A-Book Brokers Face Underestimated Risks, Adviser Warns

A strategic adviser specialising in financial services has argued that the A-book model, widely regarded as a safer approach to order handling in FX and CFD broking, carries a series of underappreciated risks that can prove as costly as those associated with B-book operations.
 

Anya Aratovskaya, Strategic Advisor

Writing as a guest contributor for LiquidityFinder, Anya Aratovskaya, Strategic Advisor, sets out five risk categories she says are routinely overlooked by brokers operating or transitioning to an agency model.

Aratovskaya opens by addressing a commonly held industry assumption, framing her argument with a principle she attributes to risk and compliance officers: risk can be reduced and reallocated, but never eliminated. She then challenges the notion that an A-book confers any inherent operational safety:

Check out the full article here

Anya Aratovskaya, Strategic Advisor commented:

"A-book is not a proxy for a good company, a sound strategy, or competent risk management. Any book can be mismanaged. Each one is only as good as the discipline behind it. On that note, I want to elaborate that the idea that running an A-book will make you sleep better is overrated."

The first risk she identifies centres on the mismatch between the leverage brokers offer clients and what prime brokers are willing to extend. Retail clients typically seek leverage ratios of 1:200 and may hold more than 500 lots over a weekend, while prime brokers often limit exposure to 1:30 on cross-currency pairs.

Anya Aratovskaya, Strategic Advisor also said:

“To stay competitive, you're offering leverage you don't actually have. That mismatch has to be funded by your balance sheet. This is why 'decent' LPs are usually backed by massive banks or hedge funds; they need the deep pockets to bridge that gap. Add in the lack of cross-margining (still an issue in 2026, somehow) and sudden weekend margin hikes, and your 'safe' A-book is just a different flavor of over-leveraged gamble.”

She also made the following comments:

The second area concerns clients who meet the legal definition of institutional counterparties but operate with retail trading expectations, a tension Aratovskaya describes as a favourite industry paradox.

"The client who is 'Institutional' on paper… but has a $500 retail brain. They legally deserve the label, but they demand 1:500 leverage and rage over 0.1 pip slippage during a news event. You're stuck providing white-glove service to people who have zero tolerance for how real markets actually function. Side note: LPs secretly love these guys - they trade institutional volumes with retail-level predictability."

On collateral management, Aratovskaya introduces the term collateral fragmentation to describe the capital efficiency problem that arises when brokers post margin across multiple liquidity providers simultaneously.
 

"Brokers love to brag about '50+ LPs.' If they do, ask them one thing: Where is the cash? Posting margin with 10+ LPs is capital suicide — it can't be redeployed easily. You're basically giving your counterparties a free loan. Very few LPs offer interest on that margin. It's dead money — and in an agency model, dead money is a slow-motion death sentence for your ROI."

Technology infrastructure forms the fourth risk category. During periods of elevated market volatility, Aratovskaya notes that liquidity providers may reduce tick delivery, accumulate order rejections, and become temporarily unreachable, with the consequences falling squarely on the broker.
 

"Aggregation, routing, and stack management are always risk points. When volatility spikes, LPs don't just widen spreads. They (can) drop ticks, pile up rejections, and ghost the bridge. Even top-tier tech providers are vulnerable to outages and routing errors. When your primary LP's feed freezes, even the best routing logic can fall short. As the LP, you eat it all: The complaints. The fill risk. The reputational damage. All while operating on razor-thin commissions. One operational hiccup can wipe out an entire month of 'safe' A-book revenue."
 

The fifth risk is flow quality. Aratovskaya argues that routing client orders externally does not remove the problem of toxic flow from a broker's book.
 

“A-book clients need just as much vetting as B-book ones. Liquidity aggregators (and experienced dealers) may tolerate (or route around) some risky behaviors. But that doesn't make the toxic flow non-toxic. An A-book doesn't magically clean your client base. If anything, the toxicity just shows up with better grooming and more complex behavior. If you don't manage flow quality, your LPs will eventually manage you — right out of their pool.”

Lastly, she also said:

"An A-book isn't a 'night saver.' It's a high-maintenance, capital-heavy, relationship-sensitive, tech-fragile monster. Running one well takes just as much discipline as running any other book — especially when institutional clients are involved."

Found this interesting? Become a member of LiquidityFinder and get daily industry news direct to your inbox — join here.