Such a collapse could have meant the loss of hundreds of thousands of restaurants. But no such extinction event, as one advocate described the pandemic’s potential impact, occurred. The number of closed establishments to date is a fraction of those early dire predictions, which were based largely on fears or a small-business survey with acknowledged flaws or just educated guesses.
Just how many restaurants have succumbed to the pandemic, with its mix of public health restrictions, government shutdowns and shared diner anxiety? That number is difficult to nail down.
The National Restaurant Association (NRA), the trade group that advocates for the industry, has an in-house economist as well as a senior vice president of research to provide reports, data and insights to members and the public at large. As the NRA lobbied Congress for relief to the battered industry, the group tried to calculate the damage already done, whether the restaurateurs were members of the association or not.
By the spring of 2021, the association had settled on a number: 90,000 closed restaurants, both temporary and permanent, which includes diners, cafes, chain outlets, taverns, bars and neighborhood restaurants. That number has been repeated many times, by media outlets and politicians, in the months since the NRA released it. That number has not been updated in more than a year, in large part because the NRA believes that, more than two years into the public health crisis, current restaurant closures may not be a result solely of the pandemic.
Is there a better way to determine — or even estimate — the number of bars and restaurants that have closed during the pandemic? Finding out means first examining the NRA’s own estimate — and how it arrived at it.
The NRA tackled the enormous task of tallying the damage by sending out email surveys to hundreds of thousands of recipients. The group then extrapolated the results — the percentage of closed restaurants — against the total number of “food services and drinking places” in the United States, as determined by the Bureau of Labor Statistics (BLS). That number, right before the pandemic, was about 660,000 establishments.
For example, in December 2020, the association estimated that nearly 17 percent of food and drink establishments had shut their doors, based on survey responses from 6,000 restaurant operators and 250 supply chain companies. The NRA then used that percentage rate to calculate that about 110,000 establishments had closed, at least temporarily.
Several months later, in April 2021, the NRA lowered its estimate to 90,000 because “many of those temporarily closed restaurants reopened,” the association mentioned in a data analysis sent to The Post. It was based on survey responses from 2,500 restaurant operators.
One thing to keep in mind about the “food services and drinking places” number, as calculated by the BLS, is that it includes many businesses that are not, strictly speaking, restaurants. The number also includes airline food-service contractors, concessionaires at sports facilities, caterers, street vendors, ice cream and food trucks, coffee shops, doughnut shops, cocktail lounges and nightclubs.
The flaws with the NRA’s methodology, as least the parts shared with The Post, are not unique to this trade association and its email surveys, say economists and analysts. The response rate was small. The respondents were self-selected, not randomly sampled. The email lists may not have reflected the makeup of the industry at large.
For its part, the NRA determined early in the crisis that it would survey more than members. For its first pandemic-related survey, in March 2020, the NRA “went with a mailing list that we had, but in very, very, very short order our mailing list grew to be restaurant operators that were not members because we weren’t limiting our pandemic information to just members,” says Vanessa Sink, director of media relations for the association.
“We were sending it to anybody that wanted to be on the mailing list,” she adds. “We have tried to get representation from all parts of the industry in those surveys, so that we know how they’re being impacted.”
The NRA’s closure numbers, of course, were also in service to a much larger goal: to secure the federal assistance to an industry that was in free fall. The estimates were not necessarily meant to be airtight.
Another way to determine the number of shuttered restaurants over any given time is to review Business Employment Dynamics (BED) statistics, which are compiled from the BLS’s quarterly employment and wage census. These statistics include a quarterly tracking of business closures under the broad “food services and drinking places” category.
Katharine Abraham, an economist with the University of Maryland, calls the BED numbers “very reliable” even though the NRA cautions they include temporary or seasonal closures, as well as permanent ones.
In 2020, according to the BED data, the total number of closures was 159,000. In the second quarter of 2020 alone, as government restrictions kicked into high gear, BED statistics indicate that more than 72,500 food and drink establishments closed, more than three times the quarterly average of about 20,300.
But those numbers don’t account for the normal, annual churn of closures. From 2011 to 2019, according to BED data, an average of more than 81,000 food and drink establishments closed every year. If you subtract that average from 159,000, you get nearly 78,000 additional closures in 2020.
The 78,000 figure, though, is still based on the larger category that includes establishments other than restaurants and bars. The BED stats are not broken down into smaller subcategories, but based on numbers crunched by The Post, restaurants and bars make up about 94 percent of the larger BLS group. That means in 2020, about 72,700 more restaurants and bars than normal closed, apparently due to the pandemic, a 95 percent jump over the average annual rate.
This is a clear sign of the pandemic’s devastating effect, representing hundreds of thousands of lost jobs that, in turn, impacted countless lives and communities.
Following a tumultuous 2020, the closure rates have stabilized over the first three quarters of 2021, according to the latest BED statistics available: about 19,500 per quarter, lower than the pre-pandemic average of 20,300.
The numbers would seem to indicate the industry has weathered the worst of the pandemic. They might even reflect a hopeful trend: that more bars and restaurants have reopened, given that the BED data include temporary closures. If you add up the BED closures so far during the pandemic, subtract the average shutdown rate and adjust the figures for just bars and restaurants, you get a total of about 70,300 establishments closed to date.
The NRA has not conducted a more recent survey to adjust its 90,000 number, in large part because the pandemic is just one factor — along with food costs, fuel costs, staff shortages and more — that may figure into a restaurant’s shuttering at this point, says Sink.
“There was a period of time where closures were because of the pandemic,” says Sink. “I don’t know that all closures could now be said that they’re because of the pandemic, unless we’re calling all of what’s happening right now the long tail of the pandemic.”
Restaurant advocates say the lower closure rates for 2021 don’t reflect what they hear from operators and probably mask a larger, looming problem: that thousands of operators, mostly independents, remain deep in debt; that they added to their debt load after assurances from the federal government they would receive relief through the Restaurant Revitalization Fund. After one round of RRF grants, Congress couldn’t muster the votes in May to replenish the fund.
“Just like we hear that the stock market is doing great and then we go to Main Street and it’s not so great, I feel the same way about the BLS numbers,” said Erika Polmar, executive director of the Independent Restaurant Coalition. “Maybe things look great on paper and maybe it looks like we’re thriving, but when you get out there and you talk to the community, it’s not good.”
With the replenishment of the RRF all but dead, the coalition is predicting that more than half of those 177,000 restaurants left out in the cold from the initial round will shut their doors permanently. If true, that would more than double the existing number of closures during the pandemic, though a certain percentage of them might have shut down even without the health crisis.
Colicchio, who predicted a 75 percent death rate at the start of the pandemic, says government assistance is the main reason there hasn’t been more widespread damage. Without money from the RRF, the Paycheck Protection Program and other programs, he says, the closure rate may not have been 75 percent, “but I think a lot more than we saw.”
Colicchio has lost one of his eight restaurants during the pandemic, and he chalks that up to a bad landlord. But like other independent restaurateurs, Colicchio has debt. He says he owes “close to a million dollars” to one landlord alone.
“They’re not pushing me, but sooner or later, they’re going to start,” Colicchio says. “We’re not out of the woods yet.”