Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Pharmaceuticals and Biotechnology1 IAIN COCKBURN University of British Columbia and National Bureau of Economic Research REBECCA HENDERSON Massachusetts Institute of Technology and National Bureau of Economic Research LUIGI ORSENIGO Universita Commerciale Luigi Bocconi GARY P. PISANO Harvard Business School The pharmaceutical industry has been by almost any measure outstandingly successful. It is one of the few high-technology industries that American firms have dominated almost since its inception, and it is one of the few in which American firms continue to have an indisputable lead. During the 1980s and 1990s, double-digit rates of growth in earnings and return on equity were the norm for most pharmaceutical companies, and the industry as a whole ranked among the most profitable in the United States.2 To what degree can this success be viewed as a triumph of U.S. public policy? This question cannot be answered definitively because the roots of the industry's success are complex, and causality cannot be attributed to any single factor with precision. A plausible case can be made, however, that in the case of the pharma- ceutical industry public policy has played a particularly important role in contrib- uting to the global success of American firms. Public policy has always played an enormously important role in shaping the pharmaceutical industry in the United States. On the supply side, public funding for health-related research supplies both new knowledge and highly trained em- ployees to pharmaceutical firms. New drugs can be sold only with the explicit ~ This paper draws on an ongoing program of work exploring the determinants of research produc- tivity in the pharmaceutical industry, funded by the Program on the Pharmaceutical Industry and the Center for Innovation in New Product Development under NSF Cooperative Agreement Number EEC-9529140. Their support is gratefully acknowledged. 2 Note that these figures are based on accounting rates of return. Figures that are recalculated to account for heavy spending by the industry on advertising and research suggest that rates of return were actually somewhat lower than the accounting figures would suggest. 363
364 U.S. INDUSTRYIN2000 approval of the federal government, an approval that is typically granted only after potential candidates have passed a series of rigorous clinical tests. On the demand side, the federal government has an enormous impact on the market for new drugs, both by virtue of its role as a major consumer of drugs through its funding of Medicare and Medicaid and through its regulation of how pharmaceu- tical firms may advertise and market their products. Public policy toward the protection of intellectual property also has a very significant effect because the pharmaceutical industry is one of the few in which intellectual property protec- tion plays a central role in product market competition. Public policy also plays a more indirect but nevertheless important role in shaping the industry through its effects on both the labor markets and the markets for new capital, particularly the market for venture capital. Taken together these policy instruments have been instrumental in building an exceptionally strong industry. Before World War II public policy played little role in shaping the industry's evolution. In the postwar period, however, the federal government's heavy investment in basic research, its support of a strong intellectual property regime, and its imposition in 1962 of tight product approval criteria combined to help create an industry whose leading firms were not only increasingly able to translate scientific advances into effective therapies but also well positioned to exploit the new opportunities opened up by the revolution in molecular biology. The molecular biology revolution made the role of public policy in shaping the industry even more important. The revolution was initially based in the uni- versities, and the size and strength of the American commitment to health-related research ensured that U.S. universities were at the frontier of the new science. But public policy also proved very important in shaping the ways in which the new science affected the pharmaceutical industry. The industry used molecular biol- ogy in two forms as a new process technology in making large molecular weight drugs and as a new research tool in searching for more conventional, small mo- lecular weight drugs. The vast majority of drugs prescribed today are "small" molecular weight drugs relatively small, simple molecules that can be synthe- sized in a test tube and that often can be taken orally. "Large" molecular weight drugs, are much, much larger. They usually cannot be directly synthesized but must be "grown" or "expressed" and cannot usually be taken orally. The first trajectory was, at least initially, unambiguously competence de- stroying and was most effectively exploited by new entrants. In the United States an institutional environment that not only supported universities in making the fundamental breakthroughs necessary to exploit the new science but also sup- ported their translation into small, flexible, aggressively funded new firms led to the birth of an entire industry segment, the biotechnology firms. At the same time the second trajectory the adoption of the tools of biotech- nology as search tools proved to be competence destroying for those pharma- ceutical firms that had not fully made the transition to "science-based" or "ratio
PHARMACEUTICALS AND BIOTECHNOLOGY 365 nal" drug discovery. It thus reinforced the dominance of the large scientifically based firms, a large majority of which were located in the United States and which owed much of their success to the U.S. public policy regimes of the 1970s and 1980s. The remainder of this chapter expands on this argument. We begin by dis- cussing the evolution of drug discovery research technology and the role of pub- lic policy in shaping U.S. success prior to the molecular biology revolution. We suggest that a number of public policies played instrumental roles in building an American industry that was among the strongest in the world. The third section lays the foundation for a discussion of the impact of public policy on the industry in the wake of the revolution in molecular biology. We suggest that molecular biology as a process technology "biotechnology" was competence destroying for the vast majority of established firms, while molecular biology used as a research tool was competence enhancing for those firms that had already made a transition to science driven, or more "rational" drug discovery. Further, we describe the ways in which the revolution shaped the evolution of the industry across the world, focusing particularly on the ways in which response in the U.S. was very different, and in many ways much more effective, than responses in Europe and Japan. Finally, we discuss the role of public policy in shaping this differential response. We suggest (as have many before us) that public policy was instrumental in laying the foundations for the explosion of vibrant "new biotechnology firms" that characterized the American response to "biotechnol- ogy." We also suggest that the ability of many of the established American firms to respond effectively to the challenges of the new science was predicated on skills that they had developed during the previous era, skills developed partly in response to an environment largely shaped by American public policy. THE PHARMACEUTICAL INDUSTRY BEFORE THE MOLECULAR BIOLOGY REVOLUTION The history of the pharmaceutical industry can be usefully divided into three major epochs. The first, corresponding roughly to the period 1850-1945, was one in which little new drug development occurred and in which the minimal research that was conducted was based on relatively primitive methods. The large-scale development of penicillin during World War II marked the emergence of the second period of the industry's evolution. This period was characterized by the institution of formalized in-house R&D programs and relatively rapid rates of new drug introduction. During the early part of the period the industry relied largely on so-called "random" screening as a method for finding new drugs, but in the 1970s the industry began a transition to "guided" drug discovery or "drug development by design," a research methodology that drew heavily on advances in molecular biochemistry, pharmacology, and enzymology. The third epoch of the industry had its roots in the 1970s but did not begin to flower until quite
366 U.S. INDUSTRYIN2000 recently as the use of the tools of genetic engineering in the production and dis- covery of new drugs has come to be more widely dispersed. Understanding the evolution of the industry in the first two periods is impor- tant because their history illustrates the role public policy played in shaping the industry and because both the industrial and institutional structure of the industry and the organizational capabilities of individual firms were molded during these early periods. Early History By almost any measure pharmaceuticals is a classic high-technology or science-based, industry. Yet drugs are as old as antiquity. For example, the Ebers Papyrus lists 811 prescriptions used in Egypt in 550 B.C. Eighteenth cen- tury France and Germany had pharmacies where pharmacists working in well- equipped laboratories produced therapeutic ingredients of known identity and purity on a small scale. Mass production of drugs dates back to 1813, when J.B. Trommsdof opened the first specialized pharmaceutical plant in Germany. Dur- ing the first half of the nineteenth century, however, standardized medicines for treating specific conditions were virtually nonexistent. A patient instead would be given a customized prescription that would be formulated at the local phar- macy by hand. The birth of the modern pharmaceutical industry can be traced to the mid- nineteenth century with the emergence of Germany and Switzerland as leaders of the new synthetic dye industry. This was due in part to the strength of German universities in organic chemistry and in part to Basel's proximity to the leading silk and textile regions of Germany and France. During the 1880s dyestuffs and other organic chemicals were discovered to have medicinal effects, such as antiception. It was thus initially Swiss and German chemical companies such as Ciba and Sandoz, Bayer and Hoescht, leveraging their technical competencies in organic chemistry and dyestuffs, that began to manufacture drugs, usually based on synthetic dyes, later in the nineteenth century. For example, the German com- pany Bayer was the first to produce salicylic acid (aspirin) in 1883. Mass production of pharmaceuticals also began in the United States and the United Kingdom in the later part of the nineteenth century, but the pattern of development was quite different from that of Germany and Switzerland. Whereas Swiss and German pharmaceutical activities tended to emerge within larger chemical-producing enterprises, the United States and the United Kingdom wit- nessed the birth of specialized pharmaceutical producers such as Wyeth (later American Home Products), Eli Lilly, Pfizer, Warner-Lambert, and Burroughs- Wellcome. Up until World War I German companies dominated the industry, producing approximately 80 percent of the world's pharmaceutical output. In the early years the pharmaceutical industry was not tightly linked to for- mal science. Until the 1930s, when sulfonamide was discovered, drug companies
PHARMACEUTICALS AND BIOTECHNOLOGY 367 undertook little formal research. Most new drugs were based on existing organic chemicals or were derived from natural sources such as herbs, and little formal testing was done to ensure either safety or efficacy. Harold Clymer, who joined SmithKline in 1939, noted: [Y]ou can judge the magnitude of [SmithKline's] R&D at that time by the fact I was told I would have to consider the position temporary since they had already hired two people within the previous year for their laboratory and were not sure that the business would warrant the continued expenditure. World War II and wartime needs for antibiotics marked the drug industry's transition to an it&D-intensive business. Alexander Fleming discovered penicil- lin and its antibiotic properties in 1928. Throughout the 1930s, however, it was produced only in laboratory-scale quantities and was used almost exclusively for experimental purposes. With the outbreak of World War II, the U.S. government organized a massive research and production effort that focused on commercial production techniques and chemical structure analysis. More than 20 companies, several universities, and the Department of Agriculture took part. Pfizer, which had production experience in fermentation, developed a deep-tank fermentation process for producing large quantities of penicillin. This system led to major gains in productivity and, more important, laid out an architecture for the process and created a framework in which future improvements could took place. The commercialization of penicillin marked a watershed in the industry's development. Due partially to the technical experience and organizational capa- bilities accumulated through the intense wartime effort to develop penicillin, as well as to the recognition that drug development could be highly profitable, phar- maceutical companies embarked on a period of massive investment in R&D and built large-scale internal R&D capabilities. At the same time there was a very significant shift in the institutional structure surrounding the industry. Whereas before the war public support for health-related research had been quite modest, after the war it boomed to unprecedented levels, helping to set the stage for a period of great prosperity. Golden Age for the Industry: 1950-1990 The period from 1950 to 1990 was a golden age for the pharmaceutical in- dustry, as the industry in general, and particularly the major U.S. players, firms such as Merck, Eli Lilly, Bristol-Myers, and Pfizer, grew rapidly and profitably. R&D spending literally exploded and with them came a steady flow of new drugs. Drug innovation was a highly profitable activity during most of this period. Statman (1983), for example, estimated that accounting rates of return on new drugs introduced between 1954 and 1978 averaged 20.9 percent (compared to a cost of capital of 10.7 percent). Between 1982 and 1992, firms in the industry grew at an average annual rate of 18 percent.
368 U.S. INDUSTRYIN2000 Several factors supported the industry's high average level of innovation and economic performance. One was the sheer magnitude of both the research oppor- tunities and the unmet needs. In the early postwar years, there were many physi- cal ailments and diseases for which no drugs existed. In every major therapeutic category, from pain killers and anti-inflammatories to cardiovascular and central nervous system products, pharmaceutical companies faced an almost completely open field. Before the discovery of penicillin, very few drugs effectively cured diseases. Faced with such a "target-rich" environment but very little detailed knowl- edge of the biological underpinnings of specific diseases, pharmaceutical compa- nies invented an approach to research now referred to as "random screening." Under this approach, natural and chemically derived compounds are randomly screened in test tube experiments and laboratory animals for potential therapeutic activity. Pharmaceutical companies maintained enormous "libraries" of chemi- cal compounds and added to their collections by searching for new compounds in places such as swamps, streams, and soil samples. Thousands, if not tens of thousands, of compounds might be subjected to multiple screens before research- ers honed in on a promising substance. Serendipity played a key role because the "mechanism of action" of most drugs, the specific biochemical and molecular pathways that were responsible for their therapeutic effect, was generally not well understood. Typically, researchers had to rely on the use of animal models as screens. For example, researchers injected compounds into hypertensive rats or dogs to explore the degree to which they reduced blood pressure. Under this regime it was not uncommon for companies to discover a drug to treat one dis- ease while searching for a treatment for another. Although random screening may seem inefficient, it worked extremely well for many years and continues to be widely employed. Several hundred chemical entities were introduced in the 1950s and 1960s, and several important classes of drug were discovered in this way, including a number of important diuretics, all of the early vasodilators, and several centrally acting agents, including reserpine and guanethidine. In the early 1970s, the industry also began to benefit more directly from the explosion in public funding for health-related research that followed the war. Between 1970 and 1995, for example, support for the National Institutes of Health (NIH), the agency through which the vast majority of federal support for health- related research is channeled, increased nearly 200 percent in real terms, to over $8.8 billion a year or 36 percent of the federal nondefense research budget, an amount roughly equal to the total research expenditure of all the U.S. pharmaceu- tical firms (Figure 1~. Before the 1970s publicly funded research was probably most important to the industry as a source of knowledge about the etiology of disease. From the middle 1970s on, however, substantial advances in physiology, pharmacology, enzymology, and cell biology the vast majority stemming from publicly funded
PHARMACEUTICALS AND BIOTECHNOLOGY 10.00 8.00 6.00 4.00 2.00 0.00 369 Nominal I Deflated - - ~_ ~' ~' T ~ 1970 1975 1980 1985 1990 1995 Year FIGURE 1 NIH total appropriations (billions of dollars). research led to enormous progress in understanding the mechanism of action of some existing drugs and the biochemical and molecular roots of many diseases. This new knowledge made it possible to design significantly more sophisticated screens. By 1972, for example, the structure of the renin angiotensive cascade, one of the systems within the body responsible for the regulation of blood pres- sure, had been clarified by publicly funded researchers, and by 1975 several com- panies had drawn on this research in designing screens for hypertensive drugs (Henderson and Cockburn, 1994~. These firms could replace ranks of hyperten- sive rats with precisely defined chemical reactions. Instead of requesting "some- thing that will lower blood pressure in rats," pharmacologists could request "something that inhibits the action of the angiotensin 2 converting enzyme." In turn, the more sensitive screens made it possible to screen a wider range of compounds. Before the late 1970s, for example, it was difficult to screen the natural products of fermentation, a potent source of new antibiotics, in whole animal models. The compounds were available in such small quantities or trig- gered such complex mixtures of reactions in living animals that it was difficult to evaluate their effectiveness. The use of enzyme systems as screens made it much easier to evaluate these kinds of compounds. It also triggered a "virtuous cycle" in that the availability of drugs whose mechanisms of action were well known made possible significant advances in the medical understanding of the natural history of several key diseases, advances that in turn opened up new targets and opportunities for drug therapy (Gambardella, 1995; Maxwell and Eckhardt,1990~. The industry's increasing reliance on advances in fundamental science dra- matically increased the importance of public sector research in shaping industry productivity. Publicly funded research was important for several reasons. First, it provided the "raw knowledge" that undergirded many key discoveries. Table 1 illustrates the increasingly close relationship between the public and private sec- tors during the period. It summarizes detailed case histories of the discovery and development of 21 drugs identified by two leading industry experts as "having had the most impact upon therapeutic practice" between 1965 and 1992. Only 5
370 a' 3 a' C) a' a' Cq a' ·_4 3 Cq be Cal a' o o a' a' a' o o Cq ·_4 C) ca ~.0 ~ ~1 ~ ~ o o ~ ~b H.~ - C) · ~ o ~ ca o=o ~ - C) o.~> C) = o .0 Cal EM C) .e ca .5 .~ o o ca o o C) o ca . . ;^ C) ca · ~ o o ca o M ~- ~ ~ ~ ~ ~ ~ cM oo oo ~ oo oo oo ~ zz~zzz ~- cM cM ~ oo ~ ~ ~ ~ ~ z~zzz~ oo ~ cM o ca ca ·t R R -~ ~ -= ~ =-= 03 ~ 2 ~ ~ ~ ~ ~ ¢ V ~ ¢ ~ O R . - ~ ~ u R R o R ~ ~ ~ O V ~ ~ ~ ~ Z ~ .~ O O C=) ca ca .~ ca 0F 00 .~ ca R .5 ~ ~ oo O oo oo oo ~ ~ oo oo oo ~ ~ ~ zzzzzz zzz M ~ ~ oo ~ ~ ~ oo ~ oo oo oo ~ oo ~z~ ca o .= 4= ~) oo ~t o ~\ ~\ v ~ ~ ~ ~ ~ ~ ~ ~ ~ ~o 4D ~ ~ ~ ~o =~mm~Z=2 ~ o R ~ ~ ~ O ~ ~ ~ .O C) ca so~ ca moom.o oo~oo ~ ~zz =~.~ - - ~ ~ ~ ~ o o - - ~ ~ ~ ~o O-~ 5 ~ ~ ¢v ~ R ~ ~ ~° ~ ~m .O .= ~ ~ ~ 2 ' A ,~ ~ ~ ~ ~ 9 s s s s s 3 ~ ~ ~ ~ ~ , i!~3.~..2 o~g s Y.Y-2 a 5 5 ~ 9 _ 5 6, _ s 2 s a 8 5 ~ ~ Ya 5~ ~ ,~ ~ ~ ~ ~ . 5 i s 8 ~ 5 5 . . .. ~ z ~
PHARMACEUTICALS AND BIOTECHNOLOGY 371 of these drugs, or 24 percent, were developed with essentially no input from the public sector. These data suggest that public sector research has become more important to the private sector over time. Table 1 groups the drugs into three classes according to the research strategy by which they were discovered: those discovered by "random screening," those discovered by "mechanism-based screening," and those discovered through fun- damental scientific advances. Broadly speaking, the degree of reliance on the public sector for the initial insight increases across the three groups, and as the industry has moved to a greater reliance on the second and third approaches, so too has the role of the public sector increased. The public sector was also impor- tant in providing highly trained employees for the private sector and in helping to sustain a "research ethos" within those private firms that aggressively embraced the new techniques and that was highly productive. Efforts to measure the rate of return to public research have been very con- tentious and dogged by a variety of difficult practical and conceptual problems (Griliches, 1994; Ward and Dranove, 1995~. However in a recent study Cockburn and Henderson (1998) suggest that differences in the effectiveness with which pharmaceutical firms access the upstream pool of knowledge created by public science correspond to differences in research productivity of as much as 30 per- cent. Zucker et al. (forthcoming) find very similar results in their study of the role of the public sector in supporting the growth of the newly founded biotech- nology firms. Although any estimate of this type must be treated with great caution, these results are consistent with the hypothesis that public sector research has been critically important to the industry's health. Most intriguingly from a public policy perspective, these authors found that a firm's connectedness to the public sector, measured by the coauthorship of scientific papers across institutional boundaries, is closely related to several other factors that enhance the productiv- ity of privately funded pharmaceutical research. These include the number of "star scientists" employed by the firm and the degree to which the firm uses a researcher' s reputation among his or her peers as a criterion for promotion.3 These results are consistent with the hypothesis that the ability to take advantage of knowledge generated in the public sector requires investment in a complex set of activities that taken together change the nature of private sector research. Thus they raise the possibility that the ways in which public research is conducted may be as important as the level of public funding. Despite their apparent importance, these new research techniques were not uniformly adopted across the industry. For any particular firm, the shift in the technology of drug research from "random screening" to one of "guided" discov- ery or "drug discovery by design" depended critically on the ability to take ad 3 The use of coauthoring behavior to measure connectedness to the public sector was pioneered by Zucker et al. (1997) in their study of the emergence of new biotechnology firms.
372 U.S. INDUSTRYIN2000 vantage of publicly generated knowledge (Cockburn and Henderson, 1996; Gambardella, 1995) and of economies of scope within the firm (Henderson and Cockburn, 1996~. Smaller firms, those farther from the centers of public re- search, and those that were most successful with the older techniques of rational drug discovery appear to have been much slower to adopt the new techniques than were their rivals (Cockburn et al.,1998; Gambardella, 1995; Henderson and Cockburn, 1994~. There was also significant geographical variation in adoption. The larger firms in the United States, the United Kingdom, and Switzerland were among the pioneers of the new technology, but Japanese and other European firms have been slow in responding to the opportunities afforded by the new science. In general, although the pharmaceutical industry is global in nature, companies from the United States, Switzerland, Germany, and the United King- dom have dominated in the postwar period. French and Italian firms have not played major international roles. Japan is the second largest pharmaceutical mar- ket in the world and is dominated by local firms, largely for regulatory reasons; but Japanese firms have to date been conspicuously absent from the global indus- try. Only Takeda ranks among the top 20 pharmaceutical firms in the world, and until relatively recently the innovative performance of Japanese pharmaceutical firms has been weak compared with their U.S. and European competitors. INSTITUTIONAL ENVIRONMENTS Institutional forces have shaped the industry in the "pre-biotechnology world," providing powerful inducements to innovation. From its inception, the evolution of the pharmaceutical industry has been tightly linked to the structure of national institutions. The pharmaceutical industry emerged in Switzerland and Germany in part, because of strong university research and training in the rel- evant scientific areas. German universities in the nineteenth century were leaders in organic chemistry, and Basel, the center of the Swiss pharmaceutical industry, was the home of the country's oldest university, long a center for medicinal and chemical study. In the United States the government's massive wartime invest- ment in the development of penicillin profoundly altered the evolution of Ameri- can industry. In the postwar era, the institutional arrangements in four key areas, the public support of basic research, intellectual property protection, procedures for product testing and approval, and pricing and reimbursement policies, have strongly influenced both the process of innovation directly and the economic returns, and thus the incentives, for undertaking such innovation. Public Support for Health-Related Research Nearly every government in the developed world supports publicly funded health-related research, but countries vary significantly in both the level of sup- port offered and in the ways in which it is spent. As reviewed earlier, public
PHARMACEUTICALS AND BIOTECHNOLOGY 373 spending on health-related research in the United States is now the second larg- est item in the federal research budget after defense and is roughly equivalent to the research budget of the entire U.S. pharmaceutical industry. Both qualitative and quantitative evidence suggests that this spending has had a significant effect on the productivity of those large U.S. firms that were able to take advantage of it (Cockburn and Henderson, 1998; Maxwell and Eckhardt, 1990; Ward and Dranove, 1995~. Public funding of biomedical research also increased dramatically in Europe in the postwar period, although the United Kingdom spent considerably less than Germany or France, and total spending did not approach American levels (Table 2~. Moreover, the institutional structure of biomedical research in continental Europe evolved quite differently from its evolution in the United States and the United Kingdom, creating an environment in which science is far less integrated with medical practice. Science does not in general confer the same status within the medical profes- sion in continental Europe as it does in the United Kingdom or the United States. Traditionally the medical profession has had less scientific preparation than is common in either the United States or the United Kingdom, and medical training and practice have focused less on scientific methods per se than on the ability to use the results of research. Moreover doctorates in the relevant scientific disci- plines have been far less professionally oriented. Historically the incentives to engage in patient care at the expense of research have been very high. France and TABLE 2 Breakdown of National Expenditures on Academic and Related Research by Main Field, 1987a Expenditure (1987 million dollars) U.K. FRG France Netherlands U.S. Japan Averageb Engineering 436 505 359 112 1966 809 14.3% 15.6% 12.5% 11.2% 11.7% 13.2% 21.6% Physical sciences 565 1015 955 208 2325 543 21.2% 20.2% 25.1% 29.7% 21.7% 15.6% 14.5% Life sciences 864 1483 1116 313 7285 1261 36.3% 30.9% 36.7% 34.7% 32.7% 48.9% 33.7% Social sciences 187 210 146 99 754 145 6.0% 6.7% 5.2% 4.6% 10.4% 5.1% 3.9% Arts and humanities 184 251 218 83 411 358 6.8% 6.6% 6.2% 6.8% 8.6% 2.8% 9.6% Other 562 573 418 143 2163 620 15.6% 20.1% 14.2% 13.0% 14.9% 14.5% 16.6% Total 2,798 4,037 3,212 958 14,904 3,736 Expenditure data are based on OECD "purchasing power parities" for 1987 calculated in early 1989. bThis represents an unweighted average for the six countries (i.e., national figures have not been weighted to take into account the differing size of countries). Source: Irvine et al. (1990, p. 219).
374 U.S. INDUSTRYIN2000 Germany have only recently implemented systems designed to free clinicians from their financial ties to patient-related activities, and, partly in consequence, within universities medically oriented research has played a marginal role com- pared with patient care. The organizational structure of medical schools tends to reinforce these dif- ferences. In continental Europe medical schools and hospitals are part of a single organizational entity, whereas in the United States and the United Kingdom medi- cal schools are generally independent of hospital administrations. This status allows them to give clear priority to their intrinsic goals of research and teaching. In principle, the European system should have some advantages. In practice, however, patient care has tended to absorb the largest fraction of time and finan- cial resources. The weakness of the research function within hospitals in continental Europe is one of the reasons that several governments have decided to concentrate bio- medical research in national laboratories rather than in medical schools. How- ever, the separation of the research from daily medical practice may have had negative effects on both the quality of the research and on the rate at which it diffuses into the medical community. Protection of Intellectual Property In many industries, successful new products quickly attract imitators. But rapid imitation of new drugs is difficult in pharmaceuticals. One reason is that pharmaceuticals has historically been one of the few industries where patents provide solid protection against imitation. Because small variants in a molecule's structure can drastically alter its pharmacological properties, potential imitators often find it hard to work around the patent. Although other firms might under- take research in the same therapeutic class as an innovator, the probability of their finding another compound with the same therapeutic properties that did not in- fringe on the original patent is usually quite small.4 The scope and efficacy of patent protection has varied significantly across countries, however. The United States and most European countries have pro- vided relatively strong patent protection in pharmaceuticals. In contrast, until recently only process technologies could be patented in Japan and in Italy; not until 1976 in Japan and 1978 in Italy did patent law offer protection for pharma- ceutical products. As a result, Japanese and Italian firms tended to avoid product R&D and to concentrate instead on finding novel processes for making existing molecules. 4This is not always the case. The history of the discovery of the ACE inhibitors provides a notable exception.
PHARMACEUTICALS AND BIOTECHNOLOGY Procedures for Product Approval 375 Pharmaceuticals are regulated products. Procedures for approval have a pro- found impact on both the cost of innovating and on firms' ability to sustain mar- ket positions once their products have been approved. Since the early 1960s most countries have steadily increased the stringency of their approval processes. The United States and the United Kingdom have adopted by far the most stringent approval process of any industrial country, followed by the Netherlands, Switzer- land, and the Scandinavian countries. Germany and especially France, Japan, and Italy have historically been much less demanding. In the United States, the 1962 Kefauver-Harris Amendments were passed after the thalidomide disaster. This law introduced a proof-of-efficacy require- ment for approval of new drugs and established regulatory controls over the clini- cal testing of new drug candidates. Specifically, the amendments required firms to provide substantial evidence of a new drug' s efficacy based on "adequate and well controlled trials." As a result, after 1962 the Food and Drug Administration (FDA) shifted from being an evaluator of evidence and research findings at the end of the R&D process to an active participant in the process itself (Grabowski and Vernon, 1983~. The effects of the 1962 law on innovative activities and market structure have been the subject of considerable debate.5 The law certainly led to large increases in the resources devoted to obtaining approval of a new drug applica- tion (NDA), and it probably caused sharp increases in both R&D costs and in the gestation times for new chemical entities (NCEs). As a consequence, the annual rate of NCE introduction declined sharply and there was a lag in the introduction of significant new drugs therapies in the United States compared with Germany and the United Kingdom. However, the creation of a stringent drug approval process in the United States may have also helped reduce rates of entry into the industry and thus may have indirectly served to protect the margins that attracted further investment in research. Although the process of development and ap- proval increased costs, it significantly increased barriers to imitation, even after patents expired. Until the Waxman-Hatch Act was passed in 1984, generic ver- sions of drugs that had gone off patent still had to undergo extensive human clinical trials before they could be sold in the U.S. market, so that it might be years before a generic version appeared even after a key patent had expired. In 1980 generics held only 2 percent of the U.S. drug market. The institutional environment surrounding drug approval in the United King- dom was quite similar to that in the United States. Regulation of product safety, which began in 1964 and was tightened with passage of the Medicine Act in 1971, relied heavily from the beginning on formal academic medicine, in particu- lar on well-controlled clinical trials, to demonstrate the safety and efficacy of new 5See, for example, Chien (1979) and Peltzman (1974).
376 U.S. INDUSTRYIN2000 drugs. Extensive documentation and high academic standards were required of all submissions. The Committee on Safety of Drugs (CSD), known as the Com- mittee on Safety of Medicines (CSM) after 1971, comprised independent aca- demic experts, voluntarily organized and supported by the industry. Based on strong cooperation among the regulatory body, the industry, and academe, the British system effectively imposed very high standards on the industry (Davies, 1967; Hancher, 1990; Thomas, 1994; Wardell, 1978~. As in the United States, the introduction of a tougher regulatory environment in the United Kingdom in 1971 was followed by a sharp fall in the number of new drugs launched in Britain and a shakeout of the industry. Several smaller, weaker firms exited the market, and the proportion of minor local products launched into the British market shrunk significantly. The strongest British firms gradually reoriented their R&D activi- ties toward the development of more ambitious, global products (Thomas, 1994~. Japan represented a very different case from either the United States or the United Kingdom. Before 1967 any drug approved for use in another country and listed in an accepted official pharmacopoeia could be sold in Japan without going through additional clinical trials or regulatory approval. At the same time non- Japanese firms were prohibited from applying for drug approval. Thus Japanese firms were simultaneously protected from foreign competition and given strong incentives to license products that had been approved overseas. Under this re- gime the primary technology strategy for Japanese pharmaceutical companies became the identification of promising foreign products to license (Reich, l990~. The Structure of the Health Care System and Systems of Reimbursement Perhaps the biggest differences in institutional environments across coun- tries was in the structure of the various health care systems. In the United States, pharmaceutical companies' rents from product innovation were further protected by the fragmented structure of health care markets and by the consequent low bargaining power of buyers. Moreover, the U.S. government does not regulate drug prices. Until the mid-1980s most U.S. companies marketed directly to phy- sicians, who largely made the key purchasing decisions by deciding which drug to prescribe. The ultimate customers, patients, had little bargaining power, even in those instances where multiple drugs were available for the same condition. Because insurance companies generally did not cover prescription drugs,6 they did not provide a major source of pricing leverage. Pharmaceutical companies were afforded a relatively high degree of pricing flexibility. This pricing flexibil- ity, in turn, contributed to the profitability of investments in drug R&D. Drug prices were also relatively high in other countries that did not have strong government intervention in prices, such as Germany and the Netherlands. In the United Kingdom, price regulation was framed as voluntary cooperation 6In 1960, only 4 percent of prescription drug expenditures were funded by third-party payers.
PHARMACEUTICALS AND BIOTECHNOLOGY 377 between the pharmaceutical industry and the Ministry of Health. This scheme let companies set their own prices, but the ministry negotiated a global profit margin with each firm designed to ensure each of them an appropriate return on capital investments, including research, made in the United Kingdom. The allowed rate of return was negotiated directly and was set higher for export-oriented firms. In general, this scheme tended to favor both British and foreign research-intensive companies that operated directly in the United Kingdom. Conversely, it tended to penalize weak, imitative firms as well as those foreign competitors, primarily the Germans, trying to enter the British market without direct innovative effort in loco (Burstall, 1985; Thomas, 19941. In Japan the Ministry of Health and Welfare set the prices of all drugs, using suggestions from the manufacturer based on the drug's efficacy and the prices of comparable products. Once fixed, however, the price was not allowed to change over the life of the drug (Mitchell et al., 1995~. Thus, whereas in many competi- tive contexts prices began to fall as a product matured, this was not the case in Japan. Because manufacturing costs often fall with cumulative experience, old drugs thus probably offered the highest profit margins for many Japanese compa- nies, further curtailing the incentive to introduce new drugs. Moreover, generally high prices in the domestic market provided Japanese pharmaceutical companies with ample profits and little incentive to expand overseas. Thus, by the time the revolution in molecular biology began to have its effect on the industry, differences in national policies across regions had already shaped industry structure to a very considerable degree. Across the world, the industry had fragmented into two groups large, highly diversified firms that were tightly connected to the public sector and quick to take advantage of the latest scientific developments and smaller, more marketing-driven firms whose research was ei- ther governed by the older paradigm of "random" search or who concentrated on making improvements to existing therapies. A disproportionate number of Ameri- can firms were of the former type. Their development was largely predicated on the complex mix of policies outlined above, and thus these policies continued to have a very significant influence in shaping the industry even as the revolution in molecular biology further transformed industry dynamics. REVOLUTION IN MOLECULAR BIOLOGY AND CHANGING COMPETENCE IN DRUG R&D If effective public policy was critical to the health of the U.S. pharmaceutical industry in the 1 980s, the revolution in genetics and molecular biology that began 40 years ago with Watson and Crick's discovery of the double helix structure of DNA and continued with Cohen and Boyer's discovery of the techniques of ge- netic engineering made effective policy even more important. The revolution had an enormous impact on the nature of pharmaceutical re- search and development and on the organizational capabilities required to intro
378 U.S. INDUSTRYIN2000 Biotechnology as a production technique: primary focus on proteins whose therapeutic properties were already | understood such asinsu~in. | 1 Scientific a dvances in The discovery of biotechnology genetics, molecular based drugs, such as the protease biology, etc inhibitors. ~> T Biotechnology as a search receptors used as screens; primary focus on small molecule discovery. FIGURE 2 The molecular biology revolution and the trajectories of commercial R&D. duce new drugs.7 Application of the new techniques initially followed two rela- tively distinct technical trajectories (see Figure 2~. One trajectory was rooted in the use of genetic engineering as a process technology to manufacture proteins whose existing therapeutic qualities were already quite well understood in large enough quantities to permit their development as therapeutic agents. The second trajectory used advances in genetics and molecular biology as tools to enhance the productivity of the discovery of conventional "small molecule" synthetic chemical drugs. More recently, as the industry has gained experience with the new technolo- gies, these two trajectories have converged, and contemporary efforts in biotech- nology are largely focused on the search for large molecular weight drugs that must be produced using the tools of genetic engineering but whose therapeutic properties are not, as yet, fully understood. Understanding the distinction between these two trajectories is of critical importance to understanding the role of public policy in the history of the indus- try because the two require quite different organizational competencies and have had quite different implications for industry structure and for the nature of com- petition across the world. In some regions, particularly the United States, the 7Biotechnology has also had far-ranging impacts on several other fields including diagnostics and agriculture. For the purposes of this paper we consider only its impact on human therapeutics.
PHARMACEUTICALS AND BIOTECHNOLOGY 379 ability to manufacture proteins in quantity triggered an explosion of entry into the industry and a proliferation of new firms. Although the success of the U.S. bio- technology industry undoubtedly had a multitude of causes, few observers doubt that a unique mix of publicly funded research and an institutional and financial climate that encouraged the formation of new firms was of key significance. The use of genetics as a tool for small molecule discovery, in contrast, appears to have reinforced the dominance of the large, global pharmaceutical firms at the expense of smaller regional players. Public policy appears to have been important in this case not only through the role that variation in access to publicly funded, leading edge research across the world has played in shaping the success of these large firms but also, and perhaps more important, through its influence as one of the factors that led to the emergence of these large firms in the first place. Biotechnology as a Process Technology Historically, most drugs have been derived from natural sources or synthe- sized through organic chemistry. Proteins, or molecules composed of long inter- locking chains of amino acids, are simply too large and complex to synthesize feasibly through traditional synthetic chemical methods. Those proteins that were used historically as therapeutic agents, notably insulin, were extracted from natu- ral sources or produced through traditional fermentation methods. But traditional fermentation processes, which were used to produce many antibiotics, could use only naturally occurring strains of bacteria, yeast, or fungi, so they were inca- pable of producing the vast majority of proteins. Cohen and Boyer's key contri- bution was the invention of a method for manipulating the genetics of a cell to induce it to produce a specific protein. This invention made it possible for the first time to produce a wide range of proteins and thus opened up an entirely new domain of search for new drugs, the vast store of proteins that the body uses to carry out a host of biological functions. The human body produces approximately 500,000 different proteins, the vast majority of whose functions are not well understood. In principle Cohen and Boyers' discovery thus opened up an enormous new arena for research. The first firms to exploit the new technology chose, however, to focus on proteins such as insulin, human growth hormone, tissue plasminogen activator (tPA), and Factor VIII proteins whose probable therapeutic effects were already relatively well understood. This knowledge greatly simplified both the process of research for the first biotechnology-based drugs and the process of gaining regulatory ap- proval. Marketing these new drugs was also easier because their effects were well known and a preliminary patient population was already in place. Thus for those firms choosing to exploit this route, the organizational capa- bilities most critical to success have been those of manufacturing and process development learning to use the new recombinant DNA techniques as a pro- duction process to produce natural or modified human proteins. The develop
380 U.S. INDUSTRYIN2000 ment of this competence created significant challenges for nearly all of the estab- lished pharmaceutical firms because it required both the creation of an enormous body of new knowledge and a fundamental shift in the ways in which manufac- turing process development was managed inside the firm. The manufacture of small molecular weight drugs is essentially a problem in chemical process R&D. It draws primarily on chemistry and chemical engineer- ing, disciplines in which there is a long history of basic scientific research. As a result much of the relevant theoretical knowledge has been codified in scientific journals and textbooks and, in searching for and selecting alternative chemical processes for the development of small molecular weight drugs, the pharmaceuti- cal firm has at its disposal a wealth of scientific laws, principles, and models that describe the structure of relationships between different variables such as pres- sure, volume, and temperature. Thus process research chemists approaching the manufacture of a small molecular weight drug can often begin their work by deriving alternative feasible synthetic routes from theory. The characteristics of the knowledge base underlying successful biotechnol- ogy process development are quite different. The major discovery underlying the field was made only in 1973, so biotechnology is in its infancy. Moreover, although basic scientific research has been extensive in molecular biology, cell biology, biochemistry, protein chemistry, and other relevant scientific disciplines, most of this work has been geared toward the problems of product "discovery" or to the identification of potentially important proteins rather than to their manu- facture. Very little basic research has been conducted on the problems of en- gineering larger-scale biotechnology processes. Thus process developers in biotechnology have little theory to guide them in the development of new manu- facturing processes. Perhaps just as important, there is a long history of practical experience with chemical processes, whereas process developers in biotechnology had initially almost no practical experience to draw on. The chemical industry emerged in the eighteenth century, and chemical synthesis has been used to produce phar- maceuticals since the late 1800s. Through this experience a large body of heu- ristics have evolved that are widely used to guide process selection, scale-up, and plant design. Most pharmaceutical firms have also developed standard oper- ating procedures for production activities such as quality assurance, process con- trol, production scheduling, changeovers, and maintenance. Experience with these routines provides concrete starting points for development and guidance about the types of process techniques that are feasible within an actual produc- tion environment. In contrast, some observers were initially skeptical that recombinantly engi- neered processes could be scaled up at all. Since 1982, when regulatory authori- ties approved recombinant insulin, the first biotechnology-based pharmaceutical to be manufactured at commercial scale, only about 25 biotechnology-based thera- peutics have been approved for marketing. When a company develops and scales
PHARMACEUTICALS AND BIOTECHNOLOGY 381 up a specific new biotechnology process, it is likely not only to be the company' s first attempt, but also the first time anyone has attempted that process. These differences imply that an organization developing a process for a pro- tein molecule needs not only new technological or scientific capabilities but also organizational capabilities different from those required for developing a manu- facturing process for a new small molecular weight compound. As Pisano (1996) put it, biotechnology process development requires the capability to "learn by doing" in the actual production environment because it is virtually impossible to "learn before doing" in the laboratory. In contrast, small molecule pharmaceuti- cal process development requires the capability to exploit the rich theoretical and empirical knowledge base of chemistry through laboratory research. Biotechnology as a Research Tool The new techniques of genetic engineering have also had a significant im- pact on the organizational competencies required to be a successful player in the pharmaceutical industry through their effect on the competencies required to dis- cover "conventional" small molecular weight drugs. However, although the adop- tion of biotechnology as a process technology was unambiguously competence destroying for incumbent pharmaceutical firms, adoption of biotechnology as a search tool was competence destroying only for those firms that had not made the transition from "random" to "guided" drug discovery. Those firms that had made the transition initially used the tools of genetic engineering as another source of "screens" with which to search for new drugs. For example, genetic engineering techniques allow researchers to clone target receptors, so that firms can screen against a "pure" target rather than against, say, a pulverized solution of rat's brains that probably contain the receptor. The new techniques also permit the breeding of rats or mice that have been genetically altered to make them particularly sensitive to interference with a particular enzy- matic pathway. Firms had to learn some new science, but for those firms that had already made the transition to guided or science-driven drug discovery, these techniques did not destroy existing competence in the way that the use of biotech- nology as a process technology did. The transition from random to guided drug discovery required the develop- ment of a large body of new knowledge and substantially new organizational capabilities in drug research. So-called random drug discovery drew on two core disciplines medicinal chemistry and pharmacology. Successful firms employed battalions of skilled synthetic chemists and pharmacologists who managed smoothly running, large-scale screening operations. Although a working knowl- edge of current biomedical research might prove useful as a source of ideas about possible compounds to test or alternative screens to try, by and large firms did not need to employ researchers at the leading edge of their field or to sustain a tight
382 U.S. INDUSTRYIN2000 connection to the publicly funded research community, and firms differed greatly in the degree to which they invested in advanced biomedical research. The ability to take advantage of the techniques of "guided search," in con- trast, required a very substantial extension of the range of scientific skills em- ployed by the firm a scientific workforce that was tightly connected to the larger scientific community and an organizational structure that supported a rich and rapid exchange of scientific knowledge across the firm (Gambardella, 1995; Henderson and Cockburn,1994~. The new techniques also significantly increased returns to the scope of the research effort (Henderson and Cockburn, 1996~. Managing the transition from random to guided drug discovery was thus not a straightforward matter. In general the larger organizations who had indulged a "taste" for science under the old regime were at a considerable advantage in adopt- ing the new techniques, while smaller firms, firms that had been particularly successful in the older regime, and firms that were much less connected to the publicly funded research community were much slower to follow their lead (Cockburn et al., 1998; Gambardella, 1995~. These differences were critical in shaping responses to the use of biotechnol- ogy as a research tool. For those firms that had already made the transition to guided drug discovery, the adoption of the tools of genetic engineering as an additional resource in the search for small molecule drugs was a fairly natural extension of the existing competence base. Molecular geneticists could be hired as one additional scientific discipline among many, and the genetically engineered screens that they provided could be easily accommodated within the existing re- search procedures. The larger, more scientifically sophisticated firms were at an enormous advantage in employing biotechnology as a research tool in the search for small molecule drugs (Zucker and Darby, 1996, 1997), and this advantage shaped national responses to the biotechnology revolution. It continues to shape responses as the two trajectories have begun to converge. PATTERNS OF INDUSTRY EVOLUTION Thus the techniques of molecular biology had dramatic implications for both the discovery of new drugs, on the one hand, and for the ways in which they were manufactured, on the other. "Biotechnology" in the popular sense, has provided an important additional source of new drugs, but, as discussed above, it is by no means the only way in which these techniques have changed the industry. Each trajectory, biotechnology-based proteins and the use of genetics as a tool in the search for conventional drugs, has been associated with different organizational regimes and patterns of industry evolution across countries. Tables 3 and 4 present some summary data that provide a preliminary picture of some of these differences. Table 3 shows the number of firms active in bio- technology across the world for the periods 1978-1986 and 1987-1993, as de- fined by their activity at the European patenting office. The United States clearly
PHARMACEUTICALS AND BIOTECHNOLOGY TABLE 3 Patent Applications at the European Patent Office, 1978-1993 383 World patentshares (%), 1978- 1993 No. of firms 1978-1986 No. of firms 1987-1993 U.S. 36.5 213 303 Japan 19.5 108 185 U.K. 5.9 39 64 Germany 12.0 45 58 France 6.0 37 52 CH 4.2 11 19 Source: European Patent Office. hosts the majority of firms, but the Japanese are also very highly represented. Table 4 illustrates the dramatic differences in institutional form. Newly founded firms are far more important in the United States and the United Kingdom than they are elsewhere, while the public sector plays a disproportionately important role in France. New firms play a negligible role in Japan, Switzerland, and Ger- many. Comprehensive data that would allow us to match trajectory to institution type is not available, but we believe that the vast majority of the new biotechnol- ogy firms initially pursued the first trajectory, or a focus on biotechnolo~v as a TABLE 4 Patent Activity in Genetic Engineering by Type of Institution ~7~ Percent of patents filed at European patent office NBFs Established corporations Universities and other research institutions 1978-1986 U.S. 43.2 34.5 22.3 Japan 0.00 87.7 12.3 Germany 0.01 81.8 17.7 U.K. 27.3 49.1 23.6 France 18.7 21.5 59.8 Switzerland 0.00 92.9 7.1 Netherlands 12.7 56.4 30.9 Denmark 0.00 93.5 6.5 Italy 0.00 95.7 4.3 1987-1993 U.S. 40.4 38.1 20.7 Japan 3.1 86.9 10.0 Germany 3.0 80.0 17.0 U.K. 23.7 44.7 31.6 France 16.7 35.0 48.3 Switzerland 4.7 89.0 6.3 Netherlands 20.0 62.5 17.5 Denmark 5.7 92.5 1.9 Source: European Patent Office.
384 U.S. INDUSTRYIN2000 process technology, while the established firms with the important exception of the Japanese firms entering the industry from fermentation-related fields largely pursued the second trajectory, or a focus on the use of biotechnology as a re- search tool in the search for small molecule drugs. Newly founded firms were initially far more successful than the established firms in bringing new biological entities to market. Zucker and Darby (1996) present an analysis of 21 new bio- logical entities approved for the U.S. market by 1994: 7 were discovered by small independent firms, 12 by small firms that were subsequently acquired, and only 2 by established pharmaceutical firms acting "in their own right." More recently, as the two trajectories have merged, intracompany agreements have proliferated, the majority between new biotechnology firms and the larger, established firms. Many companies that were initially slow to respond to the opportunities offered by the new science have attempted to "catch up" through joint research agreements or the outright purchase of promising new firms. For example, out of 95 biotechnology drugs that entered clinical trials in the United States between 1980 and 1988,15 were developed solely by pharmaceutical firms, 36 were developed solely by biotechnology firms, and 44 were developed jointly by pharmaceutical and biotechnology firms (Bienz-Tadmore et al., 1992~. Below we explore these geographical differences in more detail as a prelude to our concluding discussion of the degree to which they can be explained by differences in the institutional structure and in the public policy regime surround- ing the industry across the different regions of the world. The United States In the United States, the use of biotechnology as a process technology was the motive force behind the first large-scale entry into the pharmaceutical indus- try since the early postwar period. The first new biotechnology start-up was Genentech, founded in 1976 by Herbert Boyer, one of the scientists who devel- oped the recombinant DNA technique, and Robert Swanson, a venture capitalist. Genentech became the model for a large number of new entrants. They were primarily university spin-offs, and they were usually formed through collabora- tion between scientists and professional managers, backed by venture capital. Their specific skills resided in the knowledge of the new techniques and in the research capabilities in that area. Their goal was to apply the new scientific discoveries to commercial drug development. Entry rates soared in 1980 and remained at a very high level at least until 1985. By the beginning of 1992, there were 48 publicly traded biotechnology companies specialized in pharmaceuticals and health care and several times this number still privately held. Between 1982, when human insulin was approved, and 1992,16 biotechnol- ogy drugs were approved for the U.S. market. As is the case for small molecular weight drugs, the distribution of sales of biotechnology products is highly skewed. Three products were major commercial successes: insulin (Genentech and Eli
PHARMACEUTICALS AND BIOTECHNOLOGY 385 Lilly), tPA (Genentech in 1987), and erythropoietin (Amgen and Ortho in 1989). By 1991 more than 100 biotechnology drugs were in clinical development, and applications for 21 biotechnology drugs had been submitted to the FDA (Grabowski and Vernon, 1994). This was roughly one-third of all drugs in clini- cal trials (Bienz-Tadmore et al., 1992). Sales of biotechnology-derived therapeu- tic drugs and vaccines had reached $2 billion, and two new biotechnology firms, Genentech and Amgen, had entered the club of the top eight major pharmaceuti- cal innovators (Grabowski and Vernon, 1994). Established pharmaceuticals initially played a less direct role in this applica- tion of biotechnology, at least in the United States. Zucker and Darby (1997) show that of all the firms in their sample U.S. firms that either employed or were closely tied to "star" biotechnology scientists taking out worldwide ge- netic-sequence patents between 1980 and 1990, 81 percent were dedicated bio- technology firms. Most of the major companies invested in biotechnology R&D through collaborative arrangements, R&D contracts, and joint ventures with the newbiotechnologystart-ups(AroraandGambardella,1990;Barbanfietal.,1998; Pisano, 1991). As outlined above, the application of molecular biology to the development of protein-based drugs required a completely different set of com- petencies in both drug discovery and process development. Incumbents were thus poorly positioned to exploit the technical opportunities afforded by the new trajectory through in-house research or manufacturing. The competencies re- quired for clinical development, regulatory approval, and marketing were essen- tially the same between biotechnology and traditional synthetic drugs, however, and new firms sought out incumbents as partners who could help commercialize the fruits of their R&D. Thus, during the 1970s and 1980s, a market for know- how emerged in biotechnology with the start-up firms positioned as upstream suppliers of technology and R&D services and established firms positioned as downstream buyers who could provide capital as well as access to complemen- tary assets (Pisano and Mang, 1993). Although newly founded firms pioneered the use of genetics as a source of large molecular weight drugs, established firms led the way in the use of genetic technology as a tool for the discovery of traditional or small molecular weight drugs. The speed with which the new techniques were adopted varied enor- mously, however. For those firms that were already heavily investing in funda- mental research and in which participating in the broader scientific community was already recognized to be of value, the new knowledge presented itself as a natural extension of existing work. They might have been exploring the mechanisms of hypertension, for example. Knowledge of the genetic bases of these mechanism was a fairly easily accommodated "competence" and in general these firms moved quite quickly to adopt the new techniques (Gambardella, 1995; Zucker and Darby, 1997). Firms such as Merck, Pfizer, and SmithKline-Beecham, for example, made the transition relatively straightforwardly. Those firms that had been more firmly oriented toward the techniques of random drug design, however, found the
386 U.S. INDUSTRYIN2000 transition much more difficult. Firms that had no history of publication or of investment in basic science often found it hard to recruit scientists of adequate caliber and to create the communication patterns that the new techniques required. The new techniques probably also significantly increased returns to scope. As drug research came to rely increasingly on the insights of modern molecular biology, discoveries in one field often had implications for work in other areas, and firms that had the size and scope to capitalize on these opportunities for cross fertilization and the organizational mechanisms in place to take advantage of these opportunities reaped significant rewards. Thus one of the major impacts of the revolution in molecular biology has been to drive a wedge between those firms that have been able to absorb the new science into their research efforts and those that are still struggling to make the transition (Cockburn et al.,1997; Zucker and Darby, 1997~. Europe and Japan In Europe and Japan the exploitation of genetics as a tool to produce proteins as drugs lagged considerably behind the effort in the United States and proceeded along different lines (Orsenigo,1995~. The most striking difference, of course, is the virtual lack of specialized biotechnology start-ups in Europe and Japan, with some exceptions in the United Kingdom and isolated cases elsewhere, at least until the late-1980s. This difference is particularly striking because governments in Japan and most European countries at the community, national, and local government levels have devised a variety of measures to foster industry-university collaboration and the development of venture capital to favor the birth of new biotechnology ven- tures. To date the results of these policies have not been particularly impressive, although the increase in the rate of formation of new biotechnology-based firms in the 1990s may reflect the fact that these policies are now beginning to have an impact. Ernst and Young (1995) suggest that there are now approximately 380 biotechnology companies in Europe. Britain has the largest number of new bio- technology firms, followed by France, Germany, and the Netherlands (Escourrou, 1992; SERD,1996~. Recent data, moreover, suggest a dramatic increase of new biological firms in Germany, with different sources estimating their number in the 400 to 500 range or as more than 600 (Coombs, 1995~. Very few of these companies resemble the American prototype, however. Many of the new European firms are not involved in drug research or develop- ment but are instead intermediaries commercializing products developed elsewhere or are active in diagnostics, the agricultural sector, or the provision of instrumen- tation and/or reagents (MERIT, 1996; SERD, 1996~. Moreover, some of these companies, especially the most significant ones like Celltech and Transgene, have been founded through the direct support and involvement of governments and large pharmaceutical companies rather than through the venture capital market. The contribution of this new breed of companies to the development of Euro
PHARMACEUTICALS AND BIOTECHNOLOGY 387 pean biotechnology remains to be seen. They already seem to be suffering from the disadvantages of entering the market relatively late. Only the earliest entrants are significant innovators, and some of the most successful, like their American counterparts, have already been acquired or expect to be acquired shortly by U.S. companies. In the absence of extensive new firm founding, most of the innovation in biotechnology in mainland Europe has occurred within established firms. In France there has been significant entry, largely from firms diversifying into bio- technology and from other research institutions, while in Germany there has been almost no entry at all. Thus in mainland Europe a few firms account for a large proportion of biotechnology patents, and innovation in biotechnology rests essen- tially on the activities of a relatively small and stable group of large established companies. However, in contrast to the majority of the established American firms that adopted the techniques of genetic engineering as a manufacturing tool prima- rily through acquisition and collaboration with the small American start-ups, the European firms showed considerable variation in the methods through which they acquired the technology. The British (Glaxo, Wellcome, and to a lesser extent ICI) and the Swiss companies (particularly Hoffman La Roche, Ciba Geigy, and Sandoz) moved early and decisively in the direction pioneered by the large U.S. firms in collabo- rating with or acquiring American start-ups. Firms in the rest of Europe tended to establish a network of alliances with local research institutes, although German companies lagged somewhat behind. Hoechst signed a 10-year agreement with Massachusetts General Hospital as early as 1981, but Bayer did not enter seri- ously until 1985. In general the Germans made little progress in the field, and they are not now considered to be among the leaders in European biotechnology. In some countries such as Italy, the scientific community took the lead in the attempt to promote the commercial development of genetic engineering through the establishment of linkages and collaboration with the pharmaceutical industry. The biggest European innovators are a research institution, Institut Pasteur in France, and two companies that have not been traditional players in the pharma- ceutical industry, Gist-Brocades and Novo Nordisk. While data are difficult to obtain, it appears that almost all of the established French, Italian, German, and Japanese companies have been slow to adopt the tools of biotechnology as an integral part of their own drug research efforts. In Japan, the large food and chemical companies with strong capabilities in process technologies, such as Takeda, Kyowa Hakko, Ajinomoto, and Suntory, pioneered entry into biotechnology. Although these firms have strong competen- cies in process development, they generally lack capabilities in basic drug re- search. During the 1980s some U.S. observers expressed concern that biotech- nology would be the next industry in which Japanese firms achieved dominance, but to date that has not occurred, and there has been only limited entry into the pharmaceutical industry through biotechnology.
388 U.S. INDUSTRYIN2000 NATIONAL SYSTEMS OF INNOVATION: HOW DID PUBLIC POLICY MATTER? This brief description of the impact of the revolution in molecular biology on the pharmaceutical industry highlights the diversity of responses across the world and suggests several "stylized facts" to be explored in examining the relationship between "national systems of innovation," or the entire set of public policies and institutional constraints that shaped the evolution of any particular firm and the evolution of the industry across the world. First, why was the use of molecular biology as a production tool pioneered in the United States by small, newly founded firms, in Japan by firms diversifying into the industry from other fields, and in Europe largely by established pharmaceutical firms? Why did new en- trants play a much smaller role in the European context? Second, did national systems of innovation play a role in shaping the diffusion of the use of molecular biology as a research tool? This technology was pioneered by established phar- maceutical firms in almost every case, yet its rate of adoption varied widely across the world. The Evolution of "Biotechnology" Why the small, independently funded biotechnology start-up was initially an American phenomenon is an old question and a much discussed one. One of the reasons that it cannot be answered definitively is that the answer is to a large degree overdetermined; many factors were clearly at play, almost any one of which may have been sufficient. As the discussion has already suggested, the use of molecular biology as a production technology was a competence-destroying innovation for the vast majority of the established pharmaceutical firms. In the United States a combination of factors allowed small, newly founded firms to take advantage of the opportunity this created. These factors included a favorable financial climate, strong intellectual property protection, a scientific and medical establishment that could supplement the necessarily limited competencies of the new firms, a regulatory climate that did not restrict genetic experimentation, and, perhaps most importantly, a combination of a very strong local scientific base and academic norms that permitted the rapid translation of academic results into com- petitive enterprises. In Europe, apart from the United Kingdom, and in Japan many of these factors were not in place, and it was left to larger firms to exploit the new technology. A Strong Scientific Base and Academic Norms The majority of the American biotechnology start-ups were tightly linked to university departments, and the very strong state of American academic molecu- lar biology clearly played an important role in facilitating the wave of start-ups that characterized the 1 980s (Zucker et al. forthcoming). The strength of the local science base may also be responsible within Europe for the relative British ad
PHARMACEUTICALS AND BIOTECHNOLOGY 389 vantage and the relative German and French delay. Similarly, the weakness of Japanese industry may partially reflect the weakness of Japanese science. There seems to be little question about the superiority of the American and British sci- entific systems in the field of molecular biology, and it is tempting to suggest that the strength of the local science base explains much of the regional differences in the speed with which molecular biology was exploited as a tool for the production of large molecular weight drugs. Although this explanation might seem unsatisfying to the degree that aca- demic science is rapidly published and thus, in principle, rapidly available across the world, the American lead appears to have been particularly important because the exploitation of biotechnology in the early years required the mastery of a considerable body of tacit knowledge that could not be easily acquired from the literature (Zucker et al., 1997; Pisano, 1996~. Geographic proximity probability facilitated the transmission of this kind of tacit knowledge (Jaffe et al., 1993~. In the case of biotechnology, however, several authors have suggested that the U.S. start-ups were not simply the result of geographic proximity (Zucker et al., 1997~. These authors have suggested that the flexibility of the American academic sys- tem, the high mobility characteristic of the scientific labor market, and, in gen- eral, the social, institutional, and legal context that made it relatively straightfor- ward for leading academic scientists to become deeply involved with commercial firms were also major factors in the health of the new industry. The willingness to exploit the results of academic research commercially also distinguishes the U.S. environment from that in either Europe or Japan. This willingness has been strengthened since the late 1970s and the passage of the Bayh-Dole Act (see below), and the resulting role of universities as seedbeds of entrepreneurship has probably also been extremely important in the take-off the biotechnology industry. In contrast, links between the academy and industry, especially the relatively free exchange of personnel, appear to have been much weaker in Europe and Japan. Indeed, the efforts of several European governments were targeted pre- cisely toward strengthening industry-university collaboration, and it has been ar- gued that the rigidities of the research system of continental Europe and the large role played in France and Germany by the public, nonacademic institutions have significantly hindered the development of biotechnology in those countries. That these kinds of factors, as distinct from the strength of the science base per se, were absolutely critical to the wave of new entry in biotechnology that occurred in America in the early 1980s is given further credibility by the rate at which the use of molecular biology diffused across the world. Access to Capital It is commonly believed that lack of venture capital restricted the start-up activity of biotechnology firms outside the United States. Clearly, venture capi
390 U.S. INDUSTRYIN2000 tat, a largely American institution, played an enormous role in fueling the growth of the new biotechnology-based firms. Prospective start-ups in Europe, however, appear to have had many other sources of funds, usually through government programs. The results of several surveys also suggest that financial constraints did not constitute a significant obstacle for the founding of new biotechnology firms in Europe (Ernst and Young, 1995; MERIT, 1996; SERD, 1996~. In addition, although venture capital played a critical role in the founding of U.S. biotechnology firms, collaborations between the new firms and the larger, more established firms provided a potentially even more important source of capi- tal. Why did prospective European or Japanese biotechnology start-ups not turn to established pharmaceutical firms as a source of capital? A plausible explana- tion focuses on the market for know-how in biotechnology. The evolution of that market created many opportunities for European and Japanese companies to col- laborate with U.S. biotechnology firms. Although some U.S.-based new firms, such as Amgen, Biogen, Chiron, Genentech, and Genzyme, pursued a strategy of vertical integration from research through marketing in the U.S. market, most firms' strategies emphasized licensing product rights outside the U.S. to foreign partners. Thus to an even greater extent than many established U.S. pharmaceu- tical firms, European and Japanese firms were well positioned as partners for U.S. new biotechnology firms. Given the plethora of new U.S. firms in search of capital, European and Japanese firms interested in commercializing biotechnol- ogy had little incentive to invest in local biotechnology firms. Even in the ab- sence of other institutional barriers to entrepreneurial ventures, start-ups in Eu- rope or Japan might have been crowded out by the large number of U.S.-based firms anxious to trade non-U.S. marketing rights for capital. Intellectual Property Rights The establishment of clearly defined property rights also played a major role in making possible the explosion of new firm foundings in the United States, because the new firms, by definition, had few complementary assets that would have enabled them to appropriate returns from the new science in the absence of strong patent rights (Teece, 1986~. In the early years of biotechnology, considerable confusion surrounded the conditions under which patents could be obtained. In the first place, research in genetic engineering was on the borderline between basic and applied science. Much of it was conducted in universities or otherwise publicly funded, and the degree to which it was appropriate to patent the results of such research became the subject of bitter debate. Millstein and Kohler's groundbreaking discovery, hybridoma technology, was never patented, while Stanford University filed a patent for Boyer and Cohen's process in 1974. Boyer and Cohen renounced their own rights to the patent, but nevertheless they were strongly criticized for having being instrumental in patenting what many observers considered to be a basic
PHARMACEUTICALS AND BIOTECHNOLOGY 391 technology. Similarly a growing tension emerged between publishing research results versus patenting them. The norms of the scientific community and the search for professional recognition had long stressed rapid publication, but patent laws prohibited the granting of a patent to an already published discovery (Merton, 1973; Kenney, 1986~. In the second place the law surrounding the possibility of patenting life-formats and procedures relating to the modification of life-forms was not defined. This issue involved a variety of problems, but it essentially boiled down, first, to whether living things could be patented at all and, second, to the scope of the claims that could be granted to such a patent (Merges and Nelson, 1994). These hurdles were gradually overcome. In 1980 Congress passed the Patent and Trademark Amendments of 1980 (Public Law 96-517~. Also known as the Bayh-Dole Act, this law gave universities, and other nonprofit institutions and small businesses, the right to retain the property rights to inventions deriving from federally funded research. In 1984 Congress expanded the rights of univer- sities further, by removing certain restrictions contained in Bayh-Dole regarding the kinds of inventions that universities could own and the right of universities to assign their property rights to other parties. In 1980 the U.S. Supreme Court ruled in favor of granting patent protection to living things (Diamond v. Chakrabarty); the case involved a scientist working for General Electric who had induced genetic modifications on a Pseudomonas bacterium that enhanced its ability to break down oil. In the same year the second reformulation of the Cohen and Boyer patent for the recombinant DNA process was approved. In subsequent years, a number of patents were granted establishing the right for very broad claims (Merges and Nelson, 1994~. Finally, a one-year grace period was intro- duced for filing a patent after publication of the invention. It is often stressed that the lack of adequate patent protection was a major obstacle to the development of the biotechnology industry in Europe.8 First, the grace period available in the United States is not available in Europe; any discov- ery that has been published is not patentable. Second, the interpretation has pre- vailed that naturally occurring entities, whether cloned or uncloned, cannot be patented. As a consequence, the scope for broad claims on patents is greatly reduced, and usually process rather than product patents are granted. In 1994 the European Parliament rejected a draft directive from the European Commission that attempted to strengthen the protection offered to biotechnology. Although it is clear that stronger intellectual property protection is not unam- biguously advantageous, as the controversy surrounding NIH's decision to seek patents for human gene sequences clearly illustrated, in the early days of the industry, the United States probably reaped an advantage from its relatively stronger regime. ~ See, for instance, Ernst and Young (1995).
392 Regulatory Climate U.S. INDUSTRYIN2000 Although public opposition to genetic engineering was a significant phe- nomenon in the United States in the earliest years of the industry, it has quickly become less important, and in general the regulatory climate has been a favorable one (Kenney, 19861. In Europe, however, opposition to genetic engineering re- search by the "Green" parties is often cited as an important factor hindering the development of biotechnology, especially in Germany and other Northern Euro- pean countries, and public opposition to biotechnology is said to have been a factor behind the decision of some European companies to establish research laboratories in the United States. The Use of Molecular Biology as a Research Tool Explaining variations in the rate of adoption of molecular biology as a re- search tool across the regions of the world is, in contrast, rather more difficult. In general the techniques were adopted first by the large, globally oriented U.S., British, and Swiss firms. Adoption by the other European firms, and by the Japanese, appears to have been a much slower process. At first glance the relative strength of the local science base and the degree to which university research was connected to the industrial community appears to be as important an explanation here as it was in understanding the case of the diffusion of "biotechnology." Science in Japan and mainland Europe was argu- ably not as advanced as it was in the United States and Britain, a factor that slowed the adoption of the new techniques. Unfortunately this explanation is made much less plausible by the Swiss case. The Swiss companies established strong connections with the U.S. scientific system, suggesting that geographic proximity played a much less important role in the diffusion of molecular biology as a research tool. A second possible explanation is that diffusion was shaped by the relative size and structure of the various national pharmaceutical industries. Henderson and Cockburn (1996) have shown that between 1960 and 1990 there were signifi- cant returns to size in pharmaceutical research, and that since 1975 these returns have come primarily from the exploitation of economies of scope. They interpret this finding as suggesting that the effective adoption of the techniques of guided search and more rational drug design placed a premium on the ability to integrate knowledge within the firm and thus that the larger, more experienced firms may have been at a significant advantage in the exploitation of the new techniques. To the degree that those firms that had already adopted the techniques of "rational" drug discovery were at a significant advantage in adopting molecular biology as a research tool, the pre-existence of a strong national pharmaceutical industry with some large internationalized companies may have been a fundamental prerequi- site for the rapid adoption of molecular biology as a tool for product screening
PHARMACEUTICALS AND BIOTECHNOLOGY 393 and design. The U.S. pharmaceutical industry has traditionally been internation- ally oriented and, at least since the early 1980s, open to international competition in the domestic market. But in many European countries, such as France and Italy, the pharmaceutical industry was highly fragmented into relatively small companies engaged essentially in the marketing of licensed products and in the development of minor products for the domestic markets. Although size or global reach may have been a necessary condition, the fail- ure of the largest German and Japanese firms to adopt these techniques suggests that it was not sufficient. The largest Japanese and German firms were arguably as international and as large as the Swiss. The most plausible explanation is that institutional variables, particularly the stringency of the regulatory environment and the nature of patent regime, were also important. As mentioned earlier, there is now widespread recognition that the introduction of the Kefauver-Harris Amendments had a significant impact in inducing a deep transformation of the U.S. pharmaceutical industry, particularly through raising the cost and complexity of R&D. Partly as a result many U.S. firms were forced to upgrade their scientific capability. Similarly the two European countries, the United Kingdom and Switzerland, whose leading firms did move more rapidly to adopt the new techniques appear to have actively encouraged a "harsher" competitive environment. The British sys- tem encouraged the entry of highly skilled foreign pharmaceutical firms, espe- cially the American and the Swiss, and a stringent regulatory environment also facilitated a more rapid trend toward the adoption by British companies of institu- tional practices typical of the American and Swiss companies in particular, product strategies based on high-priced patented molecules, strong links with universities, and aggressive marketing strategies focused on local doctors. The resulting change in the competitive environment in the home market induced British firms to pursue strategies that moved away from fragmenting innovative efforts into numerous minor products toward concentration on a few important products that could diffuse widely into the global market. By the 1970s the ensu- ing transformations of British firms had led to their increasing expansion into the world markets. Lacy Glenn Thomas (1994) has suggested that the slowness with which the majority of the European firms, apart from British and Swiss firms, adopted the techniques of guided drug discovery reflected much weaker competitive pres- sures in their domestic markets. The Japanese experience also looks in many respects like that pursued in Europe outside Switzerland and the United King- dom. In Japan legal and regulatory policies combined to frame a very "soft" competitive environment that appears to have seriously slowed the adoption of modern techniques by the Japanese pharmaceutical industry. As a result of the combination of patent laws, the policies surrounding drug licensing, and the drug reimbursement regime, Japanese pharmaceutical firms had little incentive to de- velop world-class product development capabilities, and in general they concen
394 U.S. INDUSTRYIN2000 bated on finding novel processes for making existing foreign or domestically originated molecules (Mitchell et al., 1995~. Moreover, Japanese firms were pro- tected from foreign competition and simultaneously had strong incentives to li- cense products that had been approved overseas. Under this regime the predomi- nant technology strategy for Japanese pharmaceutical companies became the identification of promising foreign products to license. Mitchell et al. (1995) have noted that some of these institutional factors are beginning to change and that these changes are starting to have effects on the R&D strategies and capabilities of some but not all firms participating in the Japanese pharmaceutical sector. After 1967 foreign-originated products required clinical testing in Japan before they could be approved for sale. After 1976 drug products could be patented. After 1981 pricing policy was changed so that prices for established drugs are reviewed periodically and compared with prices of newer drugs. Together these factors have combined to increase the incentives for origi- nal research. Recent evidence suggests that the share of new chemical entities approved in the United States that originate in Japan has increased substantially, from 4 percent in the 1970s to around 25 percent in 1988 (Mitchell et al., 1995~. Nevertheless, because they lack a history of strong internal R&D, it is taking time for Japanese pharmaceutical companies to develop world class research capabili- ties. Strong domestic competition, the existence of appropriate incentive mecha- nisms toward aggressive R&D strategies, and integration into the world markets thus appear to be important explanatory variables in analyzing variations in the diffusion of the new technologies in drug screening and design across regions. Note, however, that they appear to say little about variations in diffusion across Arms. Most of the firms that rapidly adopted the new techniques were large multi- national or global companies, with a strong presence, at least as far research is concerned, in the United States and generally on the international markets. Zucker and Darby (1997) present some evidence that size alone is a reasonable predictor of adoption, at least in the United States. We suspect that this correlation reflects the fact that adoption is highly correlated with the degree to which firms have made the transition to guided drug discovery. By and large these were larger firms that had early developed a taste for science and that were able to build and sustain tight links to the public research community (Gambardella, 1995~. Here institutional factors appear to have been a necessary but not sufficient condition. To the extent that the adoption of the new techniques also involved the successful adoption of particular, academic-like forms of organization of research within companies (Henderson, 1994), and this process was in turn influenced by the proximity and availability of first-rate scientific research in universities, it was much easier for American and to a lesser extent British firms to adopt them. From this perspective, it is tempting to suggest that the origin of the Ameri- can advantage in the use of biotechnology as a research tool as well as a process
PHARMACEUTICALS AND BIOTECHNOLOGY 395 technique lies in the comparatively closer integration between industry and the academic community, compared with other countries. One might also speculate that this closer integration resulted in some degree from the strong scientific base of the American medical culture and from the adoption of tight scientific proce- dures in clinical trials. Through this mechanism, American companies might have come to develop earlier and stronger relationships with the biomedical com- munity and with molecular biologists in particular. Segregation of the research system from both medical practice and from close contact with commercial firms (as in France and possibly in Germany) has been highlighted as a major factor hindering the transition to molecular biology in these two countries (see, for in- stance, Thomas, 1994~. CONCLUSION Public policy plays a crucial role in shaping private sector productivity in many modern economies. The case of the pharmaceutical industry provides a particularly intriguing window into this process and into the importance of na- tional systems of innovation in shaping industrial structure (Nelson, 1992~. Before the revolution in molecular biology, the U.S. pharmaceutical industry was shaped by public policy choices in a number of areas. Strong support for the life sciences provided both highly trained employees and a steady stream of knowledge that was a critical input to the industry. A tough regulatory environ- ment and an intellectual property regime together increased the returns to funda- mental innovation and further combined to create a cohort of large, diversified, highly skilled firms able to manage the transition from random to science-guided discovery effectively. The revolution in molecular biology further reinforced the power of public policy in shaping the industry. In the first place, the revolution's extraordinary dependence on fundamental science meant that the large commitment the United States made to public funding became even more important. But the importance of public policy extends far beyond the simple provision of funding for research. In the case of biotechnology, or the use of molecular biology as a production technique, advances in basic science rendered obsolete several of the core com- petencies of existing firms, particularly those related to process development and manufacturing. In the United States, institutional flexibility on a wide range of dimensions led to the formation of specialized biotechnology firms that could provide these competencies and bridge the gap between basic university research, on the one hand, and clinical development of drugs on the other. Thus the new biotechnology-based firms were, in many ways, an institutional, or public policy- shaped, response to the technical opportunities created by new scientific know- how. The case of biotechnology as a research tool presents a different but comple- mentary picture. This trajectory was born within the confines of established phar
396 U.S. INDUSTRYIN2000 maceutical firms, and institutional factors appear to played a "necessary" rather than "sufficient" role in its diffusion. Pharmaceutical firms adopted biotechnol- ogy as a research tool as a way to use molecular biology to enhance the value and productivity of their existing assets and competencies, and in this sense biotech- nology tools were "competence enhancing." But they were only competence enhancing for some pharmaceutical firms those that were already oriented to- ward "high science" research and already firmly embedded in the global scien- tific community. Thus this case is one of existing institutional arrangements and structures shaping, rather than creating, the path of technical change. Forces facilitating institutional flexibility and responsiveness played a less prominent role in this domain, which may help to explain why Swiss and British firms have joined U.S. firms as leaders in the application of molecular biology to small mol- ecule discovery. We hesitate to draw any hard and fast conclusions about how public policy might best be shaped in the future to support the health of the industry. But this brief historical overview does raise several intriguing questions. First, it high- lights the extraordinarily important role of publicly funded science in supporting the industry. Most important from a policy perspective, perhaps, it highlights the fact that the ways in which this research is conducted may be as important as the level to which it is funded. The published results of publicly funded research are, with some lag, widely diffused across the world, and this kind of "output" clearly had an important impact on the industry. But our discussion suggests American industry was able to gain extraordinary benefits from this research because of the fluid nature of the boundary between public and private research institutions in the field. In the case of the larger, more established firms, this led to the creation of several exceptionally creative and flexible research organizations that were heavily influenced by the norms of "open" science. In the case of biotechnology, it led to the foundation of an extraordinary number of new firms whose energy and creativity has been the envy of the world. To the extent that efforts to realize a direct return on public investments in research lead to a weakening of the cul- ture and incentives of "open science," our results are consistent with the hypoth- esis that the productivity of the whole system of biomedical research may suffer. REFERENCES Arora, A., and A. Gambardella. (1990). "Complementarily and external linkage: the strategies of the large firms in biotechnology." Journal of Industrial Economics 37(4):361-379. Barbanfi, P., A. Gambardella, and L. Orsenigo. (1998). "The evolution of the forms of organization of innovative activities in biotechnology." Forthcoming in Biotechnology/International Journal of Technology Management. Bienz-Tadmore, B., P. Decerbo, G. Tadmore, and L. Lasagna. (1992). "Biopharmaceuticals and con- ventional drugs: Clinical Success Rates." Bio/Technology 10:521-525. Burstall, M.L. (1985). "The Community's pharmaceutical industry," Bruxelles: Commission of the European Communities.
PHARMACEUTICALS AND BIOTECHNOLOGY 397 Chien, R.I. (1979). Issues in pharmaceutical economics. Lexington, MA: Lexington Books. Cockburn, I., and R. Henderson. (1996). "Public-Private Interaction in Pharmaceutical Research" Proceedings of the National Academy of Sciences 93/23(November):12725-12730. Cockburn, I., and R. Henderson. (1998). "Absorptive Capacity, Coauthoring Behavior, and the Orga- nization of Research in Drug Discovery." Journal of Industrial Economics XLVI(2):157-182. Cockburn, I., R. Henderson, and S. Stern. (1997). "Fixed Effects and the Diffusion of Organizational Practice in Pharmaceutical Research." MIT Mimeo. Coombs, A. (1995). The European Biotechnology Yearbook. Davies, W. (1967). The Pharmaceutical Industry: A Personal Study. Oxford: Pergamon Press. Ernst and Young. (1995). "European Biotech 95: Gathering Momentum." Escourrou, N. (1992). "Les societes de biotechnologie europeennes: un reseau tres imbrique Biofutur." (July-August) :40-42. Gambardella, A. (1995). Science and Innovation in the U.S. Pharmaceutical Industry. Cambridge University Press: Cambridge. Grabowski, H., and J. Vernon. (1994). "Innovation and structural change in pharmaceuticals and biotechnology." Industrial and Corporate Change 3(2):435-450. Grabowski, H., and J. Vernon. (1983). The regulation of pharmaceuticals. Washington and London: American Enterprise Institute for Public Policy Research. Griliches, Z. (1994). "Productivity, R&D and the Data Constraint." The American Economic Review 84(1): 1-23. Hancher, L. (1990). Regulating for Competition: Government, Law and the Pharmaceutical Industry in the United Kingdom and France. Oxford: Oxford University Press. Henderson, R. (1994). "The Evolution of Integrative Competence: Innovation in Cardiovascular Drug Discovery." Industrial and Corporate Change 3(3):607-630. Henderson, R., and I. Cockburn. (1994). "Measuring Competence? Exploring Firm Effects in Phar- maceutical Research." Strategic Management Journal 15(Winter):63-84. Henderson, R., and I. Cockburn. (1996). "Scale, Scope and Spillovers: The Determinants of Research Productivity in Drug Discovery." Rand Journal of Economics 27(1):32-59 Irvine, J., B. Martin, and P. Isard. (1990). Investing in the Future: An International Comparison of Government Funding of Academic and Related Research. Aldershot, England: Edward Elgar Publishers. Jaffe, A., M. Trajtenberg, and R. Henderson. (1993). "Geographic Localization of Knowledge Spillovers as Evidenced by Patent Citations." Quarterly Journal of Economics 434:578-598 Kenney, M. (1986). Biotechnology: The Industry-University Complex. Ithaca: Cornell University Press. Maxwell, R., and S. Eckhardt. (1990). Drug Discovery: A Case Book and Analysis. Clifton, NJ: Humana Press. Merges, R., and R. Nelson. (1994). "On limiting or encouraging rivalry in technical progress: The effect of patent scope decisions," Journal of Economic Behavior and Organization 25: 1-24. MERIT. (1996). "The Organization of Innovative Activities in the European Biotechnology Industry and its Implications for Future Competitiveness." Report for the European Commission. Maastricht. Merton, D. (1973). The Sociology of Science: Theoretical and Empirical Investigation. Chicago: University of Chicago Press. Mitchell, W., T. Roehl, and R. Slattery. (1995). "Influences on R&D Growth among Japanese Phar- maceutical Firms, 1975-1990." Journal of High Technology Management Research 6(1):17-31. Nelson, R.R., ed. (1992). National Systems of Innovation. Oxford: Oxford University Press. Orsenigo, L. (1995). The Emergence of Biotechnology. London: Pinter Publishers. Peltzman, S. (1974). Regulation of Pharmaceutical Innovation: The 1962 Amendments. Washington, DC: American Enterprise Institute for Public Policy. Pisano, G. (1991). "The Governance of Innovation: Vertical Integration and Collaborative Arrange- ments in the Biotechnology Industry." Research Policy 20:237-249.
398 U.S. INDUSTRYIN2000 Pisano, G. (1996). The Development Factory. Unlocking the Potential of Process Innovation. Boston: Harvard Business School Press. Pisano, G., and P. Mang. (1993). "Collaborative Product Development and the Market for Know- How: Strategies and Structures in the Biotechnology Industry." In Research on Technological Innovation, Management and Policy, Vol 5. R. Rosenbloom and R. Burgelman, eds. Greenwich. CT: JAI Press. Reich, M. (1990) "Why the Japanese Don't Export More Pharmaceuticals: Health Policy as Indus- trial Policy." California Management Review (Winter): 124-150. SERD. (1996). "The Role of SMEs in Technology Creation and Diffusion: Implications for European Competitiveness in Biotechnology." Report for the European Commission. Maastricht, Nether- lands. Statman, M. (1983). Competition in the Pharmaceutical Industry: The Declining Profitability of Drug Innovation. Washington, DC: American Enterprise Institute. Teece, D.J. (1986). "Profiting from Technological Innovation: Implications for Integration, Collabo- ration, Licensing and Public Policy." Research Policy 15(6):185-219. Thomas, L.G. III. (1994). "Implicit Industrial Policy: The Triumph of Britain and the Failure of France in Global Pharmaceuticals." Industrial and Corporate Change 3(2):451-489. Ward, M., and D. Dranove. (1995). "The Vertical Chain of R&D in the Pharmaceutical Industry." Economic Inquiry 33 (January): 1 - 18. Wardell, W. (1978). Controlling the Use of Therapeutic Drugs: An International Comparison. Wash- ington, DC: American Enterprise Institute. Zucker, L., and M. Darby. (1996). "Costly Information in Firm Transformation, Exit, or Persistent Failure." American Behavioral Scientist 39:959-974 Zucker, L., and M. Darby. (1997). "Present at the revolution: Transformation of technical identity for a large incumbent pharmaceutical firm." Research Policy 26:429-446. Zucker, L., M. Darby, and M. Brewer. "Intellectual Human Capital and the Birth of U.S. Biotechnol- ogy Enterprises." Forthcoming, American Economic Review.