Asymmetric treatment in joint-liability loan contracts

Last registered on May 27, 2019

View Trial History

Pre-Trial

Trial Information

General Information

Title

Asymmetric treatment in joint-liability loan contracts

RCT ID

AEARCTR-0004217

Initial registration date

May 23, 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

May 27, 2019, 4:55 PM EDT

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

Locations

Country

Netherlands

Region

Primary Investigator

Name

Sigrid Suetens

Affiliation

Tilburg University

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Philine Visser

PI Affiliation

PI Name

Franscesco Carli

PI Affiliation

PI Name

Burak Uras

PI Affiliation

Additional Trial Information

Status

In development

Start date

2019-05-28

End date

2019-12-31

Keywords

Finance & Microfinance

Additional Keywords

group loan contracts, asymmetric contracts, joint liability, laboratory experiment

JEL code(s)

Secondary IDs

Abstract

Group lending is a common practice that microfinance institutions (MFIs) utilize when lending to individuals without collateral. By making borrowers co-sign, MFIs can mitigate moral hazard problems without requiring collateral. The joint liability gives incentives to borrowers to monitor each other. However, if all borrowers have the same contract terms, it is not clear that the incentives to monitor are strong enough. They may end up in a joint non-monitoring outcome with bad investments and losses for the MFI as a result. Carli and Uras (2017) show in a theoretical model that coordination on non-monitoring can be avoided by designing contracts such that one of the borrowers (a leader) finds it optimal to monitor the peer borrowers in a group loan contract, thereby inducing the peers to invest in good projects and in turn to monitor the leader's own project choice. It is an open question, however, to which extent non-monitoring is empirically relevant in the case that contract terms are symmetric. Our study aims to answer this question.

External Link(s)

Registration Citation

Citation

Carli, Franscesco et al. 2019. "Asymmetric treatment in joint-liability loan contracts." AEA RCT Registry. May 27. https://doi.org/10.1257/rct.4217-1.0

Former Citation

Carli, Franscesco et al. 2019. "Asymmetric treatment in joint-liability loan contracts." AEA RCT Registry. May 27. https://www.socialscienceregistry.org/trials/4217/history/47130

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

Intervention Start Date

2019-05-28

Intervention End Date

2019-12-31

Primary Outcomes

Primary Outcomes (end points)

the percentage of times borrowers choose to monitor; MFI profits; profits of borrowers

Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)

Secondary Outcomes (explanation)

Experimental Design

between-subjects 2x2 design:
asymmetric contracts versus symmetric contracts
strangers matching versus partners matching

Experimental Design Details

The experiment has four treatments and uses a between-subjects 2x2 design (symmetric vs. asymmetric contracts and strangers vs. partners matching). The participants play a series of the same monitoring subgame. They are matched with another (anonymous) participant and their task is to choose to monitor the other or not to monitor the other.

In the strangers treatments they are randomly matched with another participant after each round and in the partners treatments they play the game with the same partner for all rounds. In the symmetric treatments the payoffs in the game are the same for the two players, making the game a coordination game. In the asymmetric treatments the payoffs are such that for one of the two players it is a dominant strategy to monitor.

Randomization Method

participants randomly draw a card with a number and are allocated to a computer with that number; random matching is done with the computer

Randomization Unit

the randomization unit is the matching group in the strangers treatments and the pair in the partners treatments

Was the treatment clustered?

Yes

Experiment Characteristics

Sample size: planned number of clusters

38 (20 in partners and 8 in strangers)

Sample size: planned number of observations

120 (40 in partners and 80 in strangers)

Sample size (or number of clusters) by treatment arms

partners & symmetric: 20 (10 clusters)
partners & asymmetric: 20 (10 clusters)
strangers & symmetric: 40 (4 clusters)
strangers & asymmetric: 40 (4 clusters)

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Supporting Documents and Materials

IRB