As organizations have increased their efforts to downsize over the past two decades, the amount of attention being paid to how organizations deal with early retirement issues has increased as well (Feldman and Kim, 2000; LaRock, 1999; Longo, 1999). In large part, that research has focused on how organizations structure, and how older employees respond to, incentives to accept early retirement (Kim and Feldman, 1998, 2000; Weckerle and Schultz, 1999). However, there has been very little research on the penalties associated with early retirement. That is, much more is known about the effects of incentives for early retirement than is known about the disincentives associated with penalties for retiring before eligibility for full pension benefits is reached.
Moreover, much of the research to date on early retirement decisions has a significant methodology problem. That is, much of the research in this area has examined individuals' willingness or intention to retire given a hypothetical set of incentives rather than their actual retirement behavior (Talaga and Beehr, 1995). Thus, much more is known about what individuals say they would do when given a choice about early retirement than about their actual retirement decisions (Kim and Feldman, 1998).
This article examines the factors which influence employees' retirement in defined benefit pension plans in which there are penalties associated with retiring before becoming eligible for full pension benefits. Using the actual retirement decisions of 2,205 public-sector employees, the research examines the differential effects of penalizing workers for retiring "early" in terms of age (before age 65), for retiring "early" in terms of years of service (before reaching 30 service years), and for retiring "early" by leaving both before age 65 and with fewer than 30 years of service.
Below we briefly explain the retirement plan formula used by the retirement system examined in this study and highlight the key differences between defined benefit and defined contribution pension plans. Then, in the next section, we review the literature on retirement decisions and present hypotheses on the factors which predict retirement before eligibility for full pension benefits is reached. This is followed by a description of the research methodology of the study (e.g., the research site and measures of variables). The final two sections of the article present the empirical results of the study and their implications for future research and management practice in the area of retirement planning.
RETIREMENT PLAN FORMULA
In defined contribution pension plans, employees and/or employers contribute a set percentage of the employee's annual salary to a pension fund. Funds are invested among various financial instruments (e.g., bonds, money market funds, stocks); the amount of pension benefits received is highly dependent upon the wisdom of employees' own investment decisions. Thus, funds available for retirement decisions are affected by returns from investment sources in addition to years of service and retirement age.
In contrast, in defined benefit pension plans, pensions typically are calculated by an organizational formula based on annual compensation, age at retirement, and years of service. The organization (not the employee) is responsible for investing pension funds. Unlike employees in defined contribution plans, employees in defined benefit plans are typically penalized if they retire before they reach a specified age or complete a preset number of years of service (e.g., 30). Organizations with defined benefit plans typically lower the amount of pension benefits employees receive if they retire before "normal" retirement.
Like many defined benefit plan formulae, the formula for employees in the present study is a function of compensation level, length of service, and age. The annual retirement pension benefit used in this organization was calculated by the following formula:
Average Annual Final Compensation X 1. …