Goldman Sachs pointed out that about 75% of the equity value of the S&P 500 comes from the "terminal value" (that is, the forward earnings expectations ten years in the future), which is close to the peak during the internet bubble.

Written by: Bai Shuqing

Source: Wall Street View

The narrative of AI disruption is reshaping the pricing logic of U.S. stocks. Goldman Sachs' latest research shows that investor concerns about the potential impact of artificial intelligence on companies' long-term profitability have shifted market focus to the most difficult-to-quantify and most sensitive part of stock valuations—the "terminal value," namely, the forward earnings expectations ten years out.

According to Wind Trading Desk, Goldman Sachs estimated in a report released on April 27 that currently about 75% of the S&P 500 index's equity value comes from the terminal value, nearing a 25-year high, echoing the optimism during the internet bubble period. Goldman Sachs also calculated that for every 1 percentage point drop in long-term growth rate assumptions, the enterprise value of S&P 500 constituents will shrink by about 15% overall; for high-growth stocks, this impact could be as high as 29%.

Goldman Sachs stated that discussions around AI disruption—and the resulting uncertainty about terminal value—are expected to continue for at least several quarters. "The threat of disruption is likely to create sustained pressure until AI applications reach a more mature stage."

Terminal value proportion approaching 25-year high, similar to the internet bubble period

Goldman Sachs uses a revised 10-year Dividend Discount Model (DDM) to estimate that currently about 75% of the S&P 500 index's equity value is concentrated in the terminal value part, which refers to the long-term value beyond the 10-year forecasting period. This ratio is at a historically high level, highly similar to the optimistic expectations during the internet bubble in 2000.

Goldman Sachs noted in its report that the high proportion of terminal value reflects optimistic market expectations for long-term growth, but also means that valuations are very sensitive to changes in long-term growth assumptions. "Today, the proportion of terminal value in equity value is relatively high compared to historical levels, paralleling other periods when investor long-term growth expectations became increasingly optimistic, including the internet boom."

In terms of industry distribution, the terminal value proportion is much higher in high-growth, high-margin industries compared to low-growth industries. Goldman Sachs' estimates show that the terminal value constitutes about 84% of the enterprise value for high-growth stocks, about 72% for the S&P 500 overall, and about 59% for low-growth stocks.

Software sector is most affected, AI competitive pressure spreading to lightweight asset industries

The core concern regarding AI disruption is that artificial intelligence may foster low-cost competition in industries with low entry barriers, thereby suppressing the revenue growth and profit margins of existing companies. Goldman Sachs pointed out that the industries deemed most susceptible to disruption are precisely those that have grown the fastest and have the highest profit margins in the past.

The software sector is at the epicenter of this round of sell-off. According to Reuters, the S&P 500 Software and Services Index has dropped by about 17% so far this year, primarily driven by concerns that new AI tools may erode future revenue growth and profit margins. Goldman Sachs' report also indicates that a group of software stocks has fallen by 19% year-to-date, with selling pressure now spreading to other lightweight asset industries.

It is noteworthy that the recent earnings expectations for these stocks remain robust, creating a stark contrast between the significant decline in stock prices and the resilience of short-term fundamentals, highlighting a repricing of the market based on long-term growth prospects rather than pessimistic expectations for recent performance.

Long-term growth rate is the most important driver of valuation, but short-term fluctuations are dominated by recent expectations

Goldman Sachs' empirical analysis shows that, from a cross-sectional perspective, a company's long-term growth expectations are the most important determinant of its valuation multiple. For every one standard deviation increase in implied long-term growth rate, the corresponding forward price-to-earnings ratio rises by about 0.6 standard deviations (about 4 times the price-to-earnings ratio), making its importance approximately three times that of recent earnings growth, balance sheet strength, market capitalization, and earnings stability.

However, in explaining short-term valuation changes, the role of recent growth expectations and risk premiums is more pronounced. Goldman Sachs pointed out that since 1990, recent growth expectations explain about three times the quarter-to-quarter variations in valuation multiples compared to long-term growth expectations. This is because the volatility of long-term growth expectations is much lower than that of recent expectations, changing relatively slowly in the short term.

This structural characteristic implies that when the AI disruption narrative truly shakes market confidence in long-term growth, its impact on valuations will be profound and difficult to reverse.

Goldman Sachs recommends companies strengthen long-term communication, accelerate repurchases to convey confidence signals

In the face of a market environment with rising uncertainty around terminal value, Goldman Sachs believes that corporate management should take proactive action. The report points out that the importance of terminal value highlights the necessity of communicating long-term growth plans to investors, but the reality is concerning—only 5% of companies in the S&P 500 discussed financial metrics beyond five years in recent earnings calls, primarily concentrated in the utility and real estate sectors.

Goldman Sachs suggests that more management should prioritize communication on long-term prospects, including addressable market size, growth paths, and profitability forecasts, even if multi-year performance guidance itself carries some uncertainty.

Furthermore, Goldman Sachs proposed that accelerating stock buyback programs (ASR) can serve as a tool for management to convey confidence to the market. Academic research generally finds that ASR announcements often trigger positive stock price reactions, often more significant than regular buyback programs. However, Goldman Sachs also warned that large-scale buybacks can sometimes be interpreted as a lack of growth opportunities, so management should combine the scale of ASR with positive statements about future growth prospects to avoid signaling misalignment.