| dc.description.abstract | This thesis explores the relationship between local labour demand and the uptake of disability insurance (DI) in Norway. We employ a Bartik shift-share instrumental variable approach to perform an empirical analysis into the causal relationship between local labour demand and DI uptake. The empirical strategy exploits variation in industry composition between different localities in Norway and calculates local industry shares. The industry shares are interacted with the national employment growth in the different industries to construct an instrument for local labour demand and identify a causal relationship. Our main contributions to existing research consist of an exploration into the heterogeneous impact for different demographic groups. We also conduct an extensive descriptive analysis of the welfare careers of those who later become DI recipients, based on microdata.
The descriptive analysis shows that individuals who later take up DI often have a history of previous benefit uptake, and to a much larger extent than the general population take up both health-related benefits and labour market related benefits. Even ten years prior to the start of a disability insurance spell, individuals were more likely to take up both unemployment insurance and sickness benefits.
The empirical analysis finds that for the general population, a 1% decrease in local labour demand leads to a 0.852% increase in DI uptake. The effect is stronger for women than for men, possibly explained by the fact that women in general are more elastic than men when it comes to labour supply. However, the estimated elasticities by gender are not statistically significantly different from one another. For the different age groups, we uncover a U-shaped pattern with a stronger effect for those in the 30-40 and 40-50 age groups than for the under 30- and over 50 age groups. The different age group results are inconclusive as several are not statistically significant, nor are the results statistically significantly different from one another. All our empirical analysis results are consistent and robust using several robustness checks. | en_US |
