NYU Tax Policy Colloquium - Peter Orszag on "Saving Social Security"

Yesterday's Tax Policy Colloquium at NYU featured Peter Orszag of the Brookings Institution, presenting his forthcoming Journal of Economic Perspectives paper (with Peter Diamond), entitled "Saving Social Security" and available here. With the main speakers being Orszag, Jason Furman (formerly with the Kerry campaign), and myself, one could certainly argue that not all viewpoints were equally represented. Still, we tried to have an intellectually open-minded discussion.
Points of general interest that Orszag made, or that emerged from the discussion, included the following:
--If we want to increase national saving, which seems to have been a key aim of privatization proponents such as Martin Feldstein in the late 1990s budget surplus era, then a much more straightforward and fruitful maneuver than playing with budgetary language in the hope of keeping Congress in check (i.e., by making the diverted payroll revenues look more off-limits) would be to exploit the overwhelming empirical evidence that people voluntarily save a lot more when the default rules on what happens to their paychecks require them to opt out of saving rather than to opt in. Orszag noted a study in which something like 25% would opt in to a tax-deductible savings plan, but 80% would decline to opt out if the default was changed, and 75% would remain in if they were forced to express a preference rather than having the default rule decide. I noted that, in a sense, taking advantage of this to increase national saving is a bit like a (genuine) free lunch. While not "free" in one sense, since the amount people have available for current consumption actually does decline if they save more, it would be "free" in the sense of not requiring them to depart from actual consistent preferences, insofar as these can be discerned.
--Even though some versions of the individual accounts idea have respectable supporting arguments, the Bush plan as it stands has aroused opposition from a number of leading conservative or pro-privatization economists, in particular (I might add) those who happen not to be angling for Bush Administration jobs. For example, Robert Barro views the Bush plan as simply another costly expansion of entitlements that would be harder than existing Social Security to scale back. Laurence Kotlikoff, who has strongly favored privatization for some time, thinks the Bush plan is terrible and will simply increase his bete noire, the burden being placed on younger and future generations.
--It is often argued that the government can't be given big current year positive cash flows from Social Security to play with, because if the money is there it will be spent. The argument assumes, however, something that needs to be demonstrated. So long as the government can borrow, the claim "if it is there they will spend it" misconceives how spending decisions are actually made. They can spend it whether it is there or not. And it is difficult to show convincingly that Congress's budget decisions are strongly or in a consistent way affected by Social Security's effect on the current year unified budget deficit.
For example, does anyone really think Bush wouldn't have gone to war in Iraq with zero financing but for the Social Security surplus? Or that he wouldn't have done the 2003 tax cut?
Once deficits are in the several hundred billion dollar range, we tend to think of all the big numbers as the same. Politicians get credit for a smaller deficit only when they can show a salient change relative to some arbitrary baseline ("I cut it in half") or when it crosses a salient marker. For example, a President might get much more credit from reducing a $50 billion deficit to zero than from reducing it from $600 billion to $450 billion. And this view of things tends to undermine thinking that the Social Security surplus has a strong generalizable effect on other budget decisions.
"If it is there they will spend it" seems to reflect a misguided analogy between household behavior and government behavior. Liquidity-based budget constraints don't bear on governments in quite the same way as on households, since governments have the power to levy taxes tomorrow, thus expanding their ability to borrow today.