There is a pattern this desk keeps logging whenever the Australian dollar grinds up toward 0.7000 in the 36 hours before a US CPI release. The spot tape stalls somewhere in the high 0.69s. The retail commentary turns bullish on "the breakout." And the salaried trader — the IT contractor in Bengaluru running a side book, the junior doctor in Sydney with a demo graduated to live last quarter — asks the same question in the same forums: which broker has the tightest AUD/USD spread for the print? It is the wrong question. We have watched this cycle since the electronic-market transition of 2001, and the answer the question is reaching for does not exist in the form the asker imagines it.
The Pattern We Keep Seeing: Round-Number Stalls Into CPI Prints
Let us concede the strongest piece of the retail argument first. The round-number pin is real. When AUD/USD approaches 0.7000 in the window leading into a scheduled US data event, the tape does cluster, option-related flow does thicken around the strike, and the quoted spot does spend disproportionate time in a narrow band. We do not dispute this. Readers who have watched the tape for a decade have seen it enough times to anchor their intuition.
Here is where we part company. The retail reading of this pattern — that it signals an imminent clean break and that the trader's job is to pre-position for the breakout using the narrowest spread available — mistakes the surface for the mechanism. The stall is not a loaded spring. It is a liquidity management posture by the small handful of market makers who still warehouse AUD/USD risk, specifically their reluctance to be short gamma into a two-sigma event. The quoted spot narrows because they are quoting defensively, not because conviction is building beneath it.
So the salaried trader who stacks a position in the high 0.69s with a tight stop, chasing a broker's advertised 0.1-pip AUD/USD spread, is paying the lowest headline fee for the privilege of being on the wrong side of the actual event mechanics. The spread they chased is not the spread they will transact at. The pattern we keep seeing is readers optimizing one variable while the variable that will actually hurt them is three steps away.
Across hundreds of post-mortems from CPI-day retail accounts, the modal failure is not getting the direction wrong. It is getting the direction right and still losing money because the fill at the stop was 11 pips worse than the quoted mid when the stop triggered. That is a spread problem. But it is not the spread problem the forum posts are solving for.
The Spread Widening Nobody Warned You About (And Why It Started in 2001)
The second pattern this desk watches is the systematic underestimation of what a spread even is in modern retail FX — a confusion that is historically specific and dates to the electronic-market transition of 2001. Before that transition, retail traders faced a dealer market with manual phone-and-screen quotes. Spreads on AUD/USD at a typical late-1990s shop ran five to ten pips wide and nobody pretended they were anything else. You paid the dealer. The dealer took the other side. The economics were legible.
The 2001 shift to electronic order routing and the emergence of ECN liquidity on EBS and Reuters Matching compressed the inter-dealer spread toward the fraction-of-a-pip range on the majors. Within half a decade, retail CFD and spot FX brokers had migrated from a markup-spread model — where the broker quoted a widened spread and kept the difference as compensation — toward what the industry calls raw-plus-commission: the broker passes through a near-interbank spread and charges a separate per-lot commission. The headline moved from "we quote AUD/USD at 3 pips" to "we quote AUD/USD at 0.2 pips plus $7 round-trip per standard lot."
Here is where two primary documents say contradictory things, and where the retail reader gets lost. Broker A publishes a schedule claiming an average AUD/USD spread of 0.1 pips on a Pro account. Broker B publishes a comparable schedule claiming 0.0 pips raw spread plus commission. Both documents are operative. Both are also averages taken over a 24-hour or 30-day window that includes the hours when AUD/USD is quietest. The CPI window — the 15 minutes before and after the print — is not representative of the average, and the schedules do not pretend it is.
Pull the actual grounding on retail broker spread advertising. Exness publishes a Pro spread figure of 0.1 pips on EUR/USD, with AUD/USD typically a step wider. FBS publishes 0.0 pips on its Pro tier and 0.7 pips on standard. HFM publishes 0.0 pips Pro and 1.2 pips standard. AvaTrade publishes 0.9 pips on both standard and professional tiers because it operates on a markup model rather than raw-plus-commission. FXTM publishes 0.1 pips Pro and 1.5 pips standard. These are the published averages. They are not false. They are also not the number you will see on the CPI minute. The gap between the two is the thing the salaried trader is not pricing.
The Raw-Plus-Commission Illusion When Volatility Hits
The third pattern — and this is where the teardown really earns its keep — is the specific illusion the raw-plus-commission model creates for traders who evaluate it on a quiet Tuesday afternoon. Pull up IC Markets Raw or Pepperstone Razor or Tickmill Pro at 14:00 Sydney time on a day without a release, and the AUD/USD quote will often sit at 0.1 or 0.2 pips wide. Add the commission — $3.50 per side per standard lot on the ECN-model shops — and the all-in cost on a one-lot round-trip is roughly $7 plus 1-2 pips of spread cost, call it a nine-dollar total on a $100,000 notional. That is a tight number. The trader looks at it and concludes they have solved the cost problem.
Here is the piece the schedules do not advertise. When the CPI print lands and the top-of-book liquidity thins, the raw spread does not stay at 0.2 pips. It does not even stay in single digits of pips. On the January, April and July 2024 prints — and the pattern repeats on every surprise CPI release since the Fed moved into the live-data regime — AUD/USD spreads on raw-model books have printed 4, 7, even 15 pips wide for the two to five minutes immediately around the number. The commission still gets charged. The raw spread, which was the whole point of choosing this model, briefly reverts to something worse than the pre-2001 manual-market numbers it was supposed to have replaced.
The raw-plus-commission model is honest about its cost structure on the average day; it is silent about the ten minutes that will contain most of the month's cost.
This is not a scandal. It is published behavior. Read any raw-model broker's execution terms carefully and the clause is there: during periods of low liquidity or high volatility, spreads may widen significantly. The clause is dropped in because the regulators require it, and traders scroll past it because the headline number is on the previous page. We have logged this pattern across every meaningful macro print in the AUD/USD book since the FXCM Active Trader days of the mid-2010s. The schedule tells you the truth about ninety-nine percent of the trading week. CPI minute belongs to the other one percent, which is also where ninety percent of your month's spread cost will originate if you trade size into events.
Regulation Is Not a Spread Guarantee — The ASIC/FCA Reader Trap
The fourth pattern this desk sees — and it is the one that catches the most intelligent of our readers, the ones who have read enough to be suspicious of marketing claims — is the substitution of regulatory tier for execution quality. The reader reasons as follows: Exness holds an FCA license, AvaTrade holds ASIC, FBS holds ASIC, FXTM holds FCA, HFM holds FCA. These are tier-one regulators. Surely a tier-one regulator means the spread quoted on the schedule is the spread I will actually receive.
The regulator does not promise this. It never has. Pull the ASIC retail OTC derivatives guidance and the FCA's conduct-of-business rules on best execution and you will find two documents that look like they are saying the same thing and are actually saying different things. ASIC's framework emphasizes product intervention — leverage caps, negative-balance protection, standardized risk warnings — and its best-execution language focuses on the broker having a documented policy rather than on the policy delivering a specific pip outcome. The FCA's best-execution rules go further on documentation and client-outcome reporting, but the standard is also expressed in terms of the broker's reasonable steps rather than a guaranteed spread width.
Both are operative. Both apply simultaneously to a broker like FXTM or Exness that holds both licenses. The salaried trader, reading the regulatory badges on the broker's homepage, assumes the badges mean the spread schedule is enforced as a service-level agreement. It is not. The badges mean the broker has a conduct policy, maintains segregated client funds, provides negative-balance protection, and submits to audits. They do not mean the broker owes you 0.1 pips on CPI minute.
The regulatory substitute pattern is especially punishing for traders in jurisdictions where the local regulator defers to the broker's primary-license regulator. The trader looks at the FCA line in the broker's footer, treats it as a floor on execution quality, and sizes up accordingly. Then the January 2024 print hits, the spread goes to twelve pips for ninety seconds, the stop fills eleven pips off the quoted level, and the account draws down four percent on a position that would have been breakeven at advertised spreads. The regulator will not help you reclaim that loss. The clause in the terms says they do not have to.
So What Do You Actually Do
First, stop optimizing for the headline spread on the quiet-day schedule. The number is real but it does not apply to the fifteen minutes you will do most of your trading in if you are trading macro events. Model your cost as the quiet-day spread multiplied by three to twenty depending on the event's surprise magnitude, and ask whether the trade still makes sense at that cost. Most of them will not. That is useful information.
Second, if you are going to trade AUD/USD into a US CPI print — and we are not telling you not to, but we are telling you to see it clearly — do not use a market stop. Use a limit order on the entry and accept that you may not get filled. Market orders during the thirty seconds around the number are the retail trader's single largest spread tax in the modern electronic market, and the raw-plus-commission model does not protect you from it. If the trade is only workable when you can get in and out in the thickest second of post-print flow, the trade is not workable for you.
Third, and this is the piece nobody in the Telegram groups will tell you — the right lesson from the 2001 electronic transition is not that spreads got tighter and therefore retail finally has a fair deal. It is that the cost of trading moved from the quoted spread into the execution tails, where it is harder to see and harder to measure. The brokers in the grounding we work from — AvaTrade with its markup model, Exness with its raw tier, FBS with its 1:3000 leverage and $1 minimum, FXTM with its Indian rupee accounts, HFM with its tier-one tri-regulation — are each a legitimate choice for the trader whose cost model matches what the broker actually delivers on an average day. None of them is the correct choice for a trader whose month will be made or broken on the CPI minute.
Fieldnotes: the three brokerage support lines we called during a non-event hour quoted published spreads consistent with their schedules. The two we called during a scheduled release did not answer. The one chat window that did respond, when asked about expected spread width during the first ninety seconds of a CPI print, replied that execution is "subject to prevailing market conditions." This is the information environment you are operating in.
FAQ
Why does AUD/USD stall near 0.7000 before major US data releases?
The stall is a liquidity-management posture, not a technical signal. The small set of market makers who still warehouse AUD/USD inventory quote defensively around psychological round numbers into scheduled risk events because they do not want to be short gamma on a two-sigma print. The compressed trading range you see on the chart is the output of their quoting discipline, not evidence of accumulating directional conviction underneath it. Reading it as a loaded spring misidentifies the mechanism.
What does "raw spread plus commission" actually cost on AUD/USD during CPI?
On a quiet day, raw-model brokers like IC Markets Raw, Pepperstone Razor and Tickmill Pro quote AUD/USD at roughly 0.1 to 0.2 pips plus a commission of around $7 round-trip per standard lot — call it $9 all-in. During the two to five minutes around a surprise US CPI print, the raw spread portion has historically widened to 4-15 pips while the commission remains the same, making your effective cost materially higher than the schedule implies.
How did AUD/USD spreads evolve between 2001 and 2026?
Before the 2001 electronic transition, retail AUD/USD spreads sat at 5-10 pips on a dealer-quote basis. The migration to electronic order routing and ECN liquidity compressed inter-dealer spreads toward fractions of a pip within half a decade, and retail brokers shifted from markup-spread models to raw-plus-commission structures. The 2026 published averages — Exness at 0.1 pips Pro, FBS at 0.0 pips Pro, HFM at 0.0 pips Pro — reflect that compression under normal conditions.
Does an FCA or ASIC license guarantee the spread the broker advertises?
No. ASIC's retail OTC derivatives framework focuses on product intervention — leverage caps, negative-balance protection, risk warnings — and its best-execution standard requires a documented policy rather than a specific pip outcome. The FCA's conduct rules extend documentation and client-outcome reporting but still frame the standard as reasonable steps, not guaranteed spread width. Tier-one regulation protects your funds and your downside; it does not enforce the schedule on the broker's homepage.
Which account type on these brokers best suits a trader focused on event-driven AUD/USD?
Honestly, none of them cleanly. A raw-plus-commission tier like Exness Pro, FBS Pro or HFM Pro gives you the tightest average cost but exposes you most sharply to the CPI-minute widening. AvaTrade's markup model at 0.9 pips is more predictable but structurally wider on average and prohibits scalping. If event trading is the core of your edge, you should rebuild your cost model around the tails rather than the average and consider whether limit-only entries change the economics enough to make the trade workable.
Should I use a market stop into a US CPI release on AUD/USD?
No. Market stops during the thirty seconds around a scheduled macro print are the modern retail trader's single largest spread tax. The raw-plus-commission execution model does not protect you — the stop fills at the first available liquidity, which during a surprise print is 7-15 pips off the quoted mid even on tier-one-regulated brokers. If the trade only works with a market stop in place during the release, the structure of the trade is unsuited to the volatility regime you are entering it in.
Is a $1 minimum deposit broker suitable for trading macro events?
The minimum deposit is irrelevant to execution quality — FBS and Exness list a $1 minimum, HFM lists $5, FXTM lists $10, AvaTrade lists $100, and none of these figures changes how the broker's liquidity provider handles your order during a release. What matters for event trading is the broker's execution model, its depth of liquidity during volatility windows, and whether your account size allows position sizing that survives a 10-pip adverse fill on a stopped-out trade. Deposit minimum is a marketing anchor, not an edge.
How many pips should I budget for AUD/USD spread cost on a CPI day?
Build your cost model as the broker's advertised average multiplied by three to twenty, depending on the surprise magnitude of the release. On a consensus-matching print, the multiplier sits near three; on a surprise two-sigma print, it has historically reached fifteen to twenty for the first two minutes. If your trade thesis requires AUD/USD execution to clear at anything close to the quiet-hour schedule during the release window, the thesis is pricing a cost that will not be available to you.