Market making: What to check before you trust liquidity

Market making sounds simple on the surface. Quote two prices, earn the spread, and keep the book moving. Then real money and real stress arrive, market structure stops cooperating, and the function looks a good deal less mechanical than the textbook version. An investor, a selector, or an IC member does not really need another definition of what market making is. What they need is a way to read where liquidity is genuine, where it turns conditional the moment size or volatility climbs, and where an impressive-looking quote is mostly theatre.

It sits right next to execution quality, transaction costs, ETF liquidity, and the depth of listed derivatives. It sits next to the credibility of any firm that presents itself as a stable liquidity provider. A tight quote on a calm Tuesday proves very little. The harder test is whether the operator stays disciplined and adequately capitalised when the market turns openly hostile, and whether the desk keeps functioning at all once that happens.

Quick answer

Market making is the business of continuously quoting buy and sell prices so that others can trade. The investable question is one of resilience: how those quotes behave under stress, how inventory risk gets managed, and whether the depth you see on screen survives a live order of real size.

Key takeaways

Good market making shows up in live fill quality, well beyond the visible quoted spread.
Tight prices can vanish fast when volatility, size, or a correlation shock hits.
Strong operators run inventory, technology, and capital as one system. Weaker ones lean too hard on calm conditions.
For due diligence, stress behaviour deserves far more weight than normal-market marketing.

Why market making matters beyond a textbook definition
How market makers actually earn money and absorb risk
Where apparent liquidity is weaker than it looks
Why many market-making propositions disappoint under live conditions
When market making is actually worth relying on
What separates sound operators from fragile ones
Examples of live trading results diverging from theory
A practical due-diligence checklist
Warning signs that deserve skepticism

Why market making matters more than the definition suggests

Strip it back and market making just supports continuous trading by posting bids and offers. True, but incomplete. In live markets the function shapes how much friction an investor meets on the way into a position, how a product behaves when everyone reaches for the exit at once, and how much trust the displayed liquidity actually deserves.

For investors it becomes relevant in at least three settings. Strategy implementation is one, because execution costs ride on the depth and resilience of the market makers underneath. Exchange-traded products and derivatives are another, since liquidity can look abundant on screen and prove thin in real size. Manager and provider review is the third. That last one bites hardest when a firm claims a structural edge from market making, and the real story sits somewhere between controlled risk warehousing and exposure nobody is managing.

The practical issue is blunt. Quoted liquidity and usable liquidity are two different things. A market can show narrow spreads and constant updates and still turn expensive to trade the moment order size grows or conditions sour.

How market makers make money, and where the risk really sits

Most market makers aim to earn the spread between their bid and offer while managing inventory and hedging the residual risk. That sounds like a stable business. Stability is not guaranteed, and the reasons are worth spelling out.

The spread pays for several things at once: adverse selection, inventory risk, short-term volatility, funding, technology, and the ever-present chance that the next trade comes from someone better informed than you are. A market maker that buys just before prices fall, or sells just before they rise, does not merely give up the spread. It can lose several times the spread in seconds.

So the business is really about the decisions wrapped around the quote. When to widen. When to cut size. When to hedge, and when to simply stand aside until the tape calms down. The faster and more correlated the market, the more each of those calls matters.

Element	What it looks like	Why it matters in due diligence
Bid-ask spread	Difference between buy and sell quote	A narrow spread is positive, but only if it stays tradable in meaningful size
Displayed size	Quantity shown at the quote	Small displayed size can make headline spreads misleading
Inventory management	Control of accumulated long or short exposure	Poor inventory control can turn liquidity provision into directional risk
Hedging quality	Speed and cost of offsetting exposure	Weak hedging tends to show up during volatile periods
Capital resilience	Ability to stay active under stress	Some firms quote tightly until losses force them to retreat
Technology uptime	Reliability of pricing and risk systems	Operational failures often appear first as bad quotes or sudden withdrawal

Where apparent liquidity is weaker than it looks

The most common mistake in market making is to read visible quotes as dependable execution. Markets tend to look deepest exactly when they are least tested.

The gap shows up in a few recognisable ways. A quote may be firm only in very small size. A desk may refresh prices constantly and then back away the instant volatility ticks up. A product may post healthy secondary-market volume while its real liquidity leans on a much thinner underlying basket. In each case the screen stays reassuring right up to the second it does not.

Take an ETF with a one-cent displayed spread in a quiet session. An investor might conclude liquidity is excellent. If the underlying securities are less liquid, though, the true trading cost for a meaningful order can be materially wider. The market maker is intermediating underlying market risk and pricing it in real time. There is no free liquidity being conjured out of nowhere.

Options tell a similar story. Tight markets in benchmark maturities sit happily alongside very uneven quality in longer-dated or thinly-traded strikes. A provider can advertise broad coverage while most of its dependable liquidity clusters in a narrow subset of contracts.

Why many market-making propositions disappoint under live conditions

Market making as a function matters more than many readers assume, yet durable market making is far rarer than visible quoting suggests. A quote on screen is easy to admire in calm conditions. Keeping usable liquidity available when volatility rises, hedging turns expensive, inventory risk builds, and related markets stop cooperating is a different job. A much harder one. In the decks we review, the prettiest liquidity number on the page is almost always an average spread pulled from a calm quarter.

Many market-making propositions disappoint for predictable reasons:

Tight quotes get confused with durable liquidity. The Bank for International Settlements notes that market makers respond to rising volatility by widening spreads, shrinking quoted quantities, or changing quoting behaviour altogether. (bis.org)
Narrow spreads can hide liquidity risk. BIS also warns that narrow bid-ask spreads should not automatically be read as evidence that liquidity risks are low. (bis.org)
Visible depth needs a second look. CME’s liquidity work argues that book depth alone can be an incomplete and sometimes misleading guide to real tradability under stress. (cmegroup.com)
ETF and derivative liquidity often get oversimplified. In the SEC’s analysis of the August 24, 2015 episode, 302 of 1,569 ETFs experienced trading pauses during extreme volatility, a useful reminder that secondary-market liquidity can behave very differently under stress. (sec.gov)

The investor lesson is plain. Market making the activity is real, and the average claim about liquidity quality still tends to run well ahead of the underlying resilience. Plenty of providers look credible on a quiet screen and much weaker once the market genuinely tests their inventory limits, hedging discipline, and the operating model behind them.

When market making is actually worth relying on

Market making earns its keep when the liquidity function stays useful after you move past screens and into realised execution. Visibility is not enough. The provider has to stay coherent when size grows, volatility rises, or the related hedges get harder to put on.

The proposition is most compelling when the evidence points to real tradability:

Realised fills stay reasonably close to quoted liquidity across sizes. Not perfect, but coherent enough that displayed pricing is more than an optical starting point.
Inventory discipline is explicit. Strong providers can explain how inventory is bounded, when quote size is reduced, and how hedging shifts under pressure. BIS highlights that quoted spreads and quantities are inseparable from inventory and hedging costs. (bis.org)
The provider understands the product itself, well beyond the trading interface. Good market making in ETFs, options, futures, or single stocks rests on understanding the underlying hedge path and where it breaks.
Stress-period evidence exists. The ordinary day tells you little here. What counts is the opening dislocation, the macro release, the rebalance day, and the fast volatility shock, along with how quoting held up through each one.
Liquidity is honest rather than theatrical. A slightly wider but durable quote can beat an aggressive one that evaporates precisely when you need it.

The proposition weakens when the story leans too hard on average spreads, technology slogans, or calm-period screenshots. A more useful expectation is simple. This provider can still deliver credible, usable liquidity when conditions stop being easy.

What separates sound market making from fragile quoting

Good market making goes well past aggressive pricing. It is the combination of pricing discipline, hedging competence, solid infrastructure, and the balance-sheet endurance to sit through a bad week. That combination is rarer than the marketing material suggests.

Sound operators show a few recognisable traits. They understand the products they quote at a microstructural level. They can explain how they hedge, and when a hedge stops working. Their spreads, sizes, and participation profile stay coherent across regimes instead of looking brilliant only in quiet markets. And they produce reporting that lets an outside party tell normal variation apart from a process quietly deteriorating.

Fragile operators tend to compete by looking cheapest in the easiest conditions. That holds until the market turns disorderly. Then spreads gap wider, quote size disappears, or activity gets throttled because the risk controls were tuned to normality rather than stress.

Worth saying directly: the most attractive quote is rarely the most valuable liquidity provider. A slightly less aggressive but more durable market maker can be far more useful inside a real portfolio.

How live trading can disappoint even when the model or quote looks strong

Market making earns careful scrutiny because the live result so often degrades against what looked compelling on paper or on screen.

Example 1: the attractive spread that disappears in size

A manager sees consistently narrow spreads in a listed instrument and assumes implementation will be efficient. In live trading the first small slice fills near the mid. The larger slices walk through multiple levels, because displayed size was thin and replenishment was defensive. Recorded slippage comes in far worse than the quote history implied. The strategy still works in theory. Net performance drops once real capacity gets tested.

Example 2: volatility exposes inventory weakness

A liquidity provider looks highly competitive through stable periods and wins flow by quoting tightly. A sharp move arrives, hedging costs jump, inventory gets hard to neutralise, and quoting behaviour changes abruptly. The firm stays present, but only at much wider spreads and reduced size. Anyone who underwrote normal-market pricing into their performance assumptions learns that the liquidity was conditional.

Example 3: the less glamorous operator proves more investable

Another provider is never the headline cheapest. Its spreads run modestly wider, but the quote size is real, the hedging is disciplined, and the service keeps working across earnings periods, macro events, and rebalancing days. Over time realised execution comes out better, because the quoted price is more honest and less prone to evaporation. A useful reminder that investability lives in realised implementation, and the optics matter far less.

What weak operators usually hide

Weak market-making franchises rarely describe themselves as weak. The clues surface in what is missing, vague, or oddly selective.

Only calm-period evidence: average spreads are shown, while stress-period behaviour, gap days, and opening or closing auction conditions are left out.
No useful fill analysis: the discussion centres on quoting activity rather than realised execution quality by order size and market regime.
Thin discussion of hedging: they claim sophisticated risk management without explaining hedge instruments, basis risk, or inventory limits.
Too much technology language: low latency gets emphasised as if it solved capital, risk, and market-structure problems by itself.
Capacity claims without friction analysis: scalable opportunity is presented with little evidence on turnover, impact, or stress capacity.
Selective product focus: broad coverage is marketed, while the dependable liquidity is concentrated in only the easiest names or tenors.

These are not minor details. In market making, weak disclosure usually points to weak process. When a firm cannot describe how its liquidity changes under pressure, the answer is often one it would rather not give.

What to verify before relying on market making claims

When market making sits inside a manager narrative, a product structure, or a provider evaluation, the verification standard should be practical rather than theoretical. The question is whether the liquidity function stays credible when it matters, not whether the story sounds polished.

Execution evidence

Ask for realised fills by order size, time of day, and volatility regime. Screen quotes are a starting point. The metric that counts is the gap between displayed pricing and achieved execution after costs.

Stress behavior

Request evidence from the difficult sessions: macro releases, rebalance days, sharp selloffs, opening dislocations, and unusually wide correlation moves. A provider that shines only in benign conditions has limited value. A question we have learned to ask early is which instruments a desk will openly admit are outside its comfort zone.

Inventory and hedging discipline

Understand how inventory risk is bounded, how quickly hedges can be executed, and what happens when those hedges turn expensive or imperfect. Hidden fragility tends to live in this layer.

Operational resilience

Assess uptime, kill-switch design, manual intervention procedures, and escalation governance. A sophisticated pricing engine helps. A resilient operating model matters more.

Economic alignment

Clarify the incentives. Some participants optimise for flow capture, some for spread retention, some for the broader franchise relationship. Those incentives shape quoting behaviour in ways that surface during stress.

The central due-diligence mistake is to treat market making as a visible quote problem. It is a hidden resilience problem.

Practical checklist for investors and committees

The fastest way to sharpen judgment is to ask a tighter set of questions. These usually separate substance from presentation.

How does realised execution compare with quoted spreads across different order sizes?
What happened to spread, size, and fill rates during the worst five trading days of the last two years?
Which instruments or conditions are explicitly outside the provider’s comfort zone?
How is inventory risk limited, and who can override the risk controls?
What share of liquidity quality depends on one venue, one hedge, or one internal model?
How are outages, stale quotes, and erroneous fills detected and reported?
Is the firm paid in a way that rewards durable liquidity, or merely visible activity?

If the answers come back thin, heavily qualified, or framed as proprietary beyond reason, caution is warranted. Sophisticated operators can usually explain their process without giving away trade secrets.

Where to go next

This topic connects straight to execution quality, implementation costs, and the distance between visible and realisable liquidity. No internal pages extend the chain here, so the sensible next step is to look at the adjacent questions in the same spirit: how slippage is measured, how turnover changes capacity, how market stress reshapes fills, and how a reported backtest translates into live implementation.

That broader context matters, because market making rarely fails on its own. It fails at the intersection of execution, risk warehousing, and operational resilience.

Market making under real conditions

Market making deserves more respect and more skepticism than it usually gets. Respect, because functioning liquidity is genuinely hard to provide well. Skepticism, because a polished quote can hide conditional behaviour, weak hedging, or a fragile operation underneath. Judge it by realised execution, durability through stress, and how transparent the process is. If the liquidity only looks excellent in easy markets, it is not a strong advantage. It is a fair-weather one.