Liquidity

Definition

Liquidity is the ability to trade meaningful size quickly without moving the price. It has three classical dimensions: tightness (the spread), depth (size available near the touch), and resiliency (how fast the book refills after a trade). Liquidity is a property of a market at a moment — not a constant — and its disappearance under stress is one of the most consequential facts in trading.

How it works / structure

• Measures: quoted spread and depth (from the order book), Kyle’s lambda (price impact per unit of order flow, Kyle 1985), the Amihud ratio (|return| per dollar of volume, Amihud 2002) for daily data, turnover and average daily volume as coarse proxies.
• Suppliers and their limits: market makers and opportunistic traders supply liquidity when expected spread revenue exceeds adverse-selection and inventory costs; when volatility or funding stress rises, supply is withdrawn — Brunnermeier and Pedersen (2009) model the liquidity-funding spiral.
• Position-size relativity: the practical unit is position size as a multiple of typical volume; the same order is trivial in one name and market-moving in another.
• Simulation parameters: participation-rate caps (order size as % of volume), spread and impact models per ms-slippage-friction, and a liquidity screen (minimum ADV, maximum spread) before any simulated entry.

When it applies

Sizing every position (can it be exited at the modeled cost?), screening instruments for strategy eligibility, stress planning (risk-scenario-analysis should price exits at stressed, not average, liquidity), and reading market regimes — liquidity withdrawal is itself a stress signal.

Risk profile & failure modes

• Stress evaporation: quoted depth is withdrawable in milliseconds; average-condition liquidity measures overstate exit capacity exactly when exits are needed (the 2009 liquidity-spiral literature formalizes why).
• Averaged-away gaps: daily-volume screens hide intraday and session-boundary droughts (ms-sessions-auctions).
• Crowded exits: strategies sharing signals share exit times; modeled per-trader impact understates group impact.
• Two-layer instruments: ETF and option liquidity depend on hedge/underlying liquidity; screen both layers (instrument-etf, ms-option-chain).

Evidence & limits

Kyle (1985) and Amihud (2002) are the canonical impact and illiquidity measures; Amihud also documented that less liquid stocks carried higher average returns in his sample — an illiquidity premium, with the usual post-publication caveats. Brunnermeier and Pedersen (2009) is the standard model of liquidity spirals matching crisis behavior. Precise liquidity forecasting is not established; the platform treats liquidity as a screened constraint, not a predicted quantity.

Falsifiable-thesis examples

Illustrations only, not signals:

• “X’s Amihud illiquidity ratio, at level A today, will remain below 2A on every day of the next quarter” — falsified by the daily series.
• “An exit of position P at 10% participation will complete within three sessions in replay” — falsified by simulated completion time.

Cross-references

• Cost mechanics: ms-bid-ask-spread, ms-slippage-friction
• Session structure: ms-sessions-auctions
• Risk planning: risk-scenario-analysis, risk-max-drawdown-budget
• Instrument layers: instrument-etf, ms-option-chain