Changes in the Economic Returns to Attending Prestigious Institutions in Norway

ABSTRACT

The purpose of this paper is to explore whether the returns to attending a prestigious higher education institution have increased in Norway for students who graduated in 1975 to students who graduated in 1999. A handful of published studies find that the returns to attending prestigious institutions have increased in the last decades. However, they all study the USA, which has a decentralized wage bargaining system. The results in this paper shed light on whether a centralized wage bargaining system, typical for many west-European countries, prevents the returns to attending prestigious institutions from increasing. I find that attending prestigious institutions pays off economically in Norway and that the returns have increased throughout the investigated period. Another aim of this paper is to investigate whether the returns have increased more at the top of the earnings distribution than at the bottom. I find that the returns are highest at the top of the earnings distribution and have increased the most at the top. At the 90th quantile, for instance, students who attended a prestigious business school between 1975 and 1979 obtained 80% higher earnings than students attending nonprestigious institutions in the same period. Among students who graduate 20 years later, this difference is 139%.

Key words:

• prestigious institutions
• Europe
• wage bargaining
• labor market mismatch
• returns to higher education
• unconditional quantile regression

Acknowledgments

I would like to thank Arne Mastekaasa, Solveig T. Borgen, and the anonymous reviewers for their helpful comments and suggestions. The data used in this article are provided by Statistics Norway for the project Educational Careers at the Department of Sociology and Human Geography, University of Oslo.

Supplemental data

Supplemental data for this article can be accessed here.

Notes

¹ By using information on earnings between 1987 and 1998 for individuals who were 14–21 years old in 1979, Black et al. (2005) find that the returns in the USA have remained stable across the time span. However, this study captures how the returns change across time and over the course of students’ work careers.

² Decentralization of the wage bargaining system (social norms, deunionization, and performance pay) the last decades explains at least some of the growing wage inequality in the USA (Lemieux 2008).

³ If prestigious institutions influence whether students graduate or not, including only graduates may bias the returns to attending prestigious institutions.

⁴ Norwegian University of Science and Technology is included even though it obtained University status as late as 1994.

⁵ Missing observations is handled using list-wise deletion. Because this paper uses administrative data, few observations have missing on the variables. Nevertheless, to investigate whether missing leads to systematic bias, I have estimated the returns using a full-information maximum likelihood estimator (FIML), considered to be a state-of-the-art missing data technique (Enders 2010). FIML provides similar results as the main OLS model described in Equation (1), which suggests that missing does not bias the results. Results are available on request.

⁶ The effects of attending prestigious institutions are stronger when excluding controls for field of study and educational level. These results can be obtained upon request. Also available upon request are models that are estimated separately by gender. The trends are similar for men and women and I therefore estimate the model with gender as a control variable only.

⁷ Endogenous selection may bias the results in this paper as well. By excluding students who do not have positive earnings, I exclude 4.1% of the entire sample (2.6% due to missing on the earnings variable and 1.5% due to zero earnings). However, this percent is much lower than if I were to include only fulltime workers. Lindahl and Regnér (2005) reduce their sample by 25% by restricting the sample to graduates that earn at least SEK 100,000. The FIML results indicate that excluding students with zero earnings and missing on earnings does not bias the results.

⁸ For instance, for the students who graduate in 2005 to be 36 years old within 2010, they must be at least 31 years old when graduating.

⁹ Note that I control for field of study. Hence, students graduating from a prestigious business school do not obtain higher earnings because they have a business education, but because they have an education from a prestigious institution.