Health Insurance Exchanges May Be Too Small to Succeed

November 23, 2012, 6:00 am - New York Times

Health Insurance Exchanges May Be Too Small to Succeed

By DANA P. GOLDMAN, MICHAEL CHERNEW and ANUPAM JENA

Dana P. Goldman is the director of the Leonard D. Schaeffer Center for Health Policy and Economics at the University of Southern California. Michael Chernew and Anupam Jena are professors of health care policy at Harvard University.

With the re-election of President Obama, the Affordable Care Act is back on track for being carried out in 2014. Central to its success will be the creation of health-insurance exchanges in each state. Beneficiaries will be able to go a Web site and shop for health insurance, with the government subsidizing the premiums of those whose qualify. By encouraging competition among insurers in an open marketplace, the health care law aims to wring some savings out of the insurance industry to keep premiums affordable.

Certainly, it is hard to be against competition. Economic theory is clear about its indispensable benefits. But not all health care markets are composed of rational, well-informed buyers and sellers engaged in commerce. Some have a limited number of service providers; in others, patients are not well informed about the services they are buying; and in still others, the quality of the service offerings vary from provider to provider. So the question is: What effect does insurer competition have in a marketplace with so many imperfections?

The evidence is mixed, but some of it points to a counterintuitive result: more competition among insurers may lead to higher reimbursements and health care spending, particularly when the provider market - physicians, hospitals, pharmaceuticals and medical device suppliers - is not very competitive.

In imperfect health care markets, competition can be counterproductive. The larger an insurer's share of the market, the more aggressively it can negotiate prices with providers, hospitals and drug manufacturers. Smaller hospitals and provider groups, known as "price takers" by economists, either accept the big insurer's reimbursement rates or forgo the opportunity to offer competing services. The monopsony power of a single or a few large insurers can thus lead to lower prices. For example, Glenn Melnick and Vivian Wu have shown that hospital prices in markets with the most powerful insurers are 12 percent lower than in more competitive insurance markets.

So health insurance exchanges are probably welcome news for hospitals, physicians, and pharmaceutical and medical device companies throughout the United States. If health insurance exchanges divide up the market among many insurers, thereby diluting their power, reimbursement rates may actually increase, which could lead to higher premiums for consumers.

Ultimately, economic theory predicts that the effect of insurance exchanges on insurance premiums will depend on two offsetting factors. On one hand, smaller, less-consolidated insurance companies may have less bargaining power with large hospitals, physician groups and pharmaceutical companies, which traditionally command substantial market power. Reimbursements to these parties, as well as costs to insurers, may rise in a fractionated market, and if so, these costs would be passed on to consumers as higher premiums. On the other hand, exchanges may inject competition into the marketplace, reducing premiums as even the smallest insurer can market its plans, forcing larger insurers to lower their premiums to remain competitive. Which theoretical effect will dominate in reality is an open empirical question with important policy implications.

There is some evidence on how insurer market power affects premiums. Leemore Dafny, Mark Duggan, and Subramaniam Ramanarayanan have found that greater concentration resulting from an insurance merger is associated with a modest increase in premiums - suggesting that concentration may not help consumers so much - although they did report a reduction in physician earnings on average. Over all, however, the evidence is limited and mixed.

A simple analysis of the nationwide growth in premiums over the last decade is illustrative. Using 2001-10 data from the National Association of Insurance Commissioners, we examined the relationship between insurer market power (defined as the market share of the two largest companies) and changes in premiums. We found that concentration of insurer power -- hence less competition - was not significantly associated with higher premiums, as can be seen in the chart below.

Hawaii is a good example. Kaiser Permanente and Blue Cross Blue Shield together controlled more than 90 percent of the insurance market in 2001. In this highly concentrated market, the average premium rose only 72 percent over the decade, compared to an overall increase of 135 percent nationwide. By contrast, Virginia had one of the most competitive markets in 2001, with its two largest insurers controlling only 25 percent of the market, yet premiums in the state increased nearly 140 percent over the period.

Greater competition in the insurance industry -- either through health insurance exchanges or other measures -- may not lower insurance premiums. Weakening insurers' bargaining power could instead translate into higher costs for all of us in the form of higher premiums.

In financial markets, we ask if banks are too big to fail. When it comes to health care, perhaps we should ask if insurers are too small to succeed.