Dollars, cents, and mobile money: Do denomination and form of payment affect measures of risk and time preferences?

How to cite

Hoel, Jessica; Jain, Prachi; Galaty, Bridget. 2019. "Dollars, cents, and mobile money: Do denomination and form of payment affect measures of risk and time preferences?." AEA RCT Registry. AEARCTR-0004795. https://doi.org/10.1257/rct.4795-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Jessica Hoel* , Prachi Jain , Bridget Galaty

* Primary investigator

Registry ID
AEARCTR-0004795
DOI
https://doi.org/10.1257/rct.4795-1.0
First registered
Last updated
Status
On going
Trial dates
2019-10-04 – 2020-03-31
Countries
Private
JEL codes
C90, C91, D9
Registry record
http://www.socialscienceregistry.org/trials/4795

Abstract

This study plans to examine whether people express different risk and time preferences when: a) offered payment in cash or by mobile money, and b) when options are offered in either whole/round numbers or exact numbers. We expect our findings will be of methodological interest to experimental economists, but we also expect that our findings can speak to the broader question of behavioral biases that govern human behavior. This is important because studies in the literature often differ in these choices; it is unclear whether measured risk and time preferences differ across studies due to differences in context, innate preferences, or due to these often overlooked experimental design choices. If we find differences in elicited measured preferences due to form of payment or denomination, we can also explore the behavioral biases that govern how people think about money.

Intervention

The experiment will be conducted at the Busara Center for Behavioral Economics in Fall 2019. Busara’s participant pool is broadly representative of Nairobi and Kenya. Participants will complete risk, time, and form-of-payment multiple price list preference elicitation tasks. Participants also complete a questionnaire, which includes a series of demographic questions and questions about their usage of mobile money.

Experimental design

Subjects are randomly assigned to be offered payment for the risk and time tasks either in cash or by mPesa, a mobile money platform that has been in circulation in Kenya since 2007 and is widely used across the socio-economic spectrum.

To test the effect of the denomination of payment, some choices are denominated in whole or round numbers, while others are denominated in exact numbers. For example, in the risk multiple price lists, one choice offered is between a 50-50 gamble over 100 and 800 shillings or 250 shillings for sure. This is a Whole-Whole choice. Another choice offered is between a 50-50 gamble over 126 and 768 shillings or 283 shillings for sure. This is an Exact-Exact choice. These numbers were chosen to bracket the same CRRA parameter. Subjects complete 4 low stakes risk tasks (Whole-Whole, Whole-Exact, Exact-Whole, and Exact-Exact), offered in a random order, and 2 high stakes risk tasks (Whole-Exact and Exact-Exact), also in a random order. Subjects complete 4 time tasks (Whole-Whole, Whole-Exact, Exact-Whole, and Exact-Exact), offered in a random order. Additionally, half of subjects are offered time tasks that fix the sooner payment and vary the later payment, while the other half of subjects are offered time tasks that fix the later payment and vary the sooner payment.

Design details

Randomization method
Randomization of the form of payment and order of tasks is done in zTree. Randomization of whether the sooner or later payment is fixed was done at the session level by computer.
Randomization unit
The form of payment and order of tasks is randomized at the individual level. The randomization of whether the sooner or later payment is fixed is randomized at the session level.
Clusters
The treatment is not clustered.
Observations
330 subjects.
Arms
Risk Cash, Time Cash, Sooner Fixed: 41 Risk Cash, Time Cash, Later Fixed: 41 Risk Cash, Time mPesa, Sooner Fixed: 41 Risk Cash, Time mPesa, Later Fixed: 41 Risk mPesa, Time Cash, Sooner Fixed: 41 Risk mPesa, Time Cash, Later Fixed: 41 Risk mPesa, Time mPesa, Sooner Fixed: 42 Risk mPesa, Time mPesa, Later Fixed: 42 The order of risk and time tasks is randomized at the individual level.
Primary outcomes
A binary indicator that the subject chose the gamble (sooner payment). Time taken to make the choice.
Secondary outcomes
Other measures of risk and time preference will be used for robustness checks, as specified in the pre-analysis plan.

Cited by

No citation data available for this registration yet.

Keywords