Two exchanges can show identical spreads on Bitcoin perpetuals. One will function normally through a five percent move. The other will break. The difference is not technology, not regulatory standing, not brand. It is how liquidity was structured, specifically, the depth behind the top-of-book quote and the behaviour of market makers when conditions deteriorate.
This piece is about understanding that difference in practical terms: how liquidity actually works in digital asset markets, what the metrics mean under stress, and what separates genuinely liquid markets from those that only appear to be.
A market maker quotes continuously on both sides of the order book, a bid price at which they will buy, an ask price at which they will sell. The difference between those prices is the spread. The market maker captures that spread as revenue, while managing the inventory risk that builds as trades execute against their quotes.
In practice, this is not a static process. Market makers adjust their quotes continuously based on order book conditions, recent trade flow, cross-venue price signals, and their own inventory position. A market maker who has accumulated significant long exposure will shade their quotes to encourage selling and reduce their directional risk. A market maker short of inventory will do the opposite. These micro-adjustments happen within milliseconds on competitive venues.
The spread is therefore not just a transaction cost, it is a real-time signal of how comfortable market makers are with their current position and the current market conditions. A tightening spread signals confidence and competition. A widening spread signals the opposite.
Most traders and exchange operators focus on the top-of-book spread as the primary measure of liquidity quality. It is visible, comparable, and directly connected to transaction costs. But it captures only the cheapest slice of available liquidity.
The more important question is what happens behind that top-of-book quote. How much volume sits at the next price level? And the one after that? If a market shows a tight spread with minimal depth behind it, any institutional-sized order will move through those levels rapidly, filling at progressively worse prices. The spread at the moment of order entry bears no relationship to the average fill price on a large trade.
For institutional participants, slippage, the difference between the expected price and the actual fill price, is often a more significant cost than the headline spread. A market with a slightly wider spread but genuine depth across multiple price levels will produce better execution on large orders than a market with a tight spread and a shallow book.
This is why market depth analysis, examining the full order book, not just the top, is a more reliable indicator of true liquidity quality than spread comparison alone.
Digital asset markets are fundamentally fragmented. Bitcoin trades simultaneously across hundreds of venues globally. Each exchange maintains its own order book. Each has its own participant mix, fee structure, and liquidity dynamic. Prices diverge between venues continuously, and arbitrage traders work continuously to close those gaps.
This fragmentation means price discovery happens in parallel rather than in a single consolidated venue. A large buy order on one exchange moves prices there before arbitrageurs can rebalance across venues. The speed of that rebalancing, and the depth available at each venue, determines how efficiently the market processes the information embedded in that order.
For institutional participants, fragmentation requires a different execution model than traditional markets. Executing a large order on a single venue concentrates market impact. Effective execution requires splitting orders across multiple venues, routing to where genuine liquidity exists, and minimising footprint at any single exchange. Smart order routing is not optional at scale, it is a basic operational requirement.
The 24/7 nature of digital asset trading adds another dimension. Liquidity patterns shift across time zones, Asian, European, and US sessions each carry distinct characteristics. Weekend liquidity differs materially from weekday patterns. Institutions that treat all hours as equivalent will consistently encounter worse execution during lower-liquidity periods than their models anticipated.
Normal market conditions are a poor test of liquidity quality. The real test is what happens when conditions deteriorate: a macro shock, a large liquidation cascade, a sudden sharp directional move.
Under stress, the economics of market making change rapidly. Inventory risk increases. Adverse selection risk, the risk of trading against informed participants who know something you do not, increases. The rational market maker response is to widen spreads, reduce size, or withdraw entirely until conditions stabilise. None of this is irrational behaviour. It is sensible risk management at the individual level that creates a collective problem at the market level.
The question for exchange operators is not whether market makers will pull liquidity under extreme stress, some always will. The question is how the market structure behaves when they do. An exchange with genuine depth across multiple providers and price levels will thin out but remain functional. An exchange that relied on a small number of rebate-driven participants providing top-of-book quotes will see its order book hollow out rapidly.
The March 2020 COVID-19 market disruption provided a clear illustration. Digital asset markets experienced severe liquidity crunches. Prices diverged significantly between exchanges. Participants who had assumed that normal-conditions liquidity would persist into stress scenarios discovered otherwise at significant cost.
Several metrics provide meaningful signal about true liquidity quality beyond top-of-book spread.
Time-weighted average spread gives a more complete picture than a snapshot. Markets may show tight spreads most of the time but widen materially during specific periods, around large announcements, during liquidity transitions between time zones, or following significant market events. Understanding when spreads widen helps institutional participants time large executions more effectively.
Market resilience, how quickly the order book refills after a large trade, indicates sustainable depth. A market that recovers quickly to tight spreads after an institutional-sized order has genuine liquidity behind it. A market that stays wide and thin after a large trade has not.
Price dispersion across venues provides real-time signal about market stress. Wide, persistent gaps between exchange prices indicate reduced arbitrage activity, often an early indicator that liquidity conditions are deteriorating before the order books visibly reflect it.
Volume figures require careful interpretation in digital asset markets. Reported volumes can reflect wash trading on some exchanges rather than genuine activity. Cross-referencing data sources and focusing on reputable venues is essential before drawing conclusions from volume statistics.
Decentralised finance introduced a fundamentally different approach to liquidity provision. Automated market makers use mathematical formulas, most commonly a constant product formula, to determine prices based on the ratio of assets in liquidity pools. Anyone can provide liquidity by depositing token pairs and earning a share of trading fees.
AMMs democratise liquidity provision and offer guaranteed execution at mathematical prices. But they introduce dynamics that differ significantly from order-book markets: impermanent loss for liquidity providers, slippage that increases non-linearly with trade size relative to pool depth, and MEV vulnerability from the transparency of pending transactions.
Sophisticated trading firms increasingly arbitrage price differences between AMMs and centralised exchanges, which helps maintain price consistency across market structures. Newer AMM designs with concentrated liquidity, where providers focus capital within specific price ranges, bring decentralised liquidity provision closer to traditional market making in terms of capital efficiency.
Market depth refers to the volume of buy and sell orders sitting at different price levels in an exchange's order book. Deep markets can absorb large trades without significant price impact. Shallow markets, even those showing tight top-of-book spreads, move dramatically when institutional-sized orders hit the book. Depth is a more reliable indicator of true liquidity quality than headline spread comparisons.
During volatile periods, the risk economics for market makers change rapidly. Inventory risk increases. Adverse selection risk increases. The rational response is to widen spreads, reduce posted size, or withdraw until conditions stabilise. Exchanges that relied on rebate-driven, fair-weather liquidity see their order books hollow out precisely when depth is most needed. Resilient exchanges are those that designed their liquidity structure, through tiered market maker relationships, depth-incentivising fee structures, and stress-tested parameters, to hold under pressure.
Because digital assets trade simultaneously across hundreds of venues, price discovery happens in parallel rather than in a single consolidated market. Prices diverge continuously between exchanges, and arbitrage traders work to close those gaps. For institutional participants, fragmentation requires splitting large orders across multiple venues to minimise market impact, executing entirely on a single exchange concentrates impact and produces worse average fill prices than a well-routed, multi-venue approach.
Julien Gandia is the founder of Lyquid Markets, a specialist advisory firm focused on market structure, derivatives design, and liquidity architecture for exchanges and professional trading firms. His background spans fifteen years across ICAP/EBS, NEX Group, CME Group, and Binance, covering FX primary venues, perpetual derivatives, institutional market making, and exchange product architecture.
Lyquid Markets works with a small number of clients on the structural problems that determine whether a trading business functions under real market conditions.
lyquidmarkets.com | LinkedIn