Research notes · Knowledge

Do the friendships between two places predict how much they trade?

What the authors found

Bailey, Cao, Kuchler, Stroebel and Wong introduce a measure of social connectedness built from aggregated, anonymised Facebook friendship links between geographies. They show it is shaped by distance, historical ties and political boundaries, and that it in turn predicts migration, trade, job-seeking and the diffusion of products and sentiment. Social structure is measurable at scale and it is economically consequential.

What our data adds

This database holds two of the outcomes their index predicts — bilateral trade for essentially every country pair, and migrant stock by country — and none of the social structure. That is a genuinely useful thing to be able to say precisely. The trade matrix contains a million bilateral observations and can show that Mexico ships 512.7 billion dollars to the United States; it cannot show the family ties across that border that the paper argues are part of why.

What this cannot show

Social connectedness is a proprietary measure derived from a private platform's friendship graph. It is published in aggregated form for research and is not open data in the sense this platform requires, so it could not be loaded here even if the shape fitted. The honest description is that our data holds the dependent variables of the paper and none of the independent one.

The data behind it

Related notes

• When cheap imports arrive, do domestic firms innovate their way out or give up?
• Rich countries spend more on research every year. Why is measured productivity growth falling?
• How much of what gets patented rests on publicly funded science?

Was this useful?

It answered my question Partly No

No account, no comment box. One click, counted and nothing else.