William B. Vogt

Competition Among Hospitals

We examine competition in the hospital industry, in particular the effect of ownership type (for-profit, not-for-profit, government). We estimate a structural model of demand and pricing in the hospital industry in California, then use the estimates to simulate the effect of a merger. California hospitals in 1995 face an average price elasticity of demand of -4.85. Not-for-profit hospitals face less elastic demand and act as if they have lower marginal costs. Their prices are lower than those of for-profits, but markups are higher. We simulate the effects of the 1997 merger of two hospital chains. In San Luis Obispo County, where the merger creates a near monopoly, prices rise by up to 53%, and the predicted price increase would not be substantially smaller were the chains not-for-profit.

Economics of nerve block pain management after anterior cruciate ligament reconstruction: potential hospital cost savings via associated postanesthesia care unit bypass and same-day discharge.

BackgroundAnterior cruciate ligament reconstruction is a complex outpatient surgical procedure often associated with pain. Traditionally, the procedure is performed under general anesthesia and often requires the use of the PACU. Refractory pain and/or nausea/vomiting occasionally leads to an unplanned hospital admission. In this study, the authors examine the associations of nerve block analgesia for these patients and its associated reductions in PACU use, hospital admission, and hospital costs. MethodsThis was an observational, nonrandomized study in which existing data regarding patients’ day-of-surgery outcomes were merged with hospital cost data. We reviewed a consecutive sample of 948 men and women who were in good health and underwent anterior cruciate ligament reconstruction in an outpatient surgery unit between July 1995 and June 1999. ResultsThe use of nerve block analgesia was associated with reduced PACU admissions to 18% and decreased unplanned hospital admission rates from 17% to 4%. Multivariate linear regression analysis showed that patients bypassing the PACU had an associated hospital cost reduction of 12% (P = 0.0001), whereas patients who needed hospital admission had an associated hospital cost increase of 11% (P = 0.0003). ConclusionsThe use of nerve blocks for acute pain management in patients undergoing anterior cruciate ligament reconstruction is associated with PACU bypass and reliable same-day discharge. Although the cost savings for this one procedure are unlikely to generate sufficient cost savings via staffing reductions, extrapolating these results to a large volume of all types of invasive outpatient orthopedic procedures may have the potential to create significant hospital cost savings.

The Volume-Outcome Effect, Scale Economies, and Learning-by-Doing

There is a large empirical literature documenting the existence of a positive correlation between the number of times a hospital performs a given surgical procedure and the rate of good health outcomes achieved by patients at that hospital receiving that procedure. Typically, it is found that mortality is lower in hospitals that perform more of a given procedure. This result has been found for a wide variety of different procedures, time periods, and locations (David M. Shahian and Sharon L. Normand, 2001; John D. Birkmeyer et al., 2002; Ethan A. Halm, 2002). The two leading explanations for this correlation are the “practice makes perfect” and “selective referral” effects (Harold S. Luft et al., 1987). In the practice-makes-perfect hypothesis, either learning-by-doing or qualityenhancing scale economies cause large hospitals to provide better quality care, improving outcomes. In the selective-referral hypothesis, hospitals with higher quality attract greater demand and therefore have a greater volume of patients. The direction of causality matters for policy. If volume causes outcome, then policies supporting centralization of procedures in a few facilities may make sense. Hospitals that specialize in treating one or a few conditions may have the benefit of producing better outcomes. Antitrust analysis of hospital mergers should probably consider any improved outcomes when evaluating the impact of the merger. If, on the other hand, causality runs from outcomes to volumes, then these issues are no longer relevant. Two recent papers have employed instrumentalvariables techniques using newly developed instruments to address this question. Both find that the direction of causation largely runs from volume to outcome (Gaynor et al., 2004; Gautam Gowrisankaran et al., 2004). These papers use instruments that exploit the geographical distribution of patients and hospitals and the well-established fact that patients strongly prefer hospitals closer to their homes (see Gaynor and Vogt [2000] for a review). If the causation runs from volume to outcome, then it also matters for policy whether the causation is mediated through a static scale economy, in which today’s volume affects today’s outcome, or via a learning-by-doing effect, in which today’s volume affects both today’s outcome and future outcomes. In the static story, it matters less at which hospital volume is concentrated: high volume at any hospital will generate good outcomes. In the learning-by-doing story, it matters more: shifting volume from one hospital to another will involve “stranding” the experience built up at the hospital losing volume. Thus, the benefits of the concentration of volume in one hospital would develop over time, reducing the net benefit. This would tend to make new specialty hospitals, mergers, or regulatory approaches that result in consolidation of volume in one hospital less attractive than they would be in the static story. † Discussants: Paul Gertler, University of California– Berkeley; Dana Goldman, RAND Corporation; Patricia Anderson, Dartmouth College.

ENTRY AND COMPETITION IN LOCAL HOSPITAL MARKETS

We extend the entry model developed by Bresnahan and Reiss to make use of quantity information, and apply it to data on the U.S. hospital industry. The Bresnahan and Reiss model infers changes in the toughness of competition from entry threshold ratios. Entry threshold ratios, however, identify the product of changes in the toughness of competition and changes in fixed costs. By using quantity data, we are able to identify separately changes in the toughness of competition from changes in fixed costs. This model is generally applicable to industries where there are good data on market structure and quantity, but not on prices, as for example in the quinquennial U.S. Economic Census. In the hospital markets we examine, entry leads to a quick convergence to competitive conduct. Entry reduces variable profits and increases quantity. Most of the effects of entry come from having a second and a third firm enter the market.

Are Invisible Hands Good Hands? Moral Hazard, Competition, and the Second Best in Health Care Markets

The nature and normative properties of competition in health care markets have long been the subject of much debate. In this paper we consider what the optimal benchmark is in the presence of moral hazard effects on consumption due to health insurance. Intuitively, it seems that imperfect competition in the health care market may constrain this moral hazard by increasing prices. We show that this intuition cannot be correct if insurance markets are competitive. A competitive insurance market will always produce a contract that leaves consumers at least as well off under lower prices as under higher prices.

A SIMPLE MODEL OF PHARMACEUTICAL PRICE DYNAMICS

Branded pharmaceutical firms use price and promotional strategy to manage public knowledge about their drugs. We propose a dynamic theory of pharmaceutical pricing and conduct an exploratory empirical analysis inspired by the theory. Our theory predicts a pattern of increasing prices and decreasing promotional activities over a drug’s life cycle. Prices are kept low and advertising levels high early in the life cycle in order to build public knowledge about the drug. As knowledge grows, prices rise and advertising falls. If the management of this stock is important enough, this tendency of prices to rise can overwhelm the price‐decreasing effect of entry by generic competitors late in the drug life cycle. We argue that our theory of price dynamics explains empirical regularities in this industry.

Hospital prices and market structure in the hospital and insurance industries

There has been substantial consolidation among health insurers and hospitals, recently, raising questions about the effects of this consolidation on the exercise of market power. We analyze the relationship between insurer and hospital market concentration and the prices of hospital services. We use a national US dataset containing transaction prices for health care services for over 11 million privately insured Americans. Using three years of panel data, we estimate how insurer and hospital market concentration are related to hospital prices, while controlling for unobserved market effects. We find that increases in insurance market concentration are significantly associated with decreases in hospital prices, whereas increases in hospital concentration are non-significantly associated with increases in prices. A hypothetical merger between two of five equally sized insurers is estimated to decrease hospital prices by 6.7%.

Entry and Competition in Local Hospital Markets

There has been considerable consolidation in the hospital industry in recent years. Over 900 deals occurred from 1994-2000, and many local markets, even in large urban areas, have been reduced to monopolies, duopolies, or triopolies. This surge in consolidation has led to concern about competition in local markets for hospital services. We examine the effect of market structure on competition in local hospital markets -- specifically, does the hardness of competition increase with the number of firms? We extend the entry model developed by Bresnahan and Reiss to make use of quantity information, and apply it to data on the U.S. hospital industry. In the hospital markets we examine, entry leads to a quick convergence to competitive conduct. Entry reduces variable profits and increases quantity. Most of the effects of entry come from having a second and a third firm enter the market. The fourth entrant has little estimated effect. The use of quantity information allows us to infer that entry is consumer-surplus-increasing.

The Effects of Defined Benefit Pension Incentives and Working Conditions on Teacher Retirement Decisions

The retirement behavior of Pennsylvania public school teachers in 199798 and 199899, a period when state early retirement incentives were temporarily increased, is modeled using a choice framework that emphasizes both pecuniary and nonpecuniary factors of the retirement decision under a defined benefit retirement plan. We find each to have large and statistically significant effects on the decision to retire. The present value of inflation-adjusted pension benefits of a public defined benefit plan is found to be an important and sizable determinant of retirement. A

Hormone replacement therapy and cardiovascular health in the United States.

1,000 (or .4 percent) increase in the real present value of pension benefits is estimated to increase the probability of retirement for female teachers by .02 to .08 percentage points; this implies an elasticity of retirement for female teachers with respect to the present value of real pensions of between 2.0 to 3.5. These estimated defined benefit pension elasticities for female teachers are higher than for male teachers, whose comparable retirement elasticity was 1.9 to 2.5. A