The link between unemployment and real economic growth in developed countries

Ten years ago we presented a modified version of Okun law for the biggest developed economies and reported its excellent predictive power. In this study, we revisit the original models using the estimates of real GDP per capita and unemployment rate between 2010 and 2019. The initial results show that the change in unemployment rate can be accurately predicted by variations in the rate of real economic growth. There is a discrete version of the model which is represented by a piece wise linear dependence of the annual increment in unemployment rate on the annual rate of change in real GDP per capita. The lengths of the country-dependent time segments are defined by breaks in the GDP measurement units associated with definitional revisions to the nominal GDP and GDP deflator (dGDP). The difference between the CPI and dGDP indices since the beginning of measurements reveals the years of such breaks. Statistically, the link between the studied variables in the revised models is characterized by the coefficient of determination in the range from R2=0.866 (Australia) to R2=0.977 (France). The residual errors can be likely associated with the measurement errors, e.g. the estimates of real GDP per capita from various sources differ by tens of percent. The obtained results confirm the original finding on the absence of structural unemployment in the studied developed countries.

💡 Research Summary

This paper revisits a modified version of the Okun law presented ten years ago, focusing on major developed economies and their relationship between real GDP per capita and unemployment rates from 2010 to 2019. The study confirms that changes in the unemployment rate can be accurately predicted by variations in the rate of real economic growth.

The research introduces a discrete model where there is a piecewise linear dependence of the annual increment in the unemployment rate on the annual change in real GDP per capita. Country-specific time segments are defined based on breaks in GDP measurement units, associated with definitional revisions to nominal GDP and GDP deflator (dGDP). The difference between CPI and dGDP indices reveals years when such changes occurred.

Statistically, the link between these variables is characterized by a coefficient of determination ranging from R²=0.866 for Australia to R²=0.977 for France, indicating strong predictive power. Residual errors are likely due to measurement errors, as estimates of real GDP per capita can vary significantly across different sources.

The findings confirm the absence of structural unemployment in the studied developed countries, providing a robust framework for understanding and predicting the relationship between economic growth and employment levels.