Long ties, disruptive life events, and economic prosperity

Abstract

Social networks shape and reflect economic life. Prior studies have identified long ties, which connect people who lack mutual contacts, as a correlate of individuals' success within firms and places' economic prosperity. However, we lack population-scale evidence of the individual-level link between long ties and economic prosperity, and why some people have more long ties remains obscure. Here, using a social network constructed from interactions on Facebook, we establish a robust association between long ties and economic outcomes and study disruptive life events hypothesized to cause formation of long ties. Consistent with prior aggregated results, administrative units with a higher fraction of long ties tend to have higher-income and economic mobility. Individuals with more long ties live in higher-income places and have higher values of proxies for economic prosperity (e.g., using more Internet-connected devices and making more donations). Furthermore, having stronger long ties (i.e., with higher intensity of interaction) is associated with better outcomes, consistent with an advantage from the structural diversity constituted by long ties, rather than them being weak ties per se. We then study the role of disruptive life events in the formation of long ties. Individuals who have migrated between US states, have transferred between high schools, or have attended college out-of-state have a higher fraction of long ties among their contacts many years after the event. Overall, these results suggest that long ties are robustly associated with economic prosperity and highlight roles for important life experiences in developing and maintaining long ties.

Keywords: economic outcomes; social networks; structural diversity; weak ties.

Fig. 1.

(A) Fraction of long ties, measured across counties in the United States, after controlling for average degree. The adjusted (residualized) fraction of long ties across administrative units in each country is binned into 10 deciles, with lower values shown in purple and higher values shown in green. Administrative units with less than 200 Facebook users sufficiently active over the measurement period or with less than 500 individuals are shown in white. (B) The fraction of long ties and (C) the tie-strength weighted fraction of long ties predict median income in US zip codes. The binned regression plots are generated according to models (6) and (7) in Materials and Methods, which adjust for the number of nodes in the zip code, its racial composition, and average degree; the regression with weighted fraction of long ties also adjusts for the fraction of long ties and the sum of the weights. Solid lines are local smoothers of second degree; bars are 95% confidence intervals, which are cluster-robust at the county level.

Fig. 2.

(A) Fraction of long ties and (B) tie strength weighted fraction of long ties in ego networks versus individual-level variables that are positively correlated with proxies for economic outcomes for users who reside in the Top 6 most populous states in the United States. Users are matched on degree, average degree of their contacts, daily login activity, average daily login activity of their contacts, age, and gender; see Materials and Methods. The matching for the bottom row additionally incorporates the fraction of long ties and average edge weight. Values in the y-axis correspond to the difference with the baseline represented with a value of zero. Bars are 95% confidence intervals for these differences assuming independent observations; robustness to within-state dependence is illustrated by the multiple state-specific results.

Fig. 3.

Estimated average marginal change in measures of individual economic prosperity given a 1% point (0.01) increase in the fraction or weighted fraction of long ties. These are computed from random forest models fit to each state and DC and including degree, average degree of contacts, daily login activity of the user, average daily login activity of their contacts, average weight of user’s edges, age, and gender. Bars are 95% confidence intervals that are cluster-robust at the state level.

Fig. 4.

Conditional on degree, people who (A) are interstate migrants or (B) attended multiple high schools have more long ties than matched controls. Qualitatively similar relationships hold when restricting the analysis to less directly implicated ties—those within the current state (A inset) and those outside of high school (B inset). All estimates are poststratified by gender, age, and hometown county income bins.

Fig. 5.

People who attended a high school during its closure have more long ties than those who attended the same high school prior to closure. The estimates are poststratified by school and age. Bars are bootstrap 95% confidence intervals that are cluster-robust at the school level.