ESMA Leverage Cap — 8 Years On, What the Data Actually Says About 2018–2026

In August 2018 ESMA introduced product-intervention measures under Article 40 of MiFIR that became the most consequential single retail-protection reform in EU financial-services history. Eight years on, the data exists to answer the questions the original debate could only speculate about. This piece walks the evidence — what the cap achieved, what it did not, and what the next regulatory cycle should learn from the experiment.

What the 2018 intervention actually did

ESMA Decision (EU) 2018/796, effective from 1 August 2018, imposed a uniform set of retail-protection measures across all EU/EEA member states:

- **Leverage caps** by underlying: 30:1 on major currency pairs, 20:1 on minor pairs and major indices/gold, 10:1 on commodities and minor indices, 5:1 on individual equities, 2:1 on cryptocurrencies (added later by national competent authorities) - **Mandatory negative balance protection** on retail accounts, ensuring a client could not lose more than the deposited balance - **Margin close-out at 50%** of initial margin requirement — automatic position liquidation when account equity fell below the threshold - **Standardised risk warnings** including a mandatory percentage-of-retail-accounts-that-lose-money disclosure on every marketing surface - **Ban on monetary and non-monetary inducements** to retail clients (no bonuses, no rebates, no welcome credits)

The temporary intervention was extended by ESMA in 2019 and converted into national permanent measures by every member-state competent authority during 2019-2020. The framework remains substantively unchanged in 2026 with two notable national-level additions (the German BaFin ban on binary options for retail in 2019; the French AMF tightening on online forex marketing in 2021).

What the cap was meant to fix

The 2017 ESMA evidence-gathering exercise that preceded the intervention surfaced four headline concerns:

1. **Disproportionate retail losses.** Pre-intervention data from CySEC, BaFin and the FCA showed 74-89% of retail CFD accounts lost money over the prior 12 months, with median losses of EUR 1,600-2,800 per losing client and tail losses (worst 5% of clients) running into EUR 50,000+. 2. **Leverage-induced rapid account destruction.** The pre-2018 norm of 100:1-500:1 leverage on retail CFD accounts meant a 1% adverse market move erased the entire deposit. Brokers were marketing 200:1 and 400:1 leverage directly to retail clients with limited disclosure of the consequence. 3. **Loss-making clients funding broker revenue.** Internal broker data subpoenaed in several enforcement cases showed brokers earning the largest share of revenue from clients who lost their entire deposit within 6 months — an unhealthy commercial alignment between broker incentive and client outcome. 4. **Negative-balance exposure.** Pre-NBP, clients could and did end up owing the broker money after extreme market moves. The January 2015 EUR/CHF gap, which saw Alpari UK enter administration and several brokers pursuing clients for negative balances, was the defining case study.

What changed measurably after 2018

Eight years of post-intervention data is available from CySEC, BaFin, AMF, CONSOB, the FCA (which retained ESMA-equivalent rules under FCA PS19/18) and ESMA itself. The headline movements:

**Retail loss rates fell, but less than expected.** Mandatory risk-warning disclosures across the sector now report 74-89% of retail accounts losing money — within the same range as pre-intervention. The composition changed: the worst-tail losses (5th percentile and beyond) compressed substantially, because the leverage cap mathematically limits how fast an account can be destroyed. The median loss size dropped modestly. But the headline percentage of losing accounts has not moved materially. The implication is that the cap reduced the severity of bad outcomes without reducing their frequency. Trading-decision quality is the primary determinant of retail outcome; leverage is the amplifier.

**Broker revenue per active client fell sharply.** Industry estimates (drawn from public broker financial filings — Plus500 PLC, IG Group plc, CMC Markets plc — and CySEC sector reports) suggest revenue per active retail client fell 35-55% in the 12 months after 1 August 2018. The fall was driven by both lower trading volumes (clients trading smaller positions) and lower per-trade revenue (smaller spreads as brokers competed harder on the smaller volume pie). Broker EBITDA margins compressed and several mid-tier brokers exited the retail CFD segment between 2019-2021.

**Negative balance events fell to near zero.** The mandatory NBP requirement has been operationally effective. Post-2018 there has been no documented case in EU/UK retail forex of a client being pursued for a negative balance arising from a market gap. The Alpari UK 2015 case remains the last meaningful example. The infrastructure cost to brokers of operating NBP is real but bounded; the consumer-protection upside has been large.

**Offshore migration was substantial but not complete.** The largest unintended consequence. In the 12 months post-intervention, multi-jurisdiction brokers reported 15-35% of EU retail trading volume migrating to their offshore entities (Seychelles, Vanuatu, Mauritius, St Vincent, BVI). The brokers themselves disclosed the migration in capital-markets reporting. ESMA acknowledged the migration but argued — correctly, in our reading — that a client who actively chooses to route around the protection by signing a new agreement with an offshore entity has made an informed choice that regulation cannot fully prevent. The migration plateaued around 2021-2022 as awareness of the trade-off matured and as several offshore-licensed entities ran into operational trouble.

**The bonus and inducement ban has held.** Eight years on, EU/UK retail forex marketing surfaces are demonstrably cleaner than pre-2018. The "200% deposit bonus" and "trade USD 100,000 to claim USD 500" promotional patterns have disappeared from regulated entities. Offshore entities of the same brokers still run such offers but the EU retail surface is materially cleaner. This is the protection measure most directly visible to a retail-facing observer.

What the data does not support

Two pre-intervention predictions that proved overstated:

**The "leverage cap will kill the industry" prediction.** A common 2017-2018 industry argument was that the intervention would force broker consolidation, drive clients abroad, and ultimately destroy the EU retail CFD sector. Eight years later, the EU/UK retail CFD sector is smaller than its 2017 peak in absolute revenue but it remains a substantial and profitable industry. Plus500, IG, CMC Markets, and a long tail of CySEC-licensed brokers continue to operate at scale. The consolidation that did happen was largely the exit of weak operators that could not adapt to the lower-margin environment — arguably a positive outcome.

**The "clients will simply trade more to recover" prediction.** A counter-prediction was that retail clients, denied high leverage, would simply trade higher volumes to compensate, with net detrimental impact. The data does not support this. Average lot size and trading frequency per active retail client both fell post-intervention and have remained below pre-2018 levels.

What still hasn't been fixed

Three structural issues the 2018 intervention did not address, and which remain live in 2026:

**The 80% loss-rate floor.** No leverage-based or product-based intervention has been able to push retail loss rates below the high-70s. The reason is that trading-skill distribution is the dominant variable. Most retail CFD traders are net losing because they lack edge, not because they use leverage. The cap protects them from catastrophic destruction; it does not turn them into profitable traders. A second-generation intervention addressing financial-literacy and trader-suitability (more akin to the MAS [Customer Knowledge Assessment](/blog/mas-suitability-assessment-20-minute-test-retail-traders-cannot-pass-2026)) would address this lever but EU member states have not converged on a unified approach.

**The professional-account loophole.** EU retail clients can apply to be re-categorised as "elective professional" under MiFID II if they meet two of three criteria. Professional clients forfeit the leverage cap, NBP, and other retail protections. The re-categorisation framework is a legitimate regulatory feature but in practice it is used by experienced retail clients (and some inexperienced ones) to access higher leverage without leaving the EU regulatory perimeter. The volume of trading routed through elective-professional accounts is meaningful — broker disclosures suggest 8-15% of total EU retail-segment volume now sits on professional accounts. ESMA and national CAs have not tightened the re-categorisation criteria in any meaningful way since 2018.

**The offshore subsidiary path.** Most major multi-jurisdiction brokers maintain offshore entities that solicit EU residents through targeted marketing channels (search ads in languages other than the brand's primary market, paid social campaigns geo-restricted to specific EU countries). The marketing is technically not by the EU entity, but the brand is shared. EU residents who onboard through an offshore entity fall outside the EU framework entirely. ESMA has had limited tools to address this — the brand is the broker's commercial property, and the EU regulatory perimeter is narrowly defined as the legal entity holding the EU licence.

The numbers we would still like to see

Eight years on, four data sets remain harder to obtain than they should be:

1. **Comparable per-broker loss rates.** Broker disclosures all comply with the mandatory marketing-page percentage but the methodologies vary (12-month lookback vs trailing quarter; what counts as an "active client"; treatment of closed dormant accounts). A normalised comparable disclosure regime would let retail clients compare brokers on a true like-for-like basis. 2. **Migration rates to offshore entities by member state.** The offshore migration data is disclosed in aggregate by some brokers but rarely broken down by source EU country. A clearer picture would help member-state CAs target enforcement where migration is highest. 3. **Professional-account upgrade outcomes.** What proportion of elective-professional upgraders subsequently report losses larger than they would have experienced on retail terms? This is the key data point for evaluating whether the elective-professional regime is genuinely serving experienced traders or whether it has become a leverage-bypass mechanism. 4. **Outcome data by client tenure.** Loss rates by client account age (first 90 days, 90-365 days, 1-3 years, 3+ years) would let regulators target client education at the riskiest window. Anecdotally the worst losses concentrate in the first 90 days of trading; a published longitudinal study would confirm or refute the pattern.

What this means for the next regulatory cycle

Three implications worth surfacing as the MiFIR Review and the 2026-2027 ESMA work programme take shape:

**The leverage cap is not the right lever to push further down on.** Lower leverage caps would not materially reduce loss rates and would push more clients offshore. The instrument has done its work; further tightening returns diminishing benefit at rising cost.

**The next-generation intervention is on suitability and ongoing monitoring.** The MAS Customer Knowledge Assessment model — a structured pre-onboarding quiz that the client must pass to access leveraged products — is the most evidence-supported next step. EU member states would need to converge on a uniform CKA-equivalent for the regime to be effective; piecemeal adoption would create regulatory arbitrage. See [/blog/mas-suitability-assessment-20-minute-test-retail-traders-cannot-pass-2026](/blog/mas-suitability-assessment-20-minute-test-retail-traders-cannot-pass-2026) for the detail.

**The professional-account regime needs tightening.** The elective-professional criteria were calibrated in 2007 (MiFID I) and reaffirmed in 2014 (MiFID II). The financial-portfolio threshold of EUR 500,000 was material in 2007; in 2026 it is achievable for a meaningful share of EU households. Either the threshold should be uplifted to maintain its original calibration, or the other two criteria (significant trading activity, relevant industry experience) should be operationalised more strictly.

For pair-by-pair regulator detail see [/blog/cysec-vs-bafin-which-eu-regulator-better-protection](/blog/cysec-vs-bafin-which-eu-regulator-better-protection) and [/questions/what-is-cysec](/questions/what-is-cysec). For the offshore-entity picture see [/blog/why-cyprus-dominates-eu-retail-forex-tax-licence-economics-2026](/blog/why-cyprus-dominates-eu-retail-forex-tax-licence-economics-2026).

Risk warning

Trading CFDs and leveraged forex carries a high risk of losing money rapidly due to leverage. Between 74-89% of retail investor accounts lose money when trading CFDs. ESMA leverage caps reduce the speed at which an account can be destroyed but do not change the underlying probability that the trading strategy is profitable. Negative balance protection ensures you cannot lose more than your deposit on EU retail accounts but does not protect against the deposit itself.

*This article reflects ESMA Decision (EU) 2018/796 and the subsequent national-level permanent measures as of May 2026. The MiFIR Review is ongoing and may amend the retail-protection framework — verify the current rules on the ESMA and your national-competent-authority websites before relying on a specific figure.*