Market makers, disclosed: where liquidity comes from on a young venue
Every market you have ever traded opened empty. No depth, no quotes, nobody on the other side — a book with nothing in it and a question nobody has priced.
Something has to happen between that moment and the screen you actually saw. The category’s marketing skips this chapter, and it is the wrong one to skip: how a venue’s first liquidity arrived — and whether it will tell you — is one of the sharpest tests of its honesty a reader gets.
The empty book
A market’s book is a stack of standing offers — makers posting them, takers accepting them (Where a market price comes from). That mechanism has a bootstrap problem: it explains how a busy book works, not how a book becomes busy.
The first person to post a resting offer on a quiet question is doing dangerous work. They commit to a price while nobody else is quoting — no consensus to lean on, no depth to exit into — and their offer sits exposed to everyone who knows something they do not. When news lands, the fastest informed trader takes their standing offer before they can move it.
So young books do not fill up by themselves. Someone has to go first, on purpose, on both sides, all day. That role has a name: the market maker.
What a maker actually does
A market maker is an always-on counterparty. It posts a buying price and a selling price on the same question, keeps both alive through the hours when nothing is happening, and absorbs trades it did not choose — including, unavoidably, trades against people who knew more.
Why would anyone volunteer? Because the two prices differ, and the gap is the pay. A canonical model of market making showed four decades ago that this gap — the spread — is not a fee bolted on top; it is what lets an always-on counterparty survive at all (Glosten and Milgrom 1985). A dealer quoting prices to a crowd that includes better-informed traders must buy slightly low and sell slightly high just to break even, because some of every day’s flow is exactly the flow it wishes it had refused.
The empirical record puts numbers on it. On a large licensed U.S. venue, researchers measured the two roles across more than 300,000 contracts: takers give up roughly three times as much as makers do across the sample as a whole, and on the cheapest, least likely contracts — where both roles lose on average — takers lose hardest (Bürgi, Deng and Whelan 2025, working paper). Order-book anatomy from the largest on-chain venue documents a longshot spread premium — spreads widest on the improbable outcomes, where a maker is most exposed — with maker participation broad but carried by a concentrated tail (Dubach 2026, preprint).
None of that is a scandal. It is the same economics as every exchange humans have built: immediacy is a service, and the spread is its price.
Serious exchanges pay for this openly
In regulated finance, first liquidity is not left to chance, and it is not hidden. It is contracted, disclosed, and studied.
On the Paris Bourse and later Euronext, listed companies whose books were too quiet have formally hired designated market makers — firms contractually obliged to keep quoting. The research found the arrangement measurably valuable: for the less liquid listings, an abnormal price response of nearly 5% around the announcement, with young and small firms the most likely to sign up (Venkataraman and Waisburd 2007). A companion study of 74 small-cap listings that hired one found the designated maker takes losses on the days its obligation is most likely to bind — the days liquidity is hardest to provide — hard evidence that the service is real and really paid for (Menkveld and Wang 2013).
Prediction markets have their own version of this lineage, and it is more explicit still. The field’s canonical answer to the empty book is an automated market maker the venue itself can run and subsidize, quoting every outcome at every moment, with its maximum possible loss fixed and budgeted in advance (Hanson 2007). The subsidy is not a trick to be discovered — it is the published mechanism: a known, bounded cost accepted so that there is always a price, refined in the open literature for two decades.
So the existence of an operator’s own book is not the scandal either. On a young venue it is nearly unavoidable: in the beginning, there is no one else to be the maker.
The word everyone reaches for
The word readers reach for at this point is the casino one — the house.
The word is wrong in an instructive way, but the worry underneath it is the right worry: is the venue trading against me with advantages I cannot see? A book where the operator quotes with private knowledge of the flow, or hidden priority, or invisible size is not a market — it is a performance of one. The problem was never that an operator provides liquidity. The problem is an operator that provides liquidity while pretending it is a crowd.
The honest architecture
So judge the structure, not the vocabulary. Disclosed operator liquidity has a recognizable architecture, and it has three properties.
Ring-fenced. The operator’s book runs at arm’s length from the exchange: its own account, its own limits, no priority over anyone else’s orders, and no sight of anything the public book does not show. The venue’s job is to run the market; the maker’s job is to quote it. An honest design keeps a wall between the two even when the same company owns both.
Disclosed. It is named in the rulebook, not discovered by forensic accounting. A reader should be able to find, in writing: that the venue quotes markets on its own platform, under what constraints, and at whose risk.
Designed to fade. Scaffolding goes up in the open, holds the building while it cannot stand on its own, and comes down in the open. Operator liquidity on a young venue is scaffolding: its share of the book should shrink as outside makers arrive, because the venue’s long-term product is the market, not the position. An operator’s book designed to fade is infrastructure. One designed to stay hidden is a counterparty you were never told about.
How to tell
You cannot audit a venue’s servers. But disclosure has a smell test, and it costs one read of the documents.
Does the venue say, anywhere, whether it quotes markets on its own platform? A yes with details beats a no without them — on a young venue, “we have no maker” too often means “we would rather not discuss the maker.” Is the operator’s trading segregated and constrained in writing? Will anyone say how much of current volume is operator liquidity — even approximately, even just “most of it, for now”? That last answer is uncomfortable, which is the point. A venue willing to be uncomfortable in public is disclosing. A venue that answers in generalities is asking you not to look.
The sentence a venue owes you
A venue built this way has an honest sentence available to it, and it is short: we do well when you trade. The maker is paid in spread for carrying risk. The exchange is paid in fees for keeping the book fair. Both are paid by trading.
A young venue cannot promise you a deep book on day one — nobody can. What it can promise, from day one, is to tell you what is holding the book up.
Before you trade a young market — on any venue — find out what the scaffolding is, and who will say so out loud. Honest scaffolding is visible by design. The invisible kind is rarely missing — just undisclosed.
Sources
- Glosten, L. R., Milgrom, P. R. (1985). “Bid, Ask and Transaction Prices in a Specialist Market with Heterogeneously Informed Traders.” Journal of Financial Economics 14(1): 71–100. sciencedirect.com
- Hanson, R. (2007). “Logarithmic Market Scoring Rules for Modular Combinatorial Information Aggregation.” The Journal of Prediction Markets 1(1): 3–15. ideas.repec.org
- Venkataraman, K., Waisburd, A. C. (2007). “The Value of the Designated Market Maker.” Journal of Financial and Quantitative Analysis 42(3): 735–758. cambridge.org
- Menkveld, A. J., Wang, T. (2013). “How do designated market makers create value for small-caps?” Journal of Financial Markets 16(3): 571–603. sciencedirect.com
- Bürgi, C., Deng, W., Whelan, K. (2025). “Makers and Takers: The Economics of the Kalshi Prediction Market.” Working paper, UCD Centre for Economic Research WP25/19. ucd.ie
- Dubach, P. D. (2026). “The Anatomy of a Decentralized Prediction Market: Microstructure Evidence from the Polymarket Order Book.” Preprint, arXiv:2604.24366. arxiv.org