November 1, 2026 | The Oddsmaker Research

Warren Buffett once described businesses as great, good, or gruesome. His distinction was simple: great businesses generate substantial profits without requiring enormous amounts of additional capital. Gruesome businesses continually consume capital just to remain competitive.

It's a particularly relevant framework today, as some of the world's most profitable technology companies embark on the largest infrastructure spending cycle in their histories.

Consider six companies: Microsoft, Alphabet, Amazon, Meta, Oracle, and Tesla.

Together, they represent approximately $13.5 trillion in market capitalization. They generate extraordinary cash flow, but an increasingly large percentage is being reinvested rather than returned to shareholders.

USD billions, except valuation multiples. Indicative October 30, 2026 valuations and trailing financial figures.

Figures are preliminary and require reconciliation to company filings and October 30 closing valuations. Reporting periods and capex definitions may differ. The aggregate EV/FCF is a mathematical calculation, not a conventional portfolio valuation multiple. NM = not meaningful.

The numbers are striking.

These six companies generate approximately $742 billion in operating cash flow, yet spend approximately $702 billion on capital expenditures.

That leaves just $40 billion in free cash flow against nearly $14 trillion of enterprise value.

Put differently, approximately 94 cents of every dollar of operating cash flow is being reinvested.

Buffett's framework is useful because it forces investors to look beyond earnings growth.

The critical issue is mandatory versus discretionary capital expenditures.

A company investing heavily to capture a profitable new market can create enormous shareholder value. But a company forced to continually replace expensive equipment simply to maintain its position faces a very different economic reality.

This distinction is becoming increasingly important in artificial intelligence.

Data centers, GPUs, networking equipment, and power infrastructure require substantial upfront investment. Much of that equipment also faces technological obsolescence.

Today's growth capex could become tomorrow's maintenance capex.

Amazon and Oracle illustrate the problem.

Both appear to trade around 15 times EBITDA in our preliminary comparison. Yet both are generating negative free cash flow after capital expenditures.

Alphabet presents another interesting case. Its indicated enterprise value is approximately 18 times EBITDA, but nearly 170 times current free cash flow.

These are not necessarily bad businesses or bad investments. The spending could generate substantial future profits.

However, investors are increasingly being asked to underwrite future returns on capital rather than the cash these businesses generate today.

And that introduces a different kind of risk.

Scores reflect capital efficiency, not overall investment attractiveness.

Microsoft arguably has the strongest combination of recurring revenue, pricing power, and competitive advantages in this group.

Alphabet and Meta retain exceptional core businesses, but investors need to evaluate the returns on their rapidly expanding infrastructure investments.

Amazon and Oracle face particularly important questions about cash generation, financing requirements, and the long-term economics of AI infrastructure.

There are two very different possibilities.

The optimistic outcome: AI generates substantial new revenue and productivity improvements. Today's capital spending proves highly profitable, free cash flow rebounds, and the infrastructure investment cycle creates lasting competitive advantages.

The less favorable outcome: AI becomes essential, but competition and recurring equipment replacement consume much of the economic benefit. Revenue grows, yet returns on invested capital disappoint.

Both outcomes could involve tremendous technological progress. Only one necessarily produces exceptional shareholder returns.

This is why we believe the next several years will require investors to look more closely at free cash flow, incremental returns on capital, and the economic life of AI infrastructure.

The most important question isn't whether artificial intelligence will transform the economy. It almost certainly will.

The question is how much of that economic value ultimately accrues to shareholders after paying for the infrastructure required to deliver it.

History is filled with industries that changed the world without consistently rewarding the investors who financed their expansion.

Buffett understood that a great business isn't simply one that grows earnings. It's one that can grow earnings while retaining substantial cash for its owners.

With nearly $700 billion in annual capital expenditures across these six companies, that distinction deserves considerably more attention.

At The Oddsmaker, we believe the winners of the next decade will be determined not just by who benefits from AI, but by who captures the greatest economic return on the capital invested.

The Oddsmaker | Finding the Best & Worst 1% of Stocks

Financial figures are indicative and should be independently verified before publication. This commentary is for informational purposes only and does not constitute investment advice.