Electrifying the last meter: Experimental evidence from Rwanda

Last registered on September 21, 2026

Pre-Trial

Trial Information

General Information

Title
Electrifying the last meter: Experimental evidence from Rwanda
RCT ID
AEARCTR-0019583
Initial registration date
September 10, 2026

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
September 21, 2026, 8:24 AM EDT

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

Locations

Region

Primary Investigator

Affiliation
World Bank

Other Primary Investigator(s)

PI Affiliation
World Bank
PI Affiliation
World Bank

Additional Trial Information

Status
On going
Start date
2024-10-01
End date
2028-06-30
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
We estimate the optimal mix of subsidies to achieve universal electrification. In Rwanda, expensive grid connections provide reliable and scalable electricity access, while cheaper off-grid solar home systems offer basic lighting and mobile phone charging. We cross-randomize subsidies for grid and off-grid electrification across 180 villages to estimate supply and demand for grid and off-grid electrification, and their impacts on households’ subjective welfare. We apply these results to identify subsides that minimize costs, both explicit (subsidy payouts) and implicit (below marginal cost pricing), while “justly” electrifying a target number of households: they increase subjective welfare at least as much as grid electrifying the same number of households. More broadly, our approach reconciles economic surplus- and equity-based viewpoints on achieving development objectives including universal electrification.
External Link(s)

Registration Citation

Citation
Balantrapu, Tanay, Florence Kondylis and John Loeser. 2026. "Electrifying the last meter: Experimental evidence from Rwanda." AEA RCT Registry. September 21. https://doi.org/10.1257/rct.19583-1.0
Experimental Details

Interventions

Intervention(s)
Intervention Start Date
2025-06-01
Intervention End Date
2027-03-31

Primary Outcomes

Primary Outcomes (end points)
Electrification status (has an SHS, has a grid connection).
Energy usage and expenditures (electricity, mobile phone charging, lighting).
Subjective well being (overall, energy domain).
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Financial inclusion.
Housing characteristics.
Asset ownership.
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Villages were randomized into four experimental arms, with one quarter of villages assigned to each arm. Our initial sample of villages were all 193 villages in four Districts in which new electrification was planned under the EAQIP project.

(No Readyboards/No SHS) The first arm is a control arm, in which electrification proceeds as under the status quo.

(Readyboards/No SHS) In the second arm, all vulnerable households who were not already electrified (whether through a grid connection or SHS) were offered a free readyboard (if they had not already interally wired at the time of the offer) and installation, which substitutes for the need to do costly internal wiring.

(No Readyboards/SHS) In the third arm, all households in the village are offered SHS at a 90% subsidy. SHS subsidy implementation begins after readyboard implementation is completed.

(Readyboards/SHS) In the fourth arm, both readyboard program and SHS subsidies are implemented.

Randomization was stratified by lot (the unit at which new electrification was planned, with five lots total) and village size (above and below median within lot, yielding ten randomization strata). 48/48/48/49 villages were assigned to the four arms.

12 villages in one District were dropped from electrification under EAQIP, leaving 181 villages. 1 village was fully resettled before the listing, reducing our final sample to 180 villages. These villages had been assigned 46/45/45/44 to the four arms.
Experimental Design Details
Not available
Randomization Method
Randomization done in office by a computer.
Randomization Unit
Village
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
180 Villages
Sample size: planned number of observations
1973 Households
Sample size (or number of clusters) by treatment arms
No Readyboards/No SHS: 46 Villages
No Readyboards/SHS: 45 Villages
Readyboards/No SHS: 45 Villages
Readyboards/SHS: 44 Villages
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
HML IRB
IRB Approval Date
2024-07-25
IRB Approval Number
2563