Stablecoin supply contracted $13.8 billion (−4.3%) over the 90 days ending September 10, 2026, settling at $302.8 billion — 6.1% below the May 17 all-time high of $322.4 billion. It is the first sustained drawdown since the 2022 bear market. At the same time, transaction volume hit $28 trillion i...
"The increased velocity is inconsistent with the bank's previous long-term forecast" — Geoff Kendrick, Global Head of Digital Assets Research, Standard Chartered
Stablecoin supply contracted $13.8 billion (−4.3%) over the 90 days ending September 10, 2026, settling at $302.8 billion — 6.1% below the May 17 all-time high of $322.4 billion. It is the first sustained drawdown since the 2022 bear market. At the same time, transaction volume hit $28 trillion in Q1 2026 alone, a 51% quarter-over-quarter increase and a record. Settlement volumes in February and March surpassed the ACH network at $7.2 trillion and $7.5 trillion respectively.
The divergence — shrinking supply, record throughput — points to a structural shift in how stablecoins function. Velocity, the rate at which each dollar-equivalent token changes hands, has doubled from 2.6x to 6.0x monthly over two years. According to Visa's Onchain Analytics, stablecoin velocity reached 13.56 per quarter in Q4 2025, compared to 1.65 for U.S. M1 money supply. The implication is clear: fewer coins are doing more work. Reading the supply contraction as weakness mistakes a maturing payment instrument for a shrinking asset class.
This report examines the velocity divergence, its drivers, how it reshapes the supply-demand calculus for stablecoin issuers, and what it means for the $302.8 billion market.
The numbers are unambiguous. Total stablecoin market capitalization peaked at $322.4 billion on May 17, 2026, according to data aggregated by StablecoinBeat. By September 10, it stood at $302.8 billion. The 30-day net flow as of that date was −$3.23 billion (−1.04%); the 90-day figure was −$13.79 billion (−4.28%).
USDT absorbed the bulk of the contraction. Tether's supply declined roughly $3 billion in Q1 2026 alone, pushing its dominance to 58%, down from approximately 61% at the start of the year. USDC moved in the opposite direction, adding $2 billion in Q1 and reaching approximately $78 billion. The net effect: concentration is loosening. USDT and USDC still command approximately 82% of total supply, but the gap between them has narrowed.
By chain, Ethereum holds 49% of stablecoin supply and Tron 31%, together accounting for roughly 79%. USDT saw net outflows of over $7 billion from Ethereum in Q1 2026. The rebalancing suggests institutional capital rotating toward chains with deeper compliance infrastructure, while USDT's Tron-heavy footprint — dominant in emerging-market remittances — remains stable but low-turnover.
Year-over-year, supply is still up 14.3%, from $269.4 billion in August 2025 to $308.0 billion in August 2026. The contraction is recent and concentrated in mid-2026.
Q1 2026 stablecoin transaction volume reached $28 trillion, a 51% increase over the prior quarter. By comparison, total 2025 transaction volume exceeded $33 trillion for the full year, according to Artemis data. The Q1 2026 figure alone approaches the entire previous year's total.
Stablecoins' share of all crypto trading volume reached 75% in Q1 2026 — a record. Adjusted real transfers (excluding internal wallet shuffling and exchange movements) exceeded $21.5 trillion. Settlement volumes in February ($7.2 trillion) and March ($7.5 trillion) each surpassed the Automated Clearing House network's $6.8 trillion monthly throughput, per Federal Reserve data.
USDC accounted for 80% of organic volume for the first time since 2019, per the Q1 2026 Stablecoin Insider report. USDC organic volume grew 59% quarter-over-quarter. In the first half of 2026, USDC carried approximately 70% of adjusted volume versus USDT's 25%.
One caveat: bot-driven volume constituted 76% of all stablecoin transactions in Q1 2026. On Ethereum, bot activity reached 72%. This does not invalidate the velocity thesis — automated market making and arbitrage are legitimate economic activity — but it contextualizes the raw throughput figures.
Velocity measures how many times a unit of currency changes hands over a period. For stablecoins, this metric has become the critical indicator that supply alone cannot capture.
Visa's Economic Empowerment Institute, in an April 2026 analysis by economist Ezechiel Copic, measured total stablecoin velocity at 13.56 per quarter in Q4 2025, versus 1.65 for U.S. M1. Retail stablecoin velocity (transactions of $250 or less) was 0.08 — less than 1% of total activity. The research concluded that stablecoin velocity "largely captures financial system activity" rather than consumer spending.
When benchmarked against Fedwire (velocity: 93.84 per quarter), stablecoins remain a smaller-scale settlement system. But the growth trajectory is steep.
Standard Chartered's Geoff Kendrick identified stablecoin turnover running at approximately six times per month, roughly double two years ago. The bank's digital assets research team flagged that this acceleration is "inconsistent with the bank's previous long-term forecast" of a $2 trillion stablecoin market by late 2028, because the original projection assumed steady velocity. Higher turnover means fewer coins can facilitate equivalent transaction volumes — a deflationary pressure on supply growth.
The velocity divergence is not uniform across stablecoins. The market has bifurcated.
USDC, representing approximately 25% of supply, is the primary driver of velocity increases. Its sharp acceleration began around October 2025, particularly on Solana and Base, driven partly by experimental AI agent payment protocols using open-source infrastructure. In 2025, USDC moved $18.3 trillion versus USDT's $13.3 trillion, despite running on two-fifths the supply. By June 2026, the monthly split was $1.21 trillion (USDC) to $576 billion (USDT).
USDT velocity has remained comparatively stable. Its dominance in emerging markets — where stablecoins function primarily as savings instruments and remittance vehicles rather than high-frequency settlement tools — produces lower turnover by design. Tron, which hosts the majority of USDT supply, saw 54% bot activity versus Ethereum's 72%.
This bifurcation has consequences. USDC is evolving into a high-velocity institutional settlement token. USDT remains a store-of-value and transfer medium in jurisdictions with weak local currencies. Both are legitimate use cases, but they produce fundamentally different velocity profiles.
BCG and Allium's January 2026 white paper, "Stablecoin Payments: The Truth Behind the Numbers," provided the most rigorous decomposition of stablecoin activity to date.
Of the $28–62 trillion in stablecoin transfers during 2025 (the range reflects methodological differences), approximately $350–550 billion constituted genuine real-economy payments — bilateral transactions for goods and services between economically distinct parties. That figure grew approximately 60% year-over-year. Tron dominated real-payment rails at $235–375 billion, with BNB Smart Chain ($35–50 billion), Ethereum ($20–35 billion), Solana ($20–35 billion), and Polygon ($8–10 billion) capturing incremental growth.
Stablecoins' share of global payment flows remained at approximately 1%, unchanged from 2023 and 2024 despite absolute growth. The number suggests that while the payment use case is expanding, it has not yet reached inflection in a $42 trillion global payments market.
Consumer-to-business transactions hit 284.6 million in 2025, up 128% year-over-year from 124.9 million in 2024, according to a16z crypto data.
The velocity acceleration coincides with a rapid buildout of institutional stablecoin infrastructure.
Visa's stablecoin settlement volume reached a $20 billion annualized run rate by September 2026, up more than 15x year-over-year. More than 160 stablecoin-linked card programs operate on its network, with payment volume growing nearly 200% year-over-year. The network expanded from four blockchains to nine in April 2026, adding Base, Polygon, Canton Network, Arc (Circle), and Tempo (Stripe). In September 2026, Visa began sharing VisaNet settlement data with blockchain-based lenders to support stablecoin card credit infrastructure.
Stripe, which acquired stablecoin infrastructure firm Bridge for $1.1 billion in late 2024, reported $9.9 billion in Stripe Treasury transaction volume in the first half of 2026, already exceeding its 2025 full-year total.
Stripe, Visa, and Mastercard are reportedly nearing launch of a joint stablecoin platform, with Coinbase evaluating participation. Mastercard acquired stablecoin firm BVNK and announced expanded always-on stablecoin settlement.
Rain-powered stablecoin card programs grew from near zero in November 2024 to over $300 million in monthly collateral deposits by early 2026.
These are not pilot programs. The throughput levels and partnership structures indicate infrastructure deployment at scale.
The GENIUS Act, signed into law on July 18, 2025, with bipartisan support (Senate 68-30, House 308-122), is entering its implementation phase. The U.S. Treasury issued a Notice of Proposed Rulemaking on August 17, 2026, to implement Section 3 of the Act, with a comment period open through October 19, 2026.
Beginning January 18, 2027, stablecoin issuance in the United States will require either a federal or state license. All regimes mandate full reserve backing, redemption at par, and prohibit interest payments to holders.
The law's economic impact has been immediate. According to industry estimates cited by multiple analysts, the GENIUS Act reduced the market value of incumbent payment firms by approximately $300 billion (roughly 18%), with cross-border payment companies suffering the steepest declines at −27%.
Stablecoin issuers have emerged as significant holders of U.S. Treasuries. Tether alone holds approximately $141 billion in Treasury exposure as of Q1 2026, ranking it as the 17th largest global Treasury holder.
Standard Chartered's $2 trillion stablecoin market projection for late 2028 was predicated on steady velocity. The doubling of velocity over two years undermines that assumption. If the same transaction throughput can be achieved with fewer coins, the supply ceiling is structurally lower.
Citi's projection of $2–4 trillion by 2030 faces the same headwind. Financial institutions surveyed by EY-Parthenon expect stablecoins to capture 5–10% of global payments by 2030 ($2.1–4.2 trillion), but the velocity dynamic suggests this volume target may be achievable at a supply level well below $2 trillion.
Yield-bearing stablecoins — which expanded 22% in Q1 2026, contributing over 50% of net sector supply growth — add a second variable. Products like sUSDS (which attracted over $2.5 billion in Q1 inflows) and USDY (up 150% in the quarter) create holding incentives that suppress velocity for those specific tokens while concentrating high-velocity activity in non-yield instruments like USDC.
The market is segmenting: high-velocity settlement tokens, low-velocity yield instruments, and low-velocity savings/remittance tools. Each has a different relationship between supply and utility.
The stablecoin market is undergoing a phase transition. Supply contraction coinciding with volume records is the signature of a maturing payment instrument, not a declining one. Each stablecoin dollar is working harder — settling more transactions, turning over faster, and routing through increasingly institutional infrastructure.
The velocity divergence between USDC and USDT reflects the segmentation of the market into distinct economic functions: institutional settlement, yield generation, and emerging-market savings. These categories have different velocity profiles, different chain preferences, and different regulatory exposures.
For supply forecasters, the central question has shifted. It is no longer "how many stablecoins will be issued" but "how much economic throughput can each token support." At current velocity trajectories, the answer implies a market that may deliver multi-trillion-dollar settlement volumes well before supply reaches the levels previously projected. The denominator has changed.