Skip to content

Market Making

Definition

Market making is the activity of simultaneously quoting bid and offer prices for an asset and earning the spread from trades that occur against those quotes. The market maker provides liquidity in exchange for being compensated for two risks:

  • Inventory risk — accumulating a directional position the market maker did not want.
  • Adverse selection — being systematically picked off by better-informed traders (see adverse-selection).

Modern market making is almost entirely algorithmic.

Two paradigms

1. Limit Order Book (LOB) market making

Quote passive limit orders on both sides of the book; earn the spread when both sides fill. This is the dominant model on traditional exchanges (equities, futures, FX, options).

Key decisions: - Quote prices (distance from mid) — wider = less adverse selection, but lower fill probability. - Inventory management — skew quotes asymmetrically when inventory drifts. - Cancellation/replacement — respond to order-flow imbalance signals.

Foundational models: - Avellaneda & Stoikov (2008) — stochastic-control model that derives optimal bid/ask quotes from an inventory penalty and a terminal utility. - Guéant–Lehalle–Fernandez-Tapia (2013) — asymptotic expansion, closed-form solutions for high-frequency limits. - Glosten–Milgrom (1985) — information-based spread from adverse selection.

2. Automated Market Maker (AMM) liquidity provision

Deposit tokens into a smart contract with a bonding curve \(f(x, y) = K\); the contract mechanically trades on your behalf according to the curve. This is the DeFi paradigm (Uniswap, Balancer, Curve).

Key structural facts: - An LP position is a derivative of the pool assets, not a passive deposit. - LPs earn fees but incur Impermanent Loss (IL) — value lost to price divergence vs simple buy-and-hold. - LPs also lose to informed arbitrageurs via Loss-Versus-Rebalancing (LVR) — the AMM analogue of adverse selection.

See cfmm-liquidity-provision-pricing for a rigorous derivatives-pricing framework that characterises LP payoffs and IL as option strips.

Comparison

Dimension LOB market making AMM liquidity provision
Venue Traditional exchanges DeFi / smart contracts
Pricing rule Market maker sets quotes dynamically Deterministic bonding curve
Inventory control Active — skew, cancel, reprice Passive — dictated by curve
Adverse selection Glosten–Milgrom spread Loss-Versus-Rebalancing
Revenue Spread + rebates Pool fees
Capital efficiency High (fractional inventory) Lower (full capital locked)
Risk decomposition Inventory + adverse selection Impermanent Loss + LVR
Typical horizon Sub-second to minutes Minutes to weeks

Connections between the two

Despite very different machinery, both paradigms share the same structural tension: liquidity providers earn a spread/fee but lose systematically to informed flow. The microstructure theory of adverse selection (Kyle 1985; Glosten–Milgrom 1985) re-emerges in DeFi as LVR; the mathematical tools (stochastic control for LOB; option-pricing theory for AMMs) differ, but the economic intuition is the same.

In hybrid settings (crypto markets trade on both centralised order books and DEXs), arbitrageurs bridge the two — e.g., the W/USDT spot vs perpetual tick-size experiment in explainable-crypto-microstructure shows how CEX book imbalance carries information that AMM prices implicitly track.

Open questions

  • What is the optimal market-making strategy in a hybrid CEX+DEX ecosystem where the same asset trades on both venue types?
  • Can LOB-learned inventory-management policies transfer to AMM range-selection in Uniswap v3 concentrated liquidity?
  • How do recent order-book regulations (e.g., persistent-noise-creator penalties in India — see order-flow-filtration) affect market-maker economics?
  • RL and MPC approaches to LOB market making are well-studied (Nevmyvaka et al.; Hendricks–Wilcox); equivalent for dynamic range selection in AMMs is still early-stage.

Connections