Executive Summary

• Employer-sponsored insurance is the single largest source of health insurance in the United States, raising the specter that large-scale job losses will lead to a wave of uninsured Americans.
• This study combines information from the most recent Current Population Survey and the April Employment Report to estimate the source of insurance for the 20.5 million people who lost their jobs in April 2020.
• The main finding is that 51 percent (10.4 million) of the 20.5 million had employer-sponsored insurance; if these job losses persist, this segment is the greatest potential contributor to the newly uninsured.
• The remaining sources of insurance for the newly unemployed are none (i.e., they are uninsured, at 19 percent), individual-market policies (11 percent), and Medicaid (10 percent).

Introduction

Employer-sponsored insurance is the single largest source of health insurance in the United States. As a result, the gargantuan pandemic-related job losses in April 2020 – 20.5 million jobs – raise the concern that millions of individuals will lose their insurance. Although roughly 18 million of those job losses were characterized as temporary layoffs, the possibility that they may persist makes it useful to investigate the health insurance sources of those who have lost their job recently.

This information is also useful to policymakers as they seek to anticipate the needs of the uninsured over the remainder of 2020 and beyond. If the newly unemployed have predominantly employer insurance, it may be useful, for example, to focus efforts on subsidizing Consolidated Omnibus Budget Reconciliation Act of 1985 (COBRA) coverage. This focus would preserve not only the insurance arrangements, but also the care networks beneath that coverage. Alternatively, if those who lose jobs are primarily insured in the individual market or already uninsured, it may be useful instead to focus on steps to enroll individuals in individual-market policies.

Analysis and Results

This study combines information from the 2018 Current Population Survey (CPS) and the April National Employment report to shed light on this issue. Specifically, it uses the CPS to compute the fraction of workers covered by employer insurance, individual insurance, Medicaid, Medicare, and other insurance by industry (the residual workers are classified as uninsured). These computations are shown in Table 1. As seen in the first row, for example, among nonfarm workers in 2018, 51 percent had employer insurance, 11 percent were covered by individual policies, 10 percent Medicaid, 6 percent Medicare, and 4 percent other insurance. 19 percent of the workers were uninsured.

The remaining rows, however, show considerable heterogeneity in insurance arrangements. In particular, employer insurance is lowest – and uninsurance the highest – in leisure and hospitality, the sector hit hardest by the fallout of the COVID-19 pandemic. Thus, in general, the pattern of job losses across the economy will matter greatly for the ultimate rise in the number of uninsured Americans.

This study investigates the implications of the April job losses in Table 2. Specifically, it applies the information on the distribution of insurance sources in Table 1 to the job losses experienced by industry in the employment report. There are three reasons to be cautious in interpreting the results. First, the CPS data are from 2018; while the most recent data, there may have been some changes in the intervening year. Second, these individuals have not yet necessarily become uninsured. Their employers may be continuing their coverage, individual-market policies will be portable between employment and unemployment, and so forth. Third, we are assuming that the laid-off individuals have the same composition as employment in the industry as a whole. In the absence of any direct information about insurance and job losses in April, this assumption is a necessity.

What does Table 2 show? The main finding is that 10.4 million of the 20.5 had employer insurance; if these job losses persist, this group will be the greatest source of the potentially uninsured. The second largest category of job losses are among the already uninsured (3.9 million). While being uninsured in a pandemic is not particularly desirable, this situation is not a new public policy problem. The next highest category is losses among those with individual-market insurance, which does not present a particular problem with being uninsured as long as economic resources permit continuing to pay premiums. Finally, participants in the public programs Medicaid (2.0 million) and Medicare (1.1 million) are a modest part of the insurance picture among those who lost their jobs.

Conclusions

These computations are suggestive at best, and a firmer picture of the evolution of those uninsured in the United States will await further data. The results do indicate, however, that it would be most fruitful to anticipate the insurance needs of those individuals who previously had been covered by their employer.