Estimating returns to education using the genetic lottery.

Area of Science:

• Economics
• Genetics
• Sociology

Background:

• Determining the economic returns to schooling is crucial for policy and individual outcomes.
• Observational studies face confounding, while prior quasi-experimental methods focus on limited age ranges.
• Genotyped data and Mendelian randomization (MR) offer novel approaches to address confounding and estimate returns across all ages.

Purpose of the Study:

• To estimate the causal effect of an additional year of schooling on lifetime earnings.
• To utilize genetic variants as instrumental variables for education to overcome unobserved confounding.
• To triangulate findings using multiple identification strategies for robust estimation.

Main Methods:

• Mendelian randomization (MR) using genotyped data and Norwegian population registries.
• Ordinary least squares (OLS) with covariate adjustment.
• Sibling and twin fixed-effects models.

Main Results:

• MR estimates an 8.0% return to schooling (N=109,800), with sibling-MR yielding 6.3% (N=18,666).
• MR results remain robust under sensitivity analyses for potential pleiotropy.
• Higher returns are estimated via MR compared to OLS (5.9%) and fixed-effects models (5.3% siblings, 3.2% twins).

Conclusions:

• Lifetime returns to schooling are positive and substantial.
• The internal rate of return to education surpasses market interest rates.
• MR provides reliable estimates of the causal impact of schooling on earnings.