The Incidental Economist

Recently, I explained that there are many possible causes of high market concentration. Furthermore, its effect on consumer price and market entry is theoretically ambiguous. In principle, dominant firms in concentrated markets (such as those in many health insurance markets today) could benefit from economies of scale or scope. If the resulting cost savings are not retained by the dominant firms, consumers may benefit from lower prices. This could happen if there exists the plausible entry of a potential rival that would undercut the dominant firms if they raised their prices.

On the other hand,* if economies of scale or scope don’t exist or are not large, there may be no consumer benefit from market concentration. There only exists downside risk of higher prices, particularly in the presence of barriers to entry. An unpublished paper presented to the Chicago Fed in March 2010 by Hilliard, Ghosh, and Santerre (pdf of slides available) describes what is known about these factors with respect to health insurance, providing some useful references in the literature.

Demsetz (1973) argues that while the large firm‐sizes involved in concentrated markets may make collusion more likely, they may also allow exploitation of economies of scale and scope. Such concentrated markets may be beneficial to consumers, if the cost savings are passed on in the form of lower prices. There is little evidence, though, that scale and scope economies are important for health insurance. Engberg, et al. (2004) is the most recent study to reject the presence of scale economies in health insurance. …

The AMA suggests that exclusive control over health care provider networks acts as a sufficiently high entry barrier to prevent competitiveness. … [N]ew entry can … [also] be prevented if high concentration permits existing firms to prevent [rental] access to their networks in an effort to counter new business. Furthermore, contestability theory suggests that high sunk costs may deter entry by acting as an exit barrier (Baumol, et al. (1982)). … [I]mportant start‐up costs that are not recoverable include marketing and the cost of setting up provider networks (when renting is too expensive), which may be sufficient to deter entry. Perhaps the most significant barrier to entry, relatively unique to health insurers, is very high consumer switching costs. Samuelson and Zeckhauser (1988) show that status quo bias generates high insurance switching costs in general. This bias is amplified when most customers obtain health insurance through employer‐sponsored group plans with limited provider networks.

Not mentioned by the authors are barriers to entry associated with state regulation. If compliance with state rules about network adequacy or other properties of product offerings is challenging (perhaps due to lobbying by incumbent firms), that is a deterrence to entry as well.

*Economists always have another hand handy. Annoying, isn’t it?

References

Baumol, W. J., J. C. Panzar, and R. D. Willig, 1982, Contestable markets and the theory of industry structure (San Diego: Harcourt Brace Jovanovich).

Demsetz, H., 1973, Industry structure, market rivalry, and public policy, Journal of Law and Economics 16, 1‐9.

Engberg, J., D. Wholey, R. Feldman, and J. B. Christianson, 2004, The effect of mergers on firms’ costs: Evidence from the HMO industry, The Quarterly Review of Economics and Finance 44, 574‐600.

Samuelson, W., and R. Zeckhauser, 1988, Status quo bias in decision making, Journal of Risk and Uncertainty 1, 7‐59.

It’s clear to me from Ezra Klein’s reference to my “Health care is different” post that I need to say more.

Many suggest that the solution to our health care system’s problems is to be found in a more market-based approach. Consumer-directed health plans are at the center of this concept. If you make people spend more of their own money, they’ll be more prudent users of care and seek better value at lower prices. That’s how other industries work, and we don’t complain much about them. Why should health care be any different? Get the government out of the way. Give vouchers to Medicare and Medicaid beneficiaries. Let people shop for the best deal on the unfettered market. And, moreover, reduce first-dollar, third-party payment by encouraging high-deductible plans.

About now you’re thinking I disagree with the notions I just expressed. Actually I don’t. They have merit, which I recognize, accept, and support. Where I take issue is that they are not solutions to all the problems in our system. They can address, at least in part, some of the problems, though even there more is required. It’s just not that simple. (Again I strongly encourage those who thinks otherwise to read Katherine Baicker and Amitabh Chandra, “Myths and misconceptions about U.S. health insurance.”)

That brings me to the post to which Klein referred. I illustrated how health care is like other industries,

Health care markets and the airline industry both have barriers to entry. The former requires special licenses, the latter requires considerable capital and regulatory compliance.

Trust in one’s doctor is important, as is trust in one’s lawyer.

Information assymetries exist in health care as in auto repair. Your mechanic (doctor) knows more about your car (your internal organs) than you do.

Health services and health insurance products are differentiated and complex, as are other products like cell phones or personal computers.

Health professionals want a good income, as do those in other fields.

That would suggest that if we just make health care even more like any of those industries, by reducing third-party payment, government regulation, and so forth, we could harness all the characteristics of the free market–characteristics that we don’t complain about (and even like) with respect to other products and services. True!

Except that health care is different, in one crucial way. I said it, and so did Klein:

First, if you don’t get good health care, you might die. That makes it hard for individuals to say no to things, it makes it hard for insurers to resist the backlash that comes when they say no to things, and it makes it hard for government to say no to things. [Bold mine.]

This may not seem important to the wealthy, young, and healthy. But to poorer, older, and sicker individuals (e.g. those on Medicaid and Medicare) it dominates. It is very hard to be a prudent purchaser of care when you’re in the ICU or responsible for the health of a vulnerable population. That alone suggests that an emphasis of moving the cost risk to individuals isn’t sufficient. There is plenty of cost risk to spread around. Some of it should be borne by physicians and hospitals, some by insurers (public and private), and some by individuals.

The optimal allocation of risk across these entities is not likely to be the same for the young and healthy as for the old and sick. That might be OK for buying a cell phone–overpaying for or under-utilizing cell phones won’t do you much harm–but it isn’t for buying health care. Why? Because health care is different. A mistake costs far more (your life) and you really do know far less about it than the salesman or practitioner, particularly when you’re very sick.

Some reduction in costs at minimal to no reduction in health may be possible with increased personal responsibility. The RAND health insurance experiment has demonstrated that, on average. However, that fact does not hold for all sub-populations. The health of individuals with certain chronic illnesses included in the RAND HIE was harmed by increased cost sharing. The RAND HIE did not include the elderly. More recent work has shown that increased cost sharing for Medicare beneficiaries can lead to adverse outcomes and doesn’t necessarily save taxpayers money.

I’m all for using square pegs in square holes. But when the hole is round, it just doesn’t fit. Forcing it is bound to break something, or someone. And it isn’t necessary. A health policy more nuanced than “free market for all” isn’t really that much harder. It just takes a little more thinking than can be conveyed on a bumper sticker.

Some sentences I agree with:

• Health care markets and the airline industry both have barriers to entry. The former requires special licenses, the latter requires considerable capital and regulatory compliance.
• Trust in one’s doctor is important, as is trust in one’s lawyer.
• Information assymetries exist in health care as in auto repair. Your mechanic (doctor) knows more about your car (your internal organs) than you do.
• Health services and health insurance products are differentiated and complex, as are other products like cell phones or personal computers.
• Health professionals want a good income, as do those in other fields.

Every single way one can characterize health care services, insurance, or markets, one can find another product, service, or market that is similar in that way. But, how many other products, services, or markets, possess every feature of health care that contributes to its failure as a market or the cost problems associated with it? Moreover, how many do so to the same degree? Lastly, how many also involve life and death decisions?

There are reasons health care costs are hard to control. Some of them can be addressed, at least in part, by changes that would make health care markets more like those of other industries. Some of them can’t. Except in very rare cases, make a mistake with your choice of airline, lawyer, auto mechanic, or cell phone and you’ll be inconvenienced for a time, until you switch products or practitioners. Much less rarely, make a mistake with your health care and you could pay for it for or with your life.

Health care is different. (See also “Like a breath of fresh air.”)

I’ve wondered what the role of non-profit hospitals will be once nearly all Americans have health insurance. The problem of uncompensated care will largely go away. Will the non-profits retain their tax-preferred status? One way they might not is via merger with a for-profit hospital. Jenny Gold of Kaiser Health News reports that some hospitals are considering doing just that.

[H]ospital mergers and acquisitions tend to go in boom-and-bust cycles. “My guess is that this is entering a period of expansion again,” says Gerard Anderson, director of the Johns Hopkins Center for Hospital Finance and Management.

One big reason is that the new health overhaul law will eventually extend coverage to an additional 32 million people, reducing the financial burden of the DMC and other hospitals that treat a lot of uninsured patients. “Health reform gets rid of a big chunk of the uncompensated care problem,” making urban hospitals more attractive acquisition targets, says Jack Wheeler, a professor of health management and policy at the University of Michigan.

Hospitals are merging while the political focus is on increasing competition in the insurance market. If market power swings (further) in favor of hospitals, there’s really very little reason to be optimistic about a market-based solution to reducing health care costs on the private side. The only element of the new health reform law that might help is the Cadillac tax. That doesn’t kick in until 2018 and won’t have a substantial effect for at least a decade after that. I don’t think the American public is going to be satisfied with the rate of premium growth over the next two decades.

I’ll just throw in the relevant figure again (below). Moving in the direction of “A” is probably more costly than in the direction of “C” because most health care premium dollars (~85%) collected by insurers flow to providers. (In math speak: the slope between “A” and “B” is larger in magnitude than between “B” and “C”.) Thus, all other things being equal, increasing market concentration among providers leads to a higher absolute markup in prices (premiums).

By the way, this figure isn’t just based on my own imagining. I’ve shown it to many health economists and all agree it has the right shape. Also, I’ve roughed out some theory that reproduces it algebraically. I’m confident in saying that the health care market really has the qualitative characteristics the figure suggests.

The time has come to dig a bit deeper into the current vision for the future of health care in America: accountable care organizations (ACOs). ACOs represent a new model (or a new-ish version of familiar ideas) of health provider organization and payment and are the current best hope for controlling health care costs, at least for public programs. Plans for Medicare and Medicaid ACO pilot projects are included in the new health reform law. If successful, they may be implemented widely. Someday we may all receive care from ACOs.

In brief, an ACO is a group of health care providers and institutions that are collectively responsible for (and held accountable to measures of) the health of a population. An ACO has an organizational structure that permits the encouragement of improvements in quality and lower costs through payment incentives. If that was a bunch of mumbo-jumbo to you, then read my earlier post on ACOs or Michelle Andrews’ NY Times’ Prescriptions post on the topic.

There are two key elements to an ACO, one pertains to its structure the other to how it is paid. Since an ACO is responsible for the health of a population it must include a set of health care providers such that individuals seeing those providers (mostly) don’t receive care outside the ACO. This is most easily understood by counter-example. If you see three providers–X, Y, and Z–but only providers X and Y are in the ACO then that ACO isn’t accountable for all the care you receive. Some of your care, that given by provider Z, is outside the ACO. That violates the ACO concept. You can imagine that ACOs must therefore be quite large.

The reason ACOs need to include all providers caring for a population relates to the second element of an ACO, how it is paid. In contrast to the fee for service model of payment, which encourages provision of care independent of its need or quality, ACOs would receive a bundled payment based only partially on services provided. Some portion of the payment would be based on the achievement of quality and cost performance goals. The incentives of the payment system would be designed to lower costs and improve quality.

The bundled incentive payments received by an ACO would be distributed to particular providers within that ACO according to its internal governance. Given this, you can imagine that whatever entity manages the ACO would wield considerable power. Currently hospitals are the most powerful of provider entities so one might expect they would be the source of ACO management. As I’ll argue below, that’s probably a bad idea.

Abstracting from the details, ACOs include two important economic concepts: (1) provider organization and (2) incentive payments. While the first is necessary for the second it also raises the possibility of increased market power on the part of providers. If providers consolidate as they organize into ACOs–and some are already doing just that–they may gain increased market power that could be used to negotiate higher prices from insurers, seek additional rents from legislators, and game the payment system.

In this sense, the organization necessary for payment reform may offset its intended effects. This brings us back to my favorite graph that relates health care premiums (a stand in for costs in general) to concentration in the health insurance market. (Other posts that use this graph are found under the “health care market theory” tag.)

Recall that this graph is drawn for a fixed level of provider market concentration. While provider payment reform (via ACOs) may lower public costs, additional provider market power will push the system further toward the left along this curve (toward the point “A”). Private costs and premiums could rise. (Note, this is not the standard cost shifting argument. Private costs don’t rise because public payments fall. They rise due to changes in the balance of market power between providers and insurers.)

That’s no good. But there is a way around this problem. The key is to recognize that a great deal of cost in the health care system is for hospital based services while relatively little is for primary care services. Suppose primary care doctors, not hospitals, are permitted to manage ACO-like entities. That is primary care doctors could be made to be the gateway to non-emergent specialty and hospital care (sounds familiar).

The advantage of such an arrangement is that additional market power of primary care doctors would not itself drive costs up much since the services they directly provide don’t cost very much. Yet if the primary care doctor organizations were held accountable for the care their patients receive they would be responsive to the incentives of the payment system. They would refer their patients to higher quality, lower cost specialists and hospitals. One could expect costs to come down under payment reform without substantial upward pressure due to increased primary care physician market power.

The key to all of this is wrestling control out of the hands of hospitals. I can think of a few political problems with this idea.

Robert Reich has an interesting post today on the Senate public option compromise. First he reminds readers that the compromise includes (1) an expansion of Medicare for certain individuals without employer coverage and 55-64 years of age and (2) for younger individuals a system of private plans modeled on that available for federal employees (details in this NY Times article). Then Reich writes that

we still end up with a system that’s based on private insurers that have no incentive whatsoever to control their costs or the costs of pharmaceutical companies and medical providers. If you think the federal employee benefit plan is an answer to this, think again. Its premiums increased nearly 9 percent this year. And if you think an expanded Medicare is the answer, you’re smoking medical marijuana. The Senate bill allows an independent commission to hold back Medicare costs only if Medicare spending is rising faster than total health spending. So if health spending is soaring because private insurers have no incentive to control it, we’re all out of luck. Medicare explodes as well.

I agree with all that. However, Reich then goes on to explain that private insurers can’t control costs because they’re too consolidated. This actually doesn’t make much sense. Insurers, while apparently great fun to beat up on, are not the big problem. Providers are.

Though it is true that insurers with higher market power can charge higher premiums, they also obtain lower prices from providers. The optimal balance of power that results in the lowest premiums is not one in which insurers are weakest. In fact, insurers need a certain level of market power just to offset that of providers.

As I’ve written before, the vast majority of money flowing through private insurers ends up in the hands of providers. If one wants to address the health care cost problem one should follow the money to hospitals and doctors–to the prices they charge and the volume of ineffective services they provide. It is by no means certain that increased competition among insurers leads to lower prices.

That’s not to say insurers can’t play an important role. They can. The solution to taming health care costs is to provide incentives that will encourage insurers to pay providers differently, to pay for efficiency and quality, not volume. It will take reasonably powerful insurers to impose such a system on providers. The first type of insurer likely to do so is the big public one, Medicare. That there is any hope of success in this is due to the power Medicare wields from its size, which will be even larger (albeit by only a little) under the new compromise (hence the provider backlash).

To the extent that private plans can follow Medicare’s lead will be determined in large part by their market power. Blindly weakening insurers may throw the baby out with the bathwater. Of course, we haven’t even begun to fill the tub.

Later: Ezra Klein expresses similar ideas in a must read post today.

Sick of reading posts about market theory and health care costs? If so, skip this one and read this alternative post from the archives. It has nothing to do with health care costs or market theory.

Meanwhile, I’m obsessed with what market theory can tell us about health care costs. It’s a good thing because it’s my job (or part of it). Anyway, let’s revisit my favorite graph, which you might recall from prior posts on this topic. It illustrates how I think insurer market concentration relates to premiums (or health care costs in general), holding health care provider concentration fixed (the curve shifts up (down) for higher (lower) provider concentration).

While this graph explains a lot, it doesn’t explain everything. In particular it abstracts away the role played by what the insurers actually provide for the premium charged. The curve is for the average standardized premium, which means all the variations in benefit packages offered by insurers are taken into account or controlled for, adjusting the premium accordingly.

Of course benefits matter for enrollment, which is the source of insurer market power. The greater number of enrollees, the greater the power of the insurer. Most benefits are of the form “such-and-such is covered for a copayment of thus-and-so.” But one benefit is of a different flavor than all the rest. That benefit is the provider network.

The provider network is the set of providers to which policyholders have access through the insurer. Go outside that network and you’re not covered; you have to pay for the service entirely out of pocket. Of course the establishment of the provider network of an insurer is relevant to the market power of providers and the insurer. The negotiation over rates–payments from the insurer to providers for their services–includes the possibility of exclusion from the network. For a sufficient rate of pay providers will want to be in the network to have access to the volume of patients the insurer can deliver. At the same time, the insurer is only willing to pay so much to a given provider: more for a dominant provider, less for a small player.

What distinguishes the provider network from other benefits is that it exerts a positive externality on policyholders. This type of inter-group network externality is the defining feature of a two-sided market. The greater the number of providers in an insurer’s network, the more a consumer will pay that insurer for access. All other things equal, consumers like access to more providers. They’ll pay a higher premium for such access.

This (two-sided) market-theoretic aspect of health care costs is not reflected in the figure above, nor is it reflected in most theoretical motivation or empirical specification of analysis of health care markets in the literature. This is not a surprise. The foundational work on this is brand new.

Leemore Dafny, with colleagues Mark Duggan and Subramaniam Ramanarayanan, has produced yet another fascinating paper on consolidation in the health insurance industry (I mentioned another of her papers in this area in a prior post). Their latest, “Paying a Premium on Your Premium? Consolidation in the U.S. Health Insurance Industry” (NBER, October 2009) examines the effect of health insurer consolidation on premiums and health care workers’ employment and earnings.

Three of the authors’ main results are that due increases in insurer concentration: (1) Between 1998 and 2006 premiums increased 2.1 percent. (2) Between 1999 and 2002 physician earnings declined by 2 percent. And (3) over the same period health worker employment declined 2.4 percent. Let’s put these results in context.

Recall the figure I introduced in an earlier post and reproduced below. It illustrates my hypothesis about the effect of insurer concentration on premiums, for a fixed level of provider market power. In the region in which insurers have a high degree of concentration (around point “C”), I would expect additional concentration to lead to higher premiums. This is exactly what Dafny and colleagues found.

Moreover, the authors find that the additional market power held by insurers not only led to premium increases, it also led to a reduction of input prices (physician salaries) and employment (a signal of reduced demand). The first of these phenomena is consistent with monopolistic behavior while the latter two are consistent with that of a monopsonist.

In brief, the findings suggest excessive insurer market power relative to providers. But how excessive is it? That is, is the market operating closer to point “C” or to point “B”? We don’t know, but a related piece of evidence is that over the time period of the study premiums increased 100%. So the measured effect of insurer concentration on premiums–a 2.1% increase–is quite small relative to the overall increase in premiums and the degree of increase in concentration in the insurance market (the fraction of the market deemed “highly concentrated” rose from 68% to 99% in the study period).

Whether the market was operating closer to point “B” or “C,” the slope of the curve in the region it was operating is modest. Despite the large increase in concentration, the vast majority, about 98%, of the observed increase in premiums was due to other factors, including additional provider concentration, medical technology, increased consumer demand, and so on. This suggests that additional insurer concentration is not an important factor relative to all others. Note that this is a result on the change (the additional) market concentration and says nothing about the level.

Can we therefore conclude that insurer concentration is not a substantial factor in premium inflation? That would be going a bit too far. After all, this study, like all studies, has limitations. It is beyond the scope of this post to go into them but there are several reasons why the results may not be generalizable to the entire health insurance market. Some of the reasons are provided in the paper itself, and a few others were raised when this paper was presented at the 2009 Annual Health Economics Conference hosted by the BU School of Public Health (at the time of this writing some conference papers are still online at the bottom of the web page).

So I am very enthusiastic about this paper but interpret the results with caution. This measured reaction is totally normal. All studies, even good ones such as this, have limitations.  That’s why one should base conclusions on a body of work that looks at the same issue from multiple perspectives, using different data and methodology. Nevertheless, this is a good paper on an important problem using appropriate methodology. I encourage you to read the entire manuscript.

Health reform, even with a public option, will be built largely on a private insurer infrastructure. That means hundreds of billions of dollars will be pumped through insurance companies on their way to payment to providers for health care services. Given this reality, we should all want as efficient a health insurance market as possible.

This appears to be the beginning of an argument for encouraging additional competition among health insurers. Greater competition is the source of greater market efficiency, isn’t it? The answer is, it depends on the market. The simple type of idealized market taught in Econ 101 does become more efficient (in an economic welfare sense) as competition increases. But other types of markets–in particular, so called two-sided markets, among others–do not always behave this way. And health care is one of them.

Let’s first consider the best-case scenario for the consequences of greater competition in the health insurance market. About 85% of insurers’ costs are medical. The remaining 15% is for administration, marketing, management, and profit (profit itself is about 6% at the time of this writing). With additional competitive pressure insurers would compete away some of that 15%. Perfect competition would reduce profit to zero and may force efficiencies elsewhere. So, perhaps the best competition could do would be to cut in half that 15% for administration, marketing, management, and profit.

Figure 1

Now a 7.5% reduction is not chump change. It represents a lot of money, tens of billions of dollars annually. But that’s only an optimistic best case. And it pales in comparison to the cost savings that are available on the provider side, the 85% of insurers’ costs. (Hat tip: Michael Chernew.)

If we really wish to understand what would happen if the insurance market is made more competitive we have to go one step further. There is a reason the health care market is not analyzed in Econ 101. It is much more complex than, say, the market for coffee. In fact, additional competition in the health insurance market could drive costs up. Here’s a a graphical depiction of what I have in mind.

Figure 2

The figure is drawn for a fixed level of provider market concentration. The horizontal axis is insurer market concentration, and the vertical axis is premium for a standardized plan. Relative to the fixed level of provider concentration, insurer concentration is high in the region of point “C,” and premiums are above the minimum possible level. Insurers are charging above the competitive premium level because they have excessive market power. In this region, higher premiums stem from higher insurer profits and/or lack of administrative efficiency (the red zone of Figure 1).

On the other hand, insurer concentration is low relative to that of providers in the region of point “A,” and premiums are again above the minimum because insurers can’t negotiate down to the lowest possible price with providers. Providers have too much power relative to insurers and are charging prices above the competitive minimum. Insurers pass those high prices onto consumers through higher premiums. In this case, higher premiums stem from higher medical costs (the green zone of Figure 1).

As provider power weakens the whole curve shifts downward, and as provider concentration increases the whole curve shifts upward.

If we’re now at point “C” (and this is by no means certain) it may be possible to achieve lower premiums by diluting the insurance market, weakening insurers with respect to providers. But if we go too far, we’ll shoot past the optimum “B” and be no better off. On the other hand, if we’re now at point “B” we should do nothing to upset the insurer-provider balance of power. And if we’re at point “A” then it is the provider market, not the insurance market, that has excess concentration.

As far as I know, nobody knows which of the many local health care markets are operating at “A”, “B”, or “C.” In fact, I have not yet seen any studies that can tell us what mix of insurer-provider market structure achieves the optimal point “B.”

It is worth noting that what is being considered in Washington as the long-term solution to health care costs has less to do with finding an optimal mix of insurer-provider power and more to do a change in the incentives of provider payment. The goal is to shift the entire curve in the figure above downward so that in every type of market–”A,” “B,” and “C”–health care costs and premiums come down. But a downward shift due to payment reform may be offset by a movement leftward along the curve. Such an offsetting effect could be brought about by antitrust action against insurers, weakening their market power with respect to providers. A similar offset could occur through additional integration of providers under a payment reform model (ACOs, CHTs). Both of these potential offsets are included in current legislation.

There is more to say about health care costs from a market theory point of view. I will be returning to this issue and referring to Figure 2 in the near future. The bottom line, however, is that the failure of policymakers to consider the insurer-provider balance of power may prove to be a costly oversight.