Aggressive vs Passive Execution: When to Cross the Spread and When to Wait

Strip any execution decision down far enough and what’s left is a choice between two sides of the spread.

Aggressive: cross the spread now, pay the half-spread, get filled immediately, and don’t worry about whether the price is “fair.” Passive: post on your side of the book, earn the half-spread, wait for the market to come to you, and hope you fill before the price runs away.

This entire article is about that one trade-off. Urgency, spread width, the information in your hands, your inventory, the time of day — every one of them is just another weight placed on the same scale. Get the weighing right and you save 5-10 bps per trade. Get it wrong and you either miss the fill or pay through the nose.

Harris (2003, Ch. 20) supplies the framing worth memorizing: aggressive orders buy immediacy for themselves at the cost of paying the spread; passive orders provide liquidity to others at the cost of timing risk. The trade between those two is the most fundamental decision in execution.

The Scale Itself: Half-Spread Versus Timing Risk

First, weigh the two pans properly. The quoted spread on a US large-cap is typically 1 cent: on a $50 stock that's 2 bps, on a $500 stock (BRK.A) it’s 0.2 bps, and on a $5 small-cap the same 1 cent is 20 bps. Identical absolute spread, hundredfold difference in relative cost — keep that conversion running through everything below.

The spread is what you pay when you cross and what you earn when you provide. Cross to buy and you pay (offer − mid); provide by posting at the bid and you earn (mid − bid) when filled. So, conditional on both orders filling at today’s quotes:

Passive comes in half a spread cheaper — if you fill. That “if” is the other pan of the scale: a passive fill is conditional. If the price moves up before someone hits your bid, your eventual passive fill is worse than crossing would have been; if it moves down, you got a bargain. Posting passively banks a certain half-spread against an uncertain fill and an uncertain path — better in expectation, worse in variance. The whole aggressive-versus-passive question is what you’ll pay to make that variance go away.

The First Weight: Urgency

The heaviest weight on the scale is urgency, because urgency is what prices the waiting.

High urgency — a portfolio manager screaming for the fill NOW — makes waiting worthless: cross, pay the half-spread, be done. Low urgency — the order can sit for hours — makes waiting valuable: post, earn the spread, let the market come to you. Push to the limits: a trader completely indifferent to timing risk should always post passively (earning the half-spread on average); a trader maximally averse to it should always cross (killing the variance). Most institutional traders live in between, and the Almgren-Chriss optimal trajectory formalizes exactly this: the urgency parameter $\lambda$ trades expected cost against variance, quantitatively.

The Second Weight: Fill Probability

The passive half-spread isn’t free money; it gets multiplied by a fill probability. That probability is set jointly by distance from mid (farther is worse), market activity (more volume, more fills at your level), volatility (higher vol sweeps your price more often — good for fills), spread width, and where you sit in the queue.

For rough magnitude: at the touch in a liquid large-cap, fill probability runs about 50% within a minute; a few cents more aggressive than the touch, about 30%; a few cents more conservative, about 10%. The numbers shift by stock and by hour — higher at the open when volume is heavy, lower mid-day, up again at the close — but the magnitude is enough to see the point: the passive half-spread comes discounted by half or worse. The formal version of this accounting comes later.

Placing the Weights: The Standard Situations

With urgency and fill probability as the two axes, most situations read off directly.

The reasons to cross all point the same way: waiting costs more than the half-spread. Hard urgency is the obvious one — news just landed, alpha is decaying by the second, and the half-spread is a small price for certainty. A tight spread is another — 0.5 cents on a $50 stock is 1 bp, savings too small to justify any timing risk. A small order (1,000 shares against 5M ADV) won’t move the market; cross and move on. A strongly trending market — the stock is moving 1% a minute and you want to buy — turns sitting at the bid into donating 1% per minute. Closing a position, time matters more than price. And trading on information that’s about to hit the price — covered separately below, because it’s the deepest reason of all.

The reasons to post all point the other way: waiting is cheap, or is itself the point. Low urgency, an order that can sit for hours. A wide spread — 5 cents on a $50 stock is 10 bps, real money for waiting. Moderate size (10,000 shares against 1M ADV), small enough to rest without missing the market. A range-bound tape — a stock bouncing between $49.95 and $50.05 all day fills a $49.99 bid over and over with little timing risk. Patient position-building. And strategies that exist to earn the spread — for those, passive posting isn’t an execution tactic, it’s the strategy itself.

Real execution algos almost never stand at either pole. The typical liquidity seeker posts passively at the touch to earn the spread, pings dark pools for hidden size, and flips aggressive the moment an attractive hidden order shows up — choosing a side of the scale moment by moment. For moderate-size orders in liquid names, the hybrid wins most of the time: pure aggressive is too expensive on average, pure passive too dangerous in the tails.

Information: The Weight That Breaks the Scale

Everything above quietly assumes you know no more than the market does. Private information changes the arithmetic.

Aggressive with information is straightforward: you want the position before the price reflects what you know, and the half-spread is a small fee for moving first. Passive with information is a trap: you’re offering to trade at a price you yourself know is wrong. Know the stock is worth $52 and post a bid at $50 — you’ll fill, fine. Know it’s worth $50 and post an offer at $52 — nobody comes. Information makes passive posting one-sided: you fill only on the side that hurts you. So informed traders lean systematically aggressive, and passive posting makes sense only when you hold no strong information.

Harris (2003, Ch. 21) treats this under “information-motivated trading”: informed traders take liquidity because they need to move fast; uninformed traders provide it because they have time and want the spread. The result is a selection effect — the people crossing the spread are, on average, the better informed. Which is precisely why liquidity providers earn the spread at all: it’s compensation for trading against better-informed counterparties.

That selection effect has a second-order corollary: aggression is itself a signal. A trader who consistently crosses the spread is broadcasting urgency or information; other participants read the broadcast and trade ahead, so aggressive execution triggers impact beyond the bare spread cost. This is why execution algorithms randomize their aggression — cross at unpredictable moments and the signal weakens; sophisticated counterparties can’t easily infer intent. Harris (2003, Ch. 21) closes the logic: aggressive execution is informative because it costs the trader the spread — the informed will pay that cost and the uninformed won’t, so observing aggression is observing information.

Inventory: Your Own Position Is a Weight Too

A market maker’s aggression tracks inventory: long inventory means you want to sell, so post offers passively and earn the spread — never cross to buy more; short inventory means you want to buy, so post bids; flat means indifferent — aggressive when needed, passive when possible.

For an institutional execution desk, “inventory” means something else: the parent order’s remaining size. Working a 100,000-share buy with 50,000 done, you’re carrying “50,000 shares still to fill” — and that argues for more aggression, because the remaining work is real, the deadline is approaching, and the cost of missing the order is rising. Most execution algos encode this as an “inventory urgency” parameter: aggression ratchets up while the unfilled portion stays large or the deadline closes in.

Spread Width: The Savings Scale With It

Back to the spread itself, with the conversion made concrete. US large-cap, $50 stock, 1-cent spread: 2 bps — passive saves 2 bps against real timing risk; marginal. US mid-cap, $30 stock, 5-cent spread: 17 bps — worth waiting for. US small-cap, $10 stock, 10-cent spread: 100 bps — with any time at all, absolutely worth it.

The pattern: passive’s payoff scales linearly with spread width, while the timing risk doesn’t scale the same way. Hence the short rule of thumb — spread above 5 bps and time in hand, post; spread under 2 bps, the savings don’t cover the risk.

Time of Day: The Scale Moves All Session

Liquidity has an intraday rhythm, and the same order has different right answers at different hours.

The opening auction (9:30) brings heavy volume, wide spreads, and jumpy prices: aggressive is risky because the price is unstable, passive is risky because you might miss the open — which is why most institutional flow waits 5-15 minutes after the open for the market to settle. Mid-morning (10:00-11:30) has the tightest spreads and deepest liquidity; both styles work. Mid-day (11:30-14:00) volume thins and spreads widen — and wide spreads are exactly what makes passive pay. The afternoon (14:00-15:30) brings volume back; conditions are mixed. The closing auction (15:30-16:00) is a massive volume spike, especially in ETFs and large-caps, and it’s effectively a separate market with its own liquidity dynamics: aggressive orders make it into the closing print, passive ones may not.

So the same 60-minute order lands differently by start time: begun at 10:30 it works through mid-morning, where either style is fine; begun at 12:00 it works through wide mid-day spreads, where passive is distinctly more attractive.

Order Types: Writing Your Posture Into the Order

Aggression isn’t a binary, and modern order types are essentially notches along the dial. A market order is pure aggression: cross now. A limit order at the touch is pure passivity: rest at the best bid or offer. A limit inside the touch is a hybrid — more aggressive than the best, faster fills, possibly no spread earned. A limit outside the touch is more conservative still: pure passive, slow fills. Pegged orders track the touch or the mid automatically — passivity with maintenance included; a mid-point peg rests exactly at mid, earning no spread and paying none, the precise center of the scale. A discretionary order displays at the touch while hiding a more aggressive limit — another hybrid. An iceberg displays a fraction of total size, hiding the size rather than the price. The choice follows your aggression preference and the stock’s liquidity.

Putting Numbers on the Decision

For day-to-day use, five questions suffice:

1. How urgent? High → aggressive. Low → passive.
2. How wide is the spread? Wide → passive pays more. Tight → aggressive.
3. How large is the order? Small → aggressive is fine. Large → hybrid.
4. Is the market trending? Yes → aggressive. No → passive.
5. Do I have information? Yes → aggressive. No → passive is fine.

When combining, remember the weights aren’t equal: with high urgency, spread width stops mattering — cross; with low urgency and a tight spread, you have time anyway — post.

For the rigorous version, write out both expected costs. Aggressive:

where $s/2$ is the half-spread and impact is the concession from sweeping levels (near zero for a small order). Passive, bringing in the fill probability $p$, the opportunity cost of missing $\Delta_{\text{miss}}$, and the favorable price drift that typically accompanies a passive fill (the market moves toward you — that’s why you filled):

For a small order (impact near zero), the two break even when:

Solving for $p$ gives the minimum fill probability at which passive is preferred:

Actual fill probability above $p^*$: passive wins. Below: aggressive wins. Nobody runs this calculation order by order — traders use the five questions — but the math exposes the skeleton of the trade-off: passive’s value rises with fill probability and falls with the cost of missing, and the magnitudes from the fill-probability section (about 50% at the touch) are exactly what you’d plug in.

In daily practice the discipline comes down to a few lines. For the trader: default to hybrid; ratchet aggression up with urgency — hard deadline, cross now, soft deadline, start passive; match aggression to spread width; watch the clock, since the hour moves both spreads and fill probabilities; and don’t fight the trend — when the stock is running against you, cross and be done. For the algorithm designer, the same discipline becomes mechanism: aggression must be a tunable parameter (one-size-fits-all fails), must adjust itself over time (rising as the deadline nears), should be randomized within bounds (anti-gaming without sacrificing execution quality), and should be transparent — the trader ought to see what the algo is doing, and why.

References

• Harris, L. (2003). Trading and Exchanges: Market Microstructure for Practitioners. Oxford University Press. Ch. 20-21. ISBN: 978-0-19-514470-3.
• Almgren, R., & Chriss, N. (2000). “Optimal Execution of Portfolio Transactions.” Journal of Risk, 3(2), 5-39.
• Bertsimas, D., & Lo, A. W. (1998). “Optimal Control of Execution Costs.” Journal of Financial Markets, 1(1), 1-50.