No abstract

For more than a decade, industry analysts and policy makers have raised concerns about declining pharmaceutical innovation, citing declining numbers of new molecular entities (NMEs) approved in the United States each year. Yet there is little consensus on whether this is the best measure of "innovation." We examined NME approvals during 1987-2011 and propose the three distinct subcategories of NMEs--first-in-class, advance-in-class, and addition-to-class--to provide more nuanced and informative insights into underlying trends. We found that trends in NME approvals were largely driven by addition-to-class, or "me too," drug approvals, while first-in-class approvals remained fairly steady over the study period. Moreover, the higher proportion of first-in-class drug approvals over the most recent decade is an encouraging sign of the health of the industry as a whole.

Because consumer discrimination can reduce productivity, it is often impossible to tell whether differential productivity is the effect of discrimination or of differential ability. Detailed data for the sports labor market make it possible to separate consumer discrimination from ability. We use a unique approach to determine whether the entertainment value of baseball players is related to their race: we examine whether race directly affects the value of a player in the market for baseball cards. In contrast to studies that use salaries, there is no room for owner or coworker discrimination. Our evidence supports the hypothesis of consumer discrimination.* We acknowledge the helpful comments of William Dougan, Lawrence M. Kahn, Raymond Sauer, John Warner, and an anonymous referee. We alone are responsible for all errors and omissions. For an exception see Kamalich andPolachek [1982]. For a criticism of Polachek's approach, along with a reply, see Blau and Kahn [1985] and Kamalich and Polachek [1985]. discrimination by fellow workers, and discrimination by consumers. Employer discrimination, often discussed in the popular press, is highly unlikely to persist in competitive labor markets. If workers are mobile in the long run, self-sorting by workers will cause coworker discrimination to disappear in the long run. 2 A more likely source of the continuing wage gap, then, is consumer discrimination. Empirical studies indicate that such discrimination exists. Yinger [1986] found that it led to racial discrimination in Boston housing markets. The study of consumer discrimination in labor markets, however, faces a serious problem: consumer discrimination is difficult to measure. A key assumption in Becker's model is that blacks and whites are equally productive. Yet, consumer discrimination can directly reduce productivity, making it impossible to tell whether differential productivity is the effect of discrimination or of differential ability to do the job. For example, if black realtors sell fewer houses than white realtors, is it because black realtors know less about houses, or is it because house buyers and sellers discriminate against blacks? Unless ability can be measured, this question cannot be answered.Although ability is nearly impossible to measure in most labor markets, there is an important exception: the sports labor market. The appeal of sports for studies of discrimination is that it is possible to separate consumer discrimination from ability to do the work. Professional sports firms produce entertainment. The entertainment value of a sport is directly related to players' abilities and to the way consumers repond to those abilities. The detailed data on individual athletic performance permit the separation of consumer response from other measures of performance.3 For example, in a recent study of the National Basketball Arrow [1972] argued that nonconvexities in the cost of adjusting the firm's labor force could cause wage differentials arising from employer or coworker discrimination to persist ...

Warfarin, an anticoagulant commonly used to prevent and control blood clots, is complicated to use because the optimal dose varies greatly among patients. If the dose is too high, the risk of serious bleeding increases; if the dose is too low, the risk of stroke increases. We estimate the potential health benefits and resulting changes in healthcare costs should it become feasible to base personalized warfarin dosing decisions on appropriate genetic testing. Using different assumptions regarding the costs and effectiveness of genetic testing, we estimate that formally integrating genetic testing into routine warfarin therapy could allow American warfarin users to avoid 4500-22,000 serious bleeding events annually. Genetic-based therapy could also reduce the incidence of strokes among patients taking warfarin. We estimate that the additional cost per patient from integrating genetic testing into warfarin therapy could range from US$300 in our pessimistic case, to substantial healthcare cost savings in the optimistic case.

No abstract