The GM Deal and the Dividend Are Real, but Burry’s Short Casts the Longest Shadow on Micron

On the surface, Micron Technology has rarely looked healthier. The chipmaker signed its 16th long-term supply agreement on July 1 — this time with General Motors, locking in memory and storage platforms for automotive production. Three days earlier, the board declared a quarterly dividend of $0.15 per share, payable on July 21 to shareholders of record as of July 6. The numbers behind the business remain staggering: third-quarter revenue surged 345% year over year to $41.4 billion, non-GAAP earnings per share ballooned more than 1,200% to $25.11, and the stock has gained over 790% in the past twelve months.

Yet the share price is caught in a vise. Michael Burry, the investor who famously shorted the housing market in 2008, took a short position against Micron at $1,051.87 per share. He published his bet on Substack, calling the company a “capital destroyer” trapped in a “psychological bubble.” Since that entry, the stock has fallen 8.79% over seven trading sessions, closing Friday at €912.00 — equivalent to about $1,011 at current exchange rates. The slide has been sharp and fast enough to drag Micron 17.38% below its 52-week high of €1,103.80, reached on June 25.

Burry’s critique is rooted in history. He points to Micron’s record — 34 drawdowns of more than 30% over 42 years — and argues that the pure-play DRAM and NAND business is structurally prone to boom-and-bust cycles. His numbers are precise and damning: median return on capital is around 4%, return on equity roughly 7%, and free cash flow has been negative about 48% of the time. Both returns fall well below double-digit cost of capital. Even after the recent pullback, Burry notes, the stock still trades further above its 200-day moving average than at any point since 1984.

The bearish case doesn’t stop with Burry. On July 1, the Philadelphia Semiconductor Index slumped after reports emerged that Meta Platforms is looking to sell off excess data-center capacity. For investors already nervous about overheated AI infrastructure valuations, the news was a trigger. Defensive rotation hit high-beta names hard, and Micron — with annualized 30-day volatility of 116% and an RSI of 52.2 — was squarely in the crosshairs. Analysts now wonder if the mild deceleration in DRAM and NAND contract pricing (forecast at 13–18% and 10–15% quarter-over-quarter, respectively) is a natural moderation or the leading edge of a cyclical top.

That bearish narrative, however, has to coexist with a very different set of facts. The global memory market is on track to exceed $1 trillion in sales this year, driven by AI data centers that are expected to consume roughly 70% of all high-bandwidth DRAM output in 2026. Nvidia CEO Jensen Huang has repeatedly said memory chip demand will outstrip supply for several more years. And while the ex-dividend date on July 6 is a minor technical event against a $1,000-plus stock, the fact that Micron continues to pay a dividend at all — even a small one — signals management’s confidence in cash generation amid the capital-intensive cycle.

Should investors sell immediately? Or is it worth buying Micron?

The GM deal adds another layer of ballast. Automotive is becoming a genuine second revenue pillar, insulating Micron from total dependence on a handful of hyperscaler customers. The agreement is the 16th major long-term pact disclosed in the company’s most recent quarterly report, underscoring a strategy of locking in demand across multiple verticals.

Bulls argue that the current valuation — elevated though it is against historical averages — looks reasonable on forward earnings estimates for the coming quarters. They see a structural shift, not a cyclical peak. The stock’s 50-day moving average of €762.26 sits nearly 20% below Friday’s close, while the 200-day average of €391.42 is 133% below it. Those gaps are extreme by any measure, but they reflect a rate of earnings improvement that has wiped out traditional valuation anchors.

In the end, Micron is a stock caught between two incompatible stories. On one hand, there is Burry’s historical template: a capital-intensive, commodity-like business prone to violent drawdowns, now trading at a peak premium that has historically preceded collapse. On the other hand, there is a thriving AI-driven supercycle backed by real, measurable demand — and concrete agreements with automakers, cloud providers and chip customers that run for years.

The dividend date will come and go. The GM deal is signed. What matters next is whether the market will trust the demand story or the capacity-gloom narrative. The next few quarters of memory pricing will settle the argument — and Burry, for his part, is already betting on the bust.