The Washington Post has a story today saying that ending retirement savings tax expenditures would save little money. The argument is a familiar one, and one that requires an understanding of how tax expenditures are scored.
The article cites a study by the American Society of Pension Professionals and Actuaries, which finds that limiting tax expenditures for retirement saving would not save very much, despite large yearly estimates in lost revenues for the federal government. Because 401(k)s have only been around for a few decades, tax-free contributions to these plans significantly exceed taxable withdrawals, making the tax expenditure appear larger than it will save.
This may be true, but the article makes this point--that tax expenditures don't save as much as we think--for the cutting of other tax breaks as well (the title in the print version reads "Ending tax breaks won't fix budget, study says"). As the article states:
For example, if the mortgage interest deduction were eliminated, people would probably shift their investments to other tax-preferred vehicles, thereby denying the tax man a portion of his expected rewards, said Ed Kleinbard, a former director of the congressional Joint Committee on Taxation, which, along with the Treasury, estimates the value of federal tax expenditures.
It's likely that eliminating an individual tax expenditure would save less than it currently costs the government, since taxpayers could shift towards other tax preferred activities. The $1 trillion estimate that we see from Treasury or JCT simply measures how much revenue we forgo each year by having these breaks in place. It says little about how much we save from doing away with individual tax breaks.
Regardless, though, even when we talk about individual tax expenditure changes, it is clear that they will save a significant chunk of change. For example, eliminating the deduction for state and local taxes would save $97 billion in 2016 (according to CBO's Budget Options), while the estimate for the size of the deduction in 2016 is $98 billion. Or consider the charitable deduction, as we did on Friday. Simply putting a two percent of income floor on the deduction would raise $23 billion in 2016, or one-third of the size of the deduction in that year. Also, note that these are actual cost estimates, which account for the effects of taxpayers responding to these changes.
However, we know that comprehensive reform of tax expenditures would minimize the differences between static cost estimates for various expenditures and the savings potentially achieved by reducing or eliminating them. Simply put, the fewer preferences there are in the tax code, the fewer places there are for taxpayers to put their money in an effort to lower their tax burden. Thus, the shifting that might occur if we eliminate only one large tax expenditure would be less pronounced if we get many of them in one fell swoop, as some fiscal plans have proposed.
And even if potential savings don't quite reach the sum of the current cost estimates for various expenditures, the totals would still be significant, as we showed in a paper looking at some potential reforms to certain tax expenditures. Revenues will have to be part of the equation if there is to be a bipartisan deal on a long-term debt reduction package. Raising revenue by scaling back or eliminating tax expenditures would have the added benefit of simplifying the tax code. Cleaning up the code would help make our tax system fairer and more efficient, especially if the tax breaks on the chopping block are poorly targeted, overly generous, or distributionally upside-down.