رشد بهره وری و تغییر فنی مغرضانه در آموزش عالی فرانسه

This paper analyses the nature of technical change in the French labour market. Data Envelopment Analysis (DEA) is adopted to investigate productivity change in a sample of higher education leavers over the period 1999 and 2004. In a first step, the Luenberger Productivity Indicator (LPI) is used to estimate and to decompose productivity change. Following LPI, a better productivity is found for the workers in Paris and the well-qualified occupations in France. In analysing the nature of the technical change by the concept of parallel neutrality, technical progress seems to have influenced all professions. In particular, biased inputs of human capital component benefit more for the well qualified professions with an upper increase of the efficiency scores for executives and teachers. Furthermore, some evidences show the key role of “learning by doing” in the worker's adaptation to technical change. Policy implications are then derived from our results.

Assessment of education efficiency and the optimal investment of schooling have been studied since long time (e.g. Mincer, 1958). Following human capital theory, there is a lack of investments in education (Becker, 1962 and Schultz, 1961). Some frameworks however appear in the beginning of the seventies which counterbalance this idea. In particular, the overinvestment idea in education was firstly investigated by Freeman, 1971 and Freeman, 1976 and Dore (1976). Freeman found that education returns have significantly decreased in the United States. He assigned these diminishing returns to an excess of graduate offer. The results called into question the belief that college degrees mirror a profitable investment and a virtual guarantee of economic success. Freeman's work has then catalysed a large literature on overeducation topic.1 Meanwhile, several papers have showed significant overeducation rates in several developed countries (Sloane, 2003). Today, the overeducation concept can be defined as follows: a worker is considered as overeducated if his/her educational level exceeds that required for a particular job.2 Therefore, the overeducation literature asks again the assessment of education efficiency in taking surplus schooling into account. In other words, this literature suggests a diminishing of education efficiency due to the mismatch between offer and demand of skills. In this topic, Guironnet and Peypoch (2007) have proposed a new measure of overeducation3, in using efficiency scores of human capital components. This previous paper has measured overeducation as the difference between potential income — determined by the frontier of the production boundary — and real income.4 Several frameworks have used the frontier approach to explain the gap between the current and potential wage. For example, Polachek and Robst (1998)5 have considered the residual (determined by a stochastic production frontier) between potential and current wages as a lack of information about the correspondent wages and the educational level for the job. In this paper, we choose a different specification of our frontier approach in order to have a more general interpretation in terms of overeducation: a part of this phenomenon is probably due to a lack of information. One benefit of our measure is to capture the technical change which is measured upon the durable professional insertion of graduates from higher education. For this purpose, this framework studies the durable insertion of graduates of higher education. In the present paper, the aim is to extend the extant literature by investigating whether the innovation process influences the skill demand. For example, if the wage downgrading measure quantifies technical change for all professions, this framework proposes to study the effects of innovations upon each occupation: is there a stronger influence of technical progress for some professions? Which type of bias: formal or informal education “saving”? To answer to these questions, the first section presents the theoretical model to apply our measure and the decomposition of technical progress. Second section presents the data and the statistical insights. Third section presents and comments the results before to conclude.

The overall conclusion of this paper is that the increasing productivity of the youth workers is mainly driving by technological change. In particular, significant differences in the required skills between provinces and Paris are showed (excepted for civil servants), with a higher productivity in the capital city. The “learning by doing” seems furthermore to be the major determinant for the worker's adaptation to the technical progress. This result is in conformity with a better productivity in Paris, where employers have decreased their educational requirements to favour the job tenure. Nevertheless, if the professional experience seems today to be the major determinant of the productivity and that this component seems to be a substitute for schooling, it was obvious that a minimum of educational level is required. A part of schooling can be also seen as a complement to the informal education: The capacity to carry out new technologies is proportional to the educational attainment (Bartel and Lichtenberg, 1987). Following these results, the major recommendation for policy-makers would be to favour training courses in firms during the schooling, and to improve professional insertion and productivity. From the graduates' viewpoint, they must choose their courses in function of the expected economic situation and the demand of the labour market. Paris seems in particular to be an attractive place of work since students decrease schooling cost in cut off their studies and they earn quicker higher wages. In addition, such behaviours should increase job satisfaction in decreasing overeducation extent (Vieira, 2005). To confirm our results, others studies however are needed. For example, some comparisons with other labour markets of developed countries should be envisaged. Furthermore, study upon the professional path of the workers should be investigated since higher skills protect workers against their possible non-adaptability face to the further innovations. Such framework is a challenging task ahead.