This conference is supported by Grant #P30AG012810 from the National Institute on Aging and by Grant #P01AG005842 from the National Institute on Aging
It is now established that mortality and excess mortality from COVID-19 differed across racial and ethnic groups in 2020. Less is known about educational differences in mortality during the pandemic. Case and Deaton examine mortality rates by BA status within sex, age, and race/ethnic groups comparing 2020 with 2019. Mortality rates have increasingly differed by BA status in the US in recent years and there are good reasons to expect the gap to have widened further during the pandemic. Using publicly available provisional data from the National Center for Health Statistics Case and Deaton find that mortality rates increased in 2020 over 2019 for those with and without a BA, irrespective of age, sex, or race/ethnicity. Although mortality rates increased by more for those without a BA, the ratio of mortality rates for those with and without a BA changed surprisingly little from 2019 to 2020. The BA was protective against mortality prior to the pandemic, and it was equally protective during the pandemic. Among 60 groups (sex by race/ethnicity by age) that are available in the data, the ratio of mortality rates of those without a BA to those with a BA fell for more than half of the groups. Their results suggest that differences in the risk of infection were less important in structuring mortality by education than differences in the risk of death conditional on infection.
This paper was distributed as Working Paper 29328, where an updated version may be available.
The positive epidemiological externality associated with vaccines provides a rationale for subsidies. Goodkin-Gold, Kremer, Snyder, and Williams study how optimal subsidies vary with disease characteristics by integrating a standard epidemiological model into a vaccine market with rational economic agents. The researchers focus on a vaccine campaign to quell an epidemic like Covid-19 in the short run, extending the analysis to the longrun steady-state market for a vaccine against an endemic disease like measles. Across both models and across market structures ranging from competition to monopoly, Goodkin-Gold, Kremer, Snyder, and Williams find that the infection rate, marginal externality, and optimal subsidy are nonmonotonic in transmissibility, peaking for diseases that spread quickly but not so quickly as to drive all consumers to become vaccinated. The researchers study when universal vaccination emerges in equilibrium, when vaccination exhibits increasing social returns (providing an argument for concentrating a capacity-constrained campaign in few regions), and when producers' bias away from vaccines toward treatments generates deadweight loss. The general propositions are complemented by calibrations to relevant diseases.
How Did the COVID-19 Health Care Delivery Disruption Affect Medication Use among People with Chronic Conditions?
In June 2021, 22 states ended all supplemental pandemic unemployment insurance (UI) benefits, eliminating benefits entirely for over 2 million workers and reducing benefits by $300per week for over 1 million workers. Using anonymous bank transaction data and a difference-in-differences research design, Dube, Naidu, Kluender, and Stepner measure the effect of withdrawing pandemic UI on the financial and employment trajectories of unemployed workers in states that withdrew benefits, compared to workers with the same unemployment duration in states that retained these benefits. In their data through August 6, Dube, Naidu, Kluender, and Stepner find that ending pandemic UI increased employment by 4.4 percentage points while reducing UI recipiency by 35 percentage points among workers who were unemployed and receiving UI at the end of April 2021. Through the first week of August, average UI benefits for these workers fell by $278 per week and earnings rose by $14 per week, offsetting only 5% of the loss in income. Spending fell by $145 per week, as the loss of benefits led to a large immediate decline in consumption.
This paper develops a novel revealed-preference approach that uses discontinuities in budget constraints to value workplace safety. Koenig and Anelli measure weekly labor supply decisions among hourly workers who repeatedly face the decision of how many hours to work at varying levels of Covid-19 risk and leverage 21 state-specific discontinuities in unemployment insurance eligibility criteria to identify the labor supply behavior. Results show large baseline excess and missing mass at the eligibility threshold and increasing responses at higher health risks. The observed behaviour implies that workers are willing to accept 34% lower incomes to reduce the fatality rate by one standard deviation, or 1% of income for a one in a million chance of dying. Most of this cost is not priced into wages, and canonical methods to measure the value of workplace safety miss such costs.
Miller, Segal, and Spencer empirically investigate the impact of COVID-19 shutdowns on domestic violence using incident-level data on both domestic-related calls for service and crime reports on domestic violence assault from the 18 major US police departments for which both types of records are available. Although the researchers confirm prior reports of an increase in domestic calls for service at the start of the pandemic, Miller, Segal, and Spencer find that the increase preceded mandatory shutdowns, and there was an incremental decline following the government imposition of restrictions. The researchers find no evidence that domestic violence crimes increased. Rather, domestic violence assaults declined significantly during the initial shutdown period and there was no significant change in intimate partner homicides in these months. Their results fail to support claims that shutdowns increased domestic violence and suggest caution before drawing inference or basing policy on calls data alone.
This paper was distributed as Working Paper 29429, where an updated version may be available.
Reber, Kosar, and Fashaw assessed differences in nursing home utilization and quality by race and their relation to the large racial disparities in COVID-19 mortality among older adults in 2020 and early 2021. The age-adjusted pre-pandemic rate of nursing home use was 66% higher among Black people and 35% lower among Hispanic people relative to White people 55 and older. Nursing home use was a modest contributor to racial disparities in COVID-19 mortality for Black people only, accounting for about 10% of excess deaths. Reber, Kosar, and Fashaw did not find evidence that excess deaths among Black and Hispanic people were related to lower baseline nursing home quality for these populations. Compared to racial disparities in COVID-19 mortality in the overall 55 and over population, differences in COVID-19 mortality by race were notably smaller among nursing home residents. Overall, risks unrelated to the nursing home setting were the predominant driver of excess deaths for Black people and particularly for Hispanic people during 2020 and early 2021, while White people appear to have been far more protected from COVID-19 outside of nursing homes.
Given its wide-ranging impacts on health and economic resources, the COVID-19 pandemic has likely affected children in a myriad of ways. One major dimension along which children’s lives changed was the shift in access to school environments, which may have had a deleterious effect because schools provide a range of mental, social, and physical health supports. Fitzpatrick and Sadowski examine schools’ effect on child health by measuring the causal effects of school access during the fall of 2020 on child health care usage in novel electronic health records (EHR) data. The researchers examine how exposure to in-person school environments affected health outcomes of children, with a particular focus on mental and developmental health outcomes (e.g., depression and ADHD-related care). Fitzpatrick and Sadowski pair the EHR data with newly collected district-level data on the opening and operational modes (in-person or online) of school districts across the country. They examine disparities in effects across children from different socioeconomic background and across communities with different geography, density, and economic conditions. Results show that, although the pandemic itself led to increases in depression- and anxiety-related health care visits, access to in-person school environments decreased the use of services for depression and anxiety. On the other hand, treatment for ADHD fell during the initial phases of the pandemic when schools were closed, and increased when children were exposed to school environments. Fitzpatrick and Sadowski discuss the different mechanisms that may drive these results. These results contribute to understanding the myriad of ways that school closure policies affected child wellbeing during this pandemic and to better preparation of responses to future public health emergencies.