Estimating Marginal Returns to Education

Area of Science:

• Economics
• Econometrics
• Education Policy

Background:

• Estimating the marginal returns to college education is crucial for understanding its economic impact.
• Existing instrumental variables (IV) methods often rely on strong, untestable assumptions.
• Policy-induced enrollment changes offer a potential avenue for more robust estimation.

Purpose of the Study:

• To estimate the marginal returns to college for individuals induced to enroll by policy changes.
• To develop and apply methods that relax strong assumptions common in IV estimation.
• To compare these new estimates with traditional IV estimates of the return to schooling.

Main Methods:

• Utilizing economic theory to inform the estimation strategy.
• Applying local instrumental variables (LIV) estimators.
• Analyzing data on individuals induced to enroll by marginal policy changes.

Main Results:

• Marginal returns to college are estimated using a novel approach.
• Returns are found to be higher for individuals with unobservable characteristics favoring college attendance.
• The study highlights differences between LIV and traditional IV estimates.

Conclusions:

• The LIV approach provides a more nuanced understanding of college returns.
• Individual heterogeneity in unobservables significantly influences educational returns.
• Policy evaluation of educational interventions requires careful methodological consideration.