Automatic IRA Reform: How Much Will Retirement Income Go Up?

The only available answer to this question — from a 2011 study by Butrica and Johnson that uses the Urban Institute DYNASIM microsimulation model — is that after-tax per-adult family income at age seventy will rise on average by three to five percent after a lifetime of experience with automatic IRAs. And the average percentage rise would be from six to twelve percent for those in the bottom half of the pre-reform age-seventy family income distribution (Butrica-Johnson, Table 2). Are these Butrica-Johnson estimates plausible? It is difficult to answer that question because they do not explain all their assumptions in the paper, because they do not vary most of their assumptions to measure the sensitivity of their estimates, and because the DYNASIM model is not publicly available, which prevents others from using it to determine the sensitivity of their estimates to changes in assumptions. The plausibility of these DYNASIM estimates can be studied, however, using the PENSIM microsimulation model, which is comprehensively documented, extensively validated, and publicly available at no cost. The analysis described here can be reproduced and checked for accuracy by anyone interested in installing the three PSG models. ∗The analysis in this essay has been funded exclusively by PSG; the analysis does make use of PENSIM, a microsimulation model whose development has been funded by the U.S. Department of Labor since 1997, but the views expressed here do not necessarily reflect those of any government agency. Barbara A. Butrica and Richard W. Johnson, “How Much Might Automatic IRAs Improve Retirement Security for Lowand Moderate-WageWorkers?,” Urban Institute Program on Retirement Policy Brief #33, July 2011 (hereafter referred to as Butrica-Johnson). PENSIM and two related models are documented and available for download at the PSG website 〈www.polsim.com〉.