Back to History Current Version

Time preference and liquidity constraints: experimental evidence from Pakistan

Last registered on December 17, 2017

Pre-Trial

Trial Information

General Information

Title
Time preference and liquidity constraints: experimental evidence from Pakistan
RCT ID
AEARCTR-0001343
Initial registration date
June 13, 2016
Last updated
December 17, 2017, 8:39 AM EST

Locations

Region

Primary Investigator

Affiliation
World Bank

Other Primary Investigator(s)

Additional Trial Information

Status
On going
Start date
2016-03-14
End date
2018-12-31
Secondary IDs
Abstract
The idea that many poor individuals are significantly present-biased has led to a proliferation of recent development interventions with commitment features, in particular commitment savings products (Ashraf, Karlan & Yin, 2006; John, 2015). Yet if individuals integrate background income into experimental decisions over time-dated monetary payments, then apparent measurement of ``present-bias'' may in fact reflect individuals' (correct or incorrect) anticipation that they will be less liquidity-constrained in the future. To experimentally test this hypothesis among a population which has been explicitly targeted for commitment products, I conduct a field experiment with female clients of a microfinance organisation in Pakistan. I randomly vary the timing of survey participation fees in order to induce exogenous variation in these individuals’ liquidity constraints, and examine whether and how this impacts individuals' measured rate of time preference and time-consistency.
External Link(s)

Registration Citation

Citation
Cassidy, Rachel. 2017. "Time preference and liquidity constraints: experimental evidence from Pakistan." AEA RCT Registry. December 17. https://doi.org/10.1257/rct.1343-4.0
Former Citation
Cassidy, Rachel and Rachel Cassidy. 2017. "Time preference and liquidity constraints: experimental evidence from Pakistan." AEA RCT Registry. December 17. https://www.socialscienceregistry.org/trials/1343/history/47396
Experimental Details

Interventions

Intervention(s)
Intervention Start Date
2016-03-14
Intervention End Date
2016-05-28

Primary Outcomes

Primary Outcomes (end points)
Measures of time preference: near- and far- frame discount rates, stationarity, time-inconsistency, time-invariance (Halevy, 2015).
Primary Outcomes (explanation)
See pdf for detailed description of variable construction.

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
- I randomly vary the timing of survey participation fees in order to induce exogenous variation in individuals’ liquidity constraints, and examine whether and how this impacts individuals' measured rate of time preference and time-consistency. I consider both choices at baseline and revised choices at revisit, and control for measures of risk preferences and cognitive skills which could also potentially be affected by the participation fee treatment.
- I also experimentally manipulate the order of activities and near- versus far- frame to test for salience effects.
- Finally, I randomise the order in which villages are surveyed, such that I am able to exploit exogenous variation in whether individuals are surveyed before or after the onset of the wheat harvest (and its potential associated easing of liquidity constraints).
Experimental Design Details
Randomization Method
- Assignment to treatment arms was performed in office by a computer.
- Whether each respondent would receive a bonus participation fee was performed during the interview by drawing a ball from a bag.
Randomization Unit
- Randomisation into payment day and salience treatments was done at the individual level.
- Randomisation into village survey order was done at the village level (10 individuals were surveyed per village).
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
- Randomisation into payment day and salience treatments was done with all 530 individuals (who were spread across 53 villages).
- Randomisation into village survey order was done with a subset of 48 villages (5 villages were excluded from this treatment), implying 48 clusters of 10 individuals each.
Sample size: planned number of observations
530 individuals
Sample size (or number of clusters) by treatment arms
Individual-level treatment arms
The total sample of 530 individuals was cross-cut along the following three axes:
i) Paid at baseline - 263; paid at revisit - 267
ii) Salience - 252; control activities first - 278
iii) Near frame first - 259; far frame first - 271

Village-level treatment arms
A total sub-sample of 480 individuals were also randomised into a village-level treatment alongside their individual treatment status:
iv) Pre-harvest - 24 villages (240 individuals); Post-harvest - 24 villages (240 individuals)
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
University of Oxford Department of Economics Research Ethics Committee
IRB Approval Date
2014-05-09
IRB Approval Number
ECONCIA13-032
Analysis Plan

Analysis Plan Documents

Cassidy_PakistanTimeConsistency_Pre-Analysis

MD5: 331247a2836854ac6f9906e69e6a0c28

SHA1: d281fabad3e5421a1c4deda105c0adad97c6648a

Uploaded At: June 13, 2016

Post-Trial

Post Trial Information

Study Withdrawal

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

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