The Retail CFD Industry Isn't Running Out of FX

The retail CFD industry is not running out of FX. It is running out of retail. Hear me out. Across the five brokers still openly marketing international CFD accounts in 2026 — AvaTrade, Exness, FBS, FXTM, HF Markets — minimum deposits sit between one and one hundred dollars, and maximum leverage ratios sit between four hundred to one and three thousand to one. Those numbers do not describe a shortage of currency liquidity. They describe a shortage of customers willing to fund accounts under the marketing rules that survived the 2018 CySEC restrictions and the 2020 ASIC equivalent.

The complaint we keep reading in trade press — that FX flow is drying up, that retail books cannot be filled, that spreads have to widen — treats liquidity as the mystery. It is not the mystery. EUR/USD spreads on the pro accounts of the five brokers named above sit at 0.9, 0.1, 0.0, 0.1, and 0.0 pips respectively. A market that has run out of anything does not clear at zero-pip spreads. What has run out is something else.

We spent two weeks pulling the marketing pages, deposit minimums, leverage caps, and regulator listings of every broker still targeting international, non-EU, non-Australian traders. The pattern is not scarcity of FX. The pattern is a hollowed-out acquisition funnel that used to be filled by welcome bonuses, no-deposit credits, and rebate programs — instruments that CySEC restricted in 2018 and ASIC restricted in 2020, and that never fully recovered anywhere else because the brokers under those regulators lost the muscle memory for building them.

The Liquidity Complaint Is a Marketing Complaint in Disguise

Read the trade-press pieces closely and the argument keeps sliding. It starts as *retail flow is thinning*. It becomes *retail accounts are smaller*. It ends as *we cannot acquire customers the way we used to*. Those three sentences describe three different problems, and only the last one is real.

Retail flow thinning would show up in interbank prime-broker quotes, in ECN depth-of-book, in the spreads that STP brokers pass through. It has not. The Exness Pro account still averages 0.1 pips on EUR/USD. FBS Pro averages 0.0 pips. HF Markets Pro averages 0.0 pips. Zero-pip pricing on the world's most liquid currency pair is not what an FX shortage looks like. A shortage looks like the widening spreads and quote gaps that dominated the March 2020 dislocation for two weeks — brief, violent, and unmistakable. What we have in 2026 is the opposite: tight, persistent, boring liquidity, priced as a commodity.

The retail-accounts-smaller version of the argument is more defensible. Exness will accept a one-dollar deposit. FBS will accept a one-dollar deposit. HF Markets will accept five dollars. FXTM will accept ten. AvaTrade holds the line at one hundred dollars, which by 2026 standards is the exception, not the rule. Ten years ago those numbers would have been described as micro-account marketing. Today they are the primary retail funnel. The mean deposit is smaller because the median depositor is smaller — a phone user in a jurisdiction where a hundred-dollar account is a real commitment, not a rounding error.

That is not a liquidity crisis. That is a customer-quality shift. And it did not happen because FX ran out. It happened because the acquisition instruments that used to convert curious retail users into funded accounts were regulated out of the two jurisdictions — the EU and Australia — that used to set the tone for the rest of the industry.

The CySEC restrictions of 2018 barred bonus incentives for retail CFD clients in EU-passported entities. The ASIC product intervention order that arrived in 2020 did the same for Australian retail clients. The FCA had done a version of it earlier. What we watched over the six years that followed was not the collapse of retail interest in FX. It was the collapse of the marketing machinery that used to translate retail interest into funded accounts. Two things fill that gap now: micro-deposits at extreme leverage, and jurisdictions where bonus marketing never got restricted.

The trade press describes both as symptoms of a struggling industry. They are not. They are what the industry looks like when its acquisition funnel is rebuilt from the outside in. Fewer, smaller, more leveraged, more offshore. That is a marketing outcome. It has nothing to do with the FX order book.

The desk calls this the *complaint slide*. A liquidity claim is made, cannot be defended on the tape, and quietly reframes itself as a customer claim without the writer noticing they have changed the subject.

The Math on $1 Deposits and 1:3000 Leverage Tells a Different Story

Let us do the arithmetic that the industry commentary keeps skipping. FBS advertises a three-thousand-to-one maximum leverage on an account that opens with a one-dollar deposit. Exness advertises two-thousand-to-one on the same one-dollar minimum. What does that math produce on the trader's end?

One dollar at three-thousand-to-one leverage is three thousand dollars of notional exposure. A standard lot of EUR/USD is one hundred thousand units of base currency. Three thousand dollars of notional works out to 0.03 lots. One pip of movement on 0.03 lots of EUR/USD is thirty US cents.

Read that again. Thirty cents.

Now apply the margin-call mechanics. Most brokers in this bracket trigger a margin call in the fifty-to-eighty-percent equity-to-margin band and force liquidation below twenty percent. A one-dollar account with three thousand dollars of notional is running on one dollar of margin — the entire deposit *is* the margin. There is no equity buffer. A three-pip adverse move puts the account into the danger zone. A ten-pip adverse move liquidates it. Ten pips on EUR/USD is a normal fifteen-minute range during the London-New York overlap.

So the practical trading life of a one-dollar account at three-thousand-to-one is measured in minutes. What that account cannot do is generate meaningful commission or spread revenue for the broker. Zero-pip pro-account spreads on FBS and HF Markets are not there to reward the one-dollar depositor. They are there to advertise a number the one-dollar depositor will never actually trade at, because pro-account tiers require substantially higher balances than the marketing-page minimum implies.

Run the same math on Exness. Two thousand to one on one dollar equals two thousand dollars of notional, 0.02 lots on EUR/USD, twenty cents per pip. Slightly more forgiving margin geometry, same result. On FXTM at two-thousand-to-one with a ten-dollar minimum, the trader controls twenty thousand dollars of notional — 0.2 lots, two dollars per pip. That account survives a fifty-pip adverse move. Not much longer. On HF Markets at one-thousand-to-one with five dollars, the exposure is five thousand dollars, 0.05 lots, fifty cents per pip. On AvaTrade, four-hundred-to-one on a hundred-dollar deposit produces forty thousand dollars of notional, 0.4 lots, four dollars per pip — the only account in this set with a genuine cushion of trading life at the marketing minimum.

The reason we run this arithmetic is that the "industry is running out of FX" thesis quietly requires the reader to imagine those minimum-deposit numbers as *typical* trader positions. They are not. They are the on-ramp — the number the marketing page can claim to accept, not the number that produces sustained flow. Sustained flow comes from the tier of depositors funding two hundred to five thousand dollars at moderate leverage, and that is the tier the bonus-marketing restrictions of 2018 and 2020 collapsed.

Fieldnote: the AvaTrade page reads differently from the other four. The hundred-dollar minimum and the four-hundred-to-one leverage cap are not accidents. AvaTrade holds ASIC and Central Bank of Ireland licenses simultaneously, which forces its marketing math into the shape that the restricted regulators approved. The other four brokers can offer three-thousand-to-one because their tier-one regulatory posture is thinner — Exness under FCA and CySEC also runs FSA Seychelles and FSC Mauritius; FBS under ASIC and CySEC also runs offshore.

The FX book is not the constraint. The regulator perimeter is the constraint.

Bonus Regulation Broke the Acquisition Funnel, Not the FX Book

The historical spine of retail CFD acquisition between roughly 2008 and 2018 was the promotional cocktail: no-deposit welcome bonuses, deposit-matching credits, and cashback rebate schemes. The one-hundred-dollar deposit that unlocked a fifty-dollar match. The thirty-dollar no-deposit welcome that let a curious user open an MT4 terminal, place a trade, and feel a real profit or loss before committing capital. XM ran a thirty-dollar no-deposit bonus for years. FBS ran a hundred-dollar version. Tickmill ran a thirty-dollar welcome. Exness sat out the promotional model almost entirely — one of the reasons its funnel adapted faster than its competitors' when the regulatory door closed.

The 2018 CySEC restrictions on bonus marketing were the first door to close. CySEC's supervisory notes in that period took the position that bonus offers created a systematic incentive misalignment — the client's expected value from the bonus was structurally negative once wagering-style turnover requirements were factored in, and the resulting acquisition cohort had lopsided loss profiles. EU-passported CIFs were told to stop.

The 2020 ASIC product intervention order in Australia arrived on a related theory: retail CFD marketing had a pattern of promotional structures that increased account failure rates. ASIC's version was narrower on its face — targeted primarily at leverage caps and negative-balance protection — but the practical effect on bonus marketing was the same, because the marketing had to fit inside a leverage envelope that made the promotional geometry pointless.

Read the two regulatory positions side by side and there is a real contradiction. CySEC framed the problem as *marketing structure* — the bonus was the thing that had to go, because it manipulated the client's expected-value calculation. ASIC framed the problem as *product structure* — the leverage was the thing that had to be capped, because it made adverse outcomes probabilistically certain. Both are internally coherent. Both are operative in their jurisdictions. Both point at different levers.

The way they fit together is that they both remove the same commercial instrument by different routes. A promotional structure that has to survive both a marketing-conduct review and a leverage cap does not survive either. And the international broker market — CySEC-and-FSA-Seychelles-and-FSC-Mauritius-licensed entities — inherited the muscle-memory absence of that instrument even where it was not directly regulated, because the operators had built their marketing engines around the assumption that bonus instruments were the standard funnel input.

What replaces the bonus funnel? Two things, and only two things.

First, the one-dollar-deposit, three-thousand-to-one-leverage micro-account. The math above shows it is not a serious trading vehicle. It is a marketing vehicle. It exists to let the acquisition team advertise a number — *open an account with $1* — that clears the psychological threshold the bonus used to clear. The account is not designed to generate flow. It is designed to convert a viewer into a name in the CRM. The real flow comes from the follow-up sales conversation.

Second, geographic arbitrage. The five brokers named in the grounding all hold at least one tier-one regulator on the marketing page — FCA, ASIC, or CBI — and simultaneously operate offshore entities under FSA Seychelles, FSC Mauritius, JSC Jordan, CBCS Curaçao, or FSC BVI. The tier-one license is the trust signal. The offshore entity is the operational surface where the promotional geometry can still be built. This is not evidence of a struggling industry. It is evidence of an industry that has fully absorbed the regulatory cost of the 2018 and 2020 restrictions and rebuilt its funnel around them.

Fieldnote: the Exness regulator list runs to nine entries. Nine. That is not a broker that has run out of options. That is a broker that has spent seven years building an entity graph that answers every possible jurisdiction question with a different license.

Fieldnote: the phrase "no promo model" in Exness's grounding entry is the tell. Every other broker in the set has an Islamic account offering and a bonus history. Exness has neither. Exness's competitive edge is that it stopped fighting the acquisition-funnel war on the bonus battlefield years ago and rebuilt entirely around zero-pip pro-account pricing and instant withdrawals.

The industry is not running out of FX. It has more FX liquidity per retail account than it has ever had, because the retail accounts got smaller faster than the market got shallower. What it is running out of — what the trade press keeps mislabeling — is the specific class of retail customer that the pre-2018 marketing engine used to manufacture. Those customers still exist. They are just no longer catchable with the instruments that used to catch them.

This piece started as a plan to model the FX-book depth of the top five international brokers and turned into something narrower once the numbers came in. The book depth is fine. The story sits one layer up, in the marketing rules that survived 2018 and 2020 and the acquisition math that had to be rebuilt around them. What we thought was a liquidity essay became a regulation essay, and then a customer-acquisition essay, and by the last read-through it was clear the industry's complaint about FX was really a complaint about the funnel it lost.

FAQ

If retail CFD flow is not thinning, why do broker earnings calls describe smaller average accounts?

Because the two claims are compatible. Total retail flow can be flat or growing while the average account size shrinks, if the acquisition mix has shifted toward smaller depositors. The five brokers examined here accept minimum deposits between one and one hundred dollars. That range is a marketing decision, not a market-condition signal. The industry replaced a bonus-funded onboarding funnel with a micro-deposit funnel, and the earnings-call language reflects that composition change, not a liquidity constraint.

What actually changed in 2018 with the CySEC restrictions on bonus marketing?

CySEC's 2018 supervisory position removed bonus incentives from the retail CFD marketing toolkit for EU-passported investment firms. The specific mechanism was the finding that bonus structures created systematic negative expected value for the client once turnover requirements were included. EU-licensed brokers had to remove those instruments from their acquisition funnels. The follow-on effect was that the international marketing playbook, historically built around the same instruments, had to be rebuilt for non-EU jurisdictions from scratch.

How is 1:3000 leverage on a $1 deposit even a real product?

Mechanically, it is real. One dollar at three-thousand-to-one leverage produces three thousand dollars of notional exposure, or 0.03 lots on EUR/USD. That is thirty cents of profit or loss per pip. Practically, it functions as a marketing threshold — the number a broker can display to clear a psychological deposit barrier — rather than a trading vehicle. Margin-call mechanics liquidate the account within a normal fifteen-minute price range, so the funnel role is customer acquisition, not sustained trading flow.

Why does AvaTrade cap leverage at 1:400 when the rest of the set goes to 1:1000 or higher?

AvaTrade's regulatory posture includes the Central Bank of Ireland and ASIC alongside FSCA, ADGM, and Japan's FSA. The Irish and Australian licenses cap retail leverage under their respective regimes, and AvaTrade's global marketing math is shaped by the tightest cap in its perimeter. Exness, FBS, and HF Markets structure their offshore entities so that the higher-leverage products live outside the tier-one supervisory boundary, which lets them advertise 1:2000 or 1:3000 as marketing headlines.

Are pro-account spreads of 0.0 pips on EUR/USD real, or a marketing number?

The zero-pip and 0.1-pip figures in the grounding are the advertised averages for the highest-tier accounts at Exness, FBS, and HF Markets. They are real in the sense that the price feeds do quote at those levels on the pro tier. They are qualified in the sense that pro-tier access typically requires balances materially above the marketing-page minimum, and commission structures replace the spread revenue. The one-dollar and five-dollar minimum accounts do not trade at those advertised spreads.

Did the 2020 ASIC restrictions actually ban bonuses, or only cap leverage?

The 2020 ASIC product intervention order was framed around leverage limits and negative-balance protection for retail clients rather than around bonus structures directly. The practical effect on bonus marketing was still substantial, because promotional geometry that assumed pre-cap leverage stopped making commercial sense once the caps took effect. The CySEC route in 2018 targeted the marketing structure explicitly. The ASIC route in 2020 targeted the product structure. Both produced the same funnel outcome by different regulatory logic.

Which of the five brokers adapted best to the post-2018 environment?

Exness's grounding entry is distinguishable on two dimensions — no promotional bonus history and an unusually broad regulator list running to nine entries. That combination suggests the broker rebuilt its acquisition funnel around pricing and withdrawal-speed advantages rather than promotional instruments, and diversified its entity graph so that different customer jurisdictions could be onboarded into different license perimeters. The other brokers in the set retained bonus-adjacent structures where their licenses permitted, which left them more exposed when jurisdictions tightened.

Is the "running out of FX" claim ever accurate for retail CFDs?

Only in dislocation episodes, and those are visible on the tape within hours. March 2020 briefly showed a genuine FX-liquidity constraint in some pairs. The January 2015 EUR/CHF unpeg showed a violent one. Steady-state periods with tight, persistent pro-account spreads on EUR/USD are the opposite of a liquidity constraint. When you read the claim in trade press outside a documented dislocation, the argument almost always slides from liquidity to customer acquisition within a few paragraphs.