One of the studies, by University of Virginia economist John McLaren, found that the racial discrepancy remained even after controlling for income or insurance rates. Instead, Mr. McLaren found the gap was due in part to the fact that black workers are more likely to get to work via public transit, including subways and buses.
About 10.4% of black commuters take public transit, versus 3.4% of white commuters, according to the Census. After controlling for the use of public transit, Mr. McLaren finds the racial disparity in Covid-19 deaths is less pronounced.
Both studies raise the possibility that other causes could contribute to the discrepancy in deaths, such as gaps in access to paid sick leave, residential segregation and discrimination in health services. They also cited the higher likelihood that African-Americans work in essential occupations, such as health care, which have required employees to stay on the job through the pandemic.
Black people are more than 3.5 times as likely to die of Covid-19 than white people, and Latino people are nearly twice as likely to die of the disease as white people, according to researchers at Yale University and the University of Pittsburgh. Both new studies look at black deaths from the disease as a share of the overall black population rather than as a share of those infected with the disease.
The other study, by Christopher Knittel and Bora Ozaltun, both of the Massachusetts Institute of Technology, found that a 10% increase in the share of a county’s residents who use public transit versus those who telecommute raised Covid-19 death rates by 1.21 per 1,000 people when looking at counties around the U.S.—or by 0.48 per 1,000 people when focusing only on counties within individual states. In their analysis, the researchers controlled for race, income, age, climate and other characteristics.
In part, that could be because there is something specific about public transit—close proximity to strangers on crowded railcars or buses on a daily basis for extended periods, for instance—that makes people particularly vulnerable to getting sick, wrote Mr. Knittel and Mr. Ozaltun.
Counties with higher shares of people who drove or walked to work versus telecommuting also saw higher death rates, a sign that perhaps some of the greater risk simply comes from leaving the house, they found.
But transit use isn’t the whole story, both papers say. Mr. Knittel and Mr. Ozaltun find there is still a racial disparity in death rates in counties across the country when controlling for transit use.
Both papers also note that New York City accounts for a very large share of the country’s transit ridership as well as its virus-related deaths. In both cases, the authors left out New York City and reached similar results.
The transportation data used in both cases measure people’s commuting patterns before the start of the epidemic. It is possible the epidemic altered those habits or made people more likely to stay home in ways that these studies don’t measure.
The results could complicate cities’ efforts to reopen their economies. With many urban workers and residents reliant on transit, the studies suggest that reviving business activity will depend on authorities’ ability to ensure that buses and trains are safe to use.
Early indications from transit systems abroad offer hopeful signs. In Paris, transit users must wear face masks and authorities require that every other seat on trains and subways be kept open. Since May 9, only 1% of the new disease clusters—defined as three cases of common origin recorded over seven days—have come from transit, according to the French health agency.
Already, New York’s subway system has stepped up its disinfecting of trains and buses, and launched an app that tells riders on some lines how crowded railcars are. The system is also considering checking riders’ temperatures and rolling out robot cleaners.