Grabowski and Vernon

Grabowski and Vernon have investigated other periods and time cohorts of new introductions and found thaf fhey are characferized by similar patterns. ... 

To illustrate this point, Grabowski and Vernon examined the new product sales for the U.S. drug companies that spent between 300 and 500 million on their global R D in the mid-1980s (the top tier group in that period). They found that subsequent new product sales emanating from these R D efforts varied between 100 million and 3 billion (after 7 years of marker life). ... 

Grabowski and Vernon found that relatively few NCEs are marketed with effective parent lifetimes of less than 10 years. 

To be sure, simple counts of new products introduced do not encompass either their medical or their economic importance. Some introductions are invariably more important than others. Indeed, as Grabowski and Vernon (1990,1994) point out, there are major differences in sales among products so that the relevant distributions are highly skewed. Furthermore, pharmaceuticals have various indications, and the medical and/or economic importance of particular drugs can be more closely related to the number of approved indications than to the number of products. 

On this point, Grabowski and Vernon (1990,1994) estimate rates of return from investment in pharmaceutical research and development, and report values slightly higher than the associated cost of capital. In their latter study they conclude the estimated mean return on pharmaceutical industry new chemical entity (NCE) introductions forthe first half ofthe 1980s was 11.1% compared with the estimated (real) cost of capital of 10.5% over the same period (p. 404). This finding also suggests that one is unlikely to find major unexploited opportunities. Still, without determining a social optimum, one cannot make firm conclusions about the sufficiency of resources directed toward these activities. 

This framework was originally employed by Grabowski and Vernon (1986) to simulate the potential effects of the Act on R D returns when the Act was first passed. 

Grabowski and Vernon (159) also used published aggregate R D expenditure data to estimate the cost of successful drug development. Though Grabowski and Vernon did not estimate development time profiles with statistical analysis, their estimate provides another point of reference for comparison among methods, and it is also summarized here. 

R D expenditures in each year between 1962 and 1978 to product introductions in the years 1970-79 using assumptions about the application of each year s expenditures to the future years introductions. For example, Grabowski and Vernon assumed that in 1965, 10 percent of R D expenditures for NCEs was spent on drugs introduced in 1970, 10 percent on drugs introduced in 1971, etc.12... 

Whereas Hansen s total estimated cost in 1990 dollars with a 9-percent discount rate is 114.8 million for drugs entering testing in the period, Grabowski and Vernon estimated an average cost of 142 million. For NCEs approved in 1975, Grabowski and Vernon estimated cash R D outlays of 86.7 million in 1990 dollars compared with 65.5 million estimated by Hansen. 

Both studies used Kaplan-Meier survival curve analysis (2 19,225) to estimate the ultimate success rate in the NCE cohort under study. 25 Grabowski and Vernon s estimate of R D cash costs is less useful for corroborative purposes than Wiggins estimate because the... 

Three such studies are reviewed here.2Joglekar and Paterson (215), Grabowski and Vernon (160) and Virts and Weston (500) estimated the NPV of returns on R D investment in different samples of NCEs. Table 4-1 summarizes the main assumptions and findings of each study. 

Grabowski and Vernon used the sales history of NCEs introduced between 1970 and 1979 to estimate the returns on this group of NCEs. The researchers estimated the total R D cost (capitalized to the point of market approval at 9 percent) for this group of NCEs at 125 million (in 1986 dollars). Grabowski and Vernon s assumptions about production and distribution costs are similar in many respects to those of Joglekar and Paterson s, but Grabowski and Vernon included substantial extra costs in the early years of product life to cover expenditures for facilities, equipment, advertising, and promotion associated... 

Earlier studies by Baily (32), Scbwartzman (372), and Slatman (401) also e xamined returns on R D, but these studies used industry-level data on R D expenditures, production of NCEs, and sales. These studies also cover an earlier period consequently, they are not reviewed in this report. Another study by Grabowski and Vernon (157) is essentially an early version of their study reviewed here. 

NCEs. Grabowski and Vernon, on the other hand, added a substantial expense in the first 3 years of product sales to cover the additional advertising and promotion expenditures associated with product launch, but adjusted the contribution margin to reflect lower expenses in later years (154). 

Finally, assumptions about actual cash outlays for manufacturing plant and equipment vary widely among the studies. Grabowski and Vernon effectively assumed a much higher total investment than did the authors of either of the other studies. 

OTA s approach is similar to Grabowski and Vernon s (160), but OTA s assumptions vary in important respects. Where the available data are imprecise or scant, OTA used a range of estimates reflecting the best available evidence. In addition, when uncertainty was high, OTA used conservative assumptions that would tend to understate returns on R D. 

The displayed figures are estimates and may fluctuate between one leader to another and according to the size of the pharmaceutical company (accordingto DiMasi, Grabowski, and Vernon, 1995 DiMasi, Hansen, and Grabowski, 2003). 

---