Warning signs of US stock market overheating! Shiller P/E ratio approaches the 1999 bubble peak

The Shiller P/E (CAPE) has risen to nearly 40x, approaching the peak of the 1999 dot-com bubble. While the fundamentals of AI leaders are stronger than they were back then, market concentration and the risk of compressed returns require investors to remain vigilant. (Previous coverage: US SEC pauses "RWA tokenized US stocks"! Wall Street worries 24/7 trading could fragment market liquidity) (Background: The era of the one-person company is here! Anthropic releases "AI-Native Startup Handbook": You can be a tech CEO without knowing how to code) In 2026, the classic valuation metric for the US stock market, the Shiller P/E (CAPE), has climbed to a range of approximately 39.5x to 41.7x, marking its highest level in the past 25 years. This figure is just below the extreme of approximately 44x seen at the peak of the 1999 dot-com bubble. This means that, by historical standards, the current S&P 500 index has entered a "highly inflated" zone. The core driver of this valuation surge is not the mania fueled by ".com" suffixes and dreams of wireless telephony as seen in 1999, but a comprehensive repricing triggered by the AI (Artificial Intelligence) concept. Tech giants related to semiconductors, cloud computing, and AI infrastructure—led by NVIDIA, Microsoft, Amazon, and Meta—have continued to rise, with a small number of large-cap tech stocks accounting for a significant portion of the S&P 500's gains. Market observers worry that the current trend shares structural similarities with the 1999 dot-com bubble: a dominant technological narrative, the market cap inflation of a few star companies, and a widespread belief that "this time is different." However, simply labeling the current market as a "bubble" may be an oversimplification, as the corporate fundamentals supporting these high valuations are vastly different from those of that era. Unlike the tech companies of the past, many of which were unprofitable and relied solely on "user growth curves" to convince investors, today's AI leaders generally possess the following characteristics: - Strong free cash flow: NVIDIA's annual free cash flow has exceeded $80 billion, and Microsoft's cloud business (Azure) has annual revenue approaching $300 billion. - Mature business models: Unlike the "grow first, ask questions later" approach of 1999, the revenue of AI giants largely comes from subscription-based SaaS, cloud usage-based billing, and enterprise-level contracts. - High profit margins: The combined operating margin of the Magnificent Seven generally falls between 30%–35%, far exceeding the approximately 17% of the Nasdaq 100 in 1999. Furthermore, the current AI investment wave has expanded from the chip level to data centers, energy (UHV transmission, micro-nuclear reactors), power grids, and enterprise-level AI applications. This leaves bubble skeptics with an awkward question: if this is a bubble, it may be the "most fundamentally sound" bubble in history. Despite fundamentals being stronger than in 1999, analysts still point to three major risk factors: Concentration risk: If the Magnificent Seven are excluded, the performance of the remaining 493 stocks in the S&P 500 is far below expectations. This structure, where the "Seven" hold up the market, makes it particularly sensitive to pullbacks in individual giants. Interest rates and commercialization variables: The US Fed's interest rate environment (Fed Funds Rate of approximately 4%–4.25%) has not been fully priced in. If future interest rates rise above 4.5% due to a rebound in inflation, the return on capital expenditure (CapEx) for AI infrastructure may face a reassessment. At the same time, if the progress of AI commercialization falls short of expectations—such as slower enterprise adoption or delayed ROI—the pressure of valuation compression will emerge rapidly. Historical return compression: Historical data shows that after the Shiller P/E enters a historically high range (CAPE > 30), the real return (inflation-adjusted) of the US stock market over the next decade usually tends to be in the low single digits (about 1%–3%). This means that even without a "hard landing," high valuations themselves will compress future annualized returns. The Shiller P/E is not a panacea, as the US stock market did experience historical return compression after 1999; however, it also took nearly a decade for the CAPE to fall from 44x in 2000 to about 13x in 2009 to digest the valuations.