Centre for Economic Performance,1998:1-5The Demand for Post-Compulsory Education in Four European CountriesSteven McIntoshPublished by Centre for Economic Performance London School of Economics and Political ScienceAbstractThis study seeks to explain changes in the proportion of 16 year olds, 17 year olds and 18 year olds who decide to participate in post-compulsory education. Four countries are considered;Germany, the Netherlands, Sweden and England, over as much of the period from 1960 to the present day as data will allow. An initial inspection of the data reveals non-stationarity in virtually all of the series under consideration, so that a cointegration framework is appropriate.The analysis therefore proceeds using the two-stage procedure proposed by Engle and Granger.The results suggest that the key variable that explains the growth in post-compulsory education participation is the increase in prior academic attainment before the end of compulsory schooling.This is the case particularly for females, males being also influenced by the returns available to offering a higher level of education, and the level of real income available to spend on education. The level of youth unemployment seems to play only a small part in the decision as towhether to remain in education.Key words Post-Compulsory Education; Educational needs1. IntroductionThere follows an attempt to explain movements in the rate of participation in post-compulsory education in four European countries, Germany, the Netherlands, Sweden and England, during the last twenty to thirty years The motivation for the study is to seek explanations for the decision to acquire skills. The rise in demand for skilled labour has been well documented, and it is important to acquire evidence as to whether the supply of skilled labour has also increased, and if so, why.The very low-skilled group are becoming increasingly marginalised in modern labour markets. Evidence has shown that both the employment levels and earnings of the least well educated are declining relative to those of the most well educated. This is the case particularly in the UK and the US, but is also true to a lesser extent in almost all OECD countries (see Glyn,1995). It would therefore appear to be clear that, although the proportion of unskilled individuals in the workforce of all countries is falling, it is not falling fast enough to match the declining demand for such individuals. The implication is that young people leaving education at the end of compulsory schooling with no or only low-level qualifications have fewer and fewer opportunities. This does indeed seem to bethe case, as described by Blanchflower and Freeman(1996). On the assumption that there is sufficient demand for more skilled workers, it would therefore seem to be important for more young people to obtain a higher level of skills, beyond the basic level. Of course, this is no guarantee of success in the labour market, but increasingly it seems to be a pre-requisite for at least having a chance of succeeding. For these reasons, it was decided to focus on this basic level of skill acquisition. In particular, the study is an examination of the factors which influence the decision of young people to remain in education beyond the age at which it is compulsory for them to do so, and to acquire a higher level of skills through this education, before entering the labour market. The analysis considers separately 16 year olds, 17 year olds and 18 year olds.The following section describes previous studies that have investigated such issues. The variables to describe the changing participation levels in post-compulsory education are then outlined, as well as the estimation methodology to be adopted. The results of the analysis are then provided in full, and a final section concludes.2. Previous StudiesPrevious attempts to model educational choices in the UK include Pissarides (1981), and Whitfield and Wilson (1991). These two papers differ from each other, in that the earlier paper by Pissarides adopts a straight forward logistic regression, while the latter paper performs a similar logistic transformation, but within the more modern framework of cointegration analysis. Whitfield and Wilson also extend the list of explanatory variables used by Pissarides. Pissarides considers real consumption expenditure per capita, earnings of youth workers, starting salaries of graduates, the growth in adult manual earnings, the growth in adult non-manual earnings, the adult unemployment rate and the percentage of new graduates still seeking employment six months after graduation.The rationale for entering most of these variables should be clear. Per capita consumption expenditure is included to allow for the fact that education is expected to be a consumption-good,and so demand for education should rise as the general level of consumer spending rises. The other variables are included to allow for the in