News
Where Will Super Micro Computer Stock Be in 1 Year?
With shares down nearly 40% over the last 30 days alone, Super Micro Computer's (NASDAQ: SMCI) rocket-ship rally is imploding. Shares now trade at a level unseen since late January, undoing almost seven months of gains in the blink of an eye. The company is reeling after weaker-than-expected profitability. Let's explore what the next year could have in store.
Super Micro Computer's lackluster earnings
A great company doesn't always make a good investment because its valuation may price in unrealistic future expectations. This was one of the challenges Supermicro faced in its fourth-quarter earnings.
Net sales increased by 144% year over year to $5.3 billion, driven by soaring demand for the company's AI infrastructure. However, analysts were disappointed by the company's adjusted earnings per share (EPS), which came in at $6.25 compared to expectations of $8.07. While Supermicro's business is booming, profitability is under pressure because of lower gross margins, which fell from 17% to 11% year over year.
A somewhat shallow economic moat
Supermicro is known for turning Nvidia's (and other chipmakers') graphics processing units (GPUs) into ready-to-use computer servers. This niche has allowed it to piggyback off the soaring demand for AI chips as more data center clients race to build up their capacity.
However, Supermicro's "middleman" business model is quite vulnerable. For starters, the company relies on hugely expensive chip supplies from its partners. And it isn't well differentiated from other server makers like Dell Technologies or Hewlett Packard Enterprise, which operate a similar strategy. This competition could make it harder for Supermicro to pass costs to consumers or pad its margins by raising prices.
This is a very different situation from Supermicro's partner, Nvidia, which managed to increase its gross margin from 76% to 78.4% in its fiscal first quarter because it has locked clients into using its products through popular software solutions like CUDA (optimized for Nvidia hardware) and a relentless update cycle, which can keep its chips technologically ahead of the competition.
According to management, Supermicro's weaker-than-expected gross margins were mainly due to higher supply chain costs and tight inventory of key components. They expect this situation to be resolved by the end of fiscal 2025 as manufacturing partners ramp up production, but it highlights the ongoing challenge with the company's economic moat . And it could emerge again in the future.
What does the next year have in store for Supermicro?
The next 12 months are shaping up to be a make-or-break period for Supermicro. Storm clouds are gathering over the U.S. economy as unemployment ticks up and consumer spending begins to stall .
Analysts at JPMorgan predict a 35% likelihood of a U.S. recession by year's end, which could wreak havoc on the AI industry.
Typically , when macroeconomic conditions weaken , companies tighten their belts by cutting back speculative and unprofitable investments. Consumer-facing AI algorithms would likely fit into this category because they struggle to generate enough revenue or profits to justify the massive capital expenditures needed to run and train them. Supermicro could be particularly vulnerable to a downturn because its business already faces significant competition and margin pressure.
Is Supermicro stock undervalued?
With a forward price-to-earnings (P/E) multiple of just 14, Supermicro shares are shockingly cheap. For context, the S&P 500 has an average estimate of 22, while market leader Nvidia boasts 40. It's hard to see why a company growing at a triple-digit rate should trade at such a discount, especially if it manages to get its gross margins under control.
With that said, despite Supermicro's attractive discount, investors should be cautious about betting on any AI-related stock right now because of the speculative and unproven nature of the industry and the growing risk of a U.S. recession over the next 12 months.
Before you buy stock in Super Micro Computer, consider this: