Why Market Concentration Is Health Care’s Central Problem

For forty years, American health policy has generally leaned on a single load-bearing assumption: that competition among hospitals, physicians, and insurers can be harnessed to discipline prices and lift quality, sparing us the heavier hand of a regulated or single-payer system. It is very American to want the free market enterprise and the power of innovation and industry to solve our problems. But, the reality of the situation in many industries, including health care, is that competition (the primary driver of the cost and quality effects in the free market) is missing.

Competition, in most of the health care markets in the country, is a fiction. The markets that competition policy presumes simply do not exist across the bulk of American geography.

What exists instead is concentration so pervasive that “let the market work” has become less a strategy than an alibi for inaction. If we are honest about the empirical record, market concentration is not one problem among many. It is the problem, and it is the strongest available argument for abandoning the competition framework in favor of either direct regulation or drastic (and I mean drastic) efforts to increase competition—both of which require significant government intervention.

Lets consider the most recent and most striking finding. KFF’s 2024 analysis of hospital markets found that one or two health systems controlled the entire market for inpatient hospital care in nearly half, 47 percent, of U.S. metropolitan areas. In more than four of five metro areas, one or two systems controlled over 75 percent of the market, and nearly all metro areas (97 percent) were highly concentrated under federal antitrust thresholds. 97 percent! I’ll say it again, 97 percent!

This is not a health care market with a competition problem at the margins. It is the absence of competition as the baseline condition. And it is getting worse: 80 percent of metro hospital markets either became more concentrated between 2015 and 2024 or were already controlled by a single system the entire time.

This is the continuation of a long arc, not a sudden break. Cutler and Scott Morton documented in 2013 that the typical American region had only three to five consolidated health systems, with the top three share leaders capturing 77 percent of admissions and the top five capturing 88 percent. Crucially, they found that no hospital markets in the country qualified as “highly competitive,” and the hospital Herfindahl-Hirschman index (HHI), a measure of market concentration, had risen roughly 40 percent since the mid-1980s, which is the equivalent of a market shrinking from five independent firms to three.

Fulton’s parallel work, published the same year as Glied and Altman’s, found that by 2016 about 90 percent of metropolitan areas were highly concentrated for hospitals, 65 percent for specialist physicians, 39 percent for primary care physicians, and 57 percent for insurers, with 91 percent of the 382 metro areas he analyzed warranting antitrust concern for at least one of these markets. The share of primary care physicians working in hospital- or system-owned organizations climbed from 28 percent to 44 percent over just six years, a 57 percent jump, as systems absorbed independent practices.

Three independent research efforts, using different data and spanning more than a decade, converge on the same conclusion. The competitive market that competition policy assumes is the exception, not the rule.

The reflexive response is to call for more competition. We need to break up the systems, lower barriers to entry, let new players in!

But the structural forces driving concentration make this largely futile, a point Glied and Altman make with unusual candor. Inpatient hospital demand has been falling for decades. Cutler and Scott Morton note inpatient days dropped by a third between 1981 and 2011 even as the population grew and aged, and more than 15 percent of hospitals closed. Glied and Altman show that this decline has fallen hardest on midsize community hospitals, squeezed from below by nimble freestanding surgical and diagnostic centers (which face fewer regulatory hurdles, rarely unionize, and can refuse the uninsured) and from above by tertiary hospitals reaching down to capture cases that could be treated more cheaply locally.

For many of these midsize hospitals, the realistic alternatives are merger or closure and either way, competition shrinks.

Worse, the standard competition toolkit can backfire. Glied and Altman point out that lowering barriers to entry boosts competition for low-fixed-cost services like outpatient imaging, but by siphoning the profitable cases away from community hospitals, it can reduce competition for the complex services only full-service hospitals can provide.

Meanwhile, “must-have” hospitals, those perceived to deliver the best care for rare and serious conditions, where the volume-outcome relationship is real, wield decisive bargaining leverage. When these institutions form systems, they extend that leverage across community hospitals, clinics, and physician groups, forcing insurers to accept favorable rates system-wide as the price of access. Insurers shopping for narrow networks cannot escape this as the geography simply does not contain enough genuine substitutes.

This is the heart of the matter. Competition requires alternatives. Across most of America, for the services that matter most, alternatives do not exist and they cannot be magically manufactured.

The price evidence is robust and one-directional. Cutler and Scott Morton cite eight studies showing merger-driven price increases of 10 to 40 percent, and emphasize that nonprofit status offers no protection (e.g., prices rise just as much at nonprofit hospitals, which still pursue profit maximization and simply redistribute the proceeds differently).

Fulton’s review found hospital mergers in already-concentrated markets frequently raising prices by more than 20 percent. The insurer side reveals the mechanism’s cynicism. Fulton documents that higher insurer concentration does extract lower prices from hospitals and physicians, but those savings are not passed to consumers.

Instead, more concentrated insurer markets produce higher premiums for employers and individuals alike.

Concentration on both sides of the negotiating table is a wealth transfer from patients to consolidated firms, not a discipline on cost. This is a primary mechanism of what most Americans hate about health care and it is among, if not the top, problem to be solved.

The health care competition-believer’s last refuge is quality. Even if concentration raises prices, perhaps integrated systems deliver better care due to shared information technology and better care coordination. I used to believe this narrative, but I think about it differently now.

The evidence refutes this directly. Short and Ho’s analysis of 29 quality measures across thousands of hospitals found that increased market concentration was strongly associated with reduced quality across all ten patient-satisfaction measures.

Their illustration used in the paper is concrete. A merger taking a market’s HHI from 2,500 to 3,750 was associated with satisfaction declines of up to a full percentage point on measures like getting help when wanted. With fewer competitors, there is less incentive to keep patients content, and measures like clear communication about medications plausibly track real clinical safety. Now, I will note that this is not causal research and there can be many reason why this is the case.

Vertical integration fares no better. Short and Ho found that hospitals employing physicians improved quality on only a tiny subset of process measures, consistent with Glied and Altman’s observation that there is “little evidence” integrated systems improve quality, and with Fulton’s review finding that in some cases higher hospital concentration was associated with higher mortality. Structural integration on paper does not produce clinical integration in practice.

Concentration also corrodes the kind of improvement that matters most for long-run cost. Cutler and Scott Morton distinguish product innovation (which correlates with profits and survives consolidation) from process innovation (i.e., checklists, uniform protocols, more efficient ways of delivering care). Process innovation declines under market power, because dominant institutions face little pressure to undertake the difficult work of redesigning care when there is no competitor to lose patients to. The very efficiencies consolidation is sold on are the ones market power makes optional.

Here is the argument’s keystone, and it hides in plain sight in these very sources. Concentration wreaks its damage almost entirely on the private side of the system. As Cutler and Scott Morton note, public payers (i.e., Medicare and Medicaid) set prices administratively and do not negotiate (though there are other problems), so essentially every hospital accepts their rates regardless of market power.

In other words, the part of American health care that already operates on regulated administrative pricing is more insultated to the concentration problem.

Even the pro-competition authors concede the point. I, myself, am very pro competition. But we need to face reality.

Glied and Altman state flatly that antitrust enforcement alone “is unlikely to be sufficient.” It is slow, expensive, and backward-looking. It can sometimes block a merger but cannot conjure competitors into a region that can only support one tertiary hospital, and large systems can build their leverage organically when acquisition is barred.

Glied and Altman propose the following remedies: regulating prices for must-have services as a multiple of Medicare rates, requiring those facilities to accept all insurers, structuring competitive bidding across entire markets rather than within them. These are regulation, not competition. Cutler and Scott Morton arrive at the same place with tiered networks, bundled payments, and area-wide price or spending targets like Maryland’s and Massachusetts’s.

The case for competition in American health care has been tested against four decades of data. It is not working and it is robbing Americans of their livelihoods and wage growth. This is not because competition is a bad idea in the abstract, but because the structural prerequisites for it are absent across most of the country and cannot be magically restored because we want to or because politicians invoke the power of competition in their rhetoric

Concentration raises prices without bound, fails to improve quality and sometimes worsens it, dampens the process innovation that could bend the cost curve, and leaves patients with surprise bills and narrowing choices.

We can keep invoking the great powers of competition as a reason to avoid a regulated, single-payer or all-payer system. Or we can read the evidence we already have.

Unfortunately, it points, unmistakably, in one direction.