and Political Science. Bottazzi, G., G. Dosi, M. Lippi, F. Pammolli, and M. Riccaboni (2001), Innovation and Corporate Growth in the Evolution of the Drug Industry, International Journal of Industrial Organization, 19, 1161-1187. Brown. L., A. Walker, A.M. Walters, A. Harding, and L. Thurecht (2002), Funding of High Cost Biotechnology and Other Innovative Targeted Therapies Under the Pharmaceutical Benefit Scheme, University of Canberra. Chandler, A.D. (1990), Scale and Scope: The Dynamics of Industrial Capitalism (Cambridge: Harvard University Press).

References

167

Chattopadhyay, A., and T.M. Horbulyk (2004), Strategic Public Policy Towards Agricultural Biotechnology with Externalities in Developing Countries, Journal of Agricultural & Food Industrial Organization, 2, Article 6. Chaturvedi, S. (2002), Agricultural Biotechnology and New Trends in IPR Regime, Economic and Political Weekly, March 30, 1212-1222. Chaturvedi, S., and S.R. Rao (2004), Biotechnology and Development (New Delhi: Academic Foundation). Chaudhuri, S. (2004), The Pharamaceutical Industry, in S. Gokarn, A. Sen, and R.R. Vaidya (eds.) The Structure of Indian Industry (New Delhi: Oxford University Press). Chaudhuri, S., P.K. Goldberg, and P. Jia (2003), Estimating the Effects of Global Patent Protection in Pharmaceuticals: A Case of Qunolenes in India, Unpublished Manuscript. Chern, W.S., K. Rickertson, N. Tsboi, and T. Fu (2003), Consumer Acceptance and Willingness to Pay for Genetically Modified Vegetable Oil and Salmon: A Multiple Country Assessment, AgBioForum, 5, 105-122. Chung, C., and J.E. Pettigrew (1998), Economics of Soybean Biotechnology in the Livestock Industry, International Food and Agribusiness Management Review, 1, 373-385. Cockburn, I., and R.M. Henderson (2001), Scale and Scope in Drug Development: Unpacking the Advantages of Size in Pharmaceutical Research, Journal of Health Economics, 20, 1033-1057. Cohen, W., and R. Levin (1989), Innovation and Market Structure, in R. Schmalensee and R.D. Willig (eds.) Handbook of Industrial Organization (Amsterdam: Elsevier Science Publishers). Collins, W., and M. Petit (1998), Strategic Issues for National Policy Decisions in Managing Genetic Resources, World Bank Special Report 23926. Correa, C. (2001), Patents, TRIPS, and R&D Incentives: A Southern Perspective, available at www. cmhealth. org/docs/wg2_paper12. pdf Correa, C. M. (2005), The TRIPS Agreement and Transfer of Technology, in K. P. Gallaghar (ed.) Putting Development First (London: Zed Books). Cortada, J . W. , and H. E. Eraser (2005), Mapping the Future in Science Intensive Industries: Lessons from the Pharmaceutical Industry, IBM Systems Journal, 44, 163-183. Danzon, P. M. (1997), Pharmaceutical Price Regulation: National Policies vs. Global Interests (Washington: American Enterprise Institute). Danzon, P. M. (1998), The Economics of Parallel Trade, Pharmacoeconomics, 13, 293-304. Danzon, P. M., and A. Towse (2005), Theory and Implementation of Differential Pricing for Pharmaceuticals, in K. E. Mascus and J. H. Reichman (eds. ), International Public Goods and Transfer of Technology (Cambridge: Cambridge University Press). Danjon, P. M., J. McCullough, and S. Nicholson (2001), Efficiency in the Market for BiotechPharmaceutical Alliances, University of Pennsylvania.

168


Danjon, P.M., S. Nicholson, and N.S. Pereira (2005), Productivity in Pharmaceutical Biotechnology R&D: The Role of Experience and Alliances, Journal of Health Economics, 24, 317-339. Decarolis, D.M., and D.L. Deeds (1999), The Impact of Stocks and Flows of Organizational Knowledge on Firm Performance: An Empirical Investigation of the Biotechnology Industry, Strategic Management Journal, 20, 953-968. DeGarca, M., and D. Fonesca (2004), Institutions and Financial Requirements for the Emergence of Biotechnology in Brazil, Presented at the Conference on Systems of Innovation and Development of Third World Countries, Rio de Janeiro. Dhar, T., and J.D. Foltz (2003), Market Structure and Consumer Valuation in the rBST-Free and Organic Milk Markets, Working Paper FSWP 2003-2, Food System Research Group, Madison. Dickens, R.M. (2004), Biotechnology Foreseen and Forestalled: A Review Essay, Population and Development Review, 30, 755-761. DiMasi, J.A., R.W. Hansen, and H.G. Grabowski (2003), The Price of Innovation: New Estimates of Drug development Costs, Journal of Health Economics, 22, 151-185. Dixon, P., and C. Greenhalgh (2002), The Economics of Intellectual Property: A Review to Identify Themes for Future Research, Oxford Intellectual Property Research Center, Oxford. Fafchamps, M. (1992), Solidarity Networks in Preindustrial Societies: Rational Peasants with a Moral Economy, Economic Development and Cultural Change, 41, 147-174. Falck-Zapeda, J.B., G. Traxler, and R.R. Nelson (2000), Surplus Distribution from the Introduction of a Biotechnology Innovation, American Journal of Agricultural Economics, 82, 360-369. Falconi, C.A., S.W. Omamo, G.D’Toteren, and F. Iraqi (2001), An Ex ante Economic and Policy Analysis of Research on Genetic Resistance to Livestock Disease: Trypanosomosis in Africa, Agricultural Economics, 25, 153-163. Fenn, G.W., N. Liang, and S. Prowse (1995), The Economics of the Private Equity Market, Board of Governers of the Federal Reserve System. Ferreira, E.J., and L.D. Brooks (2000), On Public Versus Private Equity Placements: Pedagogical Illustrations, Financial Practice and Education, 241-248. Filson, D., and R. Morales (2005), Equity Links and Information Acquisition in Biotechnology Alliances, Journal of Economic Behavior and Organization, 59, 1-28. Fink, C. (2000), How Stronger Patent Protection in India Might Affect the Behavior of Transnational Pharmaceutical Industries, World Bank Working Paper WPS 2352. Fisher, K., and D. Byerlee (2001), Managing Intellectual Property and Commercialization in Public Research Organizations, The World Bank. Foltz, J.D., and H.H. Chang (2002), The Adoption and Profitability of rBST on Connecticut Dairy Farms, American Journal of Agricultural Economics, 84, 1021-1032.

References

169

Fontes, M. (2005), The Process of Transformation of Scientific and Technological Knowledge into Economic Value Conducted by Biotechnology Spinoffs, Technovation, 25, 339347. Frisvold, G.B., J. Sullivan, and A. Reneses (2003), Genetic Improvements in Major US Crops: The Size and Distribution of Benefits, Agricultural Economics, 28, 109-119. Fulton, M., and L. Keyowski (1999), The Producer Benefits of Herbicide Resistant Canola, AgBioForum, 2, 85-93. Fulton, M., and K. Giannakis (2001), Agricultural Biotechnology and Industry Structure, AgBioForum, 4, 137-151. Fulton, M., and K. Giannakas (2004), Inserting GM Products into the Food Chain: The Market and Welfare Effects of Different Labeling and Regulatory Regimes, American Journal of Agricultural Economics, 86, 42-60. Galhardi, R.M. (1996), Trade Implications of Biotechnology in Developing Countries: A Quantitative Assessment, Technology in Society, 18, 17-40. Ganslandt, M., K.E. Maskus, and E.V. Wong (2001), Developing and Distributing Essential Medicines to Poor Countries: The DEFEND Proposal, Unpublished Manuscript. Garca, M., and D. Fonesca (2003), Institutional and Financial Requirements for the Emergence of Biotechnology in Brazil, Unpublished Manuscript. Gerpacio, R.V. (2003), The Role of Public Sector Versus Private Sector in R&D and Technology Generation: The Case of Maize in Asia, Agricultural Economics, 29, 319-330. Gersony, N. (1998), Applying a Three-Phase, Multi-Mission Strategic Framework to Entreprenurial Biotechnology Companies, available at http://www.usabse.org/ knowledge/proceedings/1998/34-Gersony.pdf. Gesche, A., A. Huslberger, and R.M. Enstsua-Mensah (2004), Towards a Global Code of Ethics for Modern Foods and Agricultural Biotechnology, Queensland University of Technology. Giannakas, K., and M. Fulton (2002), Consumption Effects of Genetic Modification: What if the Consumers are Right?, Agricultural Economics, 27, 97-109. Giannakis, K., and A. Yiannaka (2004), The Market Potential of a New High-Oleic Soybean : An Ex Ante Analysis, AgBioForum, 7, 101-122. Giesecke, S. (2000), The Contrasting Roles of Government in the Development of Biotechnology Industry in the US and Germany, Research Policy, 29, 205-223. Gisselquist, D., and C.V. DerMeer (2001), Regulation for Seed and Fertilizer Markets, World Bank Working Paper 22817. Grabowski, H.G. (2002), Patents and New Product Development in the Pharmaceutical and Biotechnology Industries, Proceedings of a Conference on Science and Cents: Exploring the Economics of Biotechnology, available at http://www.dallasfed.org/research/ pubs/science/grabowski.pdf.

170


Graff, G., D. Zilberman, and C. Yarkin (2001), The Role of Economic Research in the Evolution of International Agricultural Biotechnology, in M. Qaim, A. Kattiger, and J. VonBraun (eds.) Agricultural Biotechnology in Developing Countries: Towards Optimizing the Benefits to the Poor (Dordrecht: Kluwer Academic Publishers). Guedj, I. (2004), Organizational Scope and Investment: Evidence from the Drug Development Strategies and Performance of Biopharmaceutical Firms, NBER Working Paper 10933. Gupta, A., and V. Chandak (2004), Agricultural Biotechnology in India: Ethics, Business, and Politics, available at http://www.iimahd.ernet.in/publications/data/ 2004_07_01anilgupta.pdf. Hall, B.H. (2002), The Financing of Research and Development, Oxford Review of Economic Policy, 16, 35-51. Hall, B.H., and J.V. Reenen (2000), How Effective are Fiscal Incentives for R&D? A Review of the Evidence, Research Policy, 29, 449-469. Hamelink, C.N., G.V. Hooijdonk, and A.P.C. Faaij (2005), Ethanol from Lignocellulosic Biomass: Techno-Economic Performance in Short-, Middle-, and Long-Term, Biomass and Bioenergy, 28, 384-410. Harhoff, D., and M. Reitzig (2004), Determinants of Opposition Against EPO Patent Grants – The Case of Biotechnology and Pharmaceuticals, International Journal of Industrial Organization, 22, 443-480. Hayhurst, R. (2005), Norway Looks to Biotech as the Oil Starts to Run Out, Biotech Focus, 10, 387-390. Henderson, R., and I. Cockburn (1996), Scale, Scope, and Spillovers: The Determinants of Research Productivity in Drug Discovery, Rand Journal of Economics, 27, 32-39. Horsch, R., and J. Montgomery (2004), Why We Partner: Collaboration Between the Private and Public Sector for Food Security and Poverty Alleviation Through Agricultural Biotechnology, AgBioForum, 7, 80-83. Huang, J., R. Hu, S. Rozelle, F. Qiao, and C.E. Pray (2001), The Case of Cotton Farmers in China, Working Paper 01-15, University of California, Davis. Available at http:// www.agecon.ucdavis.edu/facltypapers/rozelle/pdfs/cotton.pdf. Huang, J., R. Hu, C. Pray, E. Qiao, and S. Rozelle (2003), Biotechnology as an Alternative to Chemical Pesticides: A Case Study of Bt Cotton in China, Agricultural Economics, 29, 55-67. Hubbell, B.J., M.C. Marra, and G.A. Carlson (2000), Estimating the Demand for a New Technology: Bt Cotton and Insecticide Policies, American Journal of Agricultural Economics, 82, 118-132. Huffman, W.E., and R.E. Evenson (1992), Contributions of Public and Private Science and Technology to U.S. Agricultural Productivity, American Journal of Agricultural Economics, 74, 751-756.

References

171

Huygen, I., M. Veeman, and M. Lerohi (2004), Cost Implications of Alternative GM Tolerance Levels: Non-Genetically Modified Wheat in Western Canada, AgBioForum, 6, 169177. Ingley, A., J. Pavlik, and T. Smith (2002), Transgenic Protein Production as an Alternative Manufacturing Technology for Pharmaceutical Companies, Kellogg School of Management. Inouye, B. (2002), Unpacking the Pharma Biotech Engines, Polaris Institute, Ottawa. Isaac, G.E., M. Phillipson, and W.A. Kerr (2001), International Regulation of Trade in the Products of Biotechnology, Estey Center for Law and Economics in International Trade. Ishibashi, I., and T. Matsumura (2005), R&D Competition Between Public and Private Sectors, European Economic Review, to appear. Jensen, M., and Meckling, W.H. (1976), Theory of the Firm: Managerial Behavior, Agency Costs, and Ownership Structure, Journal of Financial Economics, 3, 305-360. Jorgenson, D.W. (1983), Capital Theory and Investment Behavior, American Economic Review, 53, 247-255. Just, R.E., and D.L. Heuth (1993), Multimarket Exploitation: The Case of Biotechnology and Chemicals, American Journal of Agricultural Economics, 75, 936-945. Kalaitzandokes, N. (1999), The Agricultural Knowledge System: Appropriate Roles and Interactions for the Public and Private Sectors, AgBioForum, 2, 1-4. Kapeleris, J., D. Hine, and R. Barnard (2004), Towards Definition of the Global Biotechnology Value Chain Using Cases from Australian Biotechnology SMEs, International Journal of Globalization and Small Business, 1, 79-91. Kathuria, V., and V. Tewari (2004), Venture Capitalists and Biotech Sector – Discovering the Potential, available at http://www.thehindubusinessonline.com/2004/12/29/stories/ 2004122900200900.htm Kathuria, V., and V. Tewari (2006), Vanture Capitalist’s Role in Choosing Entrepreneurs – A Study of Indian Biotechnology Industry, Presented at the First Annual Max Plank India Workshop, IISc, Bangalore. Kawasaki, S., and J. McMillan (1987), The Design of Contracts: Evidence from Japanese Subcontracting, Journal of Japanese and International Economics, 1, 327-349. Kay, N., and W. McBride (2003), Production Contracts and Productivity in the U.S. Hog Sector, American Journal of Agricultural Economics, 85, 121-133. Khush, G.S. (2004), Biotechnology: Public-Private Partnerships and IPR in the Context of Developing Countries, available at law.wustl.edu/centeris/Confpapers/ PDFWrdDoc/biotechkhush.pdf Kim, Y.J. (2003), Determinants of Inter-Firm Technology Licensing: The Case of Licensors, George Washington University, Washington.

172


Klotz-Ingram, C., S. Jaus, J. Fernando-Cornejo, and W. McBride (1999), Farm Level Production Effects Related to the Adoption of Genetically Modified Cotton for Pest Management, AgBioForum, 2, 73-84. Knight, F.H. (1933), Economic Organization (NewYork: Harper Torch Books). Krattiger, A.F. (2004), Financing the Bioindustry and Facilitating Biotechnology Transfer, IP Strategy Today, 8, 1-46. Lakshmikumaran, M., and P. Phillip (2005), Patenting Biotechnology Innovations, Asian Biotechnology and Development Review, 7, 25-41. Lanjouw, J.O. (1997), The Introduction of Pharmaceutical Product Patents in India: “Heartless Exploitation of the Poor and Suffering”, Center Discussion Paper no.775, Yale University. Lawlor, M.S. (2002), Biotechnology and Government Funding: Economic Motivation and Policy Models, available at http://www.dallasfed.org/research/pubs/science/ lawlor.pdf. Lehman, B.A. (2004), Intellectual Property and Compulsory Licensing: Pharmaceuticals and the Developing World, Twelfth Annual Conference on International Intellectual Property Law and Policy, New York. Lehrer, M., and K. Asakawa (2004), Rethinking the Public Sector: Idiosyncracies of Biotechnology Commercialization as Motors of National R&D Reform in Germany and Japan, Research Policy, 33, 921-938. Lemarie, S., and S. Marette (2002), Substitution and Complementarities in the Biotechnology and Pesticide Markets: A Theoretical Framework, in N. Kalaitzandonakes (ed.) Economic and Environmental Impacts of Agbiotech: A Global Perspective (Amsterdam: Kluwer-Plenum). Lemieux, C.M., and M.K. Wohlgenant (2001), Ex ante Evaluation of the Economic Impact of Agricultural Biotechnology: The Case of Porcine Somatotropin, American Journal of Agricultural Economics, 83, 903-914. Lence, S.H., and D.J. Hayes (2003), Impact of Biotech Grains on Market Structure and Societal Welfare, AgbioForum, 5, 85-89. Lensink, R., R.V. Molen, and S. Gangopadhayay (2003), Business Groups, Financing Constraints, and Investment: The Case of India, Journal of Development Studies, 40, 93-119. Lerner, J., and R.P. Merges (1998), The Control of Technology Alliances: An Empirical Analysis of the Biotechnology Industry, Journal of Industrial Economics, 46, 125-156. Lerner, J., M. Shane, and A. Tsai (2003), Do Equity Cycles Matter? Evidence from Biotechnology Alliances, Journal of Financial Economics, 67, 411-446. Lesser, W. (1998), Intellectual Property Rights and Concentration in Agricultural Biotechnology, AgBioForum, 1, 56-61.

References

173

Lichtenberg, E., and D. Zilberman (1986), The Econometrics of Damage Control: Why Specification Matters, American Journal of Agricultural Economics, 68, 261-273. Lichtenberg, F., and T.J. Philipson (2002), The Dual Effects of Intellectual Property Regulations: Within and Between Patent Competition in the U.S. Pharmaceutical Industry, Journal of Law and Economics, 45, 643-672. Liebskind, J.P., A.L. Oliver, L. Zucker, and M. Brewer (1996), Social Networks, Learning, and Flexibility: Sourcing Scientific Knowledge in New Biotechnology Firms, Organization Science, 7, 428-443. Lusk, J.L., L.O. House, C. Velli, S.R. Jaeger, M. Moore, B. Morrow, and W. Bruce Truill (2005), Consumer Welfare Effects of Introducing and Labeling Genetically Modified Food, Economics Letters, 88, 382-388. MacDonald, J.M., M.C. Ahearn, and D.Banker (2004), Organizational Economics in Agriculture Policy Analysis, American Journal of Agricultural Economics, 86, 744-749. Malerba, F., and I. Orsenigo (2002), Innovation and Market Structure in the Dynamics of the Pharmaceutical Industry and Biotechnology: Towards a History Friendly Model, Industrial and Corporate Change, 11, 667-703. Maltbarger, R., and N. Kalaitzandokes (2000), Direct and Hidden Costs in Identity Preserving Supply Chains, AgBioForum, 3, 236-242. Mangematin, V., S. Lemarie, J.P. Bossin, D. Catherine, F. Corolleur, R. Coronini, and M. Trommetter (2003), Development of SMEs and Heterogenity of Trajectories: The Case of Biotechnology in France, Research Policy, 32, 621-638. Maria, A. (2003), Biotechnology in India, Report by the French Embassy in India. Maria, A., and S.V. Ramani (2004), Trips and its Possible Impact on the Indian Biopharmaceutical Industry, Presented at the Conference on Market for Pharmaceuticals and the Health of Developing Nations, Toulouse. Maskus, K.E. (2001), Parallel Imports in Pharmaceuticals: Implications for Competition and Prices in Developing Countries, Report to the World Intellectual Property Organization. McGloughlin, M. (1999), Ten Reasons Why Biotechnology Will be Important to the Developing World, AgBioForum, 2, 163-174. Monsanto (2003), 2003 Monsanto Contract, Available at http://www.non-gen-farmers.com/news_details.asp?ID-613. Mizoan, S. (2000), Biotechnology, Environment, and Regulations, University of Georgia. Moon, W., and S.K. Balasubramanian (2001), Public Perceptions and Willingness to Pay a Premium for non-GM Foods in the US and UK, AgBioForum, 4, 221-231. Morse, S., R. Bennett, and Y. Ismael (2005), Comparing the Performance of Official and Unofficial Genetically Modified Cotton in India, AgBioForum, 8, 1-6. Moschini, G.C. (2001), Economic Benefits and Costs of Biotechnology Innovations in Agriculture, Working Paper 01-WP264, Iowa State University.

174


Moschini, G.C. (2001), Biotech – Who Wins? Economic Benefits and Costs of Biotechnology Innovations in Agriculture, The Estey Center Journal of International Law and Trade Policy, 2, 93-117. McMillan, G.S., F. Narin, and D.L. Deeds (2000), An Analysis of the Critical Role of Public Science in Innovation: The Case of Biotechnology, Research Policy, 29, 1-8. Newell, P. (2003), Biotech Firms, Biotech Politics: Negotiating GMOs in India, IDS Working Paper 201. Niosi, J. (2003), Alliances are Not Enough Explaining Rapid Growth in Biotechnology Firms, Research Policy, 32, 737-750. Noll, J. (2004), Comparing Quality Signals as Tools of Consumer Protection: Are Warranties Always Better than Advertisements to Promote Higher Product Quality?, International Review of Law and Economics, 24, 227-239. Noussair, C., S. Robin, and B. Ruffieux (2004), Do Consumers Really Refuse to Buy Genetically Modified Foods?, Economic Journal, 114, 102-120. OECD (2002), Genetic Inventions, Intellectual Property Rights and Licensing Practices. OECD (2004), Biotechnology for Sustainable Growth and Development. Oehmke, J. (2002), Biotechnology R&D Races, Industry Structure, and Public and Private Sector Research Orientation, AgBioForum, 4, 105-114. Oehmke, J., and C.A. Wolf (2003), Measuring Concentration in the Biotechnology R&D Industry: Adjusting for Interfirm Transfer of Genetic Materials, AgBioForum, 6, 134140. Oehmke, J., M. Mywish, and D. Weatherspoon (2001), The Effects of Biotechnology Policy on Trade and Growth, The Estey Center Journal of International Law and Trade Policy, 2, 283-306. Office of Technology Assessment (1984), Commercial Biotechnology – An International Analysis (Washington: Government Printing Office). Oh, V.M.S. (2002), Should Ethical Issues in Biotechnology Research be Decided by Physician Scientists or by Lawyers?, Annals of the Academy of Medicine, 31, 137-139. Orsenigo, L. (1989), The Emergence of Biotechnology: Institutions and Markets in Industrial Innovation (NewYork: St. Martin’s Press). Orsenigo, L., F. Pamolli, and M. Ricaboni (2001), Technological Change and Network Dynamics Lessons From the Pharmaceutical Industry, Research Policy, 30, 485-508. Palnitkar, U. (2002), Biotechnology in India, Ernst & Young, Hyderabad. Perrin, R., and L. Falginiti (2001), Dynamic Pricing of Genetically Modified Crop Traits, Presented at the 5th International Conference on Biotechnology, Science, and Modern Agriculture, Ravello, Italy. Perry, J.N. (2003), GM Crops and the Environment, available at www.rotheyhouse.co.uk.

References

175

Powell, W.W., K.W. Koput, and L. Smith-Doerr (1996), Interorganizational Collaboration and the Locus of Innovation: Networks of Learning in Biotechnology, Administrative Science Quarterly, 41, 116-145. Powell, W.W., K.W. Koput, L. Smith-Doerr, and J. Owen-Smith (1999), Network Position and Firm Performance: Organizational Returns to Collaboration in the Biotechnology Industry, in S. Andrews and D. Knoke (eds.) Networks In and Around Organizations (Greenwich: Jai Press). Pyka, A., and P.P. Saviotti (2000), Innovation Networks in Biotechnology, The SEIN Project. Qaim, M. (2001), A Prospective Evaluation of Biotechnology in Semi-Subsistence Agriculture, Agricultural Economics, 25, 165-175. Qaim, M. (2003), Bt Cotton in India: Field Trial Results and Economic Projections, World Development, 31, 2115-2127. Rai, A.K. (2005), Property Rights and Collective Action: The Case of Biotechnology Research with Low Commercial Value, in K.E. Mascus and J.H. Reichman (eds.), International Public Goods and Transfer of Technology (Cambridge: Cambridge University Press). Rai, A.K. (2005a), Open and Collaborative Research: A New Model for Biomedicine, in R. Hahn (ed.) Intellectual Property Rights in Frontier Industries: Software and Biotech (Washington: Brookings Press). Rai, A.K. (2005b), Property Rights and Collective Action: The Case of Biotechnology Research with Low Commercial Value, in K.E. Maskus and J.H. Reichman (eds.) International Public Goods and Transfer of Technology Under a Globalized Intellectual Property Regime (Cambridge: Cambridge University Press). Raina, R. (2003), Biotechnology in the National Agricultural Research System: A Case for Institutional Reform, Asian Biotechnology and Development Review, 5, 27-56. Ramani, S.V. (2002), Who is Interested in Biotech? R&D Strategies, Knowledge Base and Market Sales of Indian Biopharmaceutical Firms, Research Policy, 31, 381-398. Rangnekar, D. (2006), No Pills for Poor People? Economic and Political Weekly, 41, 409-417. Ricaboni, M., and E. Pammoli (2002), On Firm Growth in Networks, Research Policy, 31, 1405-1416. Robinson, J. (1999), Ethics and Transgenic Crops: A Review, Electronic Journal of Biotechnology, 2, 71-81. Roijakkers, N., J. Hagedoorn, and H.V. Kranenberg (2005), Dual Market Structures and the Likelihood of Repeated Ties—Evidence from Pharmaceutical Biotechnology, Research Policy, 34, 235-245. Rothaermal, F.T. (2001), Complementary Assets, Strategic Alliances, and the Incumbent’s Advantage: An Empirical Study of Industry and Firm Effects in the Biopharmaceutical Industry, Research Policy, 30, 1235-1251. Rothaermal, F.T., and D.L. Deeds (2004), Exploration and Exploitation Alliances in Biotechnology: A System of New Product Development, Strategic Management Journal, 25, 201-221.

176


Sampath, P.G. (2004), Agricultural Biotechnology: Issues for Biosafety Governance in Asian Countries, United Nations University Working Paper 2004-13. Schall, B.A. (2004), Biodiversity, Biotechnology, and the Environment, Washington University. Schimmelpfenning, D.E., C.E. Pray, and M.F. Brennan (2004), The Impact of Seed Industry Concentration on Innovation: A Study of US Biotech Market Leaders, Agricultural Economics, 30, 157-167. Shan, W., J.P. Morgan, and J. Song (1997), Foreign Direct Investment and the Sourcing of Technological Advantage: Evidence from the Biotechnology Industry, Journal of International Business Studies, 28, 267-284. Shavell, S., and T. Van Ypersele (2001), Rewards Versus Intellectual Property Rights, Journal of Law and Economics, 44, 525-547. Shoemaker, R. (2001), Economic Issues in Agricultural Biotechnology, Washington D.C., U.S. Department of Agriculture. Sonka, S., and S. Pueppke (1999), Exploring the Public’s Role in Agricultural Biotechnology Research, AgBioForum, 2, 33-36. Sporleder, J., and L.A. Moses (2004), Knowledge Capital, Intangible Assets, and Leverage: Evidence from U.S. Agricultural Biotechnology Firms, International Food and Agribusiness Management Review, 7, 26-36. Strohmenger, R., and A. Wambach (2000), Adverse Selection and Categorical Discrimination in Health Insurance Markets: The Effects of Genetic Tests, Journal of Health Economics, 19, 197-218. Stuart, T.E. (2000), Interorganizational Alliances and the Performance of Firms: A Study of Growth and Innovation Rates in a High Technology Industry, Strategic Management Journal, 21, 791-811. Subramanian, A. (1995), Putting Some Numbers on the TRIPS Pharmaceutical Debate, International Journal of Technology Management, 10, 252-265. Teece, D.J. (19890), Towards an Economic Theory of the Multiproduct Firm, Journal of Economic Behavior and Organization, 1, 237-247. Thompson, P. (2000), Food and Agricultural Biotechnology: Incorporating Ethical Considerations, Canadian Biotechnology Advisory Committee Project. Traore, N., and A. Rose (2003), Determinants of Biotechnology Utilization by the Canadian Industry, Research policy, 32, 1719-1735. Traxler, G. (2004), The Economic Impacts of Biotechnology Based Technological Innovation, ESA Working Paper No. 04-08, FAO. Vanloqueren, G., and P.A. Baret (2005), Systemic “Relevance” Assessment of Transgenic Crops: Bridging Biotechnology Regulations and Sustainable Development Policies, Unpublished Manuscript, Catholic University of Louvain. Wally, T. (2000), Outcomes Research and Biotechnology Products: A European View, Drug Information Journal, 34, 185-192.

References

177

Watal, J. (2000), Intellectual Property Rights (New Delhi: Oxford University Press). Watal, J. (1996), Introducing Product Patents in Indian Pharmaceutical Sector, World Competition, 20, 7-21. Watal, J. (2000), Pharmaceutical Patents, Prices and Welfare Losses: Policy Options for India Under WTO TRIPS Agreement, World Economy, 23, 733-752. Watal, J. (2001), Workshop on Differential Pricing and Financing of Essential Drugs, WTO Secretariat. Watal, J., and A.P. Mathai (1995), Sectoral Impact of the Uruguay Round Agreements on Developing Countries: Pharmaceutical Industry, UNIDO. Watson, J.D. (1969), The Double Helix (New York: Penguin). Weavaer, R.D. and J. Wesseler (2004), Monopolistic Pricing Power for Transgenic Crops when Technology Adopters Face Irreversible Benefits and Costs, Applied Economics Letters, 11, 969-973. Wiktorowicz, M., and R. Deber (1997), Regulating Biotechnology: A Rational Political Model of Policy development, Health Policy, 40, 115-138. Williamson, O.E. (1988), Corporate Governance and Corporate Finance, Journal of Finance, 43, 567-591. Williamson, O.E. (1991), Comparative Economic Organization: The Analysis of Discrete Structural Alternatives, Administrative Science Quarterly, 36, 269-296. Williamson, O.E. (2002), The Theory of the Firm as a Governance Structure: From Choice to Contract, Journal of Economic Perspectives, 16, 171-196. Wolf, S., and D. Zilberman (1999), Public Science, Biotechnology, and the Industrial Organization of the Agrofood System, AgBioForum, 2, 37-42. Zilberman, D., C. Yarkin, and A. Heiman (1997), Agricultural Biotechnology: Economic and International Implications, Presented at the International Agricultural Economics Association Meetings at Sacramento, California. Zilberman, D., H. Ameden, G. Graff, and M. Qaim (2004), Agricultural Biotechnology: Productivity, Biodiversity, and Intellectual Property Rights, Journal of Agricultural & Food Industrial Organization, 2, Article 3. Zucker, L.G., and M. Darby (1995), Virtuous Cycles of Productivity: Star Bioscientists and the Institutional Transformation of Industry, NBER Working Paper 5342. Zucker, L.G., and M.R. Darby (1997), Present at the Biotechnological Revolution: Transformation of Technological Identity for a Large Incumbent Pharmaceutical Firm, Research Policy, 26, 429-446. Zucker, L.G., M.R. Darby, and J. Armstrong (2002), Commercializing Knowledge: University Science, Knowledge Capture, and Firm Performance in Biotechnology, Management Science, 48, 138-153.

178


Zucker, L.G., M.R. Darby, and M.B. Brewer (1998), Intellectual Human Capital and the Birth of U.S. Biotechnology Enterprises, American Economic Review, 88, 290-306. Zylbersztajn, D., and S.G. Lazzarini (2005), On the Survival of Contracts: Assessing the Stability of Technology Licensing Agreements in the Brazilian Seed Industry, Journal of Economic Behavior and Organization, 56, 108-120.

Economics Of Biotechnology

Biotechnology

Biotechnology

Biotechnology

Biotechnology

Biotechnology

History of Modern Biotechnology

History of Modern Biotechnology

The Biotechnology of Ethanol

Microbial Fundamentals of Biotechnology

History of Modern Biotechnology

History of Modern Biotechnology

History of Modern Biotechnology

Handbook of Pharmaceutical Biotechnology

Fundamentals Of Plant Biotechnology

The Ethics of Biotechnology

History of Modern Biotechnology

Biotechnology of Cereals

Biotechnology of Extremophiles

Biotechnology of biopolymers

Biotechnology of Aroma Compounds (Advances in Biochemical Engineering Biotechnology)

Cereal biotechnology

Cereal Biotechnology

Cereal biotechnology

Pharmaceutical Biotechnology

Understanding Biotechnology

Biotechnology of Microbial Exopolysaccharides (Cambridge Studies in Biotechnology 9)

Biotechnology of Microbial Exopolysaccharides (Cambridge Studies in Biotechnology 9)

Biotechnology, 2E, Vol. 12, Modern Biotechnology

Cereal Biotechnology

Cereal Biotechnology

Economics Of Biotechnology

Biotechnology

Biotechnology

Biotechnology

Biotechnology

Biotechnology

History of Modern Biotechnology

History of Modern Biotechnology

The Biotechnology of Ethanol

Microbial Fundamentals of Biotechnology

History of Modern Biotechnology

History of Modern Biotechnology

History of Modern Biotechnology

Handbook of Pharmaceutical Biotechnology

Fundamentals Of Plant Biotechnology

The Ethics of Biotechnology

History of Modern Biotechnology

Biotechnology of Cereals

Biotechnology of Extremophiles

Biotechnology of biopolymers

Biotechnology of Aroma Compounds (Advances in Biochemical Engineering Biotechnology)

Cereal biotechnology

Cereal Biotechnology

Cereal biotechnology

Pharmaceutical Biotechnology

Understanding Biotechnology

Biotechnology of Microbial Exopolysaccharides (Cambridge Studies in Biotechnology 9)

Biotechnology of Microbial Exopolysaccharides (Cambridge Studies in Biotechnology 9)

Biotechnology, 2E, Vol. 12, Modern Biotechnology

Cereal Biotechnology

Cereal Biotechnology

Recommend Documents