Maker vs Taker Rebates

Every fee schedule you have ever read has two columns. One number for orders that rest on the book and get filled; a higher number for orders that hit what is already there. Maker and taker.

Most traders learn the definitions and stop. The interesting part is why the two numbers differ, because that reasoning is what tells you when a maker rebate should actually change how you execute — and when it is a rounding error dressed up as an edge.

The two sides of a fill

Every trade has both. When your limit order rests on the book and someone else's order hits it, you are the maker: you supplied the liquidity, and your counterparty consumed it. When you cross the spread and take an existing order, you are the taker.

A "maker rebate" is a maker fee below zero. The venue pays you for the fill rather than charging you. It is not a promotion or a claim you file — it changes the arithmetic of every fill you post, silently, at the moment of execution.

Why takers pay more

Three reasons stack, and they are worth separating because they decay at different rates.

The venue is buying liquidity. A book with nothing on it is worthless. Someone has to be persuaded to leave capital sitting at a price, exposed, waiting. The asymmetric fee is how the venue routes money from the side that consumes the resource to the side that provides it. New venues lean on this hardest, which is why launch-phase maker rebates are usually the most generous ones you will ever see and also the least durable.

Adverse selection is real. This is the part traders underrate. A resting order does not get filled at random — it gets filled precisely when someone with a reason to trade decides your price is worth taking. Your buy fills when the market is about to go down slightly more often than it fills when the market is about to go up. That is a genuine, structural cost of providing liquidity, and the rebate is partial compensation for bearing it. It is why a maker rebate is not free money even when the fill rate is perfect.

Immediacy has a price. The taker fee is what you pay to be certain your order executes now, at a price you can see. Sometimes certainty is worth far more than a few basis points — a stop that does not fill is not a stop — and the fee schedule is correctly charging you for it.

When a rebate should change your execution

Here is the calculation that actually matters. Posting instead of crossing saves you the taker fee and earns you the maker rebate, so the full swing is:

## the value of getting filled passively instead of aggressively
swing = taker_fee + maker_rebate

At a 4.5 bps taker fee and a 1 bp maker rebate, that swing is 5.5 bps per fill. On $2,000,000 of monthly notional, entirely one-sided, that is $1,100 a month of difference between two execution styles on the same venue with the same trades.

But you do not capture the swing for free. Posting costs you three things:

• Non-fills. The order that does not execute is the one where the market moved away — which is to say, the one you most wanted. If your strategy is directional and time-sensitive, missing a fill costs a multiple of any rebate.
• Queue position. Being on the book is not the same as being at the front of it. In a deep book at a busy price, your order may sit behind size that never clears.
• Adverse selection, as above. Your passive fills are systematically slightly worse than random.

So the honest rule is: a maker rebate changes your execution when your holding period is long relative to the time it takes to get filled. A trader entering a multi-day position can afford to work an order for twenty minutes and should. A trader trying to catch a five-minute move cannot, and for them the entire maker/taker discussion is theoretical.

The second honest rule: if you are already a passive trader, the rebate is not a reason to change anything — it is a reason to check which venue pays you the most for what you already do. That is a venue selection decision, not a strategy one.

Tier cliffs, and why your next dollar is mispriced

Almost every venue tiers its fees by rolling volume, usually 30 days. Cross a threshold and both columns improve for the next period.

Tiers create a discontinuity that makes the marginal value of volume completely different from the average. Consider a trader $400,000 short of a tier that improves the taker fee by 0.5 bps, who expects to trade $3,000,000 next month:

## value of crossing the cliff
saving = 3,000,000 * 0.00005 = $150 for the next month
cost   = 400,000 of extra notional at the CURRENT taker fee
       = 400,000 * 0.00045 = $180

Manufacturing the volume costs more than the tier returns, and that is before slippage. Flip the numbers — a trader who is $40,000 short rather than $400,000 — and the arithmetic reverses entirely: $18 of fees to unlock $150.

This is the single most common way traders lose money to a rewards program: trading size they would not otherwise trade in order to reach a threshold. The rule is mechanical. Compute the cost of the incremental volume at your current rate, including realistic slippage, and compare it against the saving over the period the new tier actually applies. If the tier resets monthly, you get one month of benefit, not a lifetime.

Two further details that quietly change the answer:

• The tier applies going forward, not retroactively. You do not get last month's fills repriced.
• Rolling windows decay. A tier reached on a burst of volume falls away when the burst leaves the window, so the benefit is shorter than it looks.

What to check before you switch venues for a rebate

• Which side does the rebate apply to? Some programs improve the maker column only, some rebate a share of taker fees, some do both. They suit opposite traders.
• Which markets? Rebates frequently exclude the pairs you actually trade. A perp-only rebate is worth nothing to a spot trader.
• Is it a tier or a campaign? A tier is durable. A campaign is a deadline the venue is trying to hit, which means the rate is better and the duration is shorter.
• Depth at your size. A venue paying 2 bps more on a book half as deep will cost you more than it pays on anything above a small clip. Check depth where you actually trade, not at top of book.
• Funding, on perps. If you hold perpetual positions for days, funding differentials between venues are routinely larger than every fee and rebate combined. Optimising the fee line while ignoring funding is optimising the wrong number.

The exchange comparison puts maker and taker fees side by side with live volume and open interest, and the per-venue pages under venues carry the fee notes and program detail. Start there rather than from a fee page screenshot someone posted last quarter.

The uncomfortable summary

For most traders, execution quality beats every rebate program, and the correct use of a fee comparison is to break ties between venues that are already good enough. Fee optimisation matters in direct proportion to turnover: a high-frequency, market-neutral, passive strategy lives and dies on the maker column, and a swing trader holding positions for a week should barely think about it.

Work out your monthly fee spend in dollars first. If the number is small, the entire maker/taker question is a distraction from position sizing. If it is large, the swing between posting and crossing is probably the biggest single lever you have — and it is available on the venue you already use.

More on the wider rebate landscape in crypto exchange fee rebates explained, and on putting fees, points and campaigns in the same units in how to compare crypto exchange rewards.