Why is the growth of non-USD stablecoins stagnating?

Author: Techub Selected Compilation

In this article, I will analyze in depth: how banking regulation has become the main driver of liquidity shortages in non-USD channels. My core point is: the structural failure of the global foreign exchange market outside the G7 has led to a “liquidity vacuum” in non-USD cross-border settlements. Therefore, to solve the liquidity issues of non-USD stablecoins, a DeFi-native approach to liquidity bootstrapping must be adopted. Any on-chain FX solution relying on traditional foreign exchange markets is doomed to fail.

1. The Basel III Hammer (Extended Version): LCR is a “Currency + Entity” Constraint, Not Just a “Liquid Asset” Rule

What is Basel III? Basel III is an international regulatory framework aimed at making the banking system more resilient by increasing capital adequacy, liquidity, and risk management capabilities. Its measures include: requiring higher-quality capital buffers, introducing leverage limits, and setting global liquidity standards (LCR and NSFR) to prevent excessive risk-taking and reduce the likelihood of future banking crises.

How does Basel III turn FX market making into a multi-constraint optimization problem? The main tool is the Liquidity Coverage Ratio (LCR). LCR requires banks to hold enough “High-Quality Liquid Assets” (HQLA) to survive a 30-day stress scenario. On the surface, this is a prudential safety rule; in practice, it becomes a regulatory barrier to non-USD market making.

1.1 The Core Bias of LCR: HQLA Definition Based on Market Depth Under Stress, Not “What People Actually Use”

LCR mandates banks to hold sufficient, unpledged HQLA that can be “easily and immediately” converted into cash in private markets to withstand a 30-day stress scenario.

The key is that the definition of Level 1 HQLA is very strict, mainly including:

• Cash (coins and banknotes),
• Central bank reserves (limited to the portion that the central bank allows to be used under stress),
• And eligible, tradable sovereign bonds/central bank bonds/public sector bonds/multilateral development bank bonds that meet criteria such as 0% standardized credit risk weight and can be traded in markets that are “large, deep, and active” even under stress.

This means that reserve currencies’ “safe asset systems” (like USD, JPY, EUR) are structurally favored because they have:

• Deep repo and cash markets worldwide,
• Reliable liquidity even during stress,
• And fewer convertibility and transfer restrictions.

Since only reserve currencies’ “safe asset systems” can reliably meet the strict HQLA definition, banks are incentivized to allocate their liquidity buffers mainly in USD assets. If a bank wants to trade BRL or MXN, it must hold inventories of these currencies, which are often considered “substandard assets” from the LCR perspective, forcing banks to hold additional USD assets as compensation.

1.2 “Single Currency LCR” Is a Myth; Regulators Will Watch Currency Mismatch

Basel documents explicitly state that LCR is satisfied at the “single currency” level, but simultaneously, banks and regulators should monitor the LCR of important currencies to capture currency mismatch risks. When a currency liability exceeds or equals 5% of total liabilities, that currency is considered an “important currency.”

This has two direct impacts on emerging market channels.

First, under stress scenarios, it cannot be assumed that FX can be smoothly converted. Regulators will assess banks’ ability to fund in FX markets and transfer liquidity across currencies, jurisdictions, and legal entities.

Second, even if the group-level LCR is compliant, your BRL/MXN trading desk may be viewed as a source of currency mismatch, requiring additional governance, limits, and buffers.

This is where “non-USD stablecoin demand” hits a high wall: demand exists in accounting units, but prudential liquidity requirements are based on assets deemed “credible” under stress scenarios.

If a trading desk wants to run BRL/MXN positions, it will create corresponding currency liabilities. Regulators may require the bank to hold matching liquidity assets in those currencies. But since BRL and MXN assets are not Level 1 HQLA, the trading desk faces a “governance tax”—requiring higher limits, thicker buffers, and stricter scrutiny compared to USD/EUR trades.

1.3 Critical Detail: Liquidity Transfer Restrictions Turn HQLA into “Trapped Assets”

In consolidated LCR calculations, if there is reasonable doubt about the availability of liquidity, excess liquidity should not be counted. Transfer restrictions—such as asset segregation, non-convertibility, foreign exchange controls—reduce the ability to mobilize HQLA and funds within the group. If HQLA exceeds local outflows but cannot be transferred, that surplus must be excluded from the consolidated metric.

For example, if a global bank holds large cash reserves in its Brazilian entity but, due to capital controls, asset segregation rules, or operational frictions, cannot immediately transfer that cash to London:

1. This cash cannot be included in the group-level liquidity buffer;
2. The London trading desk cannot rely on the Brazilian entity’s inventory.

Why is this critical for non-USD channels? To support a particular channel, banks often need to establish local entities, local accounts, and local liquidity. If this liquidity is deemed “trapped” (legally, operationally, or politically), banks must assume it is unavailable elsewhere. This forces a “dual funding” approach—holding trapped liquidity locally and maintaining redundant liquidity centrally—making that channel structurally more “expensive” than a USD-centric system.

1.4 HQLA Is Not “What You Can Sell,” But “What You Can Liquidate Without Breaking Solvency”

Basel points out that in some jurisdictions, there are not enough large, deep repo markets, so HQLA liquidation may depend on direct sales. In such cases, banks should exclude assets with sale barriers (e.g., fire-sale discounts eroding capital adequacy, or assets with statutory minimum holdings).

Thus, even if a sovereign bond in an emerging market looks “liquid” on a normal Tuesday, if selling it during panic would undermine the bank’s capital ratio, it may be excluded from the liquidity buffer. The trading desk will be told: “If it cannot be safely liquidated under stress, it is not part of the liquidity buffer.” This further concentrates liquidity buffers in reserve currencies.

1.5 Official Recognition: Alternative Liquidity Arrangements (ALA)

The Basel framework explicitly recognizes that some currencies lack sufficient HQLA to support their banking systems. It provides “Alternative Liquidity Arrangements” (ALA) for jurisdictions with insufficient HQLA. One pathway allows banks to use foreign currency HQLA to cover domestic currency gaps, but with discounts (at least 8% for major currencies, higher for others) and strict FX risk management requirements. This indicates two points:

• Regulators understand that “some currencies inherently lack enough HQLA capacity”;
• The official workaround is “use foreign currency HQLA but pay a price for mismatch.”

This is essentially a regulatory acknowledgment of the structural liquidity limitations of non-USD currencies. The approved solution is: “Use USD, but we will charge you for currency mismatch.” At the institutional level, this establishes USD as the ultimate backing for emerging market trading.

In summary, Basel III’s LCR imposes a clear capital tax on non-USD inventories:

• Non-USD inventories occupy balance sheet space and generate liquidity outflows (margin calls, settlement funding, stress assumptions), all requiring funding.
• LCR incentivizes banks to use HQLA that have proven liquidity under stress, which are highly concentrated in reserve currencies’ safe assets.
• When liquidity is “trapped” within entities or currencies due to legal or operational barriers, banks cannot include it centrally, increasing the “tax” on maintaining non-USD channels.

The entire system favors a “Hub-and-Spoke” model. Banks engaging in triangular arbitrage via USD are swimming along the physical laws of regulation; those attempting to directly build BRL/MXN bridges are swimming upstream.

2. Basel Market Risk / FRTB: Capital Rules Favor Currency Pairs with Strong Liquidity

Even if banks ignore inventory issues (LCR), they face a second equally stringent constraint: market risk capital.

Under Basel, especially in the Fundamental Review of the Trading Book (FRTB), banks must hold sufficient capital to absorb potential losses from market fluctuations. The framework does not treat all currencies equally but explicitly penalizes “thin markets,” creating a regulatory feedback loop that makes non-USD channels extremely capital-inefficient.

2.1 Liquidity Horizon: The “Time Tax” on Emerging Markets

Capital requirements are based on risk models (e.g., expected shortfall), which simulate potential losses during market crashes. A key input is the liquidity horizon—the assumed time needed for a bank to sell a position without triggering a price collapse.

Under the FRTB market risk framework, internal models embed liquidity horizons into risk factors. The longer the assumed holding period, the larger the potential loss and the higher the capital buffer.

If a channel does not belong to the “Designated Currency Pairs” (or primary cross-currency pairs), the framework assumes a 20-day liquidity horizon, meaning it takes twice as long to exit risk during stress compared to major currency pairs, and thus requires higher capital.

• 10 days: applies to “FX: Designated Currency Pairs,” i.e., highly liquid pairs like USD/EUR, USD/JPY.
• 20 days: applies to “FX: Other Currencies.”

Therefore, when calculating risk for BRL/MXN, the model assumes the bank needs 20 days to exit risk during a crisis, while for major pairs only 10 days. This creates a capital surcharge on non-USD inventories, making holding USD/EUR risk cheaper structurally than holding BRL/MXN risk.

2.2 Regulators Can Be More Stringent: This Is a Floor, Not a Ceiling

More importantly, 20 days is just a minimum. The FRTB explicitly allows regulators to increase the liquidity horizon based on the trading desk’s circumstances. If a regulator perceives high risk or poor liquidity, it can require even longer horizons.

This introduces regulatory uncertainty. If a bank tries to market-make in a volatile channel (e.g., NGN/ZAR), regulators can arbitrarily extend the liquidity horizon to 40 or 60 days. In other words, the more “niche emerging market risk” a channel has, the more likely it is to be assigned a longer horizon, requiring higher capital.

This uncertainty imposes an unquantifiable risk on capital planning, incentivizing banks to participate only in “designated currency pairs” with predictable, capped horizons (e.g., 10 days).

2.3 Non-Modelable Risk Factors (NMRF)

Banks prefer using internal models for capital efficiency, but to do so, they must prove the market is “genuinely existing,” via the Risk Factor Eligibility Test (RFET). To pass RFET, a risk factor (e.g., BRL/MXN rate) needs at least 24 actual price observations per year (roughly twice a month), with no gap exceeding one month between observations.

If a channel is sparse in trading, it cannot pass RFET and is classified as an NMRF. This is catastrophic at the capital level because once designated as NMRF, banks cannot use more capital-efficient models and must rely on stress scenarios.

NMRF penalties are severe. The framework states that for each NMRF, the stress liquidity horizon must be the larger of the designated horizon or 20 days.

This creates a vicious cycle for non-USD channels:

1. Low trading volume, infrequent trades;
2. Cannot pass RFET, classified as NMRF;
3. Capital requirements skyrocket, making trading unprofitable;
4. Banks withdraw, stop quoting to reduce capital costs and maintain higher leverage;
5. Liquidity further deteriorates.

This mechanism effectively kills banks’ presence in emerging market channels: before a channel is sufficiently liquid, banks cannot afford to make markets in it.