Working Paper: NBER ID: w3369
Authors: Robin L. Lumsdaine; James H. Stock; David A. Wise
Abstract: Recently many U.S. firms have offered "window" plans that provide bonuses to a group of workers if the worker retires within a specified short time span. This paper examines a window plan at a Fortune 500 firm, and addresses two main issues. First, what was the effect of the window plan on departures? Second, assuming a variety of possible firm objectives, what would be the design of an efficient window plan? These questions are addressed using the retirement model in Stock and Wise [1988a, 1988b] . The model, estimated using data for an earlier year, predicts well out-of-sample the subsequent large increase in retirements under the window plan. We find that while the firm successfully maximized departures, if its goal was to minimize either expected future wage payments or the current cost per induced retirement, the firm could have saved more with efficient plans constructed using the model. One interpretation is that the firm was primarily interested in reducing the overall size of the labor force or in retiring older employees to allow promotion of younger employees.
Keywords: Labor Force Reduction; Retirement Plans; Pension Incentives
JEL Codes: J26; J32
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| implementation of the window plan (J65) | increased retirement rates (J26) |
| window plan maximized departures (L93) | minimized expected future wage payments (J39) |
| window plan maximized departures (L93) | current costs per induced retirement (J26) |
| firm's objectives misalignment (L21) | less efficient plan designs (H21) |
| window plan implementation (C59) | facilitate promotion of younger employees (M51) |