Key Takeaways
- Bernstein maintains Outperform rating with $641 price target on Microsoft (MSFT) stock
- MSFT shares have declined 27.5% in the last six months, hovering near 52-week lows at $370.87
- Firm argues Azure margin headwinds are transitory, resulting from timing mismatch between AI infrastructure investment and revenue generation
- Acceleration in Azure growth anticipated during Q3 with continued strength through Q4
- Strong Buy consensus from Wall Street analysts with average target of $581.61
Microsoft shares have faced significant pressure over the last half-year, declining 27.5% to reach $370.87. The stock now trades close to its 52-week bottom. Despite this weakness, Bernstein remains firmly optimistic about the tech giant’s prospects.
Bernstein analyst Mark Moerdler has maintained his Outperform rating alongside a $641 price objective for MSFT shares — representing upside potential exceeding 70% from current trading levels.
Bernstein’s investment thesis centers on a timing mismatch. Microsoft has committed substantial capital to AI infrastructure buildout, which has created concern among some market participants. However, Moerdler contends this spending pattern doesn’t represent the risk many investors perceive.
The research firm’s analysis suggests the majority of capital expenditures translate into revenue-generating capacity within a six-month timeframe. This delay between investment and monetization is temporarily distorting financial metrics, creating an unfavorable near-term picture.
Bernstein conducted a detailed examination of five potential destinations for Microsoft’s capital spending: proprietary applications, complimentary Copilot access, internal infrastructure, lower-margin Azure AI services, and offline capacity awaiting deployment. Their assessment revealed a more advantageous allocation than current market pricing reflects.
A substantial portion of investments flows toward higher-margin opportunities, especially Microsoft’s proprietary software platforms and artificial intelligence solutions. Copilot specifically generates software-as-a-service style AI revenue with attractive margins after transitioning to paid subscription models.
Azure Margins Under Pressure — But Not Forever
Azure profitability metrics have experienced compression, which Bernstein recognizes. According to the firm, this stems from nascent AI workloads producing thinner margins compared to conventional cloud computing services.
As these AI workloads evolve and achieve scale, Bernstein anticipates margin improvement. The current pressure reflects Azure’s position within its AI expansion phase rather than indicating a fundamental flaw.
Research and development expenditure as a proportion of revenue has remained essentially stable. Bernstein points to this metric as evidence that Microsoft maintains financial discipline despite aggressive investment.
Microsoft delivered 16.7% revenue expansion over the trailing twelve months. The stock currently trades at a P/E multiple of 23.26, accompanied by a PEG ratio of 0.8 — valuation levels that both Bernstein and InvestingPro characterize as attractive relative to growth prospects.
Azure Growth Expected to Pick Up in Second Half
Bernstein projects Azure growth acceleration during Q3, with positive momentum extending into Q4. This forecast directly correlates with previously funded capacity installations coming operational.
Microsoft has simultaneously launched a strategic initiative — creating proprietary large-scale AI models targeting 2027 completion as alternatives to solutions from OpenAI and Anthropic.
UBS recently confirmed its Buy rating on Chevron following an electricity generation partnership with Microsoft. The collaboration involves constructing natural gas power facilities in Texas dedicated to supplying Microsoft’s AI datacenter operations.
Across the broader analyst community, 34 of 37 Wall Street analysts covering MSFT over the past three months assigned Buy ratings. The consensus price target stands at $581.61, suggesting 56% appreciation potential from present levels.
