Microsoft Corporation (MSFT)

The panel's take

Verdict: WAIT · Conviction: MEDIUM · Last price: USD 371.30 (as of 2026-06-29)

This is the panel's clearest "best value, worst tape" standoff — and, unusually, no one argues the business is broken. Microsoft fell ~33% from its ~$555 peak, but Azure reaccelerated to 40% growth, AI annual revenue hit a ~$37B run rate (up ~123%), ROE is ~34%, and the moat — Windows, Office, Azure switching costs plus Copilot upsell — is widening, not narrowing. The decline is almost entirely multiple compression (from ~35x to ~20x), not deterioration. The value and quality frameworks see that as opportunity: fair-value estimates run from $420 to $565, well above the price, and the GARP lens calls a sub-1.0 PEG "the market giving it away." The caution is about the chart and the cash: the stock sits ~18% below its long-term trend in a confirmed downtrend, and a sharply higher AI capex cycle cut free cash flow ~22% year-over-year, with ~45% of the contracted backlog tied to a single customer (OpenAI). The bulls say buy the de-rated quality now; the cautious say world-class business, wrong tape — wait for it to base. That split synthesizes to WAIT.

Key levels

Key resistance starts at ~$392 (the 50-week trend), then the ~$432 breakdown zone, up to the ~$477 level where the long-term trend signal flipped negative. Key support sits at ~$357 (the 2024 base and ~61.8% retracement), then ~$322, with a deeper historical accumulation zone near $287 (the 200-week trend). The analyst consensus target spans roughly USD 460 to 650, averaging near USD 565 — every Street target sits well above the current price, an unusually wide gap that reflects how much the analyst community reads the selloff as multiple compression rather than a broken thesis.

What legendary investors think

We ran Microsoft past a panel of six legendary investors' frameworks.

Where they agree — and where they clash

Common ground

• The fundamentals are strong and improving: Azure reaccelerated to 40%, AI revenue is at a ~$37B run rate (+123%), operating margin ~45%, ROE ~34% — among the best numbers Microsoft has reported.
• The 33% drop is a re-rating, not a breakdown: every framework, including the bears, attributes it to AI-capex multiple compression rather than business deterioration.
• The capex cycle is the real cost: a sharply higher AI build cut free cash flow ~22% year-over-year, and the depreciation ramp hasn't fully hit the income statement.
• OpenAI is both anchor and risk: ~45% of the contracted backlog, a customer now diversifying its cloud providers.

The real debate

• Cheap or just less expensive? The value and quality lenses see five-year-low multiples on a 40%-Azure compounder as a gift; the Forensic Skeptic sees a fairly-priced great business with no margin of safety until the $340s.
• Buy the de-rating or wait for the tape? The fundamental case says the price already discounts a prolonged FCF depression the data doesn't support; the macro and technical lenses say a confirmed downtrend below the 200-week trend is not to be caught.
• Does the capex pay back? Bulls frame ~$80B+ of capex as essential platform-shift investment; the skeptics note FCF stays suppressed for years before the AI revenue fully scales.

The question it comes down to: Is Microsoft a wide-moat compounder that the AI-capex scare has handed back at five-year-low multiples — or a still-richly-capitalized business whose free cash flow stays gutted long enough, and whose OpenAI dependence runs deep enough, that the broken tape is telling the truth?

The numbers

Metric	Value
Price / Market cap	USD 371.30 / ~2.77T
P/E (TTM / fwd)	~22x / ~19x (vs ~30–35x at the peak)
ROE	~34%
Operating margin	~45%
Dividend yield	~0.9%
Debt / equity	~0.30x (net cash)
Free cash flow	~USD 15.8B (Q3 FY2026, down ~22% YoY on the AI capex ramp)

Figures as of Q3 FY2026 (quarter ended March 2026) / June 2026; sourced from Microsoft IR/SEC filings, StockAnalysis, GuruFocus, MacroTrends, MarketBeat. Full-year AI capex figures vary by source (quarterly capex in the low-$30Bs; full-year estimates cited from ~$80B toward higher multi-year commitments); the consistent fact is a sharply higher build that cut free cash flow ~22% YoY. Forward multiples reflect consensus EPS; a normalized-earnings view (accounting for the depreciation ramp) yields a higher effective multiple.

The bottom line

Microsoft is the rare large-cap where almost every framework likes the business and several like the price — and the verdict still lands on patience, because the chart and the cash flow disagree with the value case for now. The bull argument is grounded and, on the latest quarter, accelerating: Azure reaccelerated to 40%, AI revenue compounds past a $37B run rate, the moat is widening through Copilot upsell across 300M+ commercial seats, ROE is ~34%, and the 33% drawdown compressed the multiple from the mid-30s to roughly 20x — handing the quality and value lenses fair-value estimates from $420 to $565, all above the price. The caution is equally concrete and, this quarter, dominant in the tape: the stock sits ~18% below its long-term trend in its worst month in a generation, a sharply higher AI capex cycle cut free cash flow ~22% with the depreciation ramp still building, and ~45% of the backlog leans on an OpenAI relationship that is diversifying away. What would tip the synthesis decisively bullish is the tape repairing — a reclaim of the long-term trend on volume — alongside free cash flow stabilizing as the capex cycle peaks, which would confirm the value frameworks are right that this is cheap rather than merely cheaper. What keeps it a wait is that, until the downtrend bases and the cash flow turns, the panel is being asked to catch a world-class business mid-fall — and the bulls who'd buy it here are outweighed, for now, by the voices saying the same company will be available with the tape on their side.