Things did not appear to go well for retail landlords in 2017. Big chains announced thousands of store closings and filed for Chapter 11 bankruptcy protection at a record rate, leading analysts to predict that hundreds of malls were on the verge of closing. The forecasts for the future proved even more grim than the present.

But that’s not how mall owners see their own prospects. Mall managers tend to bemoan what they see as sensationalized coverage and doomish predictions designed to drive traffic to news websites — and rent reductions for their retail tenants.

How bad is it really? Figuring that out isn’t as easy as you would think.

There were about 1,070,000 retail stores in the U.S. in 2015, according to census data that includes the types of stores you find in shopping malls as well as big-box stores, gas stations, and other retailers. That’s up slightly from a recession-era low of 1,063,000 in 2011. The problem is that the data lag, so it leaves out any carnage from the last two years.

Statistical modelers can use census data as a starting point. In one example, market-research provider Euromonitor used a combination of census data and company filings to estimate that the number of U.S. stores decreased by just 0.1 percent in 2017 — hardly an apocalyptic level of die-off.

There’s another way to measure recent retail closings: Ask Yelp.

Since the launch of the local review website in 2004, Yelp has compiled millions of U.S. business listings. That makes its data a useful complement to government statistics, according to Harvard economist Edward Glaeser. “Changes in the number of businesses and restaurants reviewed on Yelp can predict changes in the number of overall establishments and restaurants,” he wrote with co-authors Hyunjin Kim and Michael Luca in a Harvard Business School working paper published in October.

For some industries, including retail, leisure and hospitality, the researchers found that Yelp listings are better at predicting economic change than statistical methods of extrapolating census data. Yelp tends to be more useful in denser, more affluent and more educated areas.

So what does Yelp’s data say about the retail apocalypse? Stores closed at a faster pace than they opened in 2017, the only time that happened in the four years of data that the company provided to Bloomberg. For mall owners and others in retail, that has to sting worse than a one-star review.


Clark writes for Bloomberg.