Market Maker

A market maker posts a bid at which it will buy and an ask at which it will sell, each for a stated size. Revenue accrues only when the two sides eventually match: buying at the bid from a seller who wants immediacy, then selling at the ask to a later buyer, capturing the difference on the round trip. Because the arrivals of buyers and sellers are not synchronised, the dealer holds an unwanted position in between, and that inventory is the core operational problem.

Quote management is therefore inventory management. A dealer that has accumulated a long position will shade both its bid and its ask downward, making it more attractive to buyers and less attractive to sellers, so the flow it attracts pushes inventory back toward target. Residual exposure is often neutralised with futures, options or correlated securities rather than by waiting for offsetting order flow.

The spread compensates three distinct costs. Order processing covers technology, exchange fees, clearing and the cost of maintaining a quoting operation. Inventory holding covers the price risk and financing cost of positions the dealer is obliged to absorb; the more volatile the security and the thinner the flow, the wider this component must be. Adverse selection covers the losses incurred when the counterparty is better informed, because a dealer's passive quotes are systematically hit by traders acting on news the dealer has not yet seen.

This decomposition explains most observed spread behaviour. Large, heavily traded stocks with dispersed retail flow trade at pennies wide; small caps, distressed credits and securities around earnings or macro releases trade far wider. When volatility spikes, both the inventory and adverse-selection components rise at once, which is why spreads and quoted depth deteriorate together rather than independently.

Market-making structures differ by venue. Order-driven electronic markets match public limit orders directly, but professional liquidity providers still supply most of the resting depth, sometimes under formal obligations attached to a designated status and rebate incentives. Quote-driven markets, typical of corporate bonds, many sovereign issues, currencies and most OTC derivatives, have no central book at all: the executable price is whatever a dealer quotes on request, and a client's ability to trade size depends entirely on dealer balance sheet capacity.

Exchange-traded funds show the mechanism most clearly. Market makers quote the fund's shares intraday while arbitraging any deviation from the value of the underlying basket, creating new shares when the fund trades rich and redeeming when it trades cheap. That arbitrage, not the fund sponsor, is what keeps ETF prices tethered to net asset value, and it weakens whenever the underlying basket itself becomes hard to trade.

You should be able to move fluently between quotes and cost measures. The quoted spread is the ask minus the bid; the midquote is their average and serves as the reference fair value. Expressing the spread as a percentage of the midquote makes securities comparable. The effective spread for a single trade is twice the absolute difference between the execution price and the midquote prevailing when the order arrived, which captures both price improvement inside the quote and the slippage suffered when an order is larger than the displayed size.

Equally important are the conceptual distinctions. A broker acts as agent and is paid a commission; a dealer acts as principal and is paid through the spread, taking the other side and the risk. Liquidity itself has several dimensions, and a tight spread on a small quoted size is not the same as genuine depth. Candidates should also be able to explain why market makers are risk intermediaries rather than forecasters, and why their withdrawal of size amplifies price moves in stressed markets.