Take On Payments, a blog sponsored by the Retail Payments Risk Forum of the Federal Reserve Bank of Atlanta, is intended to foster dialogue on emerging risks in retail payment systems and enhance collaborative efforts to improve risk detection and mitigation. We encourage your active participation in Take on Payments and look forward to collaborating with you.
Federal Reserve Web Sites
Other Bank Regulatory Sites
October 17, 2016
EMV Comments That Make Me Cringe
Some aspects of the chip card implementation in the United States certainly make us frustrated. For one, the customer experience could be seen as slightly more negative because of the longer transaction time and confusion about the debit card selection menu. However, at several payments conferences I have attended recently, I have heard comments made by speakers and panelists about EMV chip cards and their technology that caused me to cringe a bit. I understand that a number of stakeholders are not proponents of EMV technology for a variety of reasons and, while some parts of their comments are factually accurate, they certainly are not "the truth, the whole truth and nothing but the truth."
Cringe #1: The United States is implementing 20-year-old-technology with EMV chip cards. Yes, the first EMV specifications were publicly released in 1995. But isn't that like saying that the gasoline-powered automobile is technology that is 130 years old? Microsoft's first release of Windows was in 1985. Do we hear complaints about it being 30-plus years old? The reality is that the EMV specifications, like practically all software development, are continually updated over the years with enhancements continuing as long as the software is still being supported. The EMV specifications are now at version 4.3, released in November 2011, with 20 supplemental bulletins issued since then and more on the way.
Cringe #2: EMV (chip) cards haven't solved the card-not-present (CNP) fraud problem. Again, this is an accurate statement. CNP card fraud is the second largest category of fraud losses in the U.S. (see the chart). But, the statement is misleading inasmuch as the EMV specifications and chip cards were never intended to address the CNP ecommerce environment. Counterfeit card fraud, whereby the criminal produces a card using data obtained from a skimmer or data breach, has been the number-one source of card-present fraud in the United States. It was this type of card fraud that the chip card was designed to target, and, from all accounts to date, it has been highly successful in doing so.
Source: Chip Cards in the United States: The PIN, PINless, Debit, Credit Conundrum, Aite Group, July 2016
Cringe #3 – Using a PIN improves the security of the chip card. While a cardholder using a PIN in lieu of a signature does clearly result in a lower level of fraud losses, the claim is somewhat of an apples and oranges comparison. The chip on the card authenticates the card itself, while the use of a PIN is intended to authenticate the cardholder performing the transaction. These are two separate types of authentication which, when combined, make the transaction more secure—a good thing. The use of a PIN should result in lower lost/stolen card fraud as it invokes two-factor authentication—something you have (card) and something you know (PIN).
Are the current EMV specifications perfect? Of course not, and that is why there are constant efforts to identify ways to improve them. But one must recall that the EMV specifications provide global interoperability and must be developed keeping that requirement in mind. What are your thoughts on the EMV specifications and how they can be improved?
By David Lott, a payments risk expert in the Retail Payments Risk Forum at the Atlanta Fed
May 16, 2016
Improving Customer Authentication: Is the PIN Past Its Prime?
The Financial Fraud Action UK recently released its Year-End 2015 Fraud Update. This report, filled with fraud-related figures from a fully EMV(chip)-migrated country, provides insight into what the future of fraud in the United States might look like as we are approximately eight months into our EMV journey. And if indeed the United Kingdom’s experience is a harbinger of things to come in the United States, then I think there will be disappointment for anyone who thought EMV by itself would be a magic bullet. After I spent time studying this report, it became evident that customer authentication is the latest low-hanging fruit and fraudsters are having a feast.
Fraud losses on payment cards in the United Kingdom (£567.5m) are approaching pre-EMV migration levels, and fraud loss rates have increased above 8 basis points (0.08%), hitting a level last seen in 2009. Diving deeper, we find that:
- As expected, card-not-present (CNP) fraud losses represent a majority of card fraud losses (70 percent). Interestingly though, ecommerce spend volume grew faster than ecommerce fraud losses in 2015, suggesting that the industry made headway in its efforts to mitigate ecommerce fraud.
- Lost and stolen card fraud (remember, the United Kingdom is a PIN environment) increased more than 24 percent in 2015, reaching levels last seen in 2006. The report highlights distraction thefts through cameras or simply shoulder surfing as methods of fraudulently obtaining PINs.
- Card ID theft fraud losses, defined as losses from spend on fraudulently opened or obtained cards through stolen personal information, increased by 28 percent and are now approaching counterfeit card levels.
- A bit of good news is that counterfeit card fraud losses remain well below pre-EMV levels and fell even further in 2015—perhaps, as the report suggests, driven partly by the increased acceptance of EMV cards in the United States.
- Beyond cards, remote banking fraud losses (losses from Internet, telephone, and mobile banking) increased by more than 134 percent during the last two years, totaling nearly £169 million.
EMV is performing exactly as expected and doing a phenomenal job of authenticating payment cards in the card-present environment. Why are fraud losses increasing in a mature EMV environment? Because customer authentication remains a challenge, as is evident by rising fraud losses from lost and stolen cards, card applications with stolen identities, and remote banking.
Whether on the front end of authenticating the user during the account opening process or the back end of authenticating the user at the time of payment, authentication measures are coming up short, and these measures include PINs and passwords. Replacing passwords has been an ongoing conversation and likely may continue to be a conversation piece rather than a prolific action item. Yet there is a growing push for the use of PINs coupled with EMV cards here in the United States. While PIN authentication is an improvement over signature authentication, it, too, has its flaws. With improvements and advancements in new technologies such as biometrics, perhaps it's time for the industry to advance beyond PINs. Because of the current signature-laden EMV environment in the U.S., the timing is perfect.
By Douglas A. King, payments risk expert in the Retail Payments Risk Forum at the Atlanta Fed
February 22, 2016
2016 Payment Predictions
In our 2015 year-end review, we promised we would provide some predictions and expectations for payments in the United States during 2016. Predictions are usually pretty…unpredictable, so by waiting a couple of months to release ours, we're hoping they will end up being more accurate than usual. Disclaimer: These predictions are through the collective wisdom of the Retail Payments Risk Forum staff and do not reflect the opinions of the Federal Reserve System or the Board of Governors. So here we go in no particular order or probability of happening.
- Cyberattacks will be the top threat to payments security: Cyberattacks and data breaches will be as robust as ever and will be the number one threat in the payments ecosystem. As retailers and financial service companies strengthen their defenses, the Risk Forum predicts that hackers will widen their focus.
- This will be the year for mobile point-of-service (POS) payments…not!: Like the broken analog clock face that is correct twice a day, we believe that those forecasting 2016 as the "year of mobile payments" (as they did in 2013, 2014, and 2015) will be a little bit right, but will still be waiting for this optimistic prediction to be fully true. While the adoption pace of mobile payments is growing because of the increasing influence of millennials, the issues of limited merchant acceptance points, fragmentation, and consumer concerns over security and privacy will remain as substantial hurdles. Major educational efforts will be launched stressing the increased security provided by mobile payments through tokenization and biometrics.
- EMV (chip card) POS migration will pick up the pace from 2015: The liability shift for POS took place October 1, 2015, and projections for both card and terminal capability missed their optimistic marks for a variety of reasons. Credit and debit card reissuance will continue during 2016 and should reach significant conversion levels by the end of the year. The Risk Forum expects the pace of merchant terminal conversions to pick up as certifications are completed and merchants targeted by counterfeit card fraudsters feel the sting of losses. However, we also think some merchant categories, such as restaurants, will continue to proceed at a tepid pace.
- ACH same-day service will not be a huge hit: The Risk Forum forecasts that the roll-out of NACHA's mandated same-day ACH service in September will, at least initially, have modest adoption because corporate originators will have to update internal systems to support faster payments, the dollar cap of $25,000 per payment, and the imposition of the interbank fee. Consumer payment applications will have modest uptake due to competing payment alternatives.
- EMV ATM liability shift will cause the number of ATMs to shrink: The implementation of chip card readers in ATMs will follow the same pattern as POS terminals did in 2015—the large ATM owners and operators will meet the October 2016 deadline but many of the small and mid-sized operators, especially those owned by nonfinancial institutions, will not and will be faced with absorbing the loss of transactions made with counterfeit cards—a fraud loss they haven't experienced in the past. Overall, the Risk Forum looks for the ATM base in the U.S. to contract by 10 to 15 percent because of financial institution mergers and the cost of EMV upgrades.
- Mobile wallet space will continue to see turbulence: 2015 saw the launch or announcement of more mobile wallets by payment stakeholders such as Samsung, Google, Chase, Capital One, Walmart, and Target. Then add the retailer and credit union consortiums (MCX CurrentC and CU Wallet) that are struggling to emerge from uncertainty. How many wallets will the consumer be willing to load on a phone and which providers do they trust to keep their payments and banking credentials safe? We believe we'll see continued turbulence in this space during 2016, with some settling of the dust by next year.
- Blockchain technology interest will accelerate: Cryptocurrencies will continue to exist in the "novelty" space, but we think large payments players will direct efforts to leveraging the distributed ledger technology for various uses and will proceed at an accelerated pace.
- Biometric technology improves, but passwords remain supreme: Despite continued cries for intervention, the user ID and password will remain the primary authentication method that consumers use to access their various applications. Biometrics technology for payment and customer authentication applications will continue to improve while decreasing in price. Fingerprint, facial recognition, and eye/iris recognition will dominate as the most-used biometrics although voice recognition will serve as a key method in certain environments such as call centers. The Risk Forum believes that the technology will continue to face critical adoption challenges due to concerns about privacy, security, and safety, but educational programs will lower this resistance.