The Demand for Status: Evidence from Platinum Credit Cards
Last registered on September 13, 2015


Trial Information
General Information
The Demand for Status: Evidence from Platinum Credit Cards
Initial registration date
September 13, 2015
Last updated
September 13, 2015 12:32 PM EDT
Primary Investigator
Bocconi University
Other Primary Investigator(s)
PI Affiliation
Harvard University
PI Affiliation
The World Bank
PI Affiliation
Sao Paulo School of Economics, FGV
PI Affiliation
UCLA Anderson School of Management
Additional Trial Information
In development
Start date
End date
Secondary IDs
Consumers might choose to purchase certain products to show to others their own economic achievements and thus gain social status. Purchasing an exclusive product may also shape one' view of oneself, regardless of the opinion of others. We consider premium credit cards, which can be demanded not only for the services and benefits that they provide, but also for social image and self-image motives. Since qualification requirements for these cards are usually high, this is a financial product generally available only to wealthy individuals. This exclusivity makes it a well-know symbol of success, which might enhance the owner's social image and self-image. We design a phone marketing experiment to disentangle the roles of self- versus social image considerations as well as instrumental benefits in explaining the demand for a platinum card.
External Link(s)
Registration Citation
Bursztyn, Leonardo et al. 2015. "The Demand for Status: Evidence from Platinum Credit Cards." AEA RCT Registry. September 13.
Former Citation
Bursztyn, Leonardo et al. 2015. "The Demand for Status: Evidence from Platinum Credit Cards." AEA RCT Registry. September 13.
Sponsors & Partners

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

Request Information
Experimental Details
We design a phone marketing experiment in which we randomize the characteristics of the product being offered and the scripts used for the call.
First, some clients are offered to replace their non-premium card by a premium card, while others are offered all of the extra services of a premium card on their existing non-premium card. Customers will thus be offered the same instrumental services: the only thing differing is whether the services are included in their present relatively nondescript card, or on the premium platinum card which might have an additional "status" component. If the share accepting the premium card is higher than the share accepting the added benefits to the non-premium card, we will have established that customers value the status associated with the exclusive premium card.
To further disentangle the nature of the demand for status -- social vs. self-image -- we proceed with additional treatments.
Some clients will be told that they were selected on the basis of their merit, while others will be told that they were chosen on the basis of luck. Both of these statements are true, since customers receiving these offers are randomly selected among those who qualify for the offer. The script mentioning merit should activate both social and self-image channels -- the opportunity to build self-image since the offer is an acknowledgment of the customer's financial success, and social image since the distinctive platinum card is well recognized by others as a marker of elite status. On the other hand, the luck script should activate only the social image aspect. If concerns about self-image are an important component of demand, the share of customers accepting the offers with the merit script should be higher than that with the luck script. A higher take up with the luck script indicates instead that self-image and social image might be substitutes: customers who received a positive shock to their self-image (being told they can have access to an exclusive good because of their economic success) are less concerned about social image considerations, and so have a lower demand for premium cards.
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
Primary Outcome: offer take-up decision
Primary Outcomes (explanation)
We ask about the customer willingness to take up the offer at the end of the call, and code this as a "yes" or "no".
If a customer decides to interrupt the call before the caller asked about her willingness to take up the offer, but after the caller described the details of the offer, then the take-up decision will be coded as "no" as well.
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
Customers will be randomly assigned into one of the following groups: (1) premium card and merit script, (2) premium card and luck script, and (3) non-premium card and luck script.
Experimental Design Details
Randomization Method
The randomization was conducted using a computer random number generator.
Randomization Unit
We randomized at the level of the individual customer. We stratify by income category (below 300 millions IDR, between 300 millions and 500 millions IDR, above 500 millions IDR) and annual fee (no annual fee or 240,000 IDR).
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
No clustering.
Sample size: planned number of observations
Sample size (or number of clusters) by treatment arms
Sample size will be: 420 customers in treatment (1) premium card and merit script, 420 customers in treatment (2) premium card and luck script, and 420 customers in treatment (3) non-premium card and luck script.
These are clients the bank will attempt to call: the final number of observations will depend on their response rate.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
With an expected answer rate of 70%, we have a MDE of 0.23 standard deviation for the comparison between treatment (2) and treatment (3), and for the comparison between treatment (1) and treatment (2). If there is no statistical difference between treatment (1) and (2), we can pool those two treatments and compare them with treatment (3): in this case we would have a MDE of 0.2 standard deviations. This is a conservative calculation of the MDE, not accounting for the added precision due to stratification on income category and annual fee.
IRB Name
UCLA Institutional Review Board
IRB Approval Date
IRB Approval Number
Post Trial Information
Study Withdrawal
Is the intervention completed?
Is data collection complete?
Data Publication
Data Publication
Is public data available?
Program Files
Program Files
Reports, Papers & Other Materials
Relevant Paper(s)