E Johansson1, P Böckerman2, A Lundqvist3. 1. Åbo Akademi University, Tuomiokirkontori 3 20500 Turku, Finland. Electronic address: edvard.johansson@abo.fi. 2. University of Jyväskylä (School of Business and Economics) Jyväskylä, Finland, Labour Institute for Economic Research and IZA Institute of Labor Economics. 3. National Institute for Welfare and Health, Helsinki, Finland.
Abstract
OBJECTIVES: This paper examines the relationship between unemployment and health using both subjective and biometric information on health status. STUDY DESIGN: Longitudinal panel data. METHODS: We compare the results of regressions of unemployment on self-reported health with those of regressions of unemployment on health as measured with biomarkers (hypertension and levels of blood glucose and C-reactive protein). Using the panel structure of our data, we account for selection bias with respect to unemployment by controlling for health before exposure to unemployment. RESULTS: We observe a striking pattern. Using self-reported health as the outcome variable, we find a link between unemployment and worse health. By contrast, we are unable to establish the same link using biometric information on health. CONCLUSION: In conclusion, our results indicate a substantial discrepancy between self-reported health and health as measured by biomarkers.
OBJECTIVES: This paper examines the relationship between unemployment and health using both subjective and biometric information on health status. STUDY DESIGN: Longitudinal panel data. METHODS: We compare the results of regressions of unemployment on self-reported health with those of regressions of unemployment on health as measured with biomarkers (hypertension and levels of blood glucose and C-reactive protein). Using the panel structure of our data, we account for selection bias with respect to unemployment by controlling for health before exposure to unemployment. RESULTS: We observe a striking pattern. Using self-reported health as the outcome variable, we find a link between unemployment and worse health. By contrast, we are unable to establish the same link using biometric information on health. CONCLUSION: In conclusion, our results indicate a substantial discrepancy between self-reported health and health as measured by biomarkers.