Market Microstructure

Market microstructure is the branch of financial economics that studies how securities are actually traded: how orders are submitted, matched and executed, who provides liquidity, and how the rules of a trading venue shape prices and costs. Where much of finance theory treats prices as if they simply appear, microstructure looks at the process by which they are formed, order by order. Its findings are used by exchanges designing their markets, by traders and brokers managing execution, and by regulators writing rules on transparency and fairness.

The field covers every traded asset, including shares, bonds, foreign exchange, derivatives and crypto-assets, though most research has been done on equity markets where data are richest. It overlaps with the practical subject of market structure, which describes the venues and intermediaries that exist in a given market and the regulation that governs them.

Microstructure research is usually organised around a few themes. The first is price discovery: how new information about an asset's value is incorporated into its price through trading. The second is liquidity: how easily and cheaply a position can be bought or sold, and what determines that. The third is transaction cost, both the visible parts such as commissions and the implicit parts such as the bid-ask spread and market impact. The fourth is market design: how different trading rules change outcomes for participants.

Markets differ in the mechanism that brings buyers and sellers together. The main forms are:

• Order-driven markets. Participants submit orders to a central order book, and a matching engine executes them by price and then time priority. Most stock exchanges work this way during continuous trading.
• Quote-driven markets. Dealers post prices at which they will buy and sell, and investors trade against them. Much bond and currency trading follows this model, often through a request for quote.
• Call auctions. Orders are collected over a period and executed together at a single price, as in an opening auction or closing auction.
• Dark and negotiated trading. Orders are not displayed before execution, as in dark pools, or are agreed bilaterally and reported afterwards.

Real markets combine these forms. A share may open in an auction, trade continuously on several competing venues during the day and close in another auction, with large orders negotiated away from the book.

A central result is that the spread between buying and selling prices compensates liquidity providers for three things: the cost of processing orders, the risk of holding market maker inventory, and the risk of trading with someone who is better informed. The last of these, known as adverse selection, was formalised in the mid-1980s in models by Albert Kyle and by Lawrence Glosten and Paul Milgrom. In these models a market maker cannot tell informed traders from uninformed ones, so it sets prices to break even on average, losing to the informed and recovering the loss from everyone else through the spread.

These models explain several observed features of markets: spreads widen before announcements, when the chance of informed trading is higher; large orders move prices more than small ones; and prices adjust to the direction of trading even when no public news has arrived. They also underpin the measurement of trading cost against benchmarks.

Practitioners apply microstructure in execution. Algorithmic trading strategies split large orders to limit their footprint, and order routing systems choose among venues according to price, fees and the likelihood of execution. High-frequency trading firms, which now supply much of the displayed liquidity in equity and futures markets, compete on speed and on models of short-term order flow.

Venues use the same knowledge when setting parameters. The tick size, the minimum price increment, affects how tight spreads can be and how much depth is displayed at each price. Fee schedules that pay rebates to liquidity providers, rules on order types, and safeguards such as a circuit breaker are all design choices with measurable effects on market quality.

Much securities regulation is applied microstructure. In the United States, Regulation NMS links competing stock venues through rules on displayed quotes. In the European Union, MiFID II sets transparency requirements before and after trades, caps on dark trading and a harmonised tick size regime. Best execution obligations require brokers to take execution quality into account when handling client orders.

Debates continue over the fragmentation of trading across many venues, the fairness of speed advantages, the growth of trading away from public exchanges, and the effect of practices such as payment for order flow on retail investors. The spread of electronic trading into bond and currency markets has extended these questions beyond equities.