Stablecoins became infrastructure

Stablecoin supply passed $321 billion in April 2026, an all-time high reached while the asset class it settles was in the middle of a 52% decline. That divergence is the clearest evidence available that dollar tokens have become infrastructure rather than a speculative instrument.

For most of their history, stablecoin balances tracked interest in trading. Supply expanded when markets rose and contracted when they fell, because the dominant use was holding trading collateral between positions. On that model, an eight-month drawdown of the size seen in 2026 should have shrunk the float substantially.

It grew instead. Total supply stood at roughly $269–270 billion in January 2026 and exceeded $321 billion by April.^1,2

Two things stand out. Supply grew by roughly $52 billion in four months during a severe price decline, and the growth was disproportionately Circle's — USDC expanded around 22% while Tether was close to flat, reducing top-two concentration from about 92% to 83%.

Why the divergence indicates use rather than speculation

Balances that exist to trade shrink when trading interest falls. Balances that exist to move money do not. The straightforward reading of 2026 is that a growing share of stablecoin float now serves payments, corporate treasury operations and cross-border settlement, which are activities indifferent to the price of bitcoin.

Supporting evidence sits in the settlement layer: Tron remains the primary network for offshore dollar settlement, a role driven by low transfer costs rather than by any characteristic that would attract speculative capital.¹ Networks selected on the basis of transaction cost are networks being used for transactions.

A settlement layer that grows while the asset class it settles halves has stopped being a derivative of that asset class.

The regulatory framework arriving

The GENIUS Act carries 2026 deadlines for stablecoin licensing, capital, custody and anti-money-laundering standards in the United States, and that framework has been progressing independently of the broader market structure legislation that remains stalled in the Senate.³ In the European Union, MiCA's transitional period ended on 1 July 2026, bringing stablecoin issuance under a completed authorisation regime across the single market.⁴

Defined capital and reserve standards change the credit character of these instruments. A dollar token is a claim on an issuer, and the analysis of that claim — what backs it, who holds the reserves, on what terms it can be redeemed — has historically depended on voluntary attestation of varying quality. A statutory framework substitutes a supervised standard for that, which is the precondition for treasury departments at conventional institutions holding these balances at scale.

The unresolved question of yield

The most consequential open issue is whether platforms may pay rewards on stablecoin balances. Banks oppose it; Coinbase earns approximately $1.35 billion annually from USDC rewards, and the question is one of three unresolved disputes holding up US market structure legislation.³

The economics are straightforward. Reserves backing a major stablecoin sit largely in short-dated government paper earning the prevailing rate. Whoever is permitted to pass that yield to holders captures the distribution. If platforms may pay it, stablecoin balances compete directly with bank deposits; if they may not, the yield accrues to issuers and distributors, and holders have an incentive to move into tokenized treasury products instead.

That second path is already visible. Tokenized US treasuries reached $15 billion by June 2026 within a $32.22 billion tokenized real-world asset market.⁵ Those instruments do what a stablecoin does and pay a yield while doing it, which makes them the natural destination for balances that do not need instant settlement.

Concentration remains the structural risk

Two issuers accounted for roughly 83% of supply as of April 2026, and the largest for around 58%.² The reserve quality question that dominated discussion of this sector for years has been substantially addressed by disclosure and by the incoming statutory regimes; the concentration question has not.

A settlement layer carrying more than $320 billion, where a majority of the float is a claim on a single issuer, has a single point of failure regardless of how well that issuer is run. Growth in Circle's share through 2026 marginally reduces the concentration, and competition from tokenized funds and from bank-issued alternatives under MiCA would reduce it further, but the structure remains materially more concentrated than the payments infrastructure it increasingly resembles.

The broader observation is that stablecoins have quietly become the part of digital assets with the clearest product-market fit, the largest institutional participation and the most advanced regulatory treatment — and they achieved that position by being useful for something other than trading.