Big data sheds light on income risk – Minneapolis Fed research

New and bigger sources of data are allowing economists to better understand patterns in income risk, according to an article published by the Federal Reserve Bank of Minneapolis.

Fatih Guvenen outlines results based on data on millions of individuals from the US Social Security Administration. One seemingly counterintuitive result is that the variance of income shocks does not vary much with the economic cycle, which seems to contradict the idea that risks grow in recessions.

However, he also finds that income risk “skews negative” in recessions, such that without the variance changing, income risk does indeed grow.

Guvenen also asks whether the data can help predict individuals’ income risk, by developing a “factor structure” to assess the different impacts of an aggregate shock on diverse groups. The data implies that – at least, based on past recessions – the lowest-income groups are hardest hit by a downturn.