Court decentralization and domestic violence against women
| Areas | : | Ethnicity, gender and citizenship |
|---|---|---|
| Researcher/s in charge | : | Raúl Andrade |
| Execution time | : | April 2008 - April 2009 |
Presentation
The objective of the study is to evaluate if increasing access to justice has positive effects on reducing domestic violence against women. The impact evaluation of a program to decentralize courts in Peru (MBJ program) will be pursued. The MBJ program installed 43 new courts in poor localities of Peru where judicial services were not available. The specific objective of the study is to measure quantitatively the impact of this program on domestic violence against women using data from the Demographic and Health Survey (Endes), specialized on women’s health.
A quasi experimental set up will be designed taking advantage of the following situation: because of a cut in the budget for the MBJ program, out of 83 planned courts only 43 were constructed. This situation provides a treatment group (43 localities with new courts) and an intent-to-treatment group (40 localities where courts were planned). The initial 83 localities were chosen on the basis of poverty levels and needs, and are similar among them. Conditionally on being among the initial 83 localities there is no reason to explain the selection of the 43 finally treated localities.
A second identification strategy is based on the presumption that the final places where MBJ were constructed were chosen on the basis of political considerations. President Fujimori had the tendency to favor localities that he considered important for his electoral ambitions by building different services, such as judicial services. This leads to the opportunity of using an instrumental variable method, modeling the construction of a court as a function of electoral results at the locality level and using the variability that this instrument generates to identify the parameter of interest.
Endes has information on violence, sexual assault, access to police offices, criminal events and decision making in the family. Endes was applied in 27 treated localities and in 30 intended to treat localities.
