Hyperscaler Valuation Gap Hits Record Low as Earnings Finally Catch Up

Hyperscaler valuations have moved closer to earnings

For years, the market gave hyperscalers a premium: more market value than their profits seemed to justify. That dynamic has now flipped. Finom's gap between market-cap weight and earnings weight fell to a record-low negative 5.22% on July 24, implying the group's share of aggregate earnings has exceeded its share of market capitalization. The debate is no longer just about AI ambition; it is about whether demand is converting into profits quickly enough to support further rerating.

Recent results have strengthened the fundamental case

Bulls long had the narrative, and the market kept paying up. Recent results have forced a more fundamental read-through. Microsoft's fiscal-year revenue increased 18% to more than $331 billion, while its commercial backlog reached $678 billion. Alphabet's Google Cloud revenue surged 82%, and its cloud operating margin expanded to 35.6%. AWS revenue rose 37% to $42.2 billion, while companywide operating income climbed 43% to $27.5 billion.

That does not remove execution risk, but it does make the case harder to dismiss as pure speculation. What matters next is whether backlog conversion, cloud growth, and margins can keep pace with the sector's capital intensity.

Why the old hyperscaler premium built up

AI-driven expectations drove the premium

The hyperscaler premium was not purely irrational. Once AI became the dominant market narrative, investors were more willing to value these companies on future optionality than on current earnings power. That mindset was reinforced by roughly $27 trillion in market value added to AI-related companies since late 2022. As prices rose quickly, recent success began to look like a base case rather than one outcome among many.

Herd behavior likely amplified the effect. In a market concentrated around a few leaders, underweighting the winners can feel riskier than overpaying for them. The result was a premium built more on expected future payoffs than on fully proven earnings power.

Why earnings changed the setup

The latest earnings wave mattered because it pushed more of that future value into present-period numbers. Goldman Sachs says spending plans for 2026 are nearly 50% higher than estimates from just six months ago, suggesting demand for AI capacity is still intensifying. At the same time, research cited by GMO finds that booms in corporate investment have been associated with lower equity returns.

That combination is the turning point. Demand looks real, but historical evidence still suggests that heavy investment phases can coincide with weaker future market returns. Once earnings catch up, stronger results can support higher prices, but they do not automatically create another round of multiple expansion.

What could reopen the gap

The bull case now depends on cash generation

The upside case is no longer just about AI leadership. It is about turning a heavier capital cycle into durable cash generation. Spending plans for 2026 are nearly 50% higher than estimates from six months ago, which points to ongoing demand for capacity. If that capacity translates into stronger AI services, cloud mix, and enterprise software uptake, investors may once again reward the group ahead of its current valuation base.

That is the real bull trigger: proof that the AI buildout can absorb unprecedented artificial-intelligence spending while producing enough profit and cash flow to reduce the market's reliance on hope.

The bear case centers on payback, not possibility

The clearest skeptic's argument is not that AI demand is imaginary. It is that the payback period may be long, capital-intensive, and sensitive to execution slippage. Even if earnings have caught up with market leadership, history still suggests that investment booms can make the next leg of multiple expansion harder to earn.

Research cited by GMO says booms in corporate investment have been associated with lower equity returns. That does not prove hyperscalers will disappoint. It does mean the market may demand clearer evidence that each extra dollar of spend is producing a measurable dollar of cash.

What to watch instead of enthusiasm

• Free-cash-flow conversion after major infrastructure launches, not just backlog headlines.
• Margin durability as AI services scale.
• The next earnings cycle for evidence that AI capex is improving cash generation rather than merely expanding depreciation and financing needs.

If execution improves, the valuation gap can reopen. If not, the market is likely to reward earnings growth more readily than another wide premium.