Why expiring tax cuts should actually expire

On Tuesday, the Senate Finance Committee held a hearing on “Extenders and Tax Reform: Seeking Long-Term Solutions.”  It’s about time!  The charade of annual or biennial debate about perpetually “expiring” tax provisions is terrible tax policy and a symbol of our failure to come to terms with budget reality.

If you need help sleeping, download the Joint Committee on Taxation’s (JCT’s) annual list of expiring tax provisions—tax laws that have built in sunset dates. The latest list is here. More than 70 provisions expired in 2011. Based on past experience, almost all will be extended after some debate about how or whether to pay for their budgetary cost.  They include very popular provisions like the research and experimentation (R&E) tax credit and the so-called “patch” that prevents tens of millions of middle-class taxpayers from falling prey to the AMT. There is a host of tax credits for energy efficiency and alternative fuels. There’s the tax deduction for state and local sales taxes. And the list includes the ethanol tax credit that provides a windfall to Midwestern farmers while contributing to higher food prices here and starvation in the rest of the world.

Given that the merits of some of these provisions are debatable (to put it nicely), subjecting them to periodic review would seem to be the epitome of frugal budgeteering. But pretending that the provisions will only be in place for a year or two makes them appear to be much more affordable than they will turn out to be.  The AMT patch and the Bush tax cuts have been extended without offsetting tax increases or spending cuts, in part justified by the relatively modest budgetary cost of a short-term extension.

University of Virginia law professor (and former JCT director) George Yin has argued that all tax breaks should be subject to annual review, as discretionary programs are, so that Congress will have to take an up or down vote on the provisions to continue them. George argues that temporary provisions are the only tax laws where the costs are fully accounted for because permanent tax cuts can cost much more outside the budget window than in the five- or ten-year period considered when the legislation is debated.  Temporary measures, in contrast, typically incur all of their cost within a few years.

Even if you buy George’s analysis of the budgetary effects (Howard Gleckman does not), there are other issues. One is that the provisions are continually extended and become a vehicle for more dumb tax laws. When I worked at the Treasury Department, we deliberately aligned temporary measures so that they would come up for renewal at the same time as the R&E credit, because we knew there would be a vote on extending that popular measure. The AMT patch has become another must-pass extender. It seems likely that the less popular measures would have a hard time surviving if they had to be voted on separately. And without the vehicle for mischief, some dubious new measures might never become law.

Furthermore, Congress’s chronic procrastination means that many temporary provisions are not extended until after they have expired. Astute calendar watchers will no doubt have noticed that 2011 came and went while the 70+ expiring provisions waited in vain for action.  There will be an extenders bill this year, but the uncertainty means that taxpayers might rationally discount the actual tax breaks and that raises the odds that they will simply provide windfalls rather than affecting behavior. Research expenditures, for example, have a very long lead time. Investors might guess that the R&E credit will be in place when the research occurs, but they can’t be sure that the provision will not change. There’s some evidence that research is relatively unresponsive to the credit, and its ephemeral nature might be part of the explanation. Similarly, taxpayers don’t know for certain whether they can take tax credits against the AMT. (The provision allowing the use of personal credits is up for renewal.) And, of course, all of those green tax incentives are doubly uncertain—they may or may not be extended and if they are, their value will depend on the size of the AMT patch (since business credits are not allowed against the AMT).

If Congress is going to spend the money to provide incentives, it shouldn’t undermine them by making them uncertain.

It might, however, make sense to make new provisions temporary. Congress might even go a step further and mandate data collection so that Treasury or other agencies could study their effectiveness. But temporary provisions should not be extended endlessly.

I propose a “three strikes and you’re out” rule.  After a provision has been extended three times, it should either be made permanent (and its cost fully offset) or it should be erased from the books.

A better solution still would be fundamental reform, which is at least hinted at by the hearing title and Chairman Baucus’s statement at the hearing.  Many of the expiring provisions would not survive a rational reworking the tax code.  RIP.