Measuring Liquidity Quality: The Metrics That Truly Matter

Every FX liquidity provider has impressive specifications: Tight spreads on major pairs, claimed depth across select exotic corridors, 24/7 availability, and fast execution times.

These are table stakes. How do you know where real liquidity actually is?

Most desks evaluate liquidity by looking at quoted spreads and claimed availability. That’s the difference between looking at the specs on a bridge and actually trying to drive 100 tons of trucks across it. When stress hits the system, you need to know that bridge will hold, and what you read on paper won’t save you.

Here’s what you actually need to know to gauge liquidity quality.

Metric 1: Bid-ask spread consistency over time

Quoted spread is what you see on the screen right now. Spread consistency is what you get over thousands of trades across different market conditions.

Research on bid-ask spreads consistently identifies spread volatility (skewness) as a critical measure of liquidity fragility. A market can have low average spreads but high spread skewness - meaning most of the time spreads are tight, but during stress events they spike dramatically. The Bank for International Settlements' research on liquidity fragility found that markets with high spread skewness are more vulnerable to sudden liquidity crises, even when average liquidity looks fine.

What to measure:

• Average spread across a full trading day, not just during London/New York overlap

• Spread volatility (standard deviation of spreads over time)

• Spread skewness (how often spreads spike vs. how often they're tight)

• Maximum observed spread during different volatility regimes

Example: Provider A quotes 1-pip average spreads on EUR/USD with occasional spikes to 15 pips during news. Provider B quotes 1.5-pip average spreads that rarely exceed 3 pips. Provider B has better liquidity quality despite the higher average—the consistency matters more than the headline number.

For exotic corridors like MXN, BRL, or PHP, spread consistency becomes more critical. A provider claiming tight spreads on the Mexican peso that suddenly quotes 50-100 basis points above the mid-market rate during volatility isn't actually providing the liquidity they advertised.

Metric 2: Execution quality at quoted prices

Seeing a price on your screen and actually executing at that price are two different things.

Fill rate at quoted price measures how often your orders execute at or better than the displayed quote. In high-quality liquidity, fill rates should be 95%+ even for reasonable order sizes. In low-quality liquidity, you'll see frequent rejections, or executions at prices meaningfully worse than what was quoted.

Slippage distribution reveals execution quality. Calculate the difference between the price you intended to trade at and the price you actually got, across all trades. Plot this as a distribution.

For high-quality liquidity: Slippage is tightly clustered around zero, with small deviations mostly due to natural market movement during order execution.

For low-quality liquidity: Slippage has a fat tail. Most trades execute close to quote, but 10-20% of trades have significant slippage, especially on larger sizes or during volatility.

Price improvement vs. price deterioration ratio is another revealing metric. Some of your trades should execute at prices better than the quote due to natural market movement in your favor. If you're only ever getting price deterioration and never price improvement, it suggests your liquidity provider is selectively executing only the trades where they have an edge—a sign of low-quality liquidity.

Metric 3: Rejection rates and last-look frequency

Last-look is a mechanism where a liquidity provider sees your order, checks current market conditions, and decides whether to accept or reject it. In theory, it protects providers from stale quotes during fast-moving markets. In practice, it's often used to cherry-pick profitable trades and reject unprofitable ones.

High rejection rates correlate with low liquidity quality. Research on FX market microstructure shows that last-look rejections impose costs on traders that aren't captured in quoted spreads, effectively widening the true cost of trading.

What to measure:

• Overall rejection rate (percentage of orders rejected)

• Rejection rate by order size (do larger orders get rejected more often?)

• Rejection rate during volatility (do rejections spike when execution needs rise?)

• Time to rejection (how long does it take to reject an order and send you back to market?)

A provider with a 5% rejection rate during normal markets, that jumps to 25% during news events, has low-quality liquidity. You're effectively trading with someone who only wants your business when they're confident they'll profit from it.

Compare this to high-quality liquidity, where rejection rates stay below 2% even during volatility, because the provider has genuine depth and doesn't need to pick and choose which trades to accept.

Metric 4: Settlement speed under load

Settlement speed matters most when markets move fast. A provider claiming 1-hour settlement that delivers on that promise when processing $1M/day, but slips to 4-hour settlement when volumes hit $50M/day, doesn't have robust infrastructure.

What to measure:

• Median settlement time across all transactions

• 95th percentile settlement time (how slow do your slowest 5% of settlements get?)

• Settlement time by transaction size (does speed degrade with larger orders?)

• Settlement time during high-volume periods (does performance hold up under load?)

OpenFX processes 90% of transactions in under 60 minutes regardless of volume—whether it's a $100K or $100M transaction. Our largest legacy competitor requires 700+ treasury operations staff for what we handle with 4 engineers. This isn't just about speed; it's about whether settlement speed degrades under stress or scales horizontally.

During volatile periods when you most need fast settlement—to minimize currency exposure, rebalance positions quickly, or respond to market moves—low-quality liquidity slows down. High-quality liquidity maintains performance.

Metric 5: Depth-of-book consistency

Quoted depth at the best bid/ask is one thing. Depth across the order book is another. Depth that persists across different market conditions is the real test.

Order book depth should be measured at multiple price levels:

• Depth at best bid/offer

• Depth within 5 pips of mid

• Depth within 10 pips of mid

• Total depth available across reasonable price range

High-quality liquidity maintains order book depth, even during volatility. The book might thin somewhat, but it doesn't collapse. The New York Fed's research on Treasury market liquidity uses order book depth as a core metric because it reveals whether market participants are willing to commit capital across prices, not just at the current best quote.

Depth replenishment speed after trades is also worth looking at. When a large order executes and clears a price level, how quickly does that level get replenished? In high-quality markets, depth returns within seconds to minutes. In low-quality markets, it can take hours … or simply never returns.

For exotic currency pairs where depth is naturally thinner, this metric becomes critical. A provider offering "deep liquidity" in PHP or BRL that shows 95% depth reduction after a moderately-sized trade isn't providing deep liquidity; they're providing surface liquidity.

Metric 6: Price accuracy and alignment

Your liquidity provider's quoted prices should align closely with the broader market, not systematically favor the provider.

Price deviation from mid-market measures how far your provider's quotes are from the true interbank mid rate. Small deviations are natural. Systematic deviations suggest the provider is quoting prices that advantage them rather than reflecting true market conditions.

For major pairs, 1-2 basis points from mid-market is typical. Exotic pairs may yield 5-10+ basis points, but watch for systematic patterns (15-20 bps consistently is a red flag). In short, it’s not about any single deviation; it’s about whether you systematically get worse prices than mid-market would justify.

Quote stability during volatile periods. Some providers' quotes become erratic during news events—prices jumping dramatically, spreads widening and narrowing rapidly, quotes refreshing constantly. This isn't a reflection of market movement; it's a reflection of the provider's inability to maintain stable pricing.

Cross-pair consistency is especially notble. If you're trading a synthetic pair (ex. BRL/MXN via USD), the implied price from the two USD-based legs should align closely with any directly quoted BRL/MXN price. Large discrepancies suggest the provider isn't actually maintaining consistent pricing across their book.

What great liquidity actually looks like

Putting these metrics together, here's what genuinely high-quality liquidity looks like by our standards:

Spread consistency: Average spreads within 20% of quoted spreads 95% of the time, even during volatility. Maximum spread spikes limited to 3-4x normal levels during extreme events, not 10-20x.

Execution quality: 98%+ fill rate at quoted prices for standard order sizes. Slippage tightly distributed around zero with minimal fat-tail events. Price improvement on 40%+ of trades due to natural market movement.

Rejection rates: Sub-2% overall rejection rate that doesn't spike during volatility. No systematic pattern of rejecting larger orders or trades during fast markets.

Settlement speed: Sub-60-minute median settlement across 90%+ of transactions, regardless of order size or volume. 95th percentile settlement under 90 minutes. No degradation during high-volume periods.

Order book depth: Visible depth at multiple price levels that persists during volatility. Depth replenishment within 1-2 minutes after trades. Book doesn't collapse during stress events.

Price accuracy: Quotes consistently within 1-2 basis points of true mid-market on major pairs, 5-10 basis points on exotics. Stable pricing during news events without erratic quote behavior.

Most providers don't share these metrics

These metrics reveal quality gaps that most providers would rather keep hidden.

It's easy to advertise "tight spreads" when you're cherry-picking the best 10% of your spread distribution. It's harder to defend "tight spreads" when someone asks for spread volatility and skewness statistics.

You can claim "deep liquidity" when markets are calm and nobody's stress-testing your order book. It's harder to explain why quoted depth evaporated completely during the last Fed announcement.

“Fast settlement" is a common headline feature. It's harder to justify why your 95th percentile settlement time is 3x your median, or why settlement speed degrades 50% when volumes exceed certain thresholds.

High-quality liquidity providers welcome these questions because their infrastructure performs well across all these metrics. Low-quality providers deflect them because the answers would reveal that their "deep liquidity" is mostly marketing.

The framework for evaluating liquidity

Don't ask: "What are your spreads on EUR/USD?" Ask: "What's your spread distribution across different volatility regimes, and what's your 95th percentile spread?"

Don't ask: "Do you offer 24/7 execution?" Ask: "What's your rejection rate during London open vs. Asian session, and how does it change during high volatility?"

Don't ask: "How fast is settlement?" Ask: "What's your median and 95th percentile settlement time by transaction size, and does performance degrade during high-volume periods?"

Don't ask: "How deep is your liquidity?" Ask: "What's your order book depth at 5 pips and 10 pips from mid, and how quickly does depth replenish after large trades?"

Providers who can answer these questions with data are the ones with genuinely high-quality liquidity. The ones who deflect or provide vague answers are the ones whose bridges look good on paper but haven't been stress-tested.

Why this matters

In calm markets, most liquidity looks roughly equivalent. The difference between high and low quality isn't visible.

But calm markets aren't when you need liquidity most. You need liquidity when:

• Central banks announce surprise rate decisions

• Geopolitical events trigger sudden currency moves

• Market infrastructure breaks down and everyone's trying to execute simultaneously

• Your business has an urgent need to move large amounts across borders quickly

These are when low-quality liquidity typically disappears. Spreads blow out, rejections spike, settlement slow or stop: The provider suddenly has very little to offer.

High-quality liquidity performs during stress. The metrics that reveal this quality aren't printed on providers' marketing materials. But they're the ones that reveal consistency, reliability, and resilience under load.

Measure accordingly.