By Quoctrung Bui and Emily Badger | August 5, 2020

The economic damage from the coronavirus is most visible in areas like Midtown Manhattan, where lunch spots have closed, businesses have gone dark and once-crowded sidewalks have emptied.

But some of the worst economic pain lies in other neighborhoods, in the places where workers who’ve endured the broadest job losses live. In corners of the Bronx, South Los Angeles or the South Side of Chicago, unemployment is concentrated to a breathtaking degree. And that means that other problems still to come — a wave of evictions, deepening poverty, more childhood hunger — will be geographically concentrated, too.

Data estimating neighborhood-level unemployment rates suggests that as many as one in three workers in these areas are jobless, deeply widening economic disparities within cities.

In New York City, it’s as if parts of the Bronx were experiencing the Great Depression while the Upper East Side faced only modest drops in employment, according to Yair Ghitza and Mark Steitz, analysts who have estimated unemployment at the census tract level based on national economic statistics over the last six months.

The federal government doesn’t report unemployment data down to the neighborhood level, so the two researchers modeled these fine-grained statistics in a way that makes themconsistent with state and national surveysThrough June, they found most neighborhoods in the Bronx had unemployment rates in excess of 20 percent, while most neighborhoods south of 95th Street in Manhattan had rates less than half that.

“What’s salient and visible right now is the businesses that are shuttered, and the office buildings that are empty,” said Ingrid Gould Ellen, a professor of urban policy and planning at N.Y.U. “What we’re not quite seeing at least the physical manifestations of yet is the really just stark decline in incomes in so many neighborhoods around the city, and in a lot of working-class neighborhoods.”

“We will see them,” she predicted, warning that concentrated distress in these neighborhoods could also have long-term consequences for the children growing up there.

Mr. Ghitza, the chief scientist at Catalist, a Democratic data firm, and Mr. Steitz, a principal at TSD Communications, have tried to solve a large multiplication problem in modeling neighborhood-level unemployment. Official government statistics estimate, for example, the share of residents in a given census tract who are women, the share who are African-American, and the share who work in food service. Using such data, Mr. Ghitza and Mr. Steitz created an educated guess of the number of Black female food-service workers in each tract, then matched those demographics to national monthly unemployment statistics on the occupations and demographic groups most severely affected in this downturn.

The approach makes it possible to gauge employment differences at a finer level of geography than what the government reports. But these estimates also come with much wider room for error than official statistics, and the researchers warn that the results should be viewed alongside other data as policymakers try to understand an economy in free fall.

The resulting maps capture the flip side of recent analyses of private-sector data showing where restaurants have cut hours or where stores have closed their doors. Those business closings have been clustered, too, often in downtown districts where office workers no longer come in, or in wealthy neighborhoods where residents have sharply reduced their spending (or where they have left town altogether).

These maps reflect, instead, where the workers who once staffed those restaurants, bars, hotels and offices commuted home at night:

The maps also highlight how the distinct nature of the coronavirus economic shock has divided cities into neighborhoods where most people can work from home and neighborhoods where most can’t. And because the latter group is disproportionately made up of Black and Hispanic workers, those lines also largely follow patterns of racial segregation, as in Chicago.

As of June, the Chicago metro area had an unemployment rate of 15.6 percent, according to the Bureau of Labor Statistics. But Mr. Ghitza and Mr. Steitz estimate that in some neighborhoods on the predominantly African-American South Side, the unemployment rate was more than double that. Wealthier neighborhoods on the North Side had unemployment rates of less than 10 percent.

In a recent analysis, Peter Ganong, an economics professor at the University of Chicago, found that workers in the lowest quintile of income have experienced three times as many job losses as workers in the highest quintile. But that’s just looking through the lens of income alone. He says layering race, age and gender could push the differences even further at the census tract level.

Jesse Rothstein, an economist who is part of a team that has been tracking the effects of the pandemic on the labor market, agrees that it’s possible for unemployment rates in some neighborhoods to barely budge while others soar across town.

“There aren’t that many food-service workers that live in Beverly Hills,” he said.

In Los Angeles, job losses appear to be most severe in South Los Angeles, in predominantly Hispanic parts of the city.

Until now, some of the worst pain of the recession has been eased in these neighborhoods by a major federal expansion of unemployment benefits, including weekly $600 supplemental payments to millions of workers. Research shows that this aid significantly lifted the spending of unemployed workers; the money from the government might well have circulated through businesses in their own neighborhoods, too.

But those jobless benefits expired a few days ago. Now, as Congress and the White House wrangle over whether and how to extend the aid, these maps offer one more insight: These are the neighborhoods where workers — and the businesses that depend on their spending — would most acutely suffer without more federal help.

Click here to explore the maps on the New York Times site >