Investments: pick-and-shovel rather than AI companies?

For investors in technology, it is tempting to look at the astonishing returns made by the winners and conveniently forget about all the failures they left in their wake.

Finding technology companies with genuine promise – and early enough in their life cycle to make money – is difficult. However, there are other ways to participate in the growth of technology that can be less volatile and unpredictable.

One tried and tested option is to bypass glory stocks, and look at the ‘pick and shovel’ manufacturers; in other words, the enablers of change rather than the those that engineer the change themselves. This investment strategy is named for the California gold rush in the 1840s and 1850s, where those that accumulated the greatest wealth were supposedly not the gold miners themselves, but those who provided them with the tools to dig.

The gold rush story may be apocryphal, but as a strategy, it has served investors relatively well through previous technological innovation cycles. During the internet boom, for example, the businesses that have endured and made money have been generally companies such as Apple, Microsoft and Alphabet that have provided cloud networks, software, mobile devices or tools to navigate the internet.

The companies that have endured during the internet boom such as Apple, Microsoft and Alphabet have provided cloud networks, software, mobile devices or tools to navigate the internet.

High-profile flops

The fate of the ‘gold miners’ has been more varied. There have been some success stories – Amazon and Meta, the company behind Facebook and Instagram, are clear examples – but many more failures, such as high-profile flops including pets.com, eToys.com and garden.com. These companies saw the early potential of the internet, but were either too early, too profligate, poorly focused or just badly managed.

There are sound reasons why the ‘picks and shovels’ approach works. When there is a major technology development, such as the internet or artificial intelligence, it takes time for use cases to emerge. In the case of the internet, it took the launch of smartphones for its potential to be fully realised. Thus investing early in specific use cases is inevitably more speculative, even if the long-term payout might be higher.

Will this approach work with artificial intelligence, the latest gold rush? Some analysts speculate that it could generate a productivity revolution. Goldman Sachs Research suggests that AI could drive a 7% increase in global GDP, equivalent to almost $7 trillion in today’s terms, and boost productivity growth by 1.5 percentage points over a 10-year period.

There are many potential uses for AI. It is already being deployed in areas such as medical diagnostics, inventory management and supply chains, in improving efficiency in agriculture, and there are many more potential fields to be explored, including manufacturing, finance, entertainment and retailing. Its impact is likely to be far-reaching and significant – but it is not yet clear where the greatest gains might be realised. No ‘killer app’ has emerged in which AI appears set to create something useful and powerful and from which investors could make outsized returns.

Opportunities in AI infrastructure

Against this backdrop, picks and shovels have definitely been the place to be, starting with the semiconductor manufacturers TSMC, Samsung and Nvidia, which designs the sophisticated chips required to power AI. While stressing that past performance is no indication of future results, it should be noted that Nvidia reported a 262% year-on-year increase in revenue in the first quarter of 2024. The sector also includes ASML, which produces the machines that manufacture semiconductors. All these companies have benefited from the rush to create AI infrastructure as companies seek to ensure they don’t get left behind.

Cloud computing providers have also been beneficiaries. AI runs on data, and the cloud providers facilitate the collection and analysis of that data. These companies have ploughed billions into ensuring they are competitive on AI, and Goldman Sachs predicts that the leading technology companies, including Microsoft, Alphabet, Meta and Amazon, could spend as much as $1 trillion in the coming years on artificial intelligence investments.

The problem for investors is that this is a story that has been comprehensively discovered by the market. The ‘magnificent seven’ tech groups – Amazon, Apple, Alphabet, Meta, Microsoft, Tesla and Nvidia – accounted for more than half of all gains in the US stock market in 2023, generating a combined 75.7% – nearly three times the S&P 500’s overall return of 24.2%.

They are also an increasingly large part of US stock market indices. The 10 largest companies in the S&P 500 accounted for 27% of the index’s total capitalisation at the end of 2023, nearly double the 14% proportion a decade earlier, according to Morgan Stanley.

In previous cycles, the pick-and-shovel manufacturers were the steady investments, with less excitement built into their prices – but this time they have become the high-profile ‘glory’ stocks.

Investors’ dilemma

This may sound like a vindication of the picks and shovels strategy, but it also holds the risk that it might not work as well this time round. In previous cycles, the pick-and-shovel manufacturers were the steady investments, with less excitement built into their prices – but this time they have become the high-profile ‘glory’ stocks. While they may still be strong companies with a promising trajectory for growth ahead, their value may already be reflected in their share price.

The dilemma for investors is whether to take their chances on the picks and shovels, where there is clear demand but prices look expensive, or hunt for the key use cases for AI, and the companies that are likely to be beneficiaries. Inevitably, this is a more speculative approach, but there will be big returns for the winners.

One trend emerging among fund managers is to target companies with large and sophisticated data sets from which AI could provide genuinely differentiated and ground-breaking insights. Examples include companies with rich legal data or research findings, which might ultimately prove to be the biggest beneficiaries in the next wave of AI.

The picks and shovels strategy is a sound approach to investing in new technology trends. However, for artificial intelligence, many investors have already had the same idea. Newcomers may need to find a new strategy to exploit the AI revolution as it develops.