Preferences for unemployment insurance and labor-market risks: over-time developments in Germany and the United States

Abstract

This article examines the relationship between labor-market risks and demand for social insurance. It looks at the over-time variations in preferences for unemployment insurance in Germany and the United States, and delineates the links with these and one's position in the labor market. The results suggest that rather than the type of human capital investment, occupational unemployment rate is explanatory for the demand for social insurance, along with income. Our analysis challenges the widespread association, in the literature, between higher specificity and higher social spending.

Keywords:

• insurance
• specificity
• risk
• unemployment rate
• social policy

Notes

Notes

1. See Hall and Soskice (2001) and Huber and Stephens (2001) among others for a detailed comparison of “coordinated” and “liberal” market economies.

2. The different outcomes of globalization in labor markets in Europe and the United States are explained by the institutional variation.

3. Iversen argues that there seems to be a misconception that de-industrialization is uniform across countries, and therefore cannot explain the cross-national variance in the speed of welfare state expansion.

4. Hiscox and Rickard (2002) use data on changes in sectoral employment patterns to gauge the speed and extent of inter-industry labor movement over time.

5. There is an extensive literature on the growing importance of worker training in US manufacturing industries during recent decades, and the related expansion in the use of fringe benefits tied to seniority as a way to encourage longer tenure among employees once they have been trained on the job.

6. Figures are taken from the OECD Benefit Systems and Work Incentives over various years.

7. The sample mean of skill specificity is 1.387 and the German mean is 1.418.

8. Histograms can be requested from the author.

9. The measure mainly captures relative skill specificity and is derived from the occupational requirements rather than what people really have and use in their jobs.

10. The insignificant variables are not presented in the tables.