Petra Todd: Advancing Empirical Economics and Program Evaluation
Petra Todd is a distinguished empirical economist whose work spans several critical domains, including labor economics, the economics of education, development, econometrics, criminology, and demography. She is most recognized for her contributions to program evaluation methods, creating frameworks to assess the impact of interventions in education and training using both experimental and nonexperimental data.
Her commitment to applying rigorous economic theory to real-world challenges is exemplified in her co-authored book with Paul Glewwe, Impact Evaluation in Developing Countries: Theory, Methods and Practice, published by the World Bank in 2020.
[ไม่มีภาพประกอบ]Key Facts
- Expertise: Specializes in program evaluation, matching methods, and regression discontinuity.
- Global Impact: Consulted on the Mexican Progresa experiment, which influenced anti-poverty programs in over 60 countries.
- Academic Contributions: Author of seminal papers and chapters in the Handbooks of Econometrics, Development Economics, Education Economics, and Labor Economics.
- Research Focus: Addresses structural inequality in education and the empirical modeling of household behaviors.
Methodological Contributions to Econometrics
Matching Methods
One of Todd's primary areas of expertise is matching methods—statistical and econometric techniques used to evaluate the impact of specific interventions. These are frequently applied to Active Labor Market programs, which are government initiatives providing training and incentives to help unemployed workers gain employment. In developing nations, these methods are essential for evaluating anti-poverty measures, such as conditional cash transfer programs. Todd and her co-authors (Heckman, Ichimura & Todd 1997, 1998) proposed new nonparametric matching estimators that remain widely utilized today.
Regression Discontinuity (RD)
Todd has also produced seminal research on regression discontinuity (RD). RD is a quasi-experimental design where a specific cut-off value of a variable determines whether a subject receives a treatment. For instance, a student might be assigned to an educational intervention only if their pre-test score falls below a certain threshold. In the influential paper Hahn, Todd & Van der Klaauw (2001), she developed new nonparametric estimators and demonstrated that RD can be interpreted as a local average treatment effect within a setting of heterogeneous treatment effects.
Addressing Inequality and Educational Policy
A significant portion of Todd's work focuses on reducing structural inequality in education, particularly in developing countries. She advocates for policies that prioritize the least-well-served students, such as providing cash incentives to poor families to encourage school attendance.
Todd served as an expert consultant for the Progresa experiment in Mexico (later renamed Oportunidades). This study randomized 506 rural villages into a conditional transfer program. The resulting evidence on improved health and schooling was pivotal in the adoption of similar programs in more than 60 countries. Additionally, she helped design the ALI experiment in Mexico, which randomized 88 high schools into a program providing incentives for improvement in mathematics curriculum tests.
[ไม่มีภาพประกอบ]Criminology and Household Behavior
Beyond education and labor, Todd has applied her econometric skills to criminology. In a study on racial profiling (Knowles, Persico & Todd 2001), she analyzed motor vehicle searches in Maryland. The data showed that African-American motorists were stopped and searched more than three times as often as others, yet drugs were found at the same rate. Todd and her co-authors argued that this equal rate of discovery suggested that the increased frequency of stops was not driven by racial bias, but rather that the higher number of stops had reduced the propensity of African-Americans to carry drugs.
Her more recent research employs dynamic discrete choice structural modeling to predict the impacts of proposed social programs before they are implemented. She also explores the empirical modeling of household behaviors—including fertility, savings, and employment—and the influence of personality traits on time allocation between spouses and educational decisions.
Summary of Research Contributions
| Research Area | Key Method/Project | Primary Application/Outcome |
|---|---|---|
| Program Evaluation | Matching Methods | Active Labor Market and anti-poverty programs |
| Econometrics | Regression Discontinuity | Local average treatment effects in education |
| Development | Progresa Experiment | Global adoption of conditional cash transfers |
| Criminology | Search Data Analysis | Testing for racial profiling in vehicle stops |
| Household Economics | Structural Modeling | Predicting impacts of new social programs |
Frequently Asked Questions
What are matching methods in economics?
Matching methods are econometric techniques used to estimate the impact of a program by comparing treated individuals to a control group with similar characteristics, often used to evaluate labor market and anti-poverty programs.
What is regression discontinuity (RD)?
Regression discontinuity is a quasi-experimental design where treatment assignment is determined by whether a variable crosses a specific threshold or cut-off value.
How did the Progresa experiment influence global policy?
By providing experimental evidence that conditional cash transfers could increase schooling and improve health, the Progresa experiment led to the implementation of similar anti-poverty programs in over 60 countries.
What is the purpose of dynamic discrete choice structural modeling?
This method is used to predict the potential impacts of social programs that do not yet exist, aiding policymakers in the design or modification of social interventions.
What did Petra Todd find regarding racial profiling in Maryland?
Her research found that while African-American motorists were searched more frequently, the rate of drug discovery was equal to that of other motorists, leading the authors to argue against the conclusion of police racism in those specific stops.