Volume 13 Issue 3 - May 2013
MRRC Holds Ninth Annual Spring Research Workshop
The ninth annual MRRC Research Workshop was held at the University of Michigan’s Ross School of Business in Ann Arbor on April 19-20, 2013. As always, this was an occasion for stimulating presentations and lively discussion of research findings and methods. The event was well attended on both days with 63 participants.
MRRC Director John Laitner was pleased with the turnout and audience participation, commenting, "The annual Research Workshop has provided a great venue for researchers to present and discuss their work, and for potential new MRRC scholars to come and get a sense of what we do."
A number of University of Michigan researchers attended, however, most scholars traveled quite a distance to participate, coming from places such as Alabama, California, Missouri, New Hampshire, New York, Pennsylvania, Wisconsin, British Columbia, and Frankfurt, Germany.
The lunchtime speaker was Charles P. Blahous III, Public Trustee for the U.S. Social Security and Medicare Trust Funds and Research Fellow at Stanford University’s Hoover Institution. He discussed the need for a shift in policy options for funding Social Security.
MRRC hosts the workshop to enable scholars to share ideas and receive feedback about their work in progress. As a result, the work is often ongoing and analyses still preliminary.
As is customary, to allow broader participation, presentations are 10 minutes each, followed by 10 minutes of open discussion. Topics included: Well-Being in Retirement/Insurance; Portfolio Choices; General Equilibrium Effects of Social Security; Social Security and its Interactions with Household Behavior; Measuring Well-Being; The Great Recession’s Effects on Household Behavior; Health and Labor Supply; What Can We Learn from Japan?; Intergenerational Interactions; and Pensions.
Maria Casanova spoke about Divorce, Remarriage and Old-Age Poverty written with Clement Joubert. They study the long-term consequences of middle-age divorces for women. Using PSID data on labor force participation, income, poverty rates, health status, and disability they find that post-retirement welfare is lower for divorcees than widows, married women, divorced men, and married men. They develop a life-cycle model of consumption, labor supply, and remarriage decisions to understand the mechanisms leading to negative post-retirement outcomes for divorcees.
Eric French presented the paper Medicaid Insurance and Old Age, coauthored with Mariacristina DeNardi and John Bailey Jones. Medicaid was originally designed to insure the poorest retirees against medical expenses. Using data on retired singles age 70 and older from the Health and Retirement Study AHEAD cohort, the authors find that while the poorest individuals are the most frequent recipients of Medicaid, wealthy individuals who live longer, and run up more expensive medical bills, sometimes claim the highest benefits. They model hypothetical changes to Medicaid benefits to determine how much people value Medicaid insurance.
Lee M. Lockwood discussed Informal Care and the Market for Long-Term Care Insurance, cowritten with Ethan M.J. Lieber. The authors ask whether informal care undermines coverage of formal home care. Despite the fact that nursing home costs are about $75,000 a year, only 10 percent of Americans own Long-Term Care Insurance. Medicaid pays for more than 2/5 of formal long-term care. About 80 percent of long-term care is informal care provided by family and friends. Lockwood and Lieber analyze data from a cash-and-carry experiment in "consumer-driven care" in states where participants can spend their Medicaid-financed home care on informal care instead of formal home care.
Marco Angrisani spoke about Understanding Heterogeneity in Household Porfolios: The Role of Cognitive Ability and Preference Parameters, coauthored with Maria Casanova. They explore the ability of cognition and individual risk preferences to explain a variety of different economic decisions. Investors with different levels of cognitive and numerical ability diverge not only in the way they allocate wealth among asset classes, but also in the way they hedge against risks.
Daniel Benjamin discussed Biased Beliefs About Random Samples: Evidence from an Integrated Experiment, coauthored with Don Moore and Matthew Rabin. They analyze experimental evidence on three biases in beliefs about random samples: the law of small numbers (also known as the gambler’s fallacy), nonbelief in the law of large numbers, and bin effects. They find evidence for all three biases in an incentivized experiment.
Raimond Maurer discussed preliminary work on the paper Family Impacts on Life-Cycle Portfolios and Retirement, coauthored with Andreas Hubener and Olivia S. Mitchell. They propose to analyze how changing family status over the life cycle influences consumption, portfolio choice, life insurance demand, and labor supply, taking into account current U.S. Social Security rules.
Christopher Tonetti discussed preliminary work on the paper Precautionary Motives and the Annuity Puzzle written with John Ameriks, Andrew Caplin, and Matthew D. Shapiro. Using data from the MINYVan survey, a collaboration between the University of Michigan, New York University, and Vanguard, the authors seek to understand the reasons people don’t annuitize their retirement savings more often. Annuitization may compete with their desires to leave a bequest or to save to cover unexpected expenses. Measuring the strength of preferences for these choices could help design better insurance products.
R. Anton Braun discussed The Joint Effects of Social Security and Medicaid on Incentives and Welfare cowritten by Karen A. Kopecki and Tatyana Koreshkova. Figure 1 shows estimated percentages of healthy and unhealthy females and males by marital status for each age group ranging from 65 to 99. These estimates are constructed by estimating survival properties for each type of individual using data from the Health and Retirement Study.
In his paper, Fiscal Policy Effects in a Heterogeneous-Agent, Overlapping-Generations Economy with an Aging Population, Dr. Shinichi Nishiyama sought to address the imbalance in government revenue caused by decreases in tax revenue even as government transfers increase because of the aging population. He models different policy changes to show how the government might be able to finance the budgetary cost of the aging population in the coming decades.
Lauren Nicholas discussed the Health, Work and Program Participation, coauthored by John Bound and Helen Levy. They investigate the possibility of errors in disability benefit decisions: Are applicants really too sick to work? Are the right applicants denied benefits? Would early retirees be able to work longer if Social Security claiming ages rise? They compare the health of Health and Retirement Study respondents claiming benefits at different ages. Preliminary results show that DI applicants and recipients use considerably more healthcare at age 65 than nonapplicants, and earlier Social Security claimers (ages 62 to 64) use more care than later claimers (ages 65 and older).
Robert Pollak spoke about Couples and Social Security Claiming, a paper cowritten by Shelly Lundberg. He stated that there has been an uneven retreat from marriage since about 1950. It is uneven in terms of race, ethnicity, and socioeconomic status. The retreat is evidenced by later ages of marriage, especially for younger cohorts, and an increased probability of cohabitation or living alone at older ages. Social Security benefits are currently tied to legal marriage. How will the marriage retreat impact benefits for those who are unmarried for long periods or never married?
Victoria Prowse presented the paper The Design of Unemployment Transfers: Evidence from a Dynamic Structural Life-cycle Model. She asks, how much does a household value the ability to adjust the labor supply of the second earner (flexible spousal labor) and how does this interact with public insurance? She builds a structural model of job search, labor supply, retirement, and savings of singles and couples throughout the life-cycle.
Stefan Staubli discussed Labor Supply Effects of Raising the Retirement Age: Evidence from Switzerland, coauthored with Rafael Lalive. They ask, how does increasing the full retirement age (FRA) affect individual labor supply, the labor supply of spouses, and savings behavior and health? Reforms in Switzerland increased the FRA for women to age 63 in 2001 and to age 64 in 2005. Preliminary results show that there is a strong effect on benefit claiming and labor market exit, but no effect on labor force exit or claiming behavior for spouses.
Arie Kapteyn talked about the paper Dimensions of Well-Being, coauthored with Jinkook Lee, Caroline Tassot, Hana Vonkova, and Gema Zamarro. The authors study the relationship between different subjective well-being concepts and scales: evaluative, experienced, and eudemonic. They ask, what is the relation between the three types of measures? How many underlying dimensions are there? Does the response scale matter? How reliable are the measures?
Jinkook Lee presented Would Delaying Social Security Eligibility Age Influence Subjective Well-Being? A Cross-Country Study of the Effect of Retirement on Subjective Well-Being, coauthored with Arie Kapteyn, Jinkook Lee, and Gema Zamarro. They exploit cross-country and within-country variation in eligibility ages for retirement pensions to study the effect that retiring later has on subjective well-being. They use internationally harmonized longitudinal data from HRS and SHARE.
Susann Rohwedder talked about the paper Income Shocks and Subjective Well-being: Evidence from Unemployment Spells during the Financial Crisis, coauthored with Michael Hurd and Caroline Tassot. The use of subjective well-being measures in many scientific surveys shows the interest of policy makers in measuring and understanding determinants of subjective well-being. The authors investigate the main channels through which income might influence subjective well-being. They use the American Life Panel, a monthly longitudinal dataset.
John Karl Scholz discussed preliminary work on the paper Assets and Liabilities of Cohorts: The Antecedents of Retirement Security, with coinvestigators J. Michael Collins and Ananth Seshadri. Using the Survey of Consumer Finances, they characterize cohort patterns of net worth and debt of American households. They aim to establish a benchmark for identifying potentially vulnerable households, and to determine the role of consumer debt. Are there correlations between distributions of net worth or debt and attitudinal and consumer behavior factors related to financial capability?
Erik Meijer gave a presentation on Early Social Security Claiming in the Great Recession, coauthored with Michael Hurd and Susann Rohwedder. What explains the increased Social Security claiming by 62 year olds in 2008 and 2009? Will this trend continue? Is earlier claiming due to increased cohort size, or unemployment induced liquidity constraints? Conversely, did the drop in wealth from these events exert pressure to retire and claim SS later? Early results indicate that the jump in number of benefit awards is largely explained by cohort size, but excess unemployment also accounts for about one percentage point higher in the claiming rate between ages 62 to 63.
David Neumark talked about the paper Did Age Discrimination Protections Help Older Workers Weather the Great Recession? researched with assistance from Patrick Button. During the Great Recession unemployment durations (but not unemployment rates) of older individuals rose far more dramatically than for other workers. Did stronger state age discrimination protections help older workers? Using monthly CPS data, the authors find that for men, stronger protections against discrimination made things worse. For women, stronger protections were associated with relatively smaller increases in unemployment durations.
Purvi Sevak discussed the paper The Great Recession, Workers with Disabilities, and Implications for Retirement Security, coauthored with Onur Altindag and Lucie Schmidt. The authors examine whether those with greater underlying risk of disability experienced worse employment outcomes during and after the recession. They find that recession-related declines in consumption were significantly larger for those with higher baseline "disability risk." "Disability risk" is not associated with larger employment decreases, yet it is associated with a greater probability of leaving one’s prior job and of involuntary job separation.
Yonatan Ben-Shalom presented Longitudinal Trends for Young Social Security Disability Awardees, a paper coauthored with David Stapleton. Disabled Adult Children (DAC) experience disability onset before age 22, and they are awarded benefits when a parent retires, dies, or qualifies for disabled worker benefits. They are likely to be on disability and in Medicare for many years. In 2007, the number of DAC beneficiaries increased faster than the increase in the number of disabled beneficiaries.
Susan Chen discussed her paper The Economic Responses of Recently Disabled Households. Studies have found that many rejected disability applicants do not return to work. Do they rely on their spouse’s earnings? Do they rely on other transfer programs? Using SIPP data and Social Security earnings records, Chen seeks to track the earnings and health status of rejected applicants during the course of 10-20 years to determine whether they use other social insurance programs.
Giovanni Gallipoli presented the paper Household Responses to Individual Shocks, written with Laura Turner. Analyzing data from the sample of Canadian men from the 1999-2010 Survey of Labour and Income Dynamics, they examine the extent to which couples can offer insurance to each other in the face of economic and health shocks. They find that when wages depend on labor supply, the main earners (mostly men) have incentive to protect accumulated human capital stock, which explains why the effects of disability onset on labor supply appear to be cumulative. Sorting into marriages based on current and expected disability status means that healthier men, and those who face lower work-limitations, are on average more likely to be married.
Selahattin Imrohoroglu talked about Achieving Fiscal Balance in Japan, a paper coauthored with Sagiri Kitao and Tomoaki Yamada. Japan has the highest debt to GDP ratio of 144 percent among advanced economies and is experiencing one of the fastest rates of population aging. Further pension reform is needed say the authors after evaluating current policy and medium demographic projections. They find significant nonpension and pension deficits and an increasingly large interest burden on the budget. Increasing female labor force participation would go a long way to improving the nonpension side of the budget.
Kenichiro Kashiwase spoke about Options for Pension Reforms in Japan, cowritten with Mashiro Nozaki and Kiichi Tokuoka. Reforms in 2004 introduced macro indexing, mandating automatic reductions to the replacement ratio in tandem with population aging. However, the government subsidy to cover the cost of the basic pension was raised from 1/3 to 1/2 by 2009. Options to reduce the subsidy would be higher pension eligibility age and better targeting. The authors also argue that women’s earnings should contribute more to the pension system.
Olivia S. Mitchell presented a preliminary version of her paper with John W. Phillips and Yong Yu, Retirement in Japan and the United States: Cross-national Comparisons using the Japanese Study of Aging and Retirement (JSTAR) and the U.S. Health and Retirement Study (HRS). The authors are in the process of examining the relative importance of health, wealth, family, and other factors in work and retirement at older ages cross-nationally.
Elena Gouskova discussed Inheritance Reporting in the Health and Retirement Study Data: Evidence of Forward Telescoping. Forward telescoping occurs when events are allocated to more recent periods than when they actually took place. Backward telescoping occurs when events are mistakenly allocated to earlier time periods. Gouskova finds that inheritance reporting in one survey wave is a good predictor of inheritance reporting in the following wave. Consistent with forward telescoping, there is a tendency to report the same inheritance more than once. She estimates that 10 percent of reported HRS inheritances are duplicates, and total transfers are overestimated by 10 percent.
Meghan Skira presented the paper Dynamic Wage and Employment Effects of Elder Parent Care. How does caregiving for an elderly parent affect a woman’s current and future labor force participation and wages? Using HRS data, she formulates a model that captures the tradeoffs between employment and caregiving, and calculates the value of caregiving. Modeling policy experiments, she finds that work leaves encourage more work during and after intensive care provision, while the caregiver allowance discourages work. In addition, work leaves do not induce a large increase in intensive caregiving, while the caregiver allowance does.
Nahid Tabatabai talked about HRS Pension Wealth and Other Measures for 1992 to 2006 with Code and Documentation, coauthored with Alan L. Gustman and Thomas L. Steinmeier. The authors measure HRS pension wealth and make the data available to users. The data files underlying the tables in the book Pensions in the Health and Retirement Study are available on the HRS website: hrsonline.isr.umich.edu/data/index.html. Code, documentation, updated wealth & disposition of pension data files are also available through HRS.
Joelle Saad-Lessler gave a talk titled Explaining the Historic Decline in Retirement Plan Coverage between 2001-2012: The Role of Bargaining Power, coauthored with Teresa Ghilarducci, Lauren Schmitz, and Anthony Bonen. Analyzing the 2001 and 2012 March Supplement to the Current Population Survey, they find that while all states show a decline in pension coverage since 2001, right-to-work states have lower average retirement plan coverage than non-right-to-work states. The authors conclude that the bargaining environment matters for workers. Proponents of so-called right-to-work laws should be mindful of how they impact pension sponsorship and retirement readiness.
Clement Joubert discussed preliminary findings from How the Design of a Pension System Influences Old Age Poverty and Gender Equity: A Study of Chile’s Private Retirement Accounts System, coauthored with Petra E. Todd. The authors develop a model of lifetime labor and saving decisions of single- and two-person households. They estimate the model using longitudinal pre-reform data on Chilean households and validate the model using post-reform (2009) data. They evaluate the short- and long-term effects of the 2008 pension reform.
Frank Stafford talked about his paper At the Corner of Main and Wall Street: Family Pension Responses to Liquidity Change and Perceived Returns, coauthored with Thomas Bridges. Using data from the Panel Study of Income Dynamics, they find that many families treat retirement savings accounts as sources of ready cash for current needs and discretionary spending rather than as sources of income in retirement.
David Powell presented the paper titled Can Liquidity Constraints Explain Early Retirement Behavior? Powell seeks to determine whether liquidity constraints are an important factor in early retirement and early Social Security claiming behavior. Using Social Security claiming behavior and labor earnings, he plans to look at individuals’ responses to tax rebates issued in 2001 and 2008.