We need to reimagine the payments rails.
Don't get me wrong. The rails are modern miracles. FedWire has been processing interbank payments for something on the order of 108 years. That's more than a century of continuously providing a payments backbone for the US. FedACH has been in the US since 1972 - before I was even born, albeit not by much, and Britain had BACS before that. Cash and Check are even longer in the tooth, but still robust solutions to national needs. That kind of continuity and reliability are incredible; the US alone processes more than $5T a day in payments over the rails, day in, and day out for decades.
I'm using the US rails as examples, but I'm speaking universally.
Payments rails meet a critical need; that's inarguable.
But...
Is that enough?
If we presuppose payments innovations add economic value (spoiler: they do), then innovation, evolution and access are the name of the game.
PWC found that increased digital payments included increased household spending, business efficiencies, and entrepreneurial growth, with business efficiency and redeployment of resources contributing to $142.4B in GDP in 2024 (source). BIS found an increase of 1 percentage point in digital payments increased GDP by 0.1 percentage point, and reduced informal (untaxed) labor participation (source). In a $31.8T economy like the US, that's decidedly non trivial.
So why do I say the rails need revolution? They serve a profoundly valuable purpose. The nation's economy runs on them. Heck, EVERY nation's economy runs on payment rails. Financial institutions have built incredible amounts of technology around them, and there's residual legislation and regulation that effectively enshrine them. They're not going anywhere - we need rails. Or do we?
What we need - what's baked into existing tech and process and law - is something that looks and functions like rails. That sounds like a distinction without meaning but I posit that it's not.
In the US alone five trillion dollars a day depend on the functionality of rails, and that kind of business needs to continue to be supported, enabled and grown. FedNow adds enormously to that innovation and potential for growth.
Great, but what are payment rails? I'm oversimplifying, but they're payment services that provide infrastructure, and a network, and offer functions including routing, clearing, settlement, rules, and standards. The collection of those things, plus a few others is basically what makes a payments rail. Each rail has tradeoffs (cost, speed, security).
Each rail's specific combination of those things creates its value proposition - and limitations - of the rail, and we've organised for decades around those. That gives us a powerful set of optionality to leverage in payments. But it also adds complexity.
Banks need to maintain the infrastructure, services, and connective fabric to manage the rails, their protocols, data formats, rules etc, and businesses have to organise around them.
So what's next?
It's hard to argue that payments isn't innovating at an unprecedented rate, from something as banal as BNPL, through to real innovations like cryptocurrency (warts and all) and CBDC. Stripe, Square etc drive further innovation. Increased appetite for global payments infrastructure is nothing new, but with the international democratisation of payments (e.g. through Stablecoin), rails as they exist today have limitations in supporting this pace and innovation.
I'd like to see an industrial revolution of the rails that would transform them into more - a true substrate of payments innovation.
Let's think of rails as a prepackaged collection of the things rails offer. That's very necessary. They offer well understood services. You could think of them as being like different model cars, say a Fit, a Civic, a CRV and an Odyssey (I drive a Honda, so I'm going to use that as my example). Great for the masses. Most of us need a car that Honda can provide.
But what if we think of the things a rail offers were actually a collection of capabilities, that coud be assembled a la carte? Payments as a service. Send the payment details, the quality of service, and the rules to apply*, and let the network operator get it to the right place, with the right constraints, and the choose the right protocol. Payments as a service. You can still offer things that look and behave like rails, but instead of actually being rails, they're one set of products in a diversified payment product portfolio. Instead of assembly line Hondas, which will continue to be the right answer for the vast majority, everyone who wants to can have a custom Ferrari. (If you want to understand that, check out the excellent Acquired Podcast Ferrari episode.)
Banks can enrich their service offerings, while giving customers more and dynamic capabilities, and continue to operate existing services while growing other aspects of the business. We democratise the business of payments, not just access to payments.
Let's take that a step further. Why not offer payments services as payments primitives - lego blocks on which payment services can be built?
Network operators have valuable tech and insights that only they can provide. Sell those services - monetise them, while allowing the industry to innovate, and everyone to benefit. Think things like Network level fraud prevention tools, alias and identity resolution (serving as a trust anchor for its own network, and any others that want to leverage trust, but not processing), deployed like AWS, Google, or Azure services - Payments as a Service.
Since we're on a journey, let's take that one more step further. Most of the things a rail does are directly applicable to new and emerging models. There's no reason that payments as we think of them now should be the only things these payment legos can be used for. Any transfer of digitised value could make use of them.
Innovation at scale, powered by trust - trust that leverages the scale of existing networks.
Rethinking payments in this way would enable better support of the leading edge; reputable services of record to support and facilitate other payment innovations like those in the crypto and digital spaces. Say, a trusted smart contract capability: smart contracts accounted for more than 45% of crypto losses in H1 2024. (source)
If a single percentage point of digital payment growth can impact the economy, it behooves us to make it easier, and safer for that growth to happen; we can lay the groundwork to make this happen.
The future of payments is bright, and it will continue to be powered by the rails. Instant Payments offer a wealth of opportunity for businesses to innovate and banks to offer new services and products. But how much more potential could we unlock by moving - and crucially thinking - beyond rails?
Disclaimer: I am not a lawyer, nor a policy wonk, but it's my understanding that in the US at least, this would require legislative and regulatory change to really push the envelope. But transformation can begin with small changes. However, the new executive order to explore options might unearth some creative apporaches (source).
* Yes, I know some rulesets are incompatible with some protocols.