I just delved into a new Dune dataset on stablecoins, and here’s what strikes me: everyone talks about $300 billion in circulation, but few understand what’s behind those numbers. Who actually holds them? How concentrated is ownership? On which chains do they operate? And most importantly — how are they actually used?

Meta just announced integration of third-party stablecoins into payments, PayPal is expanding PYUSD, Ripple launched RLUSD. Regulators and institutions are accelerating their market entry. They need answers that go far beyond mere supply figures.

Here’s what the data shows. On major chains, the 15 leading stablecoins reached $304 billion as of January 2026. That’s 49% more than a year ago. USDT accounts for $197 billion (Tether), USDC — $73 billion (Circle). Together, they control 89% of the market. By chain: Ethereum holds $176 billion (58%), Tron — $84 billion (28%), Solana — $15 billion (5%), BNB Chain — $13 billion (4%).

But here’s the interesting part: 2025 became the year of competitors. USDS grew by 376% to $6.3 billion. PYUSD soared by 753% to $2.8 billion. RLUSD experienced a staggering growth of 1803%, from $58 million to $1.1 billion. USD1 jumped from zero to $5.1 billion. But not all moved in the same direction: USD0 fell 66%, USDe ended the year with only a 23% increase after tripling.

Now, who holds them? On EVM and Solana, centralized exchanges are the largest known category, with $80 billion (a year ago it was $58 billion). Stablecoins remain infrastructure for trading and settlement. Major holders accumulate $39 billion. Yield protocols nearly doubled their holdings to $9.3 billion. Addresses of issuers grew 4.6 times to $10.2 billion — reflecting new issuance in the market. Only 23% of supply is on completely unknown addresses. For a blockchain, that’s an incredibly high level of identification.

Regarding the number of holders: 172 million unique addresses own at least one of these 15 stablecoins. USDT — 136 million addresses, USDC — 36 million, DAI — 4.7 million. The top three have truly broad distribution. The top 10 wallets of USDT and USDC control only 23-26% of the supply. But look at the others: the top 10 hold 60-99% of each. USDS has 90% concentration in ten wallets despite a turnover of $69 billion. USDF — 99%. USD0 — an extreme case: 99% in the top 10, and almost a monopoly of one or two wallets among them.

This doesn’t mean the tokens themselves are bad — some are newly launched, others intentionally created by institutional investors. But it does mean that volume data needs to be interpreted quite differently than for USDT or USDC. Concentration determines risk of disconnection, liquidity, whether “supply” reflects real demand or the behavior of a few large players.

In January 2026, stablecoin transfer volume reached $10.3 trillion — more than double the amount from a year earlier. And here’s what’s striking: distribution across chains differs radically from the supply picture. Base leads with $5.9 trillion in transfers, even though it has only $4.4 billion in circulation. Ethereum — $2.4 trillion. Tron — $682 billion. Solana — $544 billion. BNB Chain — $406 billion.

In terms of transfer volume, USDC dominates with $8.3 trillion — nearly five times more than USDT (1.7 trillion), even though its supply is 2.7 times smaller. USDC is transferred more actively and more frequently. DAI — $138 billion, USDS — $92 billion, USD1 — $43 billion.

And now, the most interesting part — how are they actually used? In January, detailed data showed: market infrastructure (trading and liquidity on DEX) — $5.9 trillion. That’s the largest usage scenario. DEX swaps — $376 billion. Together, this shows stablecoins function as trading collateral and liquidity. Interestingly, most volume is concentrated in rewards (liquidity mining), not pure demand.

Flash loans — $1.3 trillion (automated arbitrage cycles). Lending activity — $137 billion. CEX cash flows: deposits $2.24 trillion, withdrawals $2.24 trillion, internal transfers $1.51 trillion. Total — $5.99 trillion. Bridges — $28 billion. Issuer actions: minting $280 billion, burning $200 billion, re-peg adjustments $230 billion. Yield protocols — $2.7 billion. Overall, 90% of volume flows through identified categories.

Now, about transaction velocity — a metric often ignored. It’s the transfer volume divided by supply. Shows how actively a token is used as a medium of exchange. USDC on L2 and Solana has the highest velocity. On Base, the average daily turnover of USDC reaches 14 times — an astonishing figure driven by high-frequency DeFi activity. On Solana and Polygon — about once per day. On Ethereum, USDC — 0.9 times.

USDT moves fastest on BNB and Tron. On BNB, daily turnover is 1.4 times (active trading). On Tron — 0.3 times, but steady day after day, consistent with its role in cross-border payments. On Ethereum, USDT — only 0.2 times, with over $100 billion largely inactive.

USDe and USDS have low velocity — by design. USDe on Ethereum — 0.09 times, USDS — 0.5 times. Both are created as yield-bearing: USDe is embedded in sUSDe for yield from Ethena’s strategy, USDS is placed in Sky Savings Rate. A significant portion remains in contracts, lending markets, structured cycles. Low velocity here isn’t a flaw — it’s a feature.

One interesting point: blockchain activity is more important than the token itself. PYUSD on Solana has a turnover of 0.6 times per day, on Ethereum — 0.1 times. One token, vastly different usage models depending on the ecosystem.

Additionally, the dataset tracks over 200 stablecoins in more than 20 fiat currencies: euro (17 tokens, 990 million), Brazilian real (141 million), Japanese yen (13 million), Nigerian naira, Kenyan shilling, South African rand, Turkish lira, Indonesian rupiah, Singapore dollar, and others. The total volume of non-dollar stablecoins is currently $1.2 billion, but already 59 tokens are available across six continents, making up nearly 30% of all tokens in the dataset. Infrastructure for local fiat-backed stablecoins is being built right now.

All I’ve described is just the tip of the iceberg. The full dataset covers nearly 200 stablecoins across more than 30 blockchains. Each transaction is matched with its trigger and classified into one of nine activity categories. Each balance is detailed by holder type. Together, this turns noisy blockchain logs into structured data — revealing capital flows, concentration risks, participation models.

This allows answering questions I didn’t even ask: which wallets started accumulating a new stablecoin before it hit exchanges? How does concentration change in the days leading up to disconnection? What are the cross-chain transfer flows for euro-stablecoins? Correlation between issuer issuance and market pressure? This is precisely the dataset for institutional analysis, research reports, risk models, compliance monitoring. Depth is there. Start digging.