Redistributional consequences of community rating

Health Services Research, April, 1997 by Dana P. Goldman, Arlene Leibowitz, Joan L. Buchanan, Joan Keesey

Health insurers often charge higher premiums to individuals or small groups whom they expect will have higher expenditures (Health Insurance Association of America 1994). This practice, known as medical underwriting, leads to higher premiums for such individuals as the chronically ill and employees engaged in hazardous occupations (Freudenheim 1990). In extreme cases, some insurers may engage in "blacklisting," whereby businesses are completely excluded from the small group market. Formisano, Schwartz, Neale, et al. (1990) find that 29 out of 48 insurers in Wisconsin rejected some types of businesses out-of-hand. Similarly, the Governor's Office in Massachusetts reported in 1990 on the blacklisting of beauty and barber shops, florists, and taxi services (Massachusetts Governor's Office 1990). As medical science advances techniques for genetically screening medical risks, opportunities to engage in medical underwriting will increase (Huber 1992).

Perceived inequities with the status quo in medical underwriting, coupled with a concern over future developments, have spurred proposals to strengthen regulations governing the sale of insurance. A number of states - including Vermont, Maryland, Minnesota, Maine, Oregon, and New York - have already limited medical underwriting by mandating community rating of insurance policies (Kolata 1993; Rowland 1993). Similar legislation has been proposed at the federal level. President Clinton's proposed Health Security Act, for example, would have introduced community rating within regional "alliances" throughout the country.(1)

It is unclear how large each insurance market needs to be for community rating to work. However, the proposed reforms would pool risks on a much larger geographic scale than we currently observe in the non-group market, where insurers often vary premiums by zip code. Pooling on a larger geographic scale has some unintended and surprising economic consequences. To better understand these potential effects, this article uses U.S. Census data for California to simulate the effects of community rating in the state's health insurance market. First, we begin with a discussion of medical underwriting and community rating. We then present our simulation methods and discuss our results. We conclude by suggesting some ways to redress these transfers to the extent that they are unwanted.

THE PRACTICE OF MEDICAL UNDERWRITING

While medical underwriting may appear undesirable, its manifestation in insurance markets reflects competitive pressure to keep premiums low. The fundamental principle of insurance is to pool risks across a large group of beneficiaries, charging each of them the same premium to at least cover average expenses. However, with insurers competing to sell policies, companies are often driven to offer lower premiums to individuals or firms with lower than average expected healthcare use. If one insurer does not offer a competitive package, others may. Instead of letting these low-cost policies be bid away, the insurer willingly allows them to be separated from the general risk pool with a more attractive premium. After separating out these 'good' risks, the insurer is left with a more costly group that faces higher average premiums. Thus, competitive pressures force insurers to separate beneficiaries according to their systematic risk; the corporate survival of an insurer may depend on its ability to successfully identify beneficiaries with lower or higher expected costs.

Insurers have several ways of varying premiums according to systematic risk. Sometimes higher premiums are based on past claims experience, a practice known as "experience rating." Experience rating leads insurers to adjust premiums annually for firms and other group policies based on how much medical care the group has used in the previous year. However, some insurers base higher premiums not on actual claims experience, but rather on identifying an individual or small group with a larger beneficiary pool that has "typically" higher expenditures. For instance, because medical expenses vary systematically by age and gender, when insurers sell individual policies, they charge companies different premiums, based on the number of men and women and older and younger workers. In the small group insurance market, insurers often set the premium as a function of the age and sex distribution of the firm's workers. Other nonmedical factors, such as working in an occupation known to have higher medical expenses, may result in higher premiums (U.S. Library of Congress 1988).

Insurers may also associate individuals (outside the group market) with larger beneficiary pools based on the presence of certain medical conditions. It is for this reason that some insurers require new beneficiaries to reveal health conditions that are known to increase the use of medical care.(2) Under extreme conditions, the chronically ill can be singled out for individual premium increases or be denied access to insurance at any price (Freudenheim 1990).