Choice and Price Competition in Health Insurance Markets

Featured in print Bulletin on Aging & Health

Choice in health insurance markets is a double-edged sword. On the one hand, economic theory suggests that when there are more competing plans, prices will be lower and the market will be more efficient. Further, if insurers are able to offer different benefit packages, having more plans to choose from may allow consumers to select one that better matches their preferences. On the other hand, as the number of choices increases, consumers may become overwhelmed and have more difficulty finding the lowest-priced plan or the one that best suits their needs. Moreover, the existence of more plans may exacerbate problems that result from buyers having private information about their health status - for example, healthier buyers may select a high-deductible plan to avoid being pooled with expensive sickly patients, even though they would prefer to have more risk protection.

The question of how choice affects health insurance markets is very relevant for public policy. For example, under the recently enacted Medicare Part D program, most seniors can choose from among dozens of prescription drug plans, and it has not yet been established whether this is a positive or negative aspect of the program.

In "Choice, Price Competition and Complexity in Markets for Health Insurance," (NBER Working Paper 13817) researchers Richard Frank and Karine Lamiraud draw insights about the value of choice in health insurance markets from the case of Switzerland. The Swiss case offers a particularly good opportunity to study the effect of choice on price competition, since the other changes that can occur as choices increase - namely, an increase in the type of plans available and in the problems resulting from information asymmetries - are less relevant due to the nature of the health insurance market in Switzerland.

Health insurance in Switzerland is organized at the canton (region) level. Insurers offer a standardized, comprehensive benefits package that includes inpatient, outpatient, and long-term care. Each insurer sets its own premiums but must charge the same amount to all individuals of the same age group living in the same canton and cannot turn away prospective enrollees. All individuals must purchase health insurance and have 30 or more plans to choose from during the sample period examined by the authors. Individuals can switch plans easily during the annual open enrollment period; price information is readily available and switching will generally not affect an individual's access to particular doctors.

Given the nature of the Swiss health insurance market, insurers would be expected to compete actively on price. In fact, however, the market is characterized by large and persistent price differences across plans. In 2004, switching from an average-priced plan to the lowest price plan would result in a premium savings of nearly 20 percent. In one canton, the difference between the highest and lowest priced plans was nearly 80 percent of the average premium. Switching rates are also very low, about 3 percent per year.

What can explain these large premium differentials and low switching rates? Standard economic theories would include search and switching costs for buyers or price discrimination by sellers, but these explanations seem less powerful in the Swiss context. A possibly more plausible explanation draws on research in behavioral economics. That research has shown that cognitive overload and fear of making incorrect decisions often arise when decisions are complex and have high stakes. These factors can lead to poor decision making or attempts to avoid making a decision by opting for the status quo. Past research has also shown that individuals tend to exaggerate the value of items they own relative to what they would value the same item if they did not own it. The authors hypothesize that consumers may tend to underestimate the gains and exaggerate the losses from changing health insurance plans.

The authors conduct an empirical analysis of plan switching to assess the role of prices, information overload, and status quo bias in the health insurance decision-making of Swiss consumers. They use both a household survey with information on plan switching during 1997-2000 and an insurer database with information on health plans.

The authors find that switching rates are consistently lower when individuals have more plans to choose from. Interestingly, an increase in smaller "fringe" firms inhibits switching more than does an increase in larger firms. Having more choices makes consumers less likely to switch even when they expressed dissatisfaction with their current plan. These findings are all consistent with a decision overload theory. The authors also find that switching rates increase as the difference between the price of one's own plan and the other options increases, suggesting that switching behavior is responsive to price as standard economic models would suggest.

The authors show that switchers lower their premiums by an average of 16 percent relative to those who stayed in the same plan. This suggests that many consumers "leave money on the table" by not switching plans.

What is driving households' decisions, if not price? Quality is another possibility, although the regulation in the Swiss system serves to minimize quality differences across plans. One way in which firms do differ is in their reliability, as some insurers have left the market in recent years. Yet the authors find that the size of the plan's financial reserves does not have a significant effect on switching. Another major factor seems to be the desire to go with a known quantity, as 40 percent of households say they base their decision on parents' and friends' choices or on tradition and non-switchers are much more likely to be in one of the large national plans. These results may suggest a status quo bias.

The authors conclude "one implication of these results is that expanding choice to very large numbers is likely to reduce the effectiveness of consumer decision-making which may in turn result in larger markups by health insurers. At a moment in history when elderly Americans are facing large numbers of choices in private health plans and prescription drug plans our findings may offer some cautions regarding the need for decision support and mechanisms that simplify such health insurance choices."

The authors gratefully acknowledge funding from the Agency for Healthcare Research and Quality, the Alfred P Sloan Foundation, the Swiss National Science Foundation, and the Franco-American Fulbright program.