(I’m pleased to welcome back Peter Lugli, as a guest poster on the blog.  Peter is a friend, former colleague, and veteran of Prime Revenue, Ariba, Amazon, and Mastercard.  He’s also a great writer!)

THOSE OF US IN THE PAYMENTS SPACE are a bunch of hopeful nerds. Moving big bucks from A to B in new ways, looking for a killing in doing it … we eat this stuff up. Investors listen to our arm-wavy take on the next payments mousetrap because of the palpable excitement around the formula (Proximity to Massive Money Flow) + (Tiny Bps Skimmed Off Top) = Big Easy Money. “Look at Musk and Thiel”, we all say, “and the PayPal mafia they spawned… They’re building cars and rockets to Mars. Today, it’s the payment space… tomorrow, it’s space!”

In short, there’s no lacking idealists and crazy ones in the otherwise ho-hum world of payments. And yes, some do see themselves in the ultimate tribute to Crazy Ones in Apple’s iconic ad.

Alas, crazies rarely win. McKinsey points out that “[o]f more than 200 [payments] systems introduced between 1993 and 2000 […] only PayPal emerged as a standout success.” As is its wont, The Financial Times weighs in with the ring of brutal truth “Can payments eat the world? Probably not”, publishing, rightly I think, this take:

Public-market valuations across payments and merchant acquirers had tanked even before the collapsing fintech bubble started taking down banks like SVB. […] These falling valuations look more and more like the “new normal” for five major reasons: the normalization of e-commerce and small-to-medium enterprise spending, fierce competition, mixed network effects, labor-intensive investments, and IT technical debt.

Ironically, crazies that do enjoy some success grow the platforms and competitive moats of the incumbent card networks, usually through streamlined APIs and user experiences to make it easier to accept and, to a lesser extent issue, credit cards. In the end, crazy ones either get crushed or conform to the platform behemoths to get by.

But ‘getting by’ isn’t ‘crushing it’. Stripe, Block (formerly Square), Adyen, and so many others have built some cool stuff, but one is left with the sense that many are going where the puck is, not where it is going. Stripe’s down-round last year was punctuated by a more recent one this past month.  Block has received some rather unflattering short seller attention aimed at Cash App as a preferred remittance method for hitmen (hitpersons?).

I’ll waste word count with commentary on crypto and blockchain. These were supposed to change the payments world. Today, we’re realizing that (a) crypto is soooo 2021 (b) it’s not about payments in practice unless (c) you need a useful way to bribe offshore police and you’re comfortable with government-issued ankle bracelets. The jig is up. Bitcoin seems like just one big wash trade. It used to be you were someone in crypto if you could issue your own coin;  now, you’re someone in crypto if you’re arrested or on an Interpol Red Notice. The ultimate slight is that grown-ups like Jaime Dimon aren’t complaining about crypto firms stealing their lunch anymore.

Before ChatGPT and LLMs sucked the air out of the room, Big Tech’s hopeful nerds salivated in the payment space too. And failed. Google flopped, most recently their Plex checking account. Meta fell gloriously on its face with Libra (formerly Diem) and is now creating a whole new reality, perhaps because this one sucks for them. My former employer Amazon seems to be content merely to build an Amazon Pay button third parties can use to help Amazon customers buy stuff with the payment method they’ve stored on Amazon. I fell asleep writing that sentence. Last week’s announcement of a palm reader for entering payments credentials feels like, I don’t know, a BOGO of hubris and Fire Phone Redux.

Amazon Palm Reader

Microsoft has no noteworthy payment ambitions. It is busy incorporating ChatGPT into its software fabric that some fear will extinguish humanity, rendering this payments discussion moot, I guess.

And then there’s Apple, from whom we expect smart crazy stuff. How do ‘crazy ones’ handle something as unchanged as payments, I’ve wondered, which is necessarily hard, risk-laden, and fraught with roadkill? Apple Pay Later – where Apple takes on credit and risk decisioning on its own for the first time – is, I think, a way to peer into the mind of the crazies in Cupertino. More on that in a minute.

Fixing what ain’t broke is tricky…

With so much detritus, what is the payment innovator’s inconvenient truth?

Money that needs to get from point A to point B – whether in a B2B, B2C, or P2P context – does so rather well. If I could give a fractional star rating, I’d give it 3.8 stars – it sucks sometimes, but it’s okayish. Gnarly edges include things like cross-border payments, high remittance costs and fees, paper checks, fraud, AP and AR departments to manage exceptions. But darn it, The Payments System works, and the wheels of global commerce grind away.

Diners' Club Card

The credit card paradigm was born at Major’s Cabin Grill in Manhattan 75 years ago when Frank McNamara devised the Diners Club card. One can still marvel at how, say, a US-issued credit card, swiped at a Din Tai Fung checkout in Singapore, allows one to enjoy a dumpling feast without speaking the local language or getting arrested. The gang at Diners (now Discover) really came up with a good one.

The same inconvenient truth (but here, probably 3.5 stars) holds for non-card payments between businesses. We won’t double-click as much on B2B payments in this post, but, as their retail consumer brethren, B2B payments are gnarly, but they work – were that not the case, global trade would have ground to a halt centuries ago.  Companies pay, companies get paid.

… but your margin is my opportunity.

Ah, but that doesn’t mean incumbents have reason to comfortably sit around counting golden eggs and coddling geese. (And the smart ones are not sleeping well.) Players siphoning interchange from commerce have much to keep them up at night. Let’s face it – two or three hundred basis points in swipe fees is a juicy goose!

In the US, the National Retail Federation estimates that interchange yields $138 billion in fees each year, driving annual household costs up $900 in 2021. Put another way, interchange fees charged in the US were greater than the 2022 revenues of each of Chase, Bank of America, Wells Fargo, Citi, and Goldman Sachs. Lobby firm Restaurant Canada’s vice president stated, “Even in the best of times, our profit margins are around 4 percent; post-pandemic, on average it’s around 1.5 to 2 percent. We’re basically giving the credit-card companies more money than we’re making ourselves.” (emphasis added). “Take that!”, those of you saying 2% to 3% is a small price to pay for security, fraud, and payment risk mitigation.

So yes, there is money being made. Gobs of it. The credit card industry’s margin is going to be somebody’s opportunity. But… Whose? Where? How? Are we going to have to wait for Elon to do it again?

Ark’s Payments Covenant: Crazier?

Cathie Wood’s Ark ETF seems to find the payments margin opportunity, and the gang to get it there. In its Big Ideas 2023 paper, Ark points out that there are 9 steps between Buyer and Seller in a consumer payment transaction, sucking intermediary fees of roughly 2.8% of the value of the purchase. Ark believes enormous money is to be made if the steps between Buyer and Seller would be reduced from 9 to 3 (removing card networks, issuers, and acquirers, for the most part), thereby reducing the expense from 1.64% to 0.21% of the value of the transaction (leaving a massive 2.60% ‘take rate’ for its horse). If you’re looking for crazy, Cathie Wood’s group is a hot mess full of them.

Ark concludes that Block – one of its fund’s investments, and yes, that Cash App gangsta payment platform – is the horse to chow down on that juicy margin haystack. To wit, this colorful chart from its Big Ideas 2023 deck:

chart of economics of closed loop payments

But stop – wouldn’t something else be even crazier? Let’s use Ark’s own chart and color palette to reveal a third, diabolical opportunity proffered by your humble scribe:

Alternative Closed Loop Economics

In the chart here, Pretend NewCo would offer an ‘interchange’ fee of 1%, and costs of 0.21% to yield a 0.79% take rate (to borrow Ark’s terms here). Think that could ignite a new consumer-merchant payment platform? Particularly if this method were wed into a smartphone digital wallet? Would merchants paying half their net income to payment providers today actually prefer this method enough to switch from their current settlement method?

That sounds crazy.

Whither Go The Cupertino Crazy Ones?

This brings us to Apple.  Last week, Apple announced the release of its much-vaunted and quite delayed Apple Pay Later service, its version of Buy Now Pay Later (BNPL). It adds this functionality to payment capabilities that include Apple Pay (a way to link your existing debit or credit card in Apple Wallet to make purchases in stores or online), Apple Card (a Mastercard credit card issued by Goldman Sachs), and Apple Cash, a peer-to-peer payment service à la Venmo or Zelle. There’s reporting of a high-yield savings account to be added in the coming months, in partnership with Goldman Sachs.

Apple Pay Later Screenshot

It looks like Pay Later will be incorporated within the Apple Pay product in Wallet, based on its news release screenshot. At the time of writing, the Apple.com website is largely silent about Pay Later, so this feels like ‘a feature, not a product’. The release speaks to a gingerly rollout here (limited US deployment, invitation only, pre-release, purchases no more than $1,000, yada-yada).

No wonder. It looks like any Pay Later non-payment risk is on Apple’s balance sheet, unlike Apple Card’s issuer Goldman Sachs. Compared to other BNPL schemes, this writer hasn’t found anything about late payment fees, debt collection, or charges that kick in upon non-payment of a scheduled amount.

As a result, we see a lot of the ‘normal’ payments stuff happening around risk identification containment. Apple is making the credit decisioning and lending risk via its subsidiary Apple Financing, LLC. Apple Financing lawyers have been busy acquiring money transmissions and consumer lending licenses (12 by my count to date. Apple Pay Later could be rebranded “Apple Pay Lawyers”).

So far, so boring (for Apple). Is there anything new or noteworthy?

Well, yes, somewhat. Under the hood, Apple Pay Later appears to introduce a credit risk decisioning process on the device itself (iPad, iPhone), leveraging the innocuous phrase “device use patterns”. The key paragraph in the Apple Pay Later privacy policy:

“As with Apple Pay, your device may also evaluate device use patterns (for example, percent of time device is in motion, approximate number of calls per week) to help identify fraud. The information evaluated by your device is not shared with Apple or Apple Financing in a way that can be linked to you.”

Of course, the “for example” list is not exhaustive of device use patterns. In theory, device use patterns may include such minutiae as location data, app usage, call and text logs, biometric data, battery usage, screen time, internet browsing history, device orientation, gyroscope and accelerometer data, health and fitness data, ambient light sensor data, audio data, touch and gesture data, device performance data, Bluetooth data, among others.

People sitting at old style punch card machines

Put another way, if you are walking on death row in a Florida jail, multiple texts are going unanswered, and you want to use Pay Later to buy a brand new iPad for your mom, that ‘Pay Later’ tab will probably not show up on your phone at checkout (ironically, my guess is Goldman Sachs may still permission payment with Apple Card in this scenario if you’ve been pre-approved).

OK, that was dark. But companies have built entire product offerings around this data. Experian offers a Smart Digital Footprint Score, which plainly seeks to assess “consumer’s credit risk through smartphone data”. Experian’s product sheet outlines the following:

Our Smartphone Digital Footprint (SDF) Score bridges this gap of data availability from traditional credit bureaus by enabling financial institutions to credit score a wider base of applicants and grow their credit portfolio in an efficient and scalable manner. Thousands of data attributes derived using smartphone data elements are used to build credit risk models, thus allowing a credit risk score to be generated even for the unbanked and new-to-credit customers. (emphasis added).

Is Apple going here? True, the “device use patterns” referenced in the Privacy Policy are meant to identify fraud (that is, they’re in a paragraph that speaks to fraud). But is there anything preventing the use of the same information to derive – algorithmically, locally on the device, and outside the eyes of Apple and Apple Financing LLC – a credit score for Apple Pay Later decisioning? This approach could meet Apple’s commitment to privacy and security while minimizing the amount of personal data shared with external parties. That’d be crazy like a fox and perhaps serve as a model for later, and bigger, credit risk assessment paradigms at Apple.

So what about that 1% interchange? Is that where Apple is going?

Oof. Wouldn’t that be crazy? In the short term, there are some reports that’s a no-go. Card networks wouldn’t permit that, of course, and there’s some reporting that Apple has agreed to not build a competing card network in return for a 0.15% fee on every Apple Card purchase. Writes the Wall Street Journal in October 2021:

When Apple introduced Apple Pay in 2014, the iPhone had already clobbered music players, cameras, and GPS systems. Banks and card networks worried it also would displace card payments.

Banks agreed to pay Apple 0.15% of each purchase made by their credit cardholders. […]

Visa and Mastercard also agreed to give Apple an unusual concession, according to people familiar with the matter: Apple would be able to choose which issuers it would allow onto Apple Pay and which of those issuers’ cards it would accept. Visa and Mastercard generally require that entities that accept their credit cards must accept them all. Apple agreed to not develop a card network to compete against Visa and Mastercard, the people said.

As an aside, there hasn’t been any reporting I’ve seen on this agreement since the 2021 piece. Which is weird, because, uh, you know, one could argue that’s, uh, Sherman Act triggering. Or is this silence the sound of oligopolies clapping? I don’t know. Of course, if true, all players mentioned would have an interest in keeping this rather hush-hush.

Is Apple going to become a bank?

Many have wondered whether Apple was ever going to “become a bank”. At least one bank head says Apple’s already there. Perhaps over another at-risk lunch, Jamie Dimon reportedly points out:

“They’ve already got the Apple Wallet. They want to give you some kind of credit journey experience, they’re going to do merchant processing, they’re going to do merchant lending. It may not be their own balance sheet. But that’s a bank. That’s a bank. It may not have insured deposits, but it’s a bank. If you move money, hold money, manage money, lend money, that’s a bank.”

Mr. Dimon sounds perturbed. He ought to be. Data suggests 94% of US adults own a smartphone, while the Federal Reserve tells us 84% have at least one credit card. The Fed also tells us only 81% of US adults are ‘fully banked’.

I don’t think Apple’s crazy ones want it to become a bank. Given the regulatory oversight involved and the risk aversion built into the DNA of banks, that’d be nuts. There’s a difference between crazy and nuts. I think Apple’s intent has been to better solve problems that people look to banks to solve. It’s all that “skating to where the puck is going to be, not where it’s been” stuff that Gretzky proclaimed, and Jobs built into Apple M.O. But certainly, the interchange market itself (not including the broader world of business-to-business payments which we haven’t touched on here) is a major prize. That $138B interchange fee market size is about a third of Apple’s revenues globally.

What Apple Pay Later Tells Us

Apple Pay Later, in diving into the shallow end of the consumer credit risk assessment pool, appears to be a surprisingly timid step for a place full of ‘Crazy Ones’. They’re making it invitation-only. They’re limiting the ticket size. They’re limiting geographical availability. They’re pulling consumer rating agency scores. This isn’t crazy. It’s carefully measured. But perhaps there’s ‘gangsta crazy’ like Block, and ‘crazy like a fox’ in Apple’s case.


Here’s my take on all this, and one crazy suggestion.

  • Apple Pay Later is not crazy, but it is a much bigger step than it appears. Getting into consumer credit decisioning is a massive tell on intent. But it would be wrong to say Apple is “going to be a bank”. That’s skating to where the puck is. It wants to solve problems banks solve for people – only better, and probably in ways not imagined before by traditional banks. For example, a few colleagues and I busted our pick (alas, unsuccessfully) building a business around product authentication and integration to the point of sale. What if payments were as much about the things we buy, as they are about paying for them? Couldn’t an NFC-chip and camera-enabled smartphone together with a modernized payment infrastructure help with that in a way that creates new value for consumers, merchants, and brands around the stuff we buy and sell? Call it an ‘informed payment’? That’s “where the puck is going” stuff, which banks and card players are unequipped to put in place, and arguably ill-equipped to ponder.
  • Apple Pay Later is neither a feature nor a product – it is a learning opportunity. Apple will take tiny steps via Pay Later to learn and refine its credit risk model, so it can measure default risk more precisely. Inventions that result here (in-house or acquired) will create an opportunity to launch at scale while fine-tuning and minimizing risk cost. Ultimately, it can also inform a decision to syndicate payment risk default with other lenders, who by then will be salivating to participate in such a hermetically sealed risk environment. Apple can take its time here – no bank or card network has a smartphone infrastructure. There’s precedent here of course: Apple did kinda okay with its smartphone, despite being later in the game to market.
  • Public Policy. Smartphone penetration in the US far exceeds penetration of credit cards, and bank accounts. Apple can clothe itself in the garb of providing better money management, for more people, than banks do today. There is an interesting public policy play here – banking the underserved – which may pre-empt, or at least blunt, regulatory scrutiny.
  • FICO is dead in the long run. Our smartphone will become our faceless, non-human credit risk determiner. It may help, or hinder, our access to credit. But over time, neither third-party, ponderous, and untimely consumer rating agencies, nor Apple-authorized human beings will be the ones determining your credit risk profile: your phone will.
  • A 1% (or less) interchange fee, anyone? True disruption exists in building new commercial models around the interchange, which could help small and medium-sized businesses, particularly those for whom interchange is a material portion of revenues and profits. In an inflationary environment, that’s good not only for business but for cash-strapped consumers, who by some estimates today spend (in the US) $900 annually to support the current interchange platform. Apple is not there yet (and their hands may be tied?) but the pieces are being assembled (credit assessment in Apple Pay Later being a vital capability).
  • Who do you trust more? Apple has spent years building a trusted brand and image of a careful preserver of privacy, which positions it well for entry into financial services. It would be interesting to test this (and I’m sure many have privately!) whether Apple would be better trusted than banks to manage financial affairs. Would you buy a house with an Apple mortgage? Get your paycheck delivered to an Apple Pay Wallet?
  • Crazy suggestion? Deep breath. Apple could buy Discover (DFS $28B market cap at the time of writing), in order to (a) disrupt the interchange pricing model with a rock bottom interchange rate (possibly two-thirds less expensive than current rates), (b) adopt an existing merchant onboarding capability and merchant community that is Discover-enabled (c) gain a built-in payments-centric customer support infrastructure that would handle Apple payment family support issues and (d) inherit banking ‘credentials’, bona fides and FDIC coverage via Discover Bank. There’s a certain poetic craziness to this of course – should it happen, Apple will have acquired the company that came up with the credit card paradigm in the first place. It’d be neat to know if Diners’ Club inventor Frank McNamara was a Bob Dylan fan. Now that’d be crazy.

Watch this space!

Peter Lugli plugli@digitalpylon.net

Read More