The American Economic Association's registry for randomized controlled trials
Cash & Stash Evaluation
Last registered on May 15, 2017
View Trial History
Cash & Stash Evaluation
Initial registration date
May 13, 2017
May 15, 2017 11:19 AM EDT
United States of America
New York City
Contact Primary Investigator
Other Primary Investigator(s)
Yale University, Innovations for Poverty Action
Additional Trial Information
Finance & Microfinance
Consumers face countless opportunities to borrow and spend on impulse. We work with a leading check casher in New York City to develop and test a new product that facilitates impulse saving. The product links check cashing transactions to a basic savings account by prompting customers to make a deposit every time they cash a check. Using a field experiment, we estimate the effects of receiving prompts relative to the account without the prompts on savings behavior.
Karlan, Dean and Jonathan Zinman. 2017. "Cash & Stash Evaluation ." AEA RCT Registry. May 15.
Karlan, Dean, Jonathan Zinman and Jonathan Zinman. 2017. "Cash & Stash Evaluation ." AEA RCT Registry. May 15.
Sponsors & Partners
We work with a leading check casher in New York City to develop and test a new product that facilitates impulse saving: "Cash & Stash," a savings account that enables customers to make savings deposits and withdrawals from the check cashing window. The core feature of the product is a prompt to save at the moment the customer realizes a substantial inflow of liquidity. RiteCheck partners with a credit union, first BethEx and now USAlliance Financial, to offer the savings account.This evaluation assesses whether behaviorally-informed product features—well-timed reminders and incentives to save–can encourage check cashing customers to build savings balances.
Intervention Start Date
Intervention End Date
Primary Outcomes (end points)
savings accounts balances, deposits, and withdrawals, both while treatments going and after treatments stop. trying to get data on borrowing from an alternative credit bureau
Primary Outcomes (explanation)
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
RiteCheck worked with the research team to randomize 872 Cash & Stash accountholders into one of four arms (see Figure 1 and next sub-section for details on sample creation and characteristics). Account functionality is identical across the four arms, which vary only in prompts and incentives to save. The four arms are:
T0) Account only. No additional incentives to save; no prompts to save after account-opening. This arm is akin to a basic saving account where the account-issuer does not actively promote usage among existing accountholders.
T1) Prompt only. Each transaction performed by these accountholders triggered a teller pop-up window that reminded tellers to prompt the customer to save: “Would you like to make a deposit into your Cash & Stash account?”
T2) Prompt + Sweepstakes. Each transaction performed by these accountholders triggered a teller pop-up window that reminded tellers to prompt the customer to save and to promote a sweepstakes incentive: “Would you like to make a deposit into your Cash & Stash account? Each dollar in your weekly balance is a sweepstakes entry. The bigger your balance the more chances you have to win!” At the end of each week, clients earned 1 sweepstake entry per dollar in their Cash & Stash account. RiteCheck conducted drawings for small prizes ($50 cash prize), and grand prize drawings ($500 cash prize) in July 2015, January 2016, and July 2016.
T3) Prompt + Rewards points. Each transaction performed by these accountholders triggered a teller pop-up window that reminded tellers to prompt the customer to save and to promote a rewards program incentive: “Would you like to make a deposit into your Cash & Stash account? Each week you get triple points for every dollar of your balance. The more you have in your account the more points you earn!” At the end of each week, clients earned 3 points per dollar in their Cash & Stash account, up to a limit of 15,000 points per week. 7,000 points is worth roughly $1.
Experimental Design Details
Randomization occurred at the individual account level, in one of two ways:
Batch randomization: Customers who opened Cash & Stash accounts during the initial pilot period (April 2013 through December 2014) and post-promotion period (May 2015 through May 2016) were randomized by IPA staff.
In-branch, on-the-spot randomization: During the promotional period (February 2015 through April 2015), clients that visited RiteCheck branches during peak hours (10:00am to 5:00pm) were approached by hired marketers with the offer to open a Cash & Stash savings account. Clients who agreed to open the account and submit their contact information to the marketer were randomized into one of the four treatment groups. Randomization took place using tablets equipped with a mobile data collection toolkit (SurveyCTO Collect), which used a built-in random number generator.
Was the treatment clustered?
Sample size: planned number of clusters
Sample size: planned number of observations
872 check cashing clients
Sample size (or number of clusters) by treatment arms
Nothing 245, Verbal 228, Sweepstakes 185, Points 214
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials
INSTITUTIONAL REVIEW BOARDS (IRBs)
Innovations for Poverty Action
IRB Approval Date
IRB Approval Number
Post Trial Information
Is the intervention completed?
Is data collection complete?
Is public data available?
Reports and Papers