As we plunge headlong into tax filing season, let’s pause to reflect on tax-time savings initiatives—a major focus of asset-building efforts over the past decade. Through a growing number of innovative programs, low-income tax filers are encouraged to save part of their lump-sum tax refund, which they derive primarily from the earned income tax credit (EITC).

New York City’s SaveNYC pilot program offers a 50 percent match on refund dollars if the money is held in a savings account for a full 12 months. The program has been expanded to more Volunteer Income Tax Assistance sites throughout the city and has now been adopted in Newark, Tulsa, and San Antonio, under the name SaveUSA. The nonprofit organization Doorways to Dreams Fund just launched a nationwide SaveYourRefund Sweepstakes, offering cash prizes to those using IRS Form 8888 to commit their refund dollars to US savings bonds or other savings products.

As with any concerted effort to encourage workers to save, we should ask: Do these programs generate new savings? Or do they largely reward participants for doing what they would have done anyhow?

We’ll be better able to answer these questions after MDRC’s evaluation of SaveUSA underway in New York and Tulsa. In the interim, insightful evidence comes from a study by Kathryn Edin at Harvard University and Ruby Mendenhall at the University of Illinois about how low-income families use their EITC refunds. Edin and Mendenhall’s team conducted 194 in-depth interviews in Boston and Champaign-Urbana in 2007. Fully two-thirds (67 percent) used their refund to pay off back debt or overdue bills. More than one-third (39 percent) initially saved a portion of their refund, but only 21 percent still retained some of that savings after six months. Other uses included car-related expenses (27 percent) and home-related expenses (8 percent). Virtually all households used at least some of their refund for current spending needs. What’s striking about this evidence is how many of these low-income families used their tax refund t

o pay down debt. We should not be surprised if the upcoming SaveUSA impact study shows that participants hold more in savings but also face higher debt (than those not in the program), with little or no effect on net worth (in other words, no net increase in savings).

We should also ask: How do we want low-income tax filers to use their refunds? For purchasing assets, for emergency spending needs, for paying off debt, or for some combination of these? Tax-time programs are explicitly structured for asset building. The 12-month SaveUSA rule or the encouraged use of savings bond products makes it difficult to access funds for emergency expenses or for squaring one’s balance sheet. This seems needlessly restrictive. The upward mobility of low-income households rests on their ability to buffer themselves against financial shocks, avoid high-cost borrowing, and maintain their creditworthiness. Unless program rules are relaxed, tax-time savings initiatives will tend to benefit those with incomes at the upper end of the qualifying range, not those whose economic security is most at risk.