Market 'super-bubbles' are identifiable by specific characteristics like narrowing market breadth and extreme investor speculation, and they invariably revert to (or below) their mean valuation.
The Federal Reserve, particularly under Alan Greenspan and Ben Bernanke, has been a primary actor in creating asset bubbles, including the first-ever nationwide U.S. housing bubble, through overly accommodative monetary policy.
AI is a genuinely transformative technology comparable to the railroads, but the current massive capital investment in it constitutes a classic bubble that has temporarily halted a broader market decline.
Long-term global growth faces severe headwinds from declining fertility rates across the globe and the massive, under-appreciated costs associated with climate change.
Adhering to a disciplined value investing strategy is extremely painful during bubble periods, leading to significant business risk, but is ultimately vindicated when markets inevitably correct.
1977-1986
In its first nine years, Grantham's firm GMO reportedly outperformed its benchmark every single year, with an average outperformance of 8 percentage points annually.
1998-2000
During the peak of the dot-com bubble, Grantham's value-oriented stance caused GMO's assets to fall from $30 billion to $20 billion, losing over half its market share. During this period, Goldman Sachs attempted to acquire the firm.
2000-2002
As the dot-com bubble burst, Grantham's strategy was vindicated. GMO generated cumulative returns of approximately 35% while the S&P 500 declined by 50%.
c. 2002-2006
In the four years following the dot-com bust, GMO experienced massive growth, with assets under management increasing from $22 billion to $165 billion.
2022
Grantham identifies a market downturn where speculative stocks fell significantly, but claims the broader market did not decline enough to reach fair value, with the Shiller P/E remaining at a 'badly overpriced' level.
Late 2022-2023
Grantham asserts that the emergence of generative AI and the subsequent 1.5% of GDP in related capital expenditures interrupted a classic bear market and prevented a U.S. recession.
▶Market Bubble Identification and PsychologyApr–Jun 2026
Grantham outlines a recurring pattern for identifying the peak of major market bubbles, characterized by speculative stocks declining while blue-chip stocks continue to rise, a divergence seen in 1929, 2000, and 2021. He asserts that bubbles form around genuinely world-changing ideas that become overhyped, leading to irrational valuations and 'crazy' behavior, such as pre-revenue companies being valued higher than established industrial giants.
For analysts, this theme suggests that the final stage of a bubble is not a uniform ascent but a period of internal market divergence, providing a potential late-stage warning signal.
▶The Pain and Vindication of Value InvestingJun 2026
Grantham details the extreme business pressure of maintaining a value discipline during a bubble, citing GMO's experience of losing over half its market share and a third of its assets during the dot-com era. Despite this, the strategy was vindicated when the bubble burst, with GMO posting significant positive returns while the market crashed and its assets under management subsequently grew eightfold in four years.
This highlights the critical difference between being correct in an investment thesis and surviving the market's irrationality long enough to profit from it; even a correct contrarian stance carries immense career and business risk.
▶AI: A World-Changing BubbleJun 2026
Grantham presents a dual thesis on Artificial Intelligence, viewing it as a historically significant technology on par with the 19th-century railroads. However, he argues the current excitement has fueled a massive capital expenditure bubble, not in stock P/E ratios, but in the purchase of expensive AI chips by companies yet to profit from them. He claims this AI-related capex surge single-handedly prevented a U.S. recession in 2023 and interrupted the bursting of a broader market super-bubble.
This perspective separates the long-term technological impact of AI from its short-to-medium term market and economic impact, suggesting investors can believe in the technology's future while being wary of the current investment cycle's sustainability.
▶Long-Term Macroeconomic Headwinds
Grantham expresses concern over several long-term structural challenges to global growth, focusing heavily on demographics and climate change. He points to collapsing fertility rates in developed and developing nations alike, such as South Korea (0.7) and India (1.9), as a major drag. He also estimates that 25% of global GDP growth since 2000 has been consumed by preparing for or repairing climate-related damage.
These claims imply that traditional economic models may be underestimating the impact of non-financial, structural forces on future growth, requiring a broader analytical framework for long-term forecasting.