Book Description

This thesis investigates why, in spite of the high tax and opportunity costs, a substantial fraction of workers withdraw money from their employer-sponsored retirement accounts upon leaving a job. I employ three national surveys--the National Longitudinal Survey of Youth 1979 (NLSY79), the Survey of Consumer Finances (SCF) and the Survey of Income and Program Participation (SIPP)--to evaluate explanations for this behavior. Often attributed to poor decision-making, withdrawals from employer-sponsored plans may in fact serve as an important mechanism for liquidity-constrained workers to smooth consumption in response to income shocks. Results from the NLSY79 suggest that withdrawals do appear to be driven largely by need, as workers who faced jobless spells upon separation as well as those with low holdings of liquid assets, and poor access to consumer credit markets, were significantly more likely to take some money from their retirement plans when leaving an employer. Yet there was little evidence to support the hypothesis that workers who suffered an adverse job separation were more likely to withdraw money from their plans if they were also liquidity constrained. The second chapter estimates the tax sensitivity of withdrawal decisions using estimates of effective federal and state marginal tax rates generated with the NBER Taxsim program. Estimates reveal that, depending on the tax price measure used, a one percent increase in the marginal tax rate increased the probability that workers preserved the tax-deferred status of the money in their retirement plans between 10 and 30 percentage points. State-level penalties for early withdrawals were also found to be an effective policy instrument for deterring cash outs among workers under age 55. In the third chapter, I take advantage of measures of savings goals and habits, as well as more specific information on the use of cash settlements, found in the SCF to further investigate behavioral explanations for disposition of employer-sponsored plans. Workers who reported credit constraints, a short planning horizon, not saving primarily for retirement and a major future expense for which they had not yet begun saving were significantly more likely to take money from their plans.