We are delighted to share with you an in-depth research report on the Financial Exchanges & Market Infrastructure space. Depending on your feedback, we may produce similar research pieces on other industries and industry segments in the future. The goal is to help fundamental, long-term investors expand and deepen their circle of competence. The report is complemented by two podcast episodes, the first one telling the story of the industry and the second one digging deeper into the fundamentals from an investor’s standpoint. Below, you will find two PDF files for download: a research report (members only) and an accompanying slide deck (free to access). We start with a quick anecdote:
In 1997 Jeffrey Sprecher, then a power-plant developer, bought the Continental Power Exchange, a struggling electronic energy-trading platform in Atlanta, for a token sum plus the assumption of its liabilities: a dollar by the oft-told account, a thousand by others, and Sprecher has said he cannot remember which. He relaunched the platform in 2000 as Intercontinental Exchange (NYSE: ICE). ICE now owns the New York Stock Exchange, the Brent crude oil and TTF natural gas benchmark contracts, and the European Union’s carbon market. Its FY2025 net revenue of $9.9 billion was its twentieth consecutive annual record, and its market capitalization on August 31, 2026 was roughly $90 billion.
The report describes the investment case as toll roads on capitalism: an exchange collects a very small fee on a very large and recurring volume of trading, hedging, and price referencing, on a platform whose costs are mostly fixed. CME Group (Nasdaq: CME) earned just under $0.70 per contract in FY2025; on an E-mini S&P 500 future with about $300,000 of notional value, the roughly $0.61 it collects is about one-fiftieth of a basis point. Operating and EBITDA margins across the major exchange groups were 50–79% in FY2025, and derivatives volumes rise in periods of market stress, when most other businesses suffer. The fee is not guaranteed, however. Between 1993 and 1998 the all-electronic DTB, later Eurex, took the Bund futures contract from LIFFE’s London trading floor: its share of volume went from roughly 24% in 1993 to above 50% in January 1998 and above 99.9% by October 1998. The report explains why that migration happened and why no cleared derivatives franchise has migrated to a rival in the twenty-eight years since. Those two answers cover most of what an investor needs to know about competitive advantage in this industry.
Shareholder returns since these businesses became investable are the report’s starting point. Fourteen major venues have returned a median of roughly 16.7% a year since their IPO or demutualization, against about 10.5% for the S&P 500 measured from CME’s December 2002 listing. Ten thousand dollars invested at the first day’s close became roughly $1.24 million at Hong Kong Exchanges and Clearing (HKEX: 388) and $699,000 at CME, but only about $15,000 in dollar terms at the predecessors of Brazil’s B3 (B3: B3SA3), a spread of roughly 80x within one industry. The report’s argument is that the variables behind that spread, captive clearing, entry price, the shift toward data revenue, and the quality of capital allocation, were observable in advance and still are. This is the first primer in a weekly series whose purpose is to help readers expand or deepen a circle of competence, one industry at a time. Exchanges are a good place to start: the industry’s history is well documented, its unit economics are disclosed monthly, and its long-run return data go back decades.
The companion deck. Exchanges & Financial Market Infrastructure: a self-contained summary of the report in slide form.
What’s inside
The primer has two parts: a 223-page report (attached below) and a 53-slide companion deck (attached above). The report has eight parts, twenty-five chapters, and four appendices, with 26 exhibits, 28 tables, and three photographs; every figure is dated and sourced in the notes. The highlights, part by part:
Part I, Orientation. Why the industry is worth studying, how an exchange group makes money, and a map of the industry: eight venue types with different economics, the clearing and settlement layer that determines who keeps the profits, and the data and index businesses where most of the strategic activity of the last fifteen years took place. The listed universe is roughly 45 companies with a combined market capitalization of about $700bn, less than that of a single mega-cap technology company, plus several user-owned utilities that cannot be bought.
From Chapter 1: annualized total shareholder returns of fourteen venues since their first trading day. The two best performers, Euronext (Euronext Paris: ENX) and Japan Exchange Group (TSE: 8697), were also the cheapest at listing, at roughly 10.6x and 7x earnings. The table includes only survivors; an unscreened buyer of demutualized exchanges earned high-single-digit returns.
Part II, Evolution. The industry’s history from 1602 to 2026: the Dutch East India Company’s share issue and the Amsterdam exchange of 1611; the Buttonwood Agreement of 1792, whose minimum commission survived until 1975; the invention of financial futures and listed options in Chicago in the 1970s; the end of fixed commissions in New York in 1975 and London in 1986; electronic trading, demutualization, and the consolidation that followed, in which vertical and data acquisitions closed while antitrust and national politics blocked nearly every large cross-border exchange merger after 2011. Four case studies: the Eurex–LIFFE contest for the Bund, ICE’s acquisitions from 2001 to 2026, the regulatory reduction of the NYSE’s share of trading in its own listings from roughly 80% to about 22.5%, and the two modern clearing crises, the Aas default at Nasdaq Clearing in 2018 and the LME nickel crisis of March 2022.
From Chapter 5: the share of Bund futures volume traded on DTB/Eurex versus LIFFE, 1993 to 1998. The report identifies three preconditions for the migration: order flow sponsored by the challenger’s owners, a genuine all-in cost advantage, and open access. No later challenger has had all three.
Part III, Structure & Economics. Where the profits sit along the life of a trade and why: execution margins are low wherever the product can be traded elsewhere; clearing margins are high wherever positions are held in the exchange’s own clearinghouse, which earns both fees and interest on margin balances; and data and index revenue earns software-like margins. Volume drives transaction revenue and volatility drives volume; the report tests the claim that exchange volumes rise in market stress against the episodes of 2008, 2022, and April 2025 and finds that it holds. Each exchange group is a combination of six business models on a largely fixed cost base, which is why incremental margins at the major groups were 76–95% in FY2025 and why earnings fall as fast as they rise when volumes turn. The chapter on moats examines seven attempts since 1998 to take a benchmark futures franchise from its incumbent. Six failed; the seventh, FMX, backed by BGC Group (Nasdaq: BGC), is still running.
From Chapter 11: the two ownership structures for clearing. Clearing the same trade supports margins of 60–80% where the exchange owns the clearinghouse and roughly 0% where clearing is done by a user-owned utility at cost.
Part IV, The Players. Profiles of the listed companies, organized by three variables: whether clearing is captive or shared, how much revenue is recurring, and whether the franchise depends on a regulatory grant. CME, ICE, and Cboe Global Markets (Cboe: CBOE) hold the main derivatives and options franchises; London Stock Exchange Group (LSE: LSEG), Nasdaq (Nasdaq: NDAQ), and Deutsche Börse (Xetra: DB1) have shifted toward data and subscriptions; the regional exchange groups each depend on one structural feature of their home market; and India’s exchanges, which form the world’s largest derivatives market by contract volume, are the most exposed to regulatory change, with the National Stock Exchange’s IPO filed in June 2026 at a valuation of up to roughly $60bn. A chapter on the executives with long value-creation records compares Sprecher’s acquisition-led strategy at ICE with Terry Duffy’s distribution-led strategy at CME; they produced returns of about 18% and 19.6% a year respectively, and both kept clearing inside the company. A final chapter covers the unlisted institutions, DTCC, OCC, and Euroclear, whose governance determines where the listed companies’ profits form.
From Chapter 12: the share of net revenue from transaction and clearing fees at four groups in fiscal 2005, 2015, and 2025. ICE went from 88% to 49%; LSEG from 46% to 32%; Nasdaq’s share was already 18% in 2005 and Deutsche Börse’s about half, and neither changed much.
Part V, Speaking the Language. A glossary of the industry’s terms and a guide to its operating metrics, distinguishing the ones managements emphasize from the ones that matter. A chapter on accounting covers the features that make comparisons difficult: gross versus net revenue (Cboe reports $1.94 of revenue for every $1 it keeps after liquidity payments, regulatory pass-through fees, and index royalties), clearing-member collateral carried on both sides of the balance sheet (CME’s total assets grew from $137bn to $198bn in one year for this reason alone), and the gap between adjusted and GAAP earnings, which ranged from 7% to 76% at the acquisitive groups in FY2025. A chapter on the filings covers the monthly volume releases, which allow transaction revenue to be estimated before it is reported, and the CPMI-IOSCO quantitative disclosures, where the clearinghouses publish their risk figures.
Part VI, Analysis & Valuation. A six-step procedure for analyzing an exchange, which builds the revenue model product by product and normalizes volumes before capitalizing them; twenty years of valuation multiples by type of exchange, against which the sector in August 2026 sits near the middle of its historical range, at a median of 25.1x trailing earnings across eighteen companies; acquisition multiples, which rank assets in the same order as the public market does; and a decomposition of long-run returns into their sources. CME’s return of 19.6% a year since 2002 came in roughly equal parts from per-share revenue growth, a one-time increase in margins, and reinvested dividends, with the change in its multiple contributing only about two points a year. MarketAxess (Nasdaq: MKTX), which ICE agreed in July 2026 to acquire for $167.00 a share in cash, is the chapter’s example of entry price mattering as much as the quality of the franchise.
From Chapter 19: MarketAxess’s revenue, share price, earnings per share, and trailing price-earnings ratio, 2019 to 2025, with ICE’s offer at right. A shareholder who bought at the record close of December 22, 2020, about 75x that year’s earnings, and holds to ICE’s offer loses roughly 20% a year over 5.6 years, although revenue rose every year.
Part VII, The AI Inflection. What can be measured about AI in this industry as of mid-2026. There is no credible public evidence that language models make money as autonomous traders; the development that is real is order entry by software agents, which routes through exchanges rather than around them. Within the exchange operators, AI is used mainly in surveillance, financial-crime detection, and operations; Nasdaq’s Verafin unit is the only AI-related business in the sector with disclosed financials; and the matching engines themselves are largely closed to AI by design. The largest open question is whether consumption of market data by machines will raise or lower its price. The most measurable opportunity is hedging demand from the AI economy: ICE’s US power futures volumes rose 30% in 2025, and CME’s compute futures launch on October 5, 2026.
Part VIII, Risks & Debates. Six ways investors in this industry lose money, each with the indicators that give early warning: clearinghouse losses shared among members (the 2018 default at Nasdaq Clearing consumed €107m of a €166m default fund; in March 2022 the LME canceled every nickel trade executed since midnight); regulation that opens the matching business to competition; long periods of low volatility, and record-year earnings capitalized as if they were permanent; technology failures; acquisitions at multiples of 30x; and the model of clearing without intermediaries that FTX proposed and Hyperliquid now operates offshore. The final chapter sets out the seven main debates of 2026, with the bull and bear case for each: prediction markets, the US Treasury clearing mandate, tokenization, the SEC’s proposal to rescind the trade-through rule at a time when 55.7% of US equity volume already trades off-exchange, same-day options at two-thirds of SPX volume, exchange consolidation in Europe, and the volume booms in India and Hong Kong, including the NSE IPO.
The appendices. The investable universe, roughly 45 listed companies by segment with tickers, listing venues, market capitalizations, and one-line descriptions; a sixty-question due-diligence checklist ordered to match the analytical procedure of Chapter 18; a reading list of books, shareholder letters, interviews, podcasts, and data sources; and the sources for the report.
What the primer is for. A reader who works through it should be able to explain how an exchange group makes money, and why the same activity earns 60–80% margins under one ownership structure and roughly nothing under another; use the industry’s vocabulary, from ADV and RPC to open interest, net capture, and the default waterfall; read the filings without being misled by gross revenue, clearing-member balances, or adjusted earnings; appraise an exchange the way specialists do, product by product, through the cycle, against twenty years of multiples and the acquisition record; and recognize the managers and business models that have compounded value over decades, and the four observable variables that distinguished them.
How to use this primer
The report does not need to be read in order. Suggested paths:
a generalist with limited time, Parts I and VI plus the checklist in Appendix B;
a reader interested in the history, Part II with Chapter 13;
an accountant, Part V;
an analyst working on a specific company, Chapter 12, then Parts III, V, and VI in order;
a risk officer, Part VIII with Chapter 17;
a technologist, Part VII.
Every chapter ends with a “What to remember” box, and the twenty-five boxes read in sequence summarize the report. The checklist follows the order of the analytical procedure in Chapter 18; a company that fails the franchise questions in its first section rarely needs the rest.
Feedback and requests
Please reply to this email or leave a comment with what was useful, what was missing, any errors you find, and the industries you would like covered next. Member requests will inform the order of coverage.
This content is published by MOI Global for educational purposes only. It is not investment advice, an offer, or a solicitation, and no security mentioned herein is recommended for purchase or sale. The publisher and contributors may hold positions in securities discussed. Figures are believed accurate as of the dates stated but are not warranted; readers should verify all data against primary sources before relying on it. Nothing herein constitutes legal, tax, or accounting advice.