In the spring of 2017, I stood in front of 150 developers in a converted Prague warehouse and asked them a question that had nothing to do with price charts: if a system's rules are written by people you cannot see, whose rules are they really?
Nine years later, I am reading a European Central Bank opinion that asks almost the same question, only with more lawyers.
The opinion, submitted through the European System of Central Banks into the MiCA review process, argues that the reserve rule governing large stablecoin issuers should be deleted. In plain terms, that rule requires issuers above a certain size to keep 60% of their reserves in deposits at EU credit institutions. Smaller issuers face a 30% floor.
Tether CEO Paolo Ardoino has spent two years calling that requirement unsafe โ not unsafe for regulators, unsafe for users. He says the deposit floor makes tokens less safe, not more.
So here we are. The most offshore stablecoin issuer on earth and the central banking apparatus of the European Union are pulling the same lever, on the same clause, at the same moment. Almost everyone is reading it as a Tether victory. It isn't.
The machinery matters here, because the details carry the story.
MiCA โ the Markets in Crypto-Assets Regulation โ sorts stablecoins into two buckets. Electronic money tokens reference a single fiat currency. Asset-referenced tokens reference a basket or some other value. USDT lands in the asset-referenced bucket, which carries the strictest obligations in the entire framework.
Inside those obligations sits the reserve-composition rule, built on a two-tier threshold. Ordinary issuers must hold at least 30% of reserves as bank deposits. Issuers designated "significant" must hold at least 60%. The remainder can sit in other eligible assets.
Tether has never applied for a MiCA license. Not once. USDT still circulates inside the EU through exchanges, over-the-counter desks, and self-custody routes that no single supervisor controls. The ESCB's own opinion concedes that supervisors face "significant challenges" precisely because non-compliant crypto firms keep reaching European customers.
The compliant end of the market moved on without them. Revolut delisted USDT this year. USDC, Circle's product, holds the licenses and the banking relationships. European users who wanted a dollar stablecoin on a mainstream platform increasingly got one choice.
And Circle โ according to reporting over the summer โ supports the same MiCA amendment that could let Tether back in. That looks self-destructive for a compliance-first issuer, until you work out why. I'll come back to it.
In 2025 I advised an EU regulatory task force on structuring decentralized governance guidelines that protect retail participants. I sat in rooms where the difference between a mandatory floor and an advisory standard was the difference between a rule people comply with and a rule people route around. That is why I read the 30/60 thresholds as a design question rather than a political one โ and why I am not surprised that the most consequential actor in this debate is the one that never showed up to the hearing.
This is also landing in a bull market, which is precisely when reserve schedules go unread. New capital is flowing into stablecoin supply, and nobody pricing risk this quarter wants to hear about maturity bands. The consultation window closes on September 30. Until EU legislators actually revise the text, the floors remain current law and Tether remains unauthorized.
Now the technical layer, because everything above is downstream of one engineering decision.
Strip away the politics and the 60% rule does exactly one thing: it forces a stablecoin issuer to hold a large share of its liabilities as unsecured claims on commercial banks.
A stablecoin is a demand liability redeemable instantly, globally, around the clock. A bank deposit is a claim inside a fractional-reserve institution whose own liabilities are also redeemable at par. Stacking them means stacking two demand-liability structures on top of each other.
The ESCB sees this clearly. Its opinion notes that funds moving in and out with token minting and redemption are not stable deposits โ they are wholesale money that can leave overnight. A large redemption event could drain a bank's balance sheet faster than any supervisor could respond. That is not rhetorical risk language. It is correct, and it is the strongest technical argument in the entire document.
The proposed alternative is cleaner: allow reserves to be held in assets maturing within one to five business days. Short-dated sovereign paper. Money market instruments. This is already the global norm. Reserve attestations from the largest issuers are dominated by short-term Treasury bills, precisely because the instrument matches the redemption profile of the liability. Short paper can be sold into a deep market in hours. A bank deposit cannot be withdrawn in size without triggering exactly the dynamic the ECB fears.
Ardoino's counter-argument is narrower than it sounds, but it holds. EU deposit insurance caps at โฌ100,000 per depositor per bank. USDT's reserves run into the tens of billions. A deposit-heavy mandate therefore wraps an enormous exposure in an insurance envelope that covers a rounding error of it. Calling that arrangement "safer" requires a definition of safety that only survives on a slide.
Here is the part that gets lost. The 60% floor is not a safety rule. It is a funding subsidy. Banks receive a large, cheap, liquid wholesale deposit balance they never had to compete for. That is the quiet beneficiary of the current text, and it explains why an amendment framed as "relaxing rules for stablecoin issuers" is being pushed by central bankers rather than resisted by them.
There is a second ledger nobody is reading. Tether's business model is reserve-interest arbitrage: users hold zero-yield USDT, Tether invests the reserves and keeps the spread. Change the eligible-asset mix and you change the spread. Deposit rates and short-term bill yields are not the same number, and the gap is not small in a high-rate environment. A debate that presents itself as prudential supervision is also, quietly, a negotiation over who captures the float.
One more technical note, because it reframes the whole fight. The two thresholds are not calibrated to any observed redemption behavior. They are round numbers with a recent legislative provenance and no published stress model behind them. I have spent enough time around protocol parameter debates to recognize the pattern: a number gets chosen for political legibility, then acquires a patina of engineering authority it never earned. A reserve floor that was never calibrated cannot be defended on prudential grounds. It can only be defended on political ones. That is why the deletion argument is winning.
I have to name the DeFi parallel, because it is the same disease. I have spent years explaining liquidation mechanics to non-technical users, and the thing I always have to admit is that the interest rate curves in the largest lending protocols were set by governance votes, not discovered by markets. The parameters look scientific. They are chosen. MiCA's 30% and 60% are chosen too. Whether it is 30/60, or a one-to-five-day maturity band, or something else entirely, the number is a decision made by a small group of people about a system used by millions.
That is the actual subject of this article. Trust is not a feature you ship; it is a property you earn, audit by audit.
Now the contrarian angle, and it stings.
The dominant headline is that Tether and the ECB have landed on the same side. True. The dominant conclusion โ that Tether is winning โ is false, for a reason that has nothing to do with the merits.
Tether never applied for a license. The 60% rule was never binding on Tether. Deleting it takes nothing away from the ECB and gives Tether nothing it did not already have.
An issuer that never enters the perimeter cannot be constrained by rules written for the perimeter. So the ECB is not conceding anything to Tether. It is cleaning up a rule that was creating risk inside European banks โ risk that flowed from an issuer it could not regulate in the first place. Removing the deposit mandate moves dollar-stablecoin exposure out of EU bank balance sheets and into short-dated sovereign paper, where it belongs and where somebody else holds the bag. That is not a truce. It is a boundary adjustment.
Which brings me back to Circle. Support for an amendment that could readmit your largest competitor looks irrational until you notice that Circle, as a significant issuer, is subject to the same 60% floor. Relaxing it improves Circle's own reserve flexibility. The support is self-interested, not magnanimous.
And the deepest blind spot: the ECB's warning is not really about USDT at all. Its stated concern is that euro-denominated stablecoin growth could squeeze bank lending inside the euro area. USDT's reserves are dollar assets. They do not directly threaten euro-area credit creation. The eagerness to delete the deposit mandate is risk isolation, paired with a freer hand to police euro stablecoins more tightly. This is a dollar stablecoin story by accident, and a euro stablecoin story by design.
There is a temptation to read "central banks and Tether agree" as vindication for crypto. Resist it. A coincidence of positions is not an alliance, and the ECB would write a rule tomorrow that makes USDT unusable in the euro area if it believed euro bank lending were at stake.
One more uncomfortable observation for anyone who cares about governance. DAO voter turnout on major proposals still sits below 5%. "Community decision-making" in practice often means a handful of whales and venture funds moving the quorum. Here the quorum is 27 national central banks and a comment deadline. The scale changed. The structural problem did not: the people most exposed to the rule have the least access to writing it.
What should you actually watch? Not Ardoino's next post. Watch the consultation output after September 30, then watch whether the Commission drafts an amendment at all โ the ESCB opinion is advisory, and the final text belongs to the Commission and Parliament. The signal that matters is not whether Tether returns to Europe. It is whether MiCA can enforce anything against an issuer that simply declines to apply. If the answer is no, then every reserve rule in the framework is a suggestion wearing a statute's clothing.
I organized a workshop series in Prague a decade ago for developers who were drowning in speculation, and the only thing that ever worked was slowing down enough to explain the mechanism. Regulation people cannot understand is regulation they cannot follow. Reserve schedules nobody reads are reserve schedules nobody respects.
Education is the ultimate yield. Build for humans, not just nodes.