EconTalk

James Heckman on Facts, Evidence, and the State of Econometrics

16 snips
Jan 25, 2016
Nobel Laureate James Heckman shares insights on econometrics and the importance of addressing selection bias for accurate economic measurement. He discusses how labor statistics for minorities can be distorted by incarceration rates and how education shifts affect wage trends for women. Heckman critiques randomized control trials, emphasizing that while they can reduce bias, they also obscure economic interpretations. He highlights the necessity of incorporating economic theory into data analysis, advocating for humility in the face of complex empirical findings.
Ask episode
AI Snips
Chapters
Books
Transcript
Episode notes
INSIGHT

Selection Bias Forced Better Data And Theory

  • Selection bias forced economists to collect better data and incorporate economic choice into analysis.
  • Recognizing self-selection links econometrics to economic behavior and improves inference.
ANECDOTE

How Incarceration And Composition Skew Stats

  • Heckman cites incarceration removing low-wage black men from labor statistics and biasing observed wage gains.
  • He also notes female labor-force growth concentrated among higher-educated women, altering measured wage trends.
INSIGHT

Definitions Change Measured Inequality

  • Measurement units matter: comparing taxpayer units to households gives different inequality pictures.
  • Hold composition constant to avoid misleading time comparisons of inequality or incomes.
Get the Snipd Podcast app to discover more snips from this episode
Get the app