Microsoft: Short-Term Pain, Long-Term Opportunity

Key Points

Instead of chasing immediate revenue, the company is allocating scarce computing resources across three strategic areas: 1) High-margin proprietary applications, such as GitHub Copilot and Microsoft 365 Copilot. 2) Internal research, development and training of Microsoft's proprietary MAI family of AI models. 3) Enterprise hyperscale cloud customers with long-term strategic value.
While these benefits may not become fully visible over the next few quarters, they should emerge progressively over the coming years as its proprietary AI ecosystem matures.
Microsoft's balance sheet remains exceptionally strong. Although net debt is likely to increase over time, we expect the company's Net Debt-to-EBITDA ratio to remain close to zero, indicating a highly conservative level of leverage.
We reaffirm our BUY recommendation on Microsoft with a 2028 target price of USD560.
The core concern surrounding Microsoft is that its heavy capital expenditure (CAPEX) is not translating into revenue quickly enough.

Microsoft, once the undisputed leader in business software, one of the largest players in cloud computing, and the largest shareholder in OpenAI, the parent company of ChatGPT, now finds itself at a crossroads. Despite maintaining its leadership positions across these businesses, the stock has just recorded its worst monthly performance since 2000 and is down 26% year-to-date (in local currency terms), making it the weakest performer among the Magnificent Seven companies.

The concerns. Investors are rotating away from companies exposed to: (1) heavy capital expenditure (CAPEX) on AI infrastructure that erodes free cash flow (FCF), and (2) the risk of AI disrupting the software industry. Microsoft are exposed to both.

Key distinction: none of these headlines describes a business in trouble. Rather, this reflects a market repricing of a company that is undergoing a significant transition. We believe these are two very different things, and the gap between perception and reality presents an opportunity.

In this article, we examine whether Microsoft's current valuation fully reflects the anticipated margin pressure. We will also revisit our Return on Invested Capital (ROIC) assumptions to determine whether the recent sell-off represents a structural buying opportunity.

Figure 1: YTD performance

Is Microsoft really spending more and getting less

The core concern surrounding Microsoft is that its heavy capital expenditure (CAPEX) is not translating into revenue quickly enough. While all major hyperscalers have announced similarly aggressive investment plans this year, Microsoft has guided for approximately USD190 billion in CAPEX (+61% year-on-year), Amazon for around USD200 billion (+52% year-on-year), and Alphabet for roughly USD190 billion (+100% year-on-year). Despite investing at a comparable scale, investors have been disappointed by Microsoft's pace of revenue growth. Both Alphabet and Amazon reported accelerating cloud growth in the first quarter of 2026. Google Cloud delivered the strongest acceleration, with growth increasing from 48% to 63% year-on-year, while Amazon Web Services (AWS) accelerated from 24% to 28%. In comparison, Microsoft Azure's growth only edged up from 38% to 39%, reinforcing market concerns that Microsoft's AI investments are generating weaker near-term returns.

However, we believe the market is overlooking an important distinction.

Unlike its peers, Microsoft appears not to sell all its available GPU compute capacity to external AI developers. While Azure's short-term revenue could be boosted by renting raw compute power to emerging AI laboratories, Microsoft has prioritised its longer-term strategy over near-term financial optimisation.

Instead of chasing immediate revenue, the company is allocating scarce computing resources across three strategic areas: 1) High-margin proprietary applications, such as GitHub Copilot and Microsoft 365 Copilot. 2) Internal research, development and training of Microsoft's proprietary MAI family of AI models. 3) Enterprise hyperscale cloud customers with long-term strategic value. This capital allocation is consistent with reflects Microsoft's long-standing operating philosophy of prioritising durable competitive advantages over near-term optimisation.

By refusing to maximise Azure revenue through raw compute sales, Microsoft is consciously accepting temporary pressure on free cash flow (FCF) and Return on Invested Capital (ROIC) in exchange for building a proprietary AI ecosystem that reduces its long-term dependence on OpenAI. Chief Executive Officer Satya Nadella has repeatedly argued that every company must build its own "learning machine" rather than relying solely on third-party foundation models.

Latest progress.

1. Chips. Microsoft's in-house AI accelerator, MAIA, has already been deployed across Azure data centres and is now providing compute capacity to enterprise cloud customers. The company is also reportedly in discussions to supply Anthropic with compute powered by its next-generation Maia 200 chips. Over time, wider deployment of Microsoft's own silicon should lower its internal cost-to-serve, improving infrastructure efficiency and supporting margin expansion.

2. Models. Microsoft's specialised multimodal models, including MAI-Image-2.5 and MAI-Transcribe-1.5, are now available through Microsoft Foundry under standard usage-based API pricing, generating commercial revenue. More importantly, MAI-Code-1-Flash has been integrated into GitHub Copilot Business and Enterprise, representing a significant monetisation milestone by embedding Microsoft's proprietary models directly into one of its highest-margin subscription businesses. Microsoft 365 Copilot. The company now reports approximately 20 million paid enterprise seats, representing a conversion rate of slightly above 4%, up from 15 million seats and a 3.3% conversion rate in the previous quarter. At the standard list price of USD30 per user per month, these subscriptions imply an annualised revenue run-rate of approximately USD7.2 billion from the Copilot add-on alone.

The easiest way might not be the best way. An analogy may help. Rather than renting out all available kitchen capacity for immediate income, a restaurant may reserve it to develop proprietary recipes that generate higher long-term returns. Microsoft is making a similar trade-off with its AI infrastructure.

The fact that its flagship reasoning model, MAI-Thinking-1, remains in private preview while smaller models have already been commercialised through GitHub Copilot demonstrates that Microsoft is intentionally delaying near-term monetisation. Rather than opening its GPU clusters to maximise third-party API revenue, the company is dedicating scarce computing resources to internal model development and high-margin first-party applications. This reinforces our view that today's margin compression is a deliberate strategic investment rather than evidence of weak demand.

Simply put, we believe Microsoft retains meaningful room to unlock both revenue and profitability over time. While these benefits may not become fully visible over the next few quarters, they should emerge progressively over the coming years as its proprietary AI ecosystem matures. We forecast Azure revenue growth of 39% in 2026, followed by 40% in 2027 before moderating to 37% in 2028, while operating margins gradually stabilise at around 45%.

Encouragingly, Microsoft's Cloud Revenue-to-CAPEX ratio has already begun to improve, suggesting that investment efficiency is moving in the right direction. We expect this trend to strengthen further, supported by the company's substantial Remaining Performance Obligations (RPO) of approximately USD627 billion, including roughly USD157 billion of current Remaining Performance Obligations (cRPO) that are expected to be recognised within the next 12 months, representing around 25% of total RPO.

Figure 2: Cloud Revenue forecast; AI contribution increasing

Figure 3: Cloud/CAPEX

Fundamental View: Healthy Financing Capacity

Unlike several other hyperscalers, Microsoft has not tap into the debt market since 2025, relying primarily on its own cash reserves to fund its AI infrastructure investments. Instead of raising additional debt, the company has continued to repay maturing borrowings, reducing total debt from approximately USD45 billion in 2024 to around USD40 billion as of early 2026.

Microsoft also maintains a strong liquidity position, with cash and cash equivalents of approximately USD78 billion as of the latest quarter. This provides a sizeable financial buffer to support future capital expenditure.

Nevertheless, we see possibilities for Microsoft to leverage on the debt market in late 2026 or early 2027 as an increasing proportion of its finance leases convert into recognised liabilities and to secure further financial capacity to sustain this level of investment.

All in all, Microsoft's balance sheet remains exceptionally strong. Although net debt is likely to increase over time, we expect the company's Net Debt-to-EBITDA ratio to remain close to zero, indicating a highly conservative level of leverage.

The credit market reinforces this view. Microsoft's credit default swap (CDS) spreads remain the lowest among the major technology companies, reflecting investors' confidence in the company's financial strength and its ability to comfortably finance its long-term investment plans.

Figure 4: Microsoft has yet to tap into the debt market since 2025

Figure 5: Net debt rising fast, but revenue also catches up fast

Figure 6: CDS spread lowest for Microsoft

Q2 2026 Earnings Preview: Key Metrics We Monitor

Microsoft is expected to announce their Q2 2026 earnings result towards the end of July. A few key metrics that we think will impact the share prices are:

1) Azure cloud growth and margin: We expect Azure revenue growth to exceed 39% year-on-year, representing a modest acceleration from the previous quarter. While higher AI infrastructure investment is likely to place further pressure on profitability, we expect the Intelligent Cloud segment's operating margin to remain above 40%. We also expect Remaining Performance Obligations (RPO) to continue expanding, reaching close to USD650 billion, representing growth of more than 70% compared with a year earlier.

2) Capex: We expect quarterly CAPEX to exceed USD40 billion. The sequential increase should be driven primarily by higher finance lease additions, alongside approximately USD5 billion of additional costs resulting from higher component prices.

3) Core business segment: Although this is unlikely to be the primary focus of the earnings release, we expect growth in the Productivity and Business Processes segment to moderate to around 12% year-on-year as PC demand continues to normalise. However, we believe overall growth could begin to re-accelerate over the medium term, supported by Microsoft's partnership with Nvidia to reinvent Windows AI PCs. We will also be closely monitoring the number of paid Microsoft 365 Copilot seats, as this provides an important indicator of whether enterprise customers are renewing licences and expanding adoption.

Additional observation: We realised its net finance leases (finance lease additions – goods received vs paid timing) have been decreasing over the last two quarters without any significant slowdown in quarterly additions. We believe this primarily reflects delays in data centre constructions rather than weaker investment demand, with few near-term solutions available to accelerate deployment.

With that, even if reported CAPEX falls short of market expectations over the next few quarters, it would not materially change our long-term investment thesis on Microsoft.

Figure 7: net Finance leases are decreasing, but additions have not stop

A 30% discount from historical average

Following the recent sell-off, Microsoft is currently trading at approximately 19x forward earnings, around 30% below its historical average forward P/E of 27x.

Some investor concerns remain valid. In particular, there remains uncertainty over the monetisation path of the GPU capacity currently allocated to Microsoft's proprietary AI model development and whether these investments will ultimately enable the company to remain competitive against other frontier AI models.

However, even under conservative assumptions regarding the future revenue contribution from these investments, we believe Microsoft's current valuation already more than discounts these risks. The stock continues to trade at a meaningful discount to its own historical average despite the company's resilient underlying fundamentals.

Short-Term Pain, Long-Term Opportunity; Reaffirm BUY for Microsoft

The market is currently pricing Microsoft as though its AI investment cycle will permanently impair the company's profitability. In our view, investors are penalising Microsoft for taking a disciplined, long-term approach by allocating computing resources towards developing its own AI models instead of maximising short-term Azure revenue.

The underlying fundamentals tell a very different story.

We expect Azure to sustain approximately 39% revenue growth this year, while more than 80% of Fortune 500 companies are already using Azure AI services. Meanwhile, Microsoft's traditional Productivity and Business Processes segment is still expected to deliver double-digit revenue growth despite concerns that AI could disrupt the software industry.

These are not the characteristics of a business in structural decline. Rather, they are the characteristics of a high-quality company undergoing a strategic transition—one that the market is currently valuing as though it were facing a structural deterioration in its business model.

Based on our forecasts for earnings growth of 13% and 17% over the next two years, we derive a target price of USD560, based on a fair forward P/E multiple of 25x.

Accordingly, we reaffirm our BUY recommendation on Microsoft with a 2028 target price of USD560.

Table 2: Valuation table

	2025Y	2026Y	2027Y	2028Y
P/E	36.36	22.2	19.5	16.7
Earnings	13.68	16.82	19.09	22.38
Earnings Growth	24.29%	23%	13%	17%
Sales (bil USD)	281	329	385	454
Sales Growth	14.90%	16.90%	16.80%	18.10%
Fair PE				25
Target Price	390			560
Upside Potential				43.5%
Source: Bloomberg Finance L.P., iFAST compilations. Data as of 05 July 2026.

Declaration:

For specific disclosure, at the time of publication of this report, IFPL (via its connected and associated entities) and the analyst who produced this report hold a NIL position in the abovementioned securities.

This research report was prepared with the assistance of artificial intelligence (AI) tools. iFAST Financial Pte Ltd does not rely exclusively on AI for content generation; the content of this report – including all investment theses, ratings, price targets and conclusions – has been independently reviewed and verified by the research analyst(s) to ensure accuracy and professional integrity.

Select Country/Region

Microsoft: Short-Term Pain, Long-Term Opportunity

Key Points

Is Microsoft really spending more and getting less

Fundamental View: Healthy Financing Capacity

Q2 2026 Earnings Preview: Key Metrics We Monitor

A 30% discount from historical average

Short-Term Pain, Long-Term Opportunity; Reaffirm BUY for Microsoft

In this article

NASDAQ: MSFT

ISOTeam: A player in Singapore's building maintenance and estate upgrading industry

Jumbo Group: East Coast closure prompts downgrade to TRIM

SpaceX has investors excited. Here's why we're looking elsewhere

Home

About Us

Legal

Useful Links

Select Country/Region

FSM Mobile Apps

Follow Us