The next step in European ACH competition?

Yesterday saw very interesting news coming out of Europe regarding a joint venture between 6 European ACHs.  To understand why many of us have sat up and taken VERY close interest in the announcement, we need to review some recent history first. Much of this will be covered in more detail in a forthcoming report on ACHs. In the very early days of SEPA, the European Commission made many public comments. As SEPA was as much a political goal as anything, many of these were observations on how the Commission thought the market ought to develop. Given the size of the task and the perceived reluctance to the banks to do anything about SEPA, the Commission narrowed down the observations to a set of specific requirements, eventually culminating in the regulations that made migration mandatory in Euro countries. The downside is that some of the initial elements triggered some activity, but they were never fully pushed through. One such item was the Commissions perceiving their to be an over-supply in payments processors. In the Commissions view, a single market would reduce the 50+ processors to between 5 and 7. That would be enough for a competitive market, but not so many for an inefficient market. The latter stance is based on the fact that processing is broadly a fixed cost business and so the larger the volumes processed, the cheaper the cost per transaction is to process. As a result of this statement was a flurry of activity amongst the ACHs to be one of the “survivors”. It triggered a wave of mergers (Equens is a German/Dutch/Italian merger for example), near mergers (everyone courted everyone else!) and direct approaches to banks and markets to acquire them as customers and boost volumes. But whilst there were mergers, the market broadly remained unchanged. Indeed, some markets chose to build their own SEPA compliant ACH, rather than use the services of a SEPA-ready ACH. There are many reasons for this, not least ownership and control. The announcement yesterday therefore was very significant. At face value, 6 ACHs are going to collaboratively process cross-border SEPA payments. Given the tiny volumes, this isn’t exciting. However, dig deeper, and it becomes clear that Equens – arguably the largest ACH in Europe – is providing all the infrastructure and services to the new company, and the new joint entity company is registered at… Equens HQ. Those other 5 ACHs are considerably smaller – their volumes combined are still dwarfed by Equens. Secondly – it’s for cross-border SEPA payments today but mentions possibly delivering the real-time payments interoperability that’ll be required going forward. That means more ad more services that will be offered by Equens to these other ACHs. It’s particularly noteworthy as many believe that EBA Clearing has been positioning itself to provide exactly that service, and has been leading the discussions. The third point is a broader one. There has been considerably more talk in the last few months about processing, given various elements of PSD2. It’s not yet clear whether the scheme/processor split will apply to “just” card companies, particularly when some of the ACHs process cards. A number of organisations have also mooted whether the XS2A provision potentially provides a way to bypass ACHs – that is break the connection between bank and ACH. Given the range of potential impacts, it seems likely that there will at least some impact. Finally, we are aware of more than one discussion in Europe about the future of that countries ACH, particularly as they ponder on how to deliver a real-time payments solution for that country. All bets are off. The net result suggests to me that we’re entering new phase for payments processors, particulalrly ACH, which has been a relatively stable market for many years. The industry – and technology – is in a very different place than when the discussions happened in c. 2005. What made sense then may not make sense now. We believe that the announcement yesterday will be just the first of a number over the next 2 years. The phrases exciting times and ACHs can be at last mentioned in the same sentence!

Same-day ACH: is anyone excited?

This week’s NACHA vote in favor of mandatory rules changes enabling same-day ACH settlement is no surprise. Some of the press coverage suggests this represents some sort of significant achievement. Really? By March 2018 (when the network is currently expected to be able to fully support systemwide changes) I predict there will be industrywide consensus on the inadequacy of the measure. Even proponents of the measure suggest the vast majority of ACH traffic will remain the next-day float-neutral type – for good reason. The majority of payments will not see a change for the same reasons the ACH has served the industry so well for so long. Specifically: • Dependability • Low-cost With this vote, we’re now going to burden this lowest cost of payments networks with perpetual, systemic cost increases for all participants. And we’ll do so for a very small percentage of network volume. NACHA’s own estimates predict that by 2027 (I don’t make predictions that far into the future) a whopping 1.4 billion same-day payments. That’s 6% of 2014 ACH network volume – presumably a much smaller percentage of 2027 volume. NACHA estimates industrywide implementation costs of $118 million initially and $49 million annually. So, by 2027, the industry will have spent nearly $500 million so banks can offer customers a premium priced same-day payment option using the ACH when other, faster options already exist. I think the NACHA volume estimates are optimistic and find the characterization of same-day ACH as “modernizing the payment system” curious. What’s modernizing about running the same batch system a few times each day instead of once each day? If demand is for real-time payments, this initiative will be found sadly lacking. It’s like installing more pay phones as a way to compete with mobile devices. Am I missing something?

Reflections on Nacha Payments 2015

As many of you know, I’ve been on something of a world tour, which started with Nacha Payments in New Orleans in mid-April. Expect a flurry of blogs as I pass on my impressions from my travels, starting with the Nacha show. Whilst I may live in London, I have global coverage. Payments is an interesting business – whilst its perhaps one thing that unites all businesses and consumers in all countries, and we are moving increasingly to global standards, the finer details show that it is still a parochial business. I don’t mean that negatively, more that payments have evolved over decades, if not centuries, to address specific local needs. That’s why the SEPA project was so difficult – even what we meant by certain terms turned out to be not straight forward. Nacha Payments this year was in New Orleans. For those not familiar, it’s a long standing show focused primarily on the ACH business. It has a very active conference schedule – I was on a panel with Dwolla and BBVA talking about real-time payments – and an exhibition floor. The US is the single largest payments market, so it’s not surprising that the show is very US in focus. What still surprises me, even now, is the fact that in a shrinking list of exhibitors each year, every time I attend there are vendors I’ve never even heard of, let alone am familiar with their products. This demonstrates the size of the market rather than my lack of knowledge! Take-aways for me: What a difference a year makes. Last year, one senior banker said real-time payments wouldn’t happen in the US in his lifetime. This year, it’s seen as a when, no question about if. Indeed, all the talk was about the “secret” (so secret that every meeting I attended asked me about it!) meeting the key Clearing House banks were having to build out the requirements for the Clearing House solution. The general opinion is that the Clearing House is seeking to go live long before the Fed working groups even get close to having a plan for requirement gathering. More on this soon. Business is alive and well. At first glance, the exhibition floor looks notably smaller than the previous year. One thing to note is that the conference is often attended by more junior people, who also get re-certification credits for attending sessions – with so many sessions to choose from, the exhibition hall is often very quiet. Yet all the vendors reported greater numbers of good meetings, with tangible next steps – in short everyone was happy. Which several said wasn’t the case of some of the new sexier shows as Money2020. It strikes me that they’re very different events. Looking overseas. Payments are largely domestic in nature and what works in one country doesn’t always work in another. For example, check technology from the US (which writes 2/3rds of all checks globally) won’t be of interest in Finland (as checks were abolished in 1993!). As such, payment conferences tend to be very domestic in focus. Not a criticism, just pragmatic. Given the changes the US is facing, and given that many other countries have already faced many of the challenges, it was interesting to see such a noticeable increase in conversations seeking an international viewpoint. I don’t think we’re anywhere close to a global market – but the US seems to have taken a significant swing from “not invented here=not relevant” to “don’t re-invent the wheel”. See you next year, in Phoenix!

Braveheart or need a brave heart?

A somewhat topical post, if mildly parochial, but which serves to highlight something more broadly. On September 18th , a new nation could be born, as Scotland goes to the polls to vote whether it should become independent from the rest of the UK. The debate has raged for months, if not years. The latest polls suggests a Yes win, with a flurry of activity now from the No camp. Why blog about this? Well there are some interesting questions that would be raised in relation to payments that are worth highlighting here. Note that this is based on the hard line being stated primarily from the No camp before the vote, but one assumes compromises would be made if there were a Yes vote. Currency Whilst Scottish banks print and issue their own money currently (I have a Scotland-only £100 bank note on my desk), the Bank of England has categorically stated that there cannot be a shared currency. Debate rages about how quickly Scotland could join Europe, so equally the Euro is out as well Central bank Scotland would require a central bank to be able to issue a currency. The creation of one is not necessarily a technical challenge, but the funding of it in the short term might be. Big business Many big banks have already pledged to move their head quarters out of Scotland should there be a Yes vote – Royal Bank of Scotland, Lloyds (owners of Bank of Scotland) and National Australia Group (owners of Clydesdale). Note that this isn’t moving out of Scotland altogether – however, at least one unnamed source has hinted doing business in Scotland will become harder and less attractive. A new currency also means that every business doing business in or with Scotland would need to make the appropriate changes, and an equivalent changeover to the Euro switch would be required.   So what does this mean for payments? Well, nobody quite knows. On day 1, systems could keep running. But longer term, it poses some interesting questions. The options crudely are: Keep as is – people continue to clear and settle in Sterling, regardless of location. Given the hard line on currency union (or lack of), this seems unlikely Shared infrastructure – the infrastructure remains, but is dual currency, with an additional settlement site at the new central bank of Scotland. That works for Scotland-Scotland, and UK-UK (the sort codes could be mapped to allow this). This doesn’t address the cross-border issue. Parallel infrastructure – Scotland recreates all its own systems. This would allow Scotland to plan the ideal system, and with low volumes, it would be relatively cheap to buy. However, it would require every bank and every corporate to change how they process paymenst as well… very expensive! So what does this mean for payments? The fate of a nation is a big thing, and we shouldn’t trivialise it for the thing I’m interested in – payments. But it does serve to highlight how embedded payments are and how critical they are, yet the debate hasn’t mentioned them once. Without a payment system, any country would collapse in hours. Nobody is suggesting that this will happen of course – but then nobody is actually suggesting *anything*. Because payments are rarely thought about by anyone outside of payments, it’s pretty safe to assume that no-one has considered this fundamental part of how a country functions, and it will need to be addressed rapidly. I best go dust off my passport and get some Scottish visas I think!

Real-Time Payments Gathering Pace

A number of you will know that I’ve been working on real-time payments with many clients around the world, and will have seen previews of some of the information in my forthcoming reports. One chart I have shown regularly is the likely adoption curve for real-time payments. This takes a classic innovation adoption bell curve. The top of the curve is where the market has reached 50% adoption. Of course, the question then becomes how many and of how many to plot where we are today. Many of the conversations I have with clients often start with a belief that there are only a handful of schemes globally. The truth is rather different. A good but not exhaustive scan showed that there were actually 35 systems globally. Using a set of criteria, such as levels of GDP, maturity of electronic payments, presence of an RTGS system, we estimate that there are 115 countries which we believe could adopt a real-time systems. That actually puts us just over 30% market adoption. At this point, I ought to point out that there are a few fudges to this figure. For example, note that we say systems, not countries, as some countries actually more than one real-time system (India for example). But it doesn’t detract from the underlying trend. Indeed, the use of the past tense was deliberate, as yesterday saw the announcement from the Finnish Federation of Financial Services of an RFI for a real-time payment system, bringing the total to 36. We also hear rumours of several other countries in advanced discussions. This also supports our other hypothesis. A study of the adoption of RTGS systems globally proves remarkably similar to both the shape of the adoption curve but also to the timelines. If we take that adoption pattern and project forwards, it would suggest that the next 5 years will see a flurry, if not significant numbers, of other systems being announced. It would seem that we are on the cusp of a revolution in payment processing – are you ready?

NACHA Payments 2014 Roundup

After attending IPS, NACHA Payments is always a slightly strange experience. Not bad, just quite a different set-up. IPS is very international – if anything the UK is under represented – and more senior. NACHA offers much to the more junior member of staff, particularly those seeking to renew their AAP accreditation. This means that the attendance is much, much higher, but that there is a real mix of people. As a result, some of the sessions are detailed, nitty-gritty discussions, great for learning about areas I don’t usually cover. The main topic of conversation for me was real-time payments. I’ve spoken a couple of times in the past at NACHA on the topic, partly because of my involvement in the UK Faster Payments scheme, and clients will know about my forth coming series of reports on the topic. Real-time was also mentioned in numerous places across the agenda, with several friends and former colleagues speaking. The focus of my first report was also the starting point for many of my conversations – addressing the many myths that seem to pervade about real-time. These include:
  • that it’s only in the UK and Singapore (it’s not – there are at least 35 other systems globally)
  • that its new and leading edge (its not – at least one system is 40 years old)
  • that it’ll canabalise wire revenue so should only be a p2p proposition (multiple examples proving that this doesn’t have to be the case!)
Shortly before NACHA Payments, NACHA announced it’s enhanced Same Day ACH proposals which also came under great debate. It’s my belief, and shared by a growing number of people, that the Fed has decided the US *will* have a real-time payments system. As such, one group of people saw this announcement as being a response to ensure that NACHA is not bypassed in some way. Jan Estep, the CEO, of NACHA, was on one of the panel sessions, and was asked about how this attempt will be any different to the previous NACHA proposal. The vote on that proposal received a Yes from the majority of banks, but not the 75% voting majority to pass it. It’s widely believed a handful of big banks effectively blocked the proposal. To my point at the beginning about there being a large operational audience, Jan gave an excellent and detailed explanation of how this proposal differed from the last. But a number of the audience suspected that the question was rather more pointed and was really asking why the blocking banks would suddenly vote for this now. That specific question was never addressed. By the time the conference finished, I was left with the impression that the debate had turned a corner, or at least moved into a new phase. Over the last year, I’ve increasingly found that people have formulated their opinions on the subject. But as my discussions highlighted, there are a lot of misconceptions, and I’m not always sure some of the people contributing to the debate aren’t muddying the waters further. I think the next step for the Fed is to address that, and even if it stops short of compiling a list of requirements, a view on what isn’t the solution would be helpful. I understand the logic of the NACHA proposal, but I fear it’s a short-term solution to a long term problem.

Real-time Payments: Different questions, funnily enough, get different answers.

Bob recently posted some views on the same day ACH – as always, great points, well made. Somehow, in Twittersphere, some of the comments got attributed to me, and from that some of those have got re-interpreted as me being anti real-time payments. As my daughters would say, whatever! That’s not the point of this blog. What really struck me was the fact that some saw Bob and I as having different opinions. I would say that I don’t believe we do (at least not in the majority of the issues), but that we were addressing different questions, and, unsurprisingly, end up with different answers. To crudely paraphrase Bob’s post, he quite rightly points out that the business case, based on today’s business, doesn’t stack up. Secondly, he points out that consumers don’t really want real-time payments – how many of us wake up with the urge to make a payment?! Let’s pose a different question, the one I’ve been primarily discussing. If you were starting with a blank piece of paper, would you replicate what we have, or would you build something better and faster? A no brainer. Second question. The current system is roughly 25 years old – do we think that the same system will still be good enough in another 25 years? The answer is again obviously no. No-one in the industry who’s close to this thinks that this isn’t going to happen. The questions we’re really asking are when, what is the trigger, how quick and how soon (i.e. incremental improvements or big bang)? Interestingly, there seems to be less discussion on how, with ACH seeming to be the default. Whilst I’m not suggesting that ACH isn’t an option or even where the majority of other systems have developed worldwide, it’s interesting in that there are already real-time systems in the market, running primarily on card backbones. The answers to those questions are still much for debate. And who gets to answer them even less so. One noticeable difference compared to some countries is the governance of payments in the US. I believe – and please correct me if I’m wrong – that there is no single body who could regulate and dictate such a change. Equally, there is no body managing the future direction of the payments industry. Which, considering that in revenue terms, the US payments is bigger than both the US hotel and US airline industry *combined*, is both remarkable and perhaps something of a risk to the industry. As the US faces more regulation in the same way as many other regions around the world already have, a joined up, united front would seem an absolute need. We may not all agree when we need real-time, but I’d be curious to know whether we agree on the need for an overarching payments body to protect our interests going forward. This blog is written on the eve of Nacha Payments, and the real-time topic is already dominating the discussions before the event has even started. The Nacha announcement has been met with a wide range of responses, but with more than a few suggesting that Nacha has both over stated their position, and that the solution misses the point. The week is shaping up to be very interesting.    

Is the ACH the Best Path to Faster Payments?

Yesterday, NACHA issued a press release announcing initial steps towards same-day ACH. This is a second attempt at accelerating ACH payments. Rather than a “big bang”, this second attempt advocates a phased approach, inviting banks to invest in three projects instead of one. The sentiment seems worthwhile, but I’m not convinced that this is a good idea. In considering faster payments, there are many considerations. Among them: what exactly needs to be faster and who is the customer? Who stands to benefit from faster payments? What Needs to be Faster? Particularly in the case of real-time payments, it is important to distinguish: 1) Notification of payment 2) Payment guarantee/ funds availability and, 3) Settlement In my view, accelerating 1 and 2 are more important than 3 and less costly to bring about. Who is the customer? Who would stand to benefit the most? Many assert strong and growing consumer demand for faster retail payments. We see more interest than demand, particularly if costs are factored in. Celent surveyed over a thousand US consumers in August 2013. In part, we explored payment expectations. With little variation across age demographics, more consumers expect instant confirmation of payment (59%) than expect real time gross settlement (42%). Other factors weigh more heavily than speed. When I Pay Source: Celent survey of US consumers, July 2013, n=1,053 In my view, merchants and regulators are more invested in faster payments than are consumers. Faster payments mean earlier access to funds (retailers) and less systemic risk (regulators). That’s why most systemically important payment systems are RTGS. Faster payments are a certainty – in time. What’s far from certain is how it comes to be – what rails are used. Some advocate using the ACH. I disagree. Moreover, I find the current dissatisfaction with the ACH amusing. Designed as an efficient, electronic, float-neutral payment system, the ACH is highly effective at fulfilling its designed purpose. More recent demands on the ACH, while not without efficacy, have also resulted in increased cost and complexity. Same-day ACH, in my opinion, is simply not compelling. If enacted through a rules change and offered optionally at a premium price, it may succeed, but would result in precious little use. Real-time ACH would be altogether different – a fool’s errand in my opinion. The ACH works splendidly when used as designed. An analogy if I may. The NACHA press release stated: “The Network has always served as a foundation upon which we can build and innovate to meet the growing needs of today’s users and those of tomorrow.” That sounds a bit like inviting telco’s to build more phone booths in response to consumer’s demand for mobility. The “square peg in a round hole” analogy may work as well. I’d love to hear your views.

Faster…or fast enough?

My last post mentioned that I was on a panel at International Payments Summit talking about real time payments. The topic is one that has cropped up many times recently in analyst inquiry calls in the last few months. With all the activity in the market, such as the decision in Australia to implement such a system, it’s perhaps not that surprising. What is surprising though is the number of myths and misunderstandings surrounding the topic. I thought it worth highlighting – and straightening – just a few here. #1. Real time isn’t always really real-time The most frequent myth is that everything end-to-end is suddenly instant. In reality, most (though not all) real-time systems are real-time in notification and authorisation, not settlement. In fact, in some systems in certain situations, settlement can take place days later. The starting point should be what needs to happen, and at what speed. In deed, for many payments, its the certainty, not the speed that matters. #2 Real time isn’t p2p One belief is that there is a pent-up demand for real time to enable p2p (or perhaps, more accurately, a2a) transactions. The use case is often quoted to be that of splitting a dinner check – one person pays the restaurant, the rest then have to find a way to pay the payer. But in reality, how often does that happen? The default in the UK at least would be tell the restaurant how to split the check across multiple cards. Even if that weren’t the case, the numbers of times that this happens would not be large enough to justify the investment on its own. The starting point should be use case driven though: who would benefit from sending – or receiving – funds faster than the current method. In most systems so far, these have been typically b2c or c2b, or indeed, mandated so the business case isn’t the issue. #3. Real time isn’t just payments In many instances around the world, real-time systems are often running 24/7. That poses, at the very least, 2 problems. Firstly, what other systems are required to run in real time, 24/7? Fraud checking, authorisation, notification and authentication systems are amongst the obvious, but banks have found dependencies on many others, and not just in the sending side. Secondly, maintaining systems becomes far more complicated if they have to always be available, and being “always on” means that maintenance becomes more important than ever.

Is MCX Betting On QR Codes and ACH?

In my recent report on Digital Wallets, I discussed a number of players which while still keeping their cards close to their chests, have a potential to significantly influence the payments market. One of them is Apple, which made headlines recently with their patent for cash distribution without ATMs – see Bob Meara’s excellent blog and my related comments for more details. Another one is MCX (Merchant Customer Exchange), a joint mobile wallet initiative amongst a number of the US retailers. The initiative was announced in August 2012, but the details have remained scarce since. The participating retailers have been talking about their desire to have a collective voice in shaping the future of mobile payments, and protect their data and customers. They have talked about developing a wallet, but it hasn’t been clear if they also had ambitions to create a new payment scheme or would rather rely on the traditional cards in their wallet. So I was intrigued to come across an article that appears to shed a little more light on MCX ambitions in payments. Citing sources close to MCX, the article suggests that MCX is indeed planning to build a new payment system based on QR codes and ACH payments cutting the transaction costs to 4c. Two cents would go to the FI for processing an ACH payment and the other two would go to the technology partners and towards future MCX development. If it is indeed a confirmation that MCX are inclined to build alternatives to cards, then it is very interesting. However, it is still not clear how such a payments system would work:
  • Would the QR code identify the customer, the merchant’s payment request or just the merchant?
  • Would the customers be asked to register their bank account details with MCX wallet in the cloud? I can imaging this would be a big stumbling block for many consumers.
  • Will the transaction be based on ACH debit or credit?
  • If it’s debit, how will the authorisation happen? If there is no authorisation, will the fraud costs just become unacceptably high negating any savings on the interchange? There is speculation that consumers would be asked to register their debit card, which would be used for authorisation over card network rails, and then the transaction would convert into an ACH debit for clearing and settlement. If that’s the case, the overall transcation costs need to include the authorisation fee as well. And it sounds very similar to many decoupled debit propositions, most of which have failed to ignite the market so far.
  • If it’s credit, the authorisation challenge turns into the authentication challenge. One way to solve it would be to ask a customer to log-in to their bank account (e.g. through a mobile banking app) and authorise a payment to the merchant. Somebody would also need to pass a token to the customer’s bank with the payment request details. This is pretty much how Online Banking ePayments (OBeP) networks work; however, attempts to build such a network in the US (e.g. NACHA’s Secure Vault Payments) have again had limited success so far.
More questions can be raised and that’s just about the technical aspects of the solution. Commercial and other questions might prove to be just as difficult to answer. Will the banks co-operate? Will the proposed restrictions for participating retailers to accept other types of mobile payments (e.g. Isis or Google) work against MCX? Will the stated desire not to share any customer data amongst the participants limit the commercial opportunity? And will the (inevitable) delays to a project of such scale and uncertainty grind the intiative to a halt before it even has a chance to take off? Only time will tell if MCX succeeds. For now, I suggest we continue to keep an eye on its progress with a healthy dose of scepticism.