The stock market can be volatile and unpredictable. That's why the most successful investors usually buy stocks with a long-term time horizon -- thinking about how a business will be doing five or even 10 years into the future instead of next week. This strategy helps smooth out the short-term noise and lets a company's real fundamentals shine through.
For Micron Technology (MU 3.82%), the near term is great. The memory specialist is riding a wave of huge demand for its chips as tech giants rush to build generative AI data centers. That said, the future will depend on Micron's ability to translate a temporary windfall into lasting shareholder value. Let's dig deeper to see where the stock could be by 2030.
NASDAQ: MU
Key Data Points
The magic of picks and shovels
Over the last few years, the generative AI boom has borne some striking resemblances to the California gold rush in the mid-19th century. Back then, vendors who sold enabling infrastructure like pickaxes, shovels, and jeans ended up making more consistent profits than the miners who actually panned for gold. A similar situation is playing out today.
Large language models (LLMs) like OpenAI's ChatGPT and Anthropic's Claude have turned into money pits. And it is unclear when (or even if) they will become profitable because of high energy consumption and training costs. That said, despite the immediate losses, these companies continue to bet on expansion because they want to maintain their market share as the technology improves and becomes more mainstream.
Micron operates on a much safer side of the generative AI opportunity, providing the high-bandwidth (HBM) memory needed to help these algorithms function. And over the coming years, it won't matter which consumer-facing algorithms succeed or fail. Micron can still win by supplying the hardware that all of them need to function.
The near term is fantastic for Micron
Over time, the priorities of the AI industry have shifted from model training to model inference, which is how the algorithm makes deductions based on information it has already been trained on. This trend has significantly boosted demand for memory -- specifically HBM, which tends to be more difficult to produce compared to other types.
With so much production capacity being diverted to HBM, bottlenecks are emerging in other parts of the memory industry, with some experts predicting that the shortages could last until 2027. When demand outstrips supply, producers like Micron can raise prices and fully utilize their production capacity, which tends to improve efficiency. These tailwinds are already showing up in the company's earnings.
Fiscal first-quarter revenue roughly doubled to $5.3 billion, while gross margins soared from 59% to 66%. Investors should expect Micron to enjoy these positive trends while the memory shortage continues.
Image source: Getty Images.
What comes next for Micron?
We don't have to make assumptions about what will come next for Micron because management has already made its plans crystal clear. The company plans to invest an eye-popping $200 billion into expanding its HBM production capacity, with the first wafers expected in 2027.
As long as demand continues to outstrip supply, this new production capacity will almost certainly help Micron grow its revenue. And it may also give the company economies of scale advantages over its rivals, which can lead to more competitive pricing and/or better margins. That said, these investments are not without risks and opportunity costs.
For a company with a market cap of $416 billion, $200 billion is a colossal amount of money. This spending will soak up capital that could have otherwise gone to shareholders in the form of dividends or buybacks, and it runs the risk of creating another supply glut in the future if generative AI-related demand doesn't continue to grow as fast as expected.
With a forward price-to-earnings (P/E) ratio of just 11.5, Micron's current valuation seems to already price in this uncertainty. And the stock remains a compelling long-term buy. That said, investors should note that there is a non-trivial risk of the expansion strategy backfiring.
