Every Bank Should Tokenize Deposits

Welcome to Fintech Brainfood, the weekly deep dive into Fintech news, events, and analysis. You can subscribe by hitting the button below, and you can get in touch by hitting reply to the email (or subscribing then replying)

Hey Fintech Nerds 👋

Incredible week in SF. First ever Sardine Con was a huge success. Fintech Nerdcon announced its first 100+ speakers, and I recorded my favorite episode of the Tokenized podcast so far (look out for that Monday).

M0, MetaMask, and Bridge launched a dedicated stablecoin for MetaMask users. Why? Because if you own the stablecoin, you own the economics. And no, that won’t be the fragmentation nightmare some expect.

Meanwhile, everyone wants to know how tokenized deposits might work, so I took a stab at this week's 📣 Rant. The real signal for me this week was China looking at a Yuan-backed stablecoin via HK 👀. Lots to dig into there.

The CFPB has been busy. There’s a new notice for a proposed rulemaking on 1033 (that was quick), and we may be on the brink of our first-ever Fintech bailout. (More in Things to Know 👀 )

Can you believe it? We just announced our first 100 speakers for Fintech Nerdcon 🤓. This will be the most talent-dense, content-deep, human-first networking event you've ever been to. Tickets are gonna go faster than you can say 1033…

Weekly Rant 📣

Every bank should launch Tokenized Deposits

The passing of the GENIUS Act has legitimized stablecoins and created a land grab opportunity for the future of digital money. Tokenized deposits will be a massive opportunity for financial institutions to win market share in onchain finance. 

Banks already tokenize deposits. They've been doing it for years. HSBC, Citi, JPM - they all offer instant, global, 24/7 payments to their enterprise customers. The difference? It only works if you're lucky enough to bank with them on both sides.

Here's what's wild: the technology gap between JPM's internal tokenization and Circle's USDC is surprisingly small. The difference is that JPM initially built a members-only club. Circle built a public square.

Western Union announced a stablecoin last week. MetaMask launched one this month. Every payment company is building its own version of what banks already have. The question isn't whether tokenized money wins - it's whether banks participate or get bypassed entirely.

Far from being winner-take-all, stablecoins and tokenized deposits will co-exist and create a flywheel for onchain finance. They’ll solve different problems for different people, and they’ll each have features and drawbacks that make them better or worse. 

Banks: If you believe the future is onchain, then your balance sheet should be too.

What is a Tokenized Deposit?

A tokenized deposit is an on-chain representation of dollars 100% backed by bank deposits. You can think about tokenized deposits like “banks issuing a stablecoin backed by their balance sheet.” They’re essentially the same thing.

How is it different to a stablecoin?

Unlike stablecoins that are mostly backed by T-Bills and Repo, a bank deposit is backed by the creditworthiness of that bank and FDIC insurance. 

There are actually three types of fiat money that could be tokenized. They’re likely to be used by different customers to solve different problems.

• Stablecoins are digital cash onchain. They’re backed 1:1 with high-quality liquid assets (HQLA) like treasuries, repo, or bank deposits. They cannot pay yield. Users hold, send and receive stablecoins to solve friction points like hyper-inflation or moving dollars in and out of exotic regions and markets.

• Tokenized deposits are bank deposits onchain. Bank deposits are backed by the bank’s own balance sheet + FDIC (or equivalent) government-run insurance schemes. In this context, onchain means onto an open-loop, public network like Ethereum, Solana or Base.

• Tokenized reserves are government-issued onchain assets. Backed by the balance sheet of the central bank and credit-worthiness of the country (e.g. the USA). Instead of minting cash or a balance (or alongside that), they could mint a token and consider it a settlement.

I’ve written many times about the opportunity for stablecoins (which is enormous, albeit at the peak of inflated expectations). But I haven’t seen many folks think through the opportunity for tokenized deposits. That may not conform to a pure crypto ideology, but it does create a massive land grab opportunity. 

How would Tokenizing a deposit work?

The first thing to understand is that the deposit doesn’t go anywhere. When you “tokenize” the deposit, it still sits there as a balance in the bank’s system of record. That means from an accounting (ALM) perspective a bank could still lend against it, sweep it, do whatever it is they do with their balance sheet.

The tokenized deposit becomes its twin. 

The token creation then looks very similar to how tokens get created / destroyed for stablecoins through a process called minting and burning.

Minting deposit tokens:

1. A deposit is added to a minting account at a financial institution or stablecoin issuer 

2. A token in then minted using the smart contract’s mint function (e.g. USDC, or JPMD)

3. That token is sent to the wallet address of the beneficiary and the contract updates the internal balances 

4. The total supply of the token is increased by a corresponding amount

5. A transfer event is sent to the blockchain network to signal this has happened

Burning deposit tokens:

1. A token burn function is called by the smart contract with a specified amount to burn

2. The burn function reduces the senders balance by a specified amount from their wallet

3. The total supply of tokens is updated (reduced) and internal balances is updated

4. The transfer and burn event is sent to the blockchain network to signal this has happened

5. The deposit is debited from the minting account and credited to the beneficiary

If you wanted to get fancy, you could also add some clever referencing to your mint / burn process as a financial institution. E.g., If your mint account is an FBO with virtual accounts, you could use the underlying payment references during the minting / burning to tie each deposit/token back to a specific event and customer for an additional layer of auditability.

Banks are nowhere in this conversation yet.

So if banks already have the technology, why aren't they competing?

The Tokenized Deposit Land Grab

Every merchant, tech company, and money mover is asking and actively testing how these stablecoins can give them an edge in cost, speed, or revenue. In the past two weeks, Western Union announced a stablecoin, MetaMask launched a stablecoin, we saw an IPO where 50% of the proceeds were in stablecoins.

Some of this is headline-grabbing during earnings season. A lot of it is an intentional land grab for a market structure that is reshaping.

But the GENIUS Act gave banks an edge. Tokenized Deposits can pay yield. And Tokenized Deposits are deposits. Meaning banks can lend against them, sweep them, do whatever they like.

So if the market structure does change shape, banks have a window to compete.

And remember.

Banks already tokenize deposits. 

If you’re a customer of HSBC, Citi, or JP Morgan, they offer global, instant, and 24/7 payments, transacting 10s and 100s of billions weekly. If you’re a Fintech company or a global corporation, in many markets, the FX problem is solved. 

But these solutions are closed loop, only available to existing bank customers. 

They’re not interoperable. The stablecoin solutions are interoperable and more open-loop.

Tokenized Deposits could be more open loop

Once a deposit is tokenized, if it exists onchain, in theory, it could be sent to any wallet anywhere on that chain or that ecosystem. This means if JPMD was on Ethereum, Solana, or Base (and a standard smart contract):

• Anyone on that network could send or receive it

• Globally, instantly, 24/7

• You’d be able to view those transfers publicly

What’s more, these ecosystems are connecting. What was once a silo between Ethereum, Tron, Solana and Base can now be bridged. The wider Ethereum ecosystem is becoming increasingly compatible and performant.

Just as the internet began with networks, the internet turned them into a network of networks.

The same is happening with onchain finance.

Tokenized Deposits and Stablecoins will co-exist and complement each other

These things don’t compete. They solve different problems.

Stablecoins solve for the long-tail of users and use cases the existing financial system doesn’t serve well. They’re a global, instant, cash-like rail that help get in and out of exotic currencies and markets.

Tokenized deposits help the existing financial system get the same benefits of tokenization. Like access to global, instant, 24/7 and programmable payments in a more open loop way.

Today’s tokenization efforts are early, but when they’re complete, they’ll start to interact with stablecoins directly onchain:

• Offshore stablecoins go to the last mile and areas where the correspondent banking can’t reach

• Onshore stablecoins become a cash-like instrument for nonbanks and help connect offshore stablecoins back to legitimacy

• Onshore tokenized deposits can become interchangeable for stablecoins and other tokenized deposits in other currencies

• All of these can be swapped, instantly onchain.

Today, platforms like Hyperliquid, Aave, and Compound let users buy, sell, and trade tokens with incredibly low fees. This will come to stablecoins, and when it does, it will be an enormous unlock.

The world of onchain finance is becoming more like a network of networks. The problem with that is, unless we’re very careful, instant, global, 24/7 payments would be a privacy and confidentiality nightmare, and a dream for money launderers.

Not something banks could tread lightly into.

The hard bit is KYC/privacy onchain.

The way we solve these tradeoffs in TradFi is with centralized organizations protecting your data.

KYC isn’t just collecting data.

• The CIP (Customer Identification Process) has been largley solved in existing crypto. Custodial and self-custodial wallets KYC their users whenever they on-ramp/off-ramp or make a transaction above $1,000. 

• However, KYC also includes monitoring transactions, customer due diligence, and a whole swathe of processes that follow. Banks have the highest bar for getting this right and receive the biggest consequences if not.

The crypto industry solved this in a few ways

• Centralized exchanges and wallets KYC their customers

• They use services like Note Bene to ensure travel rule compliance if a transaction goes above $1,000

• They use blockchain specific transaction monitoring tools like Elliptic or TRM Labs.

But banks also need to ensure commercial and consumer privacy on those networks when it comes to deposits.

There are many possible approaches here.

• They can build their own L2s (like a VPN) where they can see the privacy / KYC tradeoffs, and then allow users to swap their deposit token and bridge to another network.

• Their deposit token smart contracts can restrict which wallets get to use the token to known, KYC’d entities.

• They can use Zero Knowledge Proofs or other privacy-preserving cryptography to not reveal sensitive PII or data.

Ultimately, this is a solvable problem. 

The market just hasn’t agreed precisely how to solve it yet.

We’re about to speedrun that process because the passing of the GENIUS Act means we’re in land grab mode for an open market, where banks have a massive threat, but just as much opportunity.

What right to win does a bank have in this new market?

Stablecoin issuers and stablecoins won’t work for everyone. Banks have:

• Brand power: People know and trust banks to be where their paycheck lands, or for large corporates with managing critical finance functions. 

• Segments: Older generations, large corporates and higher income segments often prefer the products or brands of the banks.

• Distribution (network effects): Banks already have millions of customers and deep relationships. If a new product or business model works, they can do their own version to a large audience.

Case on point. Zelle and mobile banking. Digital only apps like Cash App, Venmo and digital banks like Chime (or Revolut in Europe) developed meaningful market share. 

Banks have every right to have market share in a new technology

They just might not grow as fast as the innovators.

What models are banks using to enter the market?

All of them.

Banks like to own something themselves if they can, and if they can’t, own the consortia that owns the thing. Nobody wants to cede platform control to big tech. That’s why banks were slow to cloud migration, slow to Apple Pay, and want to own their platforms.

It’s smart.

That’s how most banking rails have been born throughout history. VISA was the BankAmericard that was ultimately opened up to more banks. SWIFT was a consortium of European banks that opened more widely. Zelle and The Clearing House (TCH) today are a consortium of the largest US banks.

It doesn’t always work. Banks tried to own mobile payments before Apple Pay ultimately happened in an ill-fated (and poorly named) effort called ISIS.

The thing with onchain finance though? When it’s done right nobody owns it. Even Coinbase’s Base is progressively decentralizing. Decentralization is a hedge against platform lock-in.

That makes deposit tokens a much more interesting opportunity.

And now because the future is uncertain, they’re trying a bit of everything. 

• Piloting onchain finance (e.g. JPMD). While this was a single transaction. It showed its possible to build a dedicated smart contract on a public network and have that link back to a deposit.

• European banks making their own stablecoin subsiduary (e.g. SocGen FORGE). EURCV is a stablecoin backed by deposits at the French Bank. This model could be followed by US Banks as they scale out their deposit token business.

• Banks are forming consortia stablecoins (EWS model). True to form, they’re trying to use the classic playbook. I think this is a bad idea. We don’t need any more consortia in banking, we have enough.

It’s a strange moment for banks.

They have to do something because the competition is in the market and moving quickly. The law is live, but the rules are not. Regulators have long memories, and today’s mistakes could be tomorrow’s fines.

Caution: The Law is Live the Rule Making is not.

A banker told me an anecdote this week.

They went to visit their state regulator to give them a friendly heads up about a new Apple Pay project. So they started out with “This is how we’re going to do Tokenization.” The regulator looked like they were having a panic attack. 

After 30 minutes of successfully talking them off the ledge, and explaining that card tokenization is very different to crypto or stablecoins it became clear some regulators are in panic mode. Later that week they visited their Fed regulator, this time they were a little more cautious in approach, but also learned the regulator is concerned about banks diving in with two feet into stablecoins.

Why is this possible when it’s law?

Because the regulator rulebook doesn’t exist yet. And the hardest thing to regulate is the thing that doesn’t yet have a checklist. But as a bank, you can be assured that checklist will be completed, and when it is, there may even be a different administration.

So any mistakes you make today could come back to haunt you. 

Banks are caught between knowing they have to make a move because their new competitors are, and regulators with long memories. 

This is why banks are taking baby steps.

Baby Steps 

You don’t move $2 trillion onto a new technology without thinking it through first.

The banks, however, have thought it through. They’ve been tokenizing deposits in their private networks for at least half a decade.

They’re now taking the first steps into true onchain finance.

There’s a lot of unknowns, but the way through that is diagram by diagram, conversation by conversation.

Because it’s land grab time, and the fintech companies, shadow banks, and merchants aren’t waiting. They’re going all out for market share.

The public steps may seem small at first, but when big banks move, they move the market.

If you’re a bank, it’s time to move from the closed loop to open loop. Get closer to the most credible onchain projects. Start with pilot programs on private L2s. Pick one use case. Build the compliance bridge. Then scale.

The next decade is going to be the biggest disruption in finance.

So welcome back from the August break.

We have work to do.

ST.

4 Fintech Companies 💸

1. RegSurety - Vanta for GENIUS Act Compliance

Regsurety helps companies operating in tokenized assets or stablecoins complete applications and submit to multiple cross-border jurisdictions through a single dashboard. AI Agents help make document prep, form filling and compliance reporting much more automated. The platform also creates proactive alerts for upcoming filing deadlines or regulatory changes. 

🧠 The founder is ex-Bitpay and Binance and has lived the pain of licensing. Many complaints about the process being “slow” can be addressed by dealing with its sheer complexity. With a gold rush for stablecoins, this is the classic shovel merchant solution.

2. Structify - The data science AI agent for finance teams

Structify helps anyone create data pipelines using natural language. Users can combine messy, disparate data sets into a single clean output, and continue to iterate on that pipeline over time to feed downstream systems or analysis. It can automate messy work like fixing format inconsistencies, removing duplicates and offers “PhD student for your use case” accuracy for enterprises. 

🧠Extract, Transform, and Load (ETL) pipelines suck. Going back to the 90s, database administrators have wrestled with multiple structured data sources and tried to make sense of them for business users. Finance is full of unstructured data like PDFs and bits of paper. Combining all of this is infuriatingly hard, but it’s also the foundation of modern data science.

3. Scalar Field - Company Research AI

Scalar Field lets traders back-test their strategies on live company data and headlines. It helps benchmark an existing, proprietary portfolio against market indices and will generate Python code to help analyze new events.

🧠 There are so many of these. What’s the moat? At first glance, this looks like Perplexity finance ++, but the video use cases show you more. The focus on back testing is a really good example, where trying out a trading strategy can be modelled in real time, using real data. This lets traders react faster.

4. Matterfact - Agentic workflows for equity analysts

Matterfact automates post-earnings workflows (like updating an Excel model), lets analysts scale deep research like “find all companies with tariff exposure,” and summarize broker reports. It includes hundreds of pre-built workflows plus the ability to create your own. 

🧠Seriously, what’s the moat? For this one, the UI pattern is more visual, and some of the imagination gap has been closed. Instead of staring at a blank chat screen, having pre-built workflows is a rapid way to improve productivity. The default to chatbox era of AI might be something we look back on as being like the CLI era of computing.

Things to know 👀

1. The CFPB’s Advanced Notice of Proposed Rulemaking on 1033 just dropped. 

That was quick. After first killing the rulemaking of the previous administration, the CFPB (after some significant pressure from the crypto lobby) has sputtered back into life. 

Focussing on 5 key questions (courtesy of Rogue CFPB) 1) what is a “representative”; 2) Can banks charge fees for data access and what are the actual costs to provide access; 3) what are the InfoSec implications; 4) what are the consumer privacy implications; 5) what is a reasonable compliance timeline.

🧠These were the exact same questions we started with a decade ago. This administration moves fast, but the lobbies and the lawsuits haven’t started yet. 

🧠 The “burden of consumers’ exercise of the rights it creates”. Sounds like we might get something on fees. The previous rule was very zero-fee focused. This is one area where the CFPB may need to weigh in, because FDX and “the market” have failed to agree on a structure that works. 

🧠When will this ever pass? The CFPB seems hopeful it can stick to the original deadlines, but there will no doubt be pushback. Delays are inevitable, I fear.

🧠I do wonder if Chase realizes they may have met their match with the crypto lobby. The thing with being the bank is sometimes you have to be the bad cop. But in hindsigh,t if they’d waited longer, increased fees by less, they might just have gotten to a new equilibrium where they get paid.

2. The CFPB is on the brink of the first-ever Fintech bailout. 

They will charge a $1 civil penalty to the now bankrupt Synapse, enabling them to access their fund for “redressing harmed consumers,” AKA, making them whole. The CFPB alleges Synapse violated the Consumer Financial Protection Act of 2010 “by failing to maintain adequate records of the location of consumers’ funds and failing to ensure those records matched the records maintained by its partnering banks.”

🧠This is a major precedent. Next time there’s some major issue with intermediaries where customers are harmed, you’d imagine this becomes something every lawyer will point to.

🧠I’m still pretty sure there’s no innocent party in this story. The amount of poor reconciliation at partners, at banks, and across the industry in the wildcat era of Fintech (2015 to 2020) was enormous. I’m frankly surprised we haven’t seen far more car crash examples in the public domain.

🧠The “novel activities” division has been closed, but the job isn’t done. This was a targeted program looking at crypto and bank/fintech partnerships. It resulted in a massive wave of regulatory actions against banks, who were frankly hit for anything the examiners could find, many of whom didn’t understand the nature of activity.

🧠The real job is to upskill regulators to understand where the material risk lives. I’m hopeful that can be done in a more open dialogue, rather than an examination posture, where examiners are looking for something bad, without knowing what to look for. That didn’t stop the risk of Synapse/Evolve. It just made good banks lives harder.

Meanwhile in geopolitics

3. China Explores Yuan-Backed Stablecoins

Beijing is weighing yuan-backed stablecoins via Hong Kong and Shanghai in a bid to boost the renminbi’s role in global finance and challenge dollar dominance.

🧠This is a fascinating signal in a market full of noise. Scott Bessant has been brazen that USD stablecoin adoption is front-door, direct, global dollarization.

🧠The elephant in the room is Tether. Hong Kong is flooded with agents who can help you move local currency anywhere in the world via tether. It’s the #1 non trading use case for stablecoins.

🧠Tether is widely used for China x Africa, or China x India trade routes. Going via SWIFT or the dollar for these corridors is slow and expensive. Tether is not, and while it’s subject to OFAC, it’s also not as controlled as the US banking system is.

🧠 China tried CBDC and it failed. This is a fascinating recognition that centralization didn’t work. I hope the European Digital Euro and Digital Sterling teams are watching and learning this lesson. (Which, according to the FT, is starting to happen in Europe at least). 

Good Reads 📚

1. Open Banking and Payments Competition

If you’ve struggled to keep up with the drama in the US open banking scene Patrick McKenzie’s opus is the only thing you need. The core point is this is not a fight over seeing your data (as it appears), it's about who controls and monetizes payments.

When Chase hiked its fees, those fees would amount to 60 to 100% of a single aggregator's revenue annually. Plaid was asked for $300 million, which would be 75% of their 2024 revenue.

🤌 is my comment. This is such a good read its delicious. I can’t do it justice.

🧠Patrick finishes encouraging the banks to innovate. He points out most Fintech nerds and VCs love Chase Sapphire. They can make great products when they compete. 

Tweets of the week 🕊

That's all, folks. 👋

Remember, if you're enjoying this content, please do tell all your fintech friends to check it out and hit the subscribe button :)

Want more? I also run the Tokenized podcast and newsletter.

(1) All content and views expressed here are the authors' personal opinions and do not reflect the views of any of their employers or employees.

(2) All companies or assets mentioned by the author in which the author has a personal and/or financial interest are denoted with a *. None of the above constitutes investment advice, and you should seek independent advice before making any investment decisions.

(3) Any companies mentioned are top of mind and used for illustrative purposes only.

(4) A team of researchers has not rigorously fact-checked this. Please don't take it as gospel—strong opinions weakly held

(5) Citations may be missing, and I’ve done my best to cite, but I will always aim to update and correct the live version where possible. If I cited you and got the referencing wrong, please reach out