skip navigation

Saturday, December 20, 2014

[ – ] Text Size [ + ]  |  Print Page

Consumers’ Use of Prepaid Cards: A Transaction-Based Analysis

Open-loop prepaid cards are becoming a mainstream payment instrument in the United States. In addition, businesses and governments increasingly view them as a less expensive alternative to checks for disbursing payroll and benefits.

There are few studies that examine how consumers actually use prepaid cards, the revenues generated, and the resulting costs cardholders incur. This information gap poses a challenge for an industry in the early stages of refining its business models. It also poses a challenge for policymakers seeking to design cost-effective consumer protections for these products.

We analyze an anonymized data set of more than 280 million transactions made on more than 3 million cards issued by Meta Payment Systems. While the data have limitations and are not necessarily representative of the entire market, they can be used to establish a number of important stylized facts about the life cycle of prepaid cards.

We report detailed statistics on the longevity of prepaid cards and the intensity of their use, including the frequency and value of spending, the composition of spending at merchants, ATM withdrawals, and reloading of value onto cards. We identify cards that are likely enrolled in direct deposit and contrast usage of those cards with other cards in the data. We calculate statistics on revenues earned via consumer fees and the composition of those fees. We also estimate interchange revenues earned and thus quantify the significance of this funding source to the prepaid business model. Read the paper PDF.

Executive Summary

Prepaid cards currently account for a relatively small share of all consumer payments in the U.S., but they are becoming an increasingly important payment instrument to consumers, businesses, and government. According to aggregate statistics from the most recent edition of the Federal Reserve System's payments study, in 2009 in the U.S., there were 6 billion prepaid card transactions, valued at more than $140 billion. Prepaid transactions are growing substantially faster than transactions made with debit and credit cards. In addition, prepaid cards are replacing the remaining paper checks used to disburse Social Security disability, unemployment insurance, food stamps, and other government benefits. Despite the evolution in the use of prepaid cards, to date, there has been limited data available to study the prepaid card business model, about which little is known and which depends significantly on how consumers actually use these cards.

The Payment Cards Center of the Federal Reserve Bank of Philadelphia and the Center for Financial Services Innovation conducted this research project using transaction-level data provided to us by Meta Payment Systems in an effort to develop a better understanding of how consumers use prepaid cards by examining their transaction behavior and the issuer revenue and cardholder costs generated by those transactions. This study focuses primarily on a specific segment—open-loop reloadable prepaid cards— which account for approximately 30 percent of the prepaid transactions identified in the Federal Reserve System's payments study. This segment is growing much more rapidly than are prepaid cards in general. Two major sub- categories of open-loop prepaid cards are the focus of this study: general-purpose cards and payroll cards. This paper presents an analysis of more than 280 million transactions made on more than 3 million prepaid cards issued by Meta Payment Systems over a six-year period in over a dozen programs. While we cannot be certain these data are representative of the experience for the industry as a whole, we believe the statistics presented here will advance our understanding of prepaid cards and the way they are used.

For the cards we study, key findings include:

  • Prepaid cards are typically short-lived products. Prepaid cards offer much of the functionality of checking accounts, but that does not mean the underlying economics are the same. A typical prepaid card in the data is active for six months or less, a small fraction of the longevity seen with consumer checking accounts. As a result, account acquisition strategy and the recovery of fixed and variable costs are likely different than for checking accounts.
  • Cards typically exhibit modest use, but there is considerable variation. For the median or typical prepaid card in the data set, the number and the value of transactions are modest. That is also true for the value of funds loaded onto the card. There is very considerable variation in the usage of cards across an entire portfolio. For many of the statistics we calculate, usage has a U-shaped pattern; that is, a sizable share of the portfolio is used very little, while a smaller share of the portfolio (usually the top 20 percent of cards) is used intensively.
  • Based on this pattern of usage, issuer revenues earned from a typical card are modest, but there is considerable variation. Because prepaid cards are essentially transactional products, the patterns just described in terms of usage translate into the revenues earned by the prepaid card issuer as well as the costs incurred by the cardholder. For the typical prepaid card in the data, issuer revenues and cardholder costs per active card month are at most $12 and generally lower in most programs. Even among the most actively used prepaid cards, issuer revenues and cardholder costs are generally less than $20 per month.
  • Direct deposit is important to the economics of prepaid cards. An extremely important distinction for the prepaid cards we study is the presence (or lack thereof) of repeated value loads that appear to reflect direct deposit. While uncommon, general purpose prepaid cards with such patterns remain active at least twice as long (or longer) and have 10 times or more purchase and other activity than other cards in the same program category. These cards also generate at least four times more revenue for the prepaid card issuer. Average monthly cardholder costs are about twice as high, because these cards are used more intensively; on a per transaction basis, cardholder costs are lower.
  • Interchange is an important source of revenue for the prepaid issuer. Our calculations for issuer revenues include estimates of the revenues earned from interchange fees on point-of-sale purchases. While not as important as cardholder fees, interchange revenues account for more than one-fifth of the issuer revenues we observe in the general-purpose programs and almost half of the revenues in the payroll programs. According to our estimates, another kind of interchange—interchange fees paid by the prepaid issuer to owners of ATMs— are also not insignificant.
  • Prepaid cards are used for both purchases and cash withdrawals. To the extent they are used by consumers, most of the prepaid cards in the data are used for both cash withdrawals and purchases of goods and services. Depending on the type of prepaid card, cash withdrawals account for about one-third to one-half of the value taken off the card. The majority of purchase transactions occur at grocery stores, fast food restaurants, and gas stations. This suggests that prepaid cards are used primarily to purchase nondurable goods. Many prepaid cards in the data set are also used to pay bills.
  • The composition of fees paid by cardholders to the prepaid issuer varies by type of card. In value terms, the most significant fees incurred by cardholders include maintenance fees (primarily in certain general-purpose card programs) and ATM withdrawal fees, which account for a sizable share of fees in all programs.
  • ATM surcharges can account for a significant share of cardholder costs. Cardholders may also pay ATM surcharges, which are set and received by the owners of the ATMs they use. ATM surcharges are not uncommon in the data we study. Depending on the type of program, surcharges account for about 15 to 40 percent of the cardholder costs we observe.

 

About the Payment Cards Center

For more than 10 years, the Payment Cards Center of the Federal Reserve Bank of Philadelphia has provided insights into developments in the markets for consumer credit and payments. The center carries out its work through an agenda of applied research and analysis as well as forums and conferences that encourage dialogue incorporating industry, academic, and public-sector perspectives. For the latest research produced by and news about the Payment Cards Center, visit its web page.

About the Center for Financial Services Innovation External Link *

About Meta ® External Link *

 

* Nothing in the text should be construed as an endorsement of any organization or its products or services.

Frequently Asked Questions

Why study prepaid cards?

Open-loop prepaid cards are becoming a mainstream payment instrument in the United States. In addition, businesses and governments increasingly view them as a less expensive alternative to checks for disbursing payroll and benefits. According to the most recent edition of the Federal Reserve System’s payments study, in 2009 there were 6 billion prepaid card transactions, valued at more than $140 billion in the U.S. Prepaid transactions are growing substantially faster than transactions on debit and credit cards. Prepaid cards are also replacing the remaining paper checks used to disburse Social Security disability, unemployment insurance, food stamps, and other government benefits.

There are few studies that examine how consumers actually use prepaid cards, the revenues generated, and the resulting costs cardholders incur. This information gap poses a challenge for an industry in the early stages of refining its business models. It also poses a challenge for policymakers seeking to design cost-effective consumer protections for these products.

What does the data set contain?

The data set contains more than 280 million anonymized prepaid card transactions made across the U.S. on more than 3 million cards that were issued in 15 prepaid card programs. The data set includes transactions on cards issued over a six-year period that encompasses the most recent business cycle. But most of the transactions we observe occurred in the last two years of the data set: 2009 and 2010.

Among three of the programs in the data set, consumers purchased their general-purpose reloadable (GPR) cards via the web. In six programs, cards were purchased through a variety of brick-and-mortar retailers, including grocery stores, pharmacies, and check cashing outlets. In one program, GPR cards were sold to consumers at the branches of a financial institution. For the remaining five card programs that include three payroll programs, one transit program, and one flexible spending account program, the enrollment of cardholders occurred via their employers. Most of the transactions in the data set occur on cards in the payroll programs.

In terms of card distribution, cardholders received personalized cards via mail in all of the programs in the data set except one. In the financial institution program, customers received their cards in the branches at which the cards were purchased.

The card activity information in the data set includes the date of the transaction, the type of transaction, the dollar amount of the transaction, a merchant category code that allows us to classify the type of retailer at which the transaction was conducted, an alphanumeric field that often contains the merchant’s name or other descriptors of the transaction, and a network code indicating the network over which the transaction was processed.

What does the data set not contain?

We do not have demographic or financial information about the cardholders. All that we know about them is their ZIP code and that they obtained a card from a specific program. Also, we don’t observe the marketing information for the programs in the data set.

We observe transaction behavior on individual prepaid cards but not for individual cardholders. It is possible that more than one prepaid card in the data set was obtained by the same consumer. While we have unique anonymized (i.e., randomized) card numbers, we do not have a unique anonymized consumer identifier that would permit us to evaluate how a given cardholder uses all of his or her cards.

Are the data representative of the prepaid market as a whole?

Most of the transactions in the data set occur on cards in the payroll programs. The composition of cards in the data is probably not representative of the prepaid market as a whole. According to data from Mercator Advisory Group, payroll cards accounted for about 14 percent of value loads on open-loop prepaid cards in 2010.

We address concerns about representativeness by presenting separate statistics on transaction behavior in each program category of prepaid cards. We believe that, controlling for card type and distribution channel, our statistics are more likely to reflect patterns that are typical for similar categories of cards offered by other issuers (or across the industry).

Geographically, where are the cards used? Does this vary by program?

There are retail GPR cards in most counties of the U.S., but the distribution is far from uniform. There is a concentration in the West and in Florida that likely results from the specific geographic footprint of the retail companies where these cards were purchased.

Cards distributed via web GPR programs appear in more counties, and the distribution of cards per 10,000 of population is more uniform. This is not surprising given that this mode of distribution does not depend on the retail footprint of brick-and-mortar storefronts. Still, the web distribution shows some relative concentration in the South and Southeast in a pattern similar to the FDIC’s survey data on underbanked consumers.

A concentration of prepaid cards in the South and Southeast is also evident among the payroll cards in the data set. This concentration is driven by the geographic location of the establishments that provide payroll cards to some of their employees. These patterns may also be influenced by variation in state laws regulating payroll cards. In some counties, there is more than one payroll card for every 100 persons. Given that these data are from a single provider, and this market is still quite young, achieving a 1 percent penetration rate is a significant milestone.

What kinds of transactions are made using the cards in the data set?

Most of the prepaid cards in the data set are used for both cash withdrawals and purchases of goods and services. Depending on the type of prepaid card, cash withdrawals account for about one-third to one-half of the value taken off the card. The majority of purchase transactions occur at grocery stores, fast food restaurants, and gas stations. This suggests that prepaid cards are used primarily to purchase nondurable goods. Many prepaid cards in the data set are also used to pay bills.

What are the cardholder costs and issuer revenues associated with the cards in the data set?

First, it is important to note that card issuer revenues and direct cardholder costs are not the same for prepaid cards. For example, in the data we study, 20 percent or more of the revenues earned by the prepaid issuer is derived from interchange fees on point-of-sale transactions. Conversely, a portion (15 to 40 percent) of the costs some prepaid cardholders incur are ATM surcharges, which are determined and received by the owner of an ATM when the ATM used is not "in network" for the prepaid card program.

For the typical prepaid card in the data set, issuer revenues and cardholder costs per active card month are at most $12 and generally lower in most programs. Even among the most actively used prepaid cards, issuer revenues and cardholder costs are generally less than $20 per month.

The mix of fees paid by cardholders to the prepaid issuer in the data varies by type of card. In value terms, the most significant fees incurred by cardholders include maintenance fees (primarily in certain general purpose card programs) and ATM withdrawal fees, which account for a sizeable share of fees in all programs.

What can we say about profitability with these data?

While we have developed estimates of revenues the prepaid issuer earns on its cards, we have no information on fixed and variable program costs that would be necessary to make any conclusions about profitability.

However, we have identified patterns of card activity that influence the profitability of prepaid cards as well as account management strategy. For example, we calculated the variation by program in periods of dormancy longer than 90 days. For payroll and the retail or web GPR cards, dormancy of this length occurs in less than a fifth of cards in the data. Higher shares of cards with such long dormancies occur among cards distributed via the financial institution (45 percent) and especially among flexible spending account cards (80 percent).

Another example is the presence (or lack thereof) of repeated value loads that appear to reflect direct deposit. While uncommon, prepaid cards with such patterns remain active for much longer and have 10 times or more purchase and other activity than other cards in the same program category. They also generate at least four times more revenue for the prepaid card issuer. Cards with direct deposit are almost certainly the most profitable accounts in these programs.

As we do with our measures of card activity, we often present statistics that convey the variation across the portfolio of cards in each of the card categories. We do that by dividing cards in each category into five groups (quintiles) sorted on the basis of some measure of intensity or longevity. This information should contribute to a better understanding of the distribution of revenues generated across a card portfolio and the implications for the profitability of the portfolio.

Payment Cards Center Prepaid Card Publications

The following publications can be downloaded free of charge by clicking the links below:

2012:

2011:

2009:

2008:

2007:

2006:

2005:

2004:


  • Last update: August 7, 2012

Contact Us

Do you have feedback about the site? Can’t find what you’re looking for? Contact us. E-mail

E-Mail Notification

Find out when updates are made to the Consumer Credit and Payments section and when topical publications and events is released.

Sign Up