Valuing charity.

Arrow asserted that a variety of institutional arrangements and observable mores of the medical profession were functional responses to the failure of the market to insure against uncertainties. But one of these norms--the ethic to provide treatment without regard to ability to pay--was also a response to the failure of the political system to assure the elderly and poor would not suffer more than others when they got sick. This ethic is strikingly different from the norm in most other areas of the economy. Automobile dealers and department stores are not expected to give away their products to the poor; neither are grocery stores or farmers. Public education is a closer analogy, reflecting the norm that all children deserve a good education. In education, however, unlike in medicine, we collectively support this norm by providing public funds to accomplish this goal rather than by relying on the private market. In 1963, physicians argued that a combination of the market and private philanthropy (including the obligations of physicians) would be sufficient to guarantee high-quality care for the elderly and the poor. Government financing, they argued, would lead to socialized medicine, impairing relationships between physicians and patients and between physicians and society. Based on his article, Arrow would not have agreed. Neither, apparently, did the public. The enactment of Medicare indicated, in part, that many people understood, even in 1965, the extent to which treatment choices and outcomes were affected by ability to pay. Events since 1965 suggest that there is some tension between insurance and ethical responses to uncertainty despite Arrow's endorsement of both. I have argued here that Medicare and Medicaid further eroded the ethic that treatment should be available without regard to ability to pay by reducing physician willingness to provide charity care and by reducing the resources available to public hospitals and the interest of private teaching hospitals in providing care to the uninsured poor. Largely independent of Medicare and Medicaid, the increasing importance of pharmaceuticals and other services delivered outside of the hospital further strengthened the connection between treatment choices and ability to pay, and the growth of capitated payment systems made this connection salient to many insured patients and their physicians. In part, then, the AMA was correct: Medicare and Medicaid have contributed to the erosion of trust in physicians as incorruptible agents for patients. Some of this trust undoubtedly was misplaced, even in 1965, and trust alone was not sufficient to guarantee widespread access to medical care or to assure that treatment provided would take true social benefits and costs into account. Medicare and Medicaid, as well as the growth of prepayment insurance plans, represent institutional responses to the failure of the 1963 norms to accomplish societal goals. Still, as we have seen, these responses create their own challenges, and we continue to search for institutions that will allow widespread insurance to coexist with the physician-patient trust that Arrow correctly identified as an important response to uncertainties and information asymmetries in the medical care market.