Does the pattern of social connections between individuals matter for macroeconomic outcomes? If so, where do differences in these patterns come from and how large are their effects? Using network analysis tools, we explore how different social network structures affect technology diffusion and thereby a country’s rate of growth. The correlation between high-diffusion networks and income is strongly positive. But when we use a model to isolate the effect of a change in social networks on growth, the effect can be positive, negative, or zero. The reason is that networks diffuse both ideas and disease. Low-diffusion networks have evolved in countries where disease is prevalent because limited connectivity protects residents from epidemics. But a low-diffusion network in a low-disease environment compromises the diffusion of good ideas. In general, social networks have evolved to fit their economic and epidemiological environment. Trying to change networks in one country to mimic those in a higher-income country may well be counterproductive.

One of the most dramatic economic transformations of the past century has been the entry of women into the labor force. While many theories explain why this change took place, we investigate the process of transition itself. We argue that local information transmission generates changes in participation that are geographically heterogeneous, locally correlated, and smooth in the aggregate, just like those observed in our data. In our model, women learn about the effects of maternal employment on children by observing nearby employed women. When few women participate in the labor force, data are scarce and participation rises slowly. As information accumulates in some regions, the effects of maternal employment become less uncertain and more women in that region participate. Learning accelerates, labor force participation rises faster, and regional participation rates diverge. Eventually, information diffuses throughout the economy, beliefs converge to the truth, participation flattens out, and regions become more similar again. To investigate the empirical relevance of our theory, we use a new county-level data set to compare our calibrated model to the time series and geographic patterns of participation.