Reforming a Broken Health Care System is No Joke: A Response to Brad Delong’s Blog Post
I read Brad Delong’s blog post “This Is a Joke, Isn’t it?” on my recent piece “McCain’s Health Credit: The Intersection of Health Policy and Tax Policy,” Tax Foundation Fiscal Fact No. 144, September 16, 2008, with great interest. In my piece I tried to make it clear that the McCain health credit proposal is no panacea. It addresses important problems with the current tax treatment of employer-based health insurance. Improved incentives can help control health care costs. But the policy also increases the tax subsidy for health care from its current level of about $3.6 trillion over ten years to about $5.0 trillion over ten years. This large increase in the aggregate subsidy might well work in the opposite direction.
There are also concerns about the effect of the policy on the employer market. This is partly mitigated by keeping the payroll tax exclusion for employer-based health insurance (i.e., not completely leveling the playing field between the group and non-group markets). Usually, proposals to change the tax treatment of health care are accompanied by other proposals to reform insurance markets to ensure sufficient pooling and to increase subsidies for those with high risk/poor health status. But these proposals also more often than not take the form of objectives rather than well defined or articulated policies.
Delong’s main point is that he does not see how the McCain health credit could possibly increase the number of newly insured by 15 million or more. The idea is simply that some, perhaps many, would have the means to purchase health insurance under the McCain health credit. Let’s consider the size of the full subsidy under the McCain proposal. First, McCain provides the $5,000 and $2,500 credits for family and individual coverage. Second, he does not completely replace the existing tax subsidy for employer-based insurance. While repealing the income tax exclusion, he retains the payroll tax exclusion, which is about one-third of the existing $3.6 trillion subsidy (over ten years). The payroll tax exclusion lowers the price of health insurance purchased through one’s employer from $1 to about 85 cents for those below the Social Security wage cap (and to just 97 cents for those above the wage cap). Consider a family purchasing a $6,000 policy. After payroll taxes, the cost of this policy would be $5,100. The McCain health credit would cover another $5,000, leaving the family with a $100 after-tax cost for the policy. At this low price, it seems likely that the family would purchase the insurance.
Now, the more important question is: Can someone really find family policies that cost in the neighborhood of $5,000? The answer depends on what the policy covers. Insurance at this price probably would not be a generous plan with first dollar coverage or low deductibles. Instead, it is much more likely to be a plan with higher deductibles that is more focused on providing true insurance against catastrophic losses rather than a more generous plan that includes a lot of prepayment for routine and predictable medical expenses. But this is precisely one of the objectives of the policy: to reduce the current tax bias that encourages people to funnel routine health expenses through insurance policies.
Searching www.ehealthinsurance.com, I found that the annual premium for the top ten picks for family coverage with two adults born in 1970 (i.e., in their late thirties) and two children ranged from $2,208 to $6,048 in Berkeley’s zip code of 94720. The deductibles ranged from $500 to $7,000. The plans with a $3,000 and $5,000 deductible had annual premiums of $5,304 and $3,828, respectively. Obviously, this casual observation is not meant to suggest that everyone can get insurance at these rates, because clearly they cannot. Those with poor health status, with high risk profiles or in high cost areas would undoubtedly pay more, and possibly a lot more. The point, however, is that the level of subsidy provided by the McCain health credit is large enough to allow many to purchase real insurance that will provide a badly needed financial backstop in times of serious trouble. It would provide a substantial benefit to many. So, now, let’s turn to where the estimate of 15 million came from.
The estimate of 15 million comes from work on similar proposals about the time the Administration put forward its proposal back in early 2007 for a new standard deduction for health insurance (SDHI) to replace the existing income and payroll tax exclusions for employer-based health insurance. At the time, estimates were produced from various sources for the original SDHI proposal and related credit proposals.1 Estimates for the SDHI proposal were prepared by the Congressional Budget Office (CBO), the Joint Committee on Taxation (JCT), the U.S. Treasury Department of the Treasury, and the Lewin Group.
These organizations estimated that the SDHI proposal would increase the number of newly insured (net of any effects related to the group market) by between 6 million and 9.2 million. The JCT and Lewin Group were at the high end with estimates of 8.5 and 9.2, respectively. The Treasury came in with a more conservative estimate of 6 to 7 million (to the chagrin of the Administration, I might add).
Estimates were also produced by the Treasury for various proposals to replace the income and payroll tax exclusions for ESI with a credit. The estimates for the credit proposals’ effect on the newly insured ranged from about 11 to 15 million. The estimates at the higher end of this range were for versions of the proposal with fewer strings attached to the excess credit (i.e., the credit amount in excess of the cost of health insurance). (Incidentally, another problem with the McCain health credit is that is requires the excess of the credit to be deposited into an HSA. I am not sure why this string needs to be attached. It would decrease the take-up and means that the link between the tax subsidy and health care spending is not completely broken, which seemed to be a key merit of the approach.)
Now, the McCain health credit differs from the proposals analyzed by the Treasury Department in one critically important respect: the McCain health credit retains the current payroll tax exclusion for employer-based health insurance. This difference should have two effects on the estimates of earlier credit proposals, both of which should lead to higher estimates.
First, by retaining the payroll tax exclusion, the McCain proposal does not fully level the playing field between the group and non-group markets. Indeed, roughly speaking, and pointed out in my piece earlier this week, it retains about one-third of the existing tax subsidy for employer-sponsored health insurance. Thus, the effects of the proposal on the group market should not be as great under the McCain proposal as compared to the earlier proposals analyzed by the Treasury Department. While this component of the Treasury estimates is not public, suffice it to say that depooling was perceived as a significant problem and was one of the explanations for Treasury’s more conservative estimates.
Second, the credit proposals analyzed by Treasury were roughly revenue neutral, but the McCain credit is not. Because the proposal retains the payroll tax exclusion, it provides a large tax cut. This tax cut, in effect, increases the federal tax subsidy for employer-based health insurance from the current $3.6 trillion over ten years to $5.0 trillion over ten years. Indeed, as shown in Figure 1 in my piece earlier this week, the value of the McCain health credit exceeds the tax value of the current income tax exclusion (where I assumed an average family premium of $14,000, not the lower $12,000 as suggested by Delong in his posting) at most income levels. The greater generosity of the McCain health credit relative to the earlier credit proposals analyzed by the Treasury should translate into the McCain health credit having a larger effect on the number of uninsured.
Of course, these are very difficult estimates to make (as suggested by the large variance in the estimates cited above for the Administration’s SDHI proposal), so it is difficult to put a precise number on the McCain health credit. But, if one finds the earlier estimates believable, then it is likely that the McCain health credit should have more potent effects simply because it is a more generous proposal and does less harm to the employer market.
Another point to emphasize in an analysis of the McCain health credit is that it is highly redistributive. That is, it redirects the tax subsidy for health care towards those with low and moderate income, which is where a lot of the uninsured happen to reside in the income distribution. I would think that many who are concerned with government aid to the less fortunate would find this, plus spending another $1.4 trillion (over ten years) on health care, to be attractive features of the McCain proposal.
Given the interest among some on the left for the credit alternative to the Administration’s SDHI proposal at the time, I was a bit surprised that Senator Obama did not propose the credit himself. Indeed, a paper analyzing the Administration’s SDHI proposal co-authored by Jason Furman (Policy Director for Senator Obama’s Presidential campaign) and others saw some merit in the original SDHI proposal and suggested that a credit alternative would address some of its limitations. 2
1. These estimates are discussed in, Robert Carroll, “The Economic Effects of the President’s Proposal for a Standard Deduction for Health Insurance,” National Tax Journal Vol. LX(3) (September 2007): pp. 419-431. In particular, see tables 1 and 2 on pages 425 and 430.
2. Len Burman, Jason Furman, Greg Leiserson and Roberton Williams, “The President’s Proposed Standard Deduction for Health Insurance: An Evaluation,” Tax Policy Center, February 14, 2007.