Image Source: Pixabay

The Continuing Market Melt-up

The third quarter sustained the upward momentum in the market, which we have characterized as a “market melt-up”. In our assessment, this trend was attributable to two primary factors.

First, the extraordinary performance of the American stock market since the Global Financial Crisis. From the end of 2009 through October 1, 2025, the S&P 500 delivered a total return of nearly 700%. While there were periods of decline, each drawdown was quickly erased as the market marched to new highs.

Second, the relentless rise of passive investing, which accelerated throughout the period. Inflows to passive funds are allocated by market capitalization, with no regard for fundamentals such as revenue, earnings, or valuation.

Together, these forces fostered a simple playbook: whenever funds became available investors funneled them into passive index funds and waited for the market to rise. Fund-flow data confirms that this behavior was widespread.

The most recent melt-up began following President Trump’s walking back of his Liberation Day tariffs. In response to tariff concerns, the S&P 500 had fallen to 4,982.77 on April 8, 2025. The next day, after President Trump announced that he was postponing the tariffs, the index surged 9.5%. The following week the market remained volatile, but by April 21 the fireworks were over — and the melt-up began.

From April 21 through October 1, there were 118 trading days. Of those, 74 days or 62.7% were positive. That alone is remarkable, since over long periods up and down days usually split close to 50-50. Even more striking, declines were shallow: only three days saw losses greater than 1%, and none exceeded 2%. In comparison, on the upside, there were 11 days with gains over 1% and four days with gains above 2%. The result was an almost riskless melt-up — an investor simply buying the index would have enjoyed a nearly monotonic increase in wealth.

Of course, such a run cannot continue indefinitely. At some point, valuations become too rich. The critical questions are: when will the melt-up end, and how sharp will the reversal be? While the rise has been gradual, the fall could be sudden if enough investors rush for the exit at once. One way to access the situation is to examine the implications of Robert Shiller’sCAPE index.

The CAPE Ratio: Extreme Valuations and Long-Term Return Implications

The CAPE (Cyclically Adjusted Price-Earnings) ratio recently moved above 40, a threshold exceeded only once before, during the 2000 tech bubble. As shown in the chart below, the CAPE didn’t even reach 40 during the height of the bubble in 1929, underscoring just how extraordinary current valuations have become.

(Click on image to enlarge)

Historical Context and Structural Changes

Some market observers question whether the CAPE ratio remains a valid measure for stock market valuation. The chart shows that the long-term average from 1900 to 2025 stands at 17.77, yet CAPE has not fallen below this average since around 1990, except for briefly during the financial crisis. This persistent elevation raises the question: are we living in a fundamentally different world that makes traditional CAPE analysis less relevant?

There is compelling evidence that we have undergone structural changes that render simple reversions to long-term historical averages potentially misleading (see our paper on Structural Changes for further analysis). However, this doesn’t mean we should abandon the measure entirely—CAPE can still provide valuable insights about long-term expected returns when properly contextualized.

Adjusting CAPE for Modern Markets

Rather than dismissing CAPE, we can make reasonable adjustments to reflect the structural changes of recent decades. Two straightforward approaches yield remarkably similar results:

Modern Era Average (1989-2025): 26.72
Long-term Trend Line (1900-2025): 25.92

The convergence of these two independent methods around 26-27 provides confidence that this range represents a more appropriate equilibrium for today’s market structure.

Mean Reversion Scenarios

Understanding these adjusted benchmarks dramatically changes the implications of potential mean reversion:

• Reversion to historical average (17.77): Would require a 55.9% market decline
• Reversion to modern average (26.72): Would require a 33.6% decline
• Reversion to trend line (25.92): Would require a 35.6% decline

Note that these declines correspond only to the averages, if the CAPE moved below the averages the declines would be even more pronounced.

Long-Term Return Implications

Importantly, CAPE and its inverse—the earnings yield—have proven to be strong predictors of long-term market returns over periods of 10 years or more. However, they are not reliable tools for short-term market timing. Stock prices can remain above or below fair value for years before eventually reverting closer to equilibrium levels.

With CAPE at approximately 40, the current earnings yield of 2.5% suggests nominal returns of roughly 4-5% annually over the next decade—well below the historical average of approximately 9%. This doesn’t predict when a correction might occur, but it does suggest that patient, long-term investors should moderate their return expectations.

Investment Implications

A full return to the pre-1990 CAPE long-run average of 17.77 seems unlikely. The economy, interest-rate environment, and market mix have all evolved in ways that justify somewhat higher average multiples. Still, by any modern benchmark—whether the post-1990 average or the long-term trend—a CAPE near 40 screams “expensive” and signals low long-run returns ahead.

This analysis reinforces our emphasis on:

• Tempering return expectations for the next decade
• Maintaining portfolio diversification across asset classes and geographies
• Remaining patient for better entry opportunities that may emerge over time

The CAPE ratio’s extreme readings don’t predict the timing of market corrections, but they do provide valuable context for long-term investment planning and return expectations.

Are AI Investments Going to Unlock a Goldmine?

It’s tempting to think that every breakthrough technology is a goldmine for investors. But that’s not how markets work. Stock prices don’t move on how much value society gets from a new technology—they move on how much profit companies can keep for themselves.

For shareholders, technology only creates value when it delivers returns on invested capital (ROIC) that exceed the cost of capital. And those excess returns don’t last unless there are barriers to entry. If competitors can pile in, profits get competed away.

History makes the point. Cars and airplanes transformed the 20th century, enriching society beyond measure. But investors? Not so much. Competition was so brutal that nearly every American automaker and airline went bankrupt—some more than once. Warren Buffett once joked that a farsighted capitalist could’ve saved future investors a lot of pain by shooting Orville Wright down at Kitty Hawk. The wealth created flowed to consumers, not shareholders.

AI may follow the same script. As a general-purpose technology, it appears difficult to protect with durable barriers to entry. If that’s the case, artificial intelligence could create enormous value for society—but not necessarily for the companies racing to commercialize it, at least not for all of them.

According to the Financial Times, investment in AI is expanding at a remarkable pace, and the companies supplying the underlying tools and infrastructure—semiconductors, cloud computing, and data centers—have so far been the biggest winners. Capital expenditures related to AI are projected to surge even further in the coming years as firms rush to secure computing capacity and remain competitive.

(Click on image to enlarge)

The question is whether these massive investments will ultimately pay off. For AI to justify its price tag, it must deliver lasting productivity gains and stronger profits across the economy. If those returns fail to appear, the capital now flooding into AI could turn from asset to liability, as expensive infrastructure depreciates and balance sheets come under pressure. Even if AI transforms daily life, fierce competition may prevent the companies making the largest bets from earning commensurate returns on their investment. These are indeed extraordinary times—but for investors, they are also times that call for caution.

More By This Author:

Reflections On Investing: What Is A “Fair” CAPE Value?
The Rise Of Passive Investing
Reflections On Investing: How Much Does AI Value AI?