Join the "AI bubble" debate, Goldman Sachs: There is no bubble yet!
```
Author: Li Xiaoyin
Source: Hard AI
The surge in AI-driven tech stocks is sparking an intense debate about a “bubble” in the market.
Recently, Goldman Sachs analysts Peter Oppenheimer, Sharon Bell and others gave a clear answer in their latest research report: Although the current market exhibits some characteristics of historic bubbles, we are not in one yet. Goldman Sachs believes that the biggest difference from past bubbles is that this round of tech stock appreciation is mainly driven by robust fundamentals and solid earnings growth, rather than pure speculative frenzy.
The report acknowledges that investment mania around transformative technology is a common cause of bubbles, and that today’s higher absolute valuations, highly concentrated market structure, and soaring capital expenditure by tech giants do have similarities to past bubbles.
However, Goldman Sachs highlights several key differences. First, the rise in technology stock prices so far has solid earnings growth as support. Second, the leading companies driving the market possess exceptionally strong balance sheets, in stark contrast to the high-leverage-driven bubbles of the past. Finally, the current AI sector competition is mainly dominated by a few incumbent giants, whereas most bubbles typically form during periods when a large number of new entrants flood in and the competitive landscape is chaotic.
Despite Goldman’s view that systemic risk is low, the bank also warns that the current extremely high level of market concentration is “unsustainable,” and that a heavy dependency on earnings growth means the market could face sharp corrections if performance falls short of expectations.
Valuations Have Risen but Remain Below Bubble Levels
Regarding the valuation issues most concerning to investors, Goldman Sachs concludes after multi-dimensional comparisons that: While technology stock valuations are high, they have not reached the extreme levels of historical bubbles.
The report compares the current “Magnificent Seven” U.S. tech stocks with historic market leaders. The data show the median expected P/E ratio for these seven companies is about 27, far below the roughly 52 times level for leaders at the peak of the 2000 tech bubble.
In addition, whether compared to Japanese banking stocks during the 1989 financial bubble or to U.S. blue chips during the 1973 “Nifty Fifty” era, the valuations of today’s tech giants appear more rational.
Looking at the PEG ratio (price-earnings ratio relative to earnings growth rate), U.S. tech stocks are still below the bubble highs of the late 1990s. Historical profit-based PEG ratios show this indicator reached 3.7 at the tech bubble’s peak, whereas it is now just 1.7.
Goldman also found via a dividend discount model that current U.S. TMT (telecom, media and technology) sector pricing implies a need for 25% annual dividend growth rates over the next 10 years. While that’s high, it’s still below the 35% growth expectation implied during the tech bubble.
Earnings-Driven Rather Than Speculation-Driven
The biggest difference from past bubbles is the core driver of this rally. Goldman emphasizes that the outstanding performance of tech stocks in recent years is more a direct reflection of their robust profitability, rather than unrealistic speculation about the future.
The report shows that since 2009, global technology sector earnings per share (EPS) growth has far outstripped that of non-tech sectors, a gap that has continued to widen since the financial crisis.
To further substantiate this point, the report compares returns now with those from the year before the 2000 tech bubble burst. During 1999–2000, much of tech stock returns came from valuation expansion, with comparatively little contribution from earnings growth.
Currently, earnings growth is the key pillar supporting stock performance, providing a more solid foundation for market stability. The report expects the combined return on equity of the Magnificent Seven tech stocks in 2025 to reach 46%, with a net profit margin of 29%, far higher than the 16% net margin of leading companies during the 2000 tech bubble.
Risk Signals Remain
Although the overall tone is optimistic, Goldman Sachs does not overlook risks lurking in the market, in particular the surge in capital expenditure and record-high market concentration.
The report notes that since the debut of ChatGPT, capital expenditure by “hyperscale computing companies” has surged, expected to reach $239 billion in 2024, more than double 2018’s figure. Historically, tech-driven over-investment often led to overcapacity and falling returns, such as in the telecom sector in the late 1990s.
The key difference is in the means of financing. Currently, tech giants’ capital expenditures are mainly funded by their ample free cash flow, rather than large-scale borrowing. Their balance sheets are exceptionally strong, with abundant cash reserves and typically negative net debt ratios. This is a stark contrast to the 1990s, when telecom companies funded investments mainly through heavy stock and bond issuance, reducing systemic risk for the financial system as a whole.
The report shows that current leading companies’ cash holdings average 2.7% of market cap, with a net debt-to-equity ratio of -22%, and are financially much healthier than leading companies during the 2000 tech bubble.
However, Goldman also notes in the report that there has been a recent rise in debt financing activities among tech companies, which is a shift worth monitoring.
Moreover, market concentration is already at historic highs. Data in the report show that the U.S. market now accounts for over 60% of global market capitalization, and the combined market value of the top five U.S. tech companies exceeds that of the EURO STOXX 50, UK, India, Japan and Canada markets combined, representing about 16% of the global public equity market.
Goldman believes this extreme concentration is “unsustainable,” but points out that this in itself does not equal a bubble. Historically, dominant industries have maintained leadership for decades. From the early 19th century to today, the financial, transportation, energy, and tech sectors have successively become market leaders, each for extended periods.
Focus on Diversification
In conclusion, Goldman Sachs summarizes that the market is not yet in a full-scale AI bubble, but investors should not let their guard down.
The report notes that if investor confidence or patience in the AI theme weakens, the risk of a market correction remains. However, since today’s corporates and banks have relatively healthy balance sheets, the likelihood of an economy-wide impact similar to previous bubble bursts is low.
In the face of high valuations and high concentration risk, Goldman’s core advice to investors is “diversification,” suggesting the following approaches:
Geographic diversification: Although markets outside the U.S. have a smaller proportion of tech stocks, this year, major indices in Europe, Japan, and China have delivered local currency returns on par with the S&P 500.
Style diversification: As the interest rate environment changes, the traditional boundaries between “value” and “growth” styles are beginning to blur, offering greater opportunity for cross-style investments.
Industry diversification: The boom in AI depends on underlying physical infrastructure, driving demand for power, energy, capital goods, and resources—which will bring growth opportunities to other sectors.
Diversification within the tech sector: While focusing on existing giants, investors should also seek the next wave of tech “superstars” that can capitalize on the current capex boom and create new products and services.
This article is from WeChat Official Account “Hard AI”. For more cutting-edge AI information, click here
Risk Warning and DisclaimerThe market has risks, and investment requires caution. This article does not constitute personal investment advice, nor does it take into account the specific investment objectives, financial situation, or needs of any individual user. Users should consider whether any opinions, viewpoints, or conclusions in this article are suitable for their particular circumstances. Investing accordingly is at your own risk. ```