Student loan reforms for German higher education: financing tuition fees

Abstract

It is generally agreed that the funding base for German universities is inadequate and perhaps the time has come for serious consideration of the imposition of nontrivial tuition charges. This article compares conventional and income contingent loans (ICLs) for financing tuition fees at German universities. Two aspects are considered: the size of repayment burdens associated with mortgage-style loans, and the time structure of revenue to the government from a hypothetical ICL. We find that tuition fees could increase considerably with the use of an ICL system similar to policy approaches used in Australia, England and New Zealand.

Keywords:

• educational finance
• student financial aid
• state and federal aid
• government expenditures on education

JEL Classifications::

• H52
• I22
• I28

Acknowledgements

We thank Kiatanantha Lounkaew and two anonymous referees for valuable comments and suggestions and gratefully acknowledge the support of the Australian Group of Eight (Go8) and the German Academic Exchange Service (DAAD).

Notes

1. According to the Federal Statistical Office (2008), the average annual public expenditure on tertiary education per student in 2005 was €7180.

2. See Barr (1993), Chapman (1997) and Greenaway and Hayes (2003).

3. See Friedman (1962) and Chapman (2006).

4. This amount is comparable to the current maximum tuition fee of £3290 (about €4000) for the 2010/2011 academic year in the UK, and similar to tuition levels in Australia, Canada and New Zealand.

5. Both the Federal Statistical Office (http://www.destatis.de) and the association of universities and other higher education institutions (Hochschulrektorenkonferenz, http://www.hochschulkompass.de) regularly publish the most recent student and university numbers on their websites.

6. The Federal Training Assistance Act (Bundesausbildungsförderungsgesetz or BAföG) that regulates student loans for financing living expenses focuses on students from low-income households. The eligibility for these student loans typically depends on the parents' income. A more detailed discussion of these loans is beyond the scope of this article.

7. Detailed information about public loans for financing tuition fees at German universities can be obtained from http://www.studis-online.de and http://www.bafoeg-aktuell.de.

8. For a full discussion of the areas of uncertainty see Barr (2001), Palacios (2004) and Chapman (2006).

9. See Chapman (2006).

10. These include New Zealand (1992), England (1997), Ethiopia (2002), Hungary (2003) and Thailand (2006, for one year only).

11. See Woodhall (1987), Ziderman (1999), Schwartz and Finnie (2002), Salmi (2003) and Baum and Schwartz (2006).

12. For example, 10 years is the typical repayment period for US Stafford loans (Chapman and Lounkaew 2010a).

13. For example, in Australia in 2011 no repayments are required until the borrower's income reaches around $(A)47,000 per annum after which 4–8% of income is collected for debt repayment, depending on the borrower's income.

14. Dynarski (1994) used the National Post-secondary Student Aid Study and found strong evidence that experiencing low earnings after leaving formal education is a strong determinant of default. Importantly, borrowers from low-income households and minorities were more likely to default, as were those who did not complete their studies.

15. These uncertainties are considered fully in Barr (2001) and Chapman (2006).

16. Palacios (2004) analyses in full this type of approach to student financing.

17. See Chapman et al. (2010) for the Thai calculations and Chapman and Lounkaew (2010a) for the US analysis.

18. The default insurance and consumption smoothing aspects of ICL are addressed convincingly in a theoretical context in Quiggin (2003).

19. That is, CPI inflation plus 3%. This is about what is charged with Stafford Loans in the USA (Chapman and Lounkaew 2010a).

20. In this scenario we assume that high school is completed at age 19, followed by a year of military or civil service.

21. This figure is about the average inflation rate for the period after the German reunification (http://www.destatis.de).

22. This fact is highlighted by the relatively low explanatory power for OLS earnings functions models. For example, Chapman and Lounkaew (2010b) found an R² of around 0.4 for Thai earnings functions: a plethora of other earnings function studies typically explain no more than 20–30% of the variance.

23. See http://www.parmentier.de/steuer/incometax.htm. Monthly gross income is calculated separately for single and married households with and without children.

24. The labour force status of debtors is irrelevant in our calculation of repayment burdens since debts are owed on the basis of time.

25. In order to check the accuracy of our income measures, we have reproduced our empirical analysis for a pooled sample of the four groups and compared the results to those of a similar analysis based on data from the German Socio-Economic Panel (GSOEP). Due to the small sample size of the GSOEP, we have pooled the years 2007–2009 instead of using a single cross-section. The empirical results obtained from the two data sources are very similar, indicating that the income measures employed in our empirical analysis are quite accurate.

26. Alternative approaches have been discussed in the literature (Geweke and Keane 2000; Moffitt and Gottschalk 2002; Meghir and Pistaferri 2004; Heckman, Lochner, and Todd 2006) and their implications for the design of student loans remain a matter of future research. A comparison of alternative approaches is beyond the scope of this article.

27. A comparison of linear and interval regression estimates suggests that there are no qualitative differences in the results between the two approaches and that quantitative differences are rather small.

28. Since our analysis samples are sufficiently large, we include the dependent variable in levels instead of logs to facilitate the prediction of age–income profiles.

29. The age–income profiles were adjusted by 1.5% per annum to capture productivity growth. A growth rate of about 1.5% is in line with the usual indicators (OECD 2008).

30. Since repayment burdens are highest at the lower end of the distribution, we have also estimated the age–income profiles at the 10th percentile. The estimates reveal very high repayment burdens at the 10th percentile in the first year of the repayment period. Specifically, we obtain repayment burdens of 26.9% for male and 23.6% for female graduates in West Germany and 27.3% for male and 30.2% for female graduates in East Germany.

31. Repayment burdens at the 10th percentile in the first year of the repayment period are 25.0% for male and 18.4% for female graduates in West Germany and 23.1% for male and 23.4% for female graduates in East Germany.

32. The difference in repayment burdens between Germany and the USA is mainly attributable to the fact that loans for living expenses are included in the calculation of repayment burdens in the USA.

33. For example, the predicted income in the first year of the repayment line. Subtracting the repayment of this group is about €8300, already well below the official poverty line. Subtracting the repayment burden of about 23% at this percentile would move individuals in this group to about €6400, which is below 40% of the median equivalence income.

34. The assumed ICL can be compared to those in operation in Australia, in which there is a threshold of income below which no repayments are required, after which given progressive proportions of income are required for repayment and these range from 4 to 8%. In the English and New Zealand schemes, there are also thresholds for repayment with the basis being a set proportion of additional income above this level (Chapman 2006).