Wage Cyclicality of New and Continuing Jobs: Evidence from Chilean Tax Records

This paper provides evidence of the effect of unemployment on individual wages, using administrative data for the universe of wage earners in Chile. We highlight three main results. First, we find that wages of newly created... [ view full abstract ]

This paper provides evidence of the effect of unemployment on individual wages, using administrative data for the universe of wage earners in Chile. We highlight three main results. First, we find that wages of newly created jobs are significantly more sensitive to aggregate unemployment than that of continuing jobs. Based on monthly frequency data on employment status, we find that such excess wage sensitivity is not exclusive of job-to-job transitions (which may reflect procyclical fluctuations in job match quality) but holds also for new jobs created from non-employment, a finding robust to considering different non-employment windows. Second, we find that the differential sensitivity of wage adjustments between continuing and new jobs is concentrated in large firms –which display both less sensitivity of continuing contracts and more sensitivity of new hires vis-\`a-vis smaller firms. Third, we find that the sensitivity of wages to unemployment is particularly large in firms that contract, compared to those that expand or keep employment levels roughly constant, hinting to potential business cycle asymmetries that could be relevant for discriminating between alternative theories behind wage adjustment through the business cycle.