Jobfinding and Wages when Longrun Unemployment is Really Long: The Case of Spain
Alfonso Alba-Ramirez, Richard B. Freeman
NBER Working Paper No. 3409
This paper uses the "Encuesta de Condiciones de Vida Y Trabajo" (EGVT) -- a survey of the labor force activity of over 61,000 persons in Spain in 1985 when unemployment exceeded 20%--to examine the effect of unemployment insurance (UI) and family status on long-run joblessness. It finds that (1) duration of joblessness is some 30' longer for those eligible for UI benefits than for those ineligible for UI; (2) the long-term unemployed are disproportionately secondary workers for whom the family serves as a form of welfare; (3) hazard rates linking the chances of job finding to duration of unemployment in the 1981-85 period of massive joblessness did not decline with duration; (4) the length of unemployment spells reduces wages moderately but has huge effect on the probability that re-employed workers take secondary sector jobs; (5) the UI eligible earn more and are more likely to gain regular full-time jobs than those ineligible for UI, congruent with the additional months of job search associated with UI. The estimated effects of duration on the hazard and on earnings are consistent with the implications of labor supply and search analysis but not with the view that long unemployment spells create a class of unemployables. Our results imply a sizeable reduction in long-term unemployment with economic recovery.
Published: Published as "Self-Employment in the Midst of Unemployment: The Case of Spain and the United States", AE, Vol. 26, no. 3 (1994): 189-204.