means tests are taxes – ACA edition

US workers are finally getting near-universal healthcare, but it comes at a price. Chicago economist Casey Mulligan runs the numbers.

Healthcare is valuable but expensive. As a result, many people believe that poor and middle-income households should pay less than full price for their healthcare, and the United States now has its Affordable Care Act (hereafter, ACA) that will soon implement such a policy.

Economics tells us that redistribution typically comes at the cost of reduced incentives to work and earn, yet some economic analyses of the ACA’s labor market effects do not even mention explicit or implicit taxes.  Others note the ACA’s employer penalties, without acknowledging that the Act also includes various implicit taxes on the employee side.  The purpose of this paper is to quantify the contributions of various ACA provisions to time series for the marginal tax rate on labor income. ….

The results are startling. [All ACA] … provisions combined raise marginal tax rates in 2015 by 10 percentage points of total compensation, on average, for about half of the nonelderly adult population and zero percentage points for the rest. ….

The ACA has not been introduced into a tax-free economy, so its marginal tax rate hikes add to marginal tax rates already in effect. I estimate that, by 2015, the average marginal after-tax share among household heads and spouses with near-median weekly earnings will have fallen to 0.50 from 0.60 in 2007, largely from the ACA but also from other expansions in safety net programs. That is a massive 17 percent reduction in the reward to working ….

Casey Mulligan, “Average Marginal Labor Income Tax Rates under the Affordable Care Act“, Supply and Demand (in that order), 29 August 2013.

The full paper, with the same title, was issued as NBER Working Paper No. 19365 (August 2013).

Healthcare is expensive everywhere, but it is abnormally expensive in the USA. ACA will not change this fact.

Professor Mulligan’s calculations include all payroll (“social security”) taxes in addition to income taxes. It is debatable whether mandated contributions to a government pension plan should be counted as taxes. If they are taxes, they are very odd, for the more one pays, the greater the retirement benefits. To me, this looks like forced saving (delayed consumption, purchase of bonds) rather than pure taxation. Also, a US household with two “near-median weekly earnings” is well-to-do, unlike the the median household, which has very small secondary spousal earnings, or none at all.

As this blog has repeatedly emphasized, means tests are equivalent to taxes. There are two ways to avoid this implicit taxation: (1) provide the benefit universally, without a means test, or (2) deny benefits – in this case healthcare – to those who cannot afford to purchase them. Mulligan – like Tea Party Republicans – prefers the second option to the first. I prefer universality, but, instead of providing all healthcare free (or with subsidies and tax breaks), I would restrict government provision to basic healthcare.

Casey B. Mulligan is author of The Redistribution Recession: How Labor Market Distortions Contracted the Economy (Oxford University Press, 2012).

Update. I have now read the full NBER paper, though probably not with enough care. Mulligan, in his conclusions (p. 23) clarifies that “the ACA’s implicit taxes will be experienced primarily by persons above the poverty line”, which is a relief to me. Even half the household heads and spouses empl0yed with weekly earnings near the median will experience no hike in implicit taxes. This is why the average hike is 10 percentage points for those near-median heads and spouses who experience higher implicit taxes, but only 5 percentage points for all near median household heads and spouses. Incidentally, a “median earner” is someone who earns $790 (2014 dollars) per week plus fringe benefits, “which is what the median nonelderly household head or spouse earned in 2007 during a week that they were working” (p. 4). It is not clear whether Mulligan is referring to median household earnings or to a median individual earnings, but his writing suggests the latter.

Now, the effect of means tests on the value of benefits received – i.e., the implicit tax – depends on the nature of the work decision taken. Mulligan looks at the results of three types of work decision: (1) moving from unemployment to employment; (2) entering the labour force; and (3) increasing weekly hours of work. Decision #3 is important because full-time employees (those working 30 hours or more a week) receive no ACA assistance for insurance premiums and out-of-pocket health expenses. Mulligan combines the three implicit taxes “into an overall statutory work incentive index” by taking a simple average (see pp. 2-3). I don’t understand why he does this, nor what such an “average effect” might mean.





Tags: ,

Comments are closed.