The SIEPR senior fellow founded the field of personnel economics.
But they rarely looked at the two combined.
So SIEPR Faculty Fellow Maria Polyakova and her colleagues set out to examine how the pandemic impacted individual livelihoods depending on where people live, as well as the age of coronavirus victims.
“Initial economic damages from the pandemic are more widespread across groups than deaths, which were primarily concentrated in a few states and among the oldest,” the researchers write in their study published in the Proceedings of the National Academy of Sciences.
The researchers — who included Victoria Udalova, a former Government Visitor at SIEPR — used all-cause mortality reports from the CDC and employment data from the Current Population Survey of the U.S. Census Bureau. The studied data from January 2011 through April 2020, calculating the results nationally and separately by state and by age group.
They found all-cause mortality was 2.4 per 10,000 individuals nationally — about 30 percent higher than all reported COVID deaths in April. The employment loss was similarly staggering, at 9.9 fewer employed per 100 individuals in the population.
The economist and assistant professor at Stanford Medicine determined initial economic damages from the early days of the pandemic in April 2020 were more widespread across geographic areas than the number of deaths, which were primarily concentrated in a few states and among the oldest of the elderly.
The two states with the largest mortality increase — New York and New Jersey — accounted for about half of the national excess mortality. The two most economically impacted states — Nevada and Michigan —accounted for only 7 percent of the national economic damages.
The findings suggest that policy responses to contain the pandemic likely involve hard tradeoffs across different demographic and geographic groups, as well as between lives lost and economic damages. Not surprisingly, the debates around these tradeoffs have been increasingly dominating the national policy discussion.
In this Q&A, Polyakova — who is a core faculty member at Stanford Health Policy — discusses research and what the implications are for clinicians and policymakers moving forward.
Your key premise is that the initial economic damage from the pandemic is more widespread across ages and geographies than the initial mortality impacts. Why did you feel it was important to report your findings by state and age?
We were intrigued that a lot of the initial research and media reports either focused on mortality impacts or the economic damage of the pandemic — but were not combining the two, even though the possible tradeoff was clearly on everyone's minds.
Looking at the differences by age was a natural starting point as most clinical reporting on the pandemic suggested the virus was disproportionately affecting older individuals. And then we got curious about what was happening by states, because everyone in April was focused on New York and we figured the economic ramifications must also be happening elsewhere.
The evidence suggests the economic impact of COVID19 is extremely diverse geographically and demographically.
Every single state experienced economic damage — with the median unemployment loss across states at 7.8 percentage points. The states with the largest employment displacement over 16 percentage points — Nevada and Michigan— experienced a 30 percent decrease in employment, yet they only accounted for 7 percent of the national employment loss.
The health damages early on, however, were highly geographically concentrated, with New York and New Jersey together accounting for 49 percent of the national change in all-cause mortality in April. They were stark outliers, experiencing more than 10 excess deaths (from any cause) per 10,000 individuals, compared to the median excess all-cause mortality across states of 0.64 deaths per 10,000 individuals.
And putting those into relative terms is even more eye-popping: The increase in all-cause mortality was 207 percent in New York and 179 percent in New Jersey. But the economic damage is far more widespread than the excess mortality impacts.
How can your study help local public health officials as well as state and federal policymakers draw conclusions about their responses to the pandemic?
I think the findings starkly illustrate the enormously difficult challenge we are likely facing as a society and underscores why it may be hard to gain uniform societal support behind any one given policy. The puzzle of how to save the lives of the elderly and keep the jobs of the young — or save loved ones in one location but affect jobs in another — is likely not the puzzle politicians were hoping to have to solve during their time in office. Finding the optimal policy is challenging, since most decisions end up hurting someone in some way.
This becomes even more complex if we consider the following nuance. We do not actually know for sure how much policymakers’ actions change things in practice, both in terms of infection rates and in terms of the economy.
Why did you focus on all-cause mortality rather than just the COVID deaths?
We believe changes in all-cause mortality is the most accurate reflection of the overall impact of the pandemic on health, at least at this point. First, with all-cause mortality you don't have to worry that some deaths were just not marked as a COVID death in the heat of the moment.
Second, all-cause mortality captures both direct and indirect mortality effects from the pandemic. Indirect effects could be negative — such as increases in mortality attributable to reduced utilization of routine and emergency healthcare services — or positive, such as declines in mortality due to the reduction of motor vehicle use. The indirect effects could also be ambiguous, like the ongoing debate over the mortality effects of an economic downturn.
Which geographic groups were hit hardest and why?
As is not surprising for April, we found that excess mortality was extremely concentrated in New York and New Jersey. The economic effects in April were felt the most in Nevada, Indiana, Massachusetts, Michigan, and Hawaii. We have done some analysis that suggests that industry composition of states — in particular exposure to tourism or services — may have played some role. But there is no single, clear-cut answer yet.
Did your research lead you to uncover any surprises?
While we expected to find the age patterns that we report, the geographic dispersion of the economic effects across states was less clear to us at the outset. It was interesting to uncover that the worst economic impacts were not felt in those states hit hardest by coronavirus deaths.
We were also not exactly sure where the national all-cause excess mortality number would fall, as there was a lot of reporting about declines in traffic fatalities, for example, so one could have also thought that the overall death rate in the country, or at least in some states, declined.
The data runs through April. Do you think the trend has held as the pandemic has progressed?
As the virus spread across states, the mortality map definitely changed. We are hoping to continue tracking how this develops once reliable, newer data becomes available.