EVALUATING CONTRADICTORY EXPERIMENTAL AND NON-EXPERIMENTAL ESTIMATES OF NEIGHBORHOOD EFFECTS ON ECONOMIC OUTCOMES FOR...

Area of Science:

• Urban Sociology
• Economics
• Public Policy

Background:

• Non-experimental studies frequently report significant neighborhood effects on adult outcomes.
• Residential mobility experiments, like Moving to Opportunity (MTO), have yielded contrasting results, questioning these neighborhood effects.
• Discrepancies between experimental and non-experimental findings necessitate rigorous comparison.

Purpose of the Study:

• To compare experimental and non-experimental estimates of neighborhood effects using data from the Moving to Opportunity (MTO) experiment and the Panel Study of Income Dynamics (PSID).
• To investigate potential reasons for discrepancies, including exposure duration, effect heterogeneity, and move disruption.
• To assess the role of selection bias in non-experimental studies of neighborhood effects.

Main Methods:

• Employed a within-study comparison design.
• Analyzed data from the Moving to Opportunity (MTO) experiment.
• Conducted a parallel analysis using the Panel Study of Income Dynamics (PSID).

Main Results:

• Observed striking similarities between non-experimental estimates derived from MTO and PSID data.
• Found no clear evidence linking estimate differences to exposure duration, non-linear effects, magnitude of change, move frequency, heterogeneity, or measurement issues.
• Results regarding MTO-induced moves being disruptive were inconsistent.

Conclusions:

• The findings suggest that selection bias may account for the robust neighborhood effects observed in non-experimental studies on adult economic outcomes.
• The study highlights the importance of considering selection mechanisms when interpreting neighborhood effects.
• Reconciling experimental and non-experimental findings is crucial for accurate policy implications.