Observation on the first line of the Nikkei 225: Highly similar to 1999, investors are being forced to participate in what may be a "once-in-a-lifetime" bull market.
Daiwa Securities stated that the factor structure of the Japanese stock market in April is almost identical to that during the 1999 internet bubble period, with heavy buying focusing on AI and semiconductor sectors.
The Japanese stock market is currently experiencing a market trend highly similar to the internet bubble, with buying interest focused on the AI and semiconductor sectors. The factor structure and index trends are almost identical to those of 1999.
According to a report by Morgan Stanley MUFG Securities released on May 14, the market environment in April 2026 is most similar to the internet bubble period since 1990. The characteristics of the market include underperformance of value factors, significant outperformance of momentum factors, and weakness in low volatility and small-cap factors a rare mirror structure between the two periods.
At the same time, both the NASDAQ 100 Index and the Philadelphia Semiconductor Index (SOX) have entered a steep upward trend since April 2026, highly similar to the late stage of the internet bubble from late 1999 to early 2000.
The report directly points out the strategic implications for investors: due to the sharp rise in stock prices recently, investors have "no choice" but to participate in this trend driven by AI and semiconductors. The report also warns that some individual stocks have shown signs of significant bubble valuations, and advises investors to be cautious of potential reversals while following the trend. With the last six months of the internet bubble period (October 1999 to March 2000) as a reference, the next six months may be a crucial time window to monitor.
The factor structures highly coincide, and quantitative indicators point to 1999
The Morgan Stanley team used Mahalanobis distance to measure the similarity of factor returns from each month since 1990, using 13 factor categories as measurement dimensions, and the conclusion clearly points to the internet bubble period.
Specifically, the months with the lowest Mahalanobis distance indicating the highest similarity to the market environment in April 2026 were concentrated between June 1999 and February 2000. With 8.0 as the historical bottom decile threshold, there were five months within the mentioned range with readings below this threshold. In comparison, the most similar recent time points were October 2025 and January 2026, in line with the market logic of the AI and semiconductor themes continuing from 2025 to present.
Looking at individual factors, the common characteristics between the two periods are clear: value factors (including 12-month forward earnings yield, book-to-market ratio, dividend yield) have all underperformed; momentum factors (12x1M momentum, EPS upward/downward revisions) have significantly outperformed; while low volatility and small-cap factors have weakened. Due to differences in definitions, there are few exceptions in the performance of quality factors between the two periods.
Market concentration hits historical extremes, buying interest narrows sharply to high beta targets
In addition to the similarity at the factor level, the structural characteristics of the current Japanese stock market are also noteworthy.
The ratio of the Nikkei 225 to TOPIX first exceeded 16 times on April 24, 2026, reaching a historic high. The continuous rise of this ratio reflects the extreme concentration of buying interest in a few large-cap, high beta targets with the weights of these targets in the Nikkei 225 components much higher than in the TOPIX.
Momentum signals also show extreme readings. The 12x1M price momentum factor returns calculated over a rolling 20 trading days reached 14% on April 28, hitting a ten-year high. Meanwhile, according to Morgan Stanley's "Beta Polarization" report, the beta values of high momentum stocks are significantly higher, while the beta values of other groups have sharply fallen below 1. In other words, the market interest is almost entirely concentrated in high beta targets, leaving the rest of the stocks significantly behind.
Index valuations have not overheated, while individual stocks show signs of bubble formation
At the valuation level, there is a clear differentiation between the current market and the internet bubble period, which is also one of the most important differences between the two.
From an index perspective, as of May 10, the forward P/E ratios of the NASDAQ 100 and SOX were 24.9 times and 21.5 times, respectively, in a range that is "not low but not too high", contrasting with the rapid valuation surge seen during the internet bubble period. This data shows that the upside potential at the index level has not been exhausted.
However, a different picture emerges when focusing on individual stocks. In a basket defined by the Morgan Stanley research team as "AI Enablers", the median and mean forward P/E ratios of the constituents have been steadily climbing, significantly higher than the levels in 2025. This indicates that valuation pressures are accumulating at the individual stock level, towards stocks with relatively smaller market caps and more concentrated industry themes, increasing the risk of crowded trades.
The report points out that the above valuation differentiation is a key variable in assessing the subsequent market trends there is still tolerance for valuations at the index level, but caution is needed regarding the crowdedness at the individual stock level.
Morgan Stanley: Investors are forced to participate, but need to manage reversal risks simultaneously
Based on the analysis above, the Morgan Stanley quant team has made clear scenario judgments and strategic recommendations for the future market.
The report believes that given the many similarities between the current market and the internet bubble, there is a significant possibility that the current trend will follow the path of the internet bubble period. Seth Carpenter, Chief Global Economist at Morgan Stanley, has qualitatively characterized the current stage of AI development as the sixth major innovation wave since the industrial revolution, alongside the IT revolution of the late 1990s, providing macro support for this structural analogy.
From a strategic perspective, the report directly points out that the steep rise in stock prices recently has effectively left investors "no choice" but to participate in this trend. If the current market does indeed follow the analysis's judgment and becomes the largest bull market since the internet bubble, this will be a "once-in-a-quarter-century" opportunity for investors, who will be forced to go with the trend.
At the same time, the report advises that the appropriate strategy should be to "participate in the trend while maintaining awareness of the potential for future reversals". The late stage of the internet bubble lasted for about six months, and the report therefore qualitatively assesses the next six months as a time window worthy of close attention, to determine whether the market is nearing a potential turning point.
This article is reproduced from "Wall Street See News", GMTEight Editor: Li Fo.
Related Articles

Bank of America's "macro picture" for the next decade: the first 5 years will see "continued inflation," followed by "super deflation" in the next 5 years?

Huachuang Securities April Overseas Monthly Observation: Energy Inflation Further Fermentation, Four Major Central Banks Cautiously Weighing Options

In the past month, the US stock market hit 14 new highs, with momentum stocks surging. Goldman Sachs has uncovered a pattern from the past 40 years: similar market trends usually lead to a decline one month later.
Bank of America's "macro picture" for the next decade: the first 5 years will see "continued inflation," followed by "super deflation" in the next 5 years?

Huachuang Securities April Overseas Monthly Observation: Energy Inflation Further Fermentation, Four Major Central Banks Cautiously Weighing Options

In the past month, the US stock market hit 14 new highs, with momentum stocks surging. Goldman Sachs has uncovered a pattern from the past 40 years: similar market trends usually lead to a decline one month later.






