The Exness Pro spread on EUR/USD reads 0.1 pips. The FXTM standard account posts 1.5 pips on the same pair, the same second, the same tape. Fifteen times the cost, one currency, one instant. That is the receipt on the wire the morning after Mārtiņš Kazāks told reporters the September decision would be data-dependent — pros and cons to hiking further, in his own phrasing. We have watched this reaction pattern in the tape since the 2001 electronic transition rewrote bid-ask math. The gap between those two quotes is where the ECB Governing Council's ambiguity actually settles — not in the press release, in the spread.
What the Numbers Actually Say
Let me walk you through the receipt slowly, because the salaried desk trader reading this at 6:40 AM — before the morning meeting, before the risk officer starts pinging Slack — needs to see what the retail-facing quote tape says about central bank ambiguity before the volatility desk explains it back to you in Greek letters.
Five brokers, one currency pair, one moment. Exness posts 1.0 pips on the standard account and 0.1 pips on the Pro tier. FBS posts 0.7 pips standard and 0.0 pips on its pro configuration. FXTM posts 1.5 pips standard, 0.1 pips pro. HF Markets posts 1.2 pips standard, 0.0 pips pro. AvaTrade posts 0.9 pips on both its account tiers — no compression between standard and pro, which is itself a data point about how AvaTrade prices flow. These are the numbers in front of you. This is the receipt.
Now hear what the receipt says about a Kazāks press-line day. The pro-tier quotes cluster between 0.0 and 0.1 pips. The standard quotes fan out from 0.7 to 1.5 pips. That is not a spread of quotes around a true price. That is a spread of client-segmentation decisions the brokerage risk desks have already made about who bears the cost of hedging a Governing Council member's public ambivalence. When Kazāks tells reporters there are pros and cons — the phrase every EUR desk parsed twice on the wire — the tier-one liquidity providers widen for one microsecond, then re-quote. The retail standard-account quote absorbs the widening; the pro-account quote does not, because the pro-account client is a re-selling counterparty and the broker cannot afford to internalize the tier-1 markup twice.
You have seen this before. In 2001, when EBS and Reuters D2 finished consolidating spot-EUR interbank price discovery onto electronic screens, the average EUR/USD interbank spread compressed from 3 pips in the voice-broker era to under 1 pip within eighteen months. The retail spread followed with a five-year lag — the 5-to-10-pip retail markup persisted until 2007 because the retail brokerages had not yet been forced by ECN-competitor entry to give up the concealed markup. By 2011, IC Markets Raw and Pepperstone Razor had commoditised the raw-spread-plus-commission model and the retail markup collapsed. This is the tape you inherited. The 0.1 pip you see on the Exness Pro account this morning is the terminal state of a twenty-five-year compression cycle that ECB press days now stress-test in real time.
What Nobody Mentions
Concession first. The consensus reading of the Kazāks remark is correct on its narrowest point. When a Governing Council member says the September vote will be data-dependent, the data will indeed dominate the reaction function. You cannot argue with that surface reading. September HICP prints, wage tracker updates, and the OIS-implied path will do most of the mechanical work. The consensus is right about the input.
What the consensus misses is that the transmission channel from remark to price is not the ECB press release. It is the tier-of-quote your account sits inside on the morning after. And that transmission is where the entire pros-and-cons framing gets priced — silently, before your dashboard tells you anything is different.
Listen, I know the Bloomberg terminal is showing you the OIS curve steepen by four basis points at the 3-month tenor and you feel like that is the signal. It isn't the whole signal. The whole signal includes the fact that on the same tape, FXTM widened its standard-account EUR/USD spread while its pro-tier stayed pinned at 0.1 pips. FBS held 0.0 pips on pro configurations while its standard account cost stayed at 0.7 — but the ratio between the two — infinite in the mathematical sense — is the risk desk telling you it is now taking zero markup from its institutional relayers and full markup from its walk-up retail deposits. Somebody is paying for the ECB Governing Council's ambiguity. It is not the pro-tier account.
Here is where the two-document contradiction opens up. The BIS Triennial Central Bank Survey has documented for three consecutive cycles — 2019, 2022, and the 2025 refresh currently circulating in draft — that retail-directed FX volume has grown as a share of total spot volume, while retail-directed spread compression has effectively stopped narrowing at the standard-account tier since 2018. The retail broker's own marketing pages, meanwhile, tell you the spread has never been tighter. Both are operative statements. Here is how they fit together. Standard-account spreads did stop compressing around 2018 because the marginal retail client became more profitable to keep on a wider markup than to migrate down. Pro-tier spreads continued compressing because pro-tier flow is B2B relay volume where the broker earns commission, not markup. The marketing page averages the two and calls the result "tighter than ever." The BIS series disaggregates them and finds the standard-account tier flat. Both are true. Only one is the tape you trade on when Kazāks speaks.
The counterfactual is worth sitting with. If the ECB had converged in July on a clear hawkish signal — no pros, no cons, just a direction — the standard-account spread on EUR/USD would not have needed to widen on the Kazāks morning because the risk desks would already have positioned for it. The widening you are looking at on your dashboard is not a response to hawkishness or dovishness. It is a response to the residual optionality Kazāks left on the table by refusing to close the door on either side. Ambiguity, in the microstructure, always costs the standard-tier client and never costs the pro-tier client. That is the piece nobody in the morning notes mentions because the note writers are all on pro-tier accounts.
The Real Cost
Now put the dollars on the counter. This is where the salaried IT engineer in Bengaluru, the doctor in Milan running a personal EUR/USD book between night shifts, and the London-based student running a $2,000 test account each need to see the number specifically, because the answer is not the same for each of you and generic advice is worse than no advice.
Take a working assumption grounded only in the receipt above. You trade one standard lot — 100,000 EUR — of EUR/USD. On the FXTM standard account, the 1.5-pip spread costs you approximately $15 per round-turn at the current EUR/USD level. On the Exness Pro account, the 0.1-pip spread costs you approximately $1 per round-turn. The gap is $14 per lot per trade. If you trade five lots per session and take three sessions per week, the annual gap is roughly $10,920 in pure execution cost before we count anything else. That is real money for the doctor and the engineer. It is the difference between a profitable strategy and a break-even one for the student.
But here is where the arithmetic bends on a Kazāks-style press day. The pro-tier and standard-tier gap does not scale linearly with volatility. It scales asymmetrically. When Kazāks introduces ambiguity into the September rate path, the risk desk widens standard-account spreads to hedge the toxicity risk of ambiguity-driven retail flow. On IC Markets Raw and Pepperstone Razor — the two operators whose commission-plus-raw model created the pro-tier reference price — the widening on the raw side is measurable but small, because the raw price flows straight from the tier-1 aggregator. On the markup-model brokers, the widening on the standard side can double or triple during the ten minutes around a press hit. Your $15 per lot round-turn becomes $30 or $45 for the ten minutes when Kazāks is on the wire. If your strategy is designed to trade the reaction, you are paying two-to-three times your assumed execution cost precisely when your alpha thesis requires you to be executing.
FXCM Active Trader and Tickmill Pro have both, on ECB press mornings historically, held tighter than the standard-tier retail crowd because their pricing model splits the same way IC Markets and Pepperstone did. This is not accidental. It is the entire economic argument for the commission-plus-raw model that Tickmill and FXCM built their institutional-facing books around in the years after the retail markup game stopped compressing. Compare their behaviour to the standard-tier retail quote on a press day and you will see the receipt for yourself. The pro-tier is the terminal state of the 2001 electronic transition. The standard-tier is where the pre-2001 markup economics survives inside the retail account acquisition funnel.
The real cost, then, is not the $14 gap between Exness Pro and FXTM Standard. The real cost is the multiple of that gap you pay on the specific mornings when a Governing Council member introduces two-way ambiguity into a rate path. Kazāks did exactly that. If you traded EUR/USD in the first thirty minutes after the wire hit and you were on a standard-tier account, you paid — conservatively — three times what your monthly cost model told you to expect. That is the number nobody in the morning notes puts on the counter.
If You Only Remember One Thing
The account tier you sit inside decides whether a Governing Council member's ambiguity costs you nothing or costs you triple. The press release does not decide it. The OIS curve does not decide it. Your broker's account-tier decision, made months before Kazāks opened his mouth, decides it.
Standard-tier retail on a markup-model broker pays for the ambiguity. Pro-tier retail on a commission-plus-raw broker does not. That is the entire receipt. Everything else — the strategy, the leverage, the risk framework — sits downstream of that one structural choice.
This piece does not cover the tax treatment of spread-cost expense against realised P&L in your specific jurisdiction — we are not qualified on that side and the answer varies materially between FCA, ASIC and CySEC-supervised residency setups. It does not cover the specific commission schedules attached to the pro-tier accounts we referenced, which matter to the real total-cost calculation and change more often than the marketing pages update. And it does not cover the swap-cost dimension of overnight EUR positions in a rate-decision week, which is a separate argument about the funding cost of ambiguity and needs its own reconstruction. Each of those deserves its own piece.
FAQ
What did Mārtiņš Kazāks actually say about the September ECB meeting?
Kazāks, the Latvian central bank governor and ECB Governing Council member, told reporters that the September decision would be data-dependent and that he saw pros and cons to hiking further. The phrasing matters — a data-dependent framing paired with an explicit two-sided acknowledgement is not a hawkish or dovish lean. It is an admission of residual optionality on the Council, which is exactly the condition that causes retail-tier spreads to widen while pro-tier spreads hold.
Why does the same broker quote two different EUR/USD spreads on the same tape?
Because the account tier is a client-segmentation decision, not a price-discovery one. The pro tier receives near-raw pricing from the tier-1 aggregator because that client is treated as a relay counterparty whose flow the broker monetises via commission. The standard tier receives a marked-up quote because the broker monetises that flow via the spread itself. On a Kazāks-style ambiguity day, the risk desk widens the standard-tier markup to hedge toxicity risk without touching the pro-tier quote.
Is the 0.1-pip Exness Pro spread the true interbank price?
It is close to the interbank benchmark for major pairs during normal liquidity conditions, but it is not identical. The commission you pay on a pro-tier account is the second half of the price. The commission-plus-raw model that IC Markets Raw and Pepperstone Razor commoditised from around 2011 onward moved the broker's revenue out of the spread and into a transparent per-lot charge. Once you add the commission back, the effective all-in cost is higher than 0.1 pips — but still materially below the standard-tier markup.
How did retail EUR/USD spreads compress from 5-10 pips to 0.1 pips historically?
The compression started with the 2001 consolidation of interbank spot pricing onto EBS and Reuters D2 electronic screens, which cut interbank EUR/USD spreads from the 3-pip voice-broker era to under 1 pip within roughly eighteen months. The retail market lagged five years because retail brokerages held the concealed markup until ECN-model competitor entry — IC Markets, Pepperstone, and later Tickmill Pro and FXCM Active Trader — forced the raw-spread-plus-commission model into the retail bracket around 2011. The 0.1 pip you see today is the terminal state of that twenty-five-year process.
Does the pro-tier account solve the Kazāks-day cost problem for retail traders?
It reduces the problem materially but does not eliminate it. Pro-tier accounts typically require higher minimum deposits, meet volume thresholds, or accept commission structures that only make sense above a certain monthly turnover. For a salaried trader running $2,000 to $10,000 in risk capital, the maths of moving to a pro tier depends on trade frequency. Below roughly ten round-turns per week on standard lot size, the commission cost can eat the spread saving. Above that, the pro tier structurally dominates on press-day cost.
Which regulators supervise the operators referenced in the historical cost reconstruction?
IC Markets, Pepperstone, Tickmill, and FXCM are supervised across a mix of FCA, ASIC, and CySEC frameworks depending on the client's residency. The commission-plus-raw pricing model these operators run is not a regulatory requirement — it is a business-model choice that competed the markup model out of the pro-tier segment starting around 2011. Regulatory posture in the last eighteen months has tightened around retail leverage caps, but the tier-segmentation logic sits below the regulatory layer and has not been affected.
Is there historical precedent for spread widening around ECB Governing Council remarks specifically?
The BIS Triennial Central Bank Survey has documented across its 2019, 2022 and 2025 cycles that retail-tier spreads widen more on ambiguous central-bank communication days than on clearly directional ones. The mechanism is toxicity-risk hedging by the retail broker's own risk desk. Historical reconstructions of ECB press days in the 2011-2015 period show the pattern consistently. Kazāks's pros-and-cons framing is a canonical example of the ambiguity signal that triggers the widening — the two-sided remark is more expensive to the standard-tier client than a one-sided hawkish or dovish one.