Maker or Taker: The Decision That Costs Basis Points

Every order you send is either adding liquidity to the book or removing it, and the maker taker fee difference is the price the venue puts on that distinction. The gap is usually measured in basis points, which makes it feel trivial. It is not trivial, because it is charged on every trade you will ever make, and it compounds with turnover rather than with conviction.

The maker taker fee split also encodes an assumption: that resting orders provide a service and aggressive orders consume one. Once you see the split that way, the choice stops being a personality trait — some traders “are” makers — and becomes a per-order calculation.

• The fee is only part of the cost. Crossing the spread and accepting slippage usually cost more than the taker fee itself.
• Maker orders are not free. They pay in adverse selection and in trades that never happen, both of which are invisible on a fee statement.
• The right choice varies by order, not by trader. A hedge and a patient entry into the same position deserve different execution.
• Turnover amplifies whatever you choose. A one basis point difference on 500,000 of monthly notional is small; on 50,000,000 it decides whether a strategy survives.

The maker taker fee model in one paragraph

A maker taker fee structure charges two different prices for the same trade depending on whether your order rests in the book before execution. Makers post orders that wait; takers send orders that execute against them immediately. Venues typically charge takers more, because a taker consumes the liquidity that makers supplied, and they sometimes charge makers a negative fee — a rebate — to keep quotes competitive.

Two consequences follow. First, the effective cost of a trade is the fee plus the spread plus slippage, and the fee is often the smallest of the three. Second, any strategy that assumes a maker fee is available must also assume its orders fill, which is a different and stronger assumption.

What you are actually paying for as a taker

You are paying for certainty of execution. A market or aggressive limit order converts your intention into a position in the same block, at a price the book has already published.

That certainty has a price beyond the fee. The quoted spread is the difference between the best bid and offer, and crossing it costs half the spread on average for a small order. On a liquid pair that might be a fraction of a basis point; on a thinner pair it can exceed the entire fee. Slippage on size is separate again, and it grows faster than order size because depth thins as you walk the book.

So the honest taker cost is: fee + half-spread + slippage. On small orders in deep markets, the fee dominates. On large orders in shallow markets, it does not.

When maker orders are cheaper, and when they are not

Resting an order avoids the taker fee and may earn a rebate, but it introduces two costs that do not appear on any statement.

Adverse selection

A resting order fills when someone decides to trade against it, and that someone usually has a reason. If your bid fills, it is disproportionately likely that the price is about to fall. This is not bad luck; it is the mechanism by which makers are compensated, and it means the realised cost of a maker fill is worse than the quoted price suggests.

The practical measure is the mark-out: the mid price a short interval after each fill, compared with your fill price. If your maker fills are consistently negative on a five-minute mark-out, the rebate is not covering the selection cost.

Non-execution

An order that does not fill is not a cheap trade; it is no trade. If your reason for the position was time-sensitive — a hedge, a risk reduction, an exit from a deteriorating position — a passive order can cost far more by failing than it saves by resting.

This is the failure mode fee comparisons hide. A missed hedge never appears in the fee line, so a maker-only policy can look efficient while leaving risk unhedged.

Comparing maker and taker across order types

The fee side is determined by how the order executes, not by what you intended.

The maker taker fee decision is a decision about which risk you would rather hold: the risk of paying too much, or the risk of not trading at all.

A worked month of maker taker fees

Two hundred trades a month at 2,500 of notional each is 500,000 of monthly volume. Assume a taker fee of 0.08% and a maker fee of 0.02% with a 0.002% rebate.

All-taker: 500,000 × 0.0008 = 400 in fees.

Mostly-maker, with a 90% fill rate: 450,000 fills as maker, so 450,000 × 0.0002 = 90 in fees, less a rebate of 450,000 × 0.00002 = 9, giving 81. The remaining 50,000 has to be taken, at 50,000 × 0.0008 = 40. Total 121.

The fee comparison says the maker route saves 279 a month, or about 3,350 a year. That is the number a fee schedule would show you.

Now price the selection cost. If maker fills are on average one basis point worse than the mid five minutes later, the cost is 450,000 × 0.0001 = 45. Net saving falls to 234 a month. Add the 10% of orders that did not fill and had to be chased at a worse price — say half a basis point on 50,000, or 25 — and the saving is about 209 a month, or 2,500 a year.

Still worth having, but the decision was made by fill rate and mark-out, not by the fee table.

Choosing per order, not per identity

1. Classify the order by purpose: risk reduction, entry, or exit. Risk reduction is usually taker work.
2. Measure the quoted spread on the pair at the times you actually trade, not at midday.
3. Estimate slippage for your size from recent book depth, and add it to the taker cost.
4. For the maker alternative, estimate a realistic fill rate from your own history rather than the theoretical one.
5. Measure mark-out on your maker fills over a consistent interval to quantify selection cost.
6. Compare the two totals per order, and default to the cheaper one only when the purpose is not time-sensitive.
7. Review monthly, because fill rates and spreads change with volatility.