Princeton economist Uwe Reinhardt published an op-ed in Friday’s New York Times suggesting that the $15 billion annual government subsidy to teaching hospitals for graduate medical education — mostly for residency programs — is inappropriate because a medical education is a private good, not a public good, the cost of which can be recouped by the doctor in the marketplace. The argument demonstrates the problem with assuming first best, “perfect competition” conditions in a decidedly second-best world riddled with regulations, price controls and subsidies at almost every turn. I’ll point out two problems with Reinhardt’s argument, although many other objections could be made.

The most obvious problem with Reinhardt’s argument is that the government, not the market, sets compensation rates for Medicare and Medicaid patients that are much lower than we see in the (relatively) unregulated private market. This suggests that without government subsidies for training, students who serve any patients with government insurance (and the majority do) should be willing to purchase only an inefficiently low amount of medical training (since they can’t recover in the market the optimal investment in training).  If regulators allowed students who purchase a suboptimal level of training to become doctors, this would be bad for all of us.  If regulators refuse to allow students unwilling to purchase the economically optimal level of training to practice medicine, because of artificial caps on reimbursements students won’t find the returns from a medical education sufficient to justify the costs, and we will have too few doctors.  This would also be problem for everyone.

Perhaps more importantly, potential medical students trying to forecast their potential income over the next 40 years lack perfect information, to say the least! Major changes in market and regulatory conditions are almost certain, but their implications are entirely uncertain. Thus, taking on huge amounts of debt now that one hopes to recover over several decades is extremely risky. Maybe the investment will pay off; maybe it won’t. Medical students already take on huge amounts of debt, even with government subsidies. If the subsidies were eliminated and, for example, medical residents were required to pay for their post-graduate training rather than earn a (modest) salary, only the most impervious to risk would invest in a medical education. Do we really want to discourage the more cautious among our bright college students (except for those who have family wealth to protect them) to avoid careers in medicine because of the riskiness of the capital investment any more than we already do today? Certainly some subsidies to partially mitigate the enormous cost of a medical education are warranted to ensure that smart people are willing to invest in becoming doctors and that they receive a sufficient amount of training, although the appropriate level of the subsidy is certainly open to debate.