“Real Time” Funds Transfer in the US

8 Feb 2011

I’ve had quite a few questions from start ups on this subject, so I thought I would address here. In the past few months we have seen press from Obopay on Star/NYCE integration that would allow for “instant” transfers, yesterday there was a press release from Cashedge/NYCE on the same subject. In my previous blog on Visa Money Transfers I discussed the top 2 fallacies: Instant and Mandatory. These same issues plague NYCE’s and Star’s PIN Debit “credit push”.

For the non-bankers, there are 5 basic payment networks that surround a typical US DDA account:

  1. PIN Debit/ATM (Interlink, Pulse, Star, NYCE)
  2. Signature Debit
  3. ACH (example The Clearing House, Jamie Dimon Chairman)
  4. FedWire (the US RTGS system run by the Fed)
  5. SVPCo (Check Images)
  6. Optional (ex. SWIFT, Western Union, …)

For further information see the FFIEC’s Examination Guide on Retail Payment Systems.

From a global perspective, we are quite a few years behind the UK and most of EMEA. While consumers in the UK can expect that 98% of domestic payments to clear in “real time”, most ACH “payments” in the US clear in the 3-5 day horizon. This blog is focused on “instant” payments. Important to note that the definition of “instant” is relative to both bank and consumer. For example, each bank has its own policy on funds availability and posting (vs clearing). A consumer could see funds posted to their account, but the funds may not be available for withdrawal. Other banks choose to show the consumers available funds to avoid confusion.

FedWire is a Real Time Gross Settlement (RTGS) system run by the federal reserve. Each bank in the US maintains funds with the federal reserve, and FedWire performs real time exchange of funds between member banks. Consumers and Commercial Businesses know of this service as a “wire”, and it is the only real time payment network in the US with universal adoption. FedWire Fees (~ $.52) are paid by the sending bank.

PIN Debit

The other “real time” payments systems (surrounding a US consumer DDA) are PIN Debit and Signature Debit. PIN Debit Networks evolved from ATMs, connecting ATM nodes to bank run authorization systems. Bank Authorization processes for PIN Debit/ATM systems are rather straight forward: validate the PIN  and funds available (I emphasize this authorization process as it is key to understanding why a “credit” is difficult to process on this debit system). PIN Debit fees are typically paid by the merchant and average around 85bps + $0.10. For ATM use, banks control fees and can assess surcharge for use of bank or foreign ATMs.

ATM Networks grew as groups of banks banded together to monetize ATM infrastructure, and further expand network into the retail POS. This expansion led to further change in structure, from bank ownership to independence. The driver of any independent network is to add volume, nodes and services. ATM Networks evolved into PIN Debit Networks, with Visa’s 1987 contract to operate Interlink  serving as a key milestone. Today, Pulse is owned by Discover, Star by First Data, Interlink by Visa (these 3 make up over 83% of PIN Debit Volume).

PIN debit networks have been working to fill the real time “hole” in DDA payment services for many years. Star’s Expedited Transfer, NYCE’s A2A Money Transfer , Visa’s VMT all attempt to EXTEND their respective PIN Debit networks to handle credit transactions. In EVERY CASE, the networks must sell their members banks and get them to extend a PIN based network (which processed only debits in a simplified authorization process), into a funds transfer service. Who owns fraud? Compliance? Reg E burdens? Consumer Support? Returns? Reporting? Integration into online banking? Statements? .. yep the banks.. Oh and by the way.. the other “benefit” to a bank is that once you implement it you can forget about those expensive wire fees. The result of their efforts is what you would expect… VERY poor adoption.

Today, PIN debit networks are looking at a very bleak future. Signature and PIN debit rates will be moving to a flat fee of $0.21 as a result of the recent Dodd-Frank Act (pending completion of the comment period). As a result, my guess is that we will likely see consolidation and bank ownership of shared PIN network infrastructure as with any commodity payment service.

Signature Debit

Signature Debit varies from PIN Debit in that it evolved from Credit Card (as opposed to ATM). Visa was and has always been the leader in signature debit penetration, a look back at this 2003 article provides much insight into the history here. Most US consumers today don’t understand why their debit card has both a PIN and signature feature… many books could be written on this subject alone… but oddly enough consumers prefer PIN (see Pulse Federal Reshttp://3dmerchant.com/blog/how-can-i-reduce-american-express-transaction-fees/erve Presentation 10/10).

In the signature debit model, transaction authorization is much more complex, with most banks leveraging either network shared infrastructure (example Visa DPS), or their credit card systems. The complexity arises as the lack of PIN requires the banks to risk score transactions in a manner akin to credit card (absent the credit risk). There are limited facilities for a credit transaction within most Signature Debit systems, and most relate to a merchant credit relating to a previous transaction (ex. Overcharge, returned merchandise). DPS, Mastercard IPS, and most other platforms perform usually perform a daily net settlement with member banks (multiple settlement files are sent throughout the day.. but netted just once) .  Just as with PIN Debit, Signature Debit is also designed as a DEBIT ONLY system…

In VMT, Visa is attempting to enhance signature debit network into a quasi RTGS transfer service by leveraging its DPS hosting and authorization role. The QUASI is very important… as DPS, Interlink or any of the debit systems are “real time” ONLY in the instruction, NOT the Settlement.  To suggest that any of these services are an actual RTGS system is a significant stretch of the imagination and thoughtful invention. A payment system cannot be faster than its underlying settlement system.  The PIN and Signature Debit Networks DO NOT SETTLE, but rather depend on existing bank settlement processes (which are daily batch runs). The “message” to post or credit a transaction to the customer can be “instant” but the funds will not clear into the customers account until settlement occurs (dependent upon each bank’s policy).

What does this all mean for real time transfers in the US?

Only FedWire offers real time transfers between all financial institutions. All other solutions have sporadic coverage unless balances are held within the same institution (bank, paypal, … ).

How should you view NYCE, STAR, Visa “credit” Capabilities?

It works at some banks, with many provisions. I estimate that combined coverage of NYCE, Star and VMT “credit” covers less than 5% of all US deposit accounts. As you can see from WSJ graphic below, the top 5 banks account for 80%% of debit volume…. given that these banks have not adopted the credit services (in debit), there is little likelihood of success.

It is likely that independent PIN debit networks will go the way of Canada’s Interac… a bank owned service.

Messages for banks

Keep bank control of transfer facilities.. new services that give consumers real time transfers compete with wire, increase fraud exposure and enable rate hoppers… Why role this out today when you will likely get it from a bank owned network in 2-3 years.. ?

PayPal Shut Down in India

11 February 2010

NYTimes article from last night

India’s Central Bank Stops Some PayPal Services‎ – 

Simply put.. paypal has no license (See RBI list ) for Operating a Payment System in India under India’s Payment and Settlement Systems Act, 2007. The RBI published Annex I circular on November 27, 2009 (RBI/2009-01/ 236).

It Appears that RBI’s central issue is with PayPal’s role as an “unlicensed” Money Transfer Service. This issue is certainly not new to PayPal (see US Regulations – Online Payment/Transfer). As highlighted in the circular above:

All cross-border inward remittances under MTSS must be accompanied by accurate and meaningful remitter information (name, address and unique identification number of each remittance like, MTCN) on funds transfer and related messages that are sent and the information should remain with the transfer or related message through the payment chain. A unique reference number, as prevalent in the country concerned, must be included.”

Further, Paypal’s “agents must”:

Indian Agents should have effective risk-based procedures in place to identify cross-border inward remittances lacking complete remitter information. The lack of complete remitter information may be considered as a factor in assessing whether a cross-border inward remittance or related transactions are suspicious and whether they should be reported to the FIU-IND. The Indian Agent should also take up the matter with the Overseas Principal if a remittance is not accompanied by detailed information of the fund remitter.

Issue for PayPal is that its “agent” in this case is its commercial bank that initiaties the domestic ACH. My guess is that they are also in a bit of hot water for allowing this connectivity in light of the Nov 2009 circular (and subsequent inaction).

In order to resolve RBI’s issues, PayPal must:

  1. Obtain an MTS License (or a Payment System License)
  2. Renegotiate terms and services with its clearing bank(s) so that they will comply with “Indian agent” responsibilities above, namely PayPal must provide detailed information on remittance (above) to clearing bank and hold information in country so that bank can perform both AML sanction screening and other SAR reporting
  3. Put the detailed technology plan in place capture and send information to bank
  4. Review Plan with regulator
  5. Obtain regulatory approval on end state plan and request exception process for operating until final (end state) is in place

India’s regulators are some of the toughest on the planet. They expect that organizations read their guidelines.. “Better to ask forgiveness than permission” is a regulatory approach that probably works best before you are public company.

Note that PayPal’s “merchant” transactions (where an eBay buyer is paid) are not covered within the MTS regs above, unfortunately for PayPal it is difficult to screen these commercial transactions from other payments,  hence the broader impact in clearing both commercial “merchant” ebay payments and P2P/Remittances. 

My related Post

Cash Replacement

NACHA on Aggregation

SEC AML/Patriot Overview Regs

US Regulations – Online Payment/Transfer

Lessons from PayPal

January 25, 2010

I was on the phone today with Jeff McConnell, a tremendous exec with a long history of leading innovation in money transfer (WU, Moneygram, iKobo, …). In some respects it’s hard for me to believe that 2002 is 8 years ago, and I was reminded of how challenged PayPal was in obtaining the proper licenses “after the fact” in its early business.

In his 2006 book The PayPal Wars, Eric Jackson did an excellent job laying out the challenges paypal faced in its early years.  In the early days after its inception in 1999, PayPal was moving toward becoming a bank, but the Internet startup decided that banking regulations were too cumbersome. “We just wanted to be able to facilitate a quick payment,” he said. “The question of how to classify PayPal lingered for some time….It’s a sort of modern-era Western Union.. really, all PayPal is doing is shifting money around on your behalf.” 

To see the “change” in PayPal’s regulatory approach, take a look at PayPal’s 2002 prospectus.

We believe the licensing requirements of the Office of the Comptroller of the Currency, the Federal Reserve Board or other federal or state agencies that regulate or monitor banks or other types of providers of electronic commerce services do not apply to us. One or more states may conclude that, under its or their statutes, we are engaged in an unauthorized banking business. In that event, we might be subject to monetary penalties and adverse publicity and might be required to cease doing business with residents of those states. A number of states have enacted legislation regulating check sellers, money transmitters or service providers to banks, and we have applied for, or are in the process of applying for, licenses under this legislation in particular jurisdictions. To date, we have obtained licenses in two states.

This 2002 regulatory view, by the Paypal exec team, was based on a position that PayPal was acting as a Third party payments aggregator (TPPA), not in need of a money transfer license. TPPA is a description used for merchants that are charging a credit card for a product or service that they do not own. TPPAs simply facilitate the exchange of money between two parties sometimes using a credit card as a funding source. Several fraud and AML incidents arose which got the attention of both federal and state organizations. It became clear that PayPal was being used for much more then payment for goods within the eBay marketplace.

In Feb of 2002, the Federal Deposit Insurance Corporation (FDIC) ruled that PayPal is not a bank, which accelerated efforts by states to pursue PayPal for violating money transfer laws (New York and Louisiana are most notable).  This could have been the death knell for PayPal, as they were operating without the proper licenses. PayPal’s “post facto” licensing efforts were greatly aided by the local political support from thousands of eBay’s buyers and sellers. Today, according to spokesperson Michael Oldenburg,  PayPal is licensed as a money transmitter in 43 states (not all states require a license), demonstrating that regulatory risk was far greater than what they articulated in the 2002 prospectus. For those interested in the legal/regulatory conundrum faced by regulators, I highly recommend:  Regulating Internet Payment Intermediaries, by Ronald J. Mann, University of Texas School of Law

For today’s “emerging” payment companies, there are 4 primary choices for operating in the US:

  1. Obtain the licenses
  2. Operate as an agent of an entity with the proper licenses
  3. Sell your software to a licensed entity
  4. Exchange non-monetary forms of value (minutes, eGold, credits, …).

Obtain the licenses

For those of you that read my Blog, you’re probably aware that I’m fairly negative on Obopay, however they do excel in obtaining US MTO licensing (https://www.obopay.com/corporate/stateLicenses.shtml) . Unfortunately, all of these US licensing effort seems for naught as they are pulling out of the US and focusing in emerging markets as the “sender pays” model does not work in developed countries (morphing from a failed US P2P effort to Remittance). Today, PayPal, Western Union, Travelex, Moneygram, MoneyBookers (soon to be NY licensed) also operate as licensed Money Transfer Organizations (MTOs).

Becoming an MTO is not for the faint of heart, as regulatory capital requirements in excess of settlement obligations (fiduciary assets) are a complex (state by state) maze. This creates a challenging dynamic where capital reserve requirements grow as payment volumes grow. As a start up this means you not only need to raise capital to start the business, but also the regulatory capital BEFORE you get the state licenses.

MoneyGram’s 2007  “investment issues” offer many insights into MTO challenges. MGI suffered an $860M+ plus loss as it shifted out of high yield asset backed securities (which lost their investment grade rating). To preserve liquidity it sold $630MM in preferred and received debt financing of approximately $500M, a situation which today leaves MGI common shareholders in a $870MM equity deficit.

Operating as an MNO is not your only choice. I’m amazed at how few companies there are attempting to develop a bank based model. Trolling the dust bin of failed financial institutions may provide a unique opportunity for a start up to acquire the “shell” of a licensed bank to develop a “payment” focused value proposition. The strategy behind Revolution Money’s acquisition by Amex gives Revolution the “best of both worlds”: an acquirer and a bank. If it were not for Amex’s bank charter (and associated regulatory capital), Revolution’s PIN based debit would be highly susceptible to NACHA aggregation restrictions if they are operating as a non-bank, operating as a type of decouple debit.

I know from my own personal experience that operating as a “payment bank” is not without challenges, not just Citi C2it.. (which stopped 2.5 yrs prior to my role), but Citi GTS which today provides many of the banking licenses for payment providers like WU, Vodafone, ZAIN, …  In addition to Citi GTS, one of Citi’s most profitable “global” retail bank businesses is NRI (Non Resident Indian) which serves affluent Indians (within the US, UK, …) with comprehensive services that cater to the needs of affluent clients. Citi also effectively up sells NRI clients services within its investment and commercial bank.

Operate as an agent

Pre-paid cards offer a “fast track” to operating a new payment service (Revolution money, Squareup, payoneer, iKobo, …). In this model the service relies on the licenses of the underlying bank (example Metabank). The legal precedent here is rather new as witnessed by May 30, 2007 finding by the First Circuit , which affirmed that the National Bank Act preempted New Hampshire regulation of the pre-paid product. In the “agent” model, it is therefore paramount that start ups seek a federally chartered partner. 

There is still substantial “risk” in this pre-paid agent model, as traditional banks and networks control the “rails” for this payment type. For example, Consumer accounts must be “funded” from either a card or DDA account. NACHA has developed new rules which significantly curtail the ability of a “payment aggregator” operating off of a current account (see NACHA Tightens Risk Management and aggregation rules) . Additionally, card networks and acquirers are much more attuned to the risks that these new payment intermediaries present.

My top vendor in the bank model is CashEdge (having been the banker who signed the agreement at Wachovia). CE is the “3rd party sender” for Citi, BAC, Wachovia, PNC and other top banks representing approximately 50% of US DDA accounts. You don’t hear about them much because they are a white label “bank friendly” service. They excel in risk management, with a team second only to paypal. Most of you in the US reading this already use their software.. but just don’t know it.  In the mobile space, I love the innovation at BlingNation.

Sell your software

This is rather straightforward. Within the mobile money space, companies such as Monitise, HyperWALLET, Fundamo, Paybox (now Sybase 365) all provide good platforms from which to build an offering. Issue for small companies is that the entities which have the necessary license have largely made significant bets here already. Of course some of the bets by big banks (some alliteration here) have been terrible, most notably Firethorne which has lost the accounts at Chase, Citi and Wachovia all in the last 8 months.

Exchange non-monetary forms of value

Beyond the scope for my discussion here. My advice is that this is a slippery slope and you will have trouble (as a payment company) attracting “A Class” capital. Look no further than the history of e-Gold for education on the issues.

U.S. GOVERNMENT SEIZES E-GOLD ACCOUNTS, OWNERS INDICTED

Summary

In writing this I cannot help but be struck by many similarities in the “unregulated growth” of PayPal and Vodafone’s MPesa. The growth of both companies was driven by an existing customer base and a value proposition which addressed clear gaps within the payment systems of their respective markets. In both cases, there was no clear regulatory authority to restrict them and once they were firmly established (through contagious adoption) it was too late to stop.

Within the EU, the ECB has developed ELMI regulations that are supported by other initiatives such as SEPA (See http://www.paysys.de/download/Krueger%20e-money%20regul.pdf).

Related posts

http://tomnoyes.wordpress.com/2009/12/16/cash-replacement-part-2/

http://www.banking.state.ny.us/legal/lo020603.htm

http://www.ecommercetimes.com/story/18211.html?wlc=1264432425