Maybe it was by now a foregone conclusion, but this week Meta (Facebook) officially abandoned its cryptocurrency project Diem. Dollar prices of key crypto assets bitcoin and ethereum are both off by a third from their highs of late 2021. A sobering time for the crypto world, perhaps. Still, at least some of the innovations born from crypto experiments will find permanent homes in the financial system.
Continuing my recent look at JP Morgan's blockchain projects has led me, thanks to comments from readers, to try to get more precise about exactly what value the bank sees in blockchain technology per se. It turns out that JPM has been making the case pretty clearly. In this post, I look at one group of use cases—payment information—that JPM has found for its private blockchain Liink. The second group of uses cases has to do with programmable contracts—a topic for a future post.
The payment information problem
Perhaps the single most generic function of a bank is to send money on its depositors' behalf. Though generic, it can be quite difficult: transfers frequently involve chains of intermediary banks, customers have little tolerance for error, and regulators require near-perfect compliance with know-your-customer rules, anti-money laundering rules, sanctions compliance and so on. There are also constraints on how data can be moved across lines of ownership and jurisdiction.
JP Morgan is a money-center bank that provides payment services to other banks and to large corporations, and so it is keenly aware of such difficulties. To get it just right, banks must constantly manage a flow of payment information that moves alongside the funds as they flow between intermediaries. The situation is not unlike the problem faced by air carriers, who transport not just travelers but also their luggage through the global air travel system. Like people and their suitcases, the funds and the payment information start and end their trip together, but they might travel along separate channels.
Blockchain for payment information
Christine Moy, global head of JPM's Liink blockchain, observes that frequently a payment might be flagged for manual verification. This could be the confirmation of an account number, or a beneficiary's address or date of birth. The status quo solution to this problem is that the banks involved make a series of phone calls. The verification steps follow the chain of correspondent banks, which might be several intermediaries long. This is time-consuming and expensive.
JPM's service uses blockchain to allow such confirmations to occur bilaterally, directly between the endpoints. For this to work, each endpoint has to operate a node in the Liink network. For a fee, a sending bank can post a request to confirm information to the service. The receiving bank can confirm that the sending bank has the correct information, receiving some of the fee.
Importantly, this happens without either bank actually transmitting privileged information. Cryptography allows the banks to agree that they have the same information, and agree that they agree, and blockchain allows it to happen without having to keep data in a central server. Stretching the metaphor a bit, it is as though your airline hired a private jet to deliver your bag directly to your destination.
In a comment that now seems to have been removed, a reader points to this stop-and-recall service from SWIFT, which again basically asks whether blockchain is really necessary for any of this: https://www.swift.com/our-solutions/swift-gpi/stop-and-recall
Is it cynical to see JPM selling JPM Coin at a premium? Is it too heterodox to consider power as a motivating force?
If the Fed were smart (will current Fed tightening be seen as a policy error, like defending the gold standard during the Great Depression?), could they compete by paying the inflation rate as interest on CBDC accounts, using as excuse the mainstream argument that monetary policy can be implemented directly on an individual level, by encouraging microfoundations to save in inflationary times? (Will JPM claim its Coin pays off higher ...?)