Tax policy colloquium on "Long-Term Objectives for Government Debt"

At yesterday's NYU Tax Policy Colloquium, my past and future co-convenor Alan Auerbach presented "Long-Term Objectives for Government Debt," a paper he wrote for a conference in Sweden (arranged by a new government entity there, the Swedish Fiscal Policy Council).

Whew, this is going to be a tough one, I thought, when I saw the first page, which starts: "Finanspolitiska radet are en myndighet som har till uppgift att gora ... " etc. Okay, just kidding there. That actually is the first bit after the title page, but it comes from the soon-to-be-published book's credits, and it's the only part that's written in Swedish rather than English.

Alan identifies the three main issues associated with budget deficits and public debt as generational equity, economic efficiency or performance, and fiscal sustainability, which he identifies with a possibility of a fiscal disaster such as default or hyper-inflation. (The fourth big issue is its political economy effects, mainly reflecting political incentives to (a) defer and/or under-specify financing for government outlays and (b) pre-commit future governments' budgets because they may not share one's currently ascendant priorities. I quibbled a bit about Alan's list, notwithstanding that I have used exactly the same list in some of my writing. (Consistency being the hobgoblin of petty minds, after all.)

On generational equity, Mihir Desai argued, and I tend these days to agree, that, while the issue is important and while better information about government policy is surely welcome, it is hard to say how the policy we are actually following compares to the optimal policy. We know so little about future generations' circumstances relative to our own, and about the actual marginal costs and benefits of shifting consumption opportunities one way or the other. Sustainability, not generational equity, is actually the big enchilada so far as telling us that U.S. fiscal policy is on a dangerously bad path is concerned.

On economic efficiency or performance, I did the quibbling. As I define this category, there isn't a lot of efficiency at issue in the standard microeconomic sense. There are stabilization / counter-cyclical fiscal policy issues, though unfortunately deficits provide a poor measure given that the composition of tax and spending changes is so important (e.g., transfers to people with high versus low propensity to consume, rewarding existing investment versus new investment). And there are issues of effects on national saving, although again here composition is important and the main complaint, if fiscal policy reduces national saving, is a positive externalities story about saving.

Alan also puts in category 2 issues of tax smoothing, or having more constant tax rates over time rather than having to raise them suddenly because one has finally woken up to the sustainability problem if taxes aren't adequate over the long run given spending levels. At the risk of being a nitpicker, however, I put this in category 3, sustainability issues.

Alan views the sustainability issues discontinuously, in terms of the very real risk of a big meltdown or credit event along the lines of a run in the bank. I agree that this is the really big and growing risk or concern about our unsustainable fiscal policy - explicit or implicit default, hyper-inflation, collapse of the banking system, and so forth. As I say in Taxes, Spending, and the U.S. Government's March Toward Bankruptcy, this actually might happen, and indeed it verges on certainty of happening IF the U.S. political system can't function adequately. (It usually has in the past, but the last seven years make one a bit less optimistic.) But I would include in this category, because they are merely lesser versions of the same problem and have the same cause, lesser versions of the harm caused by deferring any serious response to sustainability problems. Examples include:

(a) foregone tax smoothing, leading to unduly high and distortionary taxes or tax rates in the future because we didn't raise them more moderately sooner,

(b) foregone consumption smoothing, such as from steep Medicare cuts in the future that cut more into essentials because more moderate cuts, hitting less vital and valuable services, weren't imposed sooner, and

(c) tough times for future elderly people whose benefits are cut late in the game, when they can no longer react by saving more, leading to their own accentuated failures of lifetime consumption smoothing.

Anodyne though it may sound, this can actually involve really bad stuff that was completely avoidable. And again, we basically agree about all of this - more of a semantic debate regarding how to conceptualize it.

We also struggled with tough questions of fiscal measurement that lack good answers and probably always will. The fiscal gap - say $80 trillion or about 10 percent of the present value of all expected future GDP (these numbers are made-up, but probably within the realm of a reasonable estimate) - sounds alarming, but strictly speaking is just a statement about statements; that is, a measure of the degree to which a reasonable projection of the current policy path actually is impossible and will not happen. Mere words don't automatically hurt, however. Thus, in thinking about the problems (such as fiscal meltdown risk) implied by the fiscal gap, we need to think about softer variables such as degrees of pre-commitment or lost flexibility.

For example: Joe Stiglitz projects an Iraq war cost of $2 or 3 trillion. Medicare prescription drugs has an infinite horizon revenue cost of more than $20 trillion. But these two numbers aren't entirely comparable, because the Iraq costs are largely water under the bridge - little we can do about them now - whereas we could actually back off spending all that money on future Medicare prescription drugs. This is a point about irreversibility, not about present value or risk / variance. Alan's article introduces a couple of concepts such as "implicit liabilities" and "deferred tax assets" that attempt to advance thinking about these issues - which I agree they do, although this post is already too long for full further discussion to make sense here - but in dealing with soft variables through the medium of a fiscal measure we will never be entirely satisfied. And, how we think about something such as the Medicare prescription drugs dollars (assuming one agrees about the remaining flexibility) does in part depend on whether one is thinking more about fiscal meltdown issues or broader smoothing / optimization issues.