Microsoft Is Cutting 4,800 Jobs to Feed the AI Machine. Should You Still Own It?

Microsoft is cutting ~4,800 jobs (~2.1% of workforce) while guiding to ~ $190B of capex in calendar 2026 to fuel AI/Azure, with investors focused on whether the spend converts to monetization. On the latest call, Microsoft said AI ARR surpassed $37B (+123% YoY) and remaining performance obligations reached $627B (+99% YoY), while Azure grew ~40% in constant currency last quarter. Despite capex concerns (FCF yield ~2.47% and Q3 capex +84% YoY to $30.88B), the piece argues valuation looks reasonable (forward P/E ~20x) and upside hinges on Azure sustaining ~39%-40% growth and AI margins holding.

Analysis

The market is debating whether MSFT is a software compounder or an AI utility with software economics, and that distinction matters more than the headline spend. If Azure keeps scaling while the AI backlog converts, MSFT can still earn a premium multiple; if not, the business starts to look like a capital-intensive infrastructure owner whose FCF yield is too low for a 20x forward P/E. The second-order winner is not just MSFT but the picks-and-shovels layer around it—NVDA, AVGO, and ANET benefit if hyperscaler spend remains lumpy because they monetize each incremental rack even before MSFT fully proves ROI.

The real risk is time mismatch, not demand destruction. Low GPU utilization implies the bottleneck may be networking, power, and integration rather than raw compute, which means incremental dollars can keep flowing into capex without immediate margin relief. That is bearish for MSFT’s multiple over the next 1-3 months if the late-July Azure readout is merely “fine” and cloud margin slips again; it becomes a structural problem over 6-18 months if capex stays elevated while buybacks and FCF growth lag.

Contrarian view: consensus may be over-fixated on the layoff/capex tension and underweight the contracted revenue visibility that reduces downside to demand. But the stock is not obviously cheap if the capex bill keeps expanding faster than monetization, so the cleanest expression is relative strength, not blind ownership. The thesis breaks if Azure falls below the low-30s growth zone or Microsoft Cloud margin deteriorates materially again, because then the market will stop treating this as temporary investment and start discounting a lower-quality growth profile.

Key Decisions for Investors

• Long MSFT vs SPY for 1-3 months, looking for relative outperformance if the late-July Azure print confirms 39%-40% growth; cut the trade if Azure growth drops below ~35% or cloud gross margin ticks down again.
• Do not chase MSFT pre-earnings into the current range; wait for either a post-report pullback or confirmation that capex is converging toward management guide rather than overshooting it.
• Favor a basket long of AI infrastructure suppliers (NVDA/AVGO/ANET) over MSFT on any sign that hyperscaler utilization remains weak, because suppliers monetize spend faster than the platform owner.
• If MSFT rallies into earnings without a better FCF narrative, use it to harvest upside via a tactical covered-call overlay rather than add outright delta.
• Set a 6-18 month alert on MSFT free cash flow yield staying below ~3% while capex rises; that would justify rotating from MSFT into higher-ROIC software names.