Ninety-Two Percent of Nothing: Inside the MiCA Artifact Reshaping Euro Stablecoins

Ethereum, Solana, Base. Ninety-two percent. That is the entire euro stablecoin market compressed into three chain names and one integer — and it is the only hard number the story hands us. Glitch detected. Source traced. No timestamp. No total circulating supply. No issuer breakdown. No methodology note. Just a concentration ratio floating in a vacuum, attached to a regulation-shaped adjective: "rising centralization and regulatory influence." That is not a data point. That is a silhouette. And a silhouette is where forensic work begins, not where it ends. I have spent years reverse-engineering stablecoin float on both sides of the Atlantic, first as a backend engineer chasing integer overflows in pre-sale code, later as an exchange market lead building Python models against institutional flow. I know what a real market-structure signal looks like, and I know what a headline dressed up to look like one looks like. This one is the second kind. But the second kind is often more informative, because the way a market describes itself tells you what it wants you to believe. So let me tell you the thing the headline refuses to say out loud. Ninety-two percent of a small number is still a small number. The euro stablecoin market is not a growth story wearing a MiCA badge. It is a compliance story wearing a growth badge. The distance between those two sentences is roughly the entire circulating supply of the sector. Let me reconstruct the scene. The claim we are working from is thin. Three chains — Ethereum, Solana, Base — account for 92% of the euro stablecoin market. That market is described as increasingly centralized, and the driver is named as regulatory influence. That is it. Three information points, no numbers underneath them. Which means everything below is a reconstruction from code, incentive structure, and the regulatory text itself. Where I am inferring, I will say so. Where the source is silent, I will name the silence as a signal in its own right. Before any of that, a foundational correction, because I am not afraid to slow down for people who need it. A euro stablecoin is not a governance token. It is not a utility token. It is not a bet on a protocol. It is a fiat-anchored money token — one euro in, one token out, one euro back on redemption, minus fees and friction. That is the entire mechanism. There is no staking yield by default, no emissions schedule, no treasury to raid, no governance forum where holders vote on direction. Choose which chain it lives on and you have chosen a rail, not an asset. This distinction matters because almost every bad take on euro stablecoins commits the same category error. It treats them like crypto tokens and asks who is winning the token war. Nobody is winning a token war. There is no token war. There is a compliance land grab wearing crypto clothing, and the three chains are simply the address labels. Now the regulatory frame, because that is the true center of gravity here. MiCA — the Markets in Crypto-Assets regulation — is the single most consequential piece of legislation ever aimed at this sector, and it is the reason the 92% number exists at all. Its stablecoin provisions landed first, tightening through 2024, with the broader framework phasing in behind them. Under MiCA, a stablecoin pegged to a single fiat currency is classified as an Electronic Money Token — an EMT. Not a security, but not free either. EMT status is the strictest practical category a stablecoin can fall into, and it comes with a set of conditions that reshape the entire competitive field. An EMT issuer must hold an Electronic Money Institution license or a full banking license. Reserves must be segregated. Redemption must be guaranteed at par. The float cannot be redeployed into whatever the treasury desk finds interesting that quarter — it has to sit in cash or high-quality, short-dated instruments, with limits and disclosures attached. Read that list again and notice what it actually is. It is a filter. Every clause throws out a class of issuer that used to exist: the offshore shell with a multisig and a marketing budget, the algorithmic construct with no reserves at all, the gray-market operator serving euro users from a jurisdiction that never asked questions. MiCA does not merely regulate euro stablecoins. It selects who is allowed to be one. That is the mechanism behind the 92%. Not technology. Not user demand. Not network effects in the way crypto natives understand them. A licensing regime compressed the field to a handful of viable issuers, and those issuers placed their product on a small number of chains that could satisfy institutional diligence. Liquidity draining from the long tail. Logic broken for anyone still modeling this as an open market. Let me name the landscape without pretending the source gave it to me, because it did not. The recognizable euro stablecoins that clear the MiCA bar are a short list: EURC from Circle, EURS from Stasis, EURI from Banking Circle, EURCV from Societe Generale-FORGE. Each of those names carries a regulated entity behind it. Each of them exists because a bank or a licensed institution decided the compliance cost was worth the float income. That short list is the market. Everything else is commentary. Now the triopoly itself, chain by chain, because treating Ethereum, Solana, and Base as interchangeable rails is the second-most common error in circulation. They are not interchangeable. They share almost nothing except the ability to mint a compliant euro token. Ethereum is the settlement layer. It is slow at base layer — fifteen to thirty transactions per second, give or take, depending on how you count — and it charges for the privilege of finality. What it offers is economic finality, the deepest liquidity pool in existence, the largest concentration of audited DeFi contracts, and the only serious institutional comfort zone for real-world-asset settlement. When a licensed issuer wants its euro token to be taken seriously by counterparties that file reports with regulators, it deploys on Ethereum. Not because it is fast. Because it is legible to institutions. I have audited enough of this landscape to know that "legible to institutions" is the most valuable property in the entire stack, and it has nothing to do with technical merit. Solana is the performance layer. Proof of History plus Proof of Stake, a monolithic design that processes orders of magnitude more throughput than Ethereum at base layer, with fees low enough that a payment use case is not laughable. If euro stablecoins are ever going to move at the speed of consumer payments — think cross-border remittance inside the eurozone, think point-of-sale settlement, think high-frequency treasury sweeps — Solana is the rail that can physically carry it. Its trust model is different from Ethereum's: less battle-tested at the extreme, more prone to the kind of cascading liveness questions that a validator set can create when load spikes. But the throughput is real, and for a licensed issuer optimizing for compliance-plus-speed, it is a rational deployment. Base is the outlier, and the outlier is the signal. Base is an Ethereum Layer 2. An Optimistic Rollup. It executes transactions off-chain and batches them back to Ethereum for settlement, inheriting Ethereum's security claims while offering cheaper execution. It launched in 2023, which means it appeared in this market after MiCA's stablecoin provisions were already shaping the field. That is not a coincidence. That is a timing signal. Here is the uncomfortable part. Base runs on a single centralized sequencer. One operator orders transactions. Fraud proofs exist in theory, but the sequencer is the choke point, and the choke point is a company. I flagged this exact architecture in an earlier piece and I will flag it again: the trust model of a single-sequencer rollup is not the trust model of Ethereum. It borrows the settlement layer's finality while inserting a company into the critical path of ordering. When people say euro stablecoins are getting more centralized, this is one of the concrete places that centralization lives. So why does Base matter? Because of who owns it. Coinbase incubated Base. Coinbase holds the licenses, the fiat on-ramps, the KYC relationships, the banking connections, the regulatory posture of a public company that answers to the SEC. Base is not winning euro stablecoin share because its technology is superior. It is winning because its parent is a compliance distribution channel with a consumer app bolted on top. The euro stablecoin that deploys on Base is not betting on a rollup. It is betting on Coinbase's ability to move regulated euros into and out of the conventional financial system. Exchange volume anomaly flagged. Not in the price — in the plumbing. The chain that should win on pure decentralization loses. The chain that should win on pure throughput places second. The chain that wins a top-three slot is the one attached to a regulated exchange. That ordering tells you more about what the euro stablecoin market actually optimizes for than any whitepaper ever will. Now the number itself. Let me do the thing the source did not. Ninety-two percent is a concentration ratio. A concentration ratio is meaningless without its denominator and its unit. Three questions determine everything: First, what is the total market? If the entire euro stablecoin float is a few hundred million dollars or low billions, then 92% describes a rounding error in the global stablecoin market. The dollar stablecoin complex is measured in the hundreds of billions. The euro complex is measured, in any honest accounting, at a fraction of a percent of that. Ninety-two percent of a small basin is not dominance. It is the shape of a puddle. Second, what is the unit? Ninety-two percent by what — circulating supply on-chain, transaction volume, number of issuers, number of tokens deployed? These give wildly different answers. If the metric is supply, it reflects where issuers parked their mint; if it is volume, it reflects where users actually transact. The two can diverge violently. A euro token can be minted on one chain and bridged to another for use. The source does not tell us which lens it used. That silence is not neutral. The number becomes a different fact depending on the answer, and the fact that it was presented without one is itself the tell. Third, what is the internal breakdown? If Ethereum, Solana, and Base sum to 92%, no single chain necessarily dominates. The plausible spread is broad. Each chain could sit anywhere between the mid-twenties and the mid-forties, and the aggregate would barely move. That means the headline "three chains dominate" could coexist with none of them dominating. Fragmentation disguised as concentration, depending entirely on how you frame the frame. This is the part that a fast reader will skip and a slower one will catch. The story says concentrated. The arithmetic permits dispersed. Both can be true, because the market is concentrated at the issuer level and dispersed at the chain level. The issuers are few. The rails are several. Do not confuse the two axes. Now the axis that actually matters, and the one the source buries. Euro stablecoins are not a chain competition. They are an issuer competition, and the chains are commodities. Consider the economics from the issuer's side. An EMT issuer earns float income — the interest on the reserves backing the tokens. Circle earns this on EURC reserves. Banking Circle earns it on EURI. Societe Generale-FORGE earns it on EURCV. The token holder earns nothing by default. There is no yield to attract users, no emissions to farm, no governance power to accumulate. The user holds a euro, on-chain, and gets the euro. So where is the competition? It is (a) over which issuance channel captures float, and (b) over which chain captures the transactions on that float. The issuer competes for regulated banking partners, for exchange listings, for payment integrations, for corporate treasury adoption. The chain competes for deployment — but only weakly, because an issuer can deploy the same token on multiple chains at trivial marginal cost. Once you have the license, the reserves, and the compliance apparatus, adding a chain is a config change. That is the definition of a commodity rail. If a customer can be served on any of several rails with no meaningful difference in outcome, the rail has no pricing power. The moat is not in the chain. The moat is in the license, the reserves, and the banking relationships. Base did not earn its slot with clever cryptography. It inherited it from a parent that already owned the compliant distribution. There is a sharper implication here, and I will state it plainly because it is the kind of thing that gets lost in the rush to report a percentage. If an issuer can switch or add chains cheaply, then the 92% figure is not stable. It is a snapshot of where issuers currently choose to mint. One issuer adding Solana, or one issuer quietly winding down a Base deployment, reorders the entire ratio, and reorders it again a quarter later. A market whose defining statistic can be rewritten by a single corporate decision is not a consolidated market. It is a fragile one wearing the mask of consolidation. Let me show how the failure would actually propagate, because I prefer concrete mechanics to abstractions. Suppose the largest euro stablecoin issuer — whichever it is, the source does not say — decides Base's economics no longer justify the deployment. Maybe sequencer fee policy changes. Maybe the parent's strategic priorities shift. Maybe a supervisory conversation goes badly. The issuer unwinds. Within a reporting period, Base's euro stablecoin share collapses, and the aggregate 92% reshuffles across Ethereum and Solana. The headline that read "three-chain oligopoly" a quarter earlier now reads "two-chain duopoly," or "Ethereum dominance," or something else entirely. Nothing technical changed. A business decision moved a market-structure statistic by tens of percentage points. That is what concentration without a real moat looks like. It is brittle precisely where it appears strongest. The longer the tail is excluded, the more the remaining chains depend on a handful of issuer decisions, and the more volatile the ratio becomes to events that have nothing to do with user demand. Now the demand side, which is where the euro stablecoin thesis quietly dies in the dark. Ask a simple question: who actually needs a euro stablecoin? The dollar stablecoin complex has a long list of answers. Crypto trading pairs are quoted in dollars. The deepest DeFi liquidity is in dollars. Cross-border trade settlement increasingly touches dollar tokens. Dollar stables sit at the center of the crypto credit system — collateral, margin, lending, borrowing. When you need a stable unit inside this ecosystem, the default is a dollar. The euro stablecoin has a much shorter list. European traders can use euro tokens for euro-denominated pairs, but the liquidity is thin. European businesses could settle cross-border euro payments on-chain, but the incumbents — SEPA, Target2, correspondent banking — already move euros cheaply and legally for the vast majority of the volume. DeFi can use euro tokens as collateral, but a collateral asset with shallow liquidity is a liability, not a feature. And a consumer in the eurozone who wants to hold euros digitally generally does not need a blockchain to do it. They have a bank account. That is the demand vacuum. Euro stablecoins are not solving a problem that euros do not already solve. They are adding a rail to a currency that already runs on rails of remarkable quality, in a currency union whose central bank has been explicit about wanting to issue its own digital money. And the buyer for euro stablecoins is thin. The dollar stablecoin buyer is a global crypto user who wants a dollar because the crypto system runs on dollars. The euro stablecoin buyer is a narrower creature: a European institution that wants compliant on-chain euro settlement, a payment provider looking for a regulator-approved token rail, an RWA operation that needs a euro leg. Real, but small. Growing, but slowly. Not the kind of demand that produces a runaway float. Which means the 92% concentration is happening inside a niche, and the more the market concentrates, the smaller the niche looks — because the rational participants are not fighting for share, they are defending a floor. And here is where I have to slow down again, because I keep watching people model stablecoin economics as if it were token economics, and it produces nonsense every time. A stablecoin is not a token with emissions, so the usual toolkit — FDV, TVL, APR, emissions, dilution curves — does not apply. What applies is float income and redemption mechanics. The issuer's revenue is a function of reserve size, interest rates, and the spread it earns on high-quality short-dated instruments. When rates are high, the float is a cash machine and every licensed issuer wants more of it. When rates fall, the machine slows and the marginal issuer reconsiders whether MiCA compliance is worth the operating cost. The euro stablecoin market is therefore rate-sensitive in a way that rarely makes headlines, and it is sensitive to the ECB's policy path as much as to any on-chain event. Rates also determine how the euro stablecoin competes against its siblings. If the ECB cuts and dollar rates stay higher, the euro token's float earns less relative to a dollar token's float, and the incentive to build euro-denominated products weakens. The euro stablecoin sector is, in a real sense, short the euro rate cycle. That is a structural headwind the growth narrative never mentions. Now the part of the story that the source actually got right, and I want to credit it before I dismantle the framing around it. It names regulation as the driver of centralization. That is correct, and it is the most important sentence in the whole piece. MiCA does not merely license euro stablecoin issuance. It restructures who can survive in the sector. The licensing requirement eliminates the long tail of unlicensed issuers. The reserve requirements eliminate the algorithmic and the thinly-backed. The redemption guarantee eliminates the opaque. What is left is a small set of entities with EMI or banking licenses, and those entities naturally cluster on the handful of chains that institutional counterparties will actually accept. The chain-level concentration is downstream of the issuer-level concentration, which is downstream of the licensing filter. Ethereum, Solana, and Base are not competing for a prize. They are the three venues that happened to clear an institutional bar that was set by Brussels and enforced by national supervisors. This is the reframe the story needs. The 92% is not evidence that the technical field has consolidated. It is evidence that the regulatory field has. That is a crucial distinction and almost everyone reporting this number will flatten it. And while I am reframing, let me take the contrarian angle that the source avoids, because this is where a careful reader earns the right to disagree with the obvious take. The obvious take is: three chains now own euro stablecoins, this is consolidation, on-chain rails are maturing, watch for further concentration. The contrarian take is: the 92% is a compliance artifact, and the man who reads it as market validation has the causality backwards. Here is the argument. Concentration ratios usually reflect competitive victory — one product outcompeted others on merit, users voted, the winner accumulated share, and the ratio measures the resulting mountain. That is not what happened here. No euro stablecoin won a user war. The market was culled by a rulebook. What looks like concentration is a selection effect: the chains that remain are the ones whose issuers could afford the license, and the issuers that remain are the ones the license allowed. Remove the license requirement, and the concentration would disperse almost immediately, because the underlying competition is not resolved. It is suppressed. If that is right, the ratio is a regulatory measurement, not an economic one. And regulatory measurements are fragile in a specific way: they move discontinuously. Supervisory guidance changes, an enforcement action lands, a new license category appears, a transitional provision expires, and the numbers jump. A market whose structure is set by a rulebook does not evolve. It gets re-regulated, and the ratio resets. There is a second contrarian angle, and I think it is the bigger one. The euro stablecoin market's real competitor is not another chain. It is the dollar stablecoin complex, and behind that, the euro itself in the form of a central bank digital currency. Glitch detected. Source traced — to the wrong axis. Everyone analyzing euro stablecoins by chain share is fighting over the wrong map. The fight that matters is between euro denominated on-chain money and dollar denominated on-chain money, and in that fight, the euro is losing by an order of magnitude. A European user who wants a stable unit does not compare Ethereum to Solana. They compare a euro token to USDT and USDC, and the latter have deeper liquidity, more listings, more integrations, and a head start measured in years. If the choice is "hold a euro token with thin markets" versus "hold a dollar token with everything," a great many users will hold the dollar and accept the currency mismatch. That is the competition. Chain share is a subplot inside a losing side-battle. And then there is the digital euro. The ECB has been developing a central bank digital currency for years, and while the timeline is slow and contested, the direction is one-way. A retail-facing digital euro would, at minimum, occupy the same conceptual space as a euro stablecoin: digital euros, held by people and businesses, moving through regulated infrastructure. If it launches with broad access, it competes directly with the licensed euro stablecoin for the retail and payment use cases. If it launches with tight limits, it still absorbs the political air, the legislative attention, and the central bank's internal resources that might have gone to supporting a healthy private euro stablecoin ecosystem. Either way, the euro stablecoin sector is building a business in the shadow of a central bank that has reserved the right to enter it. That is the single largest structural risk in the entire category, and it does not appear anywhere in the concentration headline. The market is worried about which chain is winning. The market should be worried about whether any private euro token is winning against a public alternative that has not shipped yet. Now let me put numbers to the silence, because the source gave me none. If the euro stablecoin float is in the low hundreds of millions of dollars, then three chains holding 92% are celebrating a prize the size of a mid-tier token launch. If it is in the low single-digit billions, then it is a real but minor market that can fund a few compliant issuers and no more. Either way, the absolute scale is the story, and the ratio is the distraction. A reader who takes away "three chains dominate euro stablecoins" has learned an interesting structural fact. A reader who takes away "euro stablecoins are a mature market" has been misled, and the source, by omitting the total, invited exactly that misreading. This is why, in my own reporting, I refuse to publish a concentration figure without its denominator. I have watched too many institutional desks trade against a ratio that was arithmetically true and economically empty. The 92% is arithmetically plausible. Economically, it is a shadow. Cast the right light and it disappears. Let me also flag the methodological trap that the source cannot have avoided, because no source can avoid it without saying which one it used. If the 92% is measured by supply, it reflects issuer deployment decisions — where treasury desks parked the mint — and it will overstate the importance of the chains where issuers are most comfortable issuing. If it is measured by volume, it reflects where users actually transact, and it will favor the low-fee, high-throughput rail. If it is measured by issuer count, it will over-weight a chain that hosts many small issuers and under-weight a chain that hosts one giant. Each of these is defensible. Each produces a different answer. And a report that gives the number without the lens is not neutral — it is choosing a lens and hiding the choice. I will go further. The fact that the source felt comfortable publishing 92% without a methodology is the most revealing thing about it. It tells you the number was chosen for narrative effect, not analytical precision. In markets that matter, we would never accept this. In crypto, we accept it daily. That gap is where misinformation lives. So where does this leave the euro stablecoin market? Let me draw the map I think is actually true, with the confidence levels attached, because I do not pretend to certainty I do not have. One: the market is issuer-concentrated and chain-dispersed. Medium-high confidence. The licensing filter explains the issuer concentration, and the multi-chain deployment behavior explains the dispersion. Both can hold simultaneously. Two: Base's presence in the top three is a Coinbase compliance signal, not a technical one. Medium confidence. The architecture — single sequencer, L2, 2023 launch — makes technical leadership implausible as the explanation. The parent's regulatory and fiat infrastructure is a far better fit. Three: the true competitive threat is dollar stablecoins and the digital euro, not rival chains. Medium confidence, and rising. On-chain euro demand is structurally thin, and the central bank has not ceded the space. Four: the 92% is fragile to corporate decisions and regulatory changes. Medium confidence. Concentration without moats behaves like a brittle shell, not a rock. Five: euro stablecoins are rate-cycle short and float-income dependent. Medium-high confidence. This is mechanically true of any reserve-backed token, and it means the sector's economics live and die on ECB policy. Liquidity draining from the narrative. Logic broken for anyone still calling this a growth market. Now let me write the part that readers never forgive and writers never include: the correction to my own framing. I have been hard on the euro stablecoin thesis, and I will be harder, but not unfair. There is a real, durable case for euro stablecoins, and it is not consumer payments or speculation. It is institutional settlement. The compliant euro token is a natural leg for real-world-asset settlement inside the eurozone. Picture tokenized bonds settling against a MiCA-compliant euro token on a regulated L2, both legs inside the same supervisory perimeter, both counterparties licensed, the whole transaction auditable on-chain. That is a genuine capability. Payment providers that want a regulator-approved token for cross-border euroflows have a real need. Corporate treasurers with euro exposure and an appetite for 24/7 settlement have a real need. None of these are retail. All of them are slow to scale. But they are real, and they are growing, and they are exactly the kind of use case that concentration at the issuer level actually suits, because institutions prefer dealing with a short list of regulated names. So my contrarian position is not that euro stablecoins are worthless. It is that the story being told about them is the wrong story. The right story is "slow, compliance-driven, institutionally paced, structurally capped by weak demand and a looming central bank alternative." The wrong story is "three chains consolidating a hot market, watch the growth." I have seen enough cycles to know which of those narratives gets written and which one gets traded. And there is one more thing I want to surface, because the source called it "centralization" and moved on, and that word is doing too much work. When a euro stablecoin gets "centralized," three separate things are being compressed into one. First, the issuer is a company that can freeze balances, blacklist addresses, and reverse transactions under supervisory demand — that is non-negotiable under MiCA and it is a feature, not a bug, of a regulated money token. Second, the chain may be centralized — Base's single sequencer, for instance — meaning the ordering of transactions depends on one operator. Third, the governance is corporate, not communal — there is no DAO vote, no on-chain proposal, no community of token holders charting direction, because there are no token holders with power. These are three different centralizations with three different remedies, and calling them all "centralization" is analytically useless. The issuer centralization is required by the regulator and is permanent. The chain centralization is a design choice that can be decentralized over time. The governance centralization is a category fact — a stablecoin is a product, not a protocol, and there is nothing to decentralize. The source flattened all three into one word. The reader who takes that word seriously will misjudge which risks are fixable and which are structural. Exchange volume anomaly flagged, one more time, at the level of narrative itself. The story is being told as if it were a market-share story. It is a licensing story, a demand story, and a central-bank-shadow story wearing market-share clothing. The sooner the analysis catches up to that, the fewer desks will be surprised by the next discontinuity. Let me end with what to watch, because a forensic report that does not tell you where the next anomaly will surface is only half a report. Watch the euro stablecoin total market cap, not the chain share. A move through a meaningful threshold — say, the low single-digit billions — would signal the sector is escaping its puddle and becoming a market. Until then, every concentration figure is a description of a small basin. I will be watching Dune and the standard on-chain aggregators for the aggregate, and I am more interested in the absolute number than the ratio. Watch issuer deployment announcements. Every new chain an issuer adds, and every chain it quietly leaves, invalidates the previous quarter's ratio. The 92% will move, and it will move because a business decided to move it, not because users did. Track the corporate decisions, not the chain leaderboards. Watch the dollar-euro stablecoin growth differential. If euro tokens are growing faster in percentage terms from a tiny base, that is a story — real convergence, however slow. If euro tokens are merely holding their small slice while the dollar complex compounds, then the euro market is not growing, it is idling, and the concentration figure is measuring the idle. Watch MiCA's execution details and supervisory guidance. Regulatory frameworks are not static documents. Every clarification from a European supervisor, every new interpretive note, every enforcement action against an issuer or a chain's compliance posture can reorder the market. The framework that created the 92% can rewrite it, and probably will. And watch the digital euro. Not the speeches — the pilot, the legislation, the timeline. The ECB's CBDC is the one variable that can render this entire analysis moot, either by absorbing euro stablecoin use cases directly or by pressuring the private sector into a narrower institutional lane. If the digital euro advances, euro stablecoins do not disappear. They retreat into the corners the central bank leaves open — DeFi composability, programmability, cross-chain settlement — and their market share becomes a function of what the central bank declines to do. That is the forward-looking question this whole report reduces to. Not which of Ethereum, Solana, or Base wins the euro stablecoin market. Whether a private euro stablecoin market survives the arrival of a public one. The 92% is a snapshot of a race running inside a room whose landlord has not yet decided whether to keep the room. Ninety-two percent is a number. It is not an answer. It never was. Ninety-two percent of what — and against whom — is the question the source did not ask, and the question any serious reader now has to. Code speaks. Contracts lie. In this case, the contracts are just euro tokens, and the lie is the framing around them — the tidy ratio that hides a small market, a licensing filter, a demand vacuum, and a central bank waiting in the wings. Traced. Logged. The next anomaly is not in the number. It is in what the next quarter's number quietly omits.