Wealth inequality and household structure: a comparison between Spain and the United States

There are considerable differences between the wealth distributions of the developed countries. For example, the proportion of total wealth in the hands of the richest 1% of households, a frequently cited indicator, ranges internationally from 15% to 35%. Studying the nature of these cross-country differences may help determine the effects on wealth of various institutions, such as business regulation, welfare programmes and taxation. Moreover, it is obviously of interest for the measurement of inequality. Nowadays, the growing availability in various countries of quality data drawn from household surveys means that such comparisons are feasible. However, in the international studies available hitherto, households have usually been treated as if they were homogeneous across countries (except when studies have tried to use some equivalence-scale measure based on the number of persons in the household). This practice may be appropriate when comparing countries with relatively similar demographic structures, but in most cases the distribution of wealth is influenced by demographic factors. For example, if patterns of household formation by young persons differ in two countries, not only will the distribution of households by age differ, but so will the distribution by household size and marital status. This raises the question up to what point cross-country differences in wealth distributions also apply to comparable households and up to what point they are due to cross-country differences in household structure. Some recent studies have, in turn, considered the influence of wealth on the household structure, through marriage and divorce decisions. However, this study stresses aspects of household structure that are assumed to be associated with social values and norms. For example, the age at which young people abandon the parental home to establish their own is a key reflection of deep-rooted differences in household systems across western countries, but it is not the only one. Others are the prevalence of lone parent households and elderly persons living with their children. In this respect, the sociological literature [see Reher (1998)] identifies two clearly distinct geographical areas, one in which family ties are strong (mainly Mediterranean countries) and the other in which such ties are weak (northern Europe and the United States). In the former, children tend to leave home when they start to live as a couple and save until then, while in the latter they establish themselves independently when they reach maturity. These differences are long-standing; the earliest available data show that they date back to at least the seventeenth century. In fact, according to the first censuses, in the mid-nineteenth century in northern Europe 30-55% of young people aged 15-24 (of both sexes) left the parental home to serve in another household, while only 5-20% did so in southern Europe. In recent years there has been some convergence, but a clear division remains. This article compares the wealth distribution of Spain, a country with strong family ties, with that of the United States, where family ties are weaker. As can be seen in Table 1, the differences between these two countries, as regards summary statistics of wealth distributions, are significantly reduced when homogeneous demographic groups are compared, such as couple households whose head is aged 35-40. To identify the influence of demographic factors on the differences between the wealth distributions of Spain and the United States, the wealth distribution that would be observed in the United States if households there had the same composition as in Spain (the counterfactual distribution) is estimated, using comparable household level wealth-survey data available in both countries relating to the early 2000s.4 The next section describes the data and the household demographic structure classification adopted. The third section describes the construction of the counterfactual distribution for the United States, and presents the results of the comparisons in graphic form and using a set of summary statistics of position, dispersion and inequality. Also, the types of households that contribute most to the compositional differences are identified, and information is supplied on the differences in the wealth distributions of the two countries by comparable household groups. Lastly, the fourth section contains some final comments.