Square Root Impact Law

The expected price impact ΔP is proportional to the square root of the trade size relative to volume:

ΔP ∝ σ * √(Q / V)

• σ = daily volatility
• Q = order size (shares)
• V = average daily volume

This non-linear relationship means doubling your order size increases impact by only ~41% (√2), not 100%.

Unlike linear models, the square root law exhibits concavity — each additional share traded has a smaller marginal impact than the last.

• Small orders: Face proportionally higher impact per share due to fixed spread-crossing costs
• Large orders: Benefit from the square root dampening, making block trading economically viable
• Implication: Slicing a parent order into many small child orders is optimal, as total impact scales sub-linearly with aggregate size

First documented in a 2001 study by Torre and Ferrari, later formalized by Almgren, Thum, Hauptmann, and Li (2005) using a massive proprietary dataset from Citigroup.

Key findings:

• The 3/5 power law (earlier models) was rejected in favor of the square root exponent
• The relationship holds across equities, futures, and FX markets
• Universality suggests a deep structural origin in order book dynamics rather than asset-specific factors

The square root exponent emerges from the latent liquidity hypothesis:

• Order books are not static; hidden and iceberg orders replenish at rates proportional to volatility
• Large trades trigger a liquidity mobilization effect where new limit orders arrive to absorb the imbalance
• This dynamic replenishment creates the sub-linear scaling observed empirically
• Mathematically, it arises from the interplay between diffusive price dynamics and strategic order placement

The square root law is embedded directly into optimal execution schedules:

• Almgren-Chriss extensions: Replace linear temporary impact with square root temporary impact for more realistic trajectories
• Pre-trade cost estimation: Used by TCA platforms to forecast implementation shortfall before routing
• VWAP and TWAP algorithms: Adjust participation rates based on the non-linear impact curve to minimize slippage
• Liquidity-seeking algos: Dynamically switch between dark and lit venues when predicted impact exceeds thresholds

The basic square root model has known boundary conditions:

• Very small orders: Impact becomes linear as fixed costs (spread, fees) dominate
• Extreme participation rates: Above ~30% of daily volume, the square root relationship breaks down as the order becomes the market
• Illiquid securities: Thinly traded assets exhibit steeper exponents due to sparse latent liquidity
• Modern extensions: Incorporate intraday volume profiles, spread cross-effects, and order book resilience parameters for improved accuracy

A foundational optimal execution framework that formalizes the trade-off between market impact costs and timing risk. It models both permanent and temporary impact, using a mean-variance optimization to derive an efficient trading trajectory. The model assumes a linear temporary impact function, which contrasts with the concave nature of the Square Root Law, but provides a closed-form solution for minimizing the sum of impact costs and the volatility risk of holding the position.

The lasting, non-reverting change in an asset's equilibrium price caused by a trade that conveys new information to the market. Unlike temporary impact, this component is linear with trade size and represents the market's revised estimate of the asset's fundamental value. The Square Root Impact Law primarily models the temporary impact component, while permanent impact is often modeled separately as a linear function of signed order flow, as in Kyle's Lambda.

The transient price concession required to attract liquidity and execute a trade quickly. This cost component reverses after the order is completed as the limit order book replenishes. The Square Root Impact Law is an empirical description of this phenomenon, showing that the cost grows with the square root of trade size rather than linearly. Key characteristics include:

• Dissipates over a characteristic decay time
• Depends on current order book depth
• Is the primary target of optimal execution slicing

The comprehensive measure of execution quality, defined as the difference between the decision price and the final execution price, including both explicit commissions and implicit costs. The Square Root Impact Law is a critical input for pre-trade cost estimation models that forecast the market impact component of this shortfall. Decomposing implementation shortfall reveals contributions from:

• Delay cost: price movement before trading begins
• Market impact: the cost modeled by the Square Root Law
• Opportunity cost: the cost of unexecuted shares

A measure of market illiquidity derived from Kyle's 1985 model, representing the linear relationship between order flow imbalance and permanent price change. While the Square Root Impact Law captures the concave temporary impact, Kyle's Lambda captures the linear permanent impact. Together, they form a complete picture of price formation. A higher lambda indicates a thinner market where trades cause larger permanent price moves, often due to a higher probability of informed trading.

The fraction of total market volume that an execution algorithm targets, directly influencing the aggressiveness of the strategy. The Square Root Impact Law implies that impact costs scale with the square root of the participation rate. A higher participation rate accelerates execution but incurs disproportionately higher impact costs. Common strategies include:

• Low participation (5-10%): Minimal impact, longer horizon
• VWAP/POV: Dynamically adjusts to match volume patterns
• High urgency: Accepts higher impact to complete quickly