« Back to Results

Aging and Race

Paper Session

Monday, Jan. 4, 2021 10:00 AM - 12:00 PM (EST)

Hosted By: American Economic Association & Committee on the Status of Minority Groups in the Economics Profession
  • Chair: Perry Singleton, Syracuse University

Age and the Labor Market for Hispanics in the United States

Joanna Lahey
,
Texas A&M University
Roberto Mosquera Moyano
,
University of the Americas

Abstract

The population of Hispanics in the United States has been steadily growing over time, but little is known about what happens to Hispanic workers as they age. This study explores this issue using three different methodologies and focuses on high school graduates, the largest education group. First, using the American Community Survey (ACS) and Community Population Survey (CPS), we find no substantive differences in the probability of employment or probability of being in the labor force overall. However, wages are flatter over the lifecycle for Hispanics compared to non-Hispanics. Similarly, we find few differences in job accession and separation rates between Hispanic and non-Hispanic high school graduates. These findings only provide information about the general equilibrium in the labor market and may mask potential discrimination. The second and third studies each use laboratory experiments to look at bias directly. The second study finds a quadratic effect of age on hireability for non-Hispanic resumes, particularly for female applicants, with an increase occurring sometime in the 50s or 60s. Hispanic applicants, in contrast, have hireability ratings that decline with age, particularly for women. The third study widens the signal of Hispanic names to include combinations with Hispanic first names and takes the study to mTurk. Preliminary results suggest that Hispanics overall suffer a negligible hireability penalty, and in general, their ratings increase with age, in contrast with the second study that only signals Hispanic by last name.

Healthy Aging among Older Mexican Immigrants

Emma Aguila
,
University of Southern California
Maria Casanova
,
California State University-Fullerton

Abstract

A large literature has documented the so-called Hispanic Health Paradox (HHP), that is, the epidemiological finding that some health outcomes of Hispanics residing in the United States are comparable to, and in some cases better than, those of non-Hispanic whites. In this paper, we use nationally-representative data from the Health and Retirement Study (HRS) to examine the evidence in support of the HHP for Mexican-born men and women above age 50 living in the US. Analyses based on self-reports of diagnoses of chronic conditions and health behaviors yield evidence consistent with a weak form of the HHP for all health outcomes in the sample of males. Instead, Mexican immigrant women have significantly higher odds of diabetes than NH whites. Overall, Mexican immigrant men have a more favorable health profile, relative to their non-Hispanic white counterparts, than do Mexican immigrant women. The disparity between genders may reflect actual health differences. It may also be a data artifact due to the fact that Mexican immigrant women are more likely than men to interact with the healthcare system, and hence less likely to have undiagnosed health problems. Using indicators of chronic conditions based on biomarkers and anthropomorphic measures available in the HRS, we find evidence consistent with underdiagnoses of diabetes for both Mexican immigrant men and women and of hypertension for Mexican immigrant men. The weak form of the HHP holds for some health outcomes but not others, with Mexican immigrants of both genders displaying an advantage in hypertension, no advantage in obesity, and a disadvantage in diabetes. Moreover, the health disparities between Mexican immigrant men and women disappear once the biomarker-based measures are considered.

Incarceration and Black-White Cohort Differences in Male Employment, Earnings, and Social Security Wealth

Gary Engelhardt
,
Syracuse University
Michael Eriksen
,
University of Cincinnati

Abstract

This paper provides a birth-cohort-level empirical analysis of how the rise of institutionalization (and incarceration, in particular), especially among African-American men, has affected labor force participation, employment, and earnings. The analysis focuses on men born between 1955 and 1982 and draws on data from the 1980, 1990, 2000, and 2010 decennial censuses, combined with the 2001-2016 waves of the American Community Survey (ACS), which sample the institutionalized population. Accounting for the number of former prisoners in the non-institutionalized population and non-random selection into institutionalization, the estimates indicate that a one percentage point increase in the percent of black men who were former prisoners in a state is associated with a one-third of a percentage point reduction in the labor force participation rate and a one-half of a percentage point reduction in the employment rate for black men. Based on these estimates, the rise in the number of former prisoners in the non-institutionalized population between 1980 and 2016 can explain about one-quarter of the decline in labor-force participation and real earnings for black men over the same period, and substantially more of the decline in employment. Furthermore, there is a pronounced shift in the shape of the age-earnings profile for men who came of age when crack cocaine entered the United States and sentencing grew stricter, concentrated among men born in the mid-to-late 1960s. For these birth cohorts, the peak in the age-earnings profile shifts from the late 40s to the late 30s, and for African-American men does not return to the late 40s until the youngest sample cohorts. These same affected cohorts experienced the labor-market effects of the Great Recession in what otherwise would have been their peak earning years, further hollowing out their age-earnings profile.

The Effect of Social Security Retirement Benefits on Food-Related Hardship among Older Americans

Perry Singleton
,
Syracuse University

Abstract

This study examines the effect of Social Security Retirement Benefits on food-related hardship, particularly at the early retirement age of 62. Claims for Social Security Retirement (Old Age) Benefits spike at the early retirement age of 62, the earliest age of eligibility for retired worker benefits. Rust and Phelan (1997) attribute this spike to individuals with limited wealth and private pensions who are unable to retire before age 62. This implies that individuals who claim Social Security benefits at the early retirement age are more likely to face food-related hardship and that Social Security benefits may reduce food-related hardship at age 62 and beyond.

The Effect of Opioids on Health and Physical Mobility

Engy Ziedan
,
Tulane University

Abstract

One of the most prominent responses to the opioid crisis in the U.S. has been the creation and enhancement of prescription drug monitoring programs (PDMPs) and the regulation of pain management clinics known as pill mill laws. Evidence suggests that these policies have been mostly effective in reducing opioid prescription rates, nationally, among Medicare and Medicaid enrollees. Given the national emphasis to curtail opioid prescriptions and the variation in chronic illness requiring pain relief by age and race/ethnicity, it is important to understand the marginal productivity of opioids on health and physical functioning, overall and across demographic groups. While past literature has documented variation in opioid medical use and misuse by age and race, less in known on the marginal productivity of prescription opioids themselves on health and physical functioning. In this article, I use data from the United States Renal Data System (USRDS) on patients diagnosed with End Stage Renal Disease (ESRD) to study the effect of prescription opioids on health and physical functioning of a highly comorbid US population. I exploit variation in the timing of the adoption of PDMPs and pill mill laws to conduct a difference-in-differences analysis of the effect of these policies on prescription opioid use and measures of health (eg: anxiety, weight gain, and cholesterol levels) and measures of physical mobility. I stratify all my analyses by age and race/ethnicity.
Discussant(s)
Patrick Button
,
Tulane University
Kosali Simon
,
Indiana University
Damon Jones
,
University of Chicago
Joanna Lahey
,
Texas A&M University
Ethan Lieber
,
University of Notre Dame
JEL Classifications
  • J1 - Demographic Economics
  • J2 - Demand and Supply of Labor