Issue No. 5: How Industry Concentration Affects Health Costs

New Research on the Balance of Bargaining Power Between Insurers and Providers 

With the country poised for a new administration, Washington is talking health care –especially health-care costs. 

At this point, no one really knows how – or even whether -- the Affordable Care Act will ultimately be changed and what those changes will mean for coverage and access to health care.  But we can be sure that, no matter what we do with coverage, if we don’t address expenditures, the big problems will continue.

The anxiety over health care costs is real and immediate. Members of Congress hear about it from constituents, and so will the new administration. But this is a time for reasoned understanding. We can’t bend the health-care cost curve until we recognize what’s making it rise.

As we have noted before, pharmaceutical costs may get the headlines, but they represent only about one-eighth of health care spending. More important, if patients adhere to their prescriptions, drugs can reduce other health costs – the costs that consume a much higher proportion of spending. 

Where the Money Is

It makes sense to focus on all of the costs to really save money. The most recent statisticsfrom the Centers for Disease Control show that the U.S. spends, in round numbers, $1 trillion on hospitals, $600 billion on physicians and other professionals, and $300 billion on prescription drugs (plus another $1 trillion or so on everything else: nursing facilities, dental, home health care and the like).

If we could cut hospital and out-patient spending by 5%, the total savings would be $80 billion a year. If we could cut pharmaceutical spending by the same proportion, the savings would be $15 billion. It makes sense to go where the money is. 

That is why we were intrigued by an article on market concentration that appeared in the January issue of Health Affairs. Written by Eric Roberts, Michael Chernew and J. Michael Williams – all of the Harvard Medical School – the piece starts by pointing to “renewed concerns about the effects of market concentration on commercial health care prices” as the result of “proposed mergers among large US health insurers and growing consolidation among providers.” Four of the five largest insurers – Aetna, Anthem, Cigna, and Humana – have proposed mergers. And mergers have been sweeping the provider world, as well.

It stands to reason, the authors write, that insurers that can “channel a large number of patients to providers…negotiate lower prices with providers.” But, on the other hand, “a large body of research indicates that consolidation in the provider market leads to higher prices.”

So the authors took a close look at the effects of concentration (both from the insurer and provider sides) on prices – specifically on claims for office-based physician services. They found, first of all, that concentration gives insurers “substantial bargaining power” over providers:

Using multipayer claims for physician services provided in office setting, we estimated that – within the same provider group – insurers with market shares of 15 percent or more (average 24.5 percent), for example, negotiated prices for office visits that were 21 percent lower than prices negotiated by insurers with shares of less than 5 percent.
Then, they looked at the providers themselves. The results showed that the large providers had far more market clout than smaller providers. For example, the price of a visit to a small practice was $72 for a patient insured by a company with a market share of 5 percent to 15 percent. But the price of such a visit to a large practice was $86, or 19 percent more.

It’s important to note that the trend in recent years has been for providers to get bigger and bigger, with hospitals swallowing up doctors’ practices. For example, a study by thePhysicians Advocacy Institute found that from July 2012 to July 2015, the number of physicians employed by hospitals jumped a stunning 48 percent. While the Roberts-Chernew-Williams study did not examine hospitals, previous research found that theaverage cost of the administration of an oncology drug is roughly twice as much in a hospital outpatient setting as in a physician’s office. 

The Price Differences Are Enormous

The researchers write that “large provider groups have bargaining power that mitigates insurers’ ability to pay lower rates – or, equivalently, that requires greater insurer market power to counteract.”

Their conclusion?

Our findings suggest that mergers between large health insurers, such as those recently proposed, could lead to lower negotiated prices with health care providers.


The authors do warn that “mergers might not ultimately reduce the costs of care borne by consumers because the savings that insurers realize from negotiating lower prices might not accrue to consumers.”


The point, however, is that, if we want to control costs, we need to gather the data – as the three Harvard researchers have – and make informed policy decisions. In the case of office visits, what is so remarkable is the variance in costs, depending on the competitive situation.

For example, an insurer with a large market share (greater than 15 percent) can procure an average office-visit cost of just $70 from a small-practice provider (as defined by the researchers). But an insurer with a small market share (less than 5 percent) must pay an average of $97 when dealing with a large-practice provider. That is a huge difference. The latter visit costs 39 percent more!

Of course, we can’t automatically reorder the office-based physician-services universe to a large-insurer-small-provider norm. But policy-makers can use this information to develop smart policies. That makes a lot more sense than screaming about the price of a specialty drug that cures a disease or saves a life.

Doggie Health Care vs. Human:
Some Surprising Lessons

Veterinary health care has a disturbing similarity to human health care. And it’s not just the sophisticated care that’s given. It’s the cost.

This is the surprising finding of two researchers, Liran Einav, professor of economics at Stanford, and Amy Finkelstein, an MIT professor who won the prestigious John Bates Clark Medal. The Harvard Business Review last week published a synopsis of their recent National Bureau of Economic Research working paper, which was titled, “Is American Pet Health Care (Also) Uniquely Inefficient?”

The authors find that, over the past two decades, pet health care has followed “a similar trajectory” to human care, with a “rapid growth in spending, growing far more quickly than spending on, say, housing or entertainment.” What is odd about this is that it’s hard to blame the structure and the institutions, or suggest that the solution is government price intervention, as we do with human care. As Einav and Finkelstein write:

The usual scapegoats – an inefficient insurance system and onerous government regulation – are practically non-existent in the veterinary industry. Almost no pets have health insurance, and regulation (or government intervention more broadly) is much less prevalent in pet health care than in human health care.

The researchers wonder, then, whether something other than public policy is the cause of rising costs in human health care. “Something else is at work,” they write. But they aren’t sure what it is.

They speculate that the answer may be “the nature of technological progress” – we may be spending more on human health care (as in pet health care) because technology has so vastly improved treatment. They posit another explanation as well: in both human and animal health care, our decisions on whether to deploy heroic measures to save the life of a 95-year-old woman or a 15-year-old poodle are often driven by emotion. “The nature of the decision – trading off potential health improvements against money – may make coldly rational cost-benefit decisions unlikely.”

What’s important about all this is that it broadens the aperture of our search. If we want to understand health costs, we need to look not just at policy and politics but also at drivers that are embedded in our humanity. Perhaps we spend more on health care because health care gives us what few other products or services can: a longer, better life.