Valuation password for technology stocks in the AI frenzy: Cash flow is the final arbiter.

Warren Buffett often says that the stock market is a voting machine in the short term and a weighing machine in the long term. The soon-to-retire CEO of Berkshire Hathaway has also claimed that he doesn't understand tech stocks. However, his saying can well explain the valuation phenomena of companies such as Apple Inc. (AAPL.US), Microsoft Corporation (MSFT.US), Alphabet (GOOGL.US), Meta Platforms (META.US), and Amazon.com, Inc. (AMZN.US) in the current artificial intelligence boom. The narrative of which tech giant is leading in the AI field seems to change every month. Recently, after the release of Gemini 3, Alphabet is currently on the rise. This model, based on self-developed chips, has amazed observers in benchmark tests. A few months ago, Meta, led by Zuckerberg, attracted attention through aggressive talent poaching. It can be said that, at least in the informal AI hype rankings, Apple Inc. always lags behind. This iPhone manufacturer with a market value of 4.2 trillion dollars, whose CEO Tim Cook is not a pioneer in large language models, has largely avoided capital-intensive data center construction. As the market is immersed in the AI craze, logically speaking, Cook's aloof stance could lead to a devaluation. But in fact, the opposite is true. According to data, this Cupertino-based group has a price-earnings ratio of 34 times based on expected earnings in 2026, which is 15% higher than Amazon.com, Inc., which has the second-highest valuation among the five giants, with a price-earnings ratio of 29 times. The e-commerce group founded by Jeff Bezos itself is not the focus of the AI craze. Its stake in the large language model developer Anthropic is lower than Microsoft Corporation's stake in OpenAI. Analysts also estimate that Amazon.com, Inc.'s capital expenditures, which include data center investments, will increase by only 15% next year, while Meta's increase is close to 60%, and Microsoft Corporation and Alphabet are around 30%. In other words, the two companies with the highest valuations among the tech giants are precisely the ones least directly involved in large language models. Given the current discussions about the AI stock market bubble, this seems somewhat unusual. There is a simple explanation for this, directly derived from the traditional finance textbooks loved by Buffett: in the long run, the ability to continuously generate hard cash is key. A core difference between profit and free cash flow lies in how capital expenditures are handled: these irregular investments are amortized year by year in calculating net profit, but they are fully deducted immediately in the cash flow statement. This is why some Wall Street professionals like to say that free cash flow is a fact, while profit is just an opinion. This belief still seems to prevail in the tech investment circle today. Investors seem willing to pay a higher premium for Apple Inc.'s profits because almost all of its profits can be transformed into real cash. According to data, Cook's average annual free cash flow from 2026 to 2029 will be equivalent to 108% of earnings. In other words, analysts believe that all of Apple Inc.'s net profits can be used for dividends and buybacks. This makes sense because the company is, to some extent, just a "toll booth" operator: most of its profits come from selling iPhones and running services on them, and users have little inclination to switch. Although Cook has increased AI investments, this has had little impact on cash flow. As a result, Cook's profits are almost equivalent to bank deposits, providing the basis for its high valuation. On the other end of the spectrum is Meta, with analysts expecting its average free cash flow in the coming years to only account for half of earnings. This is partly due to Zuckerberg's enormous investment in data center capital expenditures. The company has a lower price-earnings ratio of 22. Alphabet, Amazon.com, Inc., and Microsoft Corporation are in the middle, and their valuations are similar. According to calculations, the statistical correlation between valuation multiples and the proportion of cash flow in earnings for these five companies is about 90%. Companies with higher expenses have lower valuations, and vice versa, which makes sense. Investors seem skeptical about whether data center investments can be fully recovered and would rather have cash now than future promises. At the same time, NVIDIA Corporation (NVDA.US) and Oracle Corporation (ORCL.US) seem different in this framework. The chip design company led by Jensen Huang is very similar to Apple Inc. in terms of the ratio of net income converted to free cash flow. However, its stock has a price-earnings ratio of only 24 times based on earnings in 2026, closer to Meta than other companies. One explanation is that investors lack confidence in whether NVIDIA Corporation's good times can continue. Hardware demand is usually cyclical, and if the AI hype wanes, new orders for processors will dry up and profit margins will decline. Oracle Corporation is a more peculiar exception, with a price-earnings ratio of 27 times based on next year's earnings. This company led by Larry Ellison is almost like a mirror image of Apple Inc. According to data, the company is investing so heavily in AI data centers that its free cash flow will be negative in the coming years, making its relatively healthy valuation a mystery. One explanation is that, if we don't consider cash flow, Oracle Corporation's profits are growing rapidly, making its price-earnings ratio for 2026 seem high. If we use the multiples for 2028, Ellison's company has a lower trading price than NVIDIA Corporation and Meta. The future of AI is unpredictable. But the current lesson seems to be that investors are more concerned about fundamentals than all the discussions about bubbles imply. This temporarily proves that Cook's relatively cautious stance at Apple Inc. is correct. As competitors invest more and more capital to compete for AI dominance, this iPhone maker is likely to continue focusing on returning a large amount of cash to shareholders. In the long run, this is what investors care most about.