تخصیص سرمایه انسانی و آموزش فوق العاده:اندازه گیری بهره وری فرانسوی (1987، 1999)

In the last four decades, France has experienced an increase in the numbers of students in further education. In this context, we have chosen to quantify the phenomenon of overeducation using a salary-based approach and to measure the impact of the general lengthening in the duration of studies on private sector productivity. The decomposition of the aggregate Luenberger productivity indicator into technological change and efficiency change reveals an increasing imbalance between salaries and required qualifications. Moreover, while overeducation has become less marked for intermediate and higher occupations, mismatches have increased in the case of the least skilled jobs.

In France, the total number of persons in education has increased at an unprecedented rate during the last four decades. This change is explained by the extension of compulsory education, a diversification of the training available within the educational system and a reduction in selectivity. In 2000, graduates accounted for 36% of those leaving the educational system as opposed to 15% in 1980. This increase in student numbers, combined with the employment crisis of recent decades, has transformed the operation of the labour market: statistical observation leads to the conclusion that the youngest generations of individuals leaving the educational system, who are continually better and better educated, find it increasingly difficult to find employment that matches their educational level. Today, the new cohorts of graduates, encountering increased competition when looking for employment, compensate for their lack of professional experience by a surplus of qualifications. One possible interpretation would therefore consider professional experience as a substitute for initial training in order to be a potential candidate for employment: the resulting mismatch can be considered as an adaptative process in the search of an adequate job (Sicherman, 1991). This clearly raises the issue of overeducation, i.e. employees whose level of qualification exceeds that required for the job they do. Put another way, the concept of overeducation relates to inefficiency in the use of the potential workforce. We shall measure overeducation on the basis of the wage downgrading of those leaving the educational system by applying a production frontier model. Among the advantages of this method is that it can be used to evaluate the efficiency of the different forms of human capital allocation. This study aims to answer a key question: what are the changes in productivity in respect of the increasing qualifications of labour force? On the basis of two surveys conducted by the Centre d'Etudes et de Recherches sur les Qualifications (Céreq), this paper analyzes changes in the productivity and overeducation of young graduates between 1987 and 1999. The paper unfolds as follows. The next section introduces a brief account of traditional measures of overeducation, of the production frontier model and of productivity indicators. Section 3 provides details of the data used and presents the findings. Finally, Section 4 concludes.

From 1987 to 1999, the Luenberger indicators indicated a positively correlation between productivity degradation and the profession level. However, the growth of technological progress is significant: the requirement qualification rose by 0.18 years whereas the extension of school duration increased only by 0.12 years. Finally, the main explanation of the degradation of aggregate productivity indicators comes from downward rigidities in starting wages and an over-qualification of unskilled and technician professions. More generally, education surplus would have decreased over these twelve years but wage downward would have increased since the efficiency variation obviously degraded between 1987 and 1999. The underlying growth of unemployment – observed its last decades – would be the main explanation of wage rigidities and thus would be, at the same time, structurally linked to wage downward evolution which experienced an analogous movement. In terms of policy recommendations, the unemployment of young graduates must be reduced in order to appraise all wages, in particular for the most qualified professions which perceived a significant technological progress. Following job search theory, an increased in the spread of wages may stimulate job mobility and may consequently ameliorate the individual matches: this phenomenon is particularly strong for the most qualified professions (Van den Berg, 1995).19 In addition, skill requirements for recruitment candidates must be considerably reduced for least-qualified occupations. A solution would be to set up most vocational studies, like in Germany, and a more attractive labour market in terms of income, like the United Kingdom. The main flaw of the educational system, is that it cannot undertake the necessary means to needs which are yet unknown in long period (Beduwé and Planas, 2004). In long term, if the process persists with an growing imbalance between level of initial training and wages the risk would be a brain drain which is, typically socially expensive to educate. The recent researchers' leaving from France to foreign laboratories, wouldn't it be the first step in that direction? To answer it, future research should be carried out and generalize our results for women and more occupational categories. Moreover, this perspective is attractive because following the theory of differential over-qualification (Frank, 1978) women are particularly vulnerable to overeducation. Finally, the theoretical model used here allows reallocations given a fixed level of inputs and outputs. In this context of overeducation, this might open some interesting perspectives for future research to make some recommendations on the educational policy of French regions.