By Eduardo Levy Yeyati, Professor of Economics and Public Policy, Universidad Torcuato Di Tella; Regional Adviser, Inter-American Development Bank; and Non-Resident Senior Fellow, Brookings Institution
Following the passage of the GENIUS Act (Guiding and Establishing National Innovation for U.S. Stablecoins Act) last summer, US policymakers declared victory on stablecoin regulation. The framework mandates full reserve backing—cash, U.S. Treasury securities and/or insured deposits—and treats stablecoins like digital currency boards.
But this regulatory optimism ignores what Argentina’s convertibilidad taught us in 2001: You can regulate the issuer, but you cannot control what the ecosystem does with the issued money.
The Argentina lesson
Argentina’s central bank maintained a strict 1:1 backing of base money with U.S. dollar reserves—a perfect currency board. Banks and depositors trusted the peso because they believed it was “as good as dollars”.
While the central bank limited narrow money (M1) to its reserves, commercial banks created dollar-denominated broad money far exceeding those reserves through fractional-reserve lending. When capital flight began in 2001, the reserves backing base money proved insufficient to meet the total dollar-denominated claims in the system. The currency board collapsed not because the central bank violated the 1:1 rule, but because secondary-money creation broke the credibility of the peg.
Current stablecoin regulation resembles that regime. The GENIUS Act sets strict rules for issuance but provides no framework for managing claims denominated in stablecoins once issued. It addresses what Circle or Tether does with reserves, but not what the ecosystem does with those instruments.
A few years back, we called this the stablecoin paradox: To preserve stability through currency-board-type backing, stablecoins must limit leverage and secondary-money creation—but their use in financial intermediation, whether in DeFi (decentralized finance) protocols or emerging commercial credit, inherently generates that secondary creation.
Each layer of intermediation creates claims on the underlying reserves exceeding 1:1 backing, eroding the pretense of stability.
The US bet: Regulating the issuer
The GENIUS Act requires 100-percent backing in liquid, safe assets. U.S. Secretary of the Treasury Scott Bessent has projected that the stablecoin market could reach $2 trillion by 2028. Stablecoin issuers already held approximately $130 billion in Treasury bills (T-bills) as of mid-2025—with Tether alone holding more than $100 billion, more than Germany holds in U.S. Treasuries.
This framework mandates monthly attestations and annual audits. Anti-circumvention provisions prohibit stablecoin issuers from making loans or using reserves for anything beyond the permitted asset list.
Its enforcement is not trivial: Market concentration works against transparency. Tether, with approximately 60-percent market share, operates under looser disclosure than federally regulated rivals. This dual federal-state oversight enables regulatory arbitrage toward lenient jurisdictions. Internationally, Hong Kong’s rules differ from Singapore’s, the European Union’s (EU’s) MiCA (Markets in Crypto-Assets Regulation) regime and the United Kingdom’s standards. Fragmentation invites a race to the bottom on backing and disclosure.
But even if enforcement were perfect, it would not address the other half of the stablecoin-risk challenge.
The invisible risk
There is another risk, less documented and possibly more damning—the one hinted at by the currency-board analogy. The GENIUS Act regulates what stablecoin issuers must do with reserves. It says nothing about what the ecosystem does with stablecoins once issued. This matters because stablecoins are no longer just payment instruments—they have evolved into financial infrastructure, particularly for lending.
DeFi lending protocols such as Aave, Compound and Morpho now hold more than $26 billion in stablecoins deposited by lenders who earn 5-to-7-percent annual yields. Borrowers take loans denominated in USDC (Circle) or USDT (Tether) and post crypto collateral. USDC and USDT account for about 99 percent of these platforms´ lending volume.
This is secondary-money creation denominated in a supposedly fully backed currency—precisely what happened in Argentina. While Circle and Tether maintain strict 1:1 reserve backing, DeFi protocols create credit claims exceeding those reserves. A merchant holding USDC can deposit it in Aave and earn yield, with those same dollars lent to someone else. Both parties now have claims on the same underlying reserves.
The GENIUS Act regulates stablecoin issuers, but it does not regulate the parallel credit system built on top of those stablecoins.
The GENIUS Act regulates stablecoin issuers, but it does not regulate the parallel credit system built on top of those stablecoins. Just as Argentine banks created peso-denominated deposits far exceeding the central bank’s dollar reserves, DeFi protocols are creating dollar-denominated claims far exceeding stablecoin issuers’ reserves. The currency board works at the issuer level but fails at the system level.
Documented cases of formal stablecoin-denominated credit remain limited. But history suggests intermediation does not begin with loans—it begins with balances, maturities and deferred payments. Once working capital flows through stablecoin rails, progression to commercial credit becomes structurally inevitable.
Regulatory frameworks remain focused on the traditional banking perimeter. Reserve requirements, capital adequacy ratios and deposit insurance apply to banks but not to the emerging stablecoin-based financial system. Central banks’ liquidity facilities backstop banks, but not the payment platforms operating with stablecoins. The GENIUS Act regulates stablecoin issuers; it does not govern the parallel financial system being erected on those stablecoins.
Perhaps aware of the underlying risks, China is resisting private stablecoins despite the ongoing digital-currency competition—and domestic pressure. JD.com and Ant Group have lobbied the People’s Bank of China (PBOC) to allow yuan-pegged stablecoins. Former Bank of China (BOC) Vice President Wang Yongli warned that it would be a “strategic risk” if cross-border yuan payments cannot match US-dollar stablecoin efficiency.
However, the PBOC remains focused on its e-CNY, a central bank digital currency (CBDC) that avoids private intermediation entirely. China’s resistance suggests its recognition of the design problem the United States is attempting to solve through regulation. A CBDC is central bank money; a stablecoin is a private claim on reserves. This distinction matters precisely because of the intermediation risk.
The strategic question
For banking executives, the question is direct: What are you competing against?
If stablecoins remain pure payment instruments—narrow banking rails with full reserve backing—they represent efficient competition for payment services with limited systemic risk. Banks can compete by offering higher interest rates on deposits or issuing their own fully backed stablecoins.
If stablecoins evolve into intermediation vehicles through ecosystem-credit creation, banks face competition from leveraged entities operating outside prudential oversight. Traditional tools of monetary policy and prudential supervision stop working when they cannot reach the institutions that are creating monetary claims.
Market concentration amplifies these risks. If a major issuer failed—through a loss of peg, a run or a reserve impairment—spillovers could hit short-term funding markets through a disorderly liquidation of Treasuries. TerraUSD’s collapse in 2022 wiped $40 billion from its own ecosystem and triggered broader contagion across crypto markets, demonstrating how quickly confidence unravels when currency-board discipline is breached.
Pattern recognition
The US is running an experiment at scale: Can strict reserve requirements contain the paradox when the ecosystem, not the issuer, creates the secondary money? China’s resistance to private stablecoins suggests skepticism about the answer.
Regulatory frameworks remain focused on the traditional banking perimeter, with reserve requirements, capital adequacy ratios and deposit insurance applied to banks but not to DeFi lending protocols. Central banks’ liquidity facilities support banks, but not the parallel stablecoin credit system. The GENIUS Act has solved half the problem—issuer discipline—while ignoring the half that matters more.
For banks, the choice is stark. Compete with stablecoins as payment rails or compete with them as shadow banks. The former is manageable. The latter recreates Argentina’s currency board at a digital scale—across multiple jurisdictions and largely invisible to regulators.
Currency boards repeatedly failed under pressure to expand credit beyond the backing of reserves, as monetary fundamentals would have anticipated. The digital age offers new tools. The question is whether we will recognize the pattern before the system scales beyond containment.
Disclaimer
The views expressed in this article are solely those of the author and do not represent the views of the institutions with which he is affiliated.
Dr. Eduardo Levy Yeyati is Full Professor of Economics and Public Policy at the School of Government, Universidad Torcuato Di Tella, and Regional Adviser at the Inter-American Development Bank. A former Chief Economist at the Central Bank of Argentina, he is also a Non-Resident Senior Fellow in the Global Economy and Development Program at Brookings Institution and Principal Researcher at CONICET.