Do Tax Subsidies Increase Health Insurance among the Uninsured?

Summary of working paper 9567
Featured in print Digest

Their central estimates suggest that for each 10 percent that health insurance premiums were tax subsidized, take-up went up by only 0.2 percent, a very small reaction.

In recent years, politicians have expressed concern with the growth over the past two decades in the number of people not covered by health insurance. About 41 million non-elderly Americans are without health insurance coverage, one 2002 study notes, despite efforts in the past 15 years to expand the public insurance safety net, especially for children.

Roughly one-quarter of the uninsured are offered health insurance through their job, or the job of a family member, but do not take it up. These individuals are typically considered the "low hanging fruit" of the uninsured population, as the potentially cheapest group to bring into the ranks of the insured. Moreover, increases in the number of employees who decline to take up insurance offered by their employers largely explain the drop in insurance coverage over the last two decades.

However, new work by NBER Research Associate Jonathan Gruber and Ebonya Washington indicates that government subsidies to encourage workers to take up employer-provided insurance do little to improve coverage. Moreover, the authors note in Subsidies to Employee Health Insurance Premiums and the Health Insurance Market (NBER Working Paper No. 9567), the cost to the government in lost revenues from such a tax subsidy could be huge.

To reach this conclusion, the authors examine what happened to employee take-up decisions when tax subsidies were given to federal workers in the Federal Employee Health Benefits Program (FEHBP). Roughly one-half of all employees in the United States pay their insurance premiums with pre-tax dollars. But until 1994, virtually all of the federal employees insured through FEHBP paid their insurance premiums on a post-tax basis. Then, in 1994, employees of the postal service, representing roughly one-third of all federal workers, were given the right to pay their insurance premiums on a pre-tax basis. The remaining federal employees were given this right in October 2000. This saved them tax dollars, significantly reducing the cost of taking up insurance coverage; for a typical worker in Washington, DC, the cost of taking up health insurance fell by about $1000.

These changes provide Gruber and Washington with an excellent laboratory for learning about the impact of a sizeable reduction in the after-tax price of FEHBP insurance on the take-up of that insurance by federal employees. But they find that this sizeable reduction had essentially no impact on the rate of take-up of insurance by federal employees. Their central estimates suggest that for each 10 percent that health insurance premiums were tax subsidized, take-up went up by only 0.2 percent, a very small reaction. These results confirm other papers' findings that employees are not very sensitive to health insurance contribution rates in their decision to take up health insurance coverage.

Moreover, Gruber and Washington find that subsidizing insurance coverage caused federal employees to choose more expensive health insurance plans, when they did choose coverage. This further raised the cost to the government of this intervention.

The authors conclude that "subsidizing employee premiums is unlikely to be a cost-effective avenue for increasing insurance coverage." They estimate that these new tax subsidies cost the government roughly $700 million in revenues. And they have prompted only 11,000 to 20,000 new persons - a tiny percentage of the total number of federal employees - to take up the insurance coverage. So the revenue cost for each newly insured person was $31,000 to $83,000.

-- David R. Francis