Microfinance field experiment in Bhakkar and Chakwal

Last registered on May 10, 2015

Pre-Trial

Trial Information

General Information

Title
Microfinance field experiment in Bhakkar and Chakwal
RCT ID
AEARCTR-0000684
Initial registration date
May 10, 2015

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 10, 2015, 12:59 PM EDT

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

Locations

Region

Primary Investigator

Affiliation
Department of Economics and Public Policy, Imperial College London

Other Primary Investigator(s)

PI Affiliation
Stanford University
PI Affiliation
Lahore School of Economics
PI Affiliation
Lahore School of Economics
PI Affiliation
Milan Politecnico

Additional Trial Information

Status
On going
Start date
2014-08-25
End date
2015-07-01
Secondary IDs
Abstract
We are currently running a field experiment of a novel microfinance product, inspired by the rotating structure of a ROSCA. This is a scaled-up version of the product structure presented in CSAE Working Paper WPS/2014-32 (Afzal et al, `Two Sides of the Same Rupee? Comparing Demand for Microcredit and Microsaving in a Framed Field Experiment in Rural Pakistan'). We are currently approaching the completion of fieldwork.
External Link(s)

Registration Citation

Citation
Afzal, Uzma et al. 2015. "Microfinance field experiment in Bhakkar and Chakwal." AEA RCT Registry. May 10. https://doi.org/10.1257/rct.684-1.0
Former Citation
Afzal, Uzma et al. 2015. "Microfinance field experiment in Bhakkar and Chakwal." AEA RCT Registry. May 10. https://www.socialscienceregistry.org/trials/684/history/4282
Sponsors & Partners

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

Request Information
Experimental Details

Interventions

Intervention(s)
Our design is simple, and builds on the evolved wisdom of the ROSCA idea: in short, we replicate the rotating structure of a ROSCA in an individual-liability product. This combines saving and credit into a single financial service, offered to individuals (rather than groups), and tests whether microcredit demand is motivated by savings needs.

Experimental treatments vary the type of contract offered to participants. Contracts differ by (i) timing of lump sum payment and (ii) interest rate. Assuming a product cycle of six weeks, lump sum payments are either made on week 1 or week 6 (where ‘week 1’ refers to the week immediately following the week of the offer). On any week that the lump sum is not paid, the participant is required to pay a fixed amount. The base lump sum payment is either smaller (that is, a negative interest rate), equal to (zero nominal interest rate) or larger than (positive interest) the sum of weekly fixed amounts paid.

We there have two treatments, which together can be combined in six different ways (that is, there are three possible interest rates and two possible weeks for lump sum payment). Each participant in the experiment is randomly offered one of these contracts, and must make a ‘take it or leave it’ decision whether to accept. We are interested to test (i) whether there is demand for this kind of rotating individual-liability product, and (ii) if so, how that demand varies with the terms of the contract randomly offered.

The proposed study will represent a novel and important contribution to the existing literature on microfinance under two main respects. First, almost no research in this literature considers directly the trade-off between microcredit and microsaving; that is, almost no literature studies the extent to which savings and credit may operate as substitute methods for easing credit constraints. One notable exception is the recent work of Kast and Pomeranz (2013), who show some evidence that reducing barriers to savings lowers borrowing among microfinance clients in Chile. The proposed study, by evaluating a product that embodies both savings and credit components, will be able to directly test the degree of complementarity or substitutability of credit and savings for different types of individuals. Second, while there is some suggestion that the formalisation of informal savings products can have large advantages for the poor (Dupas and Robinson, 2012), no previous research has tested directly whether the rotating structure of a ROSCA can viably be implemented as an individual microfinance product. The proposed study, by reproducing the structure of a ROSCA within an individual liability product, can test the viability of formalising one of the most common informal savings tool in the developing world, and evaluate its impact.
Intervention Start Date
2014-09-22
Intervention End Date
2015-03-07

Primary Outcomes

Primary Outcomes (end points)
These are described in detail in the formal pre-analysis plan document.
Primary Outcomes (explanation)
The construction of variables is described in detail in the formal pre-analysis plan document.

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Experimental treatments vary the type of contract offered to participants. Contracts differ by (i) timing of lump sum payment and (ii) interest rate. Assuming a product cycle of six weeks, lump sum payments are either made on week 1 or week 6 (where ‘week 1’ refers to the week immediately following the week of the offer). On any week that the lump sum is not paid, the participant is required to pay a fixed amount. The base lump sum payment is either smaller (that is, a negative interest rate), equal to (zero nominal interest rate) or larger than (positive interest) the sum of weekly fixed amounts paid.

We there have two treatments, which together can be combined in six different ways (that is, there are three possible interest rates and two possible weeks for lump sum payment). Each participant in the experiment is randomly offered one of these contracts, and must make a ‘take it or leave it’ decision whether to accept. We are interested to test (i) whether there is demand for this kind of rotating individual-liability product, and (ii) if so, how that demand varies with the terms of the contract randomly offered.
Experimental Design Details
Randomization Method
Randomization into treatment and control was done in an office at the Lahore School of Economics. Among the treated respondents, randomization into different contract offers was done by drawing cards in front of the participants.
Randomization Unit
Randomization was done at the level of the individual in each case.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
900
Sample size: planned number of observations
900
Sample size (or number of clusters) by treatment arms
450 treated, 450 control
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
We provided the following statement to the funder (the International Growth Centre) prior to the study: "...we are confident that our high power from the pilot will extend to this scaled-up study. To evaluate this, we have run a bootstrap resampling algorithm; this preserves the intracluster correlation structure of both treatments and outcomes, while allowing us to vary the hypothetical sample size. For the actual effect sizes observed in the first pilot, we can use a sample of just 300 observations and still obtain power of 96.7% for detecting sensitivity to the interest rate, and power of 94.7% for detecting sensitivity to the day of repayment (with alpha of 0.05). (For the intended participating sample of 450, the power rises to over 99% in each case.) Intuitively, this power is generated by the fact that we expose each participant respondent to three separate rotations of the treatment. "As discussed shortly, our revised design includes 450 participants and 450 controls. We intend to compare participants and controls on a variety of continuous firm-level outcomes (e.g. firm profits). Simple calculations show that we will detect an Intent To Treat effect size of 0.25 standard deviations with power of 96% (again, at alpha of 0.05)."
IRB

Institutional Review Boards (IRBs)

IRB Name
Lahore School of Economics Research Ethics Review Committee
IRB Approval Date
2015-05-07
IRB Approval Number
RERC-042015-02
Analysis Plan

Analysis Plan Documents

Microfinance Field Experiment in Bhakkar and Chakwal: Pre-Analysis Plan

MD5: 5ed34bcfe4ff51dbbb8bafacc027c30f

SHA1: cf829a993e82baddb90e51ef5f1882496638d0bb

Uploaded At: May 09, 2015

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