Skip to content

New Dallas Fed Social Distancing Index Gives Insight into COVID-19’s Economic Impact

Tyler Atkinson, Jim Dolmas, Christoffer Koch, Evan Koenig, Karel Mertens, Anthony Murphy and Kei-Mu Yi

As the human and economic toll from the COVID-19 pandemic mounts, near-term forecasts of output and employment are crucial for assessing appropriate monetary, fiscal and health policy responses. An appropriate real-time index of social distancing can provide valuable real-time insight into the economic impact.

Economic forecasts typically rely on recent monthly or quarterly data on economic activity. These data are usually available several weeks, or even months, after the period in question. Such delays are crippling. In mid-March, it became clear there was an acute need for real-time measurement of the rapid slowdown in economic activity.

A key driver of the slowdown was an increase in social distance—literally, the physical distance between people—in order to mitigate the spread of COVID-19. Many businesses sharply curtailed, or even ceased, operations due to government-mandated closures, concern for the health of workers or a lack of business, as consumers avoided social interaction.

To gain insight into the economic impact of the pandemic, we developed an index of social distancing, based on geolocation data collected from a large sample of mobile devices. Our Social Distancing Index (SDI)—described in greater detail below—spiked dramatically in mid-March, coinciding with a large drop in economic activity. Recently, the index has started to decline, even prior to the relaxation of stay-at-home orders. The drop suggests economic activity may have bottomed out and could begin to improve.

Dallas Fed Social Distancing Index

The firm SafeGraph, a geospatial data analysis firm, has provided several datasets to researchers to help society respond to COVID-19. Their Social Distancing Metric database contains aggregated, anonymized, privacy-safe data on a range of spatial behaviors of mobile devices. No single indicator in the SafeGraph dataset adequately captures all aspects of social distancing, and each is noisy and subject to idiosyncrasies.

Our index therefore summarizes the information in seven different variables, each measured daily at the county level and relative to its weekday-specific average over January–February. The variables are:

  • Fraction of devices leaving home in a day.
  • Fraction of devices away from home for three to six hours at a fixed location.
  • Fraction of devices away from home longer than six hours at a fixed location.
  • An adjusted average of daytime hours spent at home.
  • Fraction of devices taking trips longer than 16 kilometers (10 miles).
  • Fraction of devices taking trips less than 2 kilometers (1.2 miles).
  • Average time spent at locations far from home.

These variables are combined via principal component analysis, which extracts a weighted average of the seven-variable series that best explains their variation. The resulting combination is our county-level SDI. We then aggregate the county-level SDIs to the metropolitan statistical area (MSA) and state and national levels. The national SDI is plotted in the left panel of Chart 1, scaled so that January and February average zero, and the maximum of the weekly averages (the week ended April 11) is 100.

Chart 1: Social Distancing Spiked in Late March, Receded in April

Downloadable chart | Chart data

The SDI varied minimally in January through early March 2020, reflecting the regular spatial behavior of cellphone users before the pandemic. The index began to increase beyond its normal range in the second week of March and increased sharply in the week ended March 21.

By the end of March, social distancing had plateaued, and over the first half of April, it moved sideways. In the second half of April, the index began to decline, as more Americans began leaving home and taking longer trips away from home. For the average of the week ended May 2, the index was 20 percent below the mid-April peak.

Each gray line in the right panel of Chart 1 is the SDI for an MSA. The tightness of the range in these local SDIs indicates that most metro areas increased their social distancing at approximately the same time and by roughly the same amount, despite differences in the timing and scale of government interventions and the local pace of COVID-19 infections. A similar pattern holds for counties and states and suggests that stay-at-home orders only partially explain social distancing. This is corroborated by the SDI’s decline in the second half of April before stay-at-home orders were lifted, as well as by the findings of several academic studies.

The SDI is consistent with other mobility metrics that have become available since the start of the crisis. For example, our index is tightly correlated with the Google Covid-19 Mobility metrics for residential dwelling and travel for work, recreation or shopping purposes.

Relationship to Economic Activity

Without much doubt, social distancing was a major factor in the slowdown in economic activity and the sharp rise in unemployment. The left panel of Chart 2 plots the national SDI along with the Weekly Economic Index (WEI).

Chart 2: Social Distancing Index Correlated With Measures of Economic Activity

Downloadable chart | Chart data

The WEI, developed by researchers at the Dallas and New York Feds and Harvard econometrician James Stock, is an aggregate of 10 daily and weekly indicators of economic activity that is highly correlated with output growth relative to a year prior.

The steep increase in the SDI coincided with the large drop in the WEI over the weeks ended March 21 through April 11. In the second half of April, social distancing began to recede, while the WEI only slowed its decline. While it is too early tell, a continued drop in the WEI could indicate more conventional recessionary dynamics, as cautious consumers and businesses pull back from spending and hiring, amplifying the initial disruption caused by social distancing.

The right panel of Chart 2 plots the SDI along with indicators of small business activity from the scheduling software firm Homebase. Both the number of small establishments open for business as well as the number of hourly employees working at those establishments contracted sharply with the rise in the SDI. Recent weeks show signs of improvement in small business activity as social distancing begins to wane.

The impact of social distancing on economic activity is also evident when looking across counties or MSAs. We find, for example, that at the April peak of social distancing, localities engaging in 10 percent more social distancing relative to the national average saw an additional 0.6 percent of the population claiming unemployment insurance, an additional 2.8 percent reduction in small businesses employment and an additional 2.6 percent increase in small business closures.

Social Distancing and the Economic Outlook

The SDI captures what is arguably the primary driver of the large drop in economic activity and, therefore, is a key metric in forming our assessment of economic conditions and the outlook for future activity. We expect social distancing behavior to decline further in the coming weeks and months as government restrictions ease.

However, we anticipate social distancing to remain elevated as long as the threat of COVID-19 infection persists, regardless of official restrictions. As a result, economic activity is also expected to continue depressed, as the hardest-hit industries—such as leisure and travel—remain materially impacted. Of course, many uncertainties remain, chief among them the path of COVID-19 and the success or failure of efforts to develop treatments and vaccines.

About the Authors

Tyler Atkinson

Atkinson is a business economist in the Research Department at the Federal Reserve Bank of Dallas.

Jim Dolmas

Dolmas is a senior research economist in the Research Department at the Federal Reserve Bank of Dallas.

Christoffer Koch

Koch is a senior research economist in the Research Department at the Federal Reserve Bank of Dallas.

Evan Koenig

Koenig is a senior vice president in the Research Department at the Federal Reserve Bank of Dallas.

Karel Mertens

Mertens is a senior economic policy advisor in the Research Department at the Federal Reserve Bank of Dallas.

Anthony Murphy

Murphy is a senior economic policy advisor in the Research Department at the Federal Reserve Bank of Dallas.

Kei-Mu Yi

Yi is a senior vice president in the Research Department at the Federal Reserve Bank of Dallas.

Economic ConditionsLaborCOVID

Economic analysis and insights from the Federal Reserve Bank of Dallas

Subscribe to Dallas Fed Economics email alerts
Our People
Subscribe to email alerts for economic research and updates