Will The Artificial Intelligence Mania Sink Stocks?

Artificial intelligence (AI) and machine learning have existed for some time. Still, the introduction of generative AI tools, like ChatGPT, to a broader audience illustrated the potential power of this technology. As evidenced by the peak popularity of AI as a Google
GOOG
search term worldwide, interest in the technology has reached the general public. Further, this has led to a boom in AI-related stocks. Corporate executives have taken notice since AI was mentioned in 20% of first-quarter earnings calls by S&P 500 companies, according to Empirical Research Partners.

U.S. stock returns year-to-date have been dominated by mega-cap technology and AI-related stocks. For example, the iShares Robotics and Artificial Intelligence ETF (IRBO
IRBO
) has gained almost 21%, while the S&P 500 index is up 9.5% year-to-date. This ETF probably understates some gains in AI stocks, with companies like NVIDIA
NVDA

DIA
(NVDA) soaring by 166%. Meanwhile, the rest of the stock market is languishing on a relative basis. The equal-weighted S&P 500, which weights each of the 500 stocks equally rather than by market capitalization, is fractionally lower for the year.

The outperformance boost from AI has been primarily evident in large-cap stocks versus their smaller brethren. Small company stocks, as measured by the are less than 1% higher year-to-date, significantly underperforming the S&P 500. Over the last 24 months, the difference is even more stark, with the Russell 2000 down over 21% while the S&P 500 is fractionally higher.

Valuations for small companies are at a significant discount to the S&P 500. Looking at the estimated price-to-earnings of small stocks relative to large, small companies are close to their historical low aside from the technology bubble period in the late-1990s. The S&P 500 sells for over 19 times 2023 estimated earnings, while the S&P 600 sells for 13.5 times. As a note, the S&P 600 is used for small-cap stock valuation since it removes companies without profits and makes the comparison to the S&P 500 more comparable. Though small companies are generally more impacted by a recession, small-cap stock relative valuations make a compelling case that an allocation should be a long-term opportunity.

Is AI all hype, and does this tiny sliver of stocks driving stock market returns spell doom for stocks at some point? To answer the second part of the question, Empiral Research Partner crunched the numbers and found that a narrow market has no real predictive power regarding future returns. While many recall the technology stock boom of the late 1990s that led to an ugly peak-to-trough decline of almost 50% in the S&P 500, the gains sometimes broaden to other stocks.

Artificial intelligence is not a fad. It has the chance to be the most significant technological development since the internet. AI has the opportunity to increase productivity across the economy, which is immensely valuable. The real question is the pace and extent of the productivity improvement, which remains unknown.

In any case, the stock market has sniffed out the upside of this technology. Five AI-exposed mega-cap stocks are between 35% and 118% higher year-to-date. There is some fundamental basis for this, rather than just speculation, as some earnings boost from AI is beginning to be seen. For example, Microsoft
MSFT
partnered with ChatGPT and invested in the firm. Four out of five of these mega-cap tech companies have seen their 2023 earnings estimates revised higher or reduced less than the S&P 500.

In looking at other AI-exposed stocks, a similar positive earnings revision trend emerges. Only two of this group have seen earnings estimates move lower in 2023. Despite the lower earnings estimates for 2023 listed here, Marvell announced last Friday that sales of AI products would at least double this year, sending the stock soaring. NVIDIA reported blowout earnings for the previous quarter and predicted more future growth from the AI business.

The intense focus on AI stocks does not guarantee a poor outcome for the stock market, but expectations for these companies to profit from the technology have risen along with the stock prices. As seen during the internet stock bubble of the late-1990s, even a revolutionary innovation that changed the world can be overdone. While not yet in bubble territory, AI-related stocks have risen at a blistering pace, and the valuation on the mega-cap subset has returned to a significant premium to the market. These signs indicate that some caution should be taken when wading into the AI investment waters.