Lapas attēli
PDF
ePub

increase in the later 1960s in the short-term trends noted by Jadlow and the long-term trends described by Schnee.

[ocr errors][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small]

Grabowski offers more specific data on innovation in the drug industry, contained in Table 9:

[blocks in formation]

These data indicate: (1) a reinforcement of Jadlow's findings of a declining NCE output after 1962; (2) a reduction in the number of firms accounting for an NCE in each time period; and (3) increasing concentration ratios at the four-firm and eight-firm levels, and, very slightly the twenty-firm level. Perhaps this last statistic warrants emphasis: although innovational concentration has increased at the four-firm and eight-firm level, it has not increased very much at the twenty-firm level. Thus, the top twenty research-intensive firms in the industry have not increased their share of total drug innovation during this time period. The gains of the top four and eight firms apparently have come at the expense of smaller firms within the top twenty. These results are consistent with Jadlow's, for in an industry of perhaps 600 firms, those ranking 9th through 20th in terms of innovational output are among the "smallest 99 and 98 percent" of all firms.

Grabowski concludes that the drug industry has displayed what most other industries have not: "a strong shift toward greater concentration of innovational output in the U.S. in the very largest

.. firms." While he states that this shift, given the "large upward shifts in development costs," is not surprising, he also noted

30 Id. at 73 (Table 5). 31 Id. at 72 (Table 4).

that it may be a characteristic of the chemical industry in general. The 1962 Amendments seem to be an implicit factor, perhaps un important one, in this shift. The validity of Schnee's earlier observation that this trend was clearly underway before 1962 is supple · mented by Cirabowski's observation that institutional features within the chemical industry as a whole, and perhaps special ones within the drug industry, may have been important contributing factors.

Decreased Effective Patent Life

At least one factor has been a countervailing force against increasing concentration in the drug industry. Longer development and approval times between discovery and marketing have reduced the effective (commercial) life of drug patents, resulting in earlier market penetration by generic rivals. Since there have been relatively fewer drug innovations than in earlier years, it would seem plausible to expect that the average commercial life of drugs has increased. On the other hand, the increase in the length of development periods for drugs has reduced the remaining patent period after market introduction ("average effective patent life"). While the impacts on profitability from these opposing factors have not been studied, some estimates of the changes in average effective patent life over recent years have been made by Schwartzman, Statman, and Evanoff.

Schwartzman" has estimated the effective patent life of eighty NCES introduced into the United States market between 1966 and 1973. While these estimates do not compare effective drug patent life before and after 1962, they suggest trends in patent life that may have been influenced by the Amendments. The key findings, which compare NCES introduced in 1966-1969 with those introduced in 1970-1973, are presented in Table 10.

[blocks in formation]

32 D. Schwartzman, INNOVATION IN THE PHARMACEUTICAL INDUSTRY (1976) (Ch. 8:

"The Life of Drug Patents").

33 Id. at 173.

[merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][ocr errors][merged small][merged small][merged small][merged small][merged small][ocr errors]

While the data do not directly show the effects of the 1962 Amendments, they show that average effective patent life is substantially shorter than the full statutory patent life. Moreover, it has decreased for drugs in each therapeutic field, if unevenly so, between the late 1960s and early 1970s. This decrease in average effective patent life may have serious implications for the profitability of such drugs and for prices that consumers ultimately pay in the marketplace.

Statman" also has examined the effective patent life of NCES for 126 drugs introduced between 1949 and 1975, thus covering a wider period than Schwartzman and including years before 1962. Using simple regression analysis, Statman finds a continuous decline in the effective life of drug patents of .375 years for each year of introduction after 1960. Thus, expected effective patent life for NCES introduced was 16.5 years in 1960, 14.6 years in 1965, 12.7 years in 1970, 10.9 years in 1975, and 9.7 years in 1978. However, Statman's findings are tenuous for many reasons, including his assumption of a total development and regulatory period of only .5 years for 1961 and before. This figure seriously conflicts with Sarett's estimate of over two years for the post patent-premarketing period at the time. Nonetheless, Statman's data generally agrees with Schwartzman's.

Evanoff" has derived estimates of "average expected patent life ("AEPL")" for the years 1949-1975 from data in other studies pertaining to estimated development and regulatory periods. Subtracting these development and regulatory periods from the 17-year statutory patent life, Evanoff finds AEPL to have been stable at fifteen years between 1949 and 1962, and then to decline steadily to nine years in 1973, remaining at the level through 1975. Evanoff's results thus fit well with Statman's; together, they imply

34 Meir Statman, "The Effect of Patent Expiration on the Market Position of Drugs." Paper presented at Conference on Drugs and Health: Economic Issues and Policy Objectives, American Enterprise Institute, Nov. 15-16, 1979.

35 Douglas D. Evanoff, "An Econometric Model of the Ethical Pharmaceutical Industry: The Effect of Legislative Alternatives," paper presented at the annual meeting of the Southern Economic Association, Nov. 6-9, 1979.

that these shorter patent life periods may have some serious effects on market profitability

Yet the shorter average effective patent life is largely the mirror image of longer development and regulatory periods, and so is not conclusively attributable to the 1962 Amendinents. While effective patent life appears to have been stable for NCEs introduced in the years before 1962 and to have grown shorter in the years after 1962, a variety of factors have contributed to that phenomenon, including a reduction in the "patent pending" period because of greater efficiency in patent issuance. A reduced effective patent life may well lead to a reduced effective commercial life, with resulting negative effects on profitability and R&D. A solution may be revising the patent laws to remedy a feature of those laws that singularly discriminates against drugs; unlike the many products for which marketing approval is not required, large parts of the patent period for drugs are used up before the product is thoroughly tested, studied, and approved for marketing.

Summary

In conclusion, the data on increasing concentration of R&D success, whether measured by source of discovery or by source of market introduction, show trends toward higher concentration among the largest firms from the 1950s through the 1970s. The largest firms experienced a decline in their share of total drug sales in the 1950s, as industry growth diluted their prominence; yet their dominant R&D positions declined relatively less than their sales shares during this period. During the 1960s and 1970s their market shares have become stable and their relative R&D endeavors have grown substantially, when measured at the four-firm, eight-firm and twenty-firm levels in the early 1960s, and continuing at the four-firm and eight-firm levels in the late 1960s.

As R&D has become more costly and time consuming, there have been fewer NCES and fewer firms introducing NCES. Correspondingly, the average effective patent life for new drugs has declined to as low as nine or ten years.

Yet, as before, the role of the 1962 Amendments as a factor in these developments is not yet fully clear, for some of these trends began before 1962. Strong currents of increasing drug R&D concentration have been at work, but these currents also seem to have begun as early as the latter 1950s. The tentative conclusion at this point in the analysis is still unchanged: the 1962 Amendments are not clearly demonstrated to have been the only, or the major,

cause of the observed changes in the drug industry's R&D) performance and structure. Many of the effects attributed to the Amendments on closer analysis can be seen to have been under way before 1962. Yet the 1962 Amendments appear to have magnified these effects, as shown by post-1962 changes in the industry's R&D characteristics. Accordingly, one should judge them to have been significant contributing factors to the continuance and ac celeration of the structural trends described by the foregoing data.

V. IMPACTS On Firm Innovation StrategIES

The preceding two sections have described significant trends in the monetary and time costs of drug R&D and approval processes, and in certain structural features of the industry, namely R&D concentration and effective patent life. Both sets of factors directly and indirectly affect the abilities and incentives to innovate new drugs in the drug industry.

One major effect attributable at least in part to these factors has been extensively describe in the literature and already considered in this study: the decline in the rate of new drug innovation, measured either by all new drugs or NCES only. But the therapeutic impacts of the drug lag depend perhaps more on its effect on quality than on its effect on the number of new drugs. This effect has received considerable attention, which focuses on changes in the pattern of both R&D activity and in the resulting R&D out

comes.

Clymer pointed out one effect of more costly and time consuming development efforts on the pattern of new drug R&D in these terms:

Research programs today must be aimed at markedly superior, and indeed breakthrough, therapy, for it takes as long and costs as much to develop a compound representing only a slight improvement over existing therapy as it does one representing totally new therapy. All the major steps to demonstrate safety and efficacy will have to be carried out, even if only a single atom has been altered in the molecule of a standard agent. It is no longer economically sound to carry such projects through the long, costly, and sometimes just risky process if one can predict only marginal differences perhaps a slight increase of efficacy or a slight reduction in side effects."

36 Clymer, "The Changing Costs of Pharmaceutical Innovation,” supra note 12, at

« iepriekšējāTurpināt »