Into the Black Hole

“The forces of change in the U.S. payments system have never been more plentiful”, Kansas City Fed economists Fumiko Hayashi and Richard J. Sullivan remind us in a recent paper. “Payment card fraud is on the rise. Mobile banking is on the rise. Payments through new social media are swelling. Bank regulations have changed. Lines of competition in the payment card industry have been drawn and redrawn.”

The two economists have focused their attention on two “separate areas of change” — crime and competition. More specifically, Hayashi and Sullivan have looked into the evidence for fraud trends in countries where computer-chip cards have been adopted and have examined the proposition that recent debit card regulations will promote competition for merchants within the payment card industry, as well as the effects of regulations on consumer welfare and payments system efficiency. For my part, I will focus on the authors’ findings on the link between EMV adoption and fraud. Let’s take a look at what they have to tell us.

The Case for U.S. Adoption of EMV

The transition in the U.S. from the older magnetic-stripe technology to the chip-based EMV, known as chip-and-PIN in the U.K. has already begun, but will take years to complete, as both the payment cards we carry in our wallets will have to be replaced and the point-of-sale terminals will need to be retrofitted or (more likely) replaced to be able to become compatible. Once the switch is complete, “[t]he fraudsters, phishers, hackers, and pickpockets who thrive off payment card fraud may… have their work cut out for them”, Hayashi and Sullivan predict. “Compared with the magnetic-stripe cards carried by millions of consumers”, the authors remind us, “the new chip cards will offer stronger defenses against fraud. But they certainly will not put an end to it.” The authors warn against “any prolonged accommodation of older card technology during a transition to computer-chip cards”, for it can “allow fraudsters to exploit weak links in security.”

Another reason for the U.S. to quicken the pace of the transition is that, as Douglas King from the Federal Reserve Bank of Atlanta told us more than a year ago and then a Europol report confirmed his findings several months ago, once they realized just how difficult it had become to commit credit card fraud in Europe following the wholesale EMV adoption, fraudsters decided to take the stolen European cards to the U.S. and take advantage of the older mag-stripe technology while it’s still the norm there. Here is the essence of the Europol report:

Payment card fraud is a low risk and highly profitable criminal activity which brings organised crime groups originating from the EU a yearly income of around 1.5 billion euros. These criminal assets can be invested in further developing criminal techniques or can be used to finance other criminal activities or start legal businesses.

The EU is increasingly exposed to the threat of illegal transactions undertaken overseas and should develop more efficient solutions to help law enforcement authorities (LEAs) combat the fraud.

…

The majority of illegal face-to-face card transactions (skimming-related) affecting the European Union take place overseas, mainly in the United States.

So the case for switching to EMV seems compelling.

The U.K. Experience

Hayashi and Sullivan have looked into the experience of the U.K. — one of the first countries to make the transition to EMV and concluded that “[t]he benefits of the computer-chip cards were apparent as early as 2005 when fraud losses due to lost or stolen cards began to decline (see Chart 1). Both the added fraud protection due to the computer-chip and the required use of a PIN for transactions successfully limited the ability of anyone who possessed a lost or stolen card to create a fraudulent payment.”

However, the two economists second King’s findings for the spillage effect, noting that, following EMV adoption, “payment fraud soon migrated to channels in the UK with weaker authentication, such as cards still using magnetic stripes and purchases made over the Internet, mail order and telephone order (IMOTO) purchases. For backward compatibility, during the transition period, new computer-chip cards had both computer chips and magnetic stripes. Fraudsters could then make counterfeit magnetic-stripe payment cards and use them wherever merchants or ATMs still accepted the cards, especially outside the UK [read U.S.]. As a result, fraud losses on counterfeit cards in the UK grew from £97 million in 2005 to £170 million in 2008. The move to computer-chip payment cards also left authentication unchanged for IMOTO transactions, making them another attractive outlet for fraudsters. Fraud on IMOTO transactions grew rapidly, from £183 million in 2005 to £328 million in 2008.”

In the years after 2008, fraud declined due to two factors, Hayashi and Sullivan tell us. To begin with, a wider EMV adoption in Europe sharply reduced the number of locations where counterfeit cards could be used. Secondly, merchants in the U.K. increasingly began to adopt “3D Secure Systems”, which are developed by Visa and MasterCard and offered to customers and merchants as the Verified by Visa and MasterCard SecureCode services. “In 2007″, the authors inform us, “only 25 percent of respondents to a survey of UK Internet merchants reported that they accepted 3D secure payments, but the same survey found at least 59 percent of respondents accepted 3D secure in 2011. Total fraud losses on payment cards in the UK fell significantly, from a peak of £620 million in 2008 to £321 million in 2011.” Now, it should be said that, from a merchant’s perspective, the use of a 3D Secure-type of a credit card processing service carries a very significant disadvantage in the shape of an excessively high ratio of declined transactions and it is far from clear whether the trade-off is worth it. Still, as far as fraud alone is concerned, 3D Secure does work.

The Takeaway

Hayashi and Sullivan are cautiously optimistic for the implications for U.S. payment card fraud of what is going to be a years-long transition to EMV:

As the United States begins its transition to computer-chip payment cards, the country will reap the benefits of the dynamic data authentication processes that are not possible on magnetic-stripe cards but can be performed by computer-chip cards. The chip cards are also much harder to counterfeit. Some sources of payment fraud, such as counterfeit cards, will decrease. However, the experience in the UK and other countries also shows that other sources of fraud are likely to increase. Thus the prospects for reducing overall card payment authorization protocols continue, such as signatures for card payments rather than PINs, the degree of fraud reduction that can be achieved will be limited. Similarly, unless authentication protocols are improved for IMOTO transactions, such transactions will become a weak link in the defenses against fraud, and IMOTO fraud will likely increase.

The payment industry must also be alert to new forms of fraud as attackers probe for security weaknesses and exploit them. Fraudsters have strong incentives to commit payment fraud and will continue to test security measures and sometimes defeat them. Card issuers, in turn, will need to reevaluate their choices of authorization and authentication methods periodically, as new trends in fraud emerge.

Image credit: Chipandpin.co.uk.

In its second Consumers and Mobile Financial Services survey, the Federal Reserve looks once again into the reasons why a significant chunk of U.S. consumers are either unbanked or underbanked. The researchers’ findings confirmed the conclusion I drew from the first survey, which was that staying out of the traditional financial system is much more a matter of choice and lack of financial education, than it is about being shut out of it. The surprise was to see just how little progress has been made in a year when a number of excellent financial products were launched that were designed specifically for the unbanked.

And it is not as if nothing has changed in the unbanked world, far from it. Of those currently unbanked, 42 percent told the researchers that they had a bank account at some point in the past. Even more strikingly, the data from the 2011 and 2012 surveys show that 40 percent of those unbanked in December of 2011 had obtained a checking, savings, or money market account by November 2012. However, 4 percent of those who had a bank account in December 2011 no longer had one by November 2012.

Even the most casual of looks at the data tells you that, for whatever reason, the vast majority of unbanked Americans simply don’t want to have anything to do with bank services. Now, even in 2011 such attitudes were difficult to justify and that was much more strongly the case in 2012. The launches of incredibly consumer-friendly prepaid cards such as Chase Liquid and, especially, Bluebird have made the decision to stay out of the traditional financial system a very costly one. It’s about time that the unbanked gave these prepaid cards a try and left the check cashier services behind.

Why Are Consumers Unbanked?

In 2012, the share of unbanked consumers — defined as consumers who do not have a checking savings, or money market account — fell to 9.5 percent of the adult population, down from 10.8 percent in 2011. And if the use of reloadable prepaid cards is included in the definition, the share of consumers who are unbanked declined from 9.0 percent in 2011 to 7.9 percent in 2012. The share of underbanked consumers — defined as having a bank account, but using an alternative financial service such as a payroll card, payday lender, check casher or auto title loan — has fallen very slightly over the past year — from 10.2 percent in 2011 to 9.9 percent a year later. So why are there so many unbanked and underbanked Americans?

Well, the reasons given by the respondents for not having a bank account are quite revealing, as you can see in the table below. Whereas there are consumers who offer genuinely good reasons — “credit problems” and “cannot manage/balance an account” — these only account for 9 percent of all respondents. It is true that, if you’ve had credit problems in the shape of issues with previous bank accounts, no bank is likely to want to service you. And if you know that you cannot balance an account, you are probably better off not opening one up in the first place.

However, all other answers, excluding the 16 percent which are divided between the “refused to answer” and “other” groups, betray either a personal attitude towards the industry and its services — “I don’t like dealing with banks” and “I don’t need or want an account” — or misinformation (all other categories). And it’s very likely that the two are related. After all, if you thought that the banks’ fees were too high, how could you like them? But here is the thing. Leaving aside the fact that there are still plenty of free checking accounts around, compared to what are bank fees too high? Payday loans? Check-cashing services? Even if you couldn’t find a free checking account, although you would be able to if you looked for one, there is no way that a bank account would cost you more than a payday loan or a check-cashing service. And not writing enough checks, which is a reason given by 16 percent of respondents, doesn’t add to the cost. So what we see is that a combination of lack of financial education and personal prejudices is preventing Americans who may qualify for traditional banking services from using them.

Why Are Consumers Using Payday Loans?

One in ten of the respondents report ever using a prepaid loan, we learn, but only 6 percent have done so in the past year. The respective percentages for 2011 are only marginally different — 11 and 5. As we know, this is an incredibly expensive form of financing, with even the best of payday lenders charging annual interest rates of 300 percent. So why would you take out such an expensive loan? Well, here is what the borrowers are telling us:

There is no way to justify taking out a payday loan, unless you absolutely needed the money for some crucially important purpose and you tried, but could not get a more traditional type of loan. Yet, what we learn is that half of payday loan borrowers are being severely overcharged, because they find payday borrowing easier (21 percent), quicker (19 percent) or more convenient (10 percent). And then there are the two percent who felt “more comfortable” with payday loans.

The Takeaway

The good news is to be found in the changing attitudes toward prepaid cards. The researchers note that “[p]repaid cards have remained the most-used alternative financial service over the last several years” and tell us that about half of the consumers surveyed reported that they have used the product. The even better news is that many new-age prepaid cards now charge very low or no fees at all. Moreover, the best one of them — Bluebird — which was only launched in October of last year, is sold at Wal-Mart, which is arguably the best way to market anything to the unbanked. So I expect prepaid issuers to grab more market share from payday lenders, although, sadly, the latter are unlikely to disappear anytime soon.

Image credit: Bankis-on.blogspot.com.

That is the inescapable, though not unexpected, conclusion from the latest Federal Reserve Consumers and Mobile Financial Services survey and report. Prompted by the “rapid pace of developments in the area of mobile finance”, the Fed conducted the survey in November of 2012, less than a year after the first one was completed, to examine the trends in adoption and use of mobile banking and payments and compare the two surveys’ findings.

Though obstacles to consumer adoption remain, chief among them being concerns about security and a sense that the new technologies don’t really offer any benefits over existing ones, the use of mobile banking has increased by more than a third over the past year, we learn. Moreover, while still small in absolute numbers, the use of mobile phones for payments at the point of sale has increased even more rapidly. As an ever increasing number of consumers use smartphones, the researchers expect that the use of both mobile banking and mobile payments will continue to increase. Let’s take a look at their findings.

Use of Mobile Banking, Payments Up

The Fed defines mobile banking as “[s]ervices that allow consumers to obtain financial account information and conduct transactions with their financial institution” using their phones, whereas mobile payments are services that “allow consumers to make payments, transfer money, or pay for goods and services”. Both have become more widely used over the past year. Mobile banking usage has increased from 21 percent of mobile phone users and 42 percent of smartphone users in December 2011 to 28 percent of mobile phone users and 48 percent of smartphone users in November 2012.

The use of mobile payments has increased less rapidly. In December 2011, 11 percent of mobile phone users and 23 percent of smartphone users reported using mobile payments. By November 2012, the ratios had risen to 15 percent and 24 percent, respectively.

Bank Branch Visits still Preferred Way of Banking

As you can see in the figure below, visiting a bank branch is still the preferred way for consumers to interact with financial institutions, with online banking and ATM visits following closely behind. Mobile banking occupies the second-to-last spot, but usage is growing fast — up by 7 percent from December 2011 to November 2012.

Young and Educated Consumers Most Avid M-Banking Users

I suspect that it should come as no surprise to you to learn that mobile banking usage is highly correlated with age, with the 18-29 group accounting for approximately 39 percent of all mobile banking users, relative to 22 percent of mobile phone users.

Mobile banking usage is generally unrelated to household income, the researchers find, except at the tails of the income distribution. Consumers earning less than $25,000 per year are significantly less likely to use mobile banking than their share of the mobile phone user population (23 percent) would suggest, whereas the opposite is true for those earning more than $100,000 per year. Here is the full distribution:

Minorities, Hispanic users in particular, are more likely to adopt mobile banking than non-Hispanic whites, as illustrated in the table below:

Educational level, it seems, is the most reliable determinant of mobile banking adoption. As shown in the table below, more than 72 percent of all mobile banking users have at least some college education, far more than their 60 percent share of all mobile phone users.

Checking Account Balances Most Common M-Banking Activity

Checking account balances and making transaction inquiries continue to be consumers’ favorite mobile banking activities, with 87 percent of users having performed at least one of them in the past 12 months, down only slightly from 90 percent in 2011. The transfer of money between accounts is the second most used mobile banking feature — 53 percent of users have reported that they had done so in the past year, up by 11 percent compared to 2011.

Among users, the frequency with which they use mobile banking has increased over the past year, we learn. Whereas the median reported usage was up only slightly — to 6 times per month from 5 times per month in 2011, the share of those using it more than 10 times per month rose to 35 percent from 22 percent.

Half of Non-Users See no Need for Mobile Banking

The main reason why mobile phone users do not use mobile banking is that they believe their banking needs are met otherwise (54 percent). A substantial group of non-users (49 percent) are concerned about security and 47 percent don’t see any reason to use mobile banking at all.

Mobile Payments Limited but Growing

Mobile payments continue to have limited adoption, although usage is growing. Only 15 percent of mobile phone users said that they made a mobile payment in the past 12 months, up by a quarter from 12 percent a year before. The payment of bills is the most common type of mobile payments activity (42 percent), followed by making online purchases (35 percent) and both are down slightly in the past year. Using mobile phones to transfer money has become more common — nearly 30 percent of users transferred money directly to another person in the past 12 months, up from 21 percent in 2011.

Point-of-sale purchases with mobile phones have also become more common, with 9 percent of users reporting that they scanned a barcode or Quick Response (QR) code to make a payment and 9 percent saying that they used a mobile app to pay for a purchase. The share of users waving or tapping a mobile phone at a cash register to pay for a purchase has more than doubled from the previous survey, to 6 percent in 2012 from just over 2 percent in 2011.

Security Concerns, Alternatives Slow Down M-Payments Progress

The biggest reason why consumers do not use mobile payments is security, cited by 38 percent of non-user respondents. Moreover, a substantial number of consumers see little or no need for mobile payments: 36 percent have reported that it is easier to pay with other methods and 35 percent have said that they don’t see any benefit from using mobile payments.

The Takeaway

On the whole, the use of mobile banking and mobile payments is growing and there is every reason to expect that in the not-too-distant future they will become the prevalent method of doing both banking and payments. The factors limiting consumer adoption — security concerns and the sense that mobile services don’t offer any real benefits to the user over existing methods — will gradually be overcome, as technologies improve and consumers become better educated about the new services. So yes, the future of mobile does look bright.

Image credit: Smartplanet.com.

The Federal Reserve has recently released its biennial report on the volume and value, interchange fee revenue, processing costs and fraud losses related to debit card transactions in 2011. The headline news to come out of it is that the Fed has decided not to make any changes to the debit interchange fees, which retail groups were hoping would go even lower than they are now. Back in 2011, acting on a mandate from Congress, the Fed capped the debit interchange fees — the amount card issuers can collect as fees from transactions involving their debit cards — at 0.05 percent of the purchase amount plus $0.21 per transaction plus an additional $0.01 per transaction if the issuer has implemented certain fraud prevention standards. That limit — which only applied to issuers with consolidated assets of $10 billion or more — slashed the average interchange fee amount from about $0.44 to $0.23 per debit transaction — a 45-percent reduction.

But I wanted to touch on another issue covered in the Fed’s report, which has not received much attention — debit card fraud. We learn that debit fraud has declined from 2009 to 2011 for all types of transactions, excluding prepaid, which technically are not debit transactions, but are included in the Fed’s report anyway. Furthermore, we learn that most of the losses are borne by the issuers, whereas the cardholders’ losses are negligible. Let’s take a look at the numbers.

Debit Card Fraud Down

The Fed estimates that debit card fraud losses to all parties (merchants, cardholders, and issuers) amounted to $1.38 billion in 2011. About $1.13 billion of these losses resulted from signature debit card transactions, $204 million resulted from PIN debit transactions and $51 million resulted from prepaid card transactions. The average loss was approximately 8 basis points per debit card transaction, down slightly from 2009. Among covered issuers, the overall incidence of fraudulent debit card transactions fell from 0.04 percent of all purchase transactions in 2009 to 0.03 percent in 2011. Card-not-present fraud was the most common type of debit card fraud, accounting for 42 percent of fraudulent transactions, followed by counterfeit fraud — 32 percent. There were significant differences in the overall incidence and distribution of fraud by type among signature, PIN, and prepaid transactions, as the following chart illustrates.

As you see, from 2009 to 2011 fraud incidence declined significantly both in signature transactions (by 28.3 percent) and in PIN debit ones (by 40 percent). As expected, fraud incidence was much lower among PIN transactions than it was among signature (by a factor of more than seven) and prepaid transactions signature (by a factor of more than six). Card-not-present fraud was found to be the most common type of fraudulent signature transaction (accounting for 46 percent of all fraudulent signature transactions) and counterfeit fraud was the most common type of fraudulent PIN transaction (constituting 48 percent of all fraudulent PIN debit transactions). Lost and stolen fraud occurred more often in PIN transactions (representing 33 percent of all PIN-based fraud) than in signature ones (accounting for 16 percent of signature fraud).

On a per-fraudulent-transaction basis, fraud losses averaged $101 in 2011. The average loss for PIN transactions ($187) was more than twice as high as the average for signature fraud ($91), and almost three times as high as the prepaid loss average ($66).

Card Issuers Bear 60% of Debit Fraud Losses

The fraud losses reported in 2011 were shared primarily between the card issuers and the merchants, with 60 percent of the losses borne by the issuers and 38 percent borne by the merchants. The remaining 2 percent were borne by the cardholders. This breakout was roughly in line with what was reported in 2009. However, the distribution of fraud losses among issuers, merchants, and cardholders varied widely based on the authentication method used in a debit card transaction. Whereas issuers bore nearly all (96 percent) of the fraud losses associated with PIN debit card transactions, they bore 72 percent of prepaid fraud and 54 percent of signature debit losses. Here is a table with the full distribution.

As you can see, merchants bore a higher share of the fraud losses in card-not-present transactions than in card-present ones. So, while they absorbed 70 percent of all signature debit card fraud losses for card-not-present transactions, the merchants only bore 20 percent of the signature counterfeit fraud losses and 35 percent of the signature lost and stolen fraud losses. And whereas merchants only absorbed 18 percent of all PIN card-not-present fraud losses, there are very few card-not-present PIN transactions. On the other hand, issuers bore most of the losses resulting from fraudulent card-present transactions.

Overall, the issuers’ fraud losses in 2011, as a share of the total transaction value, was 4.65 basis points, the Fed tells us. Issuer fraud losses from signature debit transactions averaged 5.7 basis points, more than twice as high as the average for PIN transactions — 2.74 basis points. Average prepaid losses were 5.33 basis points per transaction.

The Takeaway

Now, it has to be said that, although cardholders bear very little of the losses resulting from debit card fraud, they can still have their money tied up for the duration of the investigation. Remember that in a debit transaction the sales amount is debited to your bank account practically right at the time the transaction takes place. Incidentally, this is an area where credit card use offers an advantage in that credit cardholders are not charged for any disputed amount until the investigation is complete. So this is a consideration to be taken into account when deciding what type of card to be used for purchases.

Image credit: Lifelock.com.

Despite the highly-publicized arrival of various new forms of person-to-person (P2P) payments in the U.S. over the past decade and a half, paper checks still dominate the niche, we learn from a recent paper by Terri Bradford and William R. Keeton from the Federal Reserve Bank of Kansas City. What is more, the number of checks written for P2P transactions is still on the rise.

Bradford and Keeton are giving us a brief overview of the pre-1990s history of P2P payments and then go on to evaluate the recently emerged digital alternatives to the check-based P2P payments. Their conclusion is that “the new P2P methods improve on paper checks in a number of ways but leave important gaps, suggesting a need for further innovation”. In this post I will highlight the researchers’ findings on the three new P2P methods, which I thought might be of interest to many of you.

PayPal’s Domination

There is no universally-accepted definition of P2P payments, but here is how the two Kansas City Fed researchers define the term:

P2P payments include payments by individuals to friends and family members and to other individuals for goods and services. The latter group is often referred to as “micro-merchants.” They include gardeners, babysitters, independent repairmen and individuals selling goods through classified ads or online auction markets such as eBay.

Up until the late 1990s of the last century, P2P payments were conducted “almost entirely by cash and check”, we are reminded. Then things started to change very fast:

In the late 1990s, the spread of the Internet and growth in online auctions gave rise to new nonbank-centric payment methods. In such payment methods, the payer initiates the payment with a nonbank company, and that company acts as a middleman between the payer and the payee. Nonbank-centric P2P payment methods had been offered since the previous century by companies such as Western Union. To make a payment, however, the payer was required to visit a brick-and-mortar branch of the company. With the spread of the Internet, it became feasible for consumers to make such payments from their personal computers. In addition, the spread of the Internet led to rapid growth in online auctions such as eBay. These online auctions increased the demand for a new P2P payment method that better satisfied the needs of buyers and sellers who did not know each other and lived in different areas.

While several nonbank intermediaries competed for the payments business in online auctions, PayPal quickly emerged the clear winner. PayPal’s success was due partly to its first-mover advantage and partly to the fact that it offered payment services below cost to build business. Growth in accounts was also facilitated by the fact that recipients had to set up a PayPal account to receive their funds. These factors allowed PayPal to widen its lead over its competitors, helping induce eBay to acquire PayPal in 2002.

PayPal’s growth has indeed been spectacular:

PayPal’s is the only non-bank centric P2P service to gain “significant traction among consumers”, the researchers tell us.

Bank-Centric and Card-Centric P2P Services

Lately, PayPal’s P2P dominance has been challenged by the emergence of two new models. The first is what Bradford and Keeton call “bank-centric” method where PayPal’s intermediary has been dispensed with. As the name implies, in a “bank-centric” transaction the user would transfer funds from her own bank account directly to that of another individual at another bank. At the end of 2011, 90 percent of U.S. commercial banks facilitated such transactions for their customers (see chart below).

Then last year Bank of America, Wells Fargo, and J.P. Morgan Chase formed clearXchange — a P2P payment platform that enables users to send money from their own checking accounts to these of other customers of the participating banks. As the Kansas City Fed researchers point out, that was an important development because, even though in its initial version the consortium was made up of only three banks, these banks “account for 29 percent of total U.S. bank, thrift, and credit union deposits”. Moreover, the platform is open for other banks to join in and it is very likely that many will do just that.

Finally, the researchers highlight the “card-centric method” where P2P payments are “processed entirely over a debit card or credit card network”. This is by far the most lightly used of the three new P2P methods and it is very likely that the services given as examples will be unknown to you. One of them is Visa’s Money Transfer (VMT), which at present is only supported in the U.S. by U.S. Bank.

The Takeaway

So what does the future hold for P2P payments? Here is the two Kansas City Fed researchers’ view:

While many new P2P payment methods have been introduced over the last decade, cash and checks remain the primary methods for making such payments, with checks occupying an especially important role. In other types of consumer payments, such as bill payments and purchases from stores, the number of payments made by check has declined sharply as electronic payment methods have become more popular. In the case of P2P payments, however, the number of payments by check has continued to grow moderately.

Bradford and Keeton proceed to give us more data and analyze the reasons for the slow adoption of the new P2P payment methods, concluding that “ample scope remains for innovation in P2P payments”. The researchers suggest that the Federal Reserve could take a more central role in promoting P2P adoption “because it has secure electronic links to all U.S. banks and a clear mission to promote a safe and efficient payment system.” That sounds logical, but somehow I just can’t imagine the Fed embarking on such a project.

Image credit: Banking4tomorrow.com.