For many people, working after beginning retirement benefit collection is a way to enhance financial security by increasing income. Existing research has shown that retirees are sensitive to the Social Security earnings test, which restricts the amount of earnings some beneficiaries can receive. However, little is known about the effects of other types of policies on post-retirement employment. Instead of restricting earnings, many public pension plans restrict the number of hours beneficiaries can work. I use return-to-work rules limiting the number of hours of employment in a state's public pension plan and administrative data on employment and retirement to determine the rules’ effects on retirement decisions and post-retirement labor supply. I find that the increases in the maximum number of hours of post-retirement employment lead to no change in retirement benefit collection and to increases in part-time work among retirees. As such, these policies appear to be binding on the labor supply decisions of some employees. These results are relevant for designing policies aimed at extending work-lives or improving the health of pension systems.

I analyze the effects of state public pension parameters on the retirement of public employees. Using a panel data set of public sector workers from 12 waves of the Health and Retirement Study, I model the probability of retirement as a function of pension wealth at early and normal retirement eligibility and Social Security coverage in the public sector job. I find that becoming eligible for early retirement, or receiving an early-out offer, significantly increases the probability of retiring. I do not find any effect of retirement wealth levels. These findings suggest that state legislative action to affect retirement decisions and reduce future pension costs would be most effective operating through plan eligibility rules and early-out incentives.

Do the retirement patterns of public-sector workers differ from those in the private sector? The latter typically face a retirement landscape with exposure to market uncertainties through defined-contribution pension plans and private saving. Public-sector workers, in contrast, are often covered by defined-benefit pension plans that encourage retirement at relatively young ages and offer financial security at older ages. We examine how private- and public-sector workers transition from full-time career employment, with a focus on the importance of gradual retirement. To our surprise, we find that the prevalence of continued work after career employment, predominantly on bridge jobs with new employers, is very similar in the two sectors, a result with important implications in a rapidly aging society.

This study documents an important role for Social Security income in workers’ retirement timing. About 40% of public school teachers are not covered by Social Security. This provides an opportunity to analyze the causal impact of Social Security on retirement timing by comparing covered and non-covered teachers. Using individual-level data from the American Community Survey, we find robust evidence of higher rates of retirement among covered teachers at Social Security eligibility ages. This pattern is confirmed using an alternative regression model of participation in the teacher labor force. These estimates suggest that, should the federal government mandate full inclusion in Social Security for all public sector workers, the retirement timing patterns of newly covered teachers and other public sector workers would likely change.

We investigated labor force and health outcomes in cities experiencing fiscal difficulties to assess how those difficulties might impact their employees. We matched 23 cities with bond downgrades and 31 cities with stable bond ratings to sampling units in the Medical Expenditure Panel Survey. Starting the year before the downgrade and for the four subsequent years, the rate of separation from local public employment fell in the cities with downgrades relative to the comparison group. Self-reported health may have worsened, but there were no statistically significant effects on health care use or spending.

The State of Washington, as part of a State Innovation Model (SIM) grant, is changing the payment model within state employee health insurance plans. The system is moving away from traditional fee-for-service reimbursement to value-based payment, through insurance design (the creation of accountable care network insurance products) and bundled payment strategies. New plans were rolled out January 2016 (enrollment occurred in late 2015), with the stated goal of getting 80% of state employees covered by plans that contain value-based purchasing within the next 5 years. The goal of payment reform is to improve member experience, member health, and cut costs. However, changing health insurance during employment can directly and indirectly change labor market outcomes. Decreasing costs of insurance could lead people to remain in the state-employment sector longer. However, it could also influence retirement timing, through changing the relative costs of insurance and through improving health.

This paper examines who switches to value-based insurance, where the insurance explicitly decreases premiums without changing out-of-pocket costs. We find that the peak age for switching insurance plans is 35–45, even among the subsample of individuals who would not need to change their usual sources of care. Second, we look at the labor market activity – both leaving the state-employee sector and retiring from state-employment – and find that younger workers with value-based insurance plans are less likely to leave state employment. Further, we find evidence of value-based insurance, available at a reduced cost to both employees and retirees, leads to a shifting downward in the distribution of retirement age. While these findings support the existence of both the price and income effects, the effect sizes are rather small.

Choices regarding the disposition of wealth at retirement can have substantial implications for retirement income security. We analyze the factors determining annuity payout option choices within the context of a public sector defined pension plan with no default annuity option. Using combined administrative records and survey data, we explore the role of individual and household characteristics as well as risk preferences, time preferences, and financial literacy. We also document retiree well-being and satisfaction with retirement decision making. The evidence is consistent with predictions over which households might benefit most from each annuity option. Comparing retirees who chose different types of annuities, we find that these groups of retirees report very different levels of well-being in retirement. All retirees report lower levels of retirement income security over time, with strong differences among those who chose different types of annuities.

Many believe that global capital markets will generate lower returns in the future versus the past. We examine how persistently lower real returns will reshape work, retirement, saving, and investment behavior of older persons using a calibrated dynamic life cycle model. In a low return regime, workers build up less wealth in their tax-qualified 401(k) accounts versus the past, claim social security benefits later, and work more. Moreover, the better-educated are more sensitive to real interest rate changes, while the least-educated alter their behavior less. Interestingly, the distribution of wealth is more uniform in periods of persistent low expected returns.

This paper compares the relative strengths of working longer vs. saving more in terms of increasing a household's affordable, sustainable standard of living in retirement. Both stylized households and actual households from the Health and Retirement Study are examined. We assume that workers commence Social Security benefits when they retire. The basic result is that delaying retirement by 3–6 months has the same impact on the retirement standard of living as saving an additional one-percentage point of labor earnings for 30 years. The relative power of saving more is even lower if the decision to increase saving is made later in the work life. For instance, increasing retirement saving by one percentage point 10 years before retirement has the same impact on the sustainable retirement standard of living as working between 1 and 2 months longer. The calculations of the relative power of working longer and saving more are done for a wide range of realized rates of returns on saving, for households with different income levels, and for singles as well as married couples. The results are quite invariant to these circumstances.