Dearden, Lorraine; (1995) Education, training and earnings in Australia and Britain. Doctoral thesis (Ph.D), UCL (University College London).

Text out.pdf Download (8MB)

Abstract

One of the key topics in the empirical education and training literature at the present time concerns how one correctly measures the causal impact of education and training on earnings. The central issue concerns how one appropriately controls for the fact that education and training outcomes are not randomly assigned across the population. Education and training outcomes are endogenous, being the result of individual choices, attributes and circumstances. Estimates of the returns to education and training which ignore this endogeneity may be biased. These biases arise because of correlation between unobserved individual characteristics such as ability or family attributes which determine education and training outcomes as well as wages. This thesis uses panel data from the Australian Longitudinal Survey (ALS), Australian Youth Survey (AYS) and British National Child Development Survey (NCDS) to estimate the economic returns to different types of education and training in Australia and Britain. These particularly rich data sets allow us to directly compare the advantages and disadvantages of the different estimation techniques which been devised to deal with the endogeneity of education and training. In particular we compare instrumental variable, fixed effect and proxy methods. Instrumental variable techniques require us to identify at least one variable which affects education or training, but not wages controlling for education and training. Fixed effect methods assume that the unobserved individual attributes are fixed (over time or within families) and use econometric models which difference out this fixed effect. Proxy methods require access to data which has explicit proxies for things like ability or family attributes. The data used in this thesis allows us to exploit each of these techniques. We find that education and training confer significant wage advantages on individuals. The actual size of the estimated returns, however, depends on the estimation procedure used. The results we obtain suggest that standard estimates which do not correct for the endogeneity of education and training generally underestimate the returns to education and training for both men and women in Australia and Britain. The thesis also examines gender wage differentials and looks at the role education and training plays in explaining these differences. We find that gender wage differentials generally decrease with education. We also find that part of the observed difference in the wages of men and women is due to the fact that men receive more work related training than women once in work.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as XMLJSONCSV

Downloads by country - last 12 months

Export as CSVJSONXML

Archive Staff Only

View Item