• Pascal’s Wager? The Uncommon Relationship Between Drug Industry Profits and Research

    Tyler Cowen has cautioned in the NY Times that cost containment measures as part of health care reform may impact the US lead in medical research and development. But how much do private sector profits contribute to the pace of medical innovation?  This question is particularly salient to the US pharmaceutical industry, which is the largest private contributor to US medical R&D, and has agreed to provide up to $80 billion in cost savings over ten years in connection with comprehensive health reform. Will this modest hit to their profits detract from their investments in research and development?

    The standard corporate finance answer is no. In a well established business with access to capital markets, the link between profits and investment in research development is weak. So long as a business’s expected profits from research and development exceed its cost of capital by some amount, its investment in research and development will chiefly be limited by the law of diminishing returns. An increase in profits will not spur additional R&D investment, because the business has already invested all it profitably can. A decrease in profits to some point still above its cost of capital will not decrease R&D investment, because foregoing profitable investment opportunities will not increase profits.

    But it turns out this is not true for pharmaceutical companies. Pharmaceutical R&D investment does vary in close relation with profits (Scherer 2001, Health Affairs).  Why? Pharmaceutical companies do not lack access to capital markets. Why do they choose to make investments in R&D out of free cash flow that they won’t make with funds from other sources?

    I think the answer is this: Patent protection assures an enormous payoff to the inventor of a blockbuster drug. But the probability is remote and in any event difficult to estimate that any given research program is going to produce a blockbuster. Thus it is difficult to impossible to estimate the risk-adjusted return on investment of such a research program relative to cost of capital. So such research programs get funded cautiously if at all from the capital markets. Free cash flow, however, is close to cost-free money. Betting it on a blockbuster with free cash flow is a lot like Pascal’s wager: if you lose, you lose nothing; if you win, you win everything.

    I had the discussion that led up to these thoughts with a Facebook friend of a certain free market bent whom I took to be making the generally incorrect assertion that R&D investment varies with profit. Though I took great pains to explain the text-book reasons why that is not generally the case, he remained incredulous that I could not appreciate the obvious relationship. And the thing is, it was completely intuitive to him, because he works for a company – Google – that is also a special case. Google also operates in an industry where a single success can create blockbuster returns through first mover advantages and network effects. Not surprisingly, Google plows a lot of its free cash flow back into R&D on projects whose odds of success are hard to quantify.  It was completely obvious to my friend why it should do so.

    The pharmaceutical R&D investment model that also applies to software has been called “virtuous rent seeking.”  But the CBO, for example, has questioned whether the private incentives for such investments, in the pharmaceutical arena at least, are so strong that they produce overinvestment from the perspective of social utility.  And of course, that is the tragedy of Pascal’s wager: whether or not he lost nothing on the bet, society as a whole perhaps lost a great deal when he turned his mind away from science to God. In the end, it is not clear that any diminution of investment in R&D by the pharmaceutical companies as a result of health care reform will actually be missed.