Intergenerational social mobility refers to the relationship between the socioeconomic status of parents and the status their children attain as adults. Removing policy-related obstacles to social mobility can be advocated on both equity and efficiency grounds: enabling everyone to reach their full potential is an important catalyst of innovation and productivity (Aghion et al. 2024). Contributing to the growing body of literature (Kenedi and Sirugue 2024, Polo et al. 2019, Saez, et al. 2014), in recent work (Causa et al. 2026) we provide new evidence on intergenerational social mobility from a cross-country comparative perspective. The study relies on a harmonised individual-level dataset of the 2023 Survey of Adult Skills (PIAAC) that enables us to deliver a timely, robust, and consistent picture of intergenerational social mobility in 29 OECD countries.

Reflecting the multidimensional nature of the concept, we look at intergenerational social mobility through several angles. Parental background is proxied by the highest educational attainment of either parent, to analyse its influence on offspring’s economic outcomes such as earnings, employment and labour force participation. The analysis also explores the mechanisms of intergenerational transmission of socioeconomic outcomes, in particular the role of education, and examines heterogeneities in parental background effects across sociodemographic groups. 

Overall effects of parental background on earnings and women’s labour market participation 

Across all countries covered, parental education strongly correlates with children’s earnings: offsprings of less educated parents tend to cluster in the lower part of the distribution, while those of highly educated parents tend to cluster in the upper part (Figure 1). But this pattern is highly heterogeneous, being markedly weaker in the Nordics than, for example, in Central European countries.

Figure 1 Glass ceilings and sticky floors: Persistence at the tails of the earnings distribution

Probability of being in the top (bottom) earnings quintile for individuals with high (low) -educated parents relative to that for individuals with low (high) educated parents

Note: Sample is employees aged 30-54. Czechia is excluded due to the small sample size (of individuals with low-educated parents). How to read: in Italy, the probability of being in the top earnings quintile is 2 times higher for individuals with high-educated parents relative to those with low-educated parents; the probability of being in the bottom earnings quintile is 1.5 times higher for individuals with low-educated parents relative to those with high-educated parents.
Source: Causa et al. (2026).

A regression analysis controlling for individual characteristics confirms the importance of parental education in influencing offspring’s earnings. Figure 2 shows that relative to the middle educated parent group, children of highly educated parents enjoy an earnings premium, whereas those of less educated parents face a penalty in most countries. The premium is largest in Israel and Poland, where sons of high-educated parents earn almost 30% more than those with middle-educated parents. Estimated earnings penalties are larger for women than men in countries like Germany and Ireland, where the earnings penalty of having low relative to middle-educated parents is around 18% for women and statistically insignificant for men.

Figure 2 Earnings premium and penalty associated with parental education (per cent)

Note: The figure shows the percentage change in earnings of individuals depending on their parent’s education, based on the elasticities estimated with log earnings regressions. The sample is men and women aged 30-54. Earnings refer to hourly earnings among employees. The premium (penalty) is the increase (decrease) in the individuals’ earnings of having high-educated (low-educated) relative to middle-educated parents. The results are based on country-by-country regressions including individual control variables: age, marital status, migration background, and regional fixed effects. Estimates for women are run with a maximum likelihood Heckman estimator, to address possible self-selection biases into employment.  A solid fill denotes statistical significance at least at the 10% level.
Source: Causa et al. (2026).

The analysis of earnings is key to studying intergenerational mobility with economic lenses, but it is conditional on individuals’ labour market participation. Yet this is not always the case, especially for women, whose employment rates range from around 85% in Sweden to around 64% in Italy (as compared to 89% and 84% for men, respectively).
Figure 3 shows that women’s participation in the labour market is significantly lower when parents, especially mothers, are low educated. Estimated effects are largest for Germany and for the US, where participation probabilities are around 17 percentage points lower for daughters of low-educated mothers relative to those of middle-educated mothers. Penalties associated with having a low-educated father tend to be significant but quantitatively smaller for most OECD countries.

Figure 3 Women’s labour market participation: The impact of fathers’ and mothers’ education

Labour market participation penalty associated with having a low-educated mother or father, percentage points

Note: Marginal fixed effects from probit estimates. The sample is women aged 30-54. Participation probability penalty from having a low-educated mother (father) relative to a middle-educated mother (father). Estimations include individual controls (age, marital status, presence of children aged under 6, migration background and regional fixed effects. Detailed regression results are reported in (Causa, Nguyen and Tanaka 2026), the online Annex A1, including estimated premia of having high-educated parents, not reported here because they tend to be weakly significant. The online Annex also provides additional complementary estimates with women’s’ employment as dependent variable.
Source: Causa et al. (2026).

The role of education in transmitting socioeconomic outcomes across generations

To shed light on the mechanisms of intergenerational social mobility with a focus on the role of human capital, we expand the econometric analysis of individuals’ labour market outcomes to include individuals’ own educational attainment, thereby estimating the direct effect of parental education on outcomes, conditional on the offspring’s own education. The results suggest that in most countries covered, education and skills development are key mechanisms of social mobility across generations, insofar as estimated parental background effects on individual earnings become either significantly smaller or no longer statistically significant when own education is accounted for. This is especially the case among women. Yet importantly, while estimated effects diminish, they often remain significant: even when individuals study at similar levels and fields of education, parental background continues to exert a significant influence on offspring’s economic outcomes, albeit reduced and of different magnitude across countries. 

These findings point to the role of educational policies to enhance intergenerational social mobility, in particular by allowing individuals from disadvantaged backgrounds to climb the social ladder. Indeed, simple correlation analysis shows that countries that spend more on childhood education and care, and, more generally, on family transfers, feature higher upward mobility, measured on the basis of estimated education and earnings penalties associated with coming from a low-educated family (Figure 4).

Figure 4 Early childhood interventions and public spending on family transfers and intergenerational social mobility

Note: The mobility metrics refer to the standardized indicators defined above, ranging between 0 and 1. Public spending on ECEC is measured as of 2000, and public spending on family transfers as average between 2000 and 2006, in order to tentatively match the period where the cohorts used in the estimation sample were children. 
Source: Causa et al. (2026).

Given the finding that education is a major mechanism of intergenerational social mobility, the final part of the empirical analysis sheds light on the link between parental and offsprings’ education. This is investigated by estimating the change in the probability of achieving a tertiary level of education given parents’ education. The results show that achieving tertiary education is positively associated with growing up in a highly educated family and negatively associated with growing up in a less educated family. Yet educational persistence is heterogeneous across countries: the Nordics are found to be most mobile; Poland, Portugal, and Italy the least mobile (Figure 5).

Figure 5 Probability premium and penalty to achieve tertiary education depending on parental education, percentage points

Note: Probit estimates. The chart reports marginal fixed effects estimates associated with parental education, that is, the percentage point difference in probabilities to achieve tertiary education between individual’s grown-up in high/low-educated families relative to those grown up in middle-educated families. Regressions control for sex, migration background, age and region of residence. All estimates are statistically significant (at least at the 10% level). Detailed regression results are reported in (Causa, Nguyen and Tanaka 2026), Annex A1 online. 
Source: Causa et al. (2026).

Policy considerations

The evidence in our study points to the key role of education as a driver of intergenerational social mobility. But the results show that this is not the whole story: achieving higher education is not always enough to narrow the economic gap (here measured in terms of labour market outcomes) between individuals coming from less and more advantaged family backgrounds. This is likely to reflect the interplay between several obstacles to upward mobility, such as accessing most rewarding educational institutions and/or fields of study, a likely reflection of financial constraints but also informational gaps; and accessing relevant cultural and social networks, training and mentoring opportunities that help young people enter and make progress in the labour market. While priorities will vary depending on country-specific context, educational and broader social policies to support the generation opportunities for individuals coming from disadvantaged backgrounds is a common priority. Broad pillars of effective policy interventions can be summarised as follows:

• Enhancing access to quality childcare and early childhood education, especially for disadvantaged families and regions (e.g. places experiencing weak economic dynamism along with a decline in the availability of essential public services).
• Avoiding educational and school practices that group students into different programmes or curricula according to proficiency level, for instance early tracking and ability grouping within classes. Well-designed targeting of educational and school resources, including incentives to attract qualified and experienced teachers in disadvantaged areas and schools, can support equity and performance in the education system.
• Stepping-up policies to support youth education and training choices and transitions from education to work, considering the evolving nature of labour demand in the context of digital and green transitions. This requires training and labour market policies to provide young people with the right skills and to reap the benefits of innovation. Quality vocational and apprenticeships systems that support the matching between individuals’ qualifications and firms’ demand for skills can play an important role in this regard.
• Using competition and framework policies to support business dynamism and to reduce barriers to entry for new firms as important tools for improving social mobility, complementing education and training policies. Unleashing talents from all socio-economic backgrounds can for instance support innovation by fostering a more dynamic competitive environment. A simple cross-country correlation analysis does indeed indicate a significantly positive correlation between intergenerational mobility and various metrics of business dynamism (Causa, Nguyen and Tanaka 2026), providing tentative support for the relevance of competition and innovation to support social mobility, and vice-versa.
• Removing barriers to geographical mobility for prospective movers to reap better economic opportunities. This includes mobility-friendly social benefits, for instance in terms of housing. Place-based policies are also needed to support “stayers”, including ensuring the availability of adequate public transport infrastructure to enhance connectivity between predominantly rural areas and more urbanised ones (see work on labour and inter-regional mobility summarised in Luu et al, 2021).

References

Aghion, P, R Blundell, and X Jaravel (2024), “Innovation and Social Mobility: Two sides of the same coin”, Social Mobility Commission 2024.

Causa, O, M Nguyen, and T Tanaka (2026), “Intergenerational social mobility across OECD countries: Does the apple fall far from the tree?”, OECD Economics Department Working Papers No. 1858.

Kenedi, G and L Sirugue (2024), “Intergenerational income mobility in France: A comparative and geographic analysis”, VoxEU.org, 12 April.

Luu, N, O Causa, M C Cavalleri, and M Abendschein (2021), “The laws of attraction: Economic drivers of inter-regional migration, the role of housing, and public policies”, VoxEU.org, 11 December. https://cepr.org/voxeu/columns/laws-attraction-economic-drivers-inter-regional-migration-role-housing-and-public

Polo, A, G L Violante, and P Acciari (2019), “And yet, it moves: Intergenerational mobility in Italy.” VoxEU.org, 13 July.

Saez, E, N Hendren, P Kline, and R Chetty (2014), “Where is the land of opportunity? Intergenerational mobility in the US”, VoxEU.org, 4 February.