Tomorrow I get on a flight to Singapore for SIBOS to speak at one session and moderate another.

SWIFT (which runs SIBOS) has about 9,000 member Banks. Correspondent banking networks have been integral to global banking and trade since the days we sent messages over telex machines. This got me thinking:

What would a world without correspondent banking look like?

There is a good news, bad news, good news story:

• Good news: global trade flows are increasing. Stats junkies can go here.  TL:DR. Alibaba is valued by investors over $150 billion because global trade flows are increasing. That is 150x more than a Unicorn.

• Bad news: the price of payments is likely to be decimated (10x or more lower). 

• Good news: if cross border payment costs plummet, volumes will increase. This has happened in every other market impacted by digitization. Millions of micro entrepreneurs globally will be able to trade cross border as easily as they trade domestically. That will lift billions out of poverty (a process already starting as per this great article by the ever inspiring Nicholas Kristof) which will be good news for the global economy.

So banks will do just fine as long as they add value beyond payment.

There are four possible scenarios:

• # 1: The current SWIFT payment rails including Correspondent Banking remains dominant. This is the least likely scenario IMHO.  It is simply too slow and expensive. It is expensive because time costs money in the markets. A multi day settlement time creates market and credit risk, which is why cross border payments are so expensive.

• # 2: A few global banks acquire or crush all the regional banks. If both the sender and receiver use the same bank it is a simple change in two ledgers controlled by the same system. This enables the global bank to reduce fees for both sender and receiver and still make a profit. This again is unlikely, because too big to fail then becomes a global problem and regulators will intervene. The economy as a whole needs a healthy number of small to medium banks.

• # 3: A new cross border payment rail emerges that cuts out the banks. It will work in “human real time” (a few seconds). It could be Blockchain based. Or it could be Credit Card based.

• # 4: A new cross border payment rail emerges that keeps the banks in the loop.  I think this the most likely scenario. It will be “human real time” and will use something like correspondent banking networks (basically cooperation agreements). Some entity still needs to mediate that and provide core messaging and identity services and this excellent interview in American Banker with Gottfried Leibbrandt of SWIFT describes how they are thinking about that role.

If you want to understand how cross border payments work today via SWIFT and correspondent banking relationships, read this excellent explainer by Richard Gendal Brown. If you have labored at the core banking coal face and can explain Nostro and Vostro accounts and Letters of Credit….move right along.

The language of Letters of Credit really takes you back to Victorian days such as “bills of lading & “payment day”. This got modernized via workflow systems and scanned documents. To a millennial, the scanner is like a fax machine – an analog relic and historical footnote. The future is clearly digital trade finance.

Actually, a lot of the Letter of Credit business has already moved over to Open Account trading. Maybe the new cross border payment rail is simply the credit card? I don’t think so. My children may remember their parents pulling out credit cards to pay for stuff, but it is hard to imagine them doing that as adults.

The practical questions for the Daily Fintech community are:

• What would a new cross border payment rail look like?

• What new models for regional banks will emerge?

This is where we come to that provocative headline that the real battle could be between big banks vs small banks. Global banks have less need for relationships with regional banks whereas payment companies will need those regional banks.

Payment companies view themselves as one layer in the stack and want to partner with those 9,000 consumer-facing banks.

I think we will still see something like correspondent banking networks (basically mutually beneficial cooperation agreements) but operating in real time (which totally changes the game as we explore in this post).

A global bank is a vertically integrated entity that can do everything from payment to customer service. At the payment level, the transaction banking level, they look like a Fintech (almost totally digital and automated). Will that part of the bank operate like an independent Fintech and partner with those 9,000 consumer-facing banks? Or will the global Bank cross subsidize and compete with those 9,000 consumer-facing banks? In other words, will big global banks be tightly coupled or loosely coupled?

I will be off air for hours in the air on route to Singapore, then looking forward to some interesting conversations next week at SIBOS – and great pepper crab.

Daily Fintech Advisers (the commercial arm of this open source research site) can help implement strategies related to the topics written about here. Contact us to start a conversation.