The Fickle Allure of Memory Stocks: Why Investors Shouldn’t Get Too Caught Up in AI Hype

The memory industry has been on a wild ride, thanks largely to the launch of ChatGPT and the subsequent surge in demand for high-bandwidth memory (HBM) chips. This boom-and-bust cycle is nothing new, but investors are getting swept up in the excitement, ignoring the lessons of history.

Executives at major players like Samsung and SK Hynix have touted the industry’s newfound stability, arguing that AI has disrupted traditional cyclicality of memory demand. However, this narrative overlooks fundamental dynamics of supply and demand in the market. The fact remains that production costs are still linked to commodity prices, making the industry vulnerable to fluctuations in global economic trends.

The introduction of new compression methods like TurboQuant is a case in point. While it has the potential to significantly reduce memory requirements for large language models, its impact on demand is far from certain. Deutsche Bank notes that investors should be prepared for continuous disruption in the AI landscape, which could have significant implications for the memory sector.

Many analysts remain bullish on Samsung and SK Hynix, estimating double-digit gains in their share prices over the next 12 months. However, this optimism is largely based on current momentum rather than fundamental changes in the industry’s underlying dynamics.

The South Korean concentration risk is another factor investors should be wary of. The country’s Kospi index has been heavily influenced by the performance of Samsung and SK Hynix stocks, which together comprise over 50% of the entire index. While some banks remain optimistic about their prospects, others advise clients to take profits and rotate into more diversified portfolios.

The memory industry’s history is characterized by enormous ups and downs. William de Gale from BlueBox Asset Management notes that “a leopard does not often change its spots.” AI has undoubtedly disrupted traditional markets, but it remains unclear whether this will lead to a structural shift in demand or simply create new challenges for investors.

What’s clear is that the memory sector remains vulnerable to fluctuations in global economic trends and supply chain disruptions. Andrew Lapping from Ranmore Fund Management cautions that investors should be cautious when investing in an industry with historically average returns on capital priced to make very high returns in future.

The next few months will be crucial in determining whether the memory industry’s current momentum can be sustained or if we’re headed for another bust. One thing is certain: investors would do well to keep a close eye on developments and not get too caught up in the AI hype.

In the short term, production costs are likely to continue being driven by commodity prices, making it challenging for manufacturers to maintain profit margins. Any disruption to global supply chains or changes in AI demand could have significant implications for memory stocks.

Investors would do well to remember that history is a useful guide when navigating complex markets. The memory industry’s boom-and-bust cycle is far from over, and it’s essential to approach this sector with caution rather than complacency. By keeping a level head and being prepared for the unexpected, investors can avoid getting caught off guard by another downturn in the market.

The allure of AI-driven growth is undeniable, but it’s crucial not to get too carried away by excitement. Jon Cunliffe from JM Finn notes that there’s scope for production to increase meaningfully over the next three years, easing supply constraints – but only if AI demand grows at a more normal pace.

While AI has undoubtedly disrupted traditional markets, it remains unclear whether this will lead to a structural shift in demand or simply create new challenges for investors. The memory industry’s history is characterized by enormous ups and downs, and investors would do well to keep a close eye on developments rather than getting too caught up in the AI hype.