Incidence of the Digital Economy and Frictional Unemployment: International Evidence

This paper is the first to quantify the relationship between the incidence of the digital economy and long-term frictional unemployment across countries. The resulting evidence indicates that there is a robust, negative partial correlation between national unemployment rates and the incidence of the digital economy, proxied by the share of the adult population that reports using the internet to pay bills. Further, the absolute values of ordinary least squares estimates of the partial correlation suggest that it might be higher for developing economies than high-income economies. Controlling for the incidence of informal employment appears to be key for removing a positive omitted- variable bias in the estimate of the partial correlation between unemployment and the digital economy, which is due to the existence of a negative bivariate correlation between unemployment and informality on the one hand, and a negative bivariate correlation between informality and the incidence of digital payment on the other hand. The results from instrumental variable estimations suggest that the partial correlation between unemployment and digital payments is negative, with the absolute value of the estimates being larger than the absolute value of the ordinary least squares estimates.