Post by Alexandra Middlewood and Mark Joslyn
Professor Middlewood teaches American government, political behavior and gender politics at Wichita State University. She recently published a paper on Gun Ownership and Women’s Political Participation.
In the classic, Bowling Alone, Robert Putnam described the decline of social capital in the United States. He defined social capital as, “…features of social life-networks, norms and trust-that enable participants to act together more effectively to pursue shared objectives.”
Activities in many areas of social life had dropped considerably since the 1960s, including club memberships, volunteerism, church attendance, parent teacher associations or just playing cards with neighbors or bowling with friends.
Putnam argued these social activities broaden participants “sense of self, developing the “I” into the “We”. But in their absence, people are disconnected from each other, the collective spirit fades and civic engagement erodes.
However, while social capital declined, it did not disappear.
Indeed, as the nation now confronts an unprecedented public health crisis, the community-mindedness that social capital fosters will be an important factor in slowing the spread of the coronavirus.
Let’s see how this may work.
To prevent or curb transmission of the coronavirus, governments’ directed people to stay at home, restrict unnecessary movement, and when in public keep 6 feet from others. Only essential businesses could remain open, and until further notice a lockdown of all other social and economic activities.
Repeatedly, officials expressed two reasons for the social distancing measures. First, the shutdowns save lives. Following guidelines protects everyone; yourself, loved ones and the most vulnerable in the community. Do not be selfish, elected officials echoed, stay inside.
Second, limiting movement buys precious time. Hospitals needed to prepare for many infected patients expected during the pandemic. Too many infections, at one time, would overwhelm hospital facilities and endanger the health of front-line workers.
The message was loud and clear. Personal sacrifice helps society.
The slogan adopted across the nation, We Are All In This Together, resonated with millions of Americans.
Because social capital encourages a collective spirt, trust in others, and strong traditions of civic engagement, we see it as a key to successful distancing. Specifically, we think compliance with social distancing guidelines will be notably better in areas of high social capital compared to lower social capital.
Tracking human movement
Unacast records data from millions of anonymous mobile phones to compare the distance that people in each county of the United States traveled before and after the outbreak. Counties and states are assigned grades based on three movement measures: (1) change of average distance traveled, (2) change of visits to non-essential retail and services, (3) decrease in human encounters.[i]
Switching to a home office, for example, strongly reduces average distance traveled and curtailing or ending trips to the hair salon, restaurant or entertainment venues decrease non-essential visits. In these instances, human encounters decline as well. [ii]
Performance on the three measures are averaged to arrive at a final grade.
On April 24, we examined the state of Kansas which includes 105 counties. On that day, the state itself received a C -. Kansas scored a C for reducing average distance traveled, a B for minimizing human encounters, and an F on lowering non-essential visits.
For the 105 counties, grades ranged from A- to D, and the median score a C.
|Grades||# of counties|
We used voter turnout to represent a county’s stock of social capital.[i]
We then compared turnout levels in strong social distancing counties (B – or better) to average and less successful counties (C + or worse) – see Figure below.
The B- or better counties exhibited significantly higher voter turnout (71%) than counties receiving lower grades (66%). Precisely the relationship we anticipated.
The range in turnout accounts for the disparity. For example, turnout does not dip below 60% for the B- or better category but falls far below that for counties receiving average to poor grades.
We cannot yet conclude voter turnout influences distancing. Rather, we first must recognize other factors that likely determine distancing success including county population, income, % with health insurance, percent minority, and of course the number of confirmed Covid-19 cases.
Look for Part 2 next week, where we will examine the impact of social capital more rigorously and discuss what the results mean for politics and public policy.
[i] Importantly, because the first two measures are rates of change, Unacast compares them to a baseline derived the same day of the week during non-Covid-19 period for a specific county. Thus in less-populated rural areas where the baseline of social distancing is naturally much lower, for example, changes in mobility caused by the lockdown are compared to the typical pattern that persisted before.
[ii] Regarding the human encounter metric. Unacast data cannot detect if two people interacted, so they simulated person-to-person encounters. Unacast instead thinks about it as, “the probability that two devices that were in the same place at the same time.” They normalized this metric to account for differences across rural and urban counties – encounters per square kilometer of land area. They also compare it to a national baseline, what they refer to as “business as usual”