Professional Documents
Culture Documents
intentionally left
blank
Copyright © 2007, New Age International (P) Ltd., Publishers
Published by New Age International (P) Ltd., Publishers
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
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
CONTENTS
PREFACE (vii)
Chapter 1 1-16
INTRODUCTION
1.1 CHANGING TECHNOLOGY ...1
1.2 MODERN BIOTECHNOLOGY ...2
1.3 ORGANIZATIONAL ISSUES ...5
1.4 PATENTS AND IPRS ...6
1.5 MARKET STRUCTURE ...9
1.6 ETHICAL AND ENVIRONMENTAL CONCERNS ...12
1.7 GOVERNMENT POLICY ...13
1.8 LOOKING AHEAD ...15
Chapter 2 17-34
ORGANIZATIONAL STRUCTURE
2.1 THE SCOPE ...17
2.2 KNOWLEDGE INTENSITY ...18
2.3 BIOTECHNOLOGY KNOWLEDGE ...20
2.4 NETWORK ORGANIZATION ...22
2.5 NATURE OF CONTRACTS ...25
2.6 SHARING FIXED COSTS ...30
2.7 ECONOMIC CONSEQUENCES ...32
2.8 OTHER ASPECTS ...33
Chapter 3 35-58
IPRS AND PATENTS
3.1 WHY PROTECTION? ...35
3.2 PATENTS AS PROTECTION ...38
(xiv) Introduction 191
Chapter 4 59-78
INVESTMENT AND FINANCING
4.1 THE ISSUES ...59
4.2 ROLE OF PUBLIC INVESTMENT ...61
4.3 R&D IN SCIENTIFIC KNOWLEDGE ...64
4.4 RISKS OF R&D ...66
4.5 COMPLEMENTARY R&D ...69
4.6 AGRICULTURAL EXTENSION SERVICES ...70
4.7 BIOPROCESSING ...72
4.8 PHYSICAL CAPITAL ...73
4.9 FINANCING CONSTRAINTS ...76
4.10 FURTHER CONSIDERATIONS ...78
Chapter 5 79-96
DEMAND, COST, AND PRODUCTIVITY
5.1 THE BACKGROUND ...79
5.2 PATTERNS OF DEMAND ...81
5.3 PRODUCTIVITY ...85
5.4 VARIABLE COSTS ...89
5.5 WELFARE EFFECTS ...93
5.6 SUMMING UP ...96
Chapter 6 97-112
MARKET STRUCTURE AND PRICING
6.1 NATURE OF MARKETS ...97
6.2 DEFINING MARKET CONCENTRATION ...99
6.3 SOURCES OF CONCENTRATION ...101
6.4 MONOPOLY POWER AND PRICING ...102
6.5 DIFFERENTIAL PRICING ...106
6.6 DYNAMIC PRICING ...110
6.7 IN RETROSPECT ...110
192 Economics of Biotechnology (xv)
Chapter 7 113-124
ETHICS AND ENVIRONMENT
7.1 ISSUES AT STAKE ...113
7.2 ETHICAL ISSUES ...114
7.3 ENVIRONMENTAL ISSUES ...115
7.4 ETHICAL ISSUES IN AGRICULTURAL BIOTECHNOLOGY ...115
7.5 ETHICAL ISSUES IN DRUG DEVELOPMENT ...118
7.6 ENVIRONMENTAL ISSUES ...119
7.7 INTERNATIONAL CONVENTIONS ...121
7.8 LESSONS AND CONTROL ...124
Chapter 8 125-135
GOVERNMENT POLICY
8.1 AN OVERVIEW ...125
8.2 SCIENTIFIC R&D ...127
8.3 SCIENTIST VS. NBF CONTRACT ...132
Chapter 9 137-141
CONCLUSION
9.1 THE TECHNOLOGY ...137
9.2 ORGANIZATIONAL ASPECTS ...138
9.3 PRODUCT PROFILES AND MARKETS ...139
9.4 NEGATIVE EFFECTS ...140
9.5 STEADY STATE ...141
APPENDICES 143-164
APPENDIX I TECHNICAL TERMS 145
APPENDIX II ECONOMIC CONCEPTS 149
APPENDIX III MATHEMATICAL BACKGROUND 157
REFERENCES 165
INDEX 179
This page
intentionally left
blank
Chapter 1
INTRODUCTION
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.
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 biotech-
nology. For instance, the ethics of using biotechnology based testing for cancer or
other severe diseases. Similarly, patenting processes was deemed inadequate be-
cause 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
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 bac-
terium 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.
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 Economics of Biotechnology
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.
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, how-
ever, substitutable to the herbicides traditionally utilized to control weeds in soy-
bean 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 Economics of Biotechnology
The economic analysis of the markets for biotechnology must reflect these features explicitly
in its modeling efforts.
Chapter 2
ORGANIZATIONAL STRUCTURE
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 Economics of Biotechnology
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.
• 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 com-
petitive 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
See, for 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.
Organizational Structure 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.
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.
Organizational Structure 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 Economics of Biotechnology
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.
• 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 orga-
nizational values. The costs of making collective decisions can be quite substan-
tial, 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 Economics of Biotechnology
• 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 organiza-
tional 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 organi-
zational dissonance rather than synergy because the organizational structures of
different NBFs may clash rather than synchronize with the goals of the large firm.
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
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.
• 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
A better 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 dis-
coveries 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
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.
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
Pm A
MC
Pw B CD
E F
MR
O
Ym 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 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 Economics of Biotechnology
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.
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 Economics of Biotechnology
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 Economics of Biotechnology
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 Economics of Biotechnology
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 con-
tracts (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.
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.
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 Economics of Biotechnology
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 Economics of Biotechnology
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
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 non-
GM 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 thera-
peutic 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
Investment and Financing 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.
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 medi-
cines 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 Economics of Biotechnology
using more expensive sources of finance. This reduces the additional value gener-
ating 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 Economics of Biotechnology
• 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.
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,
Investment and Financing 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 probabil-
ity 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 financ-
ing 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.
Investment and Financing 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 Economics of Biotechnology
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.
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
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 Economics of Biotechnology
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
∞
Max ∫ e–rt [(pE)αSβ – E – P] dt
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.
Investment and Financing 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.
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
Investment and Financing 75
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 Economics of Biotechnology
• 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 Economics of Biotechnology
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.
Chapter 5
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
Demand, Cost, and Productivity 81
84 Economics of Biotechnology
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 Economics of Biotechnology
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
Demand, Cost, and Productivity 87
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) 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 comple-
mentary 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).
(b) 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).
(c) Bt cotton is widely cultivated and its productivity extensively studied. The follow-
ing 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
Demand, Cost, and Productivity 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 technol-
ogy 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.
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 Cobb-
Douglas 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
Demand, Cost, and Productivity 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 Economics of Biotechnology
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.
Demand, Cost, and Productivity 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).
B
Un Un
Ug
O
C* c
Fig. 1
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 Economics of Biotechnology
Fig. 2
96 Economics of Biotechnology
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-to-
pay 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.
Market Structure and Pricing 97
Chapter 6
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 tech-
nologies
• 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
Market Structure and Pricing 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.
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.
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.
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 Economics of Biotechnology
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 Economics of Biotechnology
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 Economics of Biotechnology
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
A
pw 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
Max ∫ f(y1) dy1 + ∫ g(y2) dy2
0 0
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).
Market Structure and Pricing 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 prerequi-
site. In addition, the MNC needs a guarantee that its proprietary technology is safe
and that low priced drugs will not be exported. Appropriate institutional arrange-
ments 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 Economics of Biotechnology
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.
6.7 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.
Market Structure and Pricing 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.”
This page
intentionally left
blank
Chapter 7
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.
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.
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.”
Ethics and Environment 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 Economics of Biotechnology
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 non-
GM 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.
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.)
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.
• 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 ma-
terial 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).
Consider 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.
Ethics and Environment 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, hazard-
ous. 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 com-
munity dynamics. Sometimes whole sections of chromosomes are transferred in-
troducing 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.
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 sat-
isfied 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
Ethics and Environment 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 Economics of Biotechnology
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.
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 Economics of Biotechnology
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 invest-
ment 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 atmo-
sphere 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 accel-
erated 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 invest-
ment. 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.
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 com-
parison 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 in-
vestments, 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 agricul-
tural 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 availabil-
ity. 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 Economics of Biotechnology
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).
CONCLUSION
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.
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.
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.
TECHNICAL TERMS
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.
This page
intentionally left
blank
Appendix 2
ECONOMIC CONCEPTS
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
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.
• 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 capital-
labor 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 utiliza-
tion. The sunk costs are much higher. Thus, the two decisive figures in determining
152 Economics of Biotechnology
costs were rated capacity and throughput (the actual amount of output). The through-
put 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 trans-
port 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 capaci-
ties. 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 increas-
ing 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, process-
ing 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.
Appendix 2 Economic Concepts 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 require-
ments 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 produc-
tion 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 pro-
duce 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 ex-
ample, if one chemical is made from a byproduct of another, then increased pro-
duction 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 Economics of Biotechnology
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.
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.
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
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)
O
Ri X m Rr X
Fig. A.1
158 Economics of Biotechnology
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.
x1 B
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 Economics of Biotechnology
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
∫ v dt = ∫ v dt + ∫ [(∂v/∂x) – d{∂v/∂(dx/dt)}/dt] (dx/dt) dt
*
t0 t0 t0
Therefore, the derivative of
t1
∫ v(x, dx/dt) dt
t0
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
Appendix 3 Mathematical Background 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 Economics of Biotechnology
Note, incidentally, that this example is the same as the previous one. But Pontryagin’s
maximum principle is far more general.
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.
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
Appendix 3 Mathematical Background 163
(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 Economics of Biotechnology
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 Biotech-
Pharmaceutical Alliances, University of Pennsylvania.
168 Economics of Biotechnology
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, 169-
177.
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 Economics of Biotechnology
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 Economics of Biotechnology
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, 134-
140.
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 Economics of Biotechnology
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 Economics of Biotechnology
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.