- Markets Mirage
- Posts
- One week / one topic: Show me the money
One week / one topic: Show me the money
Do shareholders like delayed gratification?
What happened?
Three of the ‘Fab four’ mega-caps at the center of the ‘AI revolution’ – Microsoft, Meta and Alphabet – have just reported earnings. (Nvidia reports on May 22nd)
The common thread? They all plan massive investments in the race for AI supremacy, and therefore less money for shareholders as retained earnings, dividends or buybacks.
In total, outlined 2024 capital expenditures amount to ~$130bn: that’s $15m every hour – or $250k every minute – from just three companies.
Regardless of what one thinks about the value creation potential of AI technologies (Salvation? Hype? Both? Neither?), MSFT/META/GOOG are just too big to ignore with cumulative market cap of $6tn+ and 2023 earnings of $185bn.
(Microsoft alone is worth more than the top 27 European stocks combined).
These companies' bold ambition for AI dominance is already leading to massive investments today, which will inevitably shape investor outcomes at scale including second-order effects for suppliers, competitors and incumbents.
In a meaningful way, the AI revolution is already here for investors.
How profitable is it going to be?

Source: Olga Tarkovskiy
Our observations
Fundamentals: By broad agreement, these are world-class businesses. The 'only' question is how much you want to pay for them... Forward P/E ratios might look high, but growth-adjusted valuations* are actually lower than the S&P 500 as a whole.
Price action: Given sky-high expectations, recent volatility and dispersion reveal that investors’ patience for open-ended capex cycles is finite. The gains/losses in market cap are simply breathtaking.
Investor beliefs: Investors are still quite discerning when assessing the value-creation potential of each company’s massive investment plans – see market reaction to META’s announcement vs GOOG – so we are not detecting signs of unbridled optimism.

*: Bloomberg estimates of the current year P/E ratio divided by the long-term growth rate
So what?
These companies normally command a valuations premium given their stellar track records, and investors tend to grant them the benefit of the doubt.
For context, the best-performing 4% of US listed companies explain the net gain for the entire stock market over the last ~100 years – while all other stocks collectively only matched Treasury bills returns, but for much more risk.
Clearly, MSFT/META/GOOG very much belong in the former group – aka 'superstar companies' – but investors’ goodwill is not an infinite resource…
While we own these stocks in certain portfolios in line with specific investment objectives, we also strive to remind ourselves that nothing lasts forever as investment opportunities constantly come and go.
For a quick reminder, the Nasdaq lost 25% in 2022 as Tech stocks were suddenly considered ‘high duration’ and therefore very vulnerable to a steep hiking cycle.
Once again, we strive to maintain the delicate balance between respecting price momentum and practicing ‘patient opportunism’ as we wait for high-probability investment opportunities.
