Report to the Congress on Government-Administered, General-Use Prepaid Cards - 2017
Executive Summary
Section 1075 of the Dodd-Frank Wall Street Reform and Consumer Protection Act (Dodd-Frank Act), which added section 920 to the Electronic Fund Transfer Act (EFTA), requires the Federal Reserve Board (Board) to report annually to the Congress on the prevalence of use of general-use prepaid cards in federal, state, and local government-administered payment programs and on the interchange fees and cardholder fees charged with respect to the use of such cards.1
The Board collected 2016 data from 18 bank issuers, three nonbank financial institutions, and two federal government agencies on government programs that used prepaid cards as a method to disburse funds. Across reported programs, government offices disbursed $146 billion through prepaid cards in 2016, a 3 percent decrease from 2015. This was largely driven by decreases in prepaid card disbursements in state and local programs. Across reported program types, the largest decrease by value occurred in the Supplemental Nutritional Assistance Program (SNAP).
Bank issuers reported collecting more than $518 million in revenue across reported programs during 2016, with 63 percent from interchange fees, 31 percent from cardholder fees, and 6 percent from other sources. Issuers collected 1.1 percent of the value of purchase transactions in interchange fees for these programs in 2016, a slight decrease from 2015. Revenue from cardholder fees as a percent of funds disbursed onto prepaid cards was unchanged at 0.2 percent in 2016. Automated teller machine (ATM) withdrawal fees continued to constitute the largest source of cardholder fee revenue that issuers collected in 2016. Account servicing fees and customer service fees continued to constitute the next-largest sources of cardholder fee revenue, while overdraft fees remained the smallest source.
Background
Federal, state, and local government offices use prepaid cards to disburse funds at a lower cost than checks (or other paper-based payment instruments such as vouchers or coupons) and to provide an alternative to direct deposit for payment recipients, especially those recipients who do not have bank accounts. Government offices contract with financial institutions to issue prepaid cards, disburse program funds, and provide customer service.2
Prepaid cards are a common disbursement mechanism in the government sector. Today, all states disburse SNAP benefits exclusively through electronic benefit transfer (EBT) cards, and every state offers a prepaid card option for child support programs.3 In addition, nearly every state offers a prepaid card option for unemployment insurance and Temporary Assistance for Needy Families (TANF) programs. Maps 1-6 show prepaid card adoption across common state government programs.4
A number of government offices now mandate electronic disbursement of program funds, through either a prepaid card or direct deposit. As of March 2013, the U.S. Treasury Department requires all-electronic disbursement of funds under federal benefit programs, such as Social Security and Veterans Affairs programs.5 In addition, all states are required to replace paper vouchers with an EBT card system for the Women, Infants, and Children (WIC) Supplemental Nutrition Program by October 1, 2020.6 As of May 2017, 20 state WIC agencies have completed the conversion to EBT cards, and most other states are either in the planning or implementation phase of the conversion process.7
Survey Data and Results
The Board distributed a survey to 18 issuers to collect prevalence-of-use and fee data on federal, state, and local government-administered payment programs that used prepaid cards as a method to disburse funds in calendar year 2016.8 All issuers responded, reporting data for more than 2,500 programs.9 The Board also collected a limited amount of prevalence-of-use data from the U.S. Treasury, the U.S. Department of Agriculture, and three nonbank financial institutions that process EBT transactions.10 The data collected represent programs from all 50 states and the District of Columbia.
As a result of several methodological changes to the data collection process beginning with the data collection for calendar year 2014, many of the figures in this report are not directly comparable with corresponding figures in previous reports.11 Where possible, the Board provided a limited number of year-over-year comparisons in this report. As the Board collects additional years of data under the revised data collection process, the Board intends to provide more information on general trends in these reports regarding the government-administered prepaid card market.
Prevalence of Use
Calendar Year 2016: Prepaid Card Disbursements
For calendar year 2016, government agencies disbursed $146 billion through prepaid cards across reported programs.12 These prepaid disbursements represent approximately 2.4 percent of total government expenditures in 2016.13
Total funds disbursed through prepaid cards varied widely by program type, as shown in figure 1. State agencies administering SNAP disbursed $66 billion to EBT cards in 2016, accounting for just under half of prepaid funding across all reported programs. The Social Security Administration distributed approximately $32 billion though prepaid cards, representing about one-fifth of all reported funds disbursed through prepaid cards.14 Unemployment insurance, child support, and cash assistance programs largely accounted for the remaining one-third.15 In contrast, WIC, Veterans Affairs, tax refund, and payroll programs each accounted for less than 1 percent of total prepaid disbursements.
Figure 1. Funds disbursed through prepaid cards in 2016 by program type
Across the largest program types that provide a prepaid card disbursement option, state and local programs tend to have higher rates of prepaid card use than federal programs.16 Funds disbursed under state SNAP programs, for example, are distributed exclusively through prepaid cards. Similarly, state agencies disbursed a relatively high proportion (65.6 percent) of unemployment insurance payments through prepaid cards.17 In contrast, prepaid card disbursements represent only 3.6 percent of the $898 billion in benefit payments made under federal Social Security programs.18 Similarly, the Department of Veterans Affairs disbursed only 1.2 percent of program funds through prepaid cards.
Year-over-Year Comparisons
Across reported programs, disbursements to prepaid cards decreased 3 percent between 2015 and 2016, from $150 billion to $146 billion. As shown in table 1, this decrease reflects a $1 billion decline in prepaid card disbursements across reported federal programs and a $3 billion decline across state and local programs. The decrease in federal card funding is largely due to a shift in the timing of certain payments made under Social Security programs (payments that otherwise would have fallen on New Year's Day or the first weekend in January 2016 were shifted to December 31, 2015). The reduction in state and local card funding is, in part, attributable to a $3 billion drop in SNAP benefits from 2015 to 2016, consistent with a decline of 2 million individuals receiving SNAP benefits.19 The decline in state and local card funding is also partially attributable to a 2 percent decrease in unemployment insurance benefits disbursed by prepaid cards, consistent with continued declines in the number of individuals collecting unemployment insurance benefits. According to data obtained from the U.S. Department of Labor, the number of unemployment insurance beneficiaries dropped 2 percent from 2015 to 2016.20 The declines in state and local prepaid card funding were partially offset by modest increases in child support and cash assistance programs.
Table 1. Prepaid card funding by jurisdiction and program type, 2014-16
Billions of dollars, except as noted
| Jurisdiction/program type | 2014 | 2015 | 2016 | 2015-16 percent change |
|---|---|---|---|---|
| Total | 148.0 | 149.6 | 145.7 | -3 |
| Jurisdiction | ||||
| Federal programs | 31.2 | 34.9 | 33.9 | -3 |
| State and local programs | 116.8 | 114.7 | 111.8 | -3 |
| Program type | ||||
| SNAP | 69.6 | 69.0 | 65.7 | -5 |
| Social Security | 29.9 | 33.2 | 32.2 | -3 |
| Unemployment | 22.8 | 21.1 | 20.8 | -2 |
| Child support | 12.7 | 12.9 | 13.6 | 5 |
| Cash assistance | 9.1 | 8.3 | 8.5 | 2 |
| All other (low-value categories) | 3.9 | 5.0 | 4.9 | -2 |
Fees Collected by Issuers
Issuers receive revenue from two main sources: interchange fees and cardholder fees.21 Less commonly, issuers also receive incentive payments from payment card networks (PCNs) and fees from government offices for providing prepaid services.22 In 2016, issuers reported collecting $328 million in interchange fees, $158 million in cardholder fees, $24 million in PCN incentives, and $8 million in fees assessed to government offices.23 Figure 2 illustrates the various sources of revenue collected by issuers in 2016. Revenue from these sources decreased 1 percent across reported programs from 2015 to 2016.
Figure 2. Issuer revenue by source, 2016
Note: Percentages in pie chart do not sum to 100 due to rounding.
Interchange Fees
As shown in table 2, the interchange fees from prepaid card purchase transactions have been relatively stable for the last six years. Across all programs from 2015 to 2016, the average interchange fee per purchase transaction decreased by 1 cent (from $0.35 to $0.34), and the average purchase transaction value increased by 79 cents (from $29.50 to $30.29). The average interchange fee as a percentage of purchase transaction value declined slightly from 1.2 percent to 1.1 percent in 2016.
Table 2. Average purchase transaction value, average interchange fees, 2011-2016
| Item | 2011 1 | 2012 1 | 2013 | 2014 | 2015 | 2016 |
|---|---|---|---|---|---|---|
| Average value of purchase transaction 2 | $30.94 | $29.99 | $29.94 | $29.16 | $29.50 | $30.29 |
| Federal programs | $36.82 | $38.11 | $44.83 | $38.53 | $38.41 | $38.61 |
| State and local programs | $29.81 | $28.14 | $25.58 | $24.85 | $25.07 | $25.68 |
| Average interchange fee per purchase transaction 3 | $0.33 | $0.34 | $0.36 | $0.34 | $0.35 | $0.34 |
| Federal programs | $0.40 | $0.40 | $0.48 | $0.41 | $0.41 | $0.40 |
| State and local programs | $0.32 | $0.33 | $0.33 | $0.31 | $0.32 | $0.31 |
| Average interchange fee as % of purchase transaction value4 | 1.1% | 1.1% | 1.2% | 1.2% | 1.2% | 1.1% |
| Federal programs | 1.1% | 1.1% | 1.1% | 1.1% | 1.1% | 1.0% |
| State and local programs | 1.1% | 1.2% | 1.3% | 1.2% | 1.3% | 1.2% |
1. For calendar years 2011 and 2012, some program data reported in aggregate could not be allocated between federal and state and local programs. These data, however, are reflected in the total figure. Return to table
2. Average value of purchase transaction: Value of settled purchase transactions divided by the number of settled purchase transactions. Return to table
3. Average interchange fee per purchase transaction: Total interchange fees divided by the number of settled purchase transactions. Return to table
4. Average interchange fee as % of purchase transaction value: Total interchange fees divided by the value of settled purchase transactions. Return to table
The average interchange fee as a percentage of purchase transaction value declined from 1.1 to 1.0 percent for federal programs and from 1.3 to 1.2 percent for state and local programs in 2016. Like other year-over-year comparisons in this report, the calculations of average purchase transaction values and average interchange fees are sensitive to changes in the pool of reported programs, which differs from year to year.24
Transactions made using government prepaid cards are generally exempt from the interchange fee standards of Regulation II. The average interchange fee as a percentage of purchase transaction value for government prepaid card transactions is similar to that for all debit card transactions exempt from the interchange fee caps of Regulation II.25
Cardholder Fees
The data provided by issuers show that cardholder fee revenue as a percentage of program funds disbursed by prepaid card was unchanged between 2015 and 2016 at 0.2 percent.
In addition to negotiating cardholder fee rates, government offices often restrict the number and type of cardholder fees that an issuer can charge. Most government offices require issuers to offer cardholders a certain number of free ATM or over-the-counter (OTC) cash withdrawals and prohibit issuers from charging cardholders certain types of fees, such as monthly maintenance fees.26
Figure 3 illustrates the total revenue issuers collected and the average charge per occurrence by cardholder fee type in 2016.27 Similar to 2015, fees on ATM cash withdrawals constituted the largest source of revenue issuers received from cardholder fees. Issuers reported collecting $89.2 million in revenue from ATM withdrawal fees in 2016, charging $1.00 per occurrence on average. Account servicing fees and customer service inquiry fees continued to account for the next largest sources of cardholder fee revenue, at $42.9 million and $17.2 million, respectively. Account servicing fees remained the most expensive type of cardholder fee, at an average of $3.10 per occurrence in 2016. Revenue from penalty fees declined 50 percent year over year, from $5.8 million in 2015 to $2.9 million in 2016. Overdraft fees continued to be the smallest source of cardholder fee revenue, accounting for less than 0.3 percent of total cardholder fee revenue in 2016.28
Figure 3. Cardholder fees: revenue and average fee charged in 2016, by type of fee
Note: Size of bubble represents total revenue from fee.
Position of black dot on vertical axis represents average fee when charged.
Although figures 2 and 3 illustrate the distribution of cardholder fee revenue on an aggregate level, this distribution is not representative of any particular program. Government offices and issuers negotiate a cardholder fee schedule for each program. Furthermore, the proportion of transactions resulting in the assessment of a cardholder fee depends heavily on the type of program. As a result, there is significant heterogeneity in cardholder fees across programs that use prepaid cards to disburse funds.
Fees Paid by Issuers
Issuers pay fees to third parties when a cardholder withdraws cash from an out-of-network ATM or bank.29 In 2016, issuers reported paying approximately $61 million in fees to third parties for ATM withdrawals and $24 million in fees to third parties for OTC cash withdrawals.30
Footnotes
1. A program is considered government-administered regardless of whether a federal, state, or local government office operates the program or outsources some or all functions to third parties, so long as the program is operated on behalf of a government office. In addition, a program may be government-administered even if a federal, state, or local government office is not the source of funds for the program it administers. For example, child support programs are government-administered programs even though they are funded by individuals. Return to text
2. Occasionally, third-party program managers are involved in government prepaid card programs. Issuers often contract with program managers to provide services traditionally carried out by the issuer, for example, call center management and transaction processing. In certain cases, a program manager contracts directly with the government office and provides almost all functions that are traditionally carried out by an issuer. Return to text
3. Subsections 1075(b)-(d) of the Dodd-Frank Act amend benefits statutes such that EBT cards issued in connection with the relevant program are not subject to the interchange fee standards and exclusivity and routing restrictions in section 920 of the EFTA. The amended benefits statutes are the Food and Nutrition Act of 2008, the Farm Security and Rural Investment Act of 2002, and the Child Nutrition Act of 1966. Although EBT cards issued in connection with relevant programs are not subject to section 920 of the EFTA, the Board believes that it is appropriate to include in this report information about such EBT cards because they represent a significant portion of prepaid cards issued pursuant to government-administered programs. Return to text
4. Map data are based on publicly available information from state agency websites and the U.S. Department of Agriculture's Food and Nutrition Service. Return to text
5. See 31 CFR 208.3-4. The U.S. Treasury's rule provides for certain limited waivers from the all-electronic requirement. Waivers may be granted to recipients born on or before May 1, 1921, who were receiving payments by check on March 1, 2013. Waivers may also be granted in instances when the requirement would be overly burdensome for the recipient, for example, in cases where the recipient suffers from a mental impairment or resides in a remote geographic location lacking the infrastructure to support electronic financial transactions. Return to text
6. Healthy, Hunger-Free Kids Act of 2010, Pub. L. No. 111-296, § 352(d), 121 Stat. 3183, 3255 (2010). Return to text
7. For more information on states' progress in converting to WIC EBT card systems, see "Women, Infants and Children (WIC), WIC EBT Activities," U. S. Department of Agriculture, accessed May 2017, www.fns.usda.gov/wic/wic-ebt-activities. Return to text
8. "Government-Administered, General-Use Prepaid Card Survey--Issuer Survey," FR 3063a, OMB No. 7100-0343, Federal Reserve Board website, www.federalreserve.gov/paymentsystems/files/FR3063a_government_issuer_survey_2016.pdf. The FR 3063a survey is conducted annually to collect information for this report. The Board identified issuers to survey by consulting with relevant payment card networks. The Board reviewed the data submitted by survey respondents for completeness, consistency, and anomalous response. Where possible, the Board resolved identified issues by following up with respondents and by replacing problematic data elements with imputed values. In computing a given summary statistic for this report, the Board excluded responses with unresolved issues that affected the calculation's inputs. Although the Board makes a comprehensive effort to identify and resolve issues in the reported data, some issues may not have been identified by the time this report was published. Return to text
9. Issuers self-reported the number of programs included in their responses. Some issuers were unable to distinguish between multiple government-administered payment programs that disburse funds on the same card. The number of programs, therefore, represents a lower-bound approximation. Issuers occasionally partner with third-party program managers that contract directly with government agencies. In these cases, issuers may not have access to some data requested in the survey. Return to text
10. Program counts were not available from these organizations. Return to text
11. The Board made several material changes to the data collection process, beginning with the calendar year 2014 data collection. For example, the Board discontinued the FR 3063b survey of government agencies (government survey) and instead relied on the data collected from the FR 3063a survey of issuers (issuer survey) and publicly available government expenditure data to calculate prevalence-of-use metrics. In addition, the Board significantly revised and simplified the reporting process for the issuer survey. For previous Board reports to the Congress on government-administered, general-use prepaid cards, see "Payment Research," Federal Reserve Board website, last modified December 22, 2016, www.federalreserve.gov/paymentsystems/payres_papers.htm. Return to text
12. Funds disbursed onto prepaid cards but ultimately returned to government agencies are excluded from this figure. The value of funds returned to government agencies is de minimis (approximately 0.3 percent of gross funds disbursed through prepaid cards in 2016). Return to text
13. "Table 3.1. Government Current Receipts and Expenditures," U.S. Department of Commerce, Bureau of Economic Analysis, last modified May 26, 2017, www.bea.gov/national/pdf/SNTables.pdf. Total government expenditures include expenditures of federal, state, and local governments. Return to text
14. For the purposes of this report, Social Security programs include Old-Age, Survivors, and Disability Insurance and Supplemental Security Income. Return to text
15. Cash assistance includes TANF, Low Income Home Energy Assistance Program, child care, refugee assistance, and general assistance programs. Return to text
16. Federal programs such as Social Security have a higher proportion of enrollees that receive benefits through direct deposits to bank accounts than enrollees that receive benefits through prepaid cards. Return to text
17. "Monthly Program and Financial Data," U.S. Department of Labor (DOL), accessed June 2017, https://workforcesecurity.doleta.gov/unemploy/claimssum.asp. Disbursements for state unemployment insurance totaled $31.7 billion in 2016. Return to text
18. Information about total payments made under Social Security programs was provided by the U.S. Treasury. Return to text
19. "Supplemental Nutrition Assistance Program (SNAP)," Food and Nutrition Service, last published June 9, 2017, www.fns.usda.gov/pd/supplemental-nutrition-assistance-program-snap. Return to text
20. "Unemployment Insurance Data," DOL, accessed June 2017, https://workforcesecurity.doleta.gov/unemploy/DataDashboard.asp. For this comparison, the number of beneficiaries includes beneficiaries of regular state unemployment compensation as well as beneficiaries of extended benefit programs. Return to text
21. 12 CFR 235.2(j). An interchange fee is any fee established, charged, or received by a payment card network and paid by a merchant or a merchant acquirer for the purpose of compensating an issuer for its involvement in an electronic debit transaction. Merchant acquirers typically pass the cost of these fees on to merchants. Interchange fees, in effect, are a cost to merchants and a source of revenue to issuers. Return to text
22. Incentive payments are payments received by an issuer from a PCN with respect to debit card transactions or debit-card-related activity. Issuer incentives may be based on reaching specified volume levels, promoting the network's brand through marketing activities, converting the issuer's debit card base to a different signature network, or undertaking other activities. Incentive payments do not include payments from a network to an issuer for traditional banking services the issuer provides to the network (for example, transaction account services to the network). Issuers may also receive revenue from interest on program funds held in pooled bank accounts before the beneficiaries use them; however, the Board does not gather data on this source of revenue. Return to text
23. Consistent with previous reports, the Board calculated all
revenue figures here using data collected through the issuer survey on the population of network-branded government-administered, general-use prepaid cards (excluding EBT cards). For EBT card programs, states generally pay an issuer or processor based on the number of beneficiaries enrolled in a program per month, in part because there are no interchange fees associated with these card programs. Certain cardholder fees, such as a fee for card replacement, may also apply to EBT programs. Return to text
24. The pool of reported programs changes over time because government agencies may add or eliminate programs and may alter disbursement methods for existing programs. Return to text
25. The average interchange fee as a percentage of purchase transaction value for all debit card transactions exempt from the interchange fee caps was 1.15 percent in 2016. See "Regulation II (Debit Card Interchange Fees and Routing), Average Debit Card Interchange Fee by Payment Card Network," Board, www.federalreserve.gov/paymentsystems/regii-average-interchange-fee.htm. Network-branded government-administered, general-use prepaid card transactions constituted approximately 4 percent of all exempt transactions in 2016. Return to text
26. For definitions of the various cardholder fees mentioned in this report, see "Government-Administered, General-Use Prepaid Card Survey--Issuer Survey," FR 3063a, OMB No. 7100-0343, Board, www.federalreserve.gov/paymentsystems/files/FR3063a_government_issuer_survey_2016.pdf. Return to text
27. The Board excluded transactions for which no fee was assessed from the average fee calculations. Return to text
28. As of July 21, 2012, prepaid cards that may incur overdraft fees are ineligible for the exemption from the interchange fee standards of Regulation II. It appears that, rather than lose the exemption, issuers have largely abandoned overdraft fees. 15 USC 1693o-2(a)(7)(B). Pursuant to section 920 (a)(7)(B) of the EFTA, the exemption from the interchange fee standards of Regulation II does not apply if, on or after July 21, 2012, the issuer may charge the cardholder an overdraft fee with respect to the card, or an ATM fee for the first withdrawal per calendar month from an ATM that is part of the issuer's network. Return to text
29. Issuers pay fees to ATM operators for each ATM cash withdrawal to compensate the operator for the costs of deploying and maintaining the ATMs and of providing cash services to the issuers' cardholders. Issuers pay fees to banks for each OTC cash withdrawal to compensate the bank for the costs of staffing the teller window and providing cash services to the issuers' cardholders. In addition to ATM and OTC fees, issuers pay fees to payment card networks (such as switch, license, and connectivity fees). The Board does not survey issuers of government-administered prepaid cards regarding network fees. Across all debit cards, issuers paid networks approximately 4.2 cents per transaction in 2015. For more information, see "Regulation II (Debit Card Interchange Fees and Routing), Reports and Data Collections," Federal Reserve Board website, www.federalreserve.gov/paymentsystems/regii-data-collections.htm. Return to text
30. On average, issuers paid approximately 40 cents per ATM cash withdrawal and $2.25 per OTC cash withdrawal in 2016. Because of limited data, the Board approximated the number of ATM and OTC withdrawals resulting in a fee. Therefore, the calculations of average ATM and OTC fees paid by issuers are estimates. Return to text