Cyclical asymmetries and spatial dependence in Okun’s Law: global evidence from 163 countries

This study assesses the empirical validity, heterogeneity, and spatial dependence of Okun’s Law in a global setting. Using annual data for 163 countries over the period 1992–2023, we estimate country specific unemployment–output elasticities under two standard specifications (output gap and first difference models) and allow for cyclical asymmetries by distinguishing expansionary and recessionary phases. The results indicate that Okun’s coefficient is negative and statistically significant in most countries, although its magnitude is highly heterogeneous and varies systematically across income groups. Controlling for the common 2020 shock (COVID 19) does not meaningfully alter statistical significance for most countries, but it generates economically relevant shifts in the coefficient’s magnitude for a non negligible subset, thus improving cross country comparability. We also document pronounced asymmetry: elasticities are, on average, stronger during recessions than expansions, particularly among middle and high income economies. Moran’s I statistics reveal positive and significant spatial autocorrelation in cyclical sensitivities across alternative k nearest neighbour weighting matrices, with stronger dependence during recessions. These findings motivate the design of countercyclical labour market policies tailored to structural heterogeneity and coordinated regionally during downturns.