HOOD FTX Just Collapsed - Why Does the DOJ Now Own 7.6% of the Company?

A Year That Ended With the DOJ Owning Part of the Company

The last four posts tracked Robinhood through a year of falling revenue, a workforce reduction, and - by Q3 - the first signs of cost discipline actually working. Full-year 2022 revenue fell 25% to $1.36 billion, down from $1.82 billion in 2021, and the GAAP net loss narrowed to $1.03 billion from 2021's $3.69 billion - a smaller loss, though 2021's number was inflated by one-time IPO accounting charges that don't recur, so the comparison flatters 2022 somewhat. Full-year Adjusted EBITDA^» came in at $(94) million, a step back from the $33 million positive result in 2021, even though Q3 2022 alone was Adjusted EBITDA-positive at $47 million - a reminder that one good quarter doesn't undo three difficult ones in the same year.

But the year's defining event, disclosed in this filing as a subsequent event, has nothing to do with any of that. FTX, one of the world's largest cryptocurrency exchanges, collapsed in November 2022, and its founder Sam Bankman-Fried was indicted on criminal charges related to his involvement. As of January 6, 2023, 55,273,469 shares of Robinhood common stock - roughly 7.6% of the company - originally acquired by Emergent Fidelity Technologies, Ltd. (a holding company majority-owned by Bankman-Fried) are now held by the U.S. Department of Justice, seized in connection with his indictment. On February 8, 2023, Robinhood's board authorized the company to pursue purchasing most or all of those shares back, though the filing states plainly it "cannot predict when, or if, the share purchase will take place." A company whose entire pitch is retail-investor trust just had its cap table entangled, involuntarily, with the highest-profile fraud case in crypto's short history - and it happened not because Robinhood did anything wrong, but because a passive investor it never chose to be associated with turned out to be running one.

The crypto-winter thread that's run through every post in this series since Q1 2022 closed the year with the cleanest possible number: crypto transaction revenue fell to $202 million for the full year, down 52% from $420 million in 2021 - itself already down from a mania peak. Net Cumulative Funded Accounts grew just 1% for the year, to 23.0 million, essentially the flattest growth Robinhood has ever reported, while full-year MAU fell to 11.4 million from 17.3 million (-34%) and ARPU collapsed to $60 from $103 (-42%) - a business that, by every operational measure, spent 2022 giving back the mania-driven expansion of 2021 rather than building on it.

The Prescription

Robinhood should keep doing exactly what Q3 2022 proved works: hold operating expenses flat or falling while the revenue base stabilizes, rather than assuming growth will return on its own. Full-year opex fell 31% to $2.37 billion, the first real evidence the April 2022 reduction in force and the broader cost discipline flagged across this series' 2022 posts wasn't a one-quarter blip - it held for the back half of the year. The company should treat 2022 as proof that a leaner Robinhood can get close to breakeven at a $1.3-1.4 billion revenue run rate, and build its 2023 plan around defending that breakeven line rather than betting on a crypto or meme-stock reacceleration it has no control over.

What it should stop doing: treating operational risk controls as an afterthought relative to product launches. The $57 million Q4 2022 Processing Error (see Beyond the Usual below) - a single botched reverse-stock-split trade that cost more than the entire quarter's worth of Adjusted EBITDA gains Q3 had just delivered - is exactly the kind of self-inflicted, avoidable loss a company trying to prove cost discipline can least afford. A brokerage handling corporate actions (splits, mergers, spinoffs) correctly isn't optional infrastructure; it's the baseline of the business Robinhood is in, and this year showed a real gap in it.

Key Financial Metrics

Year ended December 31, 2022 vs. year ended December 31, 2021 - consolidated, reported in USD.

Metric	FY2022	FY2021	YoY
Total net revenues	$1,358M	$1,815M	❌ -25%
Adjusted EBITDA»	$(94)M	$33M	❌ swung to a loss
Operating income (loss) (GAAP)	$(1,011)M	$(1,641)M	✅ loss narrowed
Net income (loss)	$(1,028)M	$(3,687)M	✅ loss narrowed sharply
Cash and cash equivalents (period-end)	$6,339M	$6,253M	✅ +1%

The full-year net-loss improvement is real but needs the same caveat as prior comparisons in this series: 2021's loss included $2.05 billion of one-time convertible-note fair-value charges and $1.57 billion of IPO-triggered share-based compensation, neither of which recurred in 2022 ($654 million of share-based compensation this year, down sharply). On the more comparable Adjusted EBITDA basis, 2022 was actually worse than 2021 - $(94) million versus $33 million positive - which is the more honest single number for how the underlying business performed across the full crypto-winter year, Q3's standalone positive quarter notwithstanding. Cash grew slightly to $6.34 billion despite the ongoing operating losses, helped by the cost discipline that took hold in the second half of the year. Total assets ended the year at $23.34 billion and total stockholders' equity fell modestly to $6.96 billion from $7.29 billion, a much smaller equity erosion than 2022's headline net loss might suggest, reflecting how much of the reported loss trend improved through the back half of the year.

Key Operational Metrics

Year ended December 31, 2022 vs. year ended December 31, 2021.

Metric	FY2022	FY2021	YoY
Net Cumulative Funded Accounts	23.0M	22.7M	⚠️ +1%, essentially flat
Monthly Active Users (MAU, Dec)	11.4M	17.3M	❌ -34%
Assets Under Custody (AUC)	$62.2B	$98.0B	❌ -37%
Average Revenue Per User (ARPU)	$60	$103	❌ -42%

Net Deposits - the cleanest read on whether customers are actually adding new money, independent of market-value swings - fell to $18.4 billion for the year from $27.1 billion in 2021, a real deceleration in fresh customer money on top of the market-value damage from a down year in both equities and crypto. Robinhood reports as a single operating segment, so there's no segment breakdown to run here. The seasonality pattern flagged in every prior post in this series (stronger new-account formation typically expected in Q1) is worth keeping in mind heading into whatever Q1 2023 brings, especially after a year this flat on net account growth.

Beyond the Usual

The DOJ now holds a 7.6% stake in Robinhood, seized from a Sam Bankman-Fried entity after FTX's collapse

Emergent Fidelity Technologies, Ltd., a holding company majority-owned by FTX founder Sam Bankman-Fried, had built a roughly 7.6% stake in Robinhood (55,273,469 shares) before FTX's November 2022 collapse and Bankman-Fried's subsequent criminal indictment. As of January 6, 2023, those shares are held by the DOJ, seized in connection with the indictment. This is disclosed as a subsequent event in this filing, and Robinhood's own risk factors state plainly that "any substantial sale of the Emergent Shares or perception that such a sale might occur could cause the trading price of our Class A common stock to decline substantially" - a genuine overhang risk the company has no control over and didn't create, but now has to manage. On February 8, 2023, Robinhood's board authorized pursuing a purchase of most or all of the Emergent Shares, though the outcome and timing remain unresolved as of this filing.

A botched reverse-split trade cost Robinhood $57 million in a single day

On December 16, 2022, delays in notification from third parties and process failures within Robinhood's own brokerage systems, in connection with handling a 1-for-25 reverse stock split by NASDAQ-listed Cosmos Health, Inc., allowed customers to execute trades selling more shares than they actually held. This created a temporary short position in that ticker, which Robinhood covered out of corporate cash the same trading day, at a $57 million loss - recorded within brokerage and transaction expense and explicitly called out as its own line item in the company's Adjusted EBITDA reconciliation (the "Q4 2022 Processing Error"). Robinhood's own filing lists this, alongside the Early 2021 Trading Restrictions and the November 2021 data breach, as one of the events that "faced challenges" for the brand - a rare instance of the company itself grouping a self-inflicted operational failure with its two biggest prior reputational crises.

Crypto custody accounting confirms Robinhood is more conservative than the platforms that failed around it in 2022

The SAB 121 crypto-custody accounting adopted in Q2 2022 (see that post) closed the year with the asset/liability pair at $8.43 billion, down from $8.59 billion at Q2's introduction, tracking the broader crypto market's continued decline through the year rather than any change in Robinhood's custody model. In January 2023, the Bankruptcy Court for the Southern District of New York issued a ruling in a separate crypto-platform bankruptcy addressing how safeguarded crypto-assets are treated in insolvency - a live legal question Robinhood's own filing flags as relevant to how its own safeguarding obligation might be viewed, even though Robinhood itself, unlike FTX, has consistently stated it doesn't lend out, rehypothecate, or otherwise use customer crypto for its own purposes.

Founders now hold under a majority of shares, but still control the company

As of December 31, 2022, Robinhood's founders and their related entities hold approximately 15% of outstanding common stock, but - via Class B super-voting shares - still control over 50% of the total voting power. This is a modest decline in economic ownership from prior years as the founders' RSU-driven share issuances and any personal sales dilute the raw share count, without any change to the governance structure that keeps founder control intact regardless of that dilution.

Target Valuation Range

~$6-$11 bear-to-bull range against an $8.14 actual close - the stock ended the year almost exactly where it started Q3, but with a materially different risk profile now that a DOJ-controlled 7.6% stake sits in the register. The FTX-adjacent overhang is a real, company-specific risk this framework hadn't priced in a quarter ago, and it caps the bull case more than the operating numbers alone would.

HOOD rallied into October ($11.68) before giving most of it back - $9.59 in November, $8.14 at the December 30, 2022 close, essentially flat with Q3's $10.10. The stock closed the year down 79% from its $38.00 IPO price and down 54% from where it started calendar 2022. There's been no stock split, so this is the actual nominal price quoted at the time.

Market cap → enterprise value	FY2022 (period-end)
Share price (period-end, Dec 30, 2022 close)	$8.14
Shares outstanding (Class A + B)	892.8 million
Market capitalization	~$7.3 billion
Less: cash and cash equivalents	$6.34 billion
Interest-bearing debt	none
Enterprise value	~$0.9 billion

Peer-multiple sanity check	FY2022	Q3 2022 (prior post)
Revenue basis	full-year actual ($1.36B)	quarterly-annualized (~$1.44B)
Enterprise value	~$0.9 billion	~$2.7 billion
EV/Revenue	~0.7x	~1.9x

The multiple compressed again despite Q4's stock price barely moving, because shares outstanding kept growing (share-based compensation dilution) and cash kept growing faster than the enterprise value math credits it for - at this point enterprise value is small enough relative to cash on hand that small changes in either number swing the multiple sharply. This is now the lowest EV/Revenue multiple in the entire series, lower even than Q2 2022's crypto-crash trough. Neither P/E nor EV/EBITDA is meaningful given the continued GAAP net loss and negative full-year Adjusted EBITDA.

A simplified reverse DCF: at a ~10% discount rate, 3% terminal growth, and a 20% mature FCF margin, sustaining today's ~$0.9 billion enterprise value requires roughly $63 million of steady-state annual free cash flow, implying roughly $315 million in steady-state annual revenue - less than a quarter of FY2022's actual $1.36 billion. At this price, the market isn't really pricing Robinhood as an operating business at all; it's pricing it close to a cash shell with a call option on the business recovering, which is consistent with an enterprise value this small relative to the cash sitting on the balance sheet.

Because cash ($6.34 billion) is now large relative to the enterprise value itself, price is dominated by the cash-plus-EV buildup (market cap = EV + cash, on 892.8 million shares) rather than by the multiple alone - a bear case here has to assume continued cash burn shrinks the cushion, not just that the operating multiple compresses further from an already-near-zero base:

Scenario	Key assumption	Implied EV	Implied cash	Implied market cap	Implied price
Current (Dec 30, 2022 close)	actual market price, for reference	~$0.9 billion	$6.34 billion	~$7.2 billion	$8.14
Bear	The Emergent Shares overhang resolves badly, revenue keeps sliding through 2023, and continued cash burn shrinks the balance-sheet cushion by roughly $1.3 billion over the year	~$0 billion (market ascribes no value to operations)	~$5.0 billion	~$5.0 billion	~$6
Base	Q3's cost discipline holds through 2023, revenue and cash both stabilize near FY2022 levels, the Emergent Shares situation resolves without a disorderly sale	~$0.9 billion (unchanged)	$6.34 billion	~$7.2 billion	~$8
Bull	The DOJ/Emergent Shares situation resolves cleanly, crypto markets stabilize, and cost discipline plus a net-interest-revenue tailwind from higher rates pushes Adjusted EBITDA solidly positive, letting cash keep building while the market also assigns real value to the operating business	~$3.4 billion (~2.5x FY2022 revenue)	$6.34 billion	~$9.7 billion	~$11

Two full years into being a public company, Robinhood's stock is being priced less on its own operating trajectory than on how a legal situation it didn't create gets resolved. The FY2021 post closed by noting the growth story had already peaked; this post closes on a business that's found something closer to an operating floor (Q3's Adjusted EBITDA turn, a full-year cost base down 31%) but whose near-term stock-price range is now bounded more by the Emergent Shares overhang than by anything in its own P&L.

---