Figure 10

a. Reform effect on FJ recruitment from non-employment year-by-year. Governmental sector compared to municipal/regional sector. b. Reform effect on retained FJ year-by-year. Governmental sector compared to municipal/regional sector. Note: LPM estimates. Difference-in-differences model. The solid line indicates the effect, and the dashed lines represent a 95% confidence interval using robust standard errors. The model controls for age, gender, education, lagged degree of unemployment, annual number of sick days, annual number of visits to a doctor and diagnosed with disease in bones, muscles and connective tissue. The period before the reform (2001) is set equal to zero.