The Convergence between For-Profit and Nonprofit Hospitals in the United States

Abstract

This paper provides evidence of the growing similarity in capacity of for-profit and nonprofit hospitals. In 1960, nonprofit hospitals maintained on average more than three times as many beds per hospital as their for-profit counterparts; following a monotonic decline in relative size, by 2000, the average nonprofit hospital was only 32% larger than the typical for-profit hospital. Hospital level data for the United States indicate that the convergence was driven primarily by industry-wide effects such as entry, exit and ownership switches, rather than expansions or downsizing of existing hospitals. These findings suggest that hospitals may in fact strategically choose their ownership type (nonprofit vs. for-profit status) and hence, their regulatory environment. Accordingly, I develop a model in which firms have identical objectives but differ in their ability to benefit from a given ownership form. In contrast to the existing literature, this approach relies neither on different ownership type-specific objectives nor on market failure to generate an equilibrium in which both ownership types are chosen by a strictly positive fraction of hospitals. Changes in the economic environment alter firms' incentives to maintain a given ownership type. This in turn induces firms to modify their capacity and encourages some firms to switch their ownership type. Crowding-out of government hospitals, population growth and increasing involvement of the government in the healthcare market may account for the convergence in size. Policymakers and legislators often exert pressure on nonprofit hospitals by tying tax-exemptions to hospital-level measures of community benefits such as free care for the indigent. I argue that by omitting industry-wide effects of a hospital's tax-exempt status on price and industry output, such pressure may both lead to convergence in size and be welfare decreasing. Analysis at the state and Metropolitan Statistical Area (MSA) level as well as at the hospital level corroborate the principal theoretical predictions.