Rags or Riches? Predicting Life Outcomes from the Birth Lottery Across Five High-Income Countries

• In Event: Inequalities in Income, Wealth, and Related Outcomes

Abstract

Scholars have long examined how the “birth lottery” shapes adult outcomes and how intergenerational mobility varies across countries. While prior research has identified key family and socioeconomic factors associated with life chances, explanatory models typically show limited predictive power. This study uses an ensemble of machine learning algorithms applied to harmonized longitudinal data from the Cross-National Equivalent Files (CNEF) to assess how well early-life circumstances can predict adult outcomes in five countries: the United States, United Kingdom, Germany, Australia, and South Korea. Out-of-sample predictability is modest but meaningful, with R² values up to 0.35 (though frequently much lower). Predictability varies substantially across both outcomes and countries: poverty is most predictable in the United States, top-decile attainment is most predictable in Germany, and upward mobility is the least predictable everywhere. In additional analyses with the U.S., adding a richer set of variables such as race, wealth, and parental incarceration improves predictions only slightly. Moreover, we show that predictability is a distinct dimension of stratification from traditional measures of mobility, inequality, and opportunity. Together, these findings provide new insights into where and how the birth lottery seems to most strongly determine life chances.

Authors

• Orestes Pat Hastings, Colorado State University

• Zachary Parolin, Bocconi University