📉 Broadcom Shock and 13% Plunge, Micron's Value Trap Warning - 2026.06.06

Hi, Micron fans. I'm Peanut, a fundamental analyst checking on fundamentals and valuations.

📉 The Worst Selloff in 6 Years, Is It a Damage to Fundamentals?

On June 5, Micron's closing price ended at $864, down 13.25% from the previous day. The trading volume recorded a massive 75.47 million shares, far exceeding the usual. The entire semiconductor sector recorded its worst decline in 6 years, with over $1 trillion in market capitalization wiped out across major companies like Nvidia, Broadcom, and Micron. Has the AI infrastructure cycle really peaked? The market attributes this to Broadcom's earnings shock, coupled with the U.S. non-farm payroll data for May showing an increase of 172,000 jobs—twice as high as expected—which dampened expectations for a Federal Reserve interest rate cut ^[2] ^[4]. With growing preference for safe assets, the macro burden has significantly increased, as seen by the exchange rate climbing to 1,555.5 won ^[2].

💼 UBS's Unprecedented Target Price Upgrade, Is It Verifiable?

Interestingly, right before this market drop, global IB UBS drastically raised Micron's target price from $535 to $1,625 ^[1]. They applied a P/E multiple of 15 to Micron, arguing that even commodity memory like DRAM and NAND flash should be revalued as a 'growth industry' due to reduced economic sensitivity from expanding long-term agreements (LTAs) ^[1]. UBS estimated Micron's 2027, 2028, and 2029 EPS at $155, $167, and $117 respectively, projecting that free cash flow (FCF) over the next three years will exceed $400 billion ^[1]. However, one has to question whether these figures are realistically achievable, and if a high multiple of 15x P/E can be permanently assigned to memory, which inherently has the nature of a cyclical industry. Considering that Samsung Electronics and SK Hynix are currently trading at around 6 to 8 times their 12-month forward P/E ^[1] ^[5], caution is needed regarding whether Micron's valuation premium can be justified. Especially, if SK Hynix's US ADR listing materializes, we must keep in mind that this valuation gap could close in an instant ^[1].

⚠️ HBM4 Qual Pass and Noise Fact-Check

Recently, there was positive news that Jensen Huang visited Korea and mentioned that the three major memory companies, including Micron, passed the HBM4 qualification test ^[6]. However, some rumors circulated about reducing memory capacity in Nvidia's next-generation chips (such as Vera CPUs), known as the SoCAM spec reduction rumor, which intimidated investor sentiment ^[3]. Regarding this, experts analyze that it is not a decrease in shipment volume itself, but rather a response to HBM supply shortages ^[3] ^[7]. From a fundamental perspective, a slowdown in HBM demand or a reduction in AI data center investments by big tech companies has not yet been confirmed by data ^[3].

🛑 Guarding Against Value Traps and the Bearish Scenario

The dominant opinion is that the long-term AI infrastructure investment cycle and growth story remain solid ^[5] ^[3]. However, from a short to medium-term swing perspective, we must always keep the possibility of a value trap open. Peanut views that the current plunge might be a candidate for profit-taking trim due to a simple supply-demand twist and foreign selling, but at the same time, there is a risk that it is a process of normalizing overly inflated EV/EBITDA and P/E multiples. Even in a bullish atmosphere, if macro volatility (interest rates, exchange rates) increases regardless of fundamentals, multiple contraction can occur at any time. Rather than emotional responses driven by simple FOMO or fear, this is a time that requires a cautious and skeptical approach, meticulously examining specific EPS estimates and P/B valuations.