Saving in Dynamic Life-Cycles Models

Project Start:07/2020
Researchers:Raimond Maurer, Olivia S. Mitchell, Sehrish Usman
Category: Household Finance
Funded by:SAFE

European workers increasingly accumulate pension assets during working life, and in their golden years, spend retirement wealth in an orderly fashion. They do so against the backdrop of aging populations, the demise of traditional defined benefit plans, a persistent low-return environment, and stressed national pay-as-you-go pensions. In response, nations encourage workers to invest in tax-qualified Individual Retirement Accounts (IRAs) offered by regulated financial intermediaries such as asset managers, banks, and life insurers. German Riester plans, the recently by the European Parliament introduced Pan European Pension Product (PEPP), and the US 401(k) plans being prominent examples. A unique feature of IRA’s is that savings are incentivized by deferring taxes on workers’ contributions. In this way, most IRA‘s are taxed according to an “EET” regime: workers contribute out of pre-tax earnings, recognize pre-tax investment earnings in their IRA’s, and pay income tax on withdrawals during retirement. Yet there are economic costs of such tax incentives. First, IRA are illiquid and private households cannot use them in case of unexpected shocks (health, unemployment, divorce) during working life. Second, this policy has a large current budgetary cost due to tax-deferred IRA contributions. Hence policymakers have recently proposed to treat all future retirement contributions to a “TEE” regime, in which workers would contribute to their pensions out of after-tax income, and then no additional tax would be levied thereafter. The objective of this project is to evaluate the economic consequences of alternative IRA taxation for workers’ (a) saving/investing behavior, (b) work/retirement patterns, (c) income/asset inequality, and (d) economic welfare. We extend our prior research by developing a realistically calibrated life cycle model with endogenous retirement, work effort, portfolio allocation and location. The model takes into account capital market shocks, uncertain labor income and remaining lifetimes, and the rich institutional details relevant to the tax and social security benefit structure in Germany and the US. After matching simulated model outcomes to empirical data for the current tax-regime, we then compare model results for heterogeneous households (low/middle/high education, singles/families, etc.) for the new tax regime. The project will inform policymaker by exploring alternative strategies for IRAs under a reformed old-age security system.