A Life-Cycle Analysis of Retirement Savings and Portfolio Choices: Optimal Asset Allocation and Location with Taxable and Tax-Deferred Investment
by Zhe Li, State University of New York at Stony Brook
The expansion of defined contribution (DC) pension plans provides households with an opportunity to save and make investments in a tax-favored fashion, so that they face new choices of which types of assets and how much of each asset to hold in the conventional taxable accounts and the tax-deferred pension accounts. This project develops and estimates an intertemporal life-cycle model of optimal consumption and portfolio choice, which incorporates a tax-deferred retirement saving account together with a taxable saving account. The purpose of this study is to provide a framework for better understanding how people save for retirement, maintain an optimally diversified portfolio, and at the same time reduce the tax burden of owning financial assets. Recognizing how households allocate assets in tax-deferred and taxable accounts will provide valuable insights in the role of pension plan characteristics and the effects of policy reforms.
In this research I adopt a dynamic framework in which households receive stochastic income and simultaneously choose how much to save, which assets to hold, and how much to invest in tax-deferred and taxable accounts. Because of the important role of plan characteristics in DC pensions, employer matching policies are explicitly incorporated into the model. I find that the employer matching strategies, both in the form of cash and company stocks, directly increase the returns of the retirement wealth, and consequently boost equity holdings in the tax-deferred account and reduce those investments in the taxable account. In addition, if employer matches take the form of stocks, they tend to reduce the voluntary equity investment of the employees in the retirement account. The model is estimated using the method of simulated moments, using data mainly from the Survey of Consumer Finances (SCF), and parallel programming techniques will be applied in order to reduce computational time. This study will be extended to analyze the effects of three policies on households’ behavior: a regulation limiting employer stock-matching in DC pensions; a partial privatization of the current Social Security system; and a negative aggregate shock in the financial market.