Coinbase's 'Hold With 40% Upside' Is a Bet on the Cycle, Not the Business

Morgan Stanley just initiated Coinbase at aHold-equivalent rating with a $250 price target, saying the stock should be worthroughly 40% above where it trades. That is not the analyst being confused. It is the most honest statement of where Coinbase sits right now: plugged into a platform that keeps compounding a recurring fee base, priced as though the crypto cycle is dead forever.

Start with the tension. Analyst Michael Cyprys put a$250 price target on the stock at about $172, down roughly 57% from its 52-week high and about 24% for the year. The catch is the rating itself. Equal Weight, the bank's neutral bucket, is the only designation that lets you print a price target 40% above the market while declining to call the stock a buy. Cyprys said the reason out loud: the outcome range he models for Coinbase runs from$50 on the low end to $400 on the high. When your downside case is $50 and your fair value is $250, the honest label for the middle of that spread is not conviction; it is "wait and see."

The wait is the story. That $250 is not built on anything Coinbase earns this quarter. It is built on21 times an estimated $3.0 billion of adjusted EBITDA in 2028, a year that sits on the far side of a forecast cyclical reset in 2026, a sharp rebound in 2027, and normalization after that. In other words, the price target is a bet that crypto trading comes back, not a valuation of what the platform does while it is quiet.

What it does while it is quiet is the fact worth inspecting. The second quarter was a bad one by nearly any reading: anet loss of $359.5 million, or $1.36 a share, against a profit of $5.14 a year earlier, on revenue that fell to$1.2 billion from $1.5 billion. It was the third straight quarter that missed Wall Street budget. Transaction revenue, the fee income tied to trading, came in at $599 million. This is the surface the bears have been reading all year: a cyclically dependent exchange in a crypto trough, losing money.

But split that revenue and the surface cracks. Subscription and services revenue reached$555 million in the quarter — 48% of net revenue, up from29% back in the fourth quarter of 2024. That is the stablecoin interest, custody, staking, and platform fees that keep accruing whether or not retail is trading. Cyprys' own framing is"beyond the exchange, but not beyond the cycle," with retail trading stillabout 40% of revenue. That is exactly the right distinction, and it is the one the market keeps dropping. Even in a quarter where trading collapsed, roughly half of everything Coinbase collected did not depend on a single extra trade, and it still booked about$208 million of adjusted EBITDA and a net-cash balance sheet with roughly $8.6 billion of corporate cash on hand.

That floor is the whole point. The bear narrative treats Coinbase as a fee-taker whose business evaporates when crypto gets quiet. The numbers say the fee-taker is now roughly half the business, the quiet quarter still produced positive adjusted EBITDA and free cash flow, and the company sits in a net cash position with no survival pressure. Grocery-store downturns are a different animal from a terminal decline. This still looks like the first.

Which makes the cheap-looking multiple the trap it usually is. Coinbase shows a forward price-to-earnings ratio around 15, which for a stock down this much reads like a value signal. It is not. That forward earnings number bakes in the whole 2027 rebound Morgan Stanley is forecasting. Anchor it to what is actually being produced instead: the stock trades at well over 7 times trailing sales and roughly 26 times trailing free cash flow. The stock is not cheap on what it earns in a trough. It is cheap only if you buy the cycle-recovery thesis, and paying a future multiple for a future recovery is a bet, not a bargain.

That is where the judgment lands. The market has been pricing Coinbase as though crypto winter is structural and the platform is decaying. The math says otherwise: the recurring revenue mix has roughly doubled as a share of the business in two years, the balance sheet is net cash, and even the worst quarter in recent memory produced positive adjusted EBITDA and cash flow. The floor is real and it is rising.

What does not follow is the other extreme — that the 40% upside is yours for the taking on a hop. The re-rating to that $250 depends on trading volume coming back, which Morgan Stanley itself schedules for 2027 and prices with a range wide enough to fit a 75% drawdown. The setup is a watchlist upgrade, not a stampede: a cash-backed platform with a rising subscription floor, waiting on one defined break condition — the cycle turning — before the forward math can be believed instead of hoped for. Until then, the only number that matters is the same one the analyst printed with a straight face: a Hold with 40% upside is Wall Street admitting the market is too far down the pessimism tail, not that the storm is over.