X hits on this document





23 / 98

Similar to a method that Royalty (2000) later used, Leibowitz and Chernew (1992) observed

variation in tax rates across states to examine the impact of after-tax prices on insurance offer by

small firms (with fewer than 50 employees). Based on responses to an employer survey in 1989,

they estimated a tax-price elasticity of –2.9. However, based on variation in the premium quotes

obtained from small group insurers and statistically assigned to firms in their data, they estimated

a much smaller elasticity of employer offer with respect to differences in premium, –0.8.4

Observing firms with fewer than 100 employees, Marquis and Long (2001) estimated a still

lower premium elasticity, –0.14. Hadley and Reschovsky (2002) also found a relatively small

premium elasticity of –0.54, based on a nationally representative survey of employers in 1997

and imputed premiums. In addition, they found the smallest employers (with fewer than 10

employees) were the most price-sensitive—with an estimated elasticity of –0.63. Similarly,

Gruber and Lettau (2004) found that primarily smaller firms drove their estimates of tax price

elasticity across all firms. They estimated an offer elasticity of –0.54 and a spending elasticity of

  • 1.34 for firms with fewer than 100 employees, roughly twice the elasticity of offer among all


We found only one recent study that used a randomized controlled trial to estimate small

employers’ responsiveness to price change. Kronick et al. (2004) observed selected small

employers (with 2-50 employees) in San Diego that did not provide insurance and were offered

an experimental opportunity to buy coverage for a subsidized premium. The estimated price

elasticity of employer offer was –0.14—consistent with Marquis and Long (2001) and on the low

end of estimates from other studies. Kronick et al. (2004) concluded that small employers that

4 Also using data for small firms (fewer than 50 employees), Feldman et al. (1997) estimated extremely high premium elasticities of employer offer: –3.91 for single coverage and –5.82 for family coverage based on a 1993 survey of 2,000 employers in Minnesota. However, Hadley and Reschovsky (2002) pointed out the instrumental variable used in Feldman et al. (1997) suffered from endogeneity and resulted in upwardly biased elasticity estimates.


Document info
Document views581
Page views582
Page last viewedMon Jan 23 21:58:07 UTC 2017