Mobile phone data can quickly and accurately detect and track changes in the economy at multiple levels. In particular, a team from Massachusetts-based Northeastern University found that call detail records can be used to predict unemployment rates up to four months before the release of official reports and more accurately than using historical data alone.
According to David Laser, a distinguished professor of Political Science and Computer and Information Science, their findings are of great practical importance. They potentially facilitate the identification of macroeconomic statistics faster than traditional methods of tracking the economy. Laser and his collaborators harnessed the power of algorithms to analyse call record data from two undisclosed European countries.
Their first study focused on unemployment at the community level, where they examined the behavioural traces of a mass layoff. The findings revealed that job loss had a “systematic dampening effect” on mobility and social behaviour.
For example, the researchers found that the total number of calls made by laid-off individuals dropped 51 per cent. However, the number of calls for non-laid off residents decreased to 54 per cent.
The month-to-month churn of a laid-off person’s social network reduced. This means the fraction of contacts called in the previous month increased approximately 3.6 percentage points. They found that changes in mobility and social behaviour predicted unemployment rates. This was before the release of official reports and more accurate than traditional forecasts. The findings, published in the journal of the Royal Society Interface highlighted the potential of mobile phone data to improve forecasts of critical economic indicators.