Savings and framing effects on participation in individual retirement accounts: More evidence from tax return data

I examined the association between economic, savings, and psychological factors on participation in traditional Individual Retirement Accounts (IRAs) (1983–1985). The data were panels of tax returns representing households qualifying for the maximum IRA contribution and whose only sources of income were employment and investments. Along with traditional economic variables, my regressions included psychological factors such as framing effects based on adaptive expectations. Although both economic and psychological constructs were important in explaining savings behavior, the latter were shown as more salient. Households having less favorable than expected withholding positions increased IRA participation, a finding corroborating prior research. Savings propensity (SAVE) and past participation were the most important factors linked to IRAs. Unexpected investment income was significantly related to IRA participation, providing evidence that deductible IRA contributions represent new savings rather than reshuffled old savings. The policy implications of this study suggest that savings plans redesigned to encourage greater retirement savings should include tax benefits that are in temporal proximity to the desired savings behavior.