Thursday, April 24, 2008

Final NYU Tax Policy Colloquium of 2008

Today Jason Furman presented a paper on healthcare, rightly (I'd say) lambasting the bizarre Cogan-Hubbard-Kessler plan to combat moral hazard in healthcare by increasing it, and proposing his own plan which is hardly bullet-proof (and presumably unenactable) yet has some clear virtues.

Brad DeLong once said in a post that the big issue among healthcare economists is whether the problem is moral hazard (people over-spending because they aren't paying in full hence don't require marginal benefit equal to marginal cost) or adverse selection (failure of insurance markets to permit risk pooling and actuarially fair pricing, on balance, for diverse groups).  Jason, to his credit, sees both as problems not just one.

Cogan-Hubbard-Kessler (henceforth CHK) seem to think the big problem is moral hazard from over-insurance, largely caused by the fact that the income tax permits exclusion of employer-provided health insurance but not deduction of medical expenses, creating an incentive to over-insure.  Hence they propose to make all healthcare expenses deductible, thereby externalizing the moral hazard / over-insurance problem from employer-provided healthcare to everything.

One point that doesn't seem to have occurred to CHK is that the incentive to over-insure goes ONLY to the difference between non-deductibility and coverage at the value of the income tax exclusion.  Say the marginal tax rate (MTR) is 42%, taking account not only of federal income tax marginal rates but state & local income taxes (though they are to a degree deductible) and payroll taxes (although on the Social Security part one may accrue benefits along with tax liabilities).  Under their theory, they should predict that the co-payment required for routine expenses, to the extent these are over-insured in response to the tax incentives, is 58%.  Higher co-pays can't be explained by their theory since they go beyond the tax benefit.  And without higher co-pays than this, they have no theoretical basis for expecting moral hazard to be reduced.

Anyway, CHK want to reduce moral hazard by increasing it, in the sense that it gets peeled out of employer-provided healthcare since you get a federal co-payment based on the MTR even if it is uninsured.  But they are addressing a problem that should not exist by their lights, given the absence of any tax incentive to go beyond the federal tax saving in designing the co-pay.  So their diagnosis must be wrong in order for their prescription to seem superficially appealing.

Perhaps they want to bring the current employer-provided healthcare system more generally crashing down, but that would strike me as a bit reckless and rash.  Albeit, no more so than the fiscal implications of their plan, which (in conventional Republican style these days) would add a vast sum to the fiscal gap.  But who's counting anyway?

Jason's plan is to provide refundable credits that aren't tied to the amount you actually pay - you get it for having qualifying insurance without regard to how much you pay.  So you pay at the margin both for the amount of qualifying health insurance that you select and for outlays outside the plan.  Hence moral hazard is addressed, along with adverse selection if the plan in other respects is successful.

One perplexity posed by the paper is that it suggests that, at the margin, healthcare outlays provide zero marginal healthcare benefits, because consumers (when economizing because their share of the cost has been increased) can't choose properly between reducing the healthcare that actually provides benefits and that which is pure waste or affirmatively harmful.   This is not theoretically implausible, since consumers (myself included) are poorly informed and have to rely on doctors whose incentives and ideology may be a bit off, but the evidence for it is weak, and if it is true a much more radical response than anything Jason suggests might be in order.   It might suggest that we can't rely on consumer preferences at all here, and/or that healthcare should be taxed like pollution even if health insurance is subsidized.

Jason is not on the side of the debate that says universal mandates should definitely be used, but the answer to that one was perhaps beyond our institutional expertise as a group (which is not to say that the experts all agree).

This is it for the year, so far as the NYU Tax Policy Colloquium is concerned.  I'll miss it, albeit cherishing my newfound time and freedom.  A great year, reflecting the efforts of my co-conveners (Kevin Hassett and Mihir Desai) along with the substantial contributions of both regular and sporadic attendees.  We'll be back next January, with Alan Auerbach as my co-convener, and 13 of our speakers are already set.  But more on that later.

On Sunday I head to Israel for a week.  I'll be giving two talks which presumably will be listed on Tax Prof Blog.  One concerns my tax & accounting paper, and the other a paper to be written later this summer called "The Intellectual State of the Play in U.S. International Taxation."  Back in the USA at a horrifically early hour on Monday, May 5.  Then in mid-June I leave to teach for two weeks in Singapore again (followed by two weeks vacation in Vietnam), and before that happens I need to finish my book in progress, "The U.S. Corporate Tax - What is It, and Where Is It Headed?"  Maybe the Coen brothers will want an option on that one, since it has a bit of suspense, but in the interim the Urban Institute Press will be publishing it.  All kidding aside, I do feel good about that book, and hope it can combine informing a lay audience (such as law and business school students) with being enlightening to policymakers and serious academics. 

No comments: