Land Consolidation and Agricultural Productivity: Experimental Evidence from Kenya

Last registered on January 02, 2025

Pre-Trial

Trial Information

General Information

Title
Land Consolidation and Agricultural Productivity: Experimental Evidence from Kenya
RCT ID
AEARCTR-0015043
Initial registration date
December 17, 2024

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
January 02, 2025, 9:45 AM EST

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

Locations

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

Request Information

Primary Investigator

Affiliation
Yale University

Other Primary Investigator(s)

PI Affiliation
Yale University

Additional Trial Information

Status
In development
Start date
2025-01-01
End date
2027-12-31
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
One often-postulated driver of cross-country disparities in agricultural labor productivity is differences in farm size. This project aims to provide some of the first experimental evidence on the returns to land consolidation in low-income countries. In doing so, we also test the impact of a commonly proposed (but rarely empirically tested) solution to the “too many small farms” problem: large, less credit constrained private-sector companies could be better positioned to consolidate land than smallholder farmers themselves. We partner with a Kenya-based company that leases small, adjacent plots of land from farmers, which it then consolidates to enable investments that demand scale, including irrigation, tractors, and professional managers. Using a randomized encouragement design that randomizes borehole location, we will measure the impacts of the induced land consolidation on agricultural productivity, with the ultimate goal of exploring land consolidation in the context of a structural change model. Further, we will study the structural adjustment in labor markets that land consolidation may unlock, by collecting data on employment outcomes of farmers potentially released from their land.
External Link(s)

Registration Citation

Citation
Bergquist, Lauren and Kevin Donovan. 2025. "Land Consolidation and Agricultural Productivity: Experimental Evidence from Kenya." AEA RCT Registry. January 02. https://doi.org/10.1257/rct.15043-1.0
Experimental Details

Interventions

Intervention(s)
We partner with a Kenya-based start-up that leases small, nearby plots of land from farmers. It then consolidates these farms to enable investments that demand scale, including irrigation, tractors, and professional managers. Although the plots must be in close proximity to each other to take advantage of large, fixed investments in mechanization and management, plots do not have to be completely adjacent (e.g., irrigation water is piped specifically to the participating plots and is certified by the government not to deplete reserves, implying no positive or negative impact on uninvolved plots).
Intervention Start Date
2025-01-01
Intervention End Date
2026-12-31

Primary Outcomes

Primary Outcomes (end points)
Our key outcomes of interest are agricultural outcomes, include yields, farm revenues, and farm profits.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
We will also explore two secondary outcomes:
(1) Impacts on inputs and mechanization, to understand mechanisms
(2) Impacts on landowner labor use, to understand how land consolidation can spur a transition into market activities, non-agricultural employment, and/or urban migration – as well as the labor market frictions that may constrain this adjustment.
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
When opening a new consolidated farm, the company first evaluates the agronomic, economic, and geological conditions of a wide range of areas across the country. This process typically results in a 30-50km area that is suitable for the new farm. Within this area, the company conducts a hydraulic survey to identify 3-4 possible locations where a borehole could be drilled. The borehole is used to pump groundwater to the surface and is critical for the establishment of an irrigation system. The borehole has a 2-3km catchment radius, making it unprofitable for the company to include plots outside of this radius. The company then offers lease terms within the catchment radius. Because the company offers an above-market premium on its lease terms, it typically enjoys high take up within this radius.

For the purpose of this study, the company will randomize which of the 3-4 possible borehole locations is actually used. This will generate one treatment catchment area and 2-3 control catchment areas per each consolidated farm. Each catchment area typically contains between 30 and 50 plots of land. We will use being located in a treated catchment area as a randomized encouragement for the plot to join the consolidated farm.

The company plans to roll out 10 new farms over the next two years. We will survey farmers before and after the company enters in all treated and control catchment areas to measure agricultural productivity, employment outcomes, and any non-market activity that respond to the freed time now available to farmers and their families.
Experimental Design Details
Not available
Randomization Method
This study will employ a randomized encouragement design. Among the feasible borehole locations, the one randomly selected for treatment will be determined in office by a computer.
Randomization Unit
Borehole catchment area. Each possible borehole location has a 2-3km catchment radius, typically containing between 30 and 50 plots of land.
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
We plan to have one treated borehole and three control boreholes for each consolidated farm. We expect to have ten consolidated farmers, for a total sample size of 10 treated boreholes and 30 control boreholes. More may be included if the company expands more quickly.
Sample size: planned number of observations
Each catchment area typically contains between 30 and 50 plots of land. There are 40 catchment areas (10 treated and 30 control). We therefore exact about 2,000 plots included in our sample.
Sample size (or number of clusters) by treatment arms
500 plots coming from 10 catchment areas will be in treatment, while 1,500 plots coming from 30 catchment areas will be in control.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
Yale Human Research Protection Program Institutional Review Board
IRB Approval Date
2024-11-15
IRB Approval Number
2000038642
IRB Name
Strathmore University
IRB Approval Date
2024-10-23
IRB Approval Number
SU-ISERC2383/24