The first serious argument in most platform builds is about a single field. Someone wants a leverage number on the account, one value, easy to explain. Then the compliance officer points out that under the European retail regime a major currency pair and a single equity CFD do not carry the same ceiling, that gold sits in its own band, that crypto sits in the tightest one, and that the whole table only applies to clients categorised as retail. The single field dies about ten minutes into that conversation.
Leverage is a table keyed on three things
The keys are jurisdiction, asset class and client category. Drop any one of them and the configuration cannot express what a regulated firm actually has to do. In the EU, the retail limits introduced through ESMA product intervention and then adopted nationally are graded by asset class, with the loosest band on major currency pairs and progressively tighter bands as volatility rises. The United Kingdom operates an equivalent structure. Australia's regulator has used its product intervention power to impose a comparable set of retail CFD limits. Japan's retail FX margin rules under the Financial Services Agency are tighter still than most of Asia. Elsewhere, and in a number of offshore regimes, no statutory retail cap exists at all and the ceiling is whatever the firm's own risk policy and its liquidity provider will support.
We covered the country-by-country picture in leverage limits by country and the EU mechanism in ESMA leverage caps. What matters for the build is the shape: a matrix, versioned, with a date, not a number on an account.
Client category is the part firms get wrong
Higher limits for professional or elective professional clients are real in several regimes, and they are conditional. The conditions are set by the regulator, the firm has to assess them, evidence the assessment and warn the client what protections they give up. Treating the category as a switch a salesperson can flip is how firms end up with enforcement problems, because the file has to show why each client qualified. Professional client status and client categorisation rules both come down to the same discipline: the category is an outcome of a documented test, and the platform reads it rather than sets it.
Practically, the platform should never let leverage be raised on an account directly. It should let the account be assigned to a group, and the group carries the table. If someone wants a client on higher limits, the only way to get there is to change the client's category, which triggers the assessment and leaves a record. That constraint annoys sales teams and saves firms.
Leveraged trading carries a high risk of loss. Higher leverage does not improve a client's expected outcome, it shortens the distance to a margin close-out. Firms must take their own legal advice on which limits apply to them.
The close-out rule travels with the leverage
A leverage ceiling on its own does not describe the product. The margin close-out rule does the other half of the work: the percentage of required margin at which positions start being closed, whether it is applied per position or on the account as a whole, and how it interacts with negative balance protection. In the EU retail regime these come as a package, a standardised close-out level plus per-account negative balance protection. A platform that lets you set leverage per group but hard-codes one close-out level globally cannot implement that package for some clients and not others, which means it cannot serve two regimes at once.
What the configuration actually needs to hold
| Setting | Keyed on | Why it cannot be global |
|---|---|---|
| Maximum leverage | Jurisdiction, asset class, client category | Regimes band by asset class and category, not by account |
| Margin close-out level | Jurisdiction, client category | Some regimes prescribe a level for retail clients only |
| Negative balance protection | Jurisdiction, client category | Mandatory for retail in some regimes, absent elsewhere |
| Instrument visibility | Jurisdiction | Certain products are restricted or banned for retail clients in some countries |
| Risk warning text | Jurisdiction, language | Prescribed wording differs and some regimes require a firm-specific loss figure |
Weekend, news and instrument-level overrides
Statutory ceilings are maximums, not targets. A firm's own risk policy will often sit below them, and it should be able to move for reasons that have nothing to do with regulation: reduced leverage on a single instrument ahead of a scheduled announcement, tightened margin into a weekend, a temporary cut on a name that has become illiquid. The rule that keeps this safe is simple to state and easy to violate in code: an override may only reduce the effective ceiling, never raise it above the jurisdiction table. Build the resolution so the applicable maximum is the minimum of every layer, and the regulatory layer can never be overridden upward by a marketing decision or a support agent's fat finger.
The other half of that is timing. Changing margin requirements on open positions has an immediate effect on free margin, and a firm that tightens without notice can push clients into close-outs they did not expect. Firms that handle this well publish scheduled margin changes in advance and apply increases from a stated time, and they keep the record of what was published.
Prove what the setting was, not what it is
The question a supervisor or an ombudsman asks is almost never "what is your leverage". It is "what leverage applied to this client, on this instrument, on this date". If leverage lives as a mutable field on an account, the honest answer is that you do not know. If it lives as a dated version of a group configuration, the answer is a lookup. This is the same argument as compliance audit trails in general, and it is why we treat the leverage table in our Broker CRM as versioned configuration with an effective date rather than an editable value.
One position, stated plainly: a firm serving both a strict regime and an offshore book from a single global leverage setting is not running a simplified system, it is running an unprovable one. The extra table is a day of work. Reconstructing a year of margin decisions from trade records is not.
"Every firm that sets one global leverage number ends up with two problems: clients in strict regimes who are over the limit, and clients elsewhere who think the firm is uncompetitive. It is the same setting failing in both directions."
— Alex Onta, Executive Director, SINGUARD
Key Takeaways
- Leverage is a matrix keyed on jurisdiction, asset class and client category, never a single number on an account.
- Client category must be the output of a documented assessment, so the platform should read it rather than let anyone raise leverage directly.
- Margin close-out level and negative balance protection travel with the leverage table, because several regimes prescribe them together for retail clients.
- Overrides should only ever reduce the effective ceiling, and every table needs an effective date so past treatment can be proven.
Frequently Asked Questions
Can a firm offer higher leverage to a client who asks for it?
Only if the client falls into a category that the applicable rules allow higher limits for, and only after the firm has run and documented the assessment that regime requires. A client request is not a basis on its own, and the client must be told which protections they lose.
Does the same leverage table apply to every instrument?
No. Regimes that cap retail leverage band it by asset class, typically with the loosest limit on major currency pairs and tighter limits on volatile products. A platform that stores one ceiling per account cannot express that.
What happens to open positions when margin requirements are raised?
Higher requirements reduce free margin immediately, which can move an account closer to a close-out. Firms that manage this well announce scheduled changes in advance, apply them from a stated time, and keep the published notice on file.
About the Author
Alex Onta is an Executive Director at SINGUARD. He built eTrader, the terminal, the mobile apps, eTrader Broker, Copytrading, Business and Community, along with the worldwide clustered-server infrastructure it all runs on, with his brother Roman Onta helping on the design, and he leads that division today. Together with Roman he builds the Prop Firm CRM, the Broker CRM, Scalegram and CopySignals, and the two of them carry worldwide compliance, payment processing and international business structuring side by side. He lives and works in Dubai for most of the year. Meet the executive duo leading Singuard's five divisions.