Profs. Luigi Zingales (Univ. Chicago), Luigi Guiso (European Univ. Institute) and Paola Sapienza (Northwestern) have a new paper out called Long-Term Persistence. They test Robert Putnam’s theory in Making Democracy Work, that it was the civic regions of Italy that became wealthy rather than the wealthy regions that became civic, and that high social capital levels persisted for centuries.
Zingales et. al are curious why, despite the rise of Asian tigers, the Industrial Revolution, and the Internet Revolution, there is still a correlation of 0.23 (surprisingly high) between income per capita by countries in the 1700s and 2000.
Their findings, using more sophisticated econometrics and instrumenting of variables finds that Putnam was right that levels of social capital help explain this persistent economic growth. Using Italian data, and even instrumenting (adjusting) for factors that would have led to city-state independent regions forming back in medieval times, they find that social capital helps significantly explain the north-south economic gap in Italy. And since not all parts of the north got these city-state independent institutions, their analysis helps explain how cities that experienced independence as free city-states in the 1200s today have higher levels of social capital and wealth.
They also try a difference-of-difference approach. “We…compare the difference in social capital between the towns predicted to become free city-states in the Center North (where they did…) and predicted to become free citystates in the South (where they did not). We use the difference in social capital between towns not predicted to become free city-states in the Center North and in the South as a control for generic differences between North and South. When we do so, we find that there is much less social
capital in the South regardless (an effect that could be either driven by history, as suggested by Putnam, or geography). The difference between free city-states and not within each macro-region, however, is present only in the North. For example, Northern free city-states have 17% more nonprofit associations than similar Northern towns that were not free city-states.
“Our difference in difference estimates suggest that at least half of the gap in social capital between the North and the South of Italy can be attributed to the free city-state experience.” In the south of Italy they find validation of their instrument in that current levels of social capital do not differ between those that were expected to form city-states and those that were not and helps explain the impact of the Normans on the formation of southern city-states.
They instrument for the probability of becoming a city-state with historical factors (such as the Etruscan origin of the city and the presence of a bishop in year 1,000).
Having validated their instruments, they that one standard deviation increase in social capital increases per capita income by 21%. “This estimate vindicates [Kenneth] Arrow’s (1972) statement that much of economic backwardness is due to lack of trust and social capital.” And the authors have done some indepent modeling in another 2008 article, “Social Capital as Good Culture”, Journal of the European Economic Association that suggests that even a 2-3 generation positive experience of cooperation can have lasting impact on the intergeneration transmission of beliefs in periods of centuries, even without the survival of any legal institutions.