Back to History Current Version

A lab-in-field experiment on land trade complexity and market centralization

Last registered on November 27, 2019

Pre-Trial

Trial Information

General Information

Title
A lab-in-field experiment on land trade complexity and market centralization
RCT ID
AEARCTR-0004581
Initial registration date
August 16, 2019

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
August 20, 2019, 10:46 AM EDT

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

Last updated
November 27, 2019, 2:57 PM EST

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Region

Primary Investigator

Affiliation
Queen Mary University of London

Other Primary Investigator(s)

PI Affiliation
London School of Economics
PI Affiliation
University of Melbourne
PI Affiliation
Maastricht University & UNU-MERIT

Additional Trial Information

Status
In development
Start date
2019-08-18
End date
2020-08-31
Secondary IDs
Abstract
We study the effects of complexity and market centralization in a land trading game. The game involves groups of 18 participants in villages in Masaka, Uganda. Participants are endowed with game currency and a set of artificial land plots on a "map", and trade plots among themselves to earn returns. They earn returns if they improve the efficiency of the land allocation in the game, which can be achieved by sorting higher "ability" players to better plots and/or by consolidating fragmented plots into contiguous blocks.

Our treatment variations are
1. Map complexity: some maps are "simple" because there are many possible payoff-equivalent first-best allocations of plots. Some maps are "complex" because there are only a few. Complexity is varied by minimally altering maps to add non-trading plots. These decrease the number of possible ways to arrange contiguous blocks of plots into the map.

2. Trade centralization: trade is either "decentralized," with participants trading free-form among themselves within the village, or "centralized," taking place with all participants present in the same location at the same time.

We will study the effects of these treatments on trading efficiency, measured by the share of possible gains from trade achieved in the game, and the distribution of those gains, measured by the log-utility Atkinson index of final payoffs.
External Link(s)

Registration Citation

Citation
Bryan, Gharad et al. 2019. "A lab-in-field experiment on land trade complexity and market centralization." AEA RCT Registry. November 27. https://doi.org/10.1257/rct.4581-2.0
Experimental Details

Interventions

Intervention(s)
Intervention Start Date
2019-08-19
Intervention End Date
2019-11-29

Primary Outcomes

Primary Outcomes (end points)
Overall trading efficiency. Trading efficiency broken down into a) Trading efficiency due to defragmentation, b) Trading efficiency due to sorting, c) Trading efficiency due to exposure losses
Distribution of final payoffs measured by the log-utility Atkinson index.
Primary Outcomes (explanation)
Overall trading efficiency (share of potential gains from trade realized)
Trading efficiency due to defragmentation (share of potential gains from defragmenting land)
Trading efficiency due to sorting (share of potential gains from sorting high types to high quality land)
Trading efficiency due to exposure losses (share of potential gains foregone due to some participants holding "too much" land)
Distribution of final payoffs measured by the log-utility Atkinson index (1 minus the ratio of the geometric and arithmetic means of final payoffs)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
We play a series of land trading games, in villages, over the course of 2 weeks and three meetings. The games involve trading artificial land titles, which correspond to plots on a generated "map." Participants trade amongst themselves using game currency, and earn returns according to the gains from trade they achieve. Gains from trade can be achieved by sorting higher "ability" players onto better land, and consolidating fragmented plots into contiguous blocks. Trade is done in free-form bargaining between players. Maps can be either "simple" or "complex" and trade can be "decentralized" or "centralized."
Experimental Design Details
We play a series of land trading games, in villages, over the course of 2 weeks and three meetings. The games involve trading artificial land titles, which correspond to plots on a generated "map." Participants trade amongst themselves using game currency, and earn returns according to the gains from trade they achieve. Gains from trade can be achieved by sorting higher "ability" players onto better land, and consolidating fragmented plots into contiguous blocks. Trade is done in free-form bargaining between players. Maps can be either "simple" or "complex" and trade can be "decentralized" or "centralized."

In meeting 1 we recruit 18-22 participants per village (18 main participants and up to 4 reserves), to participate in three meetings 7 days apart. In meeting 1 we train participants in the games, and then distribute maps and initial endowments to 18 participants. They play the trading game among themselves ("decentralized") for 1 week.

1 week later at meeting 2, we collect the final endowments of the players, conduct a survey, and distribute new endowments for the following week.

1 week later at meeting 3, we collect the final endowments of the players. We then surprise them with the opportunity to continue trading for an additional period. Since they are all together we call this "centralized" trade, and the period is called the "trading day."
Randomization Method
Randomization done in office by a computer
Randomization Unit
Village and week.
Each village will play either a simple map in week 1 and a complex map in week 2, or complex in week 1 and simple in week 2.
At the end of week 2 trade, all villages play a "trading day"
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
68 villages

Each village plays a simple map followed by a complex map, or a complex map followed by a simple map. The second week's game is also followed by a trading day.
Sample size: planned number of observations
204 in total. 3 observations per village: a week 1 trading efficiency outcome, a week 2 pre-trading day efficiency outcome, and a week 2 post-trading day efficiency outcome.
Sample size (or number of clusters) by treatment arms
All villages play the simple treatment once and the complex treatment once, so 68 simple and 68 complex.

The trading day is played at the end of week 2, so we will have 34 observations for simple post-trading day and 34 for complex post-trading day.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
Research Ethics Committee, London School of Economics
IRB Approval Date
2018-05-29
IRB Approval Number
000718
IRB Name
Mildmay Uganda Research Ethics Committee (MUREC)
IRB Approval Date
2018-07-18
IRB Approval Number
0406-2018
Analysis Plan

Analysis Plan Documents

landtradecomplexity_preplan.pdf

MD5: 302c0d43c32e9eea235d4d523d52e7cf

SHA1: 1c3f8049157cec25a88530a815b81d27a6068d23

Uploaded At: November 27, 2019

Post-Trial

Post Trial Information

Study Withdrawal

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

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials