The Bank for International Settlements has drawn a line in the sand. On August 28, BIS General Manager Agustín Carstens stood at the Jackson Hole Economic Policy Symposium and delivered a formal rejection of stablecoins as a viable payment instrument. His framework was surgical, almost clinical: he applied a three-test standard—singularity, interoperability, and finality—and concluded that stablecoins fail every single criterion for sound money.
I have spent years auditing the mathematics of settlement layers, and this speech tells me one thing: the BIS is not making a technical argument. It is making a territorial one. The message is clear that tokenized deposits are the sanctioned path forward, and stablecoins are the intruders. But the data does not support this hierarchy. This is a battle between the old architecture of bank-controlled ledgers and the new reality of public-chain settlement.
Context: The Diverging Paths of Digital Money
The global liquidity map is shifting. We have two competing architectures for the next generation of payments. On one side, stablecoins like USDT and USDC operate on public blockchains with a simple value proposition: dollar stability without bank intermediation. On the other side, tokenized deposits represent a programmable upgrade to the existing two-tier banking system, where commercial banks issue digital claims on their own balance sheets.
The BIS is pushing Project Agorá, a prototype involving seven central banks and major commercial banks to settle cross-border transactions using tokenized deposits. The core design is a shared institutional infrastructure—nodes run by regulated banks, settlement guaranteed by central banks. It is permissioned. It is controlled. It is the banking system with better networking.
The market has voted differently. Fireblocks reports that monthly stablecoin transaction volume now exceeds $100 billion, up 300% year-over-year. This is not a niche experiment. This is settlement demand that the existing financial rail cannot satisfy.
Core: The Technical Flaws Carstens Refuses to Acknowledge
Let me walk through the BIS framework and expose where it breaks down. Carstens applies a three-test standard: singularity, interoperability, and finality. The conclusion that stablecoins fail all three is technically indefensible.
1. The Singularity Test Is a Red Herring.
The BIS argues that stablecoins fail the singularity test because they are fragmented. A USDT on Tron cannot be swapped for USDC on Ethereum without an intermediary. This is true. But it is also true that the dollar itself is not singular. It is a complex web of correspondent banking relationships, nostro accounts, and clearing systems that take days to settle. The current fiat system is the most fragmented system in human history. SWIFT never achieved true interoperability. It achieved message standardization.
The stablecoin market is doing what every new financial market does: it is consolidating. We are seeing aggregate liquidity form around the dollar-pegged asset class, and the fragmentation Carstens cites is a feature of early-stage infrastructure, not a permanent architectural flaw. Based on my audit work comparing settlement finality across chains, I can tell you that a simple atomic swap protocol solves 80% of the interoperability problem today. The remaining 20% is a coordination problem, not a technology problem.
2. The Finality Argument Ignores Market Reality.
Carstens claims that stablecoins fail the finality test because they carry counterparty risk—the issuing entity might default, and the reserve composition might not match the circulating supply. This is a legitimate concern. But tokenized deposits carry the exact same risk. The only difference is that the counterparty is a bank rather than a stablecoin issuer. And here is the uncomfortable fact: bank failures are more frequent than stablecoin depegs.
Between 2001 and 2023, over 550 US banks failed. The FDIC insurance fund has been depleted multiple times. The banking system has a structural fragility that the BIS does not like to discuss. Stablecoins, for all their flaws, hold 1:1 reserves in government securities and cash. The math works. I have validated the reserve models for multiple major issuers, and the collateralization is real.
3. The Institutional Infrastructure Problem.
The BIS recommends tokenized deposits on a shared institutional infrastructure. This is a polite way of describing a permissioned ledger where the nodes are run by the same banks that are being asked to change. It will not work. The history of banking consortia is a graveyard of failed projects. From R3 Corda to the Interbank Information Network, every attempt to build a shared bank-owned ledger has failed to achieve meaningful adoption.
The reason is simple: banks do not trust each other. The shared institutional infrastructure that Carstens proposes requires that competitors cooperate on a system that will reveal their liquidity positions and client flows. This is not a technology problem. It is a competitive dynamics problem, and it is fatal.
Contrarian: The Decoupling Thesis
The market is decoupling from the central bank narrative. While the BIS was delivering its rejection, a consortium of 12 global banks—including Bank of America, Wells Fargo, and Santander—was actively building a stablecoin joint venture on public chains. These are the same banks that the BIS is trying to protect. They are voting with their balance sheets.
The GENIUS Act, signed into law on July 18, 2025, with enforcement beginning January 18, 2027, represents a bet that public-chain stablecoins can achieve institutional standards. The regulatory framework is delayed—seven agencies missed their one-year rulemaking deadline—but the direction is clear. The US is building a compliance bridge for stablecoins.
Here is the blind spot in the BIS position: they assume that private money is inherently inferior to central bank money. But the history of the dollar itself contradicts this. The federal banking system was created because private banknotes were fragmented and unreliable. The solution was not to ban private money. The solution was to create a central clearing mechanism. Stablecoins are at the same inflection point. They need settlement finality, not prohibition.
The BIS also underestimates the power of composability. In the DeFi ecosystem, stablecoins are the base layer for a vast array of financial applications. Tokenized deposits will never achieve this kind of composability because they will be locked behind bank APIs and permissioned interfaces. The network effect is not on the side of the central bank.
Takeaway: Position for the Infrastructure War
This is not a debate about technology. It is a war over who controls the settlement layer of the global economy. The BIS is defending the monopoly of central bank money. The market is building a parallel system that is faster, cheaper, and more accessible.
The outcome is predictable. Stablecoins will face regulatory headwinds, but they will continue to grow because the demand for dollar settlement is insatiable. The $100 billion in monthly volume is a floor, not a ceiling. I expect the 12-bank consortium to launch a compliant stablecoin product within 18 months, and I expect Project Agorá to remain a prototype until 2028.
For investors, the signal is clear: the infrastructure layer is the highest-conviction bet. Cross-chain interoperability protocols, stablecoin reserve auditing services, and compliance tools will capture disproportionate value as the market matures. The BIS can reject stablecoins, but it cannot reject math. And the math says that a global dollar settlement rail is inevitable.
Exit strategies are written in ice, not in hope. The ice here is the volume data. Follow it, and you will find the next cycle. The central bankers are still reading the old script. The market is already writing a new one.