Sweden's Riksbank staff memo finds 1% GDP growth cuts regional unemployment about 4% after a year with weaker effects in high-friction regions

Sweden's Riksbank has published a staff memo analysing how unemployment across Swedish regions co-moves with national gross domestic product (GDP) growth over 2000–2023, testing whether Okun’s law holds at the regional level. The analysis finds a clear negative relationship, with the strongest effect on unemployment occurring about one year after GDP increases, but with marked variation across regions and groups. Using local projections on a regional panel dataset and measuring unemployment as open unemployment per inhabitant aged 16–64, a 1% increase in GDP is typically followed by around a 4% fall in open unemployment per capita after one year, with the effect diminishing in years two and three. Regions with high employment and greater mobility respond faster and more strongly, while regions with high unemployment, a large share of foreign-born residents or high population density show weaker adjustments. By group, unemployment falls more sharply for people born in Sweden and for those with at least upper-secondary education, while the relationship is much weaker for the less educated and for people born outside Europe, where a 1% GDP increase is associated with only a 2–3% decline in unemployment. The memo also reports asymmetric dynamics, with unemployment rising more in response to negative GDP changes than it falls after positive shocks of similar magnitude, especially in regions with greater frictions. The authors note the results are descriptive correlations rather than causal estimates, and the staff memo does not represent the Riksbank’s policy stance.