Maker-Taker Fees and the Book They Produce

A maker order adds to the book and waits. A taker order removes from it immediately. Venues charge these differently — takers more, makers less or negative — and the gap between the two rates does more to shape a market than any other parameter a venue controls.

Why venues pay for liquidity

A venue with no resting orders is not a market. The first problem any exchange has is attracting participants willing to quote before there is volume to justify quoting, and a fee rebate is the direct solution: pay people to post, and the book fills.

Once liquidity exists, it attracts volume, which produces taker fees, which fund the rebates. The model is self-sustaining as long as taker flow exceeds what the rebates cost, and the spread between the two rates is the venue's margin.

That is a coherent design. It is also why comparing venues on their headline taker fee alone is misleading — a venue with a high taker fee and a large rebate may have far better depth than one with a low taker fee and none.

What the rebate manufactures

The complication is that a rebate rewards posting rather than being willing to trade, and those are not the same thing.

An order posted primarily to earn a rebate is intended to be cancelled if the market moves toward it. It counts as displayed depth, it makes the book look thick, and it withdraws precisely when someone wants to use it. This is not deception; it is the rational response to being paid for presence rather than for execution.

The practical consequence is that displayed depth on a heavily rebated venue overstates what you can actually trade against, and the overstatement is largest during volatility. A trader sizing an order from the book is reading a number partly generated by an incentive.

This is measurable. Fill rate against displayed size, and how far the price moves relative to what the book predicted, both reveal the gap. Venues do not publish it; anyone trading at size can compute it for themselves.

The tier structure and who it favours

Fee schedules are tiered by volume, and the tiers span a wide range. The top tier frequently pays makers a rebate and charges takers very little, while the bottom tier pays nothing and charges several times more.

The effect is that the same trade costs materially different amounts depending on who executes it, and the difference is larger than most retail traders assume. A strategy that is marginally profitable at top-tier rates can be firmly unprofitable at bottom-tier ones, which is a large part of why published backtests rarely survive contact with a retail account.

It also means fee comparisons between venues are only meaningful at a stated volume. A venue that is cheapest at high volume can be the most expensive at low volume, and both facts appear on the same page.

Where the model does not apply

Automated market makers have no maker-taker distinction because there are no resting orders — liquidity providers earn a share of fees paid by every trade regardless of direction, and there is nothing to cancel.

That removes the phantom-depth problem entirely: pool liquidity is capital actually committed, and it cannot be withdrawn in the moment before your trade lands. It introduces the opposite issue, which is that the capital cannot be repositioned quickly when it should be.

The comparison is therefore not which venue has better fees but which failure you prefer: depth that can vanish, or depth that cannot get out of the way.

Reading a fee schedule properly

• Find your actual tier, not the headline rate. The gap between top and bottom tiers is larger than the gap between most venues.
• Note whether the maker rate is a reduced fee or a genuine rebate, since the second manufactures more phantom depth than the first.
• Measure fill rate against displayed size rather than trusting the book, particularly during volatile periods.
• Include withdrawal and funding costs. A venue with excellent trading fees and expensive withdrawals is not cheap.
• Compare total cost at your volume, and re-check it if your volume changes — the ranking moves.