Why returns from artificial Intelligence require real diversification
As noted previously, artificial intelligence (AI) has been almost the sole reason any investors have taken on risk this calendar year. The NASDAQ 100 is up 42 per cent year-to-date, and its gains are almost entirely driven by the prices of just a handful of mega-cap names that investors believe will benefit directly from AI and whose diversified revenue streams are seen as defensive against an anticipated recession. Tech and AI-exposed stocks are responsible for nearly all the gains in 2023.
As an aside, even though I had said late last year that 2023 would be a good one for stocks, I did not expect 42 per cent gains for the Nasdaq 100!
It is difficult to understate the economic impact of emerging AI. Many analysts and commentators have referred to past emergent technologies to refer investors to that potential impact. Since the Personal Computer landed on employee’s desks, successive technologies have been responsible for material and measurable gains in productivity, which have, in turn, expanded corporate profit margins, not least for the technology sector itself.
As I mentioned here a few weeks ago, there’s little doubt AI is beginning to reshape the digital landscape and will ultimately have a measurable impact on the broader economy. Still, at the moment, AI remains something of a novelty, much as the initial release of Apple’s App Store saw everyone turn their iPhones into a torch.
Companies like Meta and Google are driving to integrate AI into everyday applications, but these don’t initially appear to be revenue-enhancing.
The wide divergence between the booming share prices of AI’s major beneficiaries however and pretty much every other stock will need to be supported by more than hype. In the absence of a price to earnings (P/E) re-rating across their market, substantial growth in revenue and profits will be essential to sustain the heady gains recently experienced. And so far, there’s little being said about revenue to justify the enthusiasm. Meanwhile, wider adoption may benefit the consumer more than the companies offering the AI enhancements.
If benefits do accrue to companies, it is likely, in the initial phase of adoption, the economic value will accrue to a small cluster of companies supplying hardware, and as I mentioned above, less clear is who benefits over the longer term.
Adoption of AI by software and SaaS suppliers may serve to boost margins and profits but according to Barrenjoey, the AI-exposed stocks they cover in the U.S. have seen earnings per share (EPS) estimates revised higher by only eight per cent over each of the next two years. And yet, the share prices of those companies are up circa 50 per cent, suggesting there’s more heroism and hype than rational valuation going on.
It suggests investors buying today are paying a very high price, even for what is admittedly a genuinely transformational technology. The mistake of paying too much even for world-changing technologies is as old as investing itself and it’s bound to be repeated by investors who fail to attribute equal weight to valuation. Remembering the higher the price one pays for a stock, the lower one’s return, a high price can turn a good idea into poor investment.
I believe any retracement in share prices – which can happen without notice when a bubble has formed – would present an opportunity. This is because the time it takes for the technology to gain proper mass adoption, and for it to settle into its ultimate use cases is long, and it’s enough time for several bubbles to form and fade. We are still in the very earliest stages of an expanding AI economy. Companies will eventually grow into higher valuations, and while this sentiment can support share prices for a long time, it doesn’t make those prices immune to setbacks.
Recently, Barclays published its 2Q23 Global Outlook entitled Artificial Intelligence – Real but pricey opportunity, and in it the investment bank notes technological advancements over the last 100 years have indeed driven gains in labor productivity and expanded profit margins. Interestingly Barclays also believe U.S. dominance in AI “will likely extend, or at least preserve, its equity market leadership, and continue to attract capital inflows” from the rest of the world, presumably supporting stock prices. And as I have also noted previously, the bank cautions that U.S. market outperformance is coming at a high price now, with a near-record valuation premium and index concentration warranting diversification.
Thinking about that diversification, and noting the compressed P/E ratios of Australian small caps, sensible diversification may include some high-quality Australian small cap names with indirect exposure to the AI theme, such a Megaport (ASX:MP1), and Macquarie Technology Group (ASX:MAQ), whose share prices haven’t matched the gains of their U.S. counterparts.
The Montgomery Small Companies Fund own shares in Macquarie Technology Group and Megaport. This article was prepared 28 July 2023 with the information we have today, and our view may change. It does not constitute formal advice or professional investment advice. If you wish to trade these companies, you should seek financial advice.