Social Distancing By the Numbers: Who's Staying Home?

The New York Times has been doing some excellent reporting about the spread of COVID-19. I particularly admire their graphics, which put the message into a visual form that anyone with the eyes to see can comprehend and appreciate.

One hopes that most Americans now know that COVID-19 spreads through person-to-person contact, and that the best way to avoid contracting or spreading the virus is to avoid interacting with others in close proximity--or better still, to simply stay home. Has this message sunk in? The visualizations published in today's NYT (which are not only informative, but also beautiful), which are based on analyses of 15 million anonymous Americans' cell phone use over the past few weeks, show just how much (or little) people in each U.S. county have been curtailing their travel over the past few weeks.

The three lessons these data teach are striking and troubling.

First, there is tremendous county-by-county variation in how much people have reined in their travel. In some counties (in the light pastels and greys below), travel has ground to a near standstill, with the average daily travel declining from five miles a day to around a mile or so:

Clearly, people in those light-pastel and grey counties have stopped driving their cars and have turned instead to walking their dogs:

Second, the declines in travel are not uniformly distributed across the nation. It is particularly noteworthy that counties with stay-at-home orders in place have had much steeper reductions in travel than those without travel orders in place. People in counties with stay-at-home orders have curtailed their travel by 80% or so; those in counties without stay-at-home orders have curtailed their travel by maybe 65%. That difference of 15% might not sound like much, but it's actually a huge effect, so readily comprehensible to the naked eye that you don't even have to do any statistics on the data to appreciate the difference:

Third, the counties with stay-at-home orders are mostly concentrated in the Northeast, the West Coast, and the Midwest. Unsurprisingly, given how few stay-at-home orders are in place, the counties in which people have reduced their travel the least are concentrated in the South. In Duval County, where I grew up, people were still driving about 3.4 miles per day this past Friday, making it the third least staying-at-home large county in the Nation. (My family members in Duval County, to my great relief, have been locked down in their homes for two weeks).

These figures say all we really need to know about staying at home during this crisis: Whether you like the idea of the state or county officials ordering Americans to stay at home during outbreaks of communicable diesases (for what it's worth, the federal government arrogated that power long ago, and has exercised it with impunity, as the need has arisen, for centuries), stay-at-home orders seem to be working (bearing in mind the standard caveats about correlation vs. causation). The apparent effectiveness of stay-at-home orders at getting people to stay at home is so striking that it's almost as if people possess a tendency to heed the directives of people in positions of legitmate authority--particularly when those people have the ability to impose sanctions.

The second lesson, equally clear, is that the Southern states, along with Texas, Oklahoma, Kansas, Wyoming, and a few others, are still in for a great deal of pain.