WHY WHO MARRIES WHOM MATTERS FOR INTERGENERATIONAL INEQUALITY

‘Assortative mating’ – where ‘like marry like’ – is a major factor in the intergenerational transmission of economic status, according to new research by Professor John Ermisch and colleagues, published in the July 2006 Economic Journal. Their study finds that in both Britain and Germany, just under half of the link between parents' and children's incomes can be explained by 'sorting' in the marriage market.

The evolution of inequality over generations depends on more than just one’s own earnings; it depends on who marries whom. Thus, having well-off parents can both increase a child’s earnings, and help to attract a high-earning husband or wife.

If marriage paired off people randomly, it would greatly increase intergenerational economic mobility, because many rich children would marry poor partners. But there is substantial evidence that ‘like marry like’.

The main contribution of this study is to estimate the extent to which such ‘assortative mating’ (or ‘homogamy’) affects intergenerational economic stratification in Britain and Germany. In particular, it estimates the proportion of the correlation between parents’ and someone’s combined family income that is contributed by the correlation between parents’ and his or her partner’s income.

Using data from the British Household Panel Survey (BHPS) and the German Socio-Economic Panel (GSOEP), the research finds that assortative mating plays an important role. On average, 40-50% of the correlation between parents’ and own family income can be attributed to the person to whom one is married. This effect is driven by a strong correlation between spouses in their ‘human capital’, such as education.

For centuries, marriage has been one of the primary institutions through which social stratification has been maintained. The development of a national marriage market in London and Bath in the second half of the eighteenth century widened the pool of potentially satisfactory spouses from the point of view of upper-class parents, because it increased the number of potential spouses who would meet the necessary financial and social qualifications.

Two centuries later, research looking at a group of working-class families in London in the 1940s reported high and significant correlations between husbands and wives on a number of characteristics, including age at marriage, stature, social background and pre-marital experience of sex. Thus, there is evidence of selective mating that is likely to affect the correlation between one’s own socio-economic position and that of one’s parents.

This study compares the experiences of two countries, Britain and Germany, and also compares its findings with those of a similar study of the United States. At least three sets of institutions may produce differences in intergenerational mobility and assortative mating between countries:

• First, the educational system, which is not only likely to influence intergenerational mobility directly, but also through assortative mating. For example, Germany’s early separation of children into academic streams between the ages of 10 and 14 could produce more assortative mating than in Britain.
• Second, a country’s labour market institutions affect the returns to investment in human capital and gender differences in those returns. These in turn affect intergenerational mobility.
• Third, families differ in the weight given to the next generation’s income prospects in their decisions, and there could be differences in the average weight between countries.

In the approach taken in this research, there are two important correlations: one between parents’ own long-run (‘permanent’) income and that of their children; and the other the correlation between the parents income and that of their sons- or daughters-in-law. Estimation of these requires data on the socio-economic position of individuals, their partners and their parents.

For men and women in both countries, the research finds that 40-50% of the correlation between their own permanent family income and that of their parents can be attributed to ‘sorting’ in the marriage market. So with just under a half of one’s social position attributable to the process of who marries whom, assortative mating appears to be a major factor in the intergenerational transmission of economic status.

Germany has a higher degree of assortative mating. For example, the correlation between partners’ years of education is about 0.52 in Germany and 0.41 in Britain. Similarly, almost 50% of German couples have equal educational levels, while the equivalent figure is only 30% for British couples.

ENDS

Notes for editors: ‘Intergenerational Mobility and Marital Sorting’ by John Ermisch, Marco Francesconi and Thomas Siedler is published in the July 2006 issue of the Economic Journal.

The authors are at the Institute of Economic and Social Research (ISER) at the University of Essex.

For further information: contact John Ermisch on 01206-872335 (email: ermij@essex.ac.uk); or Romesh Vaitilingam on 0117-983-9770 or 07768-661095 (email: romesh@compuserve.com).