Date of Graduation


Document Type



This dissertation uses empirical estimates of the pension - nonpension saving trade-off suggested by the life-cycle hypothesis to test when individuals perceive their real pension benefits accruing. Pension theory presents two opposing views concerning the timing of pension benefit accrual. One view is that real pension wealth only accumulates as it is legally guaranteed. Thus, unvested workers have zero pension assets because if they quit, were fired, or the pension plan were terminated, they would be entitled to no legal benefits. Due to the structure of most pension plans, vested pension benefits accrue disproportionately near retirement. Under this "legal" view of the pension contract, real pension savings is backloaded toward the end of the career. Another view is that pension assets accumulate under an implicit contract, independent of legal liabilities. The worker anticipates continuous employment with the firm through retirement. This "implicit contract" implies a more even rate of real pension benefit accumulation. Because these two views have differing impacts on the path of wealth accumulation, empirical testing of the timing will test the implications of the two theories. Two tests are proposed. The first, based on the life cycle consumption theory which predicts that private pension contributions and nonpension saving are perfect substitutes estimated the trade-off using values of pension wealth developed from the two opposing benefit accrual theories. Using data from the 1983 Survey of Consumer Finances, no trade-off was found to exist. The second test is based on the pension theories' differing impact on the path of wealth accumulation. These paths are developed by looking at the relationship between pension participation, tenure and wealth accumulation. Once the path is developed, it is compared to the paths predicted by the different pension benefit accrual views. The results of this test show that the slope of an individual's nonpension saving curve significantly declines at higher tenure levels, implying that variation of the pension contribution--saving trade-off occurs across tenure levels. Thus, this test suggests a legal view of pension wealth.