Time‐to‐degree and the business cycle

Abstract

This paper presents the results of an empirical investigation trying to explain individual time‐to‐degree variances with business cycle fluctuations. Assuming that students determine the optimum study length at university weighing up the cost of an additional semester against the consumption benefit of studying and not yet working, the general economic environment during the study period should, in turn, influence the individual time‐to‐degree through changes in the cost level and the consumption benefit of an additional semester. The investigation, using a representative data‐set based on Swiss university graduates from 1981 to 2001, shows that changes in the unemployment rate have a significant impact on individual time‐to‐degree.

Keywords:

• time‐to‐degree
• business cycle
• consumption benefit

Acknowledgements

The authors thank the Federal Statistical Office for the permission to use the data from the Swiss graduate survey. They also thank two anonymous referees, the editor and Rainer Winkelmann for helpful comments. All remaining errors are the sole responsibility of the authors.

Notes

1. For example, graduates were asked about their parents’ level of education for the first time in 1999, and questions about exchange semesters were first incorporated in the 1991 questionnaire.

2. In the case of the discipline of law, most graduates embark upon a low‐paid internship after graduating.

3. The regression analysis was also performed using alternatively Gross Domestic Product growth instead of wage change as the explanatory variable. Due to the strong correlation (multicollinearity) between the two variables, the effects on time‐to‐degree were estimated in separate models. The results showed that neither Gross Domestic Product growth nor wage change have been significant in our models. Therefore only the models with wage growth are reported.

4. Short‐term in this context means a period of one to two years.

5. The impact measured in this study is probably that large because a relatively high number of students in the disciplines covered by the data‐set is engaged in side employment. In the humanities and social sciences, for example, 85% of the students reported side employment in 2005 (see Bundesamt für Statistik 2007).

6. In addition, all observations from survey years 1999 and 2001 are omitted in the probit estimation, because graduates with a time‐to‐degree below 10 semesters are not included in the final data‐set: Students who commenced their studies after 1992 were excluded (see Section 4.3).