Working Paper: NBER ID: w0811
Authors: Paul Taubman
Abstract: Pensions and age specific death rates are intertwined in several ways. Pensions provide a mechanism to remove the uncertainty about date of death from consumption planning. Age specific death rates determine the cost and value of pensions. In this paper, we use the Retirement History Survey to estimate reduced form functions for the probability of having a pension when the person reaches 65 and on the dollar amount of the pension. We also evaluate the effect of 15% drop in age specific death rates from 1973 to 1979 on the costs of a pension. We find that the probability of having a pension is related to education, marital status, occupation, industry and assets. The probability equation is very similar for males and females. We find that the sharp drop in death rates has only a marginal impact on the cost of providing a pension.
Keywords: pensions; mortality; health economics
JEL Codes: I3; J3
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| pensions (H55) | consumption planning (E21) |
| decrease in age-specific death rates (J11) | expected value of pensions (H55) |
| decrease in age-specific death rates (J11) | pension costs (J32) |
| education (I29) | probability of having a pension (H55) |
| marital status (J12) | probability of having a pension (H55) |
| occupation (J69) | probability of having a pension (H55) |
| industry (L89) | probability of having a pension (H55) |
| assets (G32) | probability of having a pension (H55) |
| health status (I12) | probability of having a pension (H55) |
| pension status (H55) | mortality (I12) |