estimated by Marquis et al. (2004).8 Among workers offered group insurance, a decrease in
individual premiums has very small effects on the decision to purchase individual coverage
instead of group coverage (Marquis et al. 2004).
With respect to differences among population subgroups, Marquis et al. (2004) estimated
that younger (under age 35) self-employed workers with incomes below 200 percent of the
federal poverty level (FPL) were much more price sensitive (with an elasticity estimate of –0.7 to
1.2) than older employed workers with incomes above 400 percent FPL (with an elasticity
estimate of –0.03 to –0.07). In addition, individuals with health problems appear to be less
responsive to premium changes (CBO 2005), as are married adults compared to single adults
(Gruber and Poterba 1994).9
Estimating price elasticity in the non-group market has proven to be even more difficult than
estimating the group-market elasticities. In nearly all states, premiums in the non-group market
overcome this problem, researchers have attempted to construct exogenous premium schedules
based on insurers’ offers and individual characteristics (e.g., Marquis et al. 2004), or they have
8 In contrast to studies that examined the effect of price change on the probability of being covered, Reschovsky and Hadley (2004) compared out-of-pocket spending for health care among lower-income uninsured people with the amount they would have spent on insurance and health care combined, if they had taken any of three hypothetical tax credits. Based on the CTS Household Survey, they estimated premiums for the target population, and found that nearly all eligible uninsured people would pay higher health-related costs if they took advantage of tax credits of the size considered in recent proposals. This finding offers some insight into a low tax credit might stimulate very modest take up among those low-income who are now uninsured.
9 Gruber and Poterba (1994) estimated price elasticities among single persons that were much higher (-1.8 semi-elasticity) than those among married persons (essentially zero semi-elasticity).
10 Only in New York and Vermont, which require pure community rating in the nongroup market, would premiums be unrelated to the policyholder’s own characteristics.