We test if performance scorecards can increase municipal government performance. In a field experiment in Burkina Faso, local decision makers are provided with a scorecard that provides information on national performance standards and on the municipality's actual performance along 16 indicators of service delivery and institutional capacity. Municipal officials are asked to internally assign responsibilities for each performance indicator and are made aware that their performance is being monitored.
External Link(s)
Citation
Holmlund, Marcus and Malte Lierl. 2019. "Can Scorecards Improve Municipal Government Performance? A Field Experiment." AEA RCT Registry. November 05. https://doi.org/10.1257/rct.4953-1.0.
In a field experiment in Burkina Faso, local decision makers are provided with a scorecard that provides information on national performance standards and on the municipality's actual performance along 16 indicators of service delivery and institutional capacity. Municipal officials are asked to internally assign responsibilities for each performance indicator and are made aware that their performance is being monitored.
Intervention Start Date
2016-11-01
Intervention End Date
2019-01-06
Primary Outcomes (end points)
Municipal government performance, as measured by the scorecard indicators.
Primary Outcomes (explanation)
Secondary Outcomes (end points)
Municipal officials' knowledge of performance targets and of actual performance.
Secondary Outcomes (explanation)
Experimental Design
In 70 randomly selected treatment municipalities (blocked by region), municipal performance scorecards are disseminated to key municipal officials in 2016 and in 2017. Baseline data: municipal performance in 2014, 2015, 2016. Midline data: municipal performance in 2017, Endline data: municipal performance in 2018.
Experimental Design Details
Randomization Method
Block randomization by region, equal treatment assignment probabilities across all regions.
Randomization Unit
Municipalities
Was the treatment clustered?
No
Sample size: planned number of clusters
140 municipalities
Sample size: planned number of observations
140 municipalities
Sample size (or number of clusters) by treatment arms
70 treatment municipalities, 70 control municipalities
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)