This post was originally published on this site
Illustration by Kevin Whipple
Add economic data to the list of things that won’t ever be the same after the coronavirus pandemic.
An unprecedented “sudden stop” by the global economy in the first quarter as activity ground to a near halt amid lockdowns put in place to control the spread of the virus left economic policy makers, businesses and investors virtually flying blind. Traditional economic data, including more timely series like weekly unemployment claims, offered little near-term insight into what was happening on the ground.
That’s put a spotlight on the already rapidly growing world of “alternative data.” But one category known as mobility data, particularly anonymized foot-traffic information gleaned from millions of mobile phones as well as other broader measures of individual activity, is being propelled into the mainstream and now appears likely to be among the post-pandemic economy’s most closely watched economic indicators.
“Obviously, if the economy moves from 1.2% growth to 1.3% growth, you may not be able to see that in the mobility data. But if it moves from +2% to minus 30%” it’s a different story, said Jens Nordvig, founder of Exante Data, a data-focused economics research firm. “It was incredibly useful in the middle of the crisis to literally have a real-time pulse on the economy country by country, state by state.”
Exante Data was an early adopter, using mobility-related data from a variety of sources, including Chinese social-networking company Baidu, to get a grip on the pandemic’s effect on the economy in China as it fell victim to the virus. Data from the firm was used in a widely cited Imperial College London study that tested how reopening efforts affected virus transmission rates
Investors also had reason to track the spread of the virus itself — and for that, mobility data was hard to beat, Nordvig said.
“What kind of economic data would you use to analyze how the virus gets transmitted? There’s no economic data that’s particularly good for that, but the mobile phone data is just fantastic,” said Nordvig.
“Mobility data is just literally tailored to analyze a virus,” he said.
Top-down look
Mobility data, when analyzed carefully and used alongside other data sources, is particularly well suited to measuring activity when new restrictions are imposed or relaxed, as well as around events, including demonstrations and civil unrest.
Karel Mertens is a senior economic policy adviser at the Federal Reserve Bank of Dallas. Part of a team responsible for measuring economic activity and making forecasts, Mertens and his colleagues were left scrambling as the pandemic hit.
“All the usual models and the economic statistics that we would look at were essentially almost entirely useless because they’re just way too delayed,” he said in an interview with MarketWatch.
As part of a broader effort to get a grip on what was happening on the ground, the economists turned to data provided by SafeGraph, a company that tracks raw foot-traffic data it harvests from mobile-phone apps. That data was in turn used to formulate the Dallas Fed’s popular Mobility and Engagement Index, which measures the deviation from normal mobility patterns as a result of the pandemic.
“ “It was incredibly useful in the middle of the crisis to literally have a real-time pulse on the economy country by country, state by state.” ”
Minutes of the Federal Reserve’s July 28-29 meeting showed that policy makers pointed to high-frequency indicators, including data on credit and debit card transactions and mobility indicators based on mobile phone location tracking, that suggested consumer spending growth had slowed in reaction to further spread of the virus.
“So what we’re seeing is that we monitor quite a lot of what we think of as sort of nonstandard, high-frequency data. That’s become a very important thing, even more important than usual in the work that we do,” said Federal Reserve Chairman Jerome Powell in his July 29 news conference.
Mobility data was credited with helping to pinpoint a trough in economic activity in mid-April as traditional data continued to deteriorate.
“For example, when you look at whether it’s mobility data or other alternative sources, they pretty much all consistently told you in mid-April that the economy was inflecting and that we were bottoming,” Aneta Markowska, chief financial economist at Jefferies, told MarketWatch.
The increase in mobility coincided almost to the day — April 15 — when the first batch of $1,200 stimulus checks began hitting Americans’ bank accounts, she said. That stood in contrast to official economic data at the time that remained unremittingly awful. Monthly data isn’t well suited to capturing inflection points.
It wasn’t until early June, around six weeks later, when official May readings on retail sales and employment began coming in, that the bottoming was apparent in traditional data.That’s when the May jobs report showed an unexpected 2.5 million rise in nonfarm payrolls and a drop in the unemployment rate to a still eye-watering 13.3% from 14.7%. Also, May retail sales jumped more than 18%.
More recently, the second wave of COVID-19 infections that have resulted in renewed restrictions on movement in several European countries had been visible in European mobility indicators, first showing up in Spain, Nordvig said. And in this phase of the cycle, policy restrictions are more localized, which means it’s crucial to look at mobility at the local level, Nordvig said. He added that alternative data is better suited to that task than official economic statistics, which tend to be national in scope..
SafeGraph and other mobility data providers, such as Unacast, have made their raw data available for free to academics, health officials and others in the public sphere. But the information is also highly sought by hedge funds, other investors and businesses who are willing to open their checkbooks. Alphabet Inc.’s Google has made its community mobility data, utilizing the same inputs it uses in products like Google Maps, available to the public, while Apple Inc. offers daily mobility trends reports, which reflect requests for directions in Apple Maps.
But it’s the personal mobility data based on mobile phones, which tells observers whether individuals are visiting stores, going back to offices, and maintaining social distancing that is being used most aggressively, economists and analysts said.
Mobility data isn’t a silver bullet, users noted. Like any alternative data source, it works best in conjunction with other data sets and can require a lot of scrubbing to sort the signal from the noise.
One limitation, Mertens said, is that economists “know very little about what drives the series.” The limited run of data means economists have little clue about seasonality patterns. They also don’t have much idea about how mobility data would perform in a more “normal” recession as opposed to the one that followed the pandemic, which saw a large chunk of the still-active workforce working from home.
While the data offers insights into broader macroeconomic activity, providers say businesses are also relying on it to make decisions amid the pandemic.
Retailers are “laser-focused” on determining which locations best serve consumers who remain wary of going out, said Jonathan Wolf, vice president of partnerships at SafeGraph.
Decisions on site locations — including, increasingly, decisions on stores to close down — that executives and managers used to do “by gut or anecdotally” now rely on hard data, he said.
And it can get more granular than that. While real-time credit-card spending is also a popular alternative data resource, mobility data can provide crucial depth, said Michael Recce, chief data scientist for money manager Neuberger Berman.
Credit-card data shows only that a person went into a store or restaurant and bought something. But mobility data lets the user know how many people have visited a store or mall or other venue. It’s then possible to determine how many visitors actually buy something.
For example, if a mall loses an anchor store, “does the business go to another store or go away? The mobility data tells you that much more clearly than credit-card data,” Recce said.
It also provides insight into whether a consumer was visiting a store for a specific purpose — making it a “destination” — or as a “convenience.” That’s a distinction that is important as consumers and businesses navigate social distancing.
Wolf said SafeGraph’s data can provide users a ticker-based look at such data, allowing users to compare, for instance, how well competing companies are luring customers to particular locations.
Thomas Walle, chief executive and co-founder of UnaCast, said real-estate firms are using the data to understand how individual neighborhoods are recovering. Setting rent levels has become particularly challenging as people move in response to the pandemic. “Now neighborhood activity can be a metric to understand what that rent level should be,” Walle said.
Wolf said SafeGraph has seen a significant rise in interest from investors of all stripes, including hedge funds, seeking data. But it isn’t just the financial and corporate worlds that have latched onto mobility data.
The data has also been used by public health officials to monitor compliance around lockdowns and other restrictions, offering insights into the spread of the virus.
Mobility data can provide other public health-related insights. As reported by MarketWatch, researchers at the University of California, Davis, used mobility data from SafeGraph, PlaceIQ and Google Mobility from January 2020 to April 2020 to measure whether the number of people traveled outside their home during the pandemic varied by income. They found that before the pandemic, the wealthiest were the most mobile while the poorest were the least mobile.That flipped after stay-at-home orders went into effect and nonessential businesses closed, with the wealthy staying put and the poorest traveling the most.
So will mobility data have staying power?
“I think it’s going to be part of the normal toolkit” once the health crisis is over, Markowska said. “Maybe I won’t be refreshing my spreadsheets everyday like I do now…when things stabilize maybe I’ll go to once a week. But I certainly don’t think we’re going to stop looking at these data.”
