In Table 3, we report elasticities for several subsamples, basedon the fixed-effects specification in col. (5) of Table 2. The first rowshows an elasticity for the entire sample around −1.07, consistentwith the arc elasticities in Fig. 1. There is considerable hetero-geneity by age. Those under 30 have elasticities over 40 percentlarger in magnitude than those over 40. Sicker individuals—as mea-sured by self-reported health problems and any past insurancedenials—are less price-elastic. The health-status and age results areconsistent with Strombom et al. (2002). Those in the lowest income quartile also have elasticities over 40 percent larger than those inthe highest income quartile, similar to patterns found by Marquisand Long (1995). While the pattern of elasticities generally conformto expectations, their range appears more limited than in past work,a point we discuss in Section 4.4