Social Mobility Elasticity Calculator

What Is the Social Mobility Elasticity Calculator?

The Social Mobility Elasticity Calculator works out the Intergenerational Earnings Elasticity (IGE) from a list of parent and child income pairs, the most widely used statistical measure of how much a person's economic position is inherited from their parents rather than determined independently. Enter at least five income pairs and the calculator runs a log-log regression to figure out the IGE coefficient, the correlation between generations, and the rank-rank correlation that Raj Chetty's Opportunity Insights research group and most contemporary mobility researchers now prefer for cross-country and cross-time comparison. No standalone IGE calculator existed publicly online at the time this tool was built, leaving researchers, students, and policy analysts to carry out the regression manually in a spreadsheet or statistical package every time they wanted an estimate.

Economists use IGE to compare how much equal opportunity a society provides, on the logic that in a perfectly mobile society, a child's adult income would be no more predictable from their parents' income than from a stranger's. Given that real economies sit somewhere between full mobility and full inheritance, the IGE coefficient quantifies exactly where on that spectrum a population, organisation, or program's outcomes fall.

How the IGE Regression Works

The calculation takes the natural logarithm of both parent income and child income for every pair, then carries out a simple linear regression of log child income on log parent income. The slope of that regression line, conventionally written as beta, is the IGE. Algebraically, beta equals the correlation coefficient between the two log-income series multiplied by the ratio of the child generation's log-income standard deviation to the parent generation's log-income standard deviation, which is why a population with rapidly widening income inequality between generations can show a rising IGE even if the underlying correlation in relative rank has not changed.

That said, the regression intercept (alpha) and the correlation coefficient (rho) both matter for interpreting the slope correctly. A high correlation with a low slope suggests a tight but compressed relationship, while a moderate correlation with a steep slope suggests fewer outliers but a stronger proportional pass-through of income advantage. Looking into both figures together gives a fuller picture than the IGE slope alone.

IGE vs Rank-Rank Correlation: Why Modern Research Prefers Rank-Rank

The rank-rank correlation converts each person's income into a percentile rank within their own generation, then correlates the child's rank against the parent's rank. Because rank-rank correlation depends only on relative position and not on the level or growth rate of income, it is not distorted the way the raw IGE can be when one generation's income distribution has grown faster than the other's. According to research summarised in Brookings Institution's primer on measuring relative mobility, this distinction matters most when comparing mobility across countries or time periods with very different aggregate income growth, which is exactly the situation where a raw IGE comparison can mislead.

Cross-Country Mobility: The Great Gatsby Curve and Its Limits

Economist Alan Krueger popularised the "Great Gatsby Curve," a scatterplot showing that countries with higher income inequality, measured by the Gini coefficient, also tend to have higher IGE values, implying lower mobility. The pattern is a documented cross-country correlation, not a settled causal claim. Critics have shown the relationship weakens or disappears when the comparison is restricted to similarly sized labour markets, and some of the same statistical drivers that raise measured inequality can mechanically raise the measured IGE as well. As a result, treat any single chart comparing inequality and mobility across countries as suggestive evidence to look into further, not as proof that one causes the other. If you want to measure income inequality directly for the population in your own dataset, alongside its mobility elasticity, our Gini Coefficient Calculator computes that companion statistic.

Accuracy and Limitations

The regression math in this calculator is exact for the pairs you enter, but the reliability of the IGE estimate depends heavily on sample size and how parental income was measured. Academic studies typically average parental income across five or more years specifically to avoid attenuation bias, the well-documented tendency for a single noisy year of parental income to bias the estimated IGE downward. A dataset built from fewer than 30 pairs, or from a single year of parental income, should be treated as a directional estimate rather than a precise figure.

IGE also only measures income-based mobility. It does not capture mobility in education, occupational status, or wealth, which can move quite differently from income alone, a limitation documented in the World Bank's Global Database on Intergenerational Mobility methodology notes. If your goal is to keep track of opportunity comprehensively, pair this calculator's output with separate measures for educational or occupational mobility rather than treating income elasticity as the complete picture.

The Most Common Social Mobility Reporting Mistake

The mistake I see most often is reporting a simple comparison of average outcomes between income quartiles and calling it a mobility measure, rather than running the full regression across the income range. Comparing only the bottom quartile's average outcome to the top quartile's average outcome can mask substantial persistence within each quartile and produces a much rosier picture than the underlying IGE or rank-rank correlation would show. With that in mind, whenever a report claims a program or population has "broken the link" between parent and child income, check whether that claim is based on a full regression across the distribution or just a quartile comparison. This turns up most often in program impact reports from scholarship funds, training programs, and economic development initiatives, where a quartile-average claim can sound far more conclusive than the underlying data supports.