نقش و وظایف بیمه بهداشت و درمان فرد بازنشسته در بازنشستگی زود هنگام از کارمندان بخش دولتی

Most government employees have access to retiree health coverage, which provides them with group health coverage even if they retire before Medicare eligibility. We study the impact of retiree health coverage on the labor supply of public sector workers between the ages of 55 and 64. We find that retiree health coverage raises the probability of stopping full time work by 4.3 percentage points (around 38 percent) over two years among public sector workers aged 55-59, and by 6.7 percentage points (around 26 percent) over two years among public sector workers aged 60-64. In the younger age group, retiree health insurance mostly seems to facilitate transitions to part-time work rather than full retirement. However, in the older age group, it increases the probability of stopping work entirely by 4.3 percentage points (around 22 percent).

Individuals’ decisions about entering or exiting the labor force, and about switching employers, can be strongly influenced by non-monetary employer-provided benefits. For example, employer-provided group health insurance has been documented to create “job lock,” a phenomenon in which individuals remain in jobs that come with this benefit longer than they would otherwise (see, e.g., Madrian, 1994). One form of job lock occurs among older workers, who often have an incentive to delay retirement until Medicare eligibility–at age 65–in order to maintain group health coverage.1 But one important subset of workers–consisting of those whose employers extend group health coverage to retirees in addition to current employees–does not face this incentive. However, retiree health plans are rare in the private sector. According to the Kaiser Family Foundation (2013), among large firms providing employer-sponsored coverage, just over a quarter also provide retiree coverage, and that fraction has fallen sharply over time. In contrast, most public sector employers offer retiree health plans, allowing former employees who meet certain age and service requirements to participate in a group health plan before Medicare eligibility. In this paper, we examine the impact of retiree health coverage on the retirement decisions of public sector employees.2 While a large literature has examined the impact of retiree health plans in general, most previous studies focus on the private sector. Studying the public sector is important for several reasons. First, the public sector is large, accounting for around 16 percent of total nonfarm employment.3 Second, both public sector pensions and retiree health plans are severely underfunded. As public sector employers consider making changes to these plans, it is important for them to understand how these changes might affect work incentives. Finally, public sector workers might respond differently to retirement incentives than private sector workers. For example, individuals who select into public sector employment might be more risk averse than average, indicated by their willingness to accept a lower salary in exchange for more job security. Additionally, public sector employers typically offer very generous defined benefit pension plans, which also contain strong early retirement incentives. Public sector employees are also more likely to be unionized than their private sector counterparts. Because retirement is difficult to measure directly, in our empirical analysis, we use stopping full time work as a proxy. That is, we study whether generous retiree health insurance is associated with an increased pre-Medicare full time work exit rate among public sector employees with 5 or more years of job tenure. We model the probability of shifting out of full time work at ages 55-64 as a function of the generosity of retiree health coverage, as well as controls for demographics, health, job characteristics, work history, wealth, and pension plan retirement incentives. We focus on the decision to stop work before the age of 65 because retiree health coverage is most valuable for retirees in this group. Typically, a public sector retiree health plan is the primary payer for pre-Medicare eligible retirees. When a retiree becomes eligible for Medicare, however, he or she is required to enroll in Medicare, relying on the retiree health plan only as a secondary payer (Clark and Morrill, 2010). To preview our results, we find that retiree health coverage raises the probability of stopping full time work by 4.3 percentage points (around 38 percent) over two years among public sector workers aged 55-59 and by 6.7 percentage points (around 26 percent) over two years among public sector workers aged 60-64. These effects are somewhat larger than the corresponding point estimates for private sector employees; however the differences are not statistically significant. Among 55-59 year old public sector workers, retiree health insurance appears to facilitate transitions to part time work, roughly doubling the probability of shifting from full time to part time work over a two year period. However, it does not appear to have a statistically significant impact on the probability of stopping work entirely. On the other hand, among 60-64 year old public sector workers, retiree health insurance increases the probability of stopping work entirely by 4.3 percentage points (around 22 percent) over two years among public sector workers aged 60-64. As mentioned earlier, one motivation for studying the retirement incentives in public sector retiree health plans is that public sector employers are likely to consider changes to these plans in the near future. These changes will occur against the backdrop of the Affordable Care Act (ACA). As of this year, the ACA allows all individuals to purchase health insurance on state-run exchanges. In addition, many individuals are eligible to receive subsidies towards their health insurance purchases. Because the exchanges make subsidized health insurance widely available outside of employment, the ACA effectively provides all individuals with retiree health insurance. Thus, both public and private employers offering retiree health coverage may find that retiree health insurance is less attractive to workers than it was in the past. In response, we might expect to see employers dropping retiree health coverage, resulting in a smaller compensating differential in monetary wages between public and private sector employment.4 This option might be particularly attractive to public sector employers facing financial distress from both pension and retiree health obligations. Indeed, there have already been reports of public sector employers taking steps to shift retirees onto the ACA's exchanges (Niquette and Wayne, 2013). The remainder of this paper is organized as follows. Section II provides a summary of the prior research on retiree health insurance and retirement and an overview of public sector retiree health plans. Section III describes our data and methodology. Section IV presents our results. Section V discusses the policy implications of our results in light of the ACA. Section VI concludes.

This paper adds to the evidence that employer-sponsored retiree health insurance programs lead to higher rates of stopping full time work among 55-64 year olds. In this sense, the results of this paper are generally consistent with our earlier findings on employers in the private sector (Nyce et al., 2013). There are several data limitations in the current paper. First, we lack detailed information on retiree health plan rules and eligibility requirements. Second, our measures of the retirement incentives in public sector defined benefit pension plans are not as detailed as one would like. Finally, we are unable to control fully for possible selection into jobs that offer retiree health coverage based on unobservable characteristics. Overall, however, we find the evidence convincing that retiree health plans have a significant and large impact on the probability of stopping full time work between ages 55 and 64 in both the private and public sectors. Given the growing evidence that retiree health programs lead to earlier retirement, it is interesting to note that the Affordable Care Act (ACA) of 2010 offers what amounts to universal retiree health. Under the ACA, all retirees under 65 can now purchase health coverage through the state-based exchanges, and those purchases will be subsidized for all whose income in retirement is below 400 percent of the official poverty standard. The research on retiree health programs, including this paper, suggests that the ACA may lead to earlier retirements, particularly for those in the private sector who currently do not have access to subsidized health insurance in retirement before age 65. Indeed, a recent report by the Congressional Budget Office (2014) discusses this possibility and incorporates this effect into its estimate that the ACA will cause total hours worked to decline by 1.5 to 2 percent between 2017 and 2024. This estimate of the total reduction in work also includes other channels, such as the work disincentives created by the phase-out of subsidies as income rises. Moreover, the CBO does not specify what fraction of its estimated reduction in labor supply is due to early retirements. Thus, it is difficult to determine whether our results are consistent with the CBO's. The second ramification of the ACA is that it could change the competitiveness of state and local public sector jobs relative to private sector jobs without retiree health benefits. It does so by making one of the benefits of public employment universally available. Our speculation is that many state and local employers will drop retiree health benefits as a result. It makes little sense to continue paying for something that many employees can get anyway (through the exchanges). The net effect of dropping retiree health benefits would likely improve the budget position of state and local governments even though the ACA worsens the relative attractiveness of state and local jobs. In order to restore the overall attractiveness of these jobs and therefore maintain the quality and quantity of applicant pools, other job attributes–like salary or other benefits–would have to be improved. These compensating adjustments, however, could likely be achieved for less money than the cancelled retiree health insurance.