Research

Working Papers

The Effects of Physician Practice Acquisitions on Hospital Competition (Job Market Paper)

Show Abstract Consolidation between hospitals and physician practices is a ubiquitous feature of health care markets across the US. This paper examines the impacts of hospital-physician practice acquisition on physician referrals, hospital prices, and welfare using detailed administrative claims data from the Massachusetts APCD. I find that the effects of acquisitions vary substantially across large and small hospital systems in the state. Among small and mid-size hospitals, acquisition leads to a 20 percentage point increase in within-system referral rates and a 6 - 20% increase in hospital prices. In contrast, physician practices acquired by a dominant hospital neither meaningfully shift referrals nor raise negotiated hospital prices. I then develop a bargaining model of competition and use the estimates to evaluate the effects of changes in physician practice ownership. The model demonstrates how changes in referral demand due to practice consolidation enhances the bargaining leverage of the acquiring hospital system. I then simulate the hospital acquisition of every physician practice in the state, and I find that hospital prices counterfactually would increase by .72 percent or nearly \$200 per admission, though effects are larger across non-dominant firms. Higher payments to hospitals decrease insurer surplus by 19% or approximately \$18,000 per enrollee. Patient welfare declines by 22%, or \$185 per individual, due to shifts in physician referrals post-acquisition.

Publications

The Anatomy of a Hospital System Merger: The Patient Did Not Respond Well to Treatment
with Martin Gaynor, Adam Sacarny, Raffaella Sadun, Chad Syverson. NBER Working Paper No. 29449, accepted at The Review of Economics and Statistics

Show Abstract Despite the continuing US hospital merger wave, it remains unclear how mergers change, or fail to change, hospital behavior and performance. We open the “black box” of hospital practices through a mega-merger between two for-profit chains. Benchmarking the merger’s effects against the acquirer’s stated aims, we show they achieved some of their goals, harmonizing electronic medical records and sending managers to target hospitals. Post-acquisition managerial processes were similar across the merged chain. However, these interventions failed to drive detectable gains in performance. Our findings demonstrate the importance of organizations for merger research in health care and the economy more generally.

Selected Work in Progress

Impact of Increasing Hospital Price Transparency: Collusion, Search, and Equilibrium Price, with Elena Prager, Maryam Saeedi and Robert Town.

The Effects of Hospital-Physician Integration on Referrals: Evidence from Medicare
Media coverage: Wall Street Journal

Show Abstract I examine the effect of hospital-physician practice acquisition on outpatient physician referral choice using the universe of Medicare claims data from 2007-2012. I use panel difference-in-difference regressions to document a .07 to .10 increase in the av- erage probability that an acquired physician refers to the owning hospital post-merger.This effect represents a nine-fold increase relative to the baseline probability of physi- cian referral to the owning hospital and remains stable over two years after the acquisi- tion date. Moreover, acquired physicians’ increase in referrals to the owning hospital is driven almost entirely by a reduction in their referrals to independently owned physi- cian practices.

Medical Malpractice Reform and Ob-Gyn Entry

Show Abstract The 2003 Texas Medical Malpractice Reform Act reduced punitive damage caps by 50 percent and imposed other measures that dramatically increased patients' costs of claiming malpractice. This potentially generates an incentive for physicians to enter and strategically locate in isolated areas with few competitors by lowering physician' costs of entry. I document obgyn entry and location patterns in Texas from 1997-2010 using physician counts data from the Texas Medical Board, the state licensing entity. With the passage of the law, Texas medical malpractice insurance rates for obgyns decline by 8 percent on average. This led to a 6 percent increase in obgyns per capita across the state. The probability that the most isolated counties have an obgyn increases from .166 in 1997 to .258 in 2010. Further, I estimate an entry model following Bresnahan and Reiss (1991) to show that the population required to sustain up to the first three obgyn entrants in the average Texas county is lower in the post-reform period.