Jeff Goldsmith: “Hospital mergers kill”:  A case study in reality distortion

Last summer, The Wall Street Journal published an article reporting on a study by academic economists at Yale and the University of Chicago that reviewed the impact of hospital mergers.^a The study’s authors argued that hospital mergers triggered a wave of layoffs by employers in surrounding communities. That finding caught my interest, so I read the actual study, which was published by the National Bureau of Economic Research.^b

What I read made my blood boil.

This study, which was not peer reviewed, suggested that mergers are conspiracies between hospitals to extract “economic rents” from employers in their communities without creating any value. Their sample of 304 mergers between 2010 and 2015 were said to have resulted in an average 1.2% increase in local employers’ health benefits costs (which are, in turn, 9% of their total wage expense). 

The authors argued that this one-tenth of a percent increase in employment costs resulted in a wave of immediate worker layoffs in the surrounding communities, and that one in 140 laid-off workers in those communities ultimately committed suicide or died of drug overdoses. Using the same assumptions, the authors concluded that hospital rate increases “caused” 10,000 deaths in their communities over a seven-year period.  As author and columnist Dave Barry would say, “I am not making this up.”

Why hospitals merge

Hospital strategy has long been a core professional interest of mine. In 1980, while at the University of Chicago, I wrote a book, Can hospitals survive? The new competitive health care market , which predicted a wave of industry consolidation. I am not a fan of or advocate for hospital mergers. In fact, in the ensuing 43 years, I have served as an adviser to more than a dozen community hospitals that wanted to remain independent. After reading their paper, I struggled to understand how the authors could have reached this absurd conclusion.

There are a lot of reasons why hospitals and health systems merge.  But in my consulting experience, hospital mergers are rarely “conspiracies among equals.”  Rather, in major part, they are acquisitions of struggling hospitals or health systems by a larger enterprise. In many cases, the acquired hospital or health system had been damaged by economic troubles in its community or region. The result: too many public-funded or uninsured patients whose care does not generate enough cashflow to sustain the organization.

Indeed, a study by Michael Chernew, chair of the Medicare Payment Advisory Commission, concluded that adverse payer mix was a primary contributor to hospital consolidations. Chernew wrote that “hospitals with a higher share of Medicare patients had lower and more rapidly declining profits and an increased likelihood of closure or acquisition compared with hospitals that were less reliant on Medicare.”^c

Hospital boards and management of struggling hospitals bend over backward to avoid closing, and mergers are often their only alternative.

I can also tell you from the experiences of colleagues in consulting and private-equity who work on hospital mergers that the 1.2% average post-merger rate increase to health insurers the authors found doesn’t even cover the transaction costs of the merger (including, for example, merger brokers, consultants, legal and actuarial work and systems integration), let alone generate free cash flow for the merged entities.  If hospital mergers are all about exercising market power, why not a 15% or 20% increase or more?  A mere 1.2% average increase is a pathetic exercise of market power and strongly suggests other forces at work.    

Further, the economic effect of a 1.2% health cost increase is dwarfed by the costs to a community, both in job losses and suffering, of a hospital closing.^d Hospitals are frequently the largest employer in their communities.

Also, in the real world, we know that employers will bend over backward before  parting company with experienced workers. They do many other things to cope with rising operating costs, including pressuring suppliers, improving efficiency, raising prices, reducing employee hours and increasing their health cost sharing with employees, (the last of which almost doubled nationally during the study period, according to KFF).^e None of these effects were taken into account by the authors’ econometric model.

Regarding the connection of hospital mergers to deaths of despair, the absence of a control group of comparable communities without hospital mergers damages the inference of an actual causal link. The authors made no traceable efforts to tie these deaths to the affected companies’ having laid off workers. They also did not control for the other factors that lead to deaths of despair — factors including divorce rates; broad-based declines in employment opportunities; closures of churches, schools and/or community-based services; declines in the availability of mental health services; and, crucially, the sudden arrival of deadly fentanyl in local markets. 

Drivers of concerns over hospital consolidation

Concerns over consolidation and health costs go back 50 years. A glut of hospital beds was cited by economists as a major cause of explosive post-Medicare health cost growth. Indeed, consolidation offered a chance to reduce what was then viewed as a hospital bed surplus.^f When I wrote Can hospitals survive?, about 30% of hospitals were part of systems and the country had experienced more than a decade of double-digit health cost growth.^g  I predicted, incorrectly as it turned out, that most hospitals would be part of systems by the end of the decade.

Hospital consolidation has continued more or less uninterrupted for the ensuing 40 years (with notable surges after the two big Democratic health reform initiatives by Clinton and Obama). As of 2022, about two-thirds of hospitals were part of health systems.^h And health cost growth for most of the past 15 years has been less than half of what it was when I wrote my book.ⁱ The COVID-19 pandemic raised serious concerns about whether we have enough hospital capacity, with patients being treated in tents in parking lots.^j

What accounts for the discrepancy between what my on-the-ground industry experience has taught me and the allegations of cold-hearted economic exploitation presented by the authors’ recent study? The most charitable explanation is the behavioral economics concept of confirmation bias, where strongly held prior beliefs cause an observer to screen out other possible explanations for a phenomenon.    

Confirmation bias would explain why the authors, already on record condemning hospital mergers in prior studies, single-mindedly sought to argue yet again that mergers were damaging to communities, rather than comparing the effects of a merger with those of hospital closures in comparable communities, the obvious alternative scenario.^k

However, there may have been more-direct influences. The merger study was funded by Arnold Ventures, a for-profit entity created by a multi-billionaire retired Enron gas trader named John Arnold. Arnold has mounted a multi-hundred-million-dollar campaign — and that is the correct word — to influence federal health policy, including policies that would restrict hospital mergers.

Arnold Ventures generously funds research by respected academics at places like Yale, UC Berkeley, Johns Hopkins and Harvard, which have previously advocated policy positions that support its agenda. Publication is followed by a wave of press releases, social media posts, industry conferences and Congressional hearings. The studies Arnold funds are part of a targeted campaign of policy advocacy masquerading as disinterested academic inquiry. The Wall Street Journal article on the study made no mention of Arnold’s role, but it is too significant to ignore.

A need for transparency

I strongly believe in evidence-based health policy. But the quality of that evidence is critical to making the right choices, and who sponsors research is material to its quality and value.  

The evidence also needs to pass the test of real-world plausibility. In the instance of this paper, neither the logic of how employers decide to cope with operating cost increases or the linkage of mergers to layoffs or deaths of despair meet this standard.

Based on my experience as an advocate of hospitals remaining independent, I can tell you independence isn’t always possible. So the right policy question is not, “Is consolidation increasing health costs?” Rather, it should be, “Which is more costly to society, hospital consolidation or hospital closures in struggling communities?”  

The real world of healthcare markets is far more complicated than a spindly web of artfully constructed regression equations would suggest.  

Author’s note: The research for this article was supported by the Federation of American Hospitals, but the opinions expressed herein are mine alone. Trevor Goldsmith provided research assistance, and Rich Bajner and Keith Pitts added insights into the financial factors in consolidations.

Editor’s note: For an extended discussion of this issue, on which this ccommentary is based, see Jeff Goldsmith blog post “‘Hospital Mergers Kill’: An Economists’ Exercise in Reality Distortion,” published Jan. 7 on The Health Care Blog.

Footnotes

a. Evans, M., Mollica, A., and Ulick, J., “When hospital prices go up, local economies take a hit,” The Wall Street Journal, June 23, 2024.
b. Brot-Goldberg, Z., et al., Who pays for rising health care prices? Evidence from hospital mergers, Working paper, National Bureau of Economic Research, June 2024.
c. Chernew, M, et.al., “Public payment rates for hospitals and the potential for consolidation-induced cost shifting,” Health Affairs, August 2021.
d. Holmes, G.M., et al., “The effect of rural hospital closures on community economic health,” Health Services Research, Jan. 4, 2006.
e. Kaiser Family Foundations and Health Research and Educational Trust, Employer Health Benefits Survey,  September 2016.
f. Pasley, B.H., Lago, R.J., and Marshall, N.O., “Excess acute care bed capacity and its causes: The experience of New York State,” Health Services Research, April 1995.
g. Catlin, A.C., and Cowan, C.A., History of health spending in the United States, 1960-2013, CMS, Nov. 10, 2015.
h. Godwin, J., Levinson, Z., and Neuman, T., “One or Two Health Systems Controlled the Entire Market for Inpatient Hospital Care in Nearly Half of Metropolitan Areas in 2022,” Kaiser Family Foundation, Oct. 1, 2024.
i. Altarum, Health Sector Economic Indicators, Spending Brief, July 18, 2024.
j. Abelson, R., “Covid overload: U.S. hospitals are running out of beds for patients,” The New York Times, Nov. 28, 2020.
k. Cooper, Z., et al., The price ain’t right? Hospital prices and health spending on the privately insured, Working paper, National Bureau of Economic Research, December 2015.