The Earnings Are Strong.
So Why Are Investors Selling?
Over the past couple of weeks, I have found myself looking at two sets of numbers that seem to be telling completely different stories. The first is the financial performance being reported by some of the world’s largest and most profitable businesses. Revenue is growing at impressive rates, earnings are expanding, cloud computing demand remains exceptionally strong, and companies are investing enormous amounts of capital into infrastructure they expect will support growth for years to come.
The second is what has been happening to their share prices. Despite what, in many cases, have been excellent financial results, investors have been selling. Some of the very businesses reporting record revenues and profits have seen their valuations retreat from recent highs, and that apparent contradiction is worth exploring.
It is understandable that investors become concerned when portfolio balances decline, particularly after a period of strong performance. But I have always believed that one of the most important distinctions we can make is between what is happening to the price of a business and what is happening inside the business itself. Those two things do not always move in the same direction, and right now, I think that divergence is creating some interesting opportunities.
The earnings are telling an encouraging story
Last month, I wrote about the possibility that we are beginning to see the early stages of an AI-driven productivity boom. My argument was that the enormous amounts of capital being invested in artificial intelligence would ultimately need to translate into measurable improvements in economic output, business efficiency and corporate profitability. The latest earnings season provides another important piece of that puzzle, although what interests me most is not simply the continued enthusiasm surrounding AI, but the actual financial performance of the businesses involved.
Consider four companies we own in our Global Growth portfolio: Amazon, Alphabet, Microsoft and Meta. Their latest reported quarterly revenues total approximately $486 billion, with combined revenue growth of around 20% compared with the corresponding periods last year. These are businesses already operating at an extraordinary scale, collectively generating annual revenues approaching $2 trillion, yet they are still delivering growth rates that many much smaller companies would struggle to achieve.
More importantly, the growth is not confined to revenue. Across several of these businesses, operating profits are expanding, margins remain strong and demand for their core products and services continues to increase. There are legitimate questions about valuations and the enormous capital expenditure required to support future growth, but the underlying financial results provide little evidence of the widespread deterioration that recent share price movements might otherwise suggest.
Amazon is a particularly interesting example. The company reported third-quarter revenue of $201.9 billion, representing growth of 12% from the previous year, while operating income increased approximately 61%, from $17.4 billion to $28.1 billion. That pushed its operating margin to a record 13.9%, an impressive achievement for a business operating at Amazon’s enormous scale. Profitability is expanding considerably faster than revenue, demonstrating the potential operating leverage within a business that has spent decades building its distribution networks, technology infrastructure and customer relationships.
But the most compelling numbers came from Amazon Web Services. AWS generated approximately $46 billion in quarterly revenue, an increase of 39% from the previous year and its fastest growth rate in nearly five years. Annualised, that represents a cloud computing business approaching $184 billion in revenue, yet rather than slowing as it matures, growth is accelerating. AWS also generated $17.8 billion in operating income, an increase of approximately 55%, with an operating margin approaching 39%.
I find those figures particularly encouraging because AWS is not a speculative business hoping to benefit from artificial intelligence at some point in the future. It is an established, highly profitable operation providing the computing infrastructure used by thousands of businesses around the world. AI is adding another substantial source of demand to an already enormous market, and the latest results suggest that demand is translating into meaningful revenue and earnings growth.
There is, however, an important qualification. Amazon expects approximately $220 billion in cash capital expenditure during 2026, an extraordinary commitment that is placing considerable pressure on free cash flow. Over the twelve months to September, free cash flow was negative $42.4 billion, compared with positive $14.8 billion in the preceding period. That is a substantial change, and one investors are right to examine closely.
What matters is understanding why it is happening. Amazon’s operating cash flow increased 29% to $168.3 billion over the same twelve-month period, meaning the underlying business is generating considerably more cash. Management is choosing to reinvest enormous amounts of that cash into infrastructure intended to meet growing demand. The ultimate return on those investments remains to be seen, but accelerating AWS revenue and operating profit provide encouraging evidence that the opportunity is real.
Alphabet’s latest results tell a similarly impressive story, although with a somewhat different combination of growth drivers. The company reported third-quarter revenue of $127.6 billion, an increase of 25% from the previous year, while operating income increased 30%. Its operating margin remained around 32%, demonstrating that Alphabet continues to combine substantial growth with exceptional profitability.
The standout result was Google Cloud, where revenue increased 103% to $30.7 billion. That means the business has more than doubled in twelve months and is now generating revenue at an annualised rate exceeding $120 billion. For a company already operating at this scale, that level of growth is extraordinary, particularly when considered alongside the profitability of Alphabet’s broader operations.
What I find equally interesting is the continued resilience of Google’s core Search business. Google Services revenue increased 11% to $96.3 billion, while Search and related advertising revenue increased 12%. Alphabet also reported that year-on-year growth in Search queries was the strongest it had been in five years, an important development given the concerns that artificial intelligence could undermine traditional internet search.
For several years, investors have questioned whether AI assistants would eventually replace the need to search for information through Google. It is a reasonable concern, and the competitive landscape is certainly changing. But the latest figures suggest that AI may also be expanding the ways consumers interact with information, rather than simply replacing existing behaviour. Alphabet is incorporating AI into its established products while developing new services and infrastructure, giving it multiple opportunities to benefit from the technology.
That does not mean the competitive risks have disappeared, nor does it guarantee that every dollar of infrastructure investment will produce an attractive return. But it is difficult to reconcile the suggestion that Google’s core business is facing imminent deterioration with double-digit Search revenue growth and a cloud division that has more than doubled its revenue. For long-term investors, the combination of an established, highly profitable business and a rapidly expanding second growth engine remains particularly attractive.
Microsoft provides another example of the strength we are seeing across our holdings. For its fiscal first quarter of 2027, the company reported revenue of $92.3 billion, an increase of 19% from the previous year, while operating income increased 21% to $45.9 billion. That represents an operating margin approaching 50%, meaning almost half of every dollar of revenue is converted into operating profit.
There are very few businesses in the world capable of delivering that combination of scale, growth and profitability. Microsoft has spent decades establishing relationships with corporate customers through Windows, Office, enterprise software and cloud infrastructure, and those relationships provide an enormous advantage as businesses begin incorporating AI into their operations.
An organisation already using Microsoft’s technology does not necessarily need to replace its existing systems to begin experimenting with artificial intelligence. It can adopt additional capabilities within software and infrastructure it already uses, making the transition considerably easier than introducing an entirely new technology provider. That distribution advantage is particularly valuable because developing a useful product is only one part of building a successful business. Getting millions of customers to adopt it, integrate it into their workflows and pay for it is often the more difficult challenge.
Microsoft is also investing heavily in the infrastructure required to support this demand, and investors are increasingly focused on whether the returns will justify the capital being committed. That is an entirely reasonable question, but it is worth remembering that Microsoft is not an unprofitable technology start-up attempting to establish a business model. It is an extraordinarily profitable enterprise technology company investing to expand an existing business, with substantial recurring revenue and deeply established customer relationships.
Meta’s results offer perhaps the clearest illustration of how AI can improve the economics of an existing business without necessarily requiring an entirely new product. The company reported third-quarter revenue of $64.6 billion, an increase of 26% from the previous year, while advertising impressions increased 10% and the average price per advertisement increased 14%.
That combination is particularly powerful because it suggests Meta is not simply displaying more advertisements to generate additional revenue. Advertisers are also paying more for the inventory being offered, reflecting the continued improvement in the company’s advertising systems and the value businesses are receiving from them.
AI is playing an increasingly important role in those improvements. Better recommendation systems help Meta identify the content users are most likely to engage with, while more sophisticated advertising tools can improve the relevance and effectiveness of advertisements. Higher engagement creates additional opportunities to display advertising, and better results can encourage advertisers to increase their spending.
Meta reported approximately 3.64 billion daily active people across its family of applications during September, an increase of just 3% from the previous year. Yet revenue increased 26%, suggesting the company is becoming substantially more effective at generating revenue from its existing audience. This is precisely the type of productivity improvement that makes AI commercially interesting, with technology allowing a business to generate more revenue from an established customer base.
The reported earnings figures did include a significant complication. Meta recorded a $9.9 billion charge associated with the settlement of litigation relating to teen mental health and social media, contributing to a 37% decline in reported operating income to approximately $13 billion. Excluding that charge, operating income would have been approximately $22.9 billion, representing growth of around 11%.
The settlement represents a genuine economic cost and should not be dismissed, but separating it from the underlying operating performance provides a clearer picture of the business. Meta is also investing aggressively in infrastructure, with capital expenditure reaching approximately $34.9 billion during the quarter and free cash flow turning negative. Nevertheless, the strength of its advertising revenue, improving monetisation and enormous global audience remain encouraging indicators of the underlying business.
Not every company in our portfolio needs to benefit directly from the AI infrastructure boom, and Apple provides an important example of a different investment opportunity. In its June quarter, Apple reported revenue of $109.4 billion, an increase of 16% from the previous year, while diluted earnings per share increased 29% to $2.02. The company reported record June-quarter revenue across iPhone, Mac and Services, reinforcing the strength of its ecosystem and the loyalty of its enormous customer base.
Apple’s business model differs considerably from those of the major cloud infrastructure providers. While Amazon, Alphabet, Microsoft and Meta are committing enormous amounts of capital to data centres and computing capacity, Apple’s competitive advantage is primarily built around its integrated ecosystem of hardware, software and services. Its installed base provides an established platform through which new products and capabilities can be introduced, while recurring Services revenue contributes an increasingly important source of earnings.
There are certainly challenges, including the maturity of the smartphone market and the need to demonstrate that new AI capabilities can support future demand. But Apple’s combination of customer loyalty, recurring revenue and strong cash generation illustrates why we do not want our investment strategy to depend entirely on a single technology or industry theme. We want to own businesses with durable competitive advantages and the capacity to generate attractive returns over many years, regardless of whether they are directly involved in building the next generation of AI infrastructure.
Why are investors selling when the numbers look so good?
Having considered the financial results, the obvious question is why investors have responded by selling shares in some of these businesses. I think there are several explanations, and understanding them is important because a declining share price does not necessarily indicate a deteriorating investment.
The first is expectations. Markets do not simply reward companies for producing good results; they reward companies for producing results that exceed what investors have already anticipated. Expectations for many of the world’s leading technology businesses have become extraordinarily high, particularly following several years of strong earnings growth and rising share prices. A company can report revenue growth of 25%, increase earnings by 30% and deliver record operating profits, yet still experience a declining share price if investors were expecting even more.
The second issue is valuation. As share prices have risen, investors have become increasingly willing to pay higher multiples of earnings for businesses they believe can sustain exceptional growth. That creates vulnerability when sentiment changes, because even if earnings continue increasing, a reduction in the multiple investors are willing to pay can result in a declining share price.
Consider a business earning $10 per share, with investors willing to pay 30 times those earnings. Its share price would be $300. Now imagine earnings increase 20% to $12 per share, an excellent result by almost any measure. But if investors become more cautious and are only willing to pay 22 times earnings, the share price falls to $264. The business has increased earnings by 20%, yet its share price has declined 12%.
Nothing in that example requires the underlying business to deteriorate. It simply requires investors to become less optimistic about what they are prepared to pay for its future earnings. Of course, valuation multiples cannot expand indefinitely, and there are circumstances where lower valuations are entirely justified. But this is also where opportunities can emerge, particularly when earnings continue growing while share prices decline.
The third concern, and perhaps the most prominent during this earnings season, is the enormous amount of capital being invested in artificial intelligence. Amazon alone expects to spend approximately $220 billion during 2026, while Meta expects capital expenditure of between $135 billion and $145 billion. Add the substantial investments being made by Microsoft, Alphabet and other technology companies, and the scale of spending becomes extraordinary.
Investors are understandably asking whether these businesses are spending too much. Capital expenditure is not free, and money spent building data centres or purchasing advanced processors is money that cannot immediately be distributed to shareholders. Those assets also depreciate over time, and their economic value ultimately depends on generating sufficient future revenue and profits to justify the investment.
There is a genuine risk that the industry builds too much capacity, that technology changes faster than anticipated or that returns on investment prove disappointing. We should acknowledge those possibilities, particularly given the scale of the commitments being made. But I think there is another side to the argument that deserves equal attention.
These companies are not investing hundreds of billions of dollars simply because they have nothing better to do with their money. They are responding to substantial demand for computing capacity, cloud services and increasingly sophisticated AI applications. AWS revenue growth has accelerated to nearly 40%, Google Cloud revenue has more than doubled, Microsoft continues to report strong enterprise technology demand, and Meta is already demonstrating measurable improvements in advertising monetisation.
Those are observable commercial developments, not simply projections of what might happen several years from now. The challenge for investors is determining how much of the future opportunity has already been reflected in share prices and whether the eventual returns on capital will justify today’s valuations. That is a question we take seriously, but it is not the same as concluding that the underlying investment opportunity is deteriorating.
What this means for our portfolios
Turning to our own portfolios, October has been an interesting month. We saw portfolio balances reach elevated levels during the early part of the month before selling pressure began to emerge across a number of our holdings. Naturally, some of those gains have subsequently been given back, but our overall performance has held up relatively well, particularly when viewed against the movements in some individual technology shares.
I think it is important to maintain perspective during periods like this. We do not judge the success of an investment strategy by whether a portfolio finishes every month at an all-time high, because that would be an unrealistic expectation. We judge it by the quality of the businesses we own, the earnings they generate, the prices we pay for them and their ability to compound value over time.
There will inevitably be periods when share prices move ahead of earnings, just as there will be periods when earnings continue improving while share prices move backwards. The latter can be particularly interesting because it may allow us to acquire additional exposure to businesses we already know and understand at more attractive valuations.
We do not rebalance our portfolios particularly frequently, and we certainly do not believe that every period of market volatility requires us to do something. There is considerable value in allowing exceptional businesses to compound over long periods without constantly interfering with the investment process. However, there are also times when market movements create opportunities to improve the portfolio, and I believe we may be approaching one of those periods.
Over the coming weeks, we will be reviewing individual positions across our Global Growth portfolio, particularly those businesses where recent selling has created a more attractive relationship between share price, earnings growth and our assessment of long-term value. This does not necessarily mean increasing our overall exposure to equities. In some cases, it may simply mean reallocating capital from one holding to another where we believe the potential return has become more compelling.
We may reduce positions that have performed particularly well and where valuations remain demanding, while increasing exposure to businesses whose share prices have declined despite strong operating results. In other cases, we may decide that doing nothing remains the most sensible course of action. The important point is that we are evaluating these opportunities from a position of confidence rather than concern, looking for ways to improve a strategy that continues to be supported by the performance of the underlying businesses.
The advantage of being a long-term investor
There is something worth remembering during periods like this. The stock market provides investors with a price for their investments every trading day, but the businesses themselves do not become fundamentally more or less valuable simply because their share prices move. Their value is ultimately determined by their ability to generate cash and profits over time, and those outcomes unfold over years rather than days.
A company that increases earnings by 15% annually will approximately double those earnings in five years. At 20% annual growth, earnings can increase by roughly two and a half times over the same period. Very few businesses can maintain those growth rates indefinitely, of course, but the principle remains important. Over sufficiently long periods, sustained earnings growth can become a far more powerful driver of investment returns than short-term changes in investor sentiment.
That is why we spend considerably more time studying business performance than trying to predict the next movement in the stock market. We cannot reliably tell you where the Nasdaq will finish next week or exactly when the current selling will end. But we can examine what our companies are reporting, assess whether their competitive positions are strengthening or weakening, and compare their current valuations with what we believe those businesses may be capable of earning over the years ahead.
For the past several years, investors have focused on the enormous amounts of money being invested in artificial intelligence, with the obvious question being whether that investment would eventually translate into meaningful economic returns. We are now seeing increasingly encouraging evidence. Cloud infrastructure demand is accelerating, enterprise technology spending remains strong, advertising platforms are using AI to improve commercial outcomes, and some of the world’s largest companies are generating substantial revenue and earnings growth despite their already enormous scale.
That does not mean every dollar invested in AI will produce an attractive return, nor does it mean every technology company deserves a premium valuation. And it certainly does not mean share prices cannot decline further. But it does suggest that the underlying economic opportunity remains substantial, particularly for businesses with established competitive advantages, strong customer relationships and the financial resources to invest through multiple economic cycles.
For our portfolios, that is what matters. We own businesses that we believe are exceptionally well positioned to benefit from long-term changes in technology, productivity and the global economy. Their share prices will fluctuate, and there will inevitably be periods when investors become considerably more pessimistic than the underlying financial results appear to justify. Provided those businesses continue delivering, however, those periods can create opportunities rather than simply risks.
I have always been optimistic about the long-term prospects for high-quality businesses, and the latest earnings season has done very little to change that view. If anything, it has reinforced it. While some investors are becoming increasingly focused on the recent selling, we are looking closely at the opportunities it may be creating.
Ultimately, the most important question is not whether a share price has declined from its recent high. It is whether the business is worth more than the price we are being asked to pay for it.
And right now, I think that is a particularly interesting question.
It may belong to the economy as a whole.
.png)
