The influences exerted by the economic standing of parents upon the educational achievements and subsequent economic successes of their children have been the subject of a substantial body of recent research within CEPR’s Human Resources research programme. The issues investigated include: 1) the extent to which, in general, affluence and poverty in the United Kingdom are transmitted from one generation to the next; 2) whether a UK mother’s education is a good predictor of her children’s – especially her daughters’ – educational achievements; 3) the differences in educational levels attained by the offspring of native Germans and Turkish immigrants in Germany; and, 4) via a comparison between Italy and the United States, whether state-financed education is more effective than a privately-funded system in increasing intergenerational mobility. The fruits of this research were reported by four CEPR Research Fellows at a joint CEPR/Royal Economic Society meeting in February 1997.

Work undertaken by Stephen Machin (University College London, Institute for Fiscal Studies, London, and CEPR), in collaboration with Lorraine Dearden and Howard Reed, suggests that the degree of intergenerational mobility in UK society is still very limited. Instead, the economic standing of parents is an extremely important determinant of where their children end up in the income distribution. Whether this is because of children inheriting their parents’ abilities, however, or because of differences in children’s family backgrounds, is not clear. Nonetheless, the research – which made use of the National Child Development Survey (a continuing study which follows the changing circumstances of a sample of children, born in March 1958, as they grow up) – yielded several important findings.

The first is that rich people tend to come from high-income backgrounds. Thus over half of the individuals in the top quarter of today’s earnings distribution had parents who were also in the top quarter of their generation’s earnings distribution. Similarly, poor people tend to come from disadvantaged backgrounds. For example, in the case of two fathers from a generation ago, earning £20,000 and £10,000 respectively (in today’s prices), the results indicate that the son of the richer father would grow up to earn, on average, almost £7,000 more than the son of the poorer father. Second, these trends extend also to the occupational distribution, and to employment histories. Sons of managerial or professional fathers, for example, are almost three times more likely to end up in similar occupations themselves than are the sons of semi-skilled and unskilled manual fathers. Similarly, men whose fathers had experienced spells of unemployment in the 1960s and 1970s are twice as likely, in comparison to the average for their cohort, to have been unemployed for a year or more between 1981 and 1991.

Third, looking at those children from poor and disadvantaged backgrounds who do manage to escape from poverty in adult life, the data suggest that they tend both to be more able, and to have better educational qualifications, than the children who fail to escape. Finally, and more generally, while the research clearly indicates that people’s attainments in education and the labour market are strongly related to their parents’ performance a generation ago, it seems that today’s income distribution may well reflect the pattern of inequalities extending back three or more generations.

The impact of mothers on their children’s educational achievements was the subject of research carried out by John Ermisch (University of Essex and CEPR), in collaboration with Marco Francesconi. They made use of a unique set of data, which matches mothers and their young adult children, to study the impact of family background on young people’s educational attainments. The data was derived from the first five years (1991–5) of the British Household Panel Study, a representative sample of the UK population in the 1990s.

Ermisch and Francesconi concluded that a mother’s education is a very powerful predictor of her children’s educational achievements, particularly for daughters. For example, if the mother’s highest qualification was an O-level, then the likelihood of her child obtaining a university degree was only 12%; but if the mother was a graduate herself, the likelihood of her child also getting a degree was 67%. The research also considered the impact on a young person’s educational achievements of three variables: first, whether or not they spent part of their childhood in a single-parent family; second, their parents’ economic status; and third, whether their mothers worked, and what kind of work they did.

On the single parenthood issue, almost two in five of the young adults in the sample had spent some time in a single-parent family. This experience tended to reduce the educational attainments of young men moderately, but had little effect on young women’s education. Among men, for example, the study found that the probability of obtaining a university degree fell from 22% for those who had always lived in an ‘intact’ family, to 18% for those who had spent some time in a single-parent family. Part, if not all, of this negative effect of single-parent-family experience reflected the fewer economic resources available in such families.

Parents’ economic status also proved to be relevant in that young men whose parents were homeowners achieved higher levels of education. In addition, higher family income was associated with higher educational attainments among young women. Since mothers’ education had already been taken into account, these results suggested that financing constraints were operating to affect parents’ investment in their children’s human capital, particularly beyond A-level. There was also evidence that scarcity of resources (both of money and of time) in larger families may have lowered educational attainments.

Finally, Ermisch revealed that they had found no grounds to support the view that working mothers prejudice their children’s educational attainments. Indeed, having a working mother at age 14 may even increase the odds that the child will obtain A-level qualifications or higher. It was also found that young people who were born later in their mother’s life achieve higher educational qualifications.

According to a study of the educational achievements of children in Germany, the children of native Germans acquire markedly more education than their contemporaries from other ethnic groups, and particularly the Turkish ‘guest-worker’ generation. In terms of length of time spent in school, the Germans averaged 12.1 years, while the Turks averaged 7.6 years. Moreover, 47% of the Germans obtained at least a high-school degree, compared with only 6% of the Turkish children. In terms of vocational training, the pattern of degree completion was repeated, with 45% of Germans and 17% of Turks receiving such training, and other groups – Spaniards, Greeks, Italians and Yugoslavs – falling in between.

These were some of the research findings presented by Klaus F Zimmermann (SELAPO, Universität München and CEPR), who drew on the results of a study carried out jointly with Ira Gang. Zimmermann noted that the German guest-worker programme in the 1960s had attracted many workers who became permanent migrants. Since the parent migrants were mostly blue-collar workers, the educational attainments of their children – who mostly still retained their national status – were of interest.

Zimmermann argued that the inter-group variations in educational outcomes were not simply a matter of discrimination, in that the achievement of children in schools is subject to a number of influences, including parents’ educational attainment, ethnicity, culture, gender and competition for school places. For Germans, for example, parents’ educational background played a significant role, with the father’s education being more important than the mother’s in influencing children’s educational outcomes. Parents’ education had a positive impact on both the total years of education and the level of schooling achieved. There also appeared to be a bias against vocational training since the children of highly-educated German parents generally chose forms of education other than vocational training.

By contrast, for second-generation immigrants, parental education was not a good proxy for parental influence. Migrants’ education levels bore little relationship to the educational attainment of their children. Instead, it could be argued that these parents had made their human capital investment in their children through their decision to emigrate. Moreover, although second-generation immigrants have an equal start in the German educational system, large differences nevertheless remained in human capital formation across ethnic groups. Some assimilation was evident across generations, but it was still far from being complete.

The significance of these findings lies in the fact that achievement of parity in educational attainment is a central issue in the intergenerational performance of immigrants and their families. This is especially important in economies like Germany, where a very substantial weight is placed on the formal aspects of educational degrees. Without education equivalent to that of the German system, the integration of immigrants into the German economy is rendered substantially more difficult. Hence, a higher level of school education and vocational training is a safeguard against unemployment in the German labour market. Even so, the research suggests that educational qualifications are much less rewarding for foreign workers and their children than for native Germans.

If one of the goals of a state-financed education system is to promote equal opportunity for social mobility in order to reduce the waste of talents among low-income groups, then the Italian schooling system has failed to achieve this outcome. The centralized and egalitarian structure of education financing in Italy has indeed offered a substantially similar quality of education to both rich and poor families. Despite this offer of equal opportunities, however, in comparison to the United States, Italy displays lower intergenerational mobility, in terms not only of occupations but also in terms of education levels. This is the key conclusion from research conducted by Andrea Ichino (IGIER, Università Bocconi and CEPR) together with Aldo Rustichini and Daniele Checchi.

Ichino reported that, in the United States, if a father has a college degree, then the probability of his child graduating from college increases six-fold; in Italy, the probability increases 25-fold. And, if the probability that an American son will be in a high-income group depends more on his education level than on his family background, the opposite is true in Italy. It seems that for Italians, it is better to ‘choose’ the right family than to obtain a college degree. Not surprisingly, therefore, Italy has one of the lowest shares of college graduates among OECD countries.

Ichino’s explanation for this failure is that, precisely because it offers the same quality of education to everybody, the Italian public school system prevents parents from investing in the education of their children according to their expected talent. In addition, in a country in which family networking in the labour market matters, uniform provision of education takes away from talented children of low-income groups an instrument to signal their talent and to overcome the networking advantage of children in high-income groups.

For both reasons, the Italian public education system, while reducing the cost of educational investment, also reduces the return for low-income groups. Ichino’s suggested solution to this problem is that reform of the Italian public education system should be aimed at preserving a uniform minimum quality of primary education for everybody. But, at the same time, it should be aimed at increasing the quality differentiation at higher levels of education, in order to generate the correct incentives for individual investment in education.

Stephen Machin, Lorraine Dearden and Howard Reed, ‘Intergenerational Mobility in Britain’, Economic Journal, January 1997