Higher education, causality and growth: a comparison of France and Germany before the Second World War

Abstract

Economists were long unaware of the influence of knowledge upon the growth process. With theories of human capital and theories of endogenous growth, knowledge gained a central position in the growth process. However, this has not been proved in many industrialized countries. The nature of the relations between education and growth is hence far from having been perfectly determined. It is therefore of interest to examine the links between education and economic growth. From this point of view, three questions are addressed. Does higher education play a part in the growth process? Which sector of education is involved? Is its role the same in different countries? Using the notion of causality developed by Granger, the relationship between higher education and economic growth is compared in France and Germany before the Second World War. The results show that higher education has an influence on gross domestic product but only in the case of France. Germany does not seem to obey the dominant theory whereby education is the cause of growth. Moreover, economic growth increases the number of students in Germany. This is consistent with the idea that education is a growth‐driven accompanying investment.

Notes

* LAMETA, Université Montpellier I, Faculté des Sciences Economiques, Espace Richter, Avenue de la Mer, B.P. 9606, 34054 Montpellier Cedex 1, France. Email: [email protected]‐montp1.fr

Especially Diebolt (1997) L'évolution de longue période du système éducatif allemand: 19ème et 20ème siècles, Economies et Sociétés, Cahier de l'ISMEA, Série AF (Special Issue), 23, 1–370; and Diebolt (1999) L'évolution de longue période du système éducatif en France: 19ème et 20ème siècles. Les indicateurs physiques. Research paper no. 8 (Berlin, Comparative Education Centre, Humboldt University).

Institut National de la Statistique et des Études Économiques (INSEE) Annuaires Statistiques de la France, 1951, 1966, 1948–1988; and Institut National de la Statistique et des Études Économiques (INSEE) Statistiques de l'Enseignement Supérieur.

	In other words, by using formalism:
	with n=number of variables;
	p=number of lags;
	ϕp ij=coefficient of the variable j with the lag p in the equation of the variable i.

If the series is stationary, this means that there is a convergence of the series around its mean value. So a shock will only have a transitional effect on the series. However, if the series is non‐stationary, its moments depend on time and a shock will be permanent; this means that the series will move away from its long‐run equilibrium.

If cointegrating relationships are detected, the analysis of the dynamics of the model are performed with a VECM.

The notion of human capital theory is used here even though it appeared many years later.

A variable is integrated of order d if it is necessary to differentiate it d times to stationarize it. This means that the variable is not stationary in level, but its d^th differentiation is stationary. The condition of the same order of integration for the two variables is a necessary condition.

Additional information

Notes on contributors

Magali Jaoul *

* LAMETA, Université Montpellier I, Faculté des Sciences Economiques, Espace Richter, Avenue de la Mer, B.P. 9606, 34054 Montpellier Cedex 1, France. Email: [email protected]‐montp1.fr