Family Background, Education and Achievement: A Path Model of Earnings Determinants in the U. K. and Some Alternatives

Abstract

This paper uses individual data on 6,873 male employees from the I972 General Household Survey in the U.K. to assess the extent to which differences in personal characteristics explain differences in occupational and economic success. This is done by means of a path model fitted on the following key variables: father's occupation, respondent's occupation, education, ability and earnings. In addition, alternative specifications of the same basic model are fitted involving age or experience as an additional variable and different proxies for the ability variable. The results indicate that personal characteristics explain about onethird on the variance in earnings in the U.K. and that education is an important contributor to economic and social ascent. Although this finding is in slight variance with the conclusion of a similar study by Jencks using U.S. data, it is noted that the differences in results might be due to our inadequacy in matching the exact definition of variables for the purpose of international comparisons of the kind attempted here. 'Neither family background, cognitive skill, educational attainment, nor occupational status explains much of the variation in men's incomes.' This was Jencks' conclusion after an analysis of the determinants of adult success in America that is now widely cited around the world.2 This paper reports the results of a similiar analysis for the U.K. Using data on individual characteristics of nearly 7,ooo men, I show that personal characteristics explain a higher fraction of earnings variance in the U.K. than in the U.S., and that education has a sizeable direct effect on income as well as an indirect effiect through occupation. Section I briefly describes the data and defines the basic variables used in the analysis. Section II presents a path model of the determination of earnings a la Jencks. After discussing the limitations of this model, section III uses alternative multivariate techniques and different specifications of the same basic Grariables to answer the same This content downloaded from 157.55.39.162 on Thu, 11 Aug 2016 05:43:09 UTC All use subject to http://about.jstor.org/terms George Psacharopoulos 322 central question, namely, what is the power of personal characteristics in explaining the variance of earnings and, in particular, what is the role of schooling controlling for other factors ? I also try to answer some peripheral questions on the role of ability in the earnings determining process and the existence or not of an interaction effect between ability and education. Section IV examines the extent to which social mobility relates to schooling, the dependent variable now being the social distance travelled between father's and son's occupational status. SectionV discusses the policy conclusions stemming from the empirical results.