FX Liquidity Aggregation

FX liquidity aggregation is a technology and execution strategy in which a prime broker or trading platform collects simultaneous price quotes from many different liquidity providers—banks, electronic communication networks, and other market makers—and combines them into a single best bid and best offer (BBO) for a client. When the client executes at the aggregated price, the order is routed to (and split among) the providers who quoted the best prices.

How liquidity aggregation works

The aggregation pipeline is straightforward in concept but operationally complex:

1. Data collection: The aggregator (typically a prime broker or independent platform) maintains real-time connections to the pricing feeds of 10–50 different liquidity providers.

2. Quote normalisation: Each provider may quote slightly different formats or with different conventions. The aggregator normalises these—converting all prices to the same decimal places, handling bid-ask reversals for certain pairs, and filtering out stale quotes.

3. BBO computation: The system identifies the best bid (highest price at which someone will buy) and best ask (lowest price at which someone will sell) across all providers at that instant.

4. Quote display: The aggregated BBO is shown to the client—or in algorithmic execution, used automatically to fill incoming orders.

5. Execution and routing: When the client hits the ask (sells) or lifts the bid (buys), the system must immediately route the order to the provider(s) who quoted the best price. If the best ask comes from Dealer A and the second-best from Dealer B, and the client is selling 100 million, the system may send 60 million to A (whose quote was best) and 40 million to B, capturing both liquidity points.

6. Confirmation and give-up: Each dealer confirms its portion, and the prime broker gives up the trades to their respective counterparties.

Why aggregation reduces spreads

Without aggregation, a client would have to pick one dealer and execute with it alone. That dealer’s bid-ask spread might be 2 basis points on EUR/USD (buy at 1.0800, sell at 1.0802).

With aggregation across 10 dealers, each quoting a 1.5 to 2.5 basis point spread, the aggregated BBO might be 1.0800.5 / 1.0801.2—a composite spread of just 0.7 basis points. The client saves money; the prime broker captures a markup above that aggregated price.

This benefit is largest in major, liquid pairs (EUR/USD, USD/JPY, GBP/USD) where multiple dealers compete aggressively. In exotic pairs, only 2–3 dealers may be willing to quote, so aggregation offers less improvement.

The technological challenge

Aggregation requires extremely low-latency systems. The entire process—collecting quotes, computing the BBO, displaying it to the client, receiving an execution instruction, routing to dealers, and confirming—must happen in tens to hundreds of milliseconds.

If the aggregator is slow, the quotes it displays will be stale. A client sees the BBO at 10:00:00.050 but by the time they execute at 10:00:00.150, the price may have moved. The aggregator bears the risk that the underlying quotes have changed since the composite price was calculated.

Most serious aggregators use:

• Co-location: Placing servers in the same data centres as liquidity providers to minimise network latency.
• Direct feeds: Subscribing to high-speed, proprietary price feeds instead of relying on slower public feeds.
• Caching and pre-computation: Maintaining a continuously refreshed state of BBO rather than computing it on-demand.
• Conflict monitoring: Detecting and filtering out quotes that are clearly stale or erroneous (e.g., a GBP/USD quote that is miles away from the market price).

Aggregation under stress

In volatile markets, aggregation becomes harder. Dealers widen their spreads or stop quoting altogether, reducing the number of quotes available. The aggregated BBO widens accordingly.

Dealers also use last look—the right to reject a trade after seeing the client order—more frequently under stress. A client may see an aggregated price of 1.0800 / 1.0802 but when they hit the ask, half the dealers invoke last look and cancel. The execution becomes partial or slips to a worse price.

In severe dislocations (e.g., currency crises or central bank interventions), many dealers may stop quoting entirely, and liquidity aggregation becomes merely aggregating the few remaining quotes. The benefit collapses.

Multi-dealer aggregation vs. single-dealer advantage

Some brokers and dealers, particularly the largest, argue they do not need to aggregate—they claim their own internal pricing is competitive because of their large market share and internal order flow. JPMorgan Chase, for example, handles massive volume and can offer very tight spreads on major pairs from its own inventory.

However, even large dealers use aggregation in certain scenarios: for very large trades (where their own inventory may not be sufficient), for exotic pairs, or when they want to source the best price without exposing their own mark-up.

Smaller dealers and platforms almost always aggregate because they lack the volume and balance sheet to offer competitive single-dealer pricing.

Aggregation and market impact

When a large order is routed to multiple dealers simultaneously (as aggregation does), it can fragment the market impact. Instead of one dealer absorbing a 100-million buy order, five dealers each absorb 20 million. Each dealer faces less adverse impact, and the overall execution cost is lower.

However, if dealers use the fact of aggregation against the client—by widening their quotes when they detect an aggregated request, or by imposing higher last look rejection rates—the benefit is eroded. Some dealers have been known to offer tighter quotes in single-dealer “direct” channels to incentivise clients to bypass aggregation.

Aggregation and regulation

Regulators have focused on aggregation in the context of best execution. Under ESMA’s Markets in Financial Instruments Directive (MiFID II), brokers must achieve best execution for clients. For FX, this is interpreted as offering aggregated pricing or at least demonstrating that the single-dealer pricing offered is competitive with aggregated alternatives.

Similarly, the US SEC and CFTC expect brokers to show they are not systematically routing orders to sub-optimal liquidity providers simply because of commercial relationships. The use of aggregation can serve as evidence that the broker is genuinely seeking best execution.

However, not all jurisdictions mandate aggregation, and retail clients in some markets are intentionally offered single-dealer prices to keep systems simple and regulation minimal.

Aggregation for retail brokers

Many retail FX brokers, especially smaller ones, do not operate their own aggregation platforms. Instead, they white-label (resell) aggregation services from larger prime brokers or independent aggregation providers like multi-bank venues and ECN platforms.

This means a retail client trading through a small broker is actually executing against aggregated liquidity sourced from a larger entity. The retail broker adds its own mark-up on top of the aggregated spread, earning its profit.

This tiered structure means retail clients get some benefit of aggregation, but not the full benefit—each layer (the aggregator, the retail broker) takes a mark-up.

Technical considerations and latency arbitrage

Because aggregation involves latency (time to collect quotes, compute BBO, and display to client), sophisticated traders sometimes engage in “latency arbitrage”—detecting when the displayed aggregated price is stale relative to the actual market, then placing orders to exploit the mismatch.

For example, if the aggregator shows EUR/USD at 1.0800 / 1.0801 but the true market (based on raw dealer quotes) is 1.0799.5 / 1.0801.5, a trader can buy at the displayed 1.0801 and immediately sell to the actual market at 1.0801.5, capturing 0.5 basis points with zero risk.

To counter this, aggregators have invested heavily in latency reduction and in detecting and filtering stale data. The arms race between aggregators and latency arbitrageurs is ongoing.

See also