Big Data's Role in Understanding Social Mobility

Overview of Big Data and Social Sciences

  • Key Policy Question: Why are children's chances of climbing the income ladder falling in America?
    • Challenges exist when relying solely on historical macroeconomic data to answer this.
    • Many changes over time complicate theory testing.

Evolution of Social Sciences

  • Historically, limited data available for policy questions in social sciences.
  • Transition from theoretical to empirical approaches in social sciences due to the availability of big data.
    • Empirical approach allows testing and refining theories using real-world data.
Examples of Big Data
  1. Government Data: Tax records, Medicare.
  2. Corporate Data: Data from companies like Google, Uber, retail sectors.
  3. Unstructured Data: Social media, news articles, etc., enabling the study of societal trends.

Raj Chetty's Contributions

  • Field Experiments: Chetty and Harvard team conducted extensive studies using large datasets linking children's adult income to parental income.
  • Key Results:
    • Comprehensive tax return datasets tracking socio-economic status over decades was key to new findings.
    • Notable change in the understanding of income mobility in the U.S.

Findings on Income Mobility

  • Income Mobility in the U.S.:
    • Relative intergenerational mobility is lower than in countries like Canada, Denmark, and the UK.
    • Statistics:
    • U.S. relative income mobility: 13.5%
    • Canada: 7.5%
    • Denmark: 9%
    • UK: 11.7%
  • Positive correlation between parental income and children's future income.
    • Inequality is largely inherited.
Graphical Representation
  • Poverty's Impact: Shows how income rank at age 30 is correlated with parent’s income rank, supporting the inherited inequality theory.

Decline in Absolute Income Mobility

  • Definition: The likelihood of children out-earning their parents.
    • Data shows a significant decline over decades – 50% of children born in 1980 are earning less than their parents compared to a higher rate for those born in 1940.
  • Graphs depict this decline clearly illustrating the fading American dream.

Geographic Factors in Income Mobility

  • Significant regional variations in income mobility rates are identified, especially among counties/cities.
  • Location Matters: Certain areas, particularly in the Deep South and Midwest, exhibit sluggish mobility.
The Benefits of Relocation
  • Moving from low-mobility to high-mobility areas improves life outcomes, with the greatest benefit for those who move at a younger age.
    • Policy implication: Housing vouchers can improve outcomes for children if they move before age 13.
  • Impact of Gender: Boys in low-opportunity areas face greater negative outcomes than girls.

Role of Education

  • Early Education: Strongly linked to improved outcomes; even kindergarten teachers with experience enhance future earnings significantly.
  • College Education: Serves as a great equalizer, significantly diminishing the correlation between parents' and children's income.
Barriers to College Access
  • College attendance is highly dependent on family income, contrary to the theoretical perspective of mobility.

Policy Discussions

  • Policies Focus: Moving to Opportunity and place-based investments aim to increase mobility for low-opportunity areas.
    • The effects of housing vouchers are under scrutiny to improve housing policy and social mobility outcomes.

Analyzing Moving to Opportunity Experiment

  • Research Design: Focused on the effectiveness of housing vouchers in improving economic outcomes for children.
  • Key Findings:
    • Initial results showed little impact, but recent analysis suggests children who moved young experienced economic benefits.

Future Directions

  • Need for further research on long-term outcomes and scalability of successful interventions suggested by big data analysis.
  • Address challenges in randomization and compliance in studies to ensure reliable results and conclusions for policy.
Conclusion
  • The ability to study social policies through big data has changed the landscape of social sciences, allowing for better targeted and more effective policies to improve social mobility.