Can Panel Data Designs and Estimators Substitute for Randomized Controlled Trials in the Evaluation of Environmental Policy
Event Details
Presenters
Paul J. Ferraro
Professor, Department of Economics, Andrew Young School of Policy Studies,
Georgia State University
Abstract
In environmental policy, as in other areas of social policy, randomized evaluation designs are difficult to implement and thus researchers must rely on non-experimental empirical designs to evaluate program impacts. Yet there is considerable debate about whether non-experimental designs can generate accurate estimates of program impact. Design-replication studies assess the ability of non-experimental designs to replicate unbiased (experimental) estimators of program impact. Our design-replication study uses, as a benchmark, a large-scale randomized field experiment that tested the effectiveness of messages designed to induce voluntary reductions in water consumption during a drought. We find that, in general, traditional panel data estimators are unable to replicate the estimates from the experimental design except in one case: when caliper-matching methods are used to pre-process the data before applying a fixed-effects panel data estimator. Understanding how the caliper matching changes the sample, however, will be critical for interpreting the estimates from such an estimator in other contexts. Insights for best-practice econometric analysis are offered.
Date
Thursday, October 18, 2012
12:00 - 1:30 p.m. Lunch will be provided. Location
1st Floor Conference Room
1616 P St. NW
Washington, D.C. 20036 All seminars will be in the 7th Floor Conference Room at RFF, 1616 P Street NW unless otherwise noted. Attendance is open, but involves pre-registration no later than two days prior to the event. For questions and to register to an event, please contact Khadija Hill at [email protected] (tel. 202-328-5174). Updates to our academic seminars schedule will be posted at www.rff.org/academicseminarseries.
Participants
Paul Ferraro