For decades, researchers have known that family background plays a key role in shaping one's future opportunities, including educational trajectory and career choices. Yet, despite a large existing literature on intergenerational mobility, or the extent to which economic status is transmitted between generations, one question continued to bother us: how does intergenerational mobility evolve over time, and what are the broader consequences of these changes for society and the economy?
The problem was not a lack of interest. Rather, it was a lack of data. Most existing measures of intergenerational mobility are constructed either cross-nationally or lack a time dimension, and therefore provide snapshots rather than continuous histories. This makes it difficult to study how changes in mobility at the sub-national level unfold over time or how they relate to long-run regional development.
Following a broader research agenda that Guido Neidhöfer began in 2014, with an interest in understanding how social mobility and equality of opportunity had evolved in developed and developing countries, he and Sarah McNamara first began this project in the summer of 2020, with Patrick Lehnert joining the research team after a conference discussion in 2021. As we became increasingly interested in the relationship between opportunity and innovation, a key driver of economic development, we realized that answering these questions would first require building a new type of dataset.
That realization marked the beginning of what eventually became the EUROPE-IGM-ATLAS.
What began as a methodological exercise soon evolved into a much larger undertaking. In the first stage of the project, we assembled harmonized microdata covering dozens of countries and hundreds of regions. One of the biggest challenges we faced was harmonizing information across countries with very different education systems, historical records, and changing regional boundaries. Nevertheless, with time and perseverance, we produced measures of intergenerational educational mobility, inequality, and average education for cohorts born 1940-1999 in twenty-year increments.
The second stage of the project started with its own methodological challenges. First, measures of intergenerational mobility are inherently associated with specific generations of adults and the birth cohorts of their adult children, while most economic outcomes of interest, including innovation, productivity, and other indicators of regional development, are typically observed annually. Bridging the two is not straightforward.
To resolve this, we transformed the cohort-based mobility measures into annual indicators of "effective" mobility from 1985 to 2025 that reflect the educational mobility embodied in the active population at a given point in time. We do so using five separate weighting profiles linked to innovator life-cycles, employment, and income. That is, we ask, "What level of mobility characterizes the economically active population in this year?"
Second, we faced a challenge familiar to many researchers working with regional data: the lack of consistent measures of economic growth across European regions over long time horizons. While innovation can be measured annually through patent records, comparable data for regional GDP is unavailable for our period of observation. Here Patrick Lehnert’s previous work offered a solution: he had co-developed a proxy for local economic development across both time and space, derived from daytime satellite imagery. This data allowed us to account for long-run differences in regional trajectories, helping us isolate the relationship between mobility and innovation from the broader geographic and economic conditions in which both evolve.
For the first time, it became possible to examine how educational opportunity has evolved across European regions, and to use these indicators to understand how opportunity is linked to other measures of regional development.
One of the most striking findings emerged when we began visualizing the data. Educational mobility has generally increased across Europe, but the gains have been far from uniform. Improvements are concentrated in particular clusters of regions, while others experienced much more modest change. Looking at maps of Europe over time made it clear that opportunity has a geography.
The next question was whether these differences matter beyond individual outcomes.
A growing body of research has suggested that barriers to opportunity may reduce efficiency by preventing talented individuals from obtaining educational qualifications and occupational positions that best match their abilities. If this is true, aggregating the consequences of this mismatch should imply that regions with greater educational mobility are better positioned to utilize talent. Yet, at the outset of this project, direct evidence linking mobility and innovation over time remained scarce.
Using patent data, we were able to investigate this relationship at an unprecedented scale. What surprised us most was not simply that mobility and innovation were positively related, but that the relationship appeared consistent across a highly diverse set of European regions. Even after accounting for differences in average educational attainment, regional economic growth, and time and country-specific effects, we found that regions with a higher level of intergenerational educational mobility generate more innovation.
Importantly, our findings do not imply that mobility is the sole driver of innovation. Innovation ecosystems are shaped by many factors, including universities, firms, institutions, and local economic conditions. Nevertheless, the results suggest that educational mobility may play a previously underappreciated role.
More broadly, the project demonstrates the value of the new database we built as part of this project. The EUROPE-IGM-ATLAS was created as a resource for the research community, and we hope it will be used to study a wide range of questions extending well beyond innovation. Intergenerational educational mobility influences many dimensions of social and economic life, and we hope the atlas opens up new possibilities for understanding how opportunity shapes development. You can find the current version of the EUROPE-IGM-ATLAS at: https://europe-igm-atlas.github.io/app/.
One of the most rewarding aspects of this project has been the opportunity to provide quantitative evidence on the importance of social mobility and equality of opportunity for society, beyond their ethical value, by demonstrating their contribution to innovation. Furthermore, the EUROPE-IGM-ATLAS is not the conclusion of the research, but the beginning of a much broader research agenda that we also hope to see taken up by others.