National Bank of Denmark working paper finds marginal workers’ unemployment is mainly driven by lower productivity

The National Bank of Denmark published Working Paper No. 211 on why a small share of Danish workers accounts for most unemployment, finding that repeated joblessness among “marginal workers” is explained mainly by lower productivity and non-pecuniary opportunity costs rather than higher pecuniary opportunity costs, with direct implications for how unemployment insurance should be designed. Using Danish administrative data (linked across earnings, balance sheets and other records) and a worker-typing approach, the paper identifies “stable” workers (86%) and “marginal” workers (14%), with marginal workers showing about 35% unemployment and accounting for around two-thirds of total unemployment. Empirical patterns point away from high pecuniary opportunity costs, as marginal workers have lower wealth and weaker family insurance alongside indicators consistent with lower productivity, including lower education and wages, lower worker fixed effects, more temporary jobs and more separations for economic reasons. A calibrated directed-search model that also uses consumption moments estimates marginal workers are around 25% less productive, have lower home production, and have a positive non-pecuniary utility shifter during unemployment; counterfactuals reduce marginal unemployment from 35% to about 4% when productivity is equalised with stable workers, and to about 6% when the non-pecuniary component is removed. In a stylised extension with flat unemployment benefits financed by proportional labour taxation, the welfare-maximising benefit is highest in the baseline calibration (g* = 0.093), exceeding cases where the unemployment gap is driven only by productivity (0.016), pecuniary opportunity costs (0.012) or non-pecuniary opportunity costs (0.069), with the paper noting these are illustrative model results rather than policy prescriptions.