Market Impact Curve

Direct Answer

A market impact curve is a chart or model showing the estimated relationship between order size and expected price impact, the amount a trade itself is expected to move the market price, kept separate from the ordinary cost of crossing the bid-ask spread. A commonly cited class of market-impact models, often called "square-root models," estimates that price impact grows roughly in proportion to the square root of order size relative to typical trading volume in the security. That is a widely referenced modeling approach, not a single universally precise formula: the exact functional form and its parameters vary across different published models and market conditions, and real-world impact also depends on urgency, order type, and how the order is worked over time.

Key Takeaways

• Price impact is separate from the bid-ask spread, it's the additional cost created by the trade's own size, not the cost of simply crossing between bid and ask.
• A commonly cited "square-root model" estimates impact grows roughly with the square root of order size relative to typical trading volume, meaning impact rises more slowly than order size, but still rises.
• The square-root relationship is one commonly referenced modeling approach among several published models, not a single precise, universally agreed formula.
• Real-world impact also depends on urgency, order type, and how an order is worked over time, not size relative to volume alone.
• Market impact curves are estimates built from historical relationships and simplifying assumptions, not a guarantee of what any specific order will cost.

What Is a Market Impact Curve?

A market impact curve plots expected price impact on one axis against order size, usually expressed relative to typical trading volume in that security, on the other. As order size grows, the curve shows price impact rising too, but the relationship is not assumed to be a straight line. Larger orders consume more of the resting liquidity available at and near the best price, and can also signal information to other market participants, both of which push the expected price further away from where it started.

It's important to separate two distinct costs when evaluating a trade. The bid-ask spread is the cost of crossing between the best bid and best ask, and it exists even for a single-share trade. Price impact is the additional movement caused specifically by the trade's own size, it would not exist, or would be far smaller, for a much smaller order in the same security at the same moment. A market impact curve is a way of modeling that second, size-driven cost in isolation.

How the Curve Is Built

A market impact curve is typically constructed (or estimated) with order size, often normalized as a fraction or percentage of typical daily trading volume, sometimes called participation rate, on the horizontal axis, and expected price impact, often expressed in basis points, on the vertical axis. Because the underlying relationship is being estimated rather than observed directly for every possible order size, the curve is usually the output of a model rather than a single measured line.

A commonly cited class of these models, often referred to as square-root models, estimates that price impact grows roughly in proportion to the square root of order size relative to typical trading volume. In plain terms: doubling an order's size relative to volume is commonly modeled as increasing expected impact by less than double, because the square root of 2 is smaller than 2. This produces a curve that rises steeply at first for very small orders and then flattens somewhat as size grows, rather than a straight diagonal line.

It's worth being precise about what this describes and what it doesn't. The square-root relationship is a commonly cited modeling approach used across parts of the market-microstructure and execution research literature. It is not a single universally precise formula. Different published models vary in their exact functional form, in the specific parameters (constants, scaling factors) used to fit the curve to a given security or market, and in the market conditions under which they were estimated. Two different published models can produce visibly different curves for the same security. And whatever the modeled curve says, real-world impact also depends on factors the curve alone doesn't capture, how urgently the order needs to be filled, what order type is used, and how the order is worked over time (all at once versus split into many smaller pieces).

How to Read the Curve, Illustrative Example

Hypothetical example, for education only.

The table below illustrates, in a simplified and hypothetical way, how a square-root-style impact estimate might scale as order size grows relative to a stock's typical daily volume. The specific numbers are constructed for illustration only and are not drawn from any real security or published model's calibrated parameters.

Notice the pattern: each time the size factor increases by 4×, the illustrative impact estimate only doubles, consistent with a square-root-style relationship (the square root of 4 is 2). Going from 0.25% to 1.00% of typical volume (a 4× increase in relative size) roughly doubles the illustrative estimate rather than quadrupling it. This is the core intuition behind why square-root-style models are commonly cited: they capture the idea that impact keeps rising as orders get larger, but at a decreasing rate relative to size, rather than assuming cost scales in direct one-to-one proportion with order size.

Again, these figures are illustrative only. A real published model's exact curve, its scaling constant, its precise exponent, and how it responds to a specific security's volatility and liquidity profile, would need to be calibrated and verified separately, and different models calibrated to the same security can still disagree.

How Traders and Desks Use Impact Curves

Pre-trade cost estimation

Before placing a large order, a trader or execution desk may consult an impact model to get a rough sense of expected cost beyond the quoted spread, alongside other inputs like current order book depth and recent volatility. This is commonly used as one input among several, not treated as a precise, guaranteed cost.

Comparing execution strategies

An impact curve can help frame the tradeoff between executing quickly (higher expected impact, less time exposed to price drift) and executing patiently over a longer window using strategies like TWAP, VWAP, or participation-of-volume schedules (potentially lower expected impact, more time exposed to the market moving against the position for unrelated reasons). Neither approach is universally better; the right balance depends on urgency and risk tolerance.

Sizing and splitting orders

Because impact is commonly modeled as growing with order size (even if sub-proportionally under a square-root-style approach), breaking a large order into smaller pieces worked over time is a common practice, discussed further in how to estimate slippage before entering a trade. This doesn't eliminate impact, it can reduce the modeled per-trade impact at the cost of taking longer and accepting more exposure to price movement unrelated to the order itself.

None of these applications should be read as a guarantee of outcome. Impact curves are estimates, built on modeling assumptions that may not hold precisely for any individual order, security, or moment in time.

Limitations and Common Mistakes

• Treating the square-root model as the formula. It's a commonly cited modeling approach, not a single agreed-upon equation, published models differ in exact form and calibrated parameters, and using one model's output as if it were a universal constant is a common mistake.
• Ignoring urgency and order-type effects. The curve typically models size relative to volume; it does not, by itself, capture how much urgency, order type (market vs. limit), or how the order is worked over time changes the actual realized impact.
• Confusing price impact with the full cost of a trade. Price impact is one component of total execution cost. The bid-ask spread, commissions or fees, and ordinary market movement during the execution window (sometimes tracked separately as implementation shortfall) are additional, distinct components.
• Assuming impact is symmetric and stable. Impact relationships estimated during one volatility or liquidity regime may not transfer cleanly to a different regime, a curve fit during calm, liquid conditions is not automatically valid during a volatility spike or a thin, illiquid session (see liquidity gaps and thin books).
• Using a modeled estimate as a guaranteed cost. An impact curve is a modeled estimate built from historical relationships and simplifying assumptions. It is not a guarantee of what any specific future order will actually cost.

A Curve Is a Budget, Not a Prediction

The right use of an impact model is as a budget line. It gives an order of magnitude for what moving a given size is likely to cost, which is enough to decide whether an intended trade needs breaking up, stretching out, or reducing. Treating its output as an expected fill price asks more of it than any model of this kind can deliver.

The parameters carry the uncertainty. Curves are fitted to particular securities, periods and market conditions, and applying one calibrated elsewhere to the name in front of you assumes a similarity that may not hold.

The shape itself encodes a specific claim: that cost grows more slowly than size. That is a general tendency rather than a guarantee, and in a thin book or a stressed market the relationship can be far steeper than any smooth curve suggests.

Impact also depends on how an order is worked. The same total executed patiently and executed at once produce different costs, and a model taking only size as an input cannot tell them apart.

Frequently Asked Questions

References

• SEC: Rule 605 FAQs
• CFA Institute Research and Policy Center: Investment Research
• FINRA: Understanding Order Execution
• Investor.gov: Market Order

Assumptions in this article: The numeric table in the worked example is hypothetical and constructed for illustration only; it does not represent a calibrated output of any specific published model or real security. Square-root-style market-impact models are described here as a commonly cited modeling approach with variation across published research and market conditions, not as a single precise, universally agreed formula.

Educational Disclaimer

For education only; not personalized investment, tax, or legal advice. Trading can result in substantial losses.

Market conditions, liquidity, and execution mechanics change. Verify current conditions with the relevant broker, exchange, or qualified professional before acting.