Impact Evaluation of the Subnational Transport Support Program Project

Last registered on April 05, 2018


Trial Information

General Information

Impact Evaluation of the Subnational Transport Support Program Project
Initial registration date
April 04, 2018

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
April 05, 2018, 5:51 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.



Primary Investigator

The World Bank

Other Primary Investigator(s)

PI Affiliation
The World Bank
PI Affiliation
Dartmouth College
PI Affiliation

Additional Trial Information

In development
Start date
End date
Secondary IDs
The proposed impact evaluation seeks to unveil the welfare outcomes of the Subnational Transport Support Program Project in Peru, a comprehensive rural transport project financed jointly by the IADB and the World Bank. It is expected that these interventions, while having clear results on road users, can also generate a series of impacts on the people that live in the areas of influence of the rehabilitated roads. Among these impacts, improvements in access to health, education, increases in household income, and improved access to markets and off-farm employment may directly contribute to poverty reduction and inclusive growth. We will take advantage of a prioritization ranking used by the government for the selection of roughly 95 rural roads for rehabilitation, and will create control group of similar size with rural roads that nearly missed being selected. For each treatment and control rural road, we will survey two communities, one closer to the district capital and one farther from the district capital. In each community we will survey at baseline and at endline a total of 7 households and 1 community leader, for a total of 2,600 households and 380 community leaders surveyed. We will also have a baseline, three midline, and one endline surveys measuring outcomes at the road level, including travel time, road conditions, transport costs, and local prices. In addition to the main intervention, we will use randomized controlled trials to assess the impact of two additional interventions, one that provides support services to local farmers in order to improve productivity and take advantage of lower transport costs, and one that provides capacity building to local micro-enterprises that carry out the routine maintenance of the rural roads.
External Link(s)

Registration Citation

Asher, Sam et al. 2018. "Impact Evaluation of the Subnational Transport Support Program Project ." AEA RCT Registry. April 05.
Former Citation
Asher, Sam et al. 2018. "Impact Evaluation of the Subnational Transport Support Program Project ." AEA RCT Registry. April 05.
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details


The intervention consists of a comprehensive country-wide transport program involving road rehabilitation, road improvement, and maintenance activities, supported by development windows to promote the economic development of communities and their competitiveness and integration to markets.

Intervention Start Date
Intervention End Date

Primary Outcomes

Primary Outcomes (end points)
Household Income, Household consumption, Employment/ Household Occupations, Migration, Poverty, Fixed Assets, Use of Time, Total Agricultural Output, Ag Output Marketed, Prices, Land use patterns, Access to Education, Access to Health.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Transport Costs (passenger and freight), Availability of public transport, Traffic counts and vehicle speeds, Quality of Road
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
The impact evaluation design can be divided by each of the three components :

Rural Road Rehabilitation: A non-experimental design using differences-in-differences will take advantage of the discontinuity in the selection of roads through a prioritization ranking. The treatment group consists of 95 road segments selected through this prioritization process, with a total of approximately 1,549km. The control group is composed of an equivalent number of road segments that participated in the prioritization process and nearly missed being selected, which meet the minimum requirements to benefit from the program, but were excluded from the intervention due to limited resources, since the program can cover only up to 20km of roads per province.

Development Windows: The development window components will be evaluated with an experimental methodology with a stratified random selection of 24 treatment provinces that will be treated with the development window program. The remaining 51 provinces that are part of the evaluation of the road rehabilitation program will make up the control group.

Maintenance: A stratified random selection will take place to assign existing microenterprises to the treatment group. About 200 microenterprises will be part of the treatment groups from a universe of about 600. The microenterprises that do not receive the intervention will be the control group, and will not receive any type of capacity building. It is expected that systematic differences between the control and treatment roads are fully eliminated thanks to the random assignment of entry into the program.
Experimental Design Details
Randomization Method
For the development window component and for the maintenance capacity building component, the randomization will be carried out in office of the government agency responsible for the implementation of the program, with a computer using
Randomization Unit
For the rehabilitation of rural roads (non-experimental) there will be several units of analysis depending on the outcome: road segment, community, and household. For the development window component, the randomization will be carried out at the province level. For the maintenance component, the randomization will be carried out at the firm level.
Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters
For the rehabilitation of rural roads and for the development window components, treatment will be clustered in a total of 95 treated road segments, and 95 control road segments.
Sample size: planned number of observations
180 treated communities and 1,330 households.
Sample size (or number of clusters) by treatment arms
180 control communities and 1,330 control households.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number


Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information


Is the intervention completed?
Data Collection Complete
Data Publication

Data Publication

Is public data available?

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials