Select Financial Services
Ota yhteyttä

The stable dollar buys America’s debt – and people vote with their feet

Originally published in Finnish in September 2026. Market figures and legislative status are as of August 2026.

Stablecoins are no longer a side plot of the crypto market. They are buyers of short-term US government debt, a cross-border payment rail, and the way people in emerging markets escape weak local currencies. In August 2026 the whole stablecoin market stands at roughly $290–300 billion. USDT accounts for a little over $183 billion and USDC for about $72–73 billion. The peak was seen in May at close to $321 billion. Headlines scream “collapse”, even though turnover and velocity tell a different story.

This is not speculation. This is the dollar moving without a banking licence.

Market cap is not the whole story – movement is

USDT is a warehouse. USDC is a pipe.

USDT tokens often sit in wallets and OTC inventories. USDC circulates. In the first half of 2026, USDC is estimated to have carried the majority of adjusted transfer value, even though its supply is less than half of USDT’s. One dollar of USDC turns over many times more often than one dollar of USDT. For payments, that difference is decisive.

The slight contraction in market cap since spring is not “death”. It is seasonal outflow, token rotation, and money going to work in DeFi instead of sitting idle. From $321 billion in May to about $300 billion in August is a correction, not a crash. At the same time, on-chain transfer volumes have been historically high. Money that moves is a financial system. Money that sits is a warehouse.

The growth is still structural. Fiat-backed dollar tokens hold more than 90 per cent of supply. The algorithmic experiments ended as Terra’s ghost. The market chose reserves, redemption and T-bills.

The US federal government got a new buyer – one without a vote

This is where the story turns into macro policy.

The largest issuer’s reserves (USDT) rely heavily on short-term US Treasuries and repos. Depending on the source, its Treasury exposure has been in the range of $115–141 billion. That puts it among the largest holders of US debt — compared with countries, not fintech start-ups. The USDC issuer parks most of its reserves in a government money market fund run by a large asset manager: T-bills and overnight repo, short maturities, strict redemption logic.

So the stablecoin sector is a buyer that:

The GENIUS Act (Guiding and Establishing National Innovation for U.S. Stablecoins) entered into force in 2025. In August 2026 the US Treasury opened a draft rule on its implementation. A permitted payment stablecoin in the US means in practice: cash, insured deposits, T-bills under 93 days and repo tied to them. The USDC issuer is already on the path of a federal trust charter and a MiCA licence. USDT has a gap in GENIUS eligibility: part of its reserves are gold, bitcoin, loans and other items the law does not accept for a US payment stablecoin, and its issuer is building a separate US-compliant product. The market is splitting into a global OTC dollar and a licensed, “bank-grade” dollar.

For the US Treasury market this is demand that pushes short-term rates down. The IMF and the BIS have already measured the effect. The federal government gets a buyer that wants exactly the paper the Treasury sells every week. This is not a conspiracy. It is arbitrage: a redeemable dollar needs the most liquid collateral available.

The CLARITY Act: rules, not a ban – and a yield ban that does not kill DeFi

GENIUS dealt with stablecoin issuers. The CLARITY Act (Digital Asset Market Clarity Act) is a market-structure law: who supervises what, what is a commodity, what is a security, what is a payment stablecoin.

Status in August 2026: the House passed the bill in 2025. The Senate Banking Committee advanced it in May 2026. In July Senator Lummis published a combined text. In August the Senate leadership filed cloture; the voting window opens in September. The law is not yet in force. But the market is already pricing it in.

Key opportunities:

The painful point is yield. The bank lobby wanted to kill every stablecoin reward that “resembles interest”. The compromise bans passive yield for merely holding a payment stablecoin on a platform — in other words, “leave your USDC on an exchange and earn interest that replaces a bank deposit”. It does not ban:

So CLARITY does not kill yield. It moves yield away from the “deposit-disguised” exchange balance to where the risk is visible: DeFi and separate lending products. Banks got protection for retail deposits. The market got a clearer pipe.

This is an opportunity, not a funeral. A rule that separates a means of payment from an investment product is exactly what institutional money has been waiting for. Ambiguity is more expensive than a strict but readable law.

Portability is the product

Stablecoins don’t win because they are “crypto”. They win because they cross borders at the weekend, without a correspondent bank, without a three-day T+2.

This is the portability the traditional dollar doesn’t offer without a SWIFT queue and a compliance department.

A means of payment – at the checkout, not just a trading pair

E-commerce isn’t waiting for a seminar in Brussels.

Major e-commerce platforms already offer a native USDC checkout in selected markets in North America, Europe and Asia-Pacific, on several networks (Base, Ethereum, Optimism, Polygon, Arbitrum), and the merchant can settle in local fiat. That is exactly what a growth entrepreneur wants: the checkout does not turn into a trading desk.

Other rails include crypto payment processors serving electronics and ticket retailers and online store merchants, especially in cross-border trade — USDT on TRC-20 is still the workhorse of emerging markets. Some travel agencies accept USDT/USDC directly, and VPN and hosting companies have had crypto at the checkout for years, because their customers are global and card payments can be declined for various reasons.

Geography is politics. In the US, state lines (NY, TX) block the native USDC checkout. In the EU, MiCA decides who may offer an e-money token. Hong Kong is Asia’s gateway. In Latin America and Africa the volume doesn’t come from a native checkout button but from P2P wallets, local exchange corridors and TRC-20 payments. Commerce follows where the dollar is cheapest to move, not where the central bank holds a seminar.

World trade: the dollar without a correspondent bank

B2B trade is the next layer. A supplier in Vietnam, a buyer in Germany, settlement in USDC on Base, the reserve in T-bills. The correspondent bank is cut out. This won’t replace trade finance documents overnight, but it already replaces part of advance payments, freelancer fees, software licences and commodity prepayments.

When GENIUS makes the US payment stablecoin a recognisable instrument and CLARITY clarifies the platforms, an institutional treasury can hold part of its working capital on-chain without giving its lawyer a heart attack. Tokenised T-bills and tokenised receivables are growing on the same rail. The stablecoin is the liquid gap where trade sits overnight.

The future is not “all trade moves to crypto”. The future is that dollar clearing breaks free of banking hours.

DeFi is an interest account the bank cannot monopolise

This is where the bank lobby starts to cry.

In August 2026, typical high-TVL stablecoin lending rates are not a 20 per cent DeFi summer. They are in the 2–6 per cent range at the core, higher in curated vaults:

The average “all opportunities” APY on a screener is about 4 per cent. It is a mix of the T-bill rate and a risk premium. It is not a miracle. It is the market price of someone borrowing dollars and someone else lending them.

The CLARITY ban on passive exchange yield makes this more transparent: if you want yield, you take lending risk or liquidity risk openly. That is more honest than “a reward that looks like deposit interest”. A stablecoin DeFi interest account is exactly this: USDC into a pool, interest comes from the borrower, redemption follows the protocol’s rules. No deposit guarantee. No central bank window. That is why it can pay more than a current account.

For a growth entrepreneur this is treasury management. The company’s idle dollars don’t have to sit at zero interest, as long as the risk is priced and the position can be unwound. (How these positions are taxed in a Finnish company: DeFi tax and crypto accounting.)

The MiCA fortress: what the reserve requirements really demand

MiCA is not one “stablecoin law”. It is two castles on top of each other. A single-currency token is an e-money token (EMT, Title IV). A basket-referenced token is an asset-referenced token (ART, Title III). The dollar stablecoins commerce uses — and EURC — are EMTs. The ART castle was built for Libra’s ghost. In August 2026 ESMA’s register has zero authorised ART issuers. The fortress killed the category before it was born.

A dollar stablecoin’s EU eligibility runs through the EMT gate: only an EU credit institution or an authorised electronic money institution (EMI) may issue. The USDC issuer came in via France’s ACPR. USDT stayed out. After the CASP transition in July 2026, regulated EU exchanges restricted USDT. The first wall of the fortress isn’t the reserve. It is the licence.

Two different kinds of “backing” – don’t mix them up

  1. Reserve / safeguarding – assets matching the token holders’ claims.
  2. Own funds – the issuer’s loss buffer, a cousin of bank capital.

The first protects the redeemer. The second protects against the issuer failing. MiCA requires both. GENIUS focuses on the first and sets the instrument list (cash, insured deposits, T-bills under 93 days, eligible repo). MiCA additionally dictates where the money sits and who may issue it.

EMT reserve – 30 per cent in an EU bank, the rest “safe”

Article 54 layers on top of the E-Money Directive. Funds received for EMTs must be safeguarded as follows:

In practice the 70 per cent means short government paper, central bank claims and the instrument universe defined in the EBA’s RTS — no bitcoin, no gold, no corporate loans. One-to-one coverage. Segregation from the issuer’s own balance sheet. Custody with a separate party.

This 30 per cent bank-deposit floor is the core of the fortress and its weakest point. The token’s “safety” is wired back into the very system the token was supposed to break away from. The issuer is a single depositor. EU deposit protection is €100,000 per depositor. A billion-euro reserve in an EU bank gets the same protection as a savings account. SVB taught in 2023 what happens when the “safe deposit” sits in one bank. A major crypto custodian’s warning in 2026 said the same thing in different words: MiCA builds a direct pipe from bank stress to the token.

ART reserve – the heaviest wall, an empty castle

Articles 36–38 form the backbone of Title III. The reserve must at all times at least equal redemption claims, marked to market. It may not be pledged. The custodian may not be the issuer; it can be a credit institution, an investment firm (for financial instruments) or a CASP (if the reserve is crypto). Independent audit every six months. The RTS set daily and weekly maturity floors: how much can be liquidated within one or five banking days.

This is the “right” reserve model, closer to GENIUS in instrument discipline. Nobody has bothered to build an ART issuer on it in the EU. A rule without a product is the fortress’s second definition.

Own funds: 2 per cent, 3 for significant tokens, more under stress

Article 35 (ART, with an equivalent overlay for EMTs): own funds of at least the highest of:

The supervisor can raise the requirement by 20 per cent based on a risk assessment. A stress test can add up to 40 per cent on top of the base calculation. For a significant token the buffer floor rises to 3 per cent of the reserve, capital quality moves closer to banks’ CET1, and supervision moves to the EBA.

Significance criteria (Articles 43 and 56) – meet three and you are significant:

A USD EMT the size of USDC meets these with its eyes closed. 2–3 per cent own funds on a $70 billion reserve is $1.4–2.1 billion of capital. That is not thin for a start-up. It is thin if 30 per cent of the reserve is EU bank deposits that get frozen the same morning the redemption queue starts.

For comparison: the UK FCA cut the issuer capital buffer to 1 per cent in 2026, undercutting MiCA. Regulatory competition is already under way. The fortress is not “the global standard”. It is one offer that London undercuts.

Redemption is a wall that makes the EU the door for a run

Article 49: an EMT is redeemed at any time, at par, in funds, without a redemption fee. The holder has a direct claim on the issuer. This cannot be made conditional on a threshold.

This is consumer protection. It is also the ESRB’s and the ECB’s nightmare in multi-issuance. When the same fungible token is issued by an EU entity and a third-country entity, the redeemer chooses the door with the strongest rights. The EU wrote the world’s strictest redemption right, so the EU reserve is the door everyone rushes. The Lagarde-led ESRB recommended banning multi-issuance. In July 2026 Parliament voted 390–86 to keep it, with safeguards. The fortress built itself a systemic risk and then voted not to close the gate.

The recovery and redemption plans (Articles 46–47, 55) are the paper-token version of bank resolution. The issuer must plan in advance how the token will be wound down in an orderly way. This isn’t “crypto regulation”. It is bank regulation in a token shell.

The interest ban is the fortress’s political brick

Article 50 (EMT) and Article 40 (ART): the issuer, the offeror and the service provider may not pay interest. Any benefit tied to the length of holding counts as interest.

This is not a stability requirement. It is a deposit monopoly. The token may be a means of payment. It may not be an account. CLARITY and GENIUS do the same for issuers in the US, but leave DeFi lending and activity rewards alive. MiCA extends the ban to service providers. The Commission’s 2026 review asks directly whether the ban should be opened. That is the point where the fortress either admits competition or doubles the wall.

The consequence is already visible: yield doesn’t disappear, it changes address. DeFi lending protocols don’t ask a national supervisor for permission. An EU user who wants yield leaves the EMT shell for DeFi or for outside the EU. The fortress didn’t kill yield. It outsourced it.

MiCA vs GENIUS

MiCA EMTGENIUS payment stablecoin
Who may issueEU bank or EMIFederal or state licence
Reserve instruments30% credit institution accounts + 70% HQLA in the same currencyCash, insured deposits, ≤93-day T-bills, eligible repo
Bank linkMandatory 30% floorAllowed, not required
Own funds2% / 3% for significant + stress add-onIssuer solvency separate, emphasis on reserve quality
Interest from issuerBannedBanned
Yield on platforms / in DeFiBroad ban on service providersCLARITY: passive exchange yield banned, lending products allowed
RedemptionAlways, at par, freeAt par, liquid reserves
Global tokenMulti-issuance dispute openForeign token only with technical compliance + reciprocity

GENIUS buys US debt. MiCA buys EU bank deposits. The first is a friend of the Treasury. The second is a friend of the domestic banking system. That is why “systemic risk” means something different in Brussels than in Washington. Washington fears redemption from an illiquid reserve. Brussels fears the token eating deposits — and its own redemption right draining reserves in a run.

The fortress in numbers, not slogans

In August 2026 ESMA’s EMT register lists a handful of authorised issuers (about 23 issuers, 43 white papers). The USDC issuer is in. USDT is not. Zero ART issuers. The CASP transition ended on 1 July 2026.

This fortress works: it filters. It also filters out liquidity. USDT didn’t disappear for Europeans. It disappeared from the regulated European checkout and stayed on Tron, OTC and P2P corridors. Systemic risk didn’t go away. It moved out of the supervisor’s sight.

A 30 per cent bank floor, 2–3 per cent own funds, fee-free par redemption, an interest ban and mandatory EMI status: these five bricks make MiCA a fortress. Three of them are defensible as redeemer protection (1:1, segregation, par redemption). Two are about power (the bank floor and the interest ban). The growth entrepreneur’s question isn’t whether a reserve is a good idea. It is. The question is why the reserve must sit in exactly the bank whose deposit monopoly the token threatens.

If the requirement were just T-bills and cash — like GENIUS — the fortress would be a reserve standard. Now it is also a distribution licence and an interest ban. That is why it leaks. The customer doesn’t live in the castle. They live where the dollar moves and pays.

The IMF’s dollarisation anxiety

The IMF’s concern is different, and more honest. Nigeria, Argentina, emerging markets more broadly: the dollar stablecoin is already a corridor. The IMF’s summer 2026 country analysis of Nigeria and an IMF speech in Cape Town in August say the same thing: dollar stablecoins weaken local monetary policy, get around capital controls and pull deposits out of the domestic banking system. A local “own stablecoin” can even accelerate dollarisation, because it becomes an on-ramp to global dollar liquidity.

An Argentinian who held $10,000 in pesos from 2016 to 2026 ended up with the purchasing power of about $114. In that situation, “outflow into stablecoins” isn’t speculation. It is survival.

Is this regulators whining about losing power?

Yes. Partly. And that part is the most important.

A central bank fears two things: that its money loses the competition, and that the banking system it supervises loses its deposit monopoly. The IMF fears that a weak currency no longer traps its citizens. The EU fears that the redemption rights it wrote itself make it the door for a run. US banks fear that the T-bill yield flows to the token holder instead of into deposit interest.

None of these fears is technical. They are political. People choose better money when it is available. A weak local currency doesn’t fail because stablecoins exist. It fails because it is weak. The stablecoin just makes the escape cheaper and faster.

Systemic risk is real in one scenario: mass redemption + illiquid reserves + a bank link. That is why GENIUS’s T-bill straitjacket makes more sense than MiCA’s interest ban plus a 30 per cent bank floor. Short paper, a transparent reserve, redemption at par. The risk doesn’t disappear, but it becomes interest-rate risk and a redemption queue, not a black box.

Power slipping through the fingers is exactly what free exchange does. The dollar is still a US project. The token just frees its distribution from the permission of the Fed and correspondent banks. The federal government gets a buyer for its debt. A citizen with a weak currency gets protection. A merchant gets a checkout without a 2.9 per cent card fee. DeFi gets yield that doesn’t pretend to be a deposit.

A regulator who calls it “systemic risk” every time a customer leaves a bad system isn’t defending stability. They are defending a monopoly.

The central bankers’ case deserves a hearing

The central bank defence is also worth taking seriously, because in some markets the risk is real. Systemic risk can arise precisely when people make a rational decision and move their money out of bank deposits. If, say, 10 million people decide on the same day that “a stablecoin is better than a bank account”, no single person does anything wrong. But a decision that is rational for the individual can collectively create a systemic problem. Here “consumer freedom” and “financial stability” genuinely conflict. If banks’ deposit funding shrinks significantly, lending, liquidity and the transmission of monetary policy can change.

The CLARITY and GENIUS line — rules for issuers, open competition in distribution, a ban on the passive bank disguise, separating DeFi — is the growth entrepreneur’s line. The EU’s line is to fear the token and wire it back into the bank account. The IMF’s line is to warn those whose currency can’t stand scrutiny. The market has already voted.

Those who are asking for more interference of the government are asking for more coercion and less freedom.

Ludwig von Mises

A stablecoin is not anarchy. It is a dollar that doesn’t ask permission to move. If that is frightening, the question isn’t the token. The question is why your own system can’t withstand competition.

What this means for your company

If your company invoices, pays contractors or holds part of its treasury in stablecoins, the accounting and tax treatment in Finland depends on what kind of token it is. Read Stablecoin accounting and tax in Finland, the follow-up article Why euro stablecoins lag behind, or see our services in English.

Talk to us

Select Financial Services Oy, Fredrikinkatu 61, 00100 Helsinki. Tell us what you hold, where it sits and how your company uses it — we will tell you how it should be booked and taxed in Finland.

Email hello@selectfinancial.fi
Call +358 45 1583 550