HAGGLING FOR A PATENT - WHAT A GOVERNMENT WOULD HAVE TO PAY FOR PRESCRIPTION DRUG PATENTS

In previous papers (Guell, R. and Fischbaum, M. The Milbank Quarterly 1995; 73: 2 and Applied Economics Letters in press), we established th at the allocative inefficiency in the prescription drug industry is so pervasive that some remedy is warranted. In the first paper, we estim ated a lower bound on this inefficiency at approximately $3 billion, w ith an upper bound of approximately $30 billion; all on total industry sales of approximately $50 billion. In the second paper, on a narrowe r set of drugs for which sales were $8 billion, we more precisely esti mated this dead weight loss to be $5 billion. From this we showed that a system could exist whereby a government would purchase a drug paten t from a willing seller, freely distribute it and reap significant eff iciency benefits. In those papers we considered the possibility that t he innovating firm would not want to relinquish the patent, but we ass umed that they would be indifferent between being paid the expected ne t present value of future monopoly profits and reaping those uncertain profits over time. While that may be the case, a more likely scenario would be that a negotiation would take place. In the current paper I show the bargaining range that would exist under different risk prefer ence assumptions and show that this range widens as each side becomes more risk averse and narrows if the government threatens to use its po wer of eminent domain. Lastly, I acknowledge the risk of firms 'captur ing' the government agents doing the negotiation. To conclude, I prese nt the circumstances under which the proposed agency would likely impr ove societal welfare and contrast that with the circumstances where th e presently inefficient system would be made more so by government int ervention. (C) 1997 by John Wiley & Sons, Ltd.