Escalator

Your parents’ income and wealth is of ever-growing importance in determining your own position in the lifetime income distribution.

Intergenerational mobility is a subject of a large body of literature in social science.

Key findings

National estimates of intergenerational earnings mobility

  • Previous work has shown that intergenerational income mobility in England was lower for those born in 1970 than those born in 1958. Using administrative data on the most recent birth cohorts for whom earnings data are available, we find no evidence of recovery from that decline. National estimates of mobility of cohorts born in the late 1980s looks very similar to those of the 1970 cohort and education inequalities continue to be the dominant mediator.
  •  However, these estimates paint an incomplete picture for two reasons.

Differences in income mobility across areas and across ethnic groups

  • First, there is a significant amount of heterogeneity in income mobility within England. We find important differences in mobility rates across different neighbourhoods and different ethnic groups for young people from disadvantaged backgrounds. Due to data limitations, our analysis is restricted to studying upward mobility among individuals who grew up in disadvantaged households. For men, the difference between the least and most mobile groups (Black Caribbean and Indian, respectively) is around 15 percentiles – equivalent to £8,000 or 54% of mean earnings at age 28 for men who were eligible for free school meals (FSM) in these cohorts – while between the most and least mobile Local Authorities (LAs; Nottingham in East Midlands and Havering in London), it is 17 percentiles. The difference between White and Indian women (the least and most mobile ethnic groups, respectively) is also 17 percentiles (equivalent to just over 82% of mean earnings at age 28 for FSM women in these cohorts); this is also the gap between Kingston upon Hull in Yorkshire and Redbridge in London, which are the least and most mobile LAs for women.
  • More research is required to understand what is driving these differences. While, nationally, education is a key mediator of intergenerational persistence, differences in educational attainment are not the dominant force driving differences in mobility between geographical areas or ethnic groups. While there are large differences in educational mobility across ethnic groups, these do not follow the pattern of earnings differences. While the educational attainment of ethnic minorities growing up in families eligible for FSM is often higher than that of their White majority peers, their earnings outcomes show no such advantage. Furthermore, education accounts for only a quarter of the variation in mobility across LAs for men, and just under a half of the variation across LAs for women. These gender differences are stark: for women, education can account for twice the amount of differences in upward mobility across LAs compared with men. Looking beyond education, our analysis points to several specific groups of local area characteristics that warrant further investigation in future research, including local labour market conditions and inequality (for women) and ethnic make-up and immigration status of the local population (for men).

The growing importance of ‘unearned’ sources of income

  •  Second, existing research on intergenerational mobility has tended to focus on income from earnings. We find that ‘unearned’ sources of income, here meaning returns to wealth and wealth transfers received, look to be an important driver of the persistence across generations in total lifetime income (i.e. the sum of earnings and returns to wealth and wealth transfers received). These ‘unearned’ sources of income are likely to be of growing importance over time.
  • Rank–rank measures of intergenerational wealth persistence for the UK are substantially higher than that for earned income and only around half of intergenerational wealth persistence at age 30 can be explained by differences in education and earnings between those with more or less wealthy parents. This implies that other factors are important in driving intergenerational wealth persistence and these factors are also likely to drive intergenerational persistence in lifetime income.
  • In addition to the intergenerational persistence in education and earnings, the intergenerational persistence in wealth is driven in part by those with wealthy parents saving their income at a higher rate. But also important is that those with wealthy parents attain higher returns to wealth and receive larger amounts of intergenerational financial transfers (including both gifts and inheritances).
  • While inheritances are much larger in value than intergenerational transfers made during life, the latter may have important indirect impacts on career and investment decisions. Indeed, over half of the flow of during-life transfers is reported as being used for property purchase.
  • The growing size of household wealth is resulting in a rise in inherited wealth as a share of national income and this trend is set to continue. Consequently:

     

    • Reliance on earned income as a proxy for total lifetime income is resulting in a growing downward bias in the national rank–rank persistence estimates. Once accounting for the contributions of inheritances to total lifetime income, lifetime income mobility is likely to continue to decline between the 1970s and late 1980s cohorts but earned income mobility estimates will increasingly mask this decline.
    • Differences across ethnic groups and geographical areas in wealth and intergenerational transfers will be of growing importance in driving overall lifetime income mobility. The differences in parental wealth between those whose parents live in London and the South East and those whose parents live elsewhere has grown across generations, and rates of parental homeownership are much lower among Black young people than for other ethnic groups. All else equal, the rising importance of wealth should act to move those in London and the South East further up the lifetime income distribution and make it more difficult for Black people to move up the lifetime income distribution.

New directions for intergenerational mobility research

  •  Our overview provides strong motivation for important new directions in research on intergenerational mobility in the UK:
    • First, it is critical to learn more about what is driving the big differences that we find in intergenerational transmission across groups with different ‘ascribed’ characteristics, such as ethnicity and location. Without a better understanding of these drivers, it will be challenging to design policies that improve social mobility and reduce inequalities for all.
    • Second, we cannot continue to ignore the components of lifetime income not captured by earnings measures in the study of social mobility. Otherwise, we risk generating increasingly misleading estimates that significantly understate the degree of intergenerational persistence, overlooking important drivers of it, and focusing on the wrong policy levers.
    • Finally, there is significant room to improve data and methods for integrating more sophisticated measures of lifetime income that account for important unearned components.

Introduction

Intergenerational mobility is a subject of a large body of literature in social science. It focuses on the association between parents’ and children’s economic well-being and receives significant attention because it speaks to the question of equality of opportunity (Torche, 2015), which many consider an important goal for society. Intergenerational mobility is a cross-cutting theme in the IFS Deaton Review, which is touched upon in several of the core chapters, including on education, early childhood, and race and ethnicity. Therefore, our aim is not to provide a comprehensive overview of this topic but rather to spotlight some of the newer directions in intergenerational mobility research within economics driven by changes in some key trends in the recent decades, as well as by the growing availability of administrative data.

We start by providing an overview of the different ways in which intergenerational mobility has been measured, including the transmission across generations of class, occupation, education, joblessness and welfare receipt. We then narrow our focus to earned-income-based measures of intergenerational mobility, which dominate the economics literature in this area, review existing evidence for the UK and present some new estimates for England using novel administrative data. We find that that the national trend for England in intergenerational income mobility for cohorts born in the late 1980s look very similar to those of the cohort born in 1970. Sons from the most affluent families were on average around 19 percentiles higher in the adult earnings rankings than sons from the most disadvantaged families, while for daughters this figure was significantly higher at 27 percentiles. In line with existing evidence, we find that for these cohorts most of the association between income of children and their families (71%) is accounted for by education due to a combination of educational inequalities (higher educational attainment among children from better-off families) and returns to education (higher wages among those with higher educational attainment).
The remainder of this commentary focuses on considering the two important ways in which these estimates paint an incomplete picture of intergenerational mobility trends, beyond the well-documented concerns about measurement of lifetime income. These include within-country heterogeneity, as well as the role of income sources not captured by earnings.

Until recently, the intergenerational mobility literature focused on studying national trends and drawing lessons on drivers either across all individuals (but typically sons) within countries (see Blanden, Gregg and Macmillan, 2007) or from international comparisons (see Corak, 2013). Recent studies have started to show, however, that the national picture masks important and informative differences in experiences of people within countries with similar institutional settings and policy levers (Chetty et al., 2014a, b; Li and Heath, 2016). The recognition, documentation and understanding of these differences are, as such, relevant, critical for the design of effective policies, and may help uncover a richer broad understanding of drivers of social mobility. The ability to study within-country heterogeneity in intergenerational mobility in the UK context has been greatly boosted by newly available administrative data. We summarise the handful of existing studies on this topic for the UK and present new evidence using these administrative data. We demonstrate the high degree of heterogeneity in earnings mobility that exists between individuals of different ethnicities and those living in different parts of the country, which is masked by national estimates.

Another dominant feature of the intergenerational mobility literature in economics is its focus on earned income, assuming that it is a good proxy for lifetime total income and ignoring sources of lifetime income other than earnings – including returns to wealth (i.e. capital income and capital gains) and intergenerational wealth transfers received – which we refer to as ‘unearned income’. This may be because these sources of income were relatively less important in past decades. It may also reflect the fact that they are less well documented in standard economic datasets. High-quality data containing information on these outcomes, which are also linked across generations, are rare. However, in light of the dramatic rise in the ratios of private wealth to income observed over the last 50 years across many advanced economies, a focus on earned income alone may increasingly be missing an important part of the picture. In the UK, household wealth compared to national income has more than doubled over that period (Advani, Bangham and Leslie, 2021). This growing size of wealth has come alongside a growth in the capital share of income and the size of intergenerational wealth transfers as a share of national income (Alvaredo, Garbinti and Piketty, 2017). In this commentary, therefore, we explicitly consider unearned sources of income, namely returns to wealth and wealth transfers received, as potentially important components of total income. We do this by examining the data that are available on these outcomes, albeit limited, by considering persistence in wealth as potentially informative about persistence in a wider measure of income, and by making simulations of the impact of intergenerational wealth transfers on lifetime income mobility.

Clearly, the interpretation of the trends we document and the role for policy depend crucially on the mechanisms underlying the observed associations and differences between groups. We review the large literature, which considers the role of education as a driver of earned income mobility, and we provide some new estimates – leveraging high-quality administrative data on pupil attainment at different points in their schooling – on the mediating power of education in the national estimates as well as in explaining differences across groups within England. We then consider what patterns of heterogeneity in earned income mobility across groups may be able to tell us about a wider set of drivers of intergenerational mobility beyond education. Turning to drivers of ‘unearned income’ persistence, we consider the role of intergenerational wealth transfers and returns to wealth. This analysis clearly demonstrates that drivers beyond earned income, including inheritance, are playing an increasingly important role in driving persistence of lifetime income across generations. This suggests that in the current context a focus on earned income mobility and its drivers is not enough to paint a complete picture of intergenerational persistence in lifetime income. Before concluding, we present a simulation exercise that demonstrates this.

We conclude by drawing out key policy implications of the discussion and analysis presented.