« Back to Results

Social Security Policy Interactions and Spillovers

Paper Session

Saturday, Jan. 5, 2019 10:15 AM - 12:15 PM

Atlanta Marriott Marquis, International 3
Hosted By: American Economic Association
  • Chair: Nicole Maestas, Harvard University

Intergenerational Altruism and Retirement Transfers: Evidence from the Social Security Notch

Anita Mukherjee
,
University of Wisconsin-Madison

Abstract

Old-age financial security is an important topic as life expectancies increase and resources remain scarce. As the elderly navigate this new retirement landscape, their preferences (or lack thereof) for altruism towards children will play a key role in determining the intergenerational transfers of money and time. In this paper, I contribute new evidence on the extent of altruistic preferences using quasi-random variation in Social Security benefits induced by an inflation-indexing mistake. I find support for pure altruism, as individuals who received extra money because of the mistake passed on about 6.75% to children via inter vivos transfers without receiving any care in return. On the contrary, children reduce caregiving almost monotonically with increases in parental Social Security benefits, suggesting that Social Security impacts time use among younger generations. These results have practical implications for how reforms in Social Security or related programs will be impacted by a key behavioral component of utility.

Is it Bad to be Green in a Greying Firm? An Analysis of the Impact of Postponed Retirements on Younger Workers Wage Growth?

Peter Berg
,
Michigan State University
Marissa Eckrote
,
Michigan State University
Mary Hamman
,
University of Wisconsin-La Crosse
Daniela Hochfellner
,
New York University
Matthew Piszczek
,
Wayne State University
Christopher J. Ruhm
,
University of Virginia

Abstract

Most countries with pay-as-you-go public pension systems have raised the age of eligibility for full benefits and enacted other reforms to encourage longer working lives. A large body of literature shows older individuals do work longer when pension rules change in ways that make early retirements less attractive. In this paper, we examine potential effects of policy-induced postponements of retirement among older workers on the career advancement of their younger colleagues. Using administrative employment data for all workers within the same establishment, we show gradual increases in pensionable age introduced in a 1993 reform had highly variable impacts across establishments because of pre-policy differences in worker age distributions. This variation serves as the source of identification for examining the influence of delayed retirements on younger and prime aged workers. Our data span years 1994 to 2014 and contain 44.7 million observations with information on approximately 10 million workers and employed in 45,344 establishments. The detailed information in the employment records allows us to define promotion using two methods. First, we rank occupations by the average wage of all workers across all establishments in the same occupation. We define a promotion as a change to a higher ranked occupation relative to the prior year. Using this measure, we observe over 383,000 promotions in our dataset. The second measure of promotion we use based on observing a 5 percent or larger increase in wage. Using this measure, we observe over 483,000 promotions. We compare the incidence, timing, and size of promotions among younger workers in establishments that vary in the incidence and magnitude of pension reform induced extensions of working-life among older workers.

Policy Interactions at Retirement

Leora Friedberg
,
University of Virginia
Megan Miller
,
University of Virginia

Abstract

The U.S. tax and social insurance system is extraordinarily complicated. Even when some of these policies are designed to prompt certain actions (the Social Security actuarial adjustment is intended to motivate retirement delays) and penalize others (Required Minimum Distributions are intended to discourage oversaving in retirement accounts), difficulties in understanding them may lead to bunching at salient policy thresholds, which may be an overreaction for some and an underreaction for others. Further difficulties arise when these policies interact in complicated (and, likely, unintended) ways. These policy interactions makes it difficult for people not only to know their marginal and average tax rates are for particular actions, but also how one action may affect the marginal and average tax rates they face for another action. Retirement involves policy interactions to a degree that has not been fully appreciated. We illustrate how numerous policies associated with retirement decisions interact. We then document the frequency with which policy interactions affects typical households in the Health and Retirement Study. Lastly, we analyze how people respond to policy complexity.

Peer Effects and Retirement Decisions: Evidence from Pension Reform in Germany

Mary Hamman
,
University of Wisconsin-La Crosse
Daniela Hochfellner
,
New York University
David A. Jaeger
,
City University of New York-Graduate Center
John Nunley
,
University of Wisconsin-La Crosse
Christopher J. Ruhm
,
University of Virginia

Abstract

Peers are hypothesized to influence a wide range of economic behaviors, but identification of causal relationships is difficult. We use employee-employer matched administrative data to estimate the impact of peers on retirement decisions. Using exogenous variation in pensionable ages created by a reform that affected particular birth cohorts, we find robust evidence of spillovers from the retirement behavior of the affected cohorts to the unaffected cohorts. The magnitude of spillovers is large but limited to cohorts within approximately four to five years of age. In these cohorts, a one percentage point reduction in the share of peers who are affected by the reform and retire is associated with (at least) a 0.25 percentage point reduction in the share of peers retiring who are unaffected by the reform. We conclude peer effects amplify the effects of changes in retirement incentives. Our estimates imply reforms that encourage later retirements may produce changes in the share of older workers who retire that are at least 27% larger than the estimated response to changes in own incentives.
Discussant(s)
Gopi Shah Goda
,
Stanford University
David A. Jaeger
,
City University of New York-Graduate Center
Christopher J. Ruhm
,
University of Virginia
Cynthia L. Doniger
,
Federal Reserve Board
JEL Classifications
  • J2 - Demand and Supply of Labor
  • H5 - National Government Expenditures and Related Policies