How does the tax exclusion for employer-sponsored health insurance work?
The exclusion lowers the after-tax cost of health insurance for most Americans.
Employer-paid premiums for health insurance are exempt from federal income and payroll taxes. Additionally, the portion of premiums paid by employees is typically excluded from taxable income as well. The exclusion of premiums lowers most workers’ tax bills and thus reduces their after-tax cost of coverage. This tax subsidy partly explains why most American families have health insurance coverage through employers. Other factors play a role though, notably the economies of group coverage.
ESI Exclusion is worth more to taxpayers in higher tax brackets
Because the exclusion of premiums for employer-sponsored insurance (ESI) reduces taxable income, it is worth more to taxpayers in higher tax brackets than to those in lower brackets. Consider a worker in the 15 percent income-tax bracket who also faces a payroll tax of 15.3 percent (7.65 percent paid by the employer and 7.65 percent paid by the employee). If his employer-paid insurance premium is $1,000, his taxes are $281 less than they would be if the $1,000 was paid as taxable compensation. His after-tax cost of health insurance is thus $1,000 minus $281, or $719. In contrast, the after-tax cost of a $1,000 premium for a worker in the 25 percent income tax bracket is just $626 ($1,000 minus $374). Savings on state and local taxes typically lower the after-tax cost of health insurance even more.
ESI Exclusion is costly
The ESI exclusion cost the federal government an estimated $250 billion in income and payroll taxes in 2015, making it the single largest tax expenditure. Note, too, that the open-ended nature of the tax subsidy has likely increased health care costs by encouraging the purchase of more comprehensive health insurance policies with lower cost sharing or with less tightly managed care.
Replacing the ESI exclusion with a tax credit would equalize tax benefits across taxpayers in different tax brackets, as well as between those who get their insurance through their employers and those who obtain coverage from other sources. Making the credit refundable would extend that benefit to those whose tax liability falls below the value of the credit. And designing the credit so that it does not subsidize insurance on the margin (i.e., to be a fixed dollar amount as opposed to a percentage of the premium) could lower health care costs.
 These examples assume that workers bear the full burden of employer payroll taxes. Note that the effective marginal tax rates (28.1 percent for the worker in the 15 percent income-tax bracket and 37.4 percent for the worker in 25 percent income-tax bracket) are less than the sum of the income tax and payroll tax rates (30.3 percent and 40.3 percent, respectively) because those rates are applied to compensation after the employer’s share of payroll taxes has been deducted. Thus, for example, if the employer increases compensation by $1,000, cash wages only increase by $929 [calculated as $1,000 / (1+employer payroll tax rate)], because the employer would have to pay additional employer payroll taxes of $71. The lower-wage worker’s resulting combined income and payroll tax would be 30.3 percent of $929 or $281. The higher-wage worker’s resulting combined income and payroll tax would be 40.3 percent of $929 or $374.
Joint Committee on Taxation. Estimates of Federal Tax Expenditures for Fiscal Years 2014–18.
Urban-Brookings Tax Policy Center. “Microsimulation Model, version 0515-1.”
Burman, Leonard E., and Jonathan Gruber. 2005. “Tax Credits for Health Insurance.” Tax Policy Center Issues and Options Brief 11. Washington, DC: Urban-Brookings Tax Policy Center.
Burman, Leonard E., Jason Furman, Greg Leiserson, and Roberton Williams. 2007. “The President’s Proposed Standard Deduction for Health Insurance: An Evaluation.” Washington, DC: Urban-Brookings Tax Policy Center.
Gruber, Jonathan. 2011. “The Tax Exclusion for Employer-Sponsored Health Insurance.” National Tax Journal 64(2, part 2): 511–30.