Files
Abstract
On September 30th of 2004, Merck & Co. Inc. made the rather surprising announcement that it was withdrawing Vioxx due to its increased cardiovascular risk. The withdrawal of Vioxx was surely a spectacular example of catastrophic product failure, but by no means unique. In fact, according to the FDA, 14 prescription drugs have been withdrawn from the market since 1996. In these catastrophic product failures, although the immediate financial impact is obvious with the loss of revenue, the spillover impact on other brands of the firm and on competing brands is not clear. This dissertation focuses on the financial and market impact of drug withdrawals in the U.S. market. In this dissertation, a conceptual framework was developed in order to understand the impact of catastrophic product failures. A comprehensive database were compiled, and used to empirically examine 1) the effect of drug withdrawal on parent companys stock price, 2) the spillover effects of drug withdrawal on the sales of other brands in the parent companys portfolio, 3) the spillover effects on the marketing effectiveness of the parent companys marketing programs, 4) the spillover effects on the sales of competing brands in the therapeutic class, and 5) the spillover effects on the marketing effectiveness of these competing brands. Using intervention analysis and mixed effect models, the two largest drug withdrawals between 1996 and 2003, namely the Rezulin and the Baycol withdrawals, are thoroughly analyzed. The results from modeling these drug withdrawals provide evidence for 1) significant negative effect of drug withdrawal on the parent companys share price, 2) negative spillover effects on the sales of other brands of the parent company, 3) negative spillover effects on the marketing effectiveness of the parent company, 4) positive spillover effects on the sales of competing brands, and 5) negative spillover effects on the marketing effectiveness of competing brands. The findings from these two drug withdrawals are verified by replicating the modeling exercise using several additional withdrawn drugs (i.e., Seldane, Posicor, Duract, and Raplon). The results from the replication are generally consistent with those reported in the Rezulin and Baycol withdrawals.