Esther Duflo (MIT), William Gale (The Brookings Institution), Jeffrey Liebman (Kennedy School of Government, Harvard), Peter Orszag )The Brookings Institution), & Emmanuel Saez (UC-Berkeley) have posted Saving Incentives for Low- and Middle-Income Families: Evidence from a Field Experiment with H&R Block on SSRN.  Here is the abstract:

This paper analyzes the effects of a large randomized field experiment carried out with H&R Block, offering matching incentives for IRA contributions at the time of tax preparation. About 14,000 H&R Block clients, across 60 offices in predominantly low- and middle-income neighbourhoods in St. Louis, were randomly offered a 20% match on IRA contributions, a 50% match, or no match (the control group). The evaluation generates two main findings. First, higher match rates significantly raise IRA participation and contributions. Take-up rates were 3% for the control group, 8% in the 20% match group, and 14% in the 50% match group. Average IRA contributions (including non-contributors, excluding the match) for the 20% and 50% match groups were 4 and 7 times higher than in the control group, respectively. Second, several additional findings are inconsistent with the full information, rational-saver model. In particular, we find much more modest effects on take-up and amounts contributed from the existing Saver’s Credit, which provides an effective match for retirement saving contributions through the tax code; we suspect that the differences may reflect the complexity of the Saver’s Credit as enacted, and the way in which its effective match is presented. Taken together, our results suggest that the combination of a 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. This strategy would not, however, produce contribution rates anywhere near 100%.