The hidden warning beneath record stock market highs

Investors would be wise to pay attention to what markets may be signalling. Photo by Michael Nagle/Bloomberg via Getty ImagesArticle contentOne of the most dangerous signals in investing is when market indexes tell a very different story from the underlying data.

Today, the S&P 500 sits near record highs, suggesting investors remain optimistic about economic growth, corporate profits and the future. Yet beneath the surface, the average stock is struggling. Market leadership has narrowed dramatically, participation is deteriorating and an increasing number of companies are already in bear market territory.

Roughly 60 per cent of S&P 500 constituents are down more than 20 per cent from their individual highs. Market breadth has been this weak only twice before: during the 1973–74 bear market and at the peak of the technology bubble in 1999–2000.

A welcome email is on its way. If you don't see it, please check your junk folder.

The same divergence is evident when comparing the equal-weighted S&P 500 with its traditional capitalization-weighted counterpart. The ratio between the two has fallen to one of its lowest levels since 2003. Driven almost entirely by the growing concentration of mega-cap technology and artificial intelligence stocks, the ratio is on track for a fourth consecutive annual decline, a pattern not seen since the final stages of the dot-com boom.

In many ways, this is no longer a broad-based bull market. It is a market being carried by a remarkably small group of companies tied to a single narrative: artificial intelligence.

The problem is that the sector carrying the market is also one of the largest consumers of capital. Building the AI infrastructure of the future requires enormous investment in data centres, semiconductor manufacturing, power generation and transmission networks. According to a recent Brookings Institution study, financing the AI buildout could exceed US$10 trillion between 2025 and 2032.

The challenge is that this spending does not occur in isolation. The same pool of capital required to finance AI infrastructure must also absorb massive amounts of new U.S. Treasury issuance as Washington continues to run large fiscal deficits.