If you're reading this and still haven't signed up, click the subscribe button below!

Simon Taylor is joined this week by Noah Levine, Partner at Andreessen Horowitz and our guest host; Thierry Edde, head of crypto at Deel; and William Peck, head of digital assets at WisdomTree.

🎙️ Listen to the latest episode of Tokenized here,  or📷 watch on YouTube.

Banks Chose Deposits Because a Deposit Keeps the Customer

JPMorgan, Bank of America, Citi, and Wells Fargo are building a shared tokenized deposit network through The Clearing House (TCH). The same group explored a joint bank stablecoin a year ago and dropped it. The reason they switched is structural. "Issuing a stablecoin doesn't really make a lot of sense" for a bank, Noah said, "given the limitations on what you can do with the reserves" - high-quality liquid assets, short-dated, no lending against them. A tokenized deposit, by contrast, is still a deposit. Anyone holding one has to be a customer of that bank. That makes it a retention tool: the bank keeps the relationship, and everything attached to it.

It's also a constraint. A deposit can't leave the bank perimeter; it has to be cleared or swapped, which is why a clearing house is the chosen rail. The strongest fit today is corporate treasury, where products like JPM's Kinexys and Citi Token Services already have traction. But the fragmentation cost is real even there. Deel maintains over 500 bank accounts globally, Thierry said. As Simon pointed out, dealing with Citi or JPM across different markets is effectively a different bank each time - different KYC, different logins, different cutoff times. What Deel needs, Thierry said, is tokenized deposits on an "open network" that let it settle between its own accounts instantly, around the clock - something it can't do today across entities under its own umbrella.

For payments and fintech use cases beyond that corporate perimeter, Noah was blunt -  "I'm not sure they're going to have a significant amount of opportunity to be effective in them." Banks will tell you tokenized deposits are obviously better. The crypto industry will tell you stablecoins are obviously better. Both arguments are incomplete, and the parts each side leaves out are where the real risk sits.

The Redemption Mismatch Both Sides Ignore

The sharpest exchange came over deposit insurance. With a 1:1-backed, bankruptcy-remote, regulated stablecoin, Noah asked, "do you really need FDIC insurance if you have all those protections?" Simon pushed back on the premise that 1:1 backing equals safety. That backing isn't held in cash - it's in money market funds, treasuries, and repo. So in a run, the issuer has to liquidate those into the market fast enough to honor redemptions, and the timing gap between selling the assets and producing cash is where it breaks. There's still a redemption timing mismatch here that's really important.

Will Peck added the part most people miss: the FDIC's real value is the orderly wind-down. "If something happens on Friday, Monday they're going to open the bank" - a backstop neither stablecoins nor money market funds have been stress-tested against. "Nobody's ever going to build the one tokenized deposit network to rule them all," Simon said. The conclusion is coexistence, but not because both instruments are equally good. Banks aren't building for consumers. Stablecoin issuers aren't building for corporate treasury. The deposit war has already split along that line.

Privacy Is the Last Barrier. And Why No Chain Wins Alone

If the deposit-vs-stablecoin debate is a fight over trust mechanics, privacy is the other half of that question, and the one institutions still can't get comfortable with. Canton's $355 million round, led by a16z, was the episode's clearest illustration of why. Noah listed three barriers to institutional adoption - throughput, regulatory clarity, and privacy. The first two are largely solved. Privacy isn't. That's why a confidential-by-design chain finds traction where public chains don't. His broader point cuts against the whole deposit war: blockchains are "not a winner-take-all category." Canton specializes in capital markets, Tempo in payments, each chasing massive, separate TAMs.

Deel Becomes the Account

Deel was the more consequential story. The $17.3 billion payroll platform launched DLUSD so Argentine contractors stop routing through multiple hubs just to hold dollars. Hold a dollar balance, earn rewards via Morpho, spend it on a card - all inside the same app the paycheck lands in. The blockchain is invisible; contractors see a dollar balance redeemable 1:1.

Noah's read on what this means was broader than Deel. Companies that sit in the payment flow today have a chance to extend beyond the payout and become the primary financial account. "Stablecoins are a massive catalyst for non-banks that want to expand deeper into financial services," he said. Will drew the parallel to ETFs - twenty years ago, accessing any liquid asset anywhere required new paperwork with every provider. ETFs collapsed that into a single brokerage account. Wallets can do the same thing, across currencies and asset classes, on one set of rails.

But getting there requires more than a token. Thierry was clear on what Deel's enterprise customers ask first: are those transactions private? No employer wants salary data disclosed on-chain. And Deel has the operational scars to know how hard the infrastructure is - 500 bank accounts globally, each with different KYC, each effectively a different bank. That complexity is what makes the embedded wallet attractive: it replaces the patchwork, not just the payment.

One operational detail reframes the regulatory debate. Under the GENIUS Act, the most compliant route to an earn product for Deel wasn't a bank product - it was DeFi, via Morpho, which raised $175 million this week from Paradigm, a16z, and Ribbit. Thierry, who comes from Lebanon - "a country where the banking system is completely broken," as he put it - called holding dollars and earning yield on-platform a mission "dear to my heart."

That same shift showed up in the week's other deal. Figure agreed to acquire Kiavi, an AI real-estate lender that originated $7.8 billion in 2025, for $717 million. Kiavi's book is fix-and-flip and rental loans - assets banks mostly won't touch. It's the same dynamic playing out across asset classes - when the incumbent steps back, an on-chain player takes the relationship.

So the deposit war doesn't end with stablecoin legislation. It migrates: bank rails for corporate treasury; embedded wallets for the workforce. The institution that figures out the second cohort first doesn't just win a payments flow - it wins the account.

Thanks so much for reading the Tokenized Newsletter!

Please share this edition or share it with your colleagues if you enjoyed it!

Disclaimers