تعطیلی بیمارستان و بهره وری اقتصادی

We present a new framework for assessing the effects of hospital closures on social welfare and the local economy. While patient welfare necessarily declines when patients lose access to a hospital, closures also tend to reduce costs. We study five hospital closures in two states and find that urban hospital bailouts reduce aggregate social welfare: on balance, the cost savings from closures more than offset the reduction in patient welfare. However, because some of the cost savings are shared nationally, total surplus in the local community may decline following a hospital closure.

Communities often spend public funds to attract or retain private businesses because they believe that the benefits in jobs and local spending will offset the tax expenditures.1 It is less common for a community to prop up a local business that is on the verge of bankruptcy. Hospital closures are an important exception. When a hospital closes, its patients must turn to more distant and less familiar alternatives. For these reasons, public outcry typically accompanies the announcement of a pending hospital closure. In contrast to most other industries, the outcry often spurs governments to intervene to keep ailing hospitals afloat. For example, in 1999, local officials in Quincy, Massachusetts, provided a $12.1 million bail out of Quincy Hospital to facilitate its acquisition by a nonprofit enterprise. Similarly, in 2000, officials in Tampa, Florida, authorized $3.5 million from local tax revenue to bail out Tampa General Hospital. In well-functioning markets, insolvency usually signals that a firm is inefficient, its product is in low demand, or both. Previous studies have indicated that this rule of thumb applies to hospitals. For example, in a study of closures in the mid-1990s, Lindrooth et al. (2003) found that hospitals that were destined to close had occupancy rates of around 48%, while their surviving rivals had occupancy rates that were typically over 64%. This suggests that local residents did not place much value on these hospitals and that alternative sources of care were available. The notion of bailing out a failing firm would not normally arise in the context of traditional business ventures, except perhaps as a political matter. But the hospital industry is comprised of many nonprofit and local government-owned hospitals, and industry is rife with problems related to moral hazard and adverse selection. Furthermore, the prices paid by traditional Medicare and traditional Medicaid for hospital services are generally set by fiat rather than by the market. Thus, for a sizable group of patients, prices do not necessarily adjust to supply and demand conditions. Although the prices paid by private health plans (along with Medicare and Medicaid managed care plans) are shaped by market forces, consideration of pricing alone may not result in hospital closures that are socially optimal. Consider the distortions created by the absence of the profit motive. We expect for-profit hospitals to exit markets when their costs of remaining in business exceed their ability to translate value creation into revenue (Wedig et al., 1989). This expectation likely does not apply to the nonprofit and government-owned hospitals that dominate the United States market.2Bazzoli and Andes (1995) and Duffy and Friedman (1993) lend support to this notion by showing that, in contrast to struggling for-profit hospitals, distressed nonprofit hospitals linger in the market despite financial difficulties. Other studies have shown that for-profit status is a significant predictor of exit (e.g., Ciliberto and Lindrooth, 2007, Succi et al., 1997, Wedig et al., 1989 and Williams et al., 1992). Finally, note that nonprofit hospitals usually receive oversight from community-based boards, whose interests in keeping a hospital open may diverge from both profit and welfare maximization. Specifically, a hospital board may care little about the effects of potential closures on local healthcare spending. Taken together, these results suggest that nonprofits may close less frequently than is socially optimal. Other market distortions may justify bailouts. Hospital markets are imperfectly competitive, hospitals cannot perfectly price discriminate, and some prices are regulated. Thus, the total social surplus generated by an unprofitable hospital may exceed its costs.3 For example, Medicaid payments normally meet or exceed variable costs (to encourage hospitals to admit Medicaid patients), but they often do not cover the average total cost of care. As a result, hospitals that rely on Medicaid payments may go bankrupt even when the value they create exceeds their cost of doing business. Conversely, an efficient hospital may linger even if its closure were to drive down local healthcare spending. Finally, we note that while the utility loss from a hospital closure is borne entirely by the local community, the cost savings are shared by the local government, the state government, and the federal government. Hospitals derive an average of 30% of their revenue from the federal Medicare program. As a result, the cost savings from shutting these hospitals would be shared with the federal government, though the local community presumably does not internalize the value of federal savings. Similarly, the federal and state governments will value the cost savings from the Medicaid program. Accordingly, we evaluate the merits of closures from both state and national perspectives. This discussion suggests that the merit of a particular closure or bailout is an empirical question that requires the measurement of both cost and utility effects. There is a robust literature on the former, which we employ herein. To measure the utility effects of hospital closures, we draw on previous models of the effects of changes in hospital market structure on consumer welfare (Town and Vistnes, 2001 and Capps et al., 2003 (“CDS”). Specifically, we build upon the option demand framework from CDS. CDS studied negotiations between hospitals and managed care organizations and developed an index of a managed care organization's enrollees’ willingness to pay (WTP) for the inclusion of a given hospital or set of hospitals in their network.4 We do not consider the effect of outcomes beyond what is measured and reflected in a patient's ex-ante utility function. WTP captures the full utility benefit of access, including the quality of the hospital under consideration, convenience afforded by an attractive location, and consumers’ idiosyncratic preferences for that hospital. In the current context, we require a dollar-denominated estimate of the lost utility from a hospital closure that we can compare to the attendant cost savings. To achieve this, we develop a method for computing the equivalent variation of the utility effects of a hospital closure. Weber (2009) measures the effect of access to ambulatory surgery centers on welfare using a method that more closely follows that of Small and Rosen (1981). Our approach differs in that we incorporate option-demand for the hospital's services by integrating over the distribution of expected illness for the entire population. The resulting welfare effect of closure is converted to a ‘travel time equivalent’ which is the total number of hours that would need to be driven to reduce utility by the same amount as the closure. Our results indicate that, in general, urban hospital bailouts reduce aggregate social welfare: the cost savings from the closures we studied more than offset the reduction in patient welfare. However, we also found that because some of the cost savings are shared nationally, several of the closures led to a decline in total surplus in the local community. We conclude with a discussion of the implications of hospital closures on affected populations’ access to care.

We combined cost and demand estimates to evaluate the impact of hospital closures on economic efficiency. The results, when applied to five closures in three mid-size urban markets in the late 1990s, indicate that these urban hospital closures increased total welfare. Importantly, this does not imply that most hospitals are not socially valuable. Rather it is a statement about how well market forces select which hospitals to close in a market rife with imperfections, including weak price signals and third-party payers, moral hazard, and adverse selection. Our estimates indicate that, despite these imperfections, a hospital that is unable to at least break even is also failing to create more value than cost. The conclusion is that policy-makers should, absent unique circumstances, resist the pressure for a bailout that commonly accompanies a closure announcement. However, the fact that reductions in hospital costs are shared between local and federal payers, while access issues are fully local, tilts the local community's calculus in favor of bailouts in several cases. Several caveats about the generalizability of our results apply. In each of these markets, occupancy rates varied from 55% to 63%-well below full occupancy. Thus, in the cases we studied, there is sufficient capacity at relatively proximate surrounding hospital to absorb the additional patients. The fact that there are low occupancy rates in these markets is not surprising, because closure is most likely to occur in precisely those markets that have excess capacity. We also did not account for changes in hospital prices that might result from closure. We do not think this is an issue in the markets and hospitals we study, because the closing hospitals were small and, at least at the time, each market was relatively competitive. In addition, we have not allowed for a dynamic response to closure by the other hospitals in the market beyond increasing beds. The surrounding hospitals may also change service offerings to fill a niche left void by the closure of a competitor. This may lead us to overestimate the welfare-reducing aspects of closure. Furthermore, we have not considered the effects on physicians and hospital employees. There is also a possibility that a closure could lead to lengthier travel times to emergency departments, and in doing so, worsen outcomes. Buchmueller et al. (2006) found that closures in Los Angeles led to increases in the probabilities of death from Acute Myocardial Infarction (AMI) and from unintentional injuries. The magnitude of their results is quite large: a one-mile increase from a given zip code to the nearest hospital increases the AMI mortality rate by 6.5% and the injury mortality rate by 11–20%. This result is consistent with a report by the American Heart Association that states the survival probability for cardiac arrest decreases by 7–10% for every minute without treatment (American Heart Association, 2003). The report identifies the ability to reach the victim in time to reverse cardiac arrest as the key factor. Thus, if necessary, emergency medical services should be enhanced to eliminate increases in emergency response times in the wake of a closure. Doing so would help avoid some of the adverse consequences reported by Buchmueller and colleagues. This can best be accomplished by targeting response times rather than by propping up a hospital that generates more costs than consumer surplus. The methodology that we have laid out in this paper should permit a more comprehensive assessment of the merits and drawbacks of subsidizing struggling hospitals. While our results suggest that bailouts are not socially efficient for hospitals in our sample, they also suggest that the substantial federal and state funding of local healthcare spending can drive a wedge between local policy and the social optimum. Current health reform proposals may create substantial changes in the balance of federal, state, and local funding of healthcare and increase the size of the wedge. If payments to hospitals continue to decline, more and more communities may opt to prop up their local hospitals at a cost to taxpayers nationwide.