Jane Street: How a Secretive Firm Made $39.6 Billion Being the House
On the figures. Jane Street is private and secretive; revenue, market-share, and SEBI figures below are from reporting (Bloomberg, Investing.com, Business Standard and others) and the firm’s own materials, as of mid-2026, and may be revised. The India case is disputed and ongoing.
Here is a number that should reorganize how you think about markets. In 2025, a firm most people have never heard of made roughly $39.6 billion in trading revenue — more than Goldman Sachs, Morgan Stanley, or Citigroup made trading, reportedly the largest single-year trading haul any firm has ever posted. It did this not with a quarter of a million employees like a global bank, but with about 3,500. It has no retail app, runs almost no advertising, publishes little, and is owned entirely by its partners. It is called Jane Street, and understanding what it does is one of the most clarifying things a retail trader can do — because Jane Street is, quite often, on the other side of your trade.
Key takeaways
- Jane Street made about $39.6 billion in trading revenue in 2025 — reportedly a record for any firm, more than most Wall Street banks — with only ~3,500 people.
- Its business is market making: providing liquidity and earning the spread, above all in ETFs, where it trades a reported $700+ billion a month.
- Its edge is not prediction. It’s speed, technology (famously written in OCaml), probabilistic discipline, and industrial-scale risk management — being the house, not the gambler.
- It’s not without controversy: India’s SEBI accused it of index manipulation in 2025 and froze ~$565M, which Jane Street disputes and is appealing.
What Jane Street actually does: being the house
Strip away the mystique and Jane Street is a market maker. When you want to buy an ETF or an option and someone needs to be there to sell it to you — instantly, at a fair price, whether markets are calm or in freefall — that someone is increasingly Jane Street. It quotes a price it will buy at (the bid) and a price it will sell at (the offer), and it earns the tiny difference between them, the spread. On any single trade that spread is trivial, a penny or a fraction of one. Across hundreds of millions of trades a day, on more than 200 venues in roughly 45 countries, those pennies compound into the biggest trading revenue on earth.
Founded in 2000 by a small group of traders and engineers, Jane Street began in the then-unglamorous business of ETF market making and rode that instrument’s explosion into dominance. It is now widely regarded as the world’s leading ETF liquidity provider, reportedly turning over more than $700 billion in ETFs a month and holding roughly a quarter of the US primary ETF market. It has since expanded into equities, bonds, options, commodities, and crypto — but the DNA is the same everywhere: be the reliable counterparty, earn the spread, manage the risk, repeat at enormous scale.
Jane Street doesn’t try to predict where the market is going. It gets paid for being willing to trade with everyone who thinks they know — and for managing the risk of that inventory better than anyone else.
This is the crucial distinction, and it’s the whole reason the firm belongs on a behavioral-trading blog. A retail trader is usually making a directional bet: they think price will go up, so they buy. Jane Street is usually neutral on direction: it doesn’t care much whether the market rises or falls, because it’s hedged; it profits from volume and from the spread, and it hedges its inventory so that being the counterparty to a million opinions doesn’t leave it exposed to any of them. It is, structurally, the casino rather than the gambler — and the casino’s edge is not luck, it’s math, process, and never letting a single bet threaten the house.
How it does it: technology, probability, and risk
The technology
Jane Street is as much a software company as a trading firm, and it made a famously idiosyncratic bet: it writes nearly everything in OCaml, a functional programming language almost no one else uses at scale. The logic is telling. OCaml’s expressive type system catches whole categories of bugs at compile time — before code ever touches a live market — which for a firm risking billions is a risk-management decision disguised as a language choice. The firm has invested so deeply that it maintains its own compiler variant (OxCaml) and is one of the largest forces in the OCaml world. When your losses can arrive in microseconds, correctness isn’t a nicety; it’s survival.
The culture
Jane Street hires for probabilistic thinking, not finance pedigree — mathematicians, competitive programmers, poker and puzzle enthusiasts — and trains them with trading games and betting exercises designed to teach reasoning under uncertainty and calibrated confidence. New traders learn to price things, size bets to their edge, and update on evidence. (One notable alumnus, before his own spectacular flameout, was FTX founder Sam Bankman-Fried — a reminder that the firm’s process is the edge, not any individual who passes through it.) The house style is collaborative, low-ego, and relentlessly quantitative: what’s the probability, what’s the edge, what’s the risk, what’s the right size.
The risk management
This is the part worth internalizing, because it’s the part you can actually learn from. A market maker’s existential threat is not being wrong on direction — it’s holding inventory when the market gaps and being unable to hedge. So firms like Jane Street obsess over risk limits, hedging, and never letting a single position or correlated cluster grow large enough to threaten the firm. It is the exact opposite of the concentrated, leveraged, one-big- bet blow-ups we’ve chronicled — from LTCM to the Situational Awareness fund. Jane Street’s business is engineered so that no afternoon can end it, which is precisely why it’s still here — and dominant — after 25 years.
The controversy: the India / SEBI case
A deep dive that only praised Jane Street would be incomplete, because its scale has drawn scrutiny. In July 2025, India’s market regulator, SEBI, accused the firm of manipulating Indian index levels — alleging that around weekly options-expiry days it made aggressive, concentrated trades in Nifty 50 and Bank Nifty constituents to move the index and profit from large options positions. SEBI ordered the disgorgement of about ₹4,843 crore (roughly $565 million) and imposed a temporary ban.
The two sides frame the same trades in irreconcilable ways, and the disagreement is genuinely interesting. Jane Street’s position is that it was doing exactly what market makers do — finding price dislocations and profiting by correcting them, which is supposed to improve market efficiency. SEBI’s position is that the scale, concentration, and expiry-day timing crossed the line from arbitrage into deliberate index distortion. Jane Street deposited the required sum into escrow, resumed trading in India, and appealed to the Securities Appellate Tribunal, arguing it was denied the information needed to defend itself; the matter has continued into 2026 and remains unresolved. It’s a live example of the blurry, contested boundary between aggressive-but-legal liquidity provision and prohibited manipulation.
The other giants of the trade
Jane Street is the largest, but it’s one of a small elite that quietly runs modern markets. These are the firms on the other side of an enormous share of the world’s trades.
Note what these firms have in common, because it’s the lesson. None of them got rich by predicting the market’s direction better than you. They got rich by being faster, better-priced, better-hedged, and better risk-managed — by being the disciplined counterparty to everyone else’s opinions. This is a different game than the one most retail traders think they’re playing, and confusing the two is expensive.
What this means for you as a trader
You cannot become Jane Street. You don’t have the microsecond infrastructure, the OCaml stack, or the balance sheet, and you shouldn’t try to out-speed a firm that spends more on latency than you’ll trade in a lifetime. But there are two genuinely useful takeaways, and they’re the opposite of despair.
First, know which game you’re in. When you fire off a discretionary directional trade on a whim, you are often the very flow these firms are built to profit from — the impatient, emotional counterparty whose spread and mistimed entry are someone else’s edge. That’s not a reason to quit; it’s a reason to stop volunteering to be the easy customer. Second, and more constructively: the part of Jane Street you can copy is the part that has nothing to do with technology. Their edge is probabilistic thinking, disciplined position sizing, obsessive risk management, and low-ego updating on evidence — deciding how much to bet from edge rather than emotion, and never letting one position threaten the whole. That’s not a hardware advantage. It’s a behavioral one, and it’s available to anyone willing to measure themselves honestly.
You can’t out-compute the house. But you can stop trading like its favorite customer — and start borrowing the one part of its edge that’s free: discipline.
From belief to behavior: trade less like the mark
Resources and further reading
- The numbers: Bloomberg and Investing.com on Jane Street’s record ~$39.6B 2025 trading revenue and ~$16.1B record quarter; Bloomberg on Citadel Securities’ record ~$12.2B in 2025.
- What they do: Jane Street’s own “What We Do” and technology pages, and its engineering blog on OCaml and OxCaml.
- The India case: Bloomberg and Business Standard on SEBI’s July 2025 order, the ~₹4,843 crore (~$565M) disgorgement, the lifted ban after the escrow deposit, and Jane Street’s SAT appeal.
- The landscape: reporting on the market-making field — Citadel Securities, Hudson River Trading, Jump, Susquehanna, Optiver, IMC, DRW, XTX, and Virtu.
- The behavioral companion: Gecko, Ed Thorp and the Kelly criterion — the individual-scale version of sizing from edge and never risking ruin.
Frequently asked questions
What does Jane Street actually do?
It’s a proprietary trading firm and one of the world’s largest market makers — providing liquidity and earning the spread between bid and offer, especially in ETFs, plus equities, bonds, options, commodities, and crypto, across 200+ venues in ~45 countries. It trades its own capital rather than managing outside money, and profits by being the reliable counterparty at massive scale.
How much money does Jane Street make?
About $39.6 billion in net trading revenue in 2025 — reportedly a record for any firm and more than major Wall Street banks’ trading arms — plus a record ~$16.1B quarter in early 2026, all with roughly 3,500 employees, making its revenue per head extraordinary.
Why was Jane Street banned in India?
In July 2025 SEBI accused it of manipulating index levels around options-expiry days in the Nifty 50 and Bank Nifty, ordering ~₹4,843 crore (~$565M) disgorged and a temporary ban. Jane Street says it was doing normal market making — correcting dislocations — while SEBI alleged deliberate index distortion. It deposited the sum, resumed trading, and appealed; the case continues into 2026.
Who are Jane Street’s main competitors?
Citadel Securities (the dominant US equity/retail wholesaler, ~$12.2B 2025 revenue), plus Hudson River Trading, Jump Trading, Susquehanna (SIG), Optiver and IMC, DRW, Tower Research, XTX Markets in FX, and the listed Virtu Financial — all competing to be the fastest, best-priced, best-risk-managed counterparty in the world’s markets.
Deep dive in Gecko’s trading psychology series. Jane Street is private; revenue, market-share, and regulatory figures are drawn from contemporaneous reporting (Bloomberg, Investing.com, Business Standard and others) and the firm’s own materials as of mid-2026, and may be revised. The SEBI matter is disputed and ongoing; nothing here asserts that Jane Street engaged in wrongdoing — SEBI’s allegations are contested and under appeal. Gecko is an educational and informational tool. Nothing here is financial, investment, or trading advice, or a recommendation regarding any firm or security. Trading carries substantial risk of loss.
Drop in a single statement. Gecko produces a one-page Behavioral Diagnosis ranking your costliest habits in actual dollars. Free to start. No card. No broker connection.
Short notes, usually once or twice a month. Unsubscribe in one click.
No account needed. We use your email only to send Gecko blog posts, and the link at the bottom of every email opts you out in one click.
More on the blog
- The Warsh Fed: No More Forward Guidance, and the Two Levers That Are LeftKevin Warsh has scrapped forward guidance — "not the business we should be in" — ending two decades of the Fed telegraphing its path. With guidance gone, the Fed is left with its two concrete levers: the policy rate (the price of money, at 3.50-3.75%) and the balance sheet (the quantity of reserves, ~$6.57T after QT ended). Less guidance means markets trade the data, not the Fed's words — which means more volatility, more term premium, and behavioral errors punished harder.
- When Genius Failed by Roger Lowenstein: What Two Nobel Laureates and 130-to-1 Leverage Teach TradersLong-Term Capital Management had the best credentials in finance -- two Nobel laureates, elite traders, the most sophisticated risk models on Wall Street -- and returned ~40% a year. Then it lost $4.6 billion in under four months in 1998 and had to be rescued in a Fed-organized $3.6B bailout. The cause wasn't bad analysis; it was leverage up to 130-to-1 plus models that assumed a normal-distribution world markets don't inhabit. The lesson scales to any account: being right doesn't matter if leverage denies you the time to be proven right.
- Recency Bias: Why You Expect the Market to Keep Doing What It Just DidThe moment your expectations are highest is, on average, the moment future returns are about to be lowest. Greenwood & Shleifer (2014) analyzed six independent surveys and found investor return expectations strongly positively correlated with past returns and negatively correlated with actual future returns -- investors are most bullish exactly when the math says they shouldn't be, and De Bondt & Thaler (1985) showed prior losers beat prior winners by ~25% over 3 years. Recency bias is invisible in the moment because it feels like judgment, but it leaves a clear fingerprint in your trade history.