Job Market Paper

    Does High-Speed Internet (Broadband) Technology Affect the Mental Health of Older Adults? (Presentations- MEA ’23, American University-’23, SEA ’23, APPAM ’23) (Popular Media Highlight: GROW Magazine Cover Story)

    Abstract: Recent research has shown the negative effects of social networks on younger people’s mental health. Yet, we do not know whether the same effects are present across the population, particularly for older adults, who are especially vulnerable to certain mental health conditions. I estimate the effect of broadband availability on the mental health of older adults (aged 50+) in the United States, using individual panel data and exploiting the census tract-level staggered rollout of broadband from 2010 to 2018 in the latest difference-in-differences (DID) framework. In contrast to the literature that finds harmful effects of the internet among younger populations, my results show that broadband rollout significantly reduces depression symptoms by 5.7% among older adults. The primary mechanisms driving these positive effects include an increase in self-use of the internet and virtual social connectedness and a decline in social isolation and loneliness. Improved health literacy and technological efficiency (telehealth) also partly drive the results but do not affect self-reported physical health or employment. Rural dwellers and women are the biggest beneficiaries of broadband’s positive effects on mental health. I also find some evidence of the spatial spillover effects. These positive estimates are comparable with other major life events known to negatively affect the mental health of older adults, such as job loss, recession, and the death of a spouse. This work highlights broadband’s unmeasured additional benefits to public investments, given the recent allocation of over $65 billion to broadband expansion. A back-of-the-envelope calculation suggests that broadband expansion may reduce the cost of excess Medicare spending by about $5 billion due to major depressive symptoms and social isolation among older adults.

Publications

  1. Early-Life Economic Conditions and Old-Age Mortality: Evidence from Historical County-Level Bank Deposit Data with Valentina Duque, Jason Fletcher, Hamid Noghanibehambari, Lauren Schmitz (Journal of Population Economics, January 2024)
    Abstract:This paper studies the long-run mortality effects of in-utero and early-life economic conditions. We examine how local economic conditions experienced in the Great Depression, proxied by county level banking deposits during in-utero and first years of life, can influence old-age longevity. We find that a one-standard-deviation rise in per capita bank deposits is associated with an approximately 1.7 months increase in longevity at old age. The effects are robust across a wide array of specification checks. Additional analyses comparing state-level versus county-level economic measures provide insight on the importance of controlling for local-level confounders and exploiting more granular measures in exploring the relationship between early-life conditions and later-life mortality.

  2. Discrimination in Science: Wage Gap between US and Foreign-Born Land Grant Scientists with Jeremy Foltz (R&R at American Journal of Agricultural Economics (AJAE))
    Abstract: The dominance of the US innovation and academic system relies heavily on foreign-born labor for its success. Recent literature has shown evidence of wage gaps in academia based on gender and race; however, little is known about whether a wage gap might exist for foreignborn faculty. This paper studies the wage gap between the US and foreign-born agricultural and life science faculty at 52 US Land Grant Universities (LGU) using a survey of over 1,400 scientists conducted in 2005 and 2015. We develop a framework to categorize the sources of a potential wage gap into testable categories that capture direct discrimination as well as indirect (systemic) discrimination. We find that among the tenure-track faculty, foreign-born earn about 4% or $5,200 lower annual wages even though, on average, foreign-born scientists work more hours per week and produce about 52% more journal articles than US-born scientists. The estimated wage gap is robust to a range of alternative empirical specifications. The decomposition analysis suggests that about one-third of the wage gap is due to direct discrimination, and about two-thirds is due to various types of systemic discrimination. Using our framework, we then rule in and rule out some important types of systemic discrimination. Estimates from this paper are crucial for understanding potential policies that could improve diversity, equity, and inclusion in US academia.

    3. Working Papers

    1. The Effect of Early Life Exposures to the Green Revolution on later life Aging Outcomes: Evidence from the Largest Aging Data with Lauren Schmitz, Valentina Duque, Jinkook Lee, Hunter Green, (presented at APPAM)
      Abstract: The Green Revolution (GR) is arguably the most significant shock to agricultural productivity gains that increased agriculture yields, decreased food prices, increased calorie intake, and reduced infant mortality and poverty. However, its long-term im- pacts on health are not well understood. This paper contributes to the literature in two different ways. First, we examine how early life exposure to the GR affects later life cognitive health using the largest aging data and generalized difference-in-differences (DID) estimation. Secondly, we contribute by exploring the potential pathways through which GR can affect later-life cognitive function. We find a heterogeneous treatment effect showing that exposure to the GR around the birth years significantly improved the later life cognitive function among the socially disadvantaged groups and people born in rural areas. Specifically, we find that one standard deviation increase in GR during early life improved these groups’ later-life cognitive function between 0.08 to 0.13 standard deviation. The significant improvement in schooling among these groups explains some of the positive effects on general cognitive health. We rule out that the school construction was driving any positive gain of schooling. Estimates from this paper are crucial for the policy perspective since GR is an ongoing policy yet to be adopted in several parts of other developing countries that will see a higher share of the aging population in coming years than the rest of the world.

    2. Does Broadband Technology Affect Social Security Applications?
      (Competitive grant from Retirement and Disability Research Center (RDRC) Center for Financial Security (CFS), Social Security Administration (SSA))
      Abstract: One of the key focal reseach area of SSA is to understand the structural barriers for the Social Security Disability Insurance (SSDI) in the US. It is, however, unclear whether high-speed interent (braodband) technologies affect an individual's likelihood of receiving Social Security Disability Insurance (SSDI). This paper evaluates whether availability of broadband affects the enrollment in SSDI among the older adults. I leverage the quasi-experimental staggered rollout of high-speed broadband, combined with restricted individual panel data of the Health and Retirement Study (HRS), and exploit spatial, temporal, and individual variations in broadband availability. Employing the advanced difference-in-differences (DID) estimator to capture the dynamic treatment effect, I find an increase in the probability of receiving SSDI following the introduction of high-speed broadband. Heterogeneity analysis suggests benefits for rural areas. The estimates from this paper are crucial to prioritize the resources for the low broadband regions and households without access to sufficient broadband connections.

    3. Spillover Effects of a Bicycle Policy on Girls’ Enrollment in Middle Schools
      (Recognition under Best Paper Scholarship (Honorable mention), Dept. of Economics, UW-Madison, 2018)
      Abstract: Distance to school and safety are two major challenges for the educational gender gap in developing countries. Numerous programs were introduced to address this concern in different parts of the world; however, less attention has been paid to understanding the spillover effects of those policies. This study examines the spillover effect of a grade-9 girls’ bicycle program on grade-6 girls’ enrollment in Bihar, India. I use the universe of administrative school-level data and employ a difference-in-differences approach. I find a statistically significant increase by 7.4 percent in the 6th-grade girls’ enrollment after the bicycle policy was introduced. Further, using a triple difference approach, I find the increase in enrollment is more significant in the schools that have girls’ toilets than the schools that do not have girls’ toilets. I also find that the increase in enrollment is lower for the girls who belong to socially disadvantaged groups. Using quantile regression, I show that the policy’s effect varies across the school’s size, increasing enrollment for the more prominent schools. Overall, this study sheds light on the underexplored impact of the bicycle policy on girls’ schooling preferences, emphasizing the potential benefits of combining bicycle programs with school infrastructure policies to narrow the educational gender gap.

    4. The Effect of Teacher Hiring Policy on Student Test Scores: A Multiple Regression Discontinuity Design Approach for one of the Largest Public Schooling in the World
      (selected at American Education Finance and Policy conference)
      Abstract:The economic literature on the impact of class size on student achievement has mixed evidence. However, the majority of this research has been conducted in high- income countries or small-scale settings. This study contributes to the literature by ex- amining the impact of a minimum pupil-teacher ratio (PTR) policy on student achieve- ment in rural India, where multi-grade teaching is common, with one or two teachers instructing multiple grades in one or two classrooms. The PTR policy created several thresholds for the total number of primary school children, requiring the hiring of an additional teacher beyond each of these thresholds. Using a multiple regression dis- continuity design (MRDD), this study finds that the PTR policy led to some increase in the number of teachers in primary schools at a major threshold. This increase in teachers was associated with a small but significant improvement in student test scores in Mathematics and Reading. These findings suggest that reducing class size through a PTR policy can be a cost-effective way to improve student achievement in rural India, a context where educational resources are often scarce.

    Selected Work In Progress

    1. Effect of Green Revolution on the Intergenerational Human Capital
      with Valentina Duque and Lauren Schmitz
    2. Spatially Varying Gender and Caste Inequalities in Cognition and the Contribution of Individual and Statelevel Characteristics Across India
      with Lauren Schmitz, Kanika Arora, and Shubhashrita Basu