##JPMorganRaisesMeta$820

The $820 Question: Why JPMorgan Finally Changed Its Mind on Meta

There is a particular kind of silence that falls over a trading desk when a major analyst reverses a long-held position. It is not the silence of indifference. It is the silence of recalculation, of portfolios being reweighted, of assumptions being quietly revised. That silence settled over Meta Platforms on September 10, when JPMorgan analyst Doug Anmuth moved the stock from Neutral to Overweight and lifted his price target to $820 from $640. The new target implies roughly 25 percent upside from the prior close, and it carries a weight that goes beyond the headline number. JPMorgan was not a cheerleader on this stock. It was a skeptic. And that makes the upgrade worth understanding.

The immediate catalyst is Muse, Meta's new consumer-facing AI agent, which launched just days before the note was published. On its second day of availability, Muse reached the number three position in the United States App Store, with early usage running at roughly ten times that of internal training cohorts. Anmuth's argument is not that Muse will generate meaningful revenue in the near term. He explicitly says monetization is not the near-term priority. His argument is that Muse represents the beginning of a product cycle that the market has not priced in, and that the infrastructure behind it has reached a level of sophistication that changes Meta's competitive position.

That infrastructure is the more consequential story, and it is where the upgrade finds its intellectual foundation. For much of the past year, Wall Street debated whether Meta's enormous AI spending would ever produce returns beyond incremental improvements to advertising. The bear case was straightforward: Meta was pouring tens of billions of dollars into data centers and GPUs without a clear path to a product that could justify the expense. The bull case was less specific, resting mostly on the belief that AI would make Meta's existing advertising machine more efficient. What Anmuth is now arguing is something different. He points to Meta's Superintelligence Lab, which he says has essentially delivered on its goal of reaching the AI frontier within a year, culminating in Muse Spark 1.3, which he describes as competitive with the cutting-edge models from OpenAI and Anthropic. He also points to an upcoming model internally dubbed Watermelon, which he expects to unlock further opportunities across consumer products, engagement, advertising, and internal efficiency.

The distinction matters. A company that improves its advertising targeting with AI is a better advertising company. A company that builds frontier models competitive with the leading laboratories is something else entirely. It is a company that can build new products, not just optimize existing ones. That is the shift JPMorgan is responding to, and it explains why the price target moved by nearly 30 percent in a single note.

Yet the upgrade arrives against a financial picture that is genuinely mixed. Meta's second-quarter results, reported in late July, showed revenue of $60.8 billion, a 28 percent increase year over year, driven by a 27 percent rise in advertising revenue to $59.4 billion. Ad impressions rose 14 percent and average price per ad rose 12 percent, a healthy combination of volume and pricing power. Family daily active people averaged 3.6 billion in June, with Instagram reaching 2 billion daily active users and Threads surpassing 500 million monthly users. These are the numbers of a company whose core business is performing well.

The profit picture is less comfortable. Earnings per share came in at $6.18, down 13.4 percent year over year and well below the consensus estimate of $7.22. Total costs and expenses surged 55 percent to $42 billion, a pace nearly double the revenue growth rate. The increase included $2.4 billion in legal charges and $1.18 billion in severance expenses tied to a May headcount reduction. Excluding those one-time items, operating income would have risen 9 percent, but the reported operating margin contracted to 31 percent from 43 percent a year earlier. The gap between the headline and the adjusted figure is wide, and it reflects the reality that Meta is spending heavily on infrastructure, compensation, and depreciation even as its revenue grows.

That spending is the central tension in the Meta story, and JPMorgan's own projections make it explicit. Anmuth now expects capital expenditures of $243 billion in 2027 and $284 billion in 2028, both above Wall Street consensus. He projects free cash flow pressure in the range of negative $65 billion to $70 billion in each of those years, a striking contrast with Meta's current cash generation of roughly $41 billion in levered free cash flow over the past twelve months. Meta has been meeting with European bond investors in recent weeks, exploring what would be its first debt issuance outside the dollar market, a sign that the company is preparing to finance this buildout with borrowed capital rather than relying solely on operating cash flow.

This is not a minor detail. It means Meta is entering a phase where it will be spending more than it generates, and it will be doing so in a higher interest rate environment. The Federal Reserve is expected to raise rates this week, and the cost of capital is rising across the board. A company that has historically been a cash-generating machine is now asking investors to fund a vision whose returns are years away. That is a different kind of bet than the one Meta shareholders have been asked to make in the past.

The market's reaction suggests that investors are, for now, willing to make it. Meta shares rose more than 5 percent on the day Muse launched and gained further ground after the JPMorgan upgrade. The stock is up roughly 20 percent from its August low, though it remains flat for the year, compared with a 12 percent gain for the S&P 500. That gap is central to JPMorgan's argument: the stock has room to catch up. After the upgrade, more than 90 percent of analysts tracked by Bloomberg rate Meta a buy, with the rest at hold and none at sell. The average analyst price target sits near $750, well below JPMorgan's $820 but still implying meaningful upside.

What should a careful observer watch in the weeks ahead? First, the early engagement data for Muse. The app store ranking is a useful signal, but retention and daily active usage will tell a more complete story. JPMorgan's thesis depends on Muse becoming a durable product, not a novelty. Second, the trajectory of capital expenditure guidance. Meta raised the lower end of its 2026 range to $130 billion from $125 billion, and any further increase would amplify the free cash flow concerns that Anmuth himself acknowledges. Third, the broader market environment. The Fed's decision on September 16 will set the tone for risk appetite in the weeks that follow, and a more hawkish outcome would make financing costs higher for every company in Meta's position.

The deeper truth is that Meta is being asked to prove something that few companies of its size have ever proven: that a mature, highly profitable advertising business can fund the construction of an entirely new platform without destroying the returns that made it valuable in the first place. The advertising business is still growing at a pace that would be the envy of almost any company. The AI investments are real, the models are competitive, and the distribution advantage of reaching 3.6 billion daily users is unlike anything else in the technology industry. But the path from here to $820 runs through years of negative free cash flow and uncertain product adoption. JPMorgan has decided that the journey is worth taking. The rest of the market will decide in its own time.

DYOR 🔎 NFA ✔️
$META