The Relative Effect of Family and Financial Characteristics on Educational Achievement

Britain is characterised by a low rate of post compulsory schooling compared to other European countries. To reduce this disparity, the British government has been testing an Education Maintenance Allowance (EMA) where 16 to 19-year olds are given financial support to attend schooling when the family income falls below a threshold. This paper attempts at first separating family and income effects and second estimating the impact of a financial transfer on educational attainment. Children from poorer backgrounds are generally observed to have lower educational outcomes than other youth. However, the mechanism through which household income affects the child's outcomes remains unclear. Either, poorer families are financially constrained which prevents them from investing in the human capital of their offspring, thus, policies of financial support could be efficient at reducing schooling inequality. Or, some parents may be endowed with characteristics that make them less successful on the labour market and worse at parenting. Then, direct support to the children would be more efficient than financial support at reducing inequality in schooling achievements. We propose a methodology that separates financial and familial effects. By simulating a financial support policy, we are able to maintain the observed and unobserved characteristics of the family constant, and estimate the direct effect of an education benefit on post compulsory schooling decisions. As in previous studies, we find that pupils from poorer families are less likely to invest in education. However, a financial transfer would not lead to a significant increase in schooling investment, which supports the view that the family characteristic effects dominate the financial constraint effects. These results may nevertheless be dependent on our methodology. Due to data constraints, we have allocated the financial transfer to the father income, whereas in the currently tested scheme, the EMA goes either to the child or t o the mother. These changes in the recipient of the allocation may have a large effect on our conclusion.