The “invisible” cartel of market makers

Market makers are the finance equivalent of power lines: rarely noticed when working, blamed for everything when the lights flicker. They quote buy/sell prices, stand ready to trade when others won’t, and knit together dozens of venues that don’t always talk nicely to each other. Because a small number of ultra-scaled firms handle an enormous slice of order flow—especially retail flow—many investors suspect a shadowy cartel silently steering prices.

Is that accurate? The short version: there’s concentrated power and recurring conflicts of interest in how U.S. equities and options are routed and priced, and there have been documented abuses in the past. But a standing cartel that sets prices together behind closed doors? The evidence points instead to cutthroat competition among a few giants, shaped by a market design that sometimes makes their behavior look coordinated—even when it’s mostly incentives, physics, and the same math pushing everyone in the same direction.

Let’s unpack the mechanics, the myths, and the middle ground.

What a market maker actually does

A market maker (MM) continuously posts two prices:

Bid (what it will pay to buy a security)
Ask (what it will accept to sell that security)

The spread (ask minus bid) is compensation for risk: inventory can move against them, news can gap prices, and client flow isn’t symmetrical. Good MMs also hedge—across venues, ETFs, futures, options, correlated names—so they can keep quoting tight spreads without blowing up.

Today’s MMs are algorithmic and extremely fast. They co-locate servers at exchanges, pay for premium data, and run routing logic that picks the best venue, order type, and microsecond to interact. They make fractions of a cent many, many times.

Why “cartel” talk never dies

Three structural facts feed the narrative:

Concentration
In U.S. equities and options, a handful of wholesalers and proprietary market-making firms dominate retail execution and a big chunk of displayed liquidity. Scale wins because the more flow you see, the better your pricing models get and the cheaper you can hedge.
Information asymmetry
Some firms see vast streams of marketable retail orders (via broker relationships). Even with best-execution requirements, having that flow early—plus faster/private data feeds—can be a meaningful edge.
Opaque plumbing
Retail orders often execute off-exchange (in wholesalers or internalizers) rather than on lit exchanges. That can deliver price improvement to the retail trader, but it also means price discovery can migrate away from the public quote. To outsiders, off-tape fills + tight coordination across a few big players looks… coordinated.

Put those together and you get an easy villain: “They must be meeting in a smoky back room.” The harder reality is economic: if you design a market where scale + speed + flow internalization are rewarded, you’re going to get a small circle of dominant players—even if they’re fierce competitors.

The core accusation, translated

Accusation: “They control prices.”
What’s really being alleged: Through (a) control of retail order flow, (b) speed/data advantages, and (c) rebates/fees/order types, a few firms can nudge spreads, capture most of the economics, and occasionally starve the lit market of volume—leaving the public quote less informative and giving themselves better trading economics.

That’s not the same as a cartel. A cartel is explicit collusion to fix prices or exclude rivals. What the market has is structural concentration and tacit coordination—firms reacting similarly to the same microstructure incentives.

The incentives that make behavior look coordinated

1) Payment for Order Flow (PFOF) and internalization

Many retail brokers route marketable orders to wholesalers who pay for that flow. In return, wholesalers execute at or better than the public quote and share some economics with the broker. Outcomes for retail traders can look good (you often get a tiny improvement versus the NBBO), but system-wide effects include:

Price discovery off-exchange: If the easiest orders never hit lit venues, displayed quotes can get thinner.
Routing conflicts: Brokers have to prove “best execution,” but they also earn from PFOF—so critics worry about subtle biases.
Scale moat: Wholesalers with the most flow learn the most about short-term order imbalances, reinforcing their advantage.

When several big wholesalers all chase the same flow with the same constraints, their quotes and hedging behavior can look strikingly similar. That’s shared incentives, not necessarily shared plans.

2) Maker–taker fees and order-type complexity

Exchanges pay rebates to add liquidity and charge to remove it. Broker routers, algos, and MMs all respond to those incentives. Some order types (pegged, hide-not-slide, midpoint, etc.) interact in ways that are esoteric unless you live in the weeds. If multiple players optimize for the same fee grid and matching logic, you’ll see convergent behavior that can mimic coordination.

3) Data asymmetry and speed

Public feeds (the consolidated “best bid/offer”) are slower and historically showed less depth than exchange direct feeds. Faster, richer data lets MMs lean quotes microseconds ahead of the public tape. If several top firms all buy the same fastest feeds, co-locate, and run similar latency-sensitive models, they’ll pull quotes at the same time under stress. That looks like a pact; it’s usually just similar risk controls firing simultaneously.

Where real misconduct has happened

It’s important to separate illicit acts (manipulation, deception) from aggressive but legal strategy.

Spoofing/layering: placing fake orders to move price, then canceling. This is illegal; regulators have brought cases and extracted fines or guilty pleas.
Undisclosed order-type advantages: venues have been penalized when they offered complex order handling that advantaged certain users without clear disclosure.
Routing conflicts: brokers have faced scrutiny when routing appeared to prioritize economics over best execution.

Those are targeted abuses, not proof of a permanent, industry-wide cartel. But they justify strong surveillance, fuller disclosure, and teeth behind best-execution obligations.

Do market makers “set” the S&P 500 or single-stock prices?

In the short run, market makers help form prices by posting quotes and hedging flows. In the long run, fundamentals (earnings, rates, growth) dominate. Even intraday, MMs are more thermometers than thermostats: they reflect order-flow pressure and perceived risk. Could a few firms nudge the tape around key times (e.g., close, index rebalances, options expirations) by pulling or supplying liquidity? Yes—because that’s when liquidity is tight and one big balance sheet matters. But swaps/futures/ETF arb desks, long-only funds, and macro players also matter; it isn’t a single lever.

What would a real cartel need—and why that’s unlikely

A functioning cartel would require:

Explicit coordination (agreements on spreads, rebates, volumes, or market shares).
Enforcement (punishing defectors who undercut spreads).
Cover-up across multiple venues, asset classes, and jurisdictions.
Aligned economics despite radically different tech stacks, risk appetites, and client mixes.

In a high-frequency, low-margin business, the overriding temptation is to defect—undercut your rival by half a cent or respond to flow faster. That’s why antitrust cartels usually show up in industries with sticky prices, small numbers of sellers, and slow feedback loops. Ultra-fast market making is the opposite: prices update thousands of times a second and defection is instantaneous.

So why does it still feel “rigged”?

Because outcomes can be uneven even without collusion:

Retail gets micro-improvements versus the quote, but the public quote may thin out, making institutional execution more expensive and weakening transparency.
A few firms earn most of the economics, which seems unfair even if it’s just scale and speed advantages.
In stress, the same models say “pull back,” so liquidity vanishes everywhere at once. That’s fragility, not conspiracy—but it feels the same in the moment.

The healthier skepticism: conflicts, fairness, and resilience

If “cartel” is the wrong frame, what’s the right one? Three questions:

Conflicts & transparency: Do brokers and venues clearly disclose incentives and order handling? Can investors compare routing quality easily?
Fair access: Are public data and order types robust enough that basic price discovery doesn’t require premium feeds and colocation?
Resilience: Do guardrails (circuit breakers, kill switches, position/risk limits) prevent feedback spirals when models herding becomes a stampede?

Reform that targets those questions improves outcomes without pretending you can—or should—ban speed or scale.

Practical takeaways (for different market participants)

For retail investors

Use limit orders when possible, especially for thin names or wide spreads. A 1–2 cent improvement adds up.
Understand your broker’s routing. Many publish Rule 606 reports (order-routing disclosures). Look for multiple wholesalers and lit-venue access, not a single route.
Beware market orders at the open/close. Those are volatile windows with special auctions and imbalances.

For active traders

Watch liquidity, not just price. Depth, imbalance alerts, and auction feeds matter. Tight prices with shallow depth can slip quickly.
Respect the calendar. Options expiration, index rebalances, earnings, and macro prints change microstructure dramatically.
Know venue behavior. Speed bumps, midpoint books, and hidden/pegged logic differ by exchange.

For institutions

Measure implementation shortfall, not just fill rates. Add venue-level slippage diagnostics.
Demand router transparency (fee sensitivity, dark/wholesaler preferences, reversion metrics).
Use conditional and auction mechanisms in size; don’t assume sweeping the book is cheapest.

For policymakers

Tighten best-execution measurement. Make it apples-to-apples across brokers and strategies, not marketing decks.
Keep modernizing public data. Depth, odd lots, and latency improvements narrow the fairness gap.
Stress-test the microstructure. Revisit collars, volatility pause logic, and cross-venue coordination so liquidity doesn’t vanish system-wide in tail events.
Sunlight on order types. Clear, plain-English specs and public test suites reduce hidden edges.

Where the line really is

Legitimate market making: Posting tight quotes, internalizing retail flow with price improvement, hedging aggressively and fast.
Grey-area behavior: Optimizing rebates and complex order types without clear disclosure, leaning on stale quotes around data gaps, timing fills into auctions.
Manipulation (illegal): Spoofing, layering, wash trading, collusion on spreads/volumes, abusing client information.

Concentrated power increases the stakes of grey-area behavior, which is why surveillance, audits, and disclosure need to be stronger when a few players dominate.

A balanced verdict

Is there an “invisible cartel” of market makers? No—not in the antitrust sense of a secret club fixing prices. What there is: a small cadre of hyper-scaled firms competing ferociously inside a market structure that rewards speed, data, and flow internalization. That structure can produce look-alike behavior, conflicts of interest, and fragility under stress—all the ingredients for suspicion, even without a conspiracy.

The productive path isn’t to demonize market makers as a cabal, but to tighten the plumbing:

Sharper best-execution standards and routing transparency
Stronger, faster public data
Clearer order-type disclosures
Smart guardrails for stress events
Robust enforcement against real manipulation

Do that well, and the market gets the best of what sophisticated market makers bring—tighter spreads, deeper liquidity, resilient pricing—without giving oxygen to the “invisible cartel” story.

ALSO READ: Are Forex Brokers Manipulating Prices?