Alibaba Is Down 33% This Year. Here’s What It’s Actually Worth.

A trillion-dollar internet empire, a Pentagon blacklist, a billionaire short-seller who can’t decide if he’s in or out — and a valuation model that says something more interesting than any headline.

If you’ve glanced at a stock ticker this year, you’ve probably seen Alibaba (BABA) do something dramatic. The stock is down roughly a third year-to-date, it touched a 16-month low in late June after Anthropic publicly accused the company of running a large-scale campaign to extract capabilities from its Claude models, and it’s currently caught in the crossfire of a U.S.-China standoff that has nothing to do with how many packages Taobao ships. In late June, Alibaba landed on the Pentagon’s list of companies allegedly tied to the Chinese military — and then, days later, won a temporary court reprieve that let it keep lobbying in Washington while it fights the designation.

Meanwhile, the smart money can’t agree on what to do. Cathie Wood’s ARK funds bought Alibaba shares in early June, then sold a chunk of that position three weeks later. Michael Burry — yes, the Big Short guy — added to his stake in mid-June calling Alibaba “the most advanced AI company in China,” then reportedly sold the whole thing two weeks after that, saying the move was mostly about tax-loss harvesting and that he might buy back in later. One senior Alibaba insider also executed a large multi-million-dollar stock sale in early July. None of this is the behavior of a market that has made up its mind.

So here’s the question worth actually answering: underneath all that noise, what is Alibaba worth? Not what the headlines say, not what the chart says — what does the business itself, stripped of accounting sugar and stock-market drama, justify paying?

I built a full valuation model on Alibaba a few weeks ago, running six different valuation methodologies in parallel and weighting them by how well each one fits a company like Alibaba. The stock has moved a lot since then, so let’s update it for where things stand today, and walk through what actually matters.

The Profit Number You’re Reading Is Partly Fake

Start here, because it changes everything downstream. Alibaba reported GAAP net income of about $15.4 billion for fiscal year 2026. Sounds solid. Except once you strip out the one-off stuff — mark-to-market gains on equity investments, gains and losses on asset sales, goodwill impairments, none of which represent actual cash flowing into the business — the real, recurring profit was closer to $8.8 billion. That’s a 43% haircut, and it means the “real” earnings of the business actually fell sharply year-over-year, even while the headline number looked fine.

Think of it like reporting €100,000 of profit for your small business, except €60,000 of that came from selling an old van you had sitting in the garage. Your actual business earned €40,000 — less than last year — and the van sale is masking that.

This matters enormously for the price-to-earnings ratio everyone quotes. On GAAP earnings, Alibaba trades around 19x. On the real, normalized number, it’s closer to 31x — nearly double the sector’s fair multiple once you adjust for how efficiently Alibaba is actually using its capital.

And the cash story is worse than the earnings story. Free cash flow went negative in fiscal 2026, to the tune of roughly -$6.8 billion — a $12–13 billion swing from the positive cash flow the year before. Buybacks and dividends right now are being funded out of Alibaba’s cash pile, not out of the business generating money. The pile is large (net cash in the $38–59 billion range), so there’s no emergency here — but it does mean the popular “attractive free cash flow yield” argument for owning the stock simply doesn’t hold up on today’s numbers.

Growing Fast Isn’t the Same as Creating Value

Here’s a distinction that gets lost in most retail investing conversations: growth and value creation aren’t the same thing.

Alibaba’s return on invested capital (ROIC) currently sits around 4.5%. Its cost of capital (WACC) — the minimum return it needs to earn on every dollar invested just to break even economically — is around 9%. When ROIC is below WACC, the company is, in a very real sense, destroying value with its current investment cycle. It’s spending more to build things (cloud infrastructure, AI data centers, the Quick Commerce delivery business) than those investments are currently returning.

This is the uncomfortable arithmetic behind Alibaba’s ambitious AI buildout — reportedly north of $56 billion in AI infrastructure spending planned over the next three years. It could absolutely pay off. Cloud revenue is growing north of 30% year-over-year and AI-related revenue is scaling even faster. But right now, in this specific window, the company is spending ahead of the returns showing up.

So What’s the Number? Putting a Price on the Business

This is where it gets useful. Rather than lean on any single valuation method — which is how most retail analysis goes wrong — I ran six distinct models in parallel: a two-stage earnings growth model, a free-cash-flow perpetuity model, a dividend discount model, the Lynch-Graham growth-adjusted formula, a three-phase corporate lifecycle DCF, and an asset-based “what’s it worth if everything stopped tomorrow” floor valuation. Each was weighted according to how well it actually fits a company with Alibaba’s profile — high growth, low current payout, heavy reinvestment.

Individually, the six models spit out wildly different numbers, from the mid-$20s (pure liquidation value) to over $200 (the optimistic cloud-driven DCF). That spread is itself a signal: genuine uncertainty about Alibaba is high, and anyone offering you a single confident price target is skipping a step.

Blended and weighted together, the six models land on a fair value of roughly $142 per share.

Three Roads From Here

No single fair value tells the whole story, so it’s worth sketching out how the next few years could actually unfold, and how likely each path is:

The bull case (~30% likelihood): Cloud AI monetization accelerates the way AWS did for Amazon a decade ago, Quick Commerce stops bleeding cash and reaches breakeven, and buybacks ramp back up as free cash flow normalizes. Under this path, fair value climbs toward $210–250 per share.

The base case (~45% likelihood): Revenue grows around 10% annually, margins gradually recover into the low-to-mid teens by 2028–29, and the stock re-rates modestly. This path roughly matches the $142 blended fair value above.

The bear case (~25% likelihood): Quick Commerce keeps consuming capital without a payoff, Douyin and PDD keep chipping away at e-commerce share, and margins stay structurally depressed. Fair value here falls to around $56 — which, notably, lines up almost exactly with the “minimum acceptable price” the model calculates independently using a completely different method (a discounted cash-flow floor requiring a 12% annual return). Two unrelated approaches landing in the same place is a meaningful confirmation signal, not a coincidence to wave away.

The Risk No Spreadsheet Can Price

Here’s the uncomfortable truth about owning a U.S.-listed Chinese ADR right now: the single biggest threat to this thesis isn’t competitive, it’s political. The Pentagon designation, the delisting debate that periodically resurfaces in Congress, and the broader U.S.-China tech rivalry are risks that no discounted cash flow model handles well, because they’re not driven by how well Alibaba runs its business. If Washington ever forced a wholesale delisting of Chinese ADRs, Western holders would be compelled to sell regardless of what the fundamentals said — a scenario the model estimates could shave 40–60% off fair value in a matter of weeks, with roughly a one-in-five probability over the next few years. It’s the risk that explains why Chinese tech, however well-run, tends to trade at a structural discount to comparable Western businesses.

Where Does That Leave Today’s Price?

This is where the story gets genuinely interesting. A month ago, when this model was first built, Alibaba traded around $121 — inside the fair-value range, but with an expected return (probability-weighting all three scenarios above) of only about 4% a year. That’s below what a long-term Treasury bond pays you for zero risk, which made the stock a “wait and watch” rather than a “buy.”

Since then, the stock has fallen into the mid-$90s — down roughly 20% in a matter of weeks, dragged lower by the Anthropic dispute, the Pentagon listing, and a broader sell-off across Chinese ADRs. Re-running the same three scenarios against today’s price rather than last month’s tells a different story: the probability-weighted expected return jumps from roughly 4% a year to somewhere in the 11–12% a year range — right around the threshold most disciplined value investors use as a minimum hurdle rate. The discount to blended fair value has widened from about 15% to over 30%.

That’s a meaningfully different setup than a month ago. It doesn’t erase the risks — the geopolitical overhang is arguably more live today than it was a month ago, not less, and the negative free cash flow trend still needs to reverse for the thesis to hold. But price and value have moved closer together in a way that rewards patience rather than urgency. A staged entry approach — buying a portion of a position now, with room to add more if the stock falls further toward the mid-$80s to low-$90s — fits the current numbers better than an all-or-nothing bet in either direction.

The Bottom Line

Alibaba isn’t a broken business. It has a fortress balance sheet, a credible and fast-growing AI cloud engine, and — after this year’s slide — a price that finally offers something closer to fair compensation for the risk involved. But it’s also a company currently earning less than its cost of capital, generating negative free cash flow for the first time in years, and sitting at the center of a geopolitical dispute that could move the stock 40% in either direction based on a single announcement out of Washington or Beijing.

None of that makes it a “buy” or a “sell” — those calls depend entirely on your own time horizon and how well you sleep through a stock that can lose half its value on a headline that has nothing to do with earnings. What the numbers do say is that the risk-reward has shifted meaningfully more favorable over the last month, even as the underlying uncertainty hasn’t gone anywhere. For a story this binary, position size probably matters more than conviction.

This article is for informational and educational purposes only and does not constitute investment advice. Do your own research, or consult a licensed financial advisor, before making investment decisions.