In last year’s presidential race, Barack Obama scored points when he attacked John McCain for proposing to tax people’s employer-provided health insurance as income. Conservatives argued that the business tax deduction for health insurance distorted healthcare markets, encouraging people to overconsume healthcare.
That assumes people are doing the equivalent of choosing extra colonoscopies over a vacation at the beach. It was bad policy and bad politics.
Then, in their destructive confusion over health reform, some Democrats – especially in the Senate – began reviving McCain’s idea, and even the administration signaled openness to it. Democrats were seeking money to pay for their proposal, and some argued that current tax subsidies favored “gold-plated” policies for high-income executives, thus making a tax on benefits progressive. Strong criticism from liberals and unions has for now beat back the plan as bad politics.
Now Harvard physicians David Himmelstein and Steffie Woolhandler, co-founders of Physicians for a National Health plan, have shown that the supposedly progressive rationale for taxing employer-provided health insurance is as flawed as the conservative justification.
In an article online in The New England Journal of Medicine, they show that a tax on benefits would be regressive, taking a far higher percentage of income from low- and moderate-income taxpayers than from high-income individuals.
In absolute dollar terms, the affluent gain more, but as a percentage of income, lower-income earners receive a greater benefit and thus their tax rate – if benefits were taxed – would be higher. That’s the usual definition of tax regressivity.
For example, Himmelstein and Woolhandler calculate that the mean (or average) benefit for families earning $10,000 to $19,999 and receiving employer-paid coverage is $2,115, or 13.9 percent of their income. But families earning $100,000 or more receive a mean tax benefit of $4,498, or just 2.7 percent of their income.
Of course, lower-income families are less likely to have employer-provided insurance (13.8 percent of those earning $10,000 to $19,999 compared to 61.8 percent of those earning $100,000 or more).
So when Himmelstein and Woolhandler mapped the effects of a benefit tax on the whole population, not just those whose employers provided insurance, the mean tax rate for low-income families was low, rising to a peak for middle-income families, then falling back down for the more affluent. It was still basically regressive.
They calculated the effects assuming families paid the tax, but Himmelstein says most analysts assume that if employers were taxed instead, they would simply deduct the cost from wages. If insurers were taxed, their likely strategy is less clear, but they would probably raise rates for everyone.
If only high-priced policies were taxed, people living in states with higher medical costs would be penalized, and without indexing of the cut-off for inflation, more families would be affected each year. Also, Himmelstein says, federal law generally requires employers to offer the same benefit package to all employees. So lower-paid workers at an employer (like Harvard) with comprehensive health benefits would be hit, not just high earners.
Although the plan to tax benefits now seems off the table, Himmelstein expects it to spring up again. “We’re hoping,” he says, “to drive a small stake through its heart.”
SPECIAL DEAL: Subscribe to our award-winning print magazine, a publication Bernie Sanders calls "unapologetically on the side of social and economic justice," for just $1 an issue! That means you'll get 10 issues a year for $9.95.
David Moberg, a former senior editor of In These Times, was on staff with the magazine from when it began publishing in 1976 until his passing in July 2022. Before joining In These Times, he completed his work for a Ph.D. in anthropology at the University of Chicago and worked for Newsweek. He received fellowships from the John D. and Catherine T. MacArthur Foundation and the Nation Institute for research on the new global economy.
