Robinhood Markets, Inc.

01 — Executive summaryReported net income of $573 million. Operating profit of $574 million. One of those numbers is an accounting entry.

Robinhood Markets reported total net revenues of $1.31 billion for the quarter ended 30 June 2026 — a record, and a 32% increase over the prior year. Diluted earnings per share rose 48% to $0.62. Adjusted EBITDA was $741 million, a 57% margin on revenue, which is a higher margin than any large broker has sustained through a full cycle. Net deposits were $21.7 billion, the best quarter the company has recorded. Funded customers reached 28.4 million, up 7%, and 13 separate product lines now generate more than $100 million of annualised revenue. On the numbers alone, this is the strongest quarter in the company's history.

None of that is in dispute, and none of it is the subject of this report. The question we set out to answer is narrower and less comfortable: what did the record quarter actually consist of, and how much of the earnings power it demonstrates is within the company's control?

Our answer, compressed into a sentence: Robinhood's core brokerage franchise is performing better than at any point since it listed, and the market is paying for a new engine — prediction markets — whose legality is currently being decided by courts rather than by customers. The two facts are entangled. Remove event contracts from the quarter and revenue growth falls from 32% to approximately 17%; remove the crypto decline as well and it rises again. The composition of the growth matters more than its level.

Three numbers frame everything that follows. The first is $129 million: the gain on deconsolidating Robinhood Ventures Fund I, a closed-end fund the company had listed on the New York Stock Exchange, which contributed $0.14 of the quarter's $0.62 of diluted earnings per share. Reported net income of $573 million was therefore 23% non-operating. The second is $156 million: the revenue from event contracts, up more than tenfold year over year, which made the prediction-markets business larger than the entire cryptocurrency business in the quarter. The third is −38%: the change in cryptocurrency transaction revenue, the only declining line in the company.

The event-contracts number is remarkable and deserves to be stated precisely. Robinhood processed 13.6 billion event contracts in the June quarter against 774 million options contracts. It is now, by contract count, one of the largest derivatives venues in the world. It is also, by revenue, 11.9% of the company. And on 28 August 2026 — three weeks after Robinhood reported the quarter — a unanimous three-judge panel of the Ninth Circuit Court of Appeals held that the substance of sports event contracts "is sports gambling, regardless of whether Kalshi calls them swaps", that the Commodity Exchange Act does not preempt Nevada's gaming laws, and that states may regulate the product. That ruling binds nine states directly and contradicts an April 2026 ruling from the Third Circuit which held the opposite. Roughly twenty states are in active litigation.

We are not suggesting the business disappears. We are observing that a product line generating $156 million a quarter, growing tenfold, and carrying most of the incremental revenue growth sits inside a legal question that has split two federal circuits and is now a candidate for Supreme Court review. That is a different kind of risk from the ones a broker usually carries, and it is not well captured by a discounted cash flow.

The second theme is the balance sheet, and it is the opposite of the first. Net interest revenue — the money Robinhood earns on customer cash, margin loans and securities lending — peaked at $456 million in the September 2025 quarter and has fallen in three of the four quarters since. It is still 9% higher than a year ago, but the composition has deteriorated: margin balances, which carry the highest yield and the most credit risk, have more than doubled, while Cash Sweep balances, which are the cheapest funding, remain 9% below their August 2025 level after the company moved the first $10,000 of each enrolled balance into free credit balances in February 2026. That decision funded the growth in margin lending and was accretive while rates were high. It becomes a headwind as policy rates fall, because margin loans reprice downward faster than the deposit base does.

The third theme is that the market has already begun to price the first two. The shares closed at $119.82 on 18 September 2026, a 9.12% single-day move on news that the Securities and Exchange Commission had granted a five-year, conditional "Innovation Exemption" allowing qualifying venues to trade tokenised National Market System stock. That is a genuine and consequential development for a company that has built Stock Tokens, a Layer 2 blockchain and a tokenised-equity franchise in 120 countries. It is also, at $10 a share, a $9 billion re-rating on a product whose on-chain economics remain small: Robinhood Chain has processed roughly $34.6 billion of cumulative decentralised exchange volume, but the sequencer revenue Robinhood retains is shared with launch partners on a sliding scale that caps the company's take in the low hundreds of millions.

Our conclusion is that Robinhood at $119.82 is fairly valued. We initiate with a Neutral rating and a 12-month price target of $128, 6.8% above the current price and within 1% of the $129.34 sell-side consensus. The business deserves a premium multiple: 46% GAAP operating margins, a genuine subscription layer in Gold, an asset-gathering engine that has pulled in $74.1 billion of net deposits over the trailing twelve months, and a management team that has repeatedly found new products faster than its competitors. It does not deserve a premium multiple without acknowledging that the fastest-growing piece of it is a wager.

The valuation chapter sets out a scenario range of $70 to $189 — a 2.7-fold spread that is wider than the spread we would normally accept in a report of this length. That width is the finding. The outcome for this equity over the next twelve months depends less on how well Robinhood executes than on two decisions taken by other people: one at the Supreme Court, and one at the SEC. In between those decisions, the company will keep compounding. It usually does.

02 — Company & business modelA free stock trade, and everything that was built on top of it

Robinhood Markets, Inc. was incorporated in 2013 and listed on Nasdaq in July 2021. Its founding proposition was narrow and, at the time, considered commercially suicidal: charge nothing for a stock trade and monetise the order flow instead. The company now describes itself as "a global leader in financial services offering retail brokerage, crypto, advisory, digital banking services and private markets access to a new generation of investors." That sentence contains six business models, and the reason the company is hard to value is that they are at different stages of maturity, carry different regulatory exposures, and move with different underlying variables.

It is worth being precise about how Robinhood actually earns money, because the marketing language obscures it. There are three revenue lines in the income statement, and they behave nothing alike.

Transaction-based revenue is a toll on activity. When a customer trades an option, Robinhood routes the order to a market maker, the market maker pays Robinhood for the right to trade against that flow, and Robinhood books the payment as revenue. The customer pays no commission. Economically this is a payment for order flow model, and it means Robinhood's revenue is a function of three variables it does not control: how many customers trade, how often they trade, and how wide the spread is that market makers are willing to pay for. In the June 2026 quarter this line produced $776 million, 59.3% of total revenue.

Net interest revenue is a spread on a balance sheet. Robinhood holds customer cash, lends against securities in margin accounts, and runs a securities-lending book. It earns a rate on those assets and pays a rate on its funding, and books the difference. This line produced $389 million, 29.7% of revenue. It is the most bank-like part of the company and the part most exposed to the interest-rate cycle, which is why the section on it later in this report is uncomfortable reading.

Other revenue produced $143 million, 10.9% of the total, and is the most interesting line in the accounts. It is dominated by two things: Robinhood Gold subscriptions, at $50 million a quarter and growing, and the service fee Robinhood earns for administering the federal government's Trump Accounts programme, which is contracted on a cost-plus basis. This is the line that converts a cyclical brokerage into something closer to a software business, and at 10.9% of revenue it is not yet large enough to change the character of the company.

Around those three lines sits an increasingly wide set of products. Robinhood Gold, at $5 a month or $50 a year, bundles higher cash rates, margin at a discount, retirement matching, and access to the Gold Card. Robinhood Retirement holds $34.5 billion of assets across roughly 1.9 million accounts. Robinhood Strategies, a managed portfolio product launched in 2025, passed 300,000 funded customers and nearly $2 billion of assets. Robinhood Banking had over $3 billion of deposits from more than 240,000 customers. The Gold Card crossed one million customers and $17 billion of annualised purchase volume. TradePMR, acquired in 2025, runs an adviser network with $50 billion of assets. Bitstamp, also acquired in 2025, handles institutional crypto. WonderFi brought Canada. Rothera, a joint venture with Susquehanna International Group, is a CFTC-licensed exchange and clearinghouse.

The strategic logic is coherent and worth stating plainly, because it explains most of the company's decisions. Robinhood's original insight was that a generation of investors who grew up on mobile interfaces would not accept the friction of a traditional brokerage. Having acquired those customers cheaply, the company's job is to hold them as their financial lives get more complicated — from a first fractional share to an individual retirement account to a margin loan to a mortgage. Every product launch, every acquisition, and every regulatory filing should be read against that single objective: increase the number of things a customer does with Robinhood, because a customer who does four things with Robinhood does not leave.

The measure of whether it is working is not revenue. It is the ratio of investment accounts to funded customers, which stood at 29.9 million accounts against 28.4 million customers at the end of June 2026, and the net deposit rate, which tells you whether the money customers bring in exceeds the money they take out. Both have been positive and both are covered in detail later.

03 — The financial recordFour years of compounding, and the first deceleration

The shape of Robinhood's financial history is unusual among recent technology listings in that the losses came first and were large. The company lost $3.69 billion in 2021 and $1.03 billion in 2022 — a cumulative $4.7 billion — before reaching profitability in 2024. The 2021 loss was inflated by the collapse in the value of the company's own convertible notes and by stock-based compensation recognised around the IPO, but the 2022 loss was real and reflected the end of the pandemic trading boom.

What followed was a genuine turnaround. Revenue fell 25% in 2022 to $1.36 billion, then grew 37% in 2023 to $1.87 billion, 58% in 2024 to $2.95 billion, and 52% in 2025 to $4.47 billion. Net income went from a $541 million loss in 2023 to $1.41 billion of profit in 2024 and $1.88 billion in 2025. Adjusted EBITDA margin expanded from 48% in 2024 to 56% in 2025. By any standard, that is one of the more complete operational recoveries of the last five years in US financial services.

The important question is whether that trajectory is a trend or a cycle, and the honest answer is that it is mostly a cycle that management has learned to harvest. Robinhood's revenue is a function of retail engagement, which is itself a function of market volatility and the attractiveness of speculative assets. The 2024 and 2025 growth was driven by an extraordinary period for both: equity markets rose, options volumes set records, crypto rallied hard into the US election and the first half of 2025, and prediction markets went from a curiosity to a mass consumer product. Every one of those was favourable, and none was caused by Robinhood.

The first crack appeared in the first quarter of 2026. Revenue of $1.07 billion was 15% higher than the prior year but 17% lower than the preceding quarter, and net income of $346 million was 43% below the December quarter. Adjusted EBITDA margin fell to 50% from 59%. Crypto revenue fell 47% year over year. The company's own explanation — lower short-term interest rates, weaker securities lending, and a softer crypto market — was accurate and was also a description of things outside its control.

Then the second quarter recovered sharply, and the recovery came almost entirely from one product. Revenue of $1.31 billion was a record. Transaction revenue rose 44% year over year. But within that, options grew 29%, equities grew 95%, and event contracts grew more than tenfold, while crypto fell 38%. Strip event contracts out and transaction revenue growth falls from 44% to roughly 8%.

Two features of the table deserve comment. The first is the seasonality, which is more pronounced than at a traditional broker. Robinhood's customers are younger, trade smaller positions, and are concentrated in options and short-dated products. Their activity peaks in the third quarter, when equity and options volumes are seasonally strong, and again in the fourth quarter on year-end positioning. The first quarter is reliably the weakest. Any year-over-year comparison drawn from a single quarter is therefore unreliable, and any analyst extrapolating a Q2 growth rate into a full year is making an error.

Before the mix, the margin. Adjusted EBITDA margin has run between 50% and 59% across the last six quarters, and the two troughs are instructive: they both occur in first quarters, when revenue is seasonally weakest and the cost base does not flex. A company with a 57% margin and fixed costs has enormous operating leverage, and leverage is a two-sided instrument.

The second is that the revenue mix is shifting in a direction that reduces earnings quality. In the June 2025 quarter, 36.1% of revenue came from net interest — a balance-sheet business with contractual, if rate-sensitive, revenue. In the June 2026 quarter that share was 29.7%. The balance has shifted towards transaction revenue, which is the least predictable line in the company, and within transaction revenue towards event contracts, which is the least legally settled. Robinhood has grown revenue and simultaneously made its revenue less durable. That is the central tension of this report.

04 — Anatomy of a record quarterFour products, four different futures

The June 2026 quarter produced $776 million of transaction-based revenue, 44% more than a year earlier. That single number conceals four businesses moving in four different directions, and the most useful thing an analyst can do with Robinhood's income statement is take it apart.

Options produced $342 million, up 29%. This remains the backbone of the company and the reason it exists. Robinhood is, by contract count, one of the largest retail options venues in the world: 774 million contracts traded in the quarter, up 50%, with 13.9 million contracts a day in August. The economics are excellent — options are high-margin, the flow is valuable to market makers, and the customer acquisition cost was paid years ago. The risk is that the option customer is a certain type of person: self-directed, active, and disproportionately male and young. That cohort is not infinite, and it is already the company's heaviest user.

Equities produced $129 million, up 95% — the fastest growth of any legacy line. The driver is not new customers but higher activity per customer: equity notional volumes were $956 billion in the quarter, up 85%, against a customer base that grew 7%. Average daily equity volume in August was $16.0 billion, up 68% year over year. Some of this is the launch of Legend, the company's active-trader platform, which passed $100 million of annualised revenue about eighteen months after launch, and some is the addition of short selling and index options. Both point the same way: Robinhood is successfully moving upmarket from the casual investor to the semi-professional trader.

Event contracts produced $156 million, up more than tenfold. This is the new engine and it gets its own section below, because it deserves one.

Cryptocurrency produced $100 million, down 38%. It is the only line in the company that shrank, and the decline is not a Robinhood problem — it is a market problem, covered in section 09.

What makes the quarter interesting is not any of these individually. It is that the growth rate of the company now depends on the two most legally and cyclically exposed lines. Options, the stable core, grew 29% — respectable, but not enough on its own to produce a 32% revenue increase. Equities grew 95% but from a base of $66 million, so it contributed $63 million of incremental revenue. Event contracts contributed $140 million. Crypto subtracted $61 million.

The same composition problem appears in revenue per customer, and it is the most useful single sanity check on the growth story. Annualised revenue per funded customer was $187 in the June 2026 quarter — below the $191 recorded in the December 2025 quarter, and only 24% higher than a year earlier, despite a 32% increase in platform assets. Revenue per customer at Robinhood is a cycle, not a compound curve. It rises when markets are volatile and customers trade, and it falls when they are not.

There is a fourth observation to make, and it concerns the nature of the revenue rather than its size. Options and equities revenue is a payment for order flow: it depends on market makers continuing to value Robinhood's retail flow. That is a mature, litigated, and well-understood business, and it is not going away. Event-contract revenue is also economically a spread, but the product itself is a binary wager with a defined settlement, which puts it in a different category of legal risk. Crypto revenue is a take rate on trading volumes in an asset class that has now had two 60%-plus drawdowns in four years.

A company whose incremental growth comes from its two most fragile lines is not necessarily a bad investment. It is, however, a company whose multiple should be lower than one whose growth comes from recurring revenue — and Robinhood trades at a higher multiple than most recurring-revenue software businesses of comparable size. That mismatch is the valuation problem set out in section 20.

05 — Net interest and the rate cycleThe spread that peaked in September 2025

Net interest revenue is the part of Robinhood that most closely resembles a bank, and it is behaving like one. The line earned $456 million in the September 2025 quarter, its high-water mark, and has since produced $411 million, $359 million and $389 million. The June 2026 figure is 9% higher than a year earlier, which sounds healthy until it is set against the growth in the assets that generate it: the margin book rose 127% over the same period, and cash and deposits rose 34%.

The arithmetic is straightforward. Net interest revenue is the product of interest-earning assets and the spread earned on them. Assets grew substantially. The spread compressed more. Robinhood attributes the compression to two things in every one of its recent filings: lower short-term interest rates and reduced securities lending activity. Both are visible in the data. Securities lending revenue fell from $53 million in August 2025 to $36 million in August 2026, a 32% decline, and the net securities-lending line actually went negative in August, at minus $8 million.

The composition of interest-earning assets has also changed in a way that raises the risk profile. In February 2026, Robinhood changed its brokerage High-Yield Cash programme so that the first $10,000 of each enrolled customer's balance is held as a free credit balance rather than swept into a programme account. More than $6 billion moved from Cash Sweep into cash and deposits. The stated purpose was to fund the growth of margin lending, and it worked: margin balances went from $12.5 billion in August 2025 to $21.5 billion in August 2026.

This is a trade that made sense at the time and is now working against the company. Robinhood swapped a low-cost, rate-insensitive funding source for a higher-yielding asset funded by deposits that reprice with the market. While the federal funds rate was at its peak, the spread on margin loans comfortably exceeded the cost of the deposits funding them. As the rate falls — the upper bound of the target range has come down from 4.33% in the first half of 2025 to 3.51% by the second quarter of 2026 — margin loan yields reprice downward, customer cash rates reprice with them, and the net effect depends on the beta of each side. Historically, retail deposit betas are lower than loan betas on the way down, which is favourable. But Robinhood is not a bank with a large sticky deposit franchise; its customers move cash to wherever it earns most, and it competes with money-market funds that pass through rates almost one-for-one.

There is a second, less discussed consequence. Margin lending carries credit risk, and Robinhood books a provision for credit losses against it. That provision was $56 million in the June 2026 quarter, up 100% year over year, and $92 million in the first half against $114 million for the whole of 2025. The provision is growing faster than the revenue it protects. A doubling of the margin book in a period when equity markets rose sharply has been profitable; the same book in a 30% drawdown would be a different experience, and the company's risk framework has not yet been tested by one at this scale.

06 — Event contractsThirteen billion contracts, and a legal question

Event contracts are binary options on real-world outcomes: whether a team wins a game, whether a data release comes in above or below a threshold, whether an election goes a particular way. On Robinhood they trade under the futures permissions of the customer's account, are cleared through a CFTC-regulated venue, and are priced between $0.01 and $0.99, settling at $1 or $0. Robinhood earns revenue on each contract traded and on the spread.

The product has been transformative for the company's growth rate and almost invisible in its public positioning. In the June 2026 quarter, customers traded 13.6 billion event contracts, up more than tenfold year over year, generating $156 million of revenue. For scale: that is more revenue than cryptocurrencies produced, roughly half of what options produced, and it came from a standing start. In July 2026, 6.1 billion contracts traded; in August, 4.7 billion. Even after the August decline, the run-rate is running well ahead of the second quarter.

The economics of the product deserve attention because the contract count is misleading. Thirteen billion contracts sounds like an enormous business; $156 million of revenue implies roughly 1.1 cents of revenue per contract. That is because most event contracts are short-dated, cheap instruments bought in size, and because Robinhood's share of the economics is a fraction of the spread. The revenue per contract will improve if customers move towards higher-priced, longer-dated contracts, which is exactly what the company is trying to encourage by adding non-sports markets.

The strategic rationale for the product is sound and mirrors the options business. It monetises engagement. A customer who checks an app to see whether their position on a football game settled is a customer who opens the app daily, and app opens correlate with deposits. Robinhood's own disclosure that roughly 40% of new funded customers sign up for Gold in the quarter in which they join suggests the funnel is working.

Then there is the legal question, and it is not a small one. On 28 August 2026, a unanimous panel of the Ninth Circuit Court of Appeals upheld a Nevada district court decision allowing Nevada's gaming regulators to oversee Kalshi's sports event contracts. The court held that the "substance of the sports event contracts offered on Kalshi's DCM is sports gambling, regardless of whether Kalshi calls them swaps," and that the Commodity Exchange Act does not preempt state gaming law. It described as "disingenuous" the argument that the products were not sports betting when the company had used that framing in its own marketing.

That ruling conflicts directly with an April 2026 decision of the Third Circuit, which held that New Jersey could not regulate Kalshi under federal law. Two circuits, opposite conclusions, on a question of substantial commercial importance: that is the classic profile of a case the Supreme Court takes. Nine states sit within the Ninth Circuit and are bound by its holding. Roughly twenty states are in active litigation. Four — Nevada, Massachusetts, Michigan and Washington — have already obtained orders restricting Kalshi's operations.

Robinhood's own exposure is indirect but real. It distributes event contracts through Rothera, its joint venture with Susquehanna International Group, which is a CFTC-licensed designated contract market and clearinghouse. It also agreed in September 2026 to take minority stakes in Crypto.com and its prediction-market spinout OG.com, and to list Crypto.com's event contracts, a deal explicitly framed as expanding prediction-market infrastructure. The company is therefore not a bystander to this litigation; it is the largest retail distribution channel for the product at issue, and it is adding capacity rather than hedging.

We are not predicting that event contracts are shut down. There are three plausible outcomes and only one of them is terminal. In the first, the Supreme Court affirms federal preemption, the product continues to scale nationally, and event contracts become a multi-billion-dollar revenue line. In the second — arguably the most likely — the Court declines to hear the case or issues a narrow ruling, the patchwork persists, and Robinhood geofences the states that restrict the product, losing perhaps a fifth of its addressable market. In the third, the Ninth Circuit's reasoning prevails broadly, states regulate aggressively, and the product becomes a state-by-state licensing business with a fraction of its current economics.

The point for a valuation is not which outcome is most likely. It is that the outcome is decided by courts and legislatures rather than by customers, that the range of outcomes is wide, and that the market is currently capitalising the first scenario. Event contracts contributed 11.9% of Q2 2026 revenue but a much larger share of incremental growth. A reasonable investor should want a discount for that.

07 — Gold and the subscription turnThe best business in the company is still the smallest

Robinhood Gold costs $5 a month, or $50 a year. For that, subscribers get a higher interest rate on idle cash, a discount on margin borrowing, a 1% retirement contribution match, and access to the Gold Card. It is the most strategically important product the company has launched, because it is the only one that converts a transactional relationship into a recurring one.

Gold subscribers reached 4.8 million at the end of June 2026, up 39% year over year, and the penetration rate reached 17% of funded customers. Roughly 40% of new funded customers signed up in the quarter in which they joined. Gold subscription revenue was $50 million in the quarter, up 32%, and the broader "other revenue" line it sits inside reached $143 million, up 54%.

Set against the size of the company, the subscription business is still small. At 4.8 million subscribers paying $50 a year, the annualised run-rate is approximately $240 million, or 4.6% of the $5.2 billion of revenue consensus expects in 2026. That is not enough to change the character of the income statement. It is, however, enough to change the character of the customer relationship, and the reason is behavioural rather than financial: a subscriber who has paid an annual fee has an incentive to consolidate assets at the platform where the fee is already sunk.

The evidence that this is working is in the deposit data. Robinhood's stated net deposit growth rate has run at 22% to 28% of platform assets on an annualised basis for six consecutive quarters, which is a faster asset-gathering rate than any large US retail broker. Charles Schwab and Interactive Brokers, both of which have far larger asset bases, grow net new assets at substantially lower percentages. Robinhood is taking share of retail wallet at a rate that is genuinely unusual, and Gold is a meaningful part of the mechanism.

The risks in this line are modest but worth naming. The first is that Gold's value proposition is partly a function of interest rates: the higher cash rate it offers is worth more when the federal funds rate is 4% than when it is 2%. As rates fall, the marginal reason to pay for Gold weakens. The second is that the Gold Card, which has crossed one million customers and $17 billion of annualised purchase volume, is a credit product with a credit-loss tail; the provision for credit losses is already growing faster than revenue. The third is that the company has not disclosed Gold churn, and a subscription business that does not disclose churn is a subscription business whose churn should be assumed to be unflattering.

None of these changes the conclusion. Gold is a good business, it is growing, and it is the single clearest piece of evidence that management can build products beyond the trading app. It is simply not yet large enough to make the company's earnings less cyclical, and it will not be for several years.

08 — Net depositsThe number that matters most, and the number that moves most

If one metric captures whether Robinhood is winning, it is net deposits: the money customers move onto the platform, less the money they take off it. It is a better measure than revenue, which moves with markets, and a better measure than customer count, which moves with marketing spend. It measures the thing the company actually sells, which is a place to keep your money.

The record is strong. Net deposits were $68.1 billion in 2025, a 35% growth rate against the prior year's asset base. They were $17.7 billion in the March 2026 quarter, $21.7 billion in the June quarter, $5.6 billion in July and $4.0 billion in August. Over the twelve months to August 2026, net deposits were $74.1 billion, an annual growth rate of 24% against the August 2025 asset base.

Two things about this series deserve scrutiny. The first is the volatility. The quarterly number fell from $21.7 billion to $5.6 billion to $4.0 billion across three consecutive periods. That is not a collapse in the franchise; it is a reflection of how net deposits are calculated. Customer cash that leaves to pay taxes, buy a house, or simply sit in a bank account counts as a deposit outflow, and Robinhood's customer base is younger and more likely to be moving money around than Schwab's. The right way to read the series is on a trailing-twelve-month basis, which is why we have shown it alongside the quarterly figure.

The second is the composition. Deposits can arrive as cash, as securities transferred from another broker, or as acquired assets from an acquisition. Robinhood's disclosures since June 2025 include Bitstamp in net deposits, since March 2026 include TradePMR, and since June 2026 include WonderFi. Since July 2026 they include Trump Account contributions. The $74.1 billion trailing figure therefore includes money that arrived by acquisition rather than by competitive win, and the organic rate is lower than the headline. The company has not quantified the split, and any investor relying on the headline figure should assume a meaningful portion of recent growth is inorganic.

The relationship between deposits and platform assets is the crux of the investment case. Total platform assets reached $383.7 billion in August 2026, up 26% year over year. That growth came from two sources: $74.1 billion of net deposits, and market appreciation. In the first quarter of 2026, net deposits of $17.7 billion were more than offset by market depreciation, and platform assets fell from $324 billion to $307 billion. That is the mechanical proof that a business valued on assets under custody is levered to markets it does not control, and it is the reason the "asset gatherer" framing of this company needs to be held at arm's length.

Robinhood's guidance — a net deposit growth rate above 20% of beginning platform assets — is the single assumption most of the bull case rests on. If it holds, revenue compounds in the mid-teens even with flat monetisation per customer. If it breaks, the entire growth story reduces to whatever the market does, which is not a thesis.

09 — Crypto and BitstampDown 60% from the peak, and nobody is talking about it

Cryptocurrency was, for a period, Robinhood's most exciting business and the principal reason the stock re-rated in 2024 and early 2025. It is now the company's smallest major revenue line and the only one that is shrinking. Crypto transaction revenue was $253 million in the March 2025 quarter and $100 million in the June 2026 quarter, a decline of 60%.

The decline has three components, and separating them matters. The first is asset prices and sentiment: crypto trading volumes across the industry fell through late 2025 and the first half of 2026, and Robinhood's crypto notional volumes went from $82 billion in the December 2025 quarter to $66 billion and then $40 billion. The second is the closure of the retail spot market relative to institutional venues: the Robinhood app's own crypto volumes fell 46% year over year in August 2026, while Bitstamp's fell 30%, meaning the retail-facing business is deteriorating faster than the institutional one. The third is that the company has not lost share; the whole market contracted.

The Bitstamp acquisition, completed in June 2025, was intended to solve a structural problem. The Robinhood app's crypto business was retail-only, volatile, and dependent on a small number of assets. Bitstamp brought institutional flow, a global licence footprint, and roughly $10 billion a month of notional volume. On the numbers, it has done what it was bought to do: Bitstamp contributed $22 billion of the $40 billion of crypto notional volume in the June 2026 quarter, and its revenue is less correlated with retail sentiment than the app's. It has not, however, offset the decline in the retail business, because nothing could.

There is an important accounting point here that is easy to miss. Robinhood's crypto revenue is a take rate on notional volume, and the take rate has been remarkably stable at roughly 25 to 30 basis points. This means the revenue decline is almost entirely a volume phenomenon rather than a pricing phenomenon. That is good news in the sense that the business has not been commoditised; it is bad news in the sense that Robinhood has no lever to pull. It cannot raise prices to compensate, and it cannot manufacture volume.

The macro backdrop at the time of writing is genuinely mixed. Total crypto market capitalisation rose 17.6% in September 2026 to $2.70 trillion, and bitcoin posted a 24.8% seven-day gain — a move in the top 1% of weekly swings since 2020 — driven by a rates trade rather than by fundamentals. Binance Research's own analysis notes that the short squeeze which produced the move has largely played out, and that the next leg depends on fresh demand rather than positioning. Meanwhile the Federal Reserve's policy direction has turned hawkish, with rate-hike odds near 60%.

We would not underwrite a recovery. The more useful observation is about the shape of the company's revenue: crypto contributed $61 million less revenue in Q2 2026 than in Q2 2025, and event contracts contributed $140 million more. The company replaced a cyclical, sentiment-driven revenue line with a legally contested one. It is not obvious that this is an improvement in earnings quality, even though it produced a better quarter.

10 — Tokenisation and the Innovation ExemptionThirty-seven billion of volume, and a rounding error of revenue

On 17 September 2026, the Securities and Exchange Commission issued an order granting temporary, conditional exemptive relief — the agency called it the "Innovation Exemption" — from the definition of "exchange" in the Securities Exchange Act of 1934, to permit qualifying "Tokenized Securities Venues" to trade tokenised National Market System stock through permissioned automated market makers and liquidity pools. The exemptions expire five years after publication. The order is open for public comment.

Robinhood's shares rose 9.12% the following day, adding roughly $9 billion of market capitalisation. The reaction was rational in direction and, we think, excessive in magnitude. To explain why requires separating three things the market is treating as one: the legal permission, the product, and the revenue.

The legal permission. The conditions attached to the exemption are more restrictive than the headline suggests, and three of them matter. First, a Tokenized Securities Venue must verify that the tokenised stock it lists "provides holders the same rights and privileges as does traditional NMS stock of an equivalent class." Robinhood's current Stock Tokens are issued by Robinhood Assets (Jersey) Limited as tokenised debt securities under Regulation S, give holders economic exposure but no voting rights, and are not available to US persons. They do not satisfy that condition today. Second, before listing a tokenised stock tokenised by an unaffiliated third party, the venue must give written notice to the issuer of the underlying stock and an opportunity to object — a direct response to the AMC dispute discussed in section 19. Third, trading in a tokenised stock must be halted concurrently with any halt in the underlying stock on its primary listing exchange. There are also limits on the number of symbols and the volume traded.

This is a permission to experiment inside a fence, not an opening of the market. The SEC's own framing — "while temporary" and "while the Commission considers the need for additional action" — makes clear that the durable rulemaking is years away, and that the exemption is a bridge.

The product. Robinhood's tokenisation franchise is further along than the exemption. Robinhood Chain, an Ethereum Layer 2 network for tokenised real-world assets, launched its public mainnet on 1 July 2026. By early September it had processed roughly 576 million transactions across about 12.3 million addresses, with around $34.6 billion of cumulative decentralised exchange volume and roughly $791 million of bridged total value locked. More than 190 Stock Tokens are live, available to eligible users in more than 120 countries through Robinhood Wallet, with cumulative stock-token notional volume above $3 billion. Robinhood Earn, a decentralised lending product, is live in-app. Perpetual futures have launched in the EU.

The revenue. This is where the story and the accounts diverge, and it is the most important thing in this section. Robinhood's own monthly metrics footnote discloses that "Robinhood Chain revenue is shared with its launch partners, with Robinhood retaining 50% of sequencer revenue until a total of approximately $50M; then 70% until a cumulative total of approximately $150M; and finally 85% of any additional sequencer revenue in excess of approximately $150M." The company also states that Chain volumes are not included in crypto trading volumes, which means this activity is not yet visible in the revenue lines at all.

So consider the arithmetic. Robinhood Chain has generated roughly $34.6 billion of cumulative decentralised exchange volume and $3 billion of stock-token notional volume — call it $37.6 billion of on-chain activity — and the company's retained economics are capped in the low hundreds of millions across the entire life of the network to date. Against $5.2 billion of consensus revenue for 2026, that is a rounding error. The take rate on on-chain activity, from the company's own disclosure, is on the order of a fraction of one percent.

We are not dismissing the franchise. Tokenised equity is a genuinely large opportunity, the SEC has now given it a legal pathway, and Robinhood has the distribution to be a leading venue: 28.6 million funded customers, a wallet in 120 countries, and the only consumer brand in the space. Coinbase is pursuing the same opportunity and, on some measures, is ahead — it already offers one-to-one redemption for underlying shares and distributes dividends in cash, whereas Robinhood's tokens increase token quantities instead. The competition will be real.

The point is about price, not promise. A 9% single-day re-rating on a regulatory exemption for a product whose contribution to revenue is currently immaterial, and which will not be permitted in its current form under the exemption's own conditions, is a statement about sentiment rather than about cash flows. Investors who bought that day are paying for a five-year option on a business that does not yet exist in the accounts.

11 — The Trump Accounts mandateA government contract, a cost-plus margin, and seven million children

In April 2026, the US Treasury named Robinhood as the brokerage for Trump Accounts — a new tax-advantaged investment account for American children — and as sole initial trustee, working with Bank of New York Mellon. The accounts launched on 4 July 2026. By the time of the second-quarter earnings call, over seven million children had been signed up and nearly $1.5 billion had been deposited.

This is a remarkable development for a company that spent the first decade of its existence in conflict with regulators. It is also, on the numbers, a small business with an unusual economic structure that is frequently misread in both directions.

The contract is cost-plus. Robinhood builds and operates the user interface and the account infrastructure; BNY provides the financial plumbing. Robinhood earns a service fee, and the company has disclosed that the work is "contracted on a cost-plus basis with a small margin, so revenues are expected to exceed costs." In other words, this is not a profit centre. It is a break-even-plus arrangement that produces revenue roughly equal to the expense it requires.

The financial footprint is visible in two places. On the revenue side, other revenue rose 54% year over year to $143 million in the June quarter, and the company attributes the increase to "Trump Account service revenues and increased Robinhood Gold subscription revenues." On the cost side, general and administrative expenses rose 51% to $199 million, and the company cites "expenses related to Trump Accounts and Rothera." The company also revised its 2026 non-GAAP expense guidance upward by $100 million in April 2026 specifically to build and support the Trump Accounts user interface.

So why does it matter? Three reasons, none of them in the current income statement.

First, distribution. Seven million sign-ups is a customer acquisition channel of a size Robinhood could not buy. The accounts are excluded from funded customers and from investment accounts, which is why the customer count grew only 7% despite the launch. But the accounts belong to children, and the parents and guardians who open them are the exact demographic Robinhood targets. The company has not disclosed conversion rates from Trump Account openers to funded customers, and until it does, any valuation credit for this is speculative.

Second, legitimacy. Being selected as the federal government's brokerage is a regulatory imprimatur that a company with Robinhood's enforcement history could not have obtained five years ago. It changes how the company is perceived by institutional counterparties, by state regulators, and by the politicians who will decide the fate of the prediction-market and tokenisation questions. That has value even if the contract itself does not.

Third, and least comfortably: political entanglement. Robinhood is now operationally dependent on a programme named after and championed by a sitting president, administered by a Treasury whose leadership changes with administrations. The programme's funding, its account limits, and its continued existence are policy variables. A company that has been handed a national mandate by one administration should assume the mandate is reviewed by the next. We do not think this is a near-term risk — the programme is popular and has bipartisan appeal as a children's savings vehicle — but it is a new category of exposure for this equity, and it is not one that appears in any risk factor Robinhood has historically filed.

12 — Banking, cards and advisoryThirteen lines above $100 million, and the ones that are not yet

Robinhood disclosed in the June 2026 quarter that thirteen business lines have reached $100 million or more in annualised revenue. That is a genuinely impressive statistic for a company that was, five years ago, a single-product trading app. It is also a statistic that deserves to be unpacked, because the thirteen lines are not thirteen businesses of equal quality.

Robinhood Banking had over $3 billion of deposits from more than 240,000 funded customers at the end of the second quarter, with roughly 40% signed up for direct deposit. Direct deposit is the metric that matters: a customer who receives their salary into a Robinhood account has made the platform their primary financial relationship, and those customers have materially higher deposit balances and lower churn. The business is early — $3 billion against Schwab's $400 billion-plus of bank deposits — but the direction is right, and the strategic value of a direct deposit relationship is far greater than the spread income it generates today.

The Gold Card crossed one million customers with more than $17 billion of annualised purchase volume, and the credit-card business as a whole passed $100 million of annualised revenue. The Platinum Card began rolling out. This is the most double-edged of the adjacent bets. Credit cards are a good business when credit is good and a bad business when it is not, and Robinhood is issuing cards to a customer base with limited credit history in a period when consumer credit conditions are normalising. The provision for credit losses rose 100% year over year to $56 million in the June quarter, and $92 million in the first half against $114 million for all of 2025. The company is provisioning appropriately relative to the portfolio, but the portfolio has not been through a recession.

Robinhood Retirement held $34.5 billion of assets, up 82% year over year, across roughly 1.9 million funded accounts. This is, in our view, the most valuable adjacent business, because retirement assets are the stickiest money in financial services. A customer who rolls a 401(k) into a Robinhood IRA does not leave for a better app. The 1% match on contributions is a straightforward subsidy to acquire those assets, and it is working.

Robinhood Strategies passed 300,000 funded customers and nearly $2 billion of assets, growing from 285,000 customers and $1.6 billion in the prior quarter. This is the company's managed-portfolio product and its answer to the robo-advisers, with the important difference that it is designed to be sold to customers who already trade actively rather than to first-time investors. The assets are small; the strategic logic is that a managed portfolio is the first step towards charging an advisory fee rather than a transaction fee, which would be a genuine improvement in revenue quality.

TradePMR, acquired in 2025, runs the Robinhood Advisor Network with $50 billion of assets under management, serving registered investment advisers. This is a business-to-business channel that gives Robinhood a route to assets it will never reach through consumer marketing, and it is the only part of the company that looks like a traditional custody franchise.

Futures and index options are growing quickly from a small base, and short selling rolled out in November 2025. Rothera, the CFTC-licensed exchange joint venture with Susquehanna, has traded over 3.5 billion contracts since its June 2026 launch and is the venue through which event contracts are cleared. Legend, the active-trader platform, passed $100 million of annualised revenue roughly eighteen months after launch.

The pattern across all of these is consistent: Robinhood launches products that monetise engagement and grow them quickly, and none of them is yet large enough to change the cyclical character of the company. The thirteen lines above $100 million are thirteen bets on the same customer, which is a legitimate strategy and also a concentration risk. If the active retail trader stops trading, thirteen business lines decline together.

13 — Quality of earningsClean books, one asterisk, and one thing to watch

Quality-of-earnings analysis at most financial companies produces a long list of complaints. It does not here, and that is worth saying plainly before discussing the one genuine issue.

Robinhood's revenue recognition is straightforward. There is no channel stuffing, because there is no channel. There are no large capitalised costs being amortised into future periods, because the business is not capital-intensive — capital expenditure was $26 million over the trailing twelve months against $4.9 billion of revenue. There is no serial restructuring programme being excluded from adjusted figures year after year; the June 2026 reduction in force produced a $23 million charge and it was disclosed as such. There are no related-party revenue arrangements of significance. The reported revenue is the cash that came in, and the reported expenses are the cash that went out, with the usual accruals.

The one asterisk is the $129 million gain on the deconsolidation of Robinhood Ventures Fund I. RVI is a closed-end fund that Robinhood listed on the New York Stock Exchange; the company held approximately 52% of it as of 31 March 2026, and when it ceased to consolidate the fund in the June quarter it recognised a gain of $106 million in adjusted EBITDA terms and $129 million in net income terms, equivalent to $0.14 of the $0.62 of diluted earnings per share. Reported net income of $573 million therefore included 23% that had nothing to do with the operating business. Excluding it, net income was approximately $444 million, still 15% higher year over year.

We would not make much of a single non-recurring item. The reason we flag it is that it lands in a quarter where the company's narrative is about record profitability, and it is the kind of item that gets quietly folded into a growth story. Robinhood also holds a retained interest in RVI at fair value of $435 million, which will produce further mark-to-market gains or losses in future periods. Those will be non-operating too.

The thing to watch is more structural. The gap between GAAP and non-GAAP profitability at Robinhood is driven principally by stock-based compensation, which was $105 million in the June quarter, up 35% year over year, and $305 million for all of 2025. That is a real cost, it is recurring, and it is excluded from adjusted EBITDA. Total adjusted operating expenses and stock-based compensation were $641 million against GAAP operating expenses of $734 million, a 15% gap. The company's 2026 guidance is given on the adjusted basis, and the gap is not shrinking.

There is one more item that deserves scrutiny, and it is the one that would matter most in a downturn: the provision for credit losses. It doubled year over year to $56 million in the June quarter, and it is excluded from the company's non-GAAP expense guidance. Robinhood is growing a margin book and a credit card portfolio simultaneously, both of which generate provisions that scale with the book rather than with revenue. In a benign credit environment these are small. In an adverse one they are not, and because they are excluded from guidance, the company's expense outlook does not include them.

14 — Capital and the June convertsA free option written 46% above the market

In June 2026, Robinhood issued $2.2 billion of 0.00% convertible senior notes due 2029 in a private placement to qualified institutional buyers, upsized from an initial $2.0 billion on full exercise of the purchase option. The notes pay no interest. The initial conversion price is $174.42 per share, a 46% premium to the closing price on the offering date. Net proceeds were approximately $2.169 billion.

The structure is worth understanding because it is unusually favourable to existing shareholders and because it changes how the balance sheet should be read. Robinhood used approximately $290 million of the proceeds to repurchase 2.743 million Class A shares, and $123.2 million to buy capped-call transactions. The capped calls generally offset dilution on conversion up to a cap initially set at approximately $237.85 per share. Taking the capped calls and the concurrent repurchase together, the company states that it anticipates no net dilution from the transaction until the share price exceeds $303.95 — approximately 154% above the price at the time of issuance.

In plain terms: Robinhood borrowed $2.2 billion for free, used part of it to buy back stock, bought an option structure that neutralises dilution up to nearly $238, and will not issue a net new share to a noteholder unless the stock more than doubles and a half from here. If the stock does not double, the company keeps $2.2 billion of interest-free capital and the notes are repaid or refinanced. This is the cheapest equity-linked capital a company of this size can raise, and it is evidence that the market's appetite for Robinhood paper is strong.

The buyback programme has also been used at far lower prices. Cumulative repurchases since the programme began in the third quarter of 2024 total $1.3 billion, representing 27 million Class A shares at an average price of approximately $47. In the June 2026 quarter alone, Robinhood repurchased $414 million of stock — 4.4 million shares at an average of approximately $94. The board refreshed the authorisation to $1.5 billion in March 2026, expected to be used over roughly three years.

There is a legitimate criticism here. Buying back stock at $94 and $119 when the shares traded at $47 eighteen months earlier is a poor use of capital relative to the alternative of buying at $47, and it is worth asking whether the programme is being run as capital allocation or as a signal. On the other hand, the company's cash position is ample — $5.4 billion of cash and cash equivalents at the end of June 2026, up from $4.3 billion at the end of 2025 — and total stockholders' equity was $9.5 billion with an accumulated deficit reduced to $1.2 billion from $2.2 billion.

On the balance sheet, one point needs making emphatically because third-party data providers get it wrong. Several widely used financial data services report Robinhood as carrying $22.93 billion of debt and a debt-to-equity ratio of 2.40, which would make it one of the most levered companies in the index. That figure is total liabilities, and the overwhelming majority of it is customer-related — payables to customers, segregated cash held on their behalf, and the corresponding assets on the other side of the balance sheet. Robinhood's actual borrowings are $2.17 billion of long-term debt, which is the convertible notes, plus modest credit facilities. On a corporate basis the company carries net cash. Any screen that flags Robinhood as highly levered is screening a broker's customer float against a corporate capital structure, which is a category error.

The genuine capital-structure issue is share count. Shares outstanding were 899.1 million, up 0.89% year over year, having fallen 0.33% quarter over quarter on the buybacks. Diluted weighted-average shares were 912 million against basic of 899 million. Robinhood is not meaningfully diluting shareholders, which is more than can be said for most companies at this stage of maturity. The Class B shares, 109 million of them, carry supervoting rights and are held principally by the founders; any investor buying the Class A stock should be aware that control is not for sale.

15 — Competition I: the brokersNobody is losing customers, and that is the problem

Robinhood competes in a market where commission-free trading is no longer a differentiator, because it is no longer a feature. Every major US retail broker offers zero-commission equities and options, fractional shares, and a mobile app that works. The competitive question has therefore moved from price to three other things: the quality of the trading experience for active customers, the breadth of the financial relationship, and the willingness to offer products that incumbents consider reputationally risky.

Robinhood's position on the first is stronger than it is given credit for. Legend, the desktop-grade trading platform launched in 2024, passed $100 million of annualised revenue within about eighteen months, and equity notional volumes grew 85% year over year in the June quarter against a customer base that grew 7%. That combination — volume per customer rising far faster than customer count — is what a successful move upmarket looks like. Interactive Brokers, which has historically owned the sophisticated retail trader, is the direct competitor here, and Robinhood is now credibly competing for that customer rather than only for the beginner.

On the second, the breadth of the relationship, Robinhood is still the challenger and is closing fast. The relevant comparison is not who has the most assets — Charles Schwab and Fidelity have orders of magnitude more — but who is adding assets fastest as a percentage of the base. Robinhood's net deposit growth rate has run at 22% to 28% of beginning platform assets on an annualised basis for six consecutive quarters. No large incumbent grows net new assets at that percentage rate, because their bases are too large to move that quickly. This is the arithmetic of being small, and it is a genuine advantage while it lasts.

On the third, the willingness to offer contested products, Robinhood has a structural advantage and a structural vulnerability at the same time. Schwab and Fidelity will not, as a matter of brand policy, list sports event contracts. That is why Robinhood and Kalshi have the prediction-markets market largely to themselves at the retail level. It is also why Robinhood is the defendant in waiting if the legal environment turns, and why a firm with a 200-year-old franchise to protect can afford to wait for clarity while Robinhood cannot.

The competitive dynamic that matters most over the next three years is not any of these. It is that the entire retail brokerage industry is being repriced around crypto and prediction markets, and the incumbents have deeper pockets and better regulatory relationships. Interactive Brokers has been expanding its global footprint and has moved into prediction markets and crypto. Schwab has signalled interest in digital assets as the regulatory picture clarifies. Coinbase is building a brokerage. The barrier that protected Robinhood — being the only modern interface — has fallen.

There is one competitive fact that is unambiguously in Robinhood's favour, and it is demographic. The average Robinhood customer is materially younger than the average Schwab or Fidelity customer, and wealth transfers from the older cohort to the younger one over the next two decades. A broker that owns the 25-to-40-year-old investor today owns the deposit and advisory revenue of that investor for the next thirty years. This is why Schwab bought TD Ameritrade and why every incumbent has launched a mobile-first product: they are all trying to buy their way to a younger customer. Robinhood already has one, and paid almost nothing to acquire it.

The bearish reading of the same fact is that the younger customer has less money. Robinhood's annualised revenue per funded customer is $187. Schwab's is a multiple of that. The gap is not a reflection of pricing; it is a reflection of assets. Robinhood's bet is that its customers get richer, and it is a bet that will take a decade to settle.

16 — Competition II: crypto and prediction marketsTwo markets, two very different competitive positions

Robinhood's two newest growth engines face opposite competitive situations, and treating them as a single "alternative assets" theme is a mistake.

In cryptocurrency, Robinhood is a small player in a market dominated by specialists. Coinbase is the reference US venue with a far larger institutional and retail crypto franchise, a deeper balance sheet, and a product set — custody, staking, USDC economics, a Layer 2 of its own — that Robinhood does not attempt to match. The Bitstamp acquisition gave Robinhood institutional credibility, but $10 billion a month of Bitstamp volume is a fraction of what the leading venues handle. Robinhood's crypto advantage is distribution, not infrastructure: it can route 28.6 million existing customers into crypto without paying to acquire them. That is a real advantage and it has a ceiling, because crypto-native customers go where the assets and the yields are, not where the app is nicest.

The competitive position is also deteriorating in an important, under-discussed way. Robinhood's app crypto volumes fell 46% year over year in August 2026 while Bitstamp's fell 30%. The retail business is shrinking faster than the institutional one, which suggests Robinhood's crypto customers are not merely trading less in a down market — they are leaving, or trading elsewhere. When the market recovers, those customers may or may not come back to Robinhood rather than to a crypto-native venue that has been improving its product while Robinhood was focused elsewhere.

In prediction markets, the competitive situation is genuinely favourable and the risk is not competitive at all. The market is essentially a duopoly between Kalshi and Polymarket at the venue level, with Robinhood as the largest retail distribution channel for Kalshi's contracts through its Rothera joint venture. In September 2026 Robinhood agreed to take minority stakes in Crypto.com and its prediction-market spinout OG.com and to list Crypto.com's event contracts, which broadens supply without adding a competitor to the distribution layer. Traditional sportsbooks — DraftKings and Flutter in particular — are the parties with the most to gain from a restriction, and their shares rose 7% and 6% respectively on the day the Ninth Circuit ruled.

This asymmetry is worth stating clearly. In crypto, Robinhood faces better-capitalised specialists and is losing relative ground. In prediction markets, Robinhood faces a fragmented supply base and is the indispensable distribution channel. The market capitalises both as growth. Only one of them is defensible on competitive grounds, and it is the one with the legal problem.

There is a final competitive consideration that applies to both. Robinhood's product advantage has always been speed of launch, and speed of launch is not a moat. Every product Robinhood has shipped in the last three years — event contracts, Gold Card, Legend, retirement matching, banking — has been replicated by at least one competitor within twelve months. The company's durable advantages are its brand with a specific demographic, its cost of customer acquisition, and its regulatory licences. Everything else is execution, and execution advantages decay.

17 — Regulation I: the circuit splitTwo courts, opposite answers, and roughly twenty states in between

The most important legal question hanging over Robinhood's growth is whether event contracts are federally regulated financial instruments or state-regulated gambling. As of September 2026, the federal appellate courts have answered that question both ways.

On 6 April 2026, the Third Circuit Court of Appeals ruled that New Jersey could not regulate Kalshi's platform under federal law, accepting the argument that the contracts are swaps subject to exclusive CFTC oversight under the Commodity Exchange Act. On 28 August 2026, a unanimous three-judge panel of the Ninth Circuit Court of Appeals ruled the opposite way in the Nevada case, holding that the substance of sports event contracts is sports gambling regardless of characterisation, that the Commodity Exchange Act does not preempt state gaming law, and that Nevada's regulators may oversee the product. The court also sent the question of Kalshi's election contracts back to the district court.

Several details of the Ninth Circuit decision make it more consequential than a routine appellate loss. The panel was unanimous. It was composed of three judges appointed by the same president whose administration's CFTC has been the industry's principal federal ally — and it ruled against the position that agency has taken. The court's language was pointed: it called Kalshi's litigation position "disingenuous" given the company's own marketing. And the ruling binds federal judges in nine states, including California and Arizona, where a district court had previously enjoined Arizona's attorney general from proceeding against Kalshi. That injunction now rests on reasoning the Ninth Circuit has rejected.

The practical landscape is a patchwork. Roughly twenty states are in active litigation over prediction markets. Four — Nevada, Massachusetts, Michigan and Washington — have obtained court orders restricting Kalshi's operations. Kalshi agreed to geofence Nevada by 12 August 2026 and then found itself facing a claim of non-compliance and daily penalties of $120,000, which it disputes. The CFTC, meanwhile, has claimed exclusive oversight and challenged regulatory activity in nine states including New York.

A circuit split on a question of this commercial magnitude is one of the most reliable predictors of Supreme Court review. The Court is not obliged to take the case, and it may wait for additional circuits to weigh in, but the conflict is squarely presented and the industry's economic stakes are large. A decision, if certiorari is granted in the current term, would likely arrive in the first half of 2027. If the Court declines, the patchwork persists and the Ninth Circuit's rule governs in the western states indefinitely.

For Robinhood specifically, the exposure is a function of geography it has not disclosed. The company does not break out event-contract revenue by state, and it has not said what proportion of its customers sit in the nine states bound by the Ninth Circuit or the four with restricting orders. California and Arizona alone are material retail markets. An investor cannot size this risk from public disclosure, which is itself a reason to apply a discount rather than a premium.

We would frame the probability distribution as follows. A favourable Supreme Court ruling that confirms federal preemption: perhaps 30%. A narrow ruling or denial of certiorari that leaves the patchwork in place: perhaps 45%. An adverse ruling that broadly validates state regulation: perhaps 25%. In the middle scenario the product survives with a materially smaller addressable market; in the adverse scenario it becomes a licensing business with a fraction of its current economics. The market, on our reading, is pricing something closer to the first scenario than to the distribution above.

18 — Regulation II: crypto and tokenised equityA bill that has not passed, and an exemption that has

Robinhood's regulatory exposure extends beyond prediction markets into two adjacent areas: the market-structure legislation that would govern crypto, and the SEC's new framework for tokenised equities. The two are moving in opposite directions.

The CLARITY Act — formally the Digital Asset Market Clarity Act — is the most consequential piece of crypto legislation in a decade and it has not passed. The House approved it in July 2025 by 294 votes to 134. The Senate Banking Committee advanced its version in a bipartisan 15–9 vote on 14 May 2026, but the bill was shelved before the August recess without a cloture motion, and it needs roughly seven Democratic votes to reach the 60-vote threshold. Prediction-market odds of passage fell from above 80% in February 2026 to approximately 34% by late July. The single blocking issue is an ethics provision: Democrats want restrictions on officials profiting from digital assets they regulate, enforced by more than the Justice Department alone, and the Republican text negotiated with the White House was rejected within hours of release.

Two windows remain in 2026: a roughly three-week Senate session in September, competing with appropriations deadlines and avoided by legislators before elections, and a year-end attachment to must-pass legislation, which solves the floor-time problem but not the votes problem. If neither produces a result, the bill must be reintroduced in the next Congress in January 2027, and analysts warn that missing 2026 could delay market-structure legislation by years.

In the meantime, crypto is governed by an SEC–CFTC joint interpretive release from March 2026, which names sixteen digital assets as digital commodities and places staking, mining and airdrops outside securities law. That is agency policy, revocable by a vote, not statute. The GENIUS Act, which is real law, covers payment stablecoins only, and its implementing agencies have already missed their one-year rulemaking deadline — a useful base rate for how quickly the CLARITY Act's registration regimes would actually arrive. On any realistic schedule, the functioning framework the industry wants lands somewhere between 2028 and 2029.

The Innovation Exemption, by contrast, is real and immediate. The SEC's 17 September 2026 order grants qualifying Tokenized Securities Venues temporary relief from the exchange definition, subject to conditions, for five years, and solicits comment on further action. It is a genuine opening. It is also narrower than the market's reaction implied, for the reasons set out in section 10: the same rights and privileges condition that Robinhood's Regulation S debt-instrument tokens do not currently satisfy, the notice-and-objection requirement for third-party tokenisation, the volume and symbol limits, and the concurrent-halt requirement.

There is a second regulatory thread that matters more than it appears. In January 2026 the SEC delineated three frameworks for securities tokenisation, spanning direct issuer-tokenised securities, third-party custodial structures, and synthetic instruments that merely replicate price movements. Robinhood's Stock Tokens fall in the third category, and the Innovation Exemption's rights-and-privileges condition is aimed squarely at moving products out of it. If the SEC's eventual rulemaking requires genuine shareholder rights, Robinhood will have to rebuild the product — which it has already begun to signal it will do, having announced in-kind redemption and voting rights for Stock Tokens in September 2026.

The European comparison is instructive and unflattering to the US timeline. Europe's MiCA reached full enforcement across all twenty-seven member states on 1 July 2026, with a single framework, authorised service providers and passporting. The UK, Japan, Singapore, Hong Kong and the UAE have operating regimes. Robinhood's international expansion — Singapore, Canada via WonderFi, the EU perpetuals launch, UK crypto plans — is in part an arbitrage on that divergence, and it is a rational one. But it also means the company's growth increasingly depends on regulatory regimes it does not influence, in jurisdictions where it has less political capital than it does at home.

19 — Issuer consent and the AMC affairA four-day feud that defined a market

In the first week of September 2026, Adam Aron, the chief executive of AMC Entertainment, publicly attacked Robinhood for issuing a token that tracked AMC's share price without AMC's involvement. His language was unrestrained — "contemptible, outrageous, disgusting, detestable, inexcusable, vile" — and his demand was that Robinhood "CEASE AND DECIST." He threatened to take the matter to the SEC. AMC shares rose close to 20% in premarket trading on the second day of the dispute.

Vlad Tenev's initial response was two words: "What's the concern?" His chief legal officer, Dan Gallagher, a former SEC commissioner, was more expansive: "We know a little something about the U.S. securities laws and will not 'DECIST.' Send your lawyers and we'll educate them." Tenev then appeared on CNBC and argued the substantive position: issuers govern the rights of the securities they sell, but not every financial product built around them, and third-party instruments referencing a listed share should not require issuer consent.

Stripped of the adjectives, Aron's complaint had three parts, and each deserves an answer rather than a dismissal. First, that the tokens could complicate AMC's ability to raise capital, because a parallel market trading on its ticker without its consent could fragment its shareholder base. Second, that token holders get no shareholder rights — no vote, no place on the register — while carrying economic exposure to the shares. Third, that an offshore entity is monetising a US company's ticker while bearing none of the compliance cost that gives that ticker its meaning.

On the facts, Robinhood's narrow defence is correct. Its Stock Tokens are issued by Robinhood Assets (Jersey) Limited, a private company incorporated in Jersey, as tokenised debt securities under Regulation S, the cross-border safe harbour for offers outside the United States to non-US persons. Each token is an ERC-20 contract priced through on-chain Chainlink feeds, giving economic exposure without legal or beneficial rights in the underlying company. The instruments are not registered under US securities law and may not be offered to US persons. They are not AMC shares, and Robinhood never said they were.

That defence does not answer Aron's actual objection, which is a claim about market structure rather than about disclosure. He is arguing that an offshore entity can create a tradable claim referencing a public company's equity, sell it to people who will never appear on the register, and thereby create a market in that company's name that the company cannot police. That is a novel question, and the fact that the answer is not obvious is the point.

Robinhood then conceded on the product features while holding the legal line. Tenev announced on 14 September 2026 that in-kind redemption and voting rights were coming for Stock Tokens. Johann Kerbrat, who leads crypto, said one-to-one redemption for underlying shares is in active development, with voting on the roadmap, pointing to Say by Robinhood — the company's shareholder-engagement platform — as existing infrastructure. Coinbase made a parallel announcement on voting rights for its own tokenised equity products the same week, and Coinbase already offers one-to-one redemption and cash dividend distribution, which puts it ahead of Robinhood on features.

Not everyone was satisfied. Carlos Domingo, chief executive of the tokenisation platform Securitize, argued that these products "are not 'stocks'" and that marketing them as stock tokens creates investor confusion. He flagged a technical problem that has no clean solution: extending shareholder rights to tokens that transfer freely across blockchain addresses requires knowing who the beneficial owner is, and free transfer is precisely what makes the instrument useful. He also noted that Robinhood's dividend treatment — increasing token quantities rather than paying cash — diverges from standard equity conventions. Graham Rodford of the UK-regulated exchange Archax drew the line differently again, distinguishing between placing an actual share on a blockchain and issuing a separate instrument that tracks one.

There is a market-integrity detail in this episode that deserves more attention than it received. One AMC-linked token pair traded at roughly sixty times AMC's reference share price, a gap that demonstrates how thin token markets can detach from the assets they track. If the whole premise of tokenised equity is that it delivers the economics of the underlying share, a pricing dislocation of that magnitude is evidence that the mechanism does not always work — and that the investors absorbing the difference were retail.

We draw three conclusions from the affair. First, Robinhood's legal position on issuer consent is defensible and, given the SEC's own notice-and-objection condition in the Innovation Exemption, is likely to be substantially preserved. Second, the product will have to change: rights and redemption are coming because the regulator has effectively required them, not because Robinhood chose to offer them. Third, the reputational cost of this dispute is real. Robinhood spent a decade arguing that it democratises finance and is now being publicly accused by a listed company of running a "quasi-fake market." That framing will be repeated in the next regulatory proceeding, and it will not help.

20 — ValuationThe multiple depends entirely on which earnings you use

Robinhood closed at $119.82 on 18 September 2026, a market capitalisation of $107.7 billion on 899.1 million shares. Enterprise value is $106.8 billion. The stock has risen 9.12% in a single day and sits 27% above its 200-day moving average of $94.90. The trailing price-to-earnings ratio is 53.1; the forward ratio on consensus adjusted earnings is 47.2. Price-to-sales is 21.8 times trailing revenue and 18.1 times forward revenue.

Those are software multiples on a brokerage. The question is whether they are justified, and the answer turns on four things that must be examined separately rather than blended into a single "fintech" multiple.

First, which earnings? Robinhood's reported earnings are unusually sensitive to definition. Trailing twelve-month GAAP net income was $2.07 billion, or $2.26 per diluted share. Trailing twelve-month adjusted EBITDA was $2.78 billion. Consensus expects $2.54 of adjusted earnings per share in 2026 and $3.38 in 2027. The gap between the GAAP and adjusted figures is principally stock-based compensation, which is a real cost. Using trailing GAAP earnings, the stock trades at 53 times. Using 2027 consensus adjusted earnings, it trades at 35 times. Both are correct descriptions of the same price.

Second, what growth is embedded? Consensus expects revenue to grow 16% in 2026 to $5.2 billion and 27% in 2027 to $6.6 billion, with adjusted earnings per share growing 9% in 2026 and 33% in 2027. The 2027 acceleration is the load-bearing assumption in the entire valuation, and it requires three things to happen at once: event contracts to keep growing despite the legal cloud, crypto to recover, and net deposits to keep compounding at 20%-plus. If any one of the three fails, the 2027 number is not achievable and the multiple compresses.

Third, what is the right multiple? The honest answer is that it depends on the durability of the revenue, and Robinhood's revenue is less durable than its margin suggests. A company with 46% GAAP operating margins and 27% expected revenue growth would normally command 40 to 50 times forward earnings. But that is the multiple for a business whose growth is recurring. The comparable set for Robinhood is not software; it is high-growth financial intermediaries, which historically trade at 20 to 30 times forward earnings because their revenue is levered to markets. We think the truth sits between the two: the subscription layer, the retirement assets and the direct-deposit relationships deserve a software-like multiple; the transaction and event-contract revenue deserves a broker-like one.

Fourth, what is priced in? At 40 times earnings the current price requires $3.00 of earnings per share. Consensus expects $3.38 in 2027. So the market is not demanding a miracle; it is demanding that the 2027 consensus is met and that the multiple holds at a level normally reserved for recurring-revenue businesses. That is a fair description of a fairly priced stock, and it is the arithmetic behind our Neutral rating.

The scenario range — $70 to $189 — is wider than we would normally accept, and the width is itself the conclusion. A stock whose twelve-month outcome spans 2.7-fold depending on two decisions taken by other people is a stock that should be sized accordingly. It is also a stock where the distribution is not symmetric in the way the price suggests: the bull case requires an affirmative court ruling and a crypto recovery, while the bear case requires only that the current legal trend continues and that crypto stays where it is.

One further observation on the price action itself. The stock rose 9.12% on 18 September on a regulatory exemption for a product that contributes immaterial revenue today. It rose 1.6% on 14 September when Deutsche Bank raised its target from $136 to $138 — an increase of under 2% that moved a $107 billion company by $1.7 billion. This is a stock with a beta of 2.34 and annualised volatility near 74%, which means it trades on narrative rather than on estimate revisions. Investors who buy it should understand that they are buying a high-beta instrument whose news flow is regulatory and political, not operational.

21 — The bull caseSix arguments we take seriously

The bull case for Robinhood is stronger than its critics allow, and it does not rest on the speculative parts of the story. Here are the six arguments we find most persuasive, in descending order of strength.

One: the operating leverage is real and unusual

A 57% adjusted EBITDA margin and a 46% GAAP operating margin on a retail brokerage is extraordinary. It reflects a business whose costs are largely fixed and whose revenue scales with activity, and it means that incremental revenue drops to the bottom line at a very high rate. In the June 2026 quarter, revenue rose 32% and adjusted EBITDA rose 35%. In the March 2026 quarter, revenue fell 17% and adjusted EBITDA fell 30%. The leverage cuts both ways, but over a full cycle a business that converts a third of incremental revenue into EBITDA will compound earnings faster than revenue, and that is what has happened: revenue grew 52% in 2025 and adjusted EBITDA grew 76%.

Two: the asset-gathering engine is genuinely differentiated

Net deposits of $74.1 billion over the trailing twelve months, a 24% annual growth rate against the prior-year asset base, is a rate no incumbent can match. It reflects a product set that is comprehensive enough to be a primary financial relationship for a young customer — brokerage, retirement, banking with direct deposit, a credit card, a managed portfolio — and a brand that this cohort trusts. Assets under custody are the best leading indicator of future revenue in this industry, and Robinhood is accumulating them faster than anyone.

Three: Gold converts activity into subscription revenue

Four point eight million subscribers, up 39%, at 17% penetration, with roughly 40% of new customers signing up in their first quarter. Subscription revenue is worth more than transaction revenue at the same level because it is predictable and because it changes customer behaviour. The company has not disclosed churn, which is a gap, but the growth rate and the attach rate are both consistent with a product that customers value rather than a promotional giveaway.

Four: the balance sheet is unencumbered

$5.4 billion of corporate cash, $2.2 billion of zero-coupon convertible notes that do not dilute until the stock exceeds $303.95, share count down quarter over quarter, and an accumulated deficit halved in six months. Robinhood is not financing its growth with debt or with equity, and it has the capacity to make acquisitions — Indonesia, WonderFi and TradePMR were all bolt-ons funded from the balance sheet. This is a company with options.

Five: the product velocity is a genuine competitive advantage

Thirteen business lines above $100 million of annualised revenue, in a company that listed five years ago with one product. Banking, cards, retirement, advisory, futures, index options, short selling, event contracts, tokenised equities and a Layer 2 blockchain have all been launched or acquired since 2024. Competitors replicate each individual product within a year, but they do not replicate the rate at which Robinhood ships, and over a decade that compounds into a broader platform than any incumbent can assemble organically.

Six: the demographic position is a twenty-year asset

Robinhood owns the younger investor. Revenue per funded customer of $187 is low today because those customers have not accumulated assets yet, not because Robinhood prices poorly. As wealth transfers from the baby-boom cohort to the millennial and Gen Z cohorts over the next two decades, the broker that already has the relationship captures the assets without paying to acquire them again. This is the argument that justifies a premium multiple, and it is the one we find hardest to rebut.

22 — The bear caseSix arguments we cannot dismiss

The bear case does not require believing that Robinhood is a bad company. It requires only that the price reflects a growth rate that the company cannot control.

One: the fastest-growing revenue line is legally contested

Event contracts contributed $156 million in the June quarter and drove most of the incremental revenue growth. The Ninth Circuit has ruled that states may regulate the product as gambling, the Third Circuit has ruled the opposite, and roughly twenty states are litigating. Four have already obtained restricting orders. Robinhood does not disclose the geographic composition of this revenue, so the exposure cannot be sized from public information. A product that is 12% of revenue and most of the growth, whose legality is undecided, should not be capitalised at the same multiple as a subscription.

Two: crypto is structurally declining and Robinhood is losing relative ground

Crypto revenue is down 60% from its peak. App notional volumes fell 46% year over year in August against a 30% decline at Bitstamp, meaning the retail-facing business is deteriorating faster than the institutional one. The take rate has been stable, so there is no pricing lever to pull. When the market recovers, crypto-native venues with better infrastructure will compete for the same customers. Robinhood's crypto advantage is distribution, and distribution does not retain a customer whose assets live elsewhere.

Three: net interest revenue is a spread business in a cutting cycle

The line peaked at $456 million in the September 2025 quarter. Margin balances have more than doubled and cash and deposits are up 34%, yet net interest revenue grew only 9% year over year, because spreads compressed and securities lending revenue collapsed 32%. In February 2026 the company deliberately moved its cheapest funding — more than $6 billion of Cash Sweep balances — into free credit balances to fund margin lending. That trade works while rates are high. It works against the company as they fall, and the credit provisions on the margin book are already doubling year over year.

Four: the growth rate is decelerating and consensus has not fully adjusted

Revenue growth has gone from 52% in 2025 to 15% in Q1 2026 and 32% in Q2 2026 against an easy comparison. Consensus expects 16% for the full year 2026. The company's own expense guidance of $2.675 to $2.775 billion on a non-GAAP basis excludes credit losses and restructuring, and H1 2026 GAAP operating expenses annualise to $2.78 billion. If revenue growth lands at the low end and expenses at the high end, the earnings base that justifies a 47 times multiple does not materialise.

Five: the tokenisation revenue is immaterial and the market has already paid for it

Robinhood Chain has generated $34.6 billion of cumulative DEX volume and $3 billion of stock-token volume, but the sequencer revenue the company retains is shared with launch partners on a sliding scale capped in the low hundreds of millions, and the company states that Chain volumes are not included in crypto trading volumes — meaning the activity is not in the revenue lines at all. The stock rose 9% on a regulatory exemption for this product. That is $9 billion of market capitalisation for a business whose current contribution to revenue is not separately disclosed because it is not material.

Six: the company is systematically exposed to rules written after it acts

Prediction markets, tokenised equities, crypto market structure, the Trump Accounts mandate — in each case Robinhood builds first and the rules are set afterwards by courts, legislatures and agencies. That has been enormously profitable. It also means the company's largest risks are exogenous, binary, and not manageable by execution. For a stock trading at 53 times trailing earnings, that is the wrong risk profile.

23 — Risk matrixRanked by probability times impact

The matrix below is a Farstar assessment rather than a company disclosure. Positions are judgements, and reasonable analysts will place some of them differently. What matters is the clustering: the four highest-ranked risks are all exogenous, and three of them touch revenue lines that together account for roughly a quarter of the total.

Two mitigants in that table deserve scepticism. Geofencing is an operational control, not a legal defence: it reduces the number of customers who can trade a restricted product, but it does not change the fact that the company is distributing it, and the Nevada dispute has already produced allegations that Kalshi failed to comply with a geofencing order. And the "subscription revenue layer" as a mitigant against a normalisation in retail activity is a $240 million annual run-rate against $5.2 billion of revenue — helpful, not protective.

24 — CatalystsWhat we are watching, and when

25 — ConclusionNeutral at $128, on a business that is performing and a price that assumes it will keep performing

Robinhood in September 2026 is a better company than at any point in its history. It has 28.6 million funded customers, $384 billion of platform assets, a 57% adjusted EBITDA margin, thirteen business lines above $100 million of annualised revenue, no net corporate debt, and a share count that is falling rather than rising. It has turned a business that lost $4.7 billion in two years into one that earned $2.07 billion over the last twelve months. Management's record of finding new products before competitors is, on the evidence, the best in retail financial services.

None of that is the issue. The issue is that the composition of the growth has changed in a way the price does not reflect. Two years ago Robinhood's incremental revenue came from options and net interest — mature, litigated, well-understood businesses. Today it comes from event contracts, whose legality is being decided by two courts that disagree, and from equities and options activity that is levered to a market that has been unusually favourable. Meanwhile the most bank-like and predictable line in the company peaked four quarters ago and the crypto business has halved.

We are not bearish. The base case in this report is a $129 stock, and the base case is the most likely single outcome. The subscription layer is real, the retirement and deposit franchises are genuinely valuable, the demographic position is a twenty-year asset, and the balance sheet gives management options that most competitors do not have. If the Supreme Court resolves the prediction-market question favourably, this is a $189 stock on our numbers, and the company will have earned it.

But the range is $70 to $189, and the difference between those two numbers is not execution. It is a judicial ruling and a regulatory decision, taken by people who do not work for Robinhood and are not accountable to its shareholders. When a company's twelve-month outcome is determined by two external binary events, the correct response is not to avoid the equity. It is to refuse to pay a premium for the favourable branch before it is decided.

At $119.82, after a 9% single-day move on a regulatory exemption for a product whose contribution to revenue is not yet separately disclosed, the market is paying for the favourable branch. We initiate coverage with a Neutral rating and a 12-month price target of $128. We would revisit that rating in either direction — below $95, where the base case works at a broker-like multiple, or above $150 following a favourable resolution of the prediction-market question, where the risk would finally be priced rather than pending.

26 — AppendixFinancial summary tables

Annual income statement

Quarterly operating metrics

Scenario valuation detail

27 — SourcesPrimary and secondary references

Company filings and releases

• Robinhood Markets, Inc., Form 10-K for the fiscal year ended 31 December 2025, filed with the SEC.
• Robinhood Markets, Inc., Form 10-Q for the quarter ended 30 June 2026.
• Robinhood Markets, Inc., Form 10-Q for the quarter ended 31 March 2026.
• Robinhood Markets, Inc., Second Quarter 2026 Results, 29 July 2026.
• Robinhood Markets, Inc., First Quarter 2026 Results, 28 April 2026.
• Robinhood Markets, Inc., Fourth Quarter and Full Year 2025 Results, 10 February 2026.
• Robinhood Markets, Inc., Reports August 2026 Operating Data, 10 September 2026.
• Robinhood Markets, Inc., Reports July 2026 Operating Data, August 2026.
• Robinhood Markets, Inc., Closes Offering of $2.2 Billion of 0.00% Convertible Senior Notes Due 2029, 25 June 2026.
• Robinhood Markets, Inc., Q2 2026 earnings call transcript, 29 July 2026.
• Robinhood Newsroom, product and corporate announcements, April–September 2026.

Regulatory and legal

• US Securities and Exchange Commission, SEC Issues "Innovation Exemption" to Facilitate the Trading of Tokenized NMS Stock and Request for Comment, Release No. 2026-90, 17 September 2026.
• Statement of SEC Chairman Paul S. Atkins on the Innovation Exemption, 17 September 2026.
• Kalshi v. Nevada Gaming Control Board, US Court of Appeals for the Ninth Circuit, decision of 28 August 2026.
• US Court of Appeals for the Third Circuit, decision on New Jersey's authority over Kalshi event contracts, 6 April 2026.
• Digital Asset Market Clarity Act (CLARITY Act), H.R. 3633, 119th Congress; Senate Banking Committee markup, 14 May 2026.
• SEC and CFTC joint interpretive release on digital assets, March 2026.
• European Union, Markets in Crypto-Assets Regulation (MiCA), full application from 1 July 2026.
• FINRA disciplinary actions concerning Robinhood Financial and Robinhood Securities, 2025–2026.

Third-party research and industry data

• Binance Research, Monthly Market Insights, September 2026, on crypto market capitalisation, bitcoin volatility and trading volume composition.
• DefiLlama, on-chain data for Robinhood Chain, total value locked and cumulative DEX volume, September 2026.
• CoinDesk Research, on tokenised equity inflows and real-world-asset issuance, 2026.
• Synergy and Mordor Intelligence, US online trading platform and securities brokerage market analysis, 2026.
• StockBrokers.com, broker platform comparison, 2026.
• Reporting on the AMC–Robinhood stock-token dispute and the tokenisation market structure debate, September 2026.
• Coverage of the Crypto.com and OG.com minority stake transaction, September 2026.
• Third-party market data aggregators for consensus price targets, valuation multiples and balance-sheet aggregation, September 2026.

Market data

• Share price, market capitalisation, share count and valuation multiples as of the close on 18 September 2026 ($119.82).
• Consensus price targets and coverage counts as compiled by S&P Global Market Intelligence, 29 analysts, September 2026.
• Federal Reserve target range history and market-implied policy path, September 2026.

28 — DisclosureConflicts, limitations, and revision policy