Matching Incentives Raise Saving

Featured in print Digest

Higher match rates do significantly raise both the probability that the taxpayers will participate in an IRA and the size of their contribution.

Considerable attention has been paid recently to the fact that American consumers on average currently have a negative savings rate - they spend more than their income - or a savings rate close to that. As a result, policymakers, economists, researchers, and others have been debating the merits of various plans or suggestions to encourage people to set aside more money, thereby creating the savings that can provide capital for business and for building homes.

As it is, low- and middle-income families in particular appear to be saving little either for retirement or for any other purpose. Moreover, only a small percentage of families with income below $40,000 are covered under employer provided pensions. Further, these families are extremely unlikely to contribute to tax-advantaged Individual Retirement Accounts (IRAs). Indeed, in 2001 this income group had median financial wealth (half more, half less) outside of retirement accounts of a mere $2,200.

In Saving Incentives for Low- and Middle-Income Families: Evidence from a Field Experiment with H&R Block (NBER Working Paper No. 11680), co-authors Esther Duflo, William Gale, Jeffrey Liebman, Peter Orszag, and Emmanuel Saez analyze the effect of offering matching incentives to taxpayers if they decide to make contributions to an IRA at the time of tax preparation. This was done in a large, randomized field experiment carried out with the help of H&R Block, the nation's largest tax preparation firm. About 14,000 H&R Block clients, seeking tax help at 60 offices in predominantly low- and middle-income neighborhoods in the St. Louis metro area, were randomly offered either a 20 percent match on a contribution to an IRA, a 50 percent match, or no match at all.

These researchers find that the higher match rates do significantly raise both the probability that the taxpayers will participate in an IRA and the size of their contribution. Take-up rates were 3 percent for the control group getting no matching contribution, 8 percent for those getting a 20 percent match, and 14 percent for the 50 percent match group. Average IRA contributions (including for those who decided not to put money in the IRA, but excluding the "matches") for the 20 percent and 50 percent match groups were 4 and 7 times higher than in the control group, respectively. With matches included, IRA deposits were 4.5 and 10 times higher than with no match.

The authors also note that the help of H&R Block tax professionals, employing a computerized program, played a key role in the savings decisions of their clients. Those tax preparers who had been successful in steering clients in an IRA program prior to the start of the match experiment generated much higher take-up rates during the experiment.

What is also intriguing is that those tax filers who took advantage of the incentives were not gaming the system by contributing to an IRA, getting a match, and withdrawing the money very quickly afterwards. Even four months after the end of the experiment, about 90 percent of the differential effects of match rates on contributions were still present.

Another finding is that the experimental program stimulated proportionately far more savings than the existing Saver's Credit, a federal government program first implemented in tax year 2002 for tax returns filed in 2003, and scheduled to expire after 2006 (tax returns filed in 2007). This program also provides matching incentives for low- and middle-income tax filers. It is a non-refundable tax credit on the first $2,000 (for each spouse) contributed to various IRAs or voluntary pension plans (Keogh, 401k, 403b, SIMPLE IRA, and the like) of as much as 50 percent for those with the lowest "adjusted gross income" and 20 or 10 percent for those with higher incomes. For example, a low-income tax filer contributing $1,000 to a savings or pension plan would receive a $500 tax credit. Thus the out-of-pocket cost would be only $500, effectively a 100 percent match rate. However, many low-income tax filers would benefit little or not at all because they have little or no tax liability because of standard or itemi zed deductions, personal exemptions, and use of other non-refundable tax credits, such as the child tax credit.

Use of the Savings Credit is so modest that analysts might conclude that matching incentives are unlikely to represent an effective policy option to improve the financial security of future retirees. But the five authors suspect that the modest take-up and amounts contributed by taxpayers through the Saver's Credit may reflect the program's complexity and the hard-to-decipher way in which its effective match is presented.

They conclude that a savings program combining a significant, clear, and understandable match for saving, easily accessible savings vehicles, the opportunity to use part of an income tax refund to save, and professional assistance "could generate a significant increase in contributions to retirement accounts, including among middle- and low-income households.

-- David R. Francis