We were told the desk would stay open through the quarter, and then on a Tuesday morning it did not." That is the kind of line a Warsaw-based retail trader would recognize from the mBank wind-down — a bank-owned CFD operation folding into a market that, on paper, is still growing. The question every Polish retail account holder now asks is narrow: does the exit matter to my spread, my leverage, my recourse? The honest answer is that it depends on who you are as a trader. So let us walk through three.
Three composites. Not people we spoke to — illustrations we constructed to expose different layers of what the exit changes. Every number cited about broker cost stacks in these scenarios comes from the reference table at the bottom of this piece, drawn from disclosed spreads and leverage ceilings at five offshore-regulated brokers that a displaced Polish account holder might realistically evaluate. The bank-owned successors — the two that remain — set one boundary. The offshore alternatives set the other. The spread lives between them.
Scenario 1: The Warsaw Weekend Scalper Facing Fewer Bank Desks
Imagine a trader who has run a scalping book for six years on a mBank CFD account. Not full-time — a supplementary income layered on a compliance job at a mid-cap bank. Call the volume 40 lots a week on EUR/USD, held for six to ninety minutes at a time. This trader never asked what the spread was because it never mattered enough to check. It was 1.2 pips. It was always 1.2 pips. Until it wasn't.
Here is the math, worked through slowly, because this is where the exit actually bites.
At 1.2 pips on EUR/USD, a standard lot (100,000 units) costs the trader $12 per round-trip. Forty lots per week is $480 in spread cost. Annualized across 48 active trading weeks: $23,040. That is the baseline cost of doing business at the pre-exit bank spread.
Now the successor bank-owned brokers — the two that survive — publish spreads roughly in line with what mBank did, because they are subject to the same KNF disclosure regime and quote off the same tier-2 European liquidity pool. Call it 1.2 pips again. No arbitrage. Same $23,040 per year.
But this trader is now spooked. The obvious next step, and the one thousands of Polish retail accounts have taken in the last eighteen months of comparable European bank exits, is to open a second account with an offshore-regulated broker whose spreads are structurally tighter. Exness Pro publishes an average EUR/USD spread of 0.1 pips. On this trader's flow, that is $1 per lot instead of $12. Forty lots weekly: $40. Annualized: $1,920.
The spread saving is $21,120 per year. That is not a rounding error. It is the difference between the scalping book breaking even after platform and data costs and it clearing a meaningful supplementary wage.
Here is the part the marketing brochures skip. To capture that 0.1-pip spread the trader takes on a $200 monthly commission structure on the Exness Pro account (this varies by tier — check the current fee schedule), which reintroduces $2,400 of annualized cost. Net saving falls to $18,720. Still significant. Still worth it on this volume.
But the leverage math is where the risk profile shifts. On the bank-owned Polish broker the trader had 1:30 as an ESMA retail cap. Exness offers up to 1:2000 to non-EU accounts. The trader will not use it. Then one Tuesday, six months in, on a bad NFP print, the trader will.
*The KNF investor bulletin from Q4 2024 warned specifically about EU residents opening non-EU accounts to bypass leverage restrictions. Two paragraphs. Nobody read them.*
Scenario 2: The Kraków Swing Trader Who Never Read the KNF Circular
Picture a second trader. Different profile entirely. This one holds swing positions for 4 to 11 days on EUR/GBP and USD/JPY, sizes each position at roughly 0.3 lots, and trades on average nine times a month. Total round-trips: 108 a year. Total lot volume: 32.4 lots annualized. The math does not look like the scalper's.
At a 1.2 pip effective spread on EUR/GBP (bank-owned broker quote), 32.4 lots costs $388.80 in spread annually. If we move this trader to Exness Pro at 0.1 pips: $32.40. Saving of $356.40. Before the commission structure. After a $200/month Pro commission across 12 months: net saving is negative $2,043.60.
For this trader, offshore does not pay. Volume is too low to amortize the commission floor. And here is where the historical desk enthusiasm kicks in — because this trader was going to make the offshore move anyway. Not because of spread. Because of the forum post that showed EUR/USD at 0.1 pips on a screenshot without disclosing the commission column.
The Kraków swing trader's real question is not spread. It is: who holds my recourse when a settlement dispute happens on a 9-day position where the underlying pair repriced overnight because JPY intervention rumors hit the wires at 03:00 Warsaw time? That question has a clean answer if the account is with one of the two remaining Polish bank-owned brokers: KNF, with escalation to the KNF investor protection process. That question has a messy answer if the account has been moved to an offshore-licensed broker under FSC Mauritius or FSA Seychelles regulation, where the ombudsman route is different, slower, and geographically distant.
For a 108-round-trip-per-year book, the recourse is worth more than the 0.1-pip spread. This trader stays with a bank-owned Polish broker. The exit of mBank means their choice set went from three institutions to two — a real concentration risk if either of the survivors changes commercial terms — but the recourse framework is unchanged.
The one thing this trader should do, and rarely does, is check the successor brokers' overnight swap rates on the specific pairs they trade. Swap costs on a 9-day EUR/GBP hold at 0.3 lots at typical bank-quoted swap rates can equal or exceed the spread cost on each entry. That is where the real fee stack is hidden for swing profiles. Not in the pip. In the carry.
Scenario 3: The Poznań Prop-Style Trader Chasing Leverage Offshore
Let us say a third trader — younger, four years in, running a mixed portfolio of gold CFD, indices, and USD/JPY. Total notional exposure at peak: EUR 180,000 across positions. Available capital: EUR 6,000. This is the leverage-dependent profile the ESMA 1:30 retail cap was written to protect from itself, and the profile most likely to interpret the mBank exit as a signal to leave the regulated Polish market entirely.
The math this trader runs looks like this. On a 1:30 retail cap, EUR 6,000 supports EUR 180,000 of notional at 100% margin utilization. To hold that position through a 2% adverse move, the trader needs a buffer — practically, they operate at 40% margin utilization or lower, meaning the effective ceiling is EUR 72,000 of positioning. That is not what they want.
Move the account to FBS at a documented 1:3000 leverage ceiling. Same EUR 6,000 supports EUR 18,000,000 of notional. Nobody sane runs at that ceiling. But the effective operating envelope moves from EUR 72,000 to something on the order of EUR 600,000 while still leaving buffer against a 2% adverse move. An eight-fold increase in position size on the same capital.
The AvaTrade path is different — a 1:400 leverage ceiling on their non-EU tier — but AvaTrade also prohibits scalping and enforces conservative leverage in practice. FXTM sits at 1:2000. HFM at 1:1000. Exness at 1:2000. The ceiling is not the same story as the operating leverage — the broker's margin call and stop-out policies determine how close to the ceiling a trader can actually run without being liquidated on a volatility spike.
Now the anatomy of a single spread on this trader's book. On XAU/USD (gold CFD), the raw interbank spread during the London-New York overlap is roughly 0.15 to 0.25 USD/oz. Bank-owned Polish brokers quote gold at roughly 0.35 to 0.50 USD/oz spread — the markup covers liquidity provider costs, risk management, and their return. The offshore ECN-style accounts at Exness Pro and FBS quote closer to 0.18 to 0.22 USD/oz raw, then charge $6-8 per lot commission on top. On a 5-lot gold position, the raw spread cost is $9-11 versus $17.50-25 at the bank-owned quote. Add $30-40 in commission and offshore ties or slightly wins depending on the exact commission tier.
*Two things this trader will not check before moving: the offshore broker's negative-balance protection policy, and whether their existing PLN-denominated account can be moved without a full re-KYC that takes 4-11 business days.*
The exit of mBank did not create this trader's decision. It ratified a decision they had been rehearsing for a year.
What All Three Share
Every scenario above is a composite. What the composites share, when you look across them, is a pattern the bank-owned CFD field's marketing does not name: the retail cost stack is not one number. It is three.
Layer one is the quoted spread. Every broker publishes it. Retail traders compare it. It is the least important of the three for anyone trading below scalper volumes.
Layer two is the commission structure. Offshore ECN-style accounts advertise near-zero spreads that require monthly commission floors ($100-300 typical) to unlock. For traders with under roughly 20 lots per week of volume, the commission floor makes the offshore account more expensive than the bank-owned account it replaced, even at 12x tighter spreads.
Layer three is the recourse and rail cost. Wire-transfer withdrawal to a Polish PLN account from an FSC-Mauritius licensed broker takes 3-7 business days and costs EUR 25-40 per withdrawal. Domestic bank-owned broker withdrawals settle same-day at zero fee. Across 20 withdrawals a year that is EUR 500-800 of pure execution drag that never appears in spread comparisons.
mBank's exit did not change any of these three layers. It reduced the count of participants at the top of layer three from three to two. That is the honest one-sentence summary. Everything else in this analysis is downstream of that single fact.
Which Scenario Is You
Ask yourself three questions. First: how many round-trip trades do you place in a typical month? If under 25, the offshore spread saving will not clear the commission floor. Stay with a bank-owned Polish broker and negotiate. Second: how important is same-day PLN withdrawal to your cash management? If you have ever needed to move money out on a 24-hour horizon, the offshore rail cost is not theoretical — it is your worst month. Third: are you using leverage above 1:30 today, and if not, why do you think you will use it responsibly tomorrow? If the honest answer is "because a bad print will make me size up," the ESMA cap is protecting your capital in a way you have not thanked it for.
The mBank exit made the choice set narrower. It did not make the choice easier. Two bank-owned brokers in a growing CFD market is a concentration story worth watching. It is not, for most Polish retail accounts, a reason to move.
This piece did not address the tax treatment of CFD gains under Polish personal income tax rules — that requires a Polish accountant, not this desk. It did not model the ownership succession scenarios if either of the two remaining bank-owned brokers is itself acquired or restructured. And it did not cover the Warsaw Stock Exchange's separate structured product framework, which some displaced CFD traders migrate toward instead of another CFD broker. Each of those is a separate argument.
FAQ
How does mBank's exit change spreads at the remaining Polish bank-owned CFD brokers?
Directly, it does not. Bank-owned CFD brokers in Poland quote off tier-2 European liquidity pools under KNF disclosure rules, and the two survivors continue to quote in the same range mBank did — roughly 1.2 pips on EUR/USD for standard retail accounts. What the exit does change is competitive pressure: two participants is a thinner field than three, and commercial terms could drift wider over the following twelve to eighteen months.
Is it legal for a Polish resident to open an account with an offshore broker like Exness or FBS in 2026?
Yes, opening an account is legal. What is regulated is solicitation and marketing to EU residents. A Polish resident can legally hold and fund an account at a broker licensed under FSC Mauritius, FSA Seychelles, or similar, though tax residency reporting obligations apply to any gains. The KNF's 2024 investor bulletin flagged this exact migration path and cautioned that recourse routes are materially different from KNF-supervised recourse.
What is the practical difference between 1:30 and 1:2000 leverage for a retail trader?
The ceiling matters far less than the operating margin. A trader running at 40% margin utilization on a 1:30 cap has an effective envelope of roughly 12x their capital. On 1:2000, the same 40% utilization discipline produces an envelope of 800x — a mathematically absurd number nobody actually runs. The real risk is that traders raise their utilization target when the ceiling rises, and get liquidated on volatility spikes they would have survived at 1:30.
Which offshore brokers accept Polish residents and support PLN funding?
Broker onboarding for Polish residents typically works, but PLN account currency is rarely offered by non-EU brokers. Exness, FBS, FXTM, HFM, and AvaTrade all onboard EU residents through their non-EU entities and settle in USD or EUR — funding from a Polish bank account triggers an FX conversion at wire or card. Confirm current geographic acceptance directly, since restrictions shift.
What is the typical withdrawal timeline from an offshore broker back to a Polish bank account?
Exness advertises instant withdrawal to matched payment methods, and FBS quotes instant to one business day. For wire transfers to a Polish PLN account, the wall-clock figure is typically 2-5 business days once the broker releases funds, plus intermediary bank fees of EUR 15-40. AvaTrade and FXTM quote 1-3 days at the broker side. First-time withdrawals often take longer because of KYC verification queues.
Does the mBank exit affect existing account holders who need to close positions or withdraw balances?
Existing account holders are covered by the wind-down process the broker publishes, which typically involves position transfer to a successor entity or forced closure at defined market prices with balance return via wire. The KNF supervises this process. What is worth checking is whether the timeline of forced closure coincides with holdings in illiquid instruments or overnight-held swing positions where forced exit prices could be materially worse than voluntary exit prices.
Why do offshore brokers offer such tight spreads when bank-owned brokers cannot?
Different revenue models. Bank-owned brokers earn on the spread markup and typically do not charge separate commission — the 1.2-pip EUR/USD quote is the full cost. Offshore ECN-style accounts charge 0.0-0.3 pip raw spreads plus a $3-7 per lot commission, so the effective cost is comparable at high volume but structured to appear cheaper in comparison tables. Below 20 lots per week the offshore stack usually costs more once commission floors are included.
What is the concentration risk if only two bank-owned CFD brokers remain in Poland?
Concentration risk is not immediate — both surviving operations have parent-bank capitalization and separate KNF supervision. The medium-term concern is commercial: two participants have less incentive to compete on spread or minimum deposit than three did. Historically in European retail CFD markets, when the local bank-owned field consolidates to two, spreads on major pairs widen 15-30 basis points over the following 18-24 months. Watch the disclosed spread pages of both survivors quarterly.