ETHICAL DRUG INDUSTRY RETURN ON INVESTMENT*

Abstract

THE ETHICAL DRUG industry has been under continuous study by governmental investigators and academic researchers for some time. The focal point of many of these studies is the industry's relatively high rate of return on investment (ROI). High rates are often cited as an indication, if not proof, of noncompetitive behavior. However, there is little agreement regarding the definition of ROI, the determinants of ROI, and the proper method for making inter-industry ROI comparisons. These issues were examined and a model was developed to study the ethical drug industry's return on investment. Regression techniques were utilized to test the investment and structural characteristics as potential determinants of ROT. The hypothesized association with ROI was supported for the two risk measures and for growth in demand but rejected for the concentration ratio, average company size, and R and D intensity. The three measures of investment characteristics were combined into a cross-sectional, multiple regression equation giving a model of ROI which was used to study the drug industry's rate of return. Data for the industry were introduced into the model to obtain a forecast of rate of return with variance, skewness, and growth held constant. Comparing the forecast with the observed rate showed observed ethical drug industry average ROI to be below the rate that could result given the particular set of investment characteristics exhibited by the drug industry sample. Therefore, the hypothesis was accepted that ethical drug industry average ROI was the same as allmanufacturing industries average ROI after consideration of investment characteristics. There are several possible explanations for why this model can explain statistically the drug industry ROI: First, data were utilized only for firms primarily engaged in the manufacture of ethical drug products rather than data for both ethical and proprietary drug firms. Second, more recent data may reflect changed economic relationships, both within the drug industry and between the drug and other industries. Third, the model formulated included more variables than prior models. Fourth, explicit consideration was made of possible statistical errors in both the regression equation and in the ROI forecast.