Author: Rita

Trend Guide

Micron Technology's latest earnings report gave Wall Street a surprise: revenue for the May quarter reached $41 billion, a staggering year-on-year increase of 346%. But more astonishing than the numbers themselves is Micron's guidance for the August quarter, projected to reach $50 billion, a sequential growth of 21%. The company's gross margin hit 85% and operating margin 81%, both setting historical records.

The storage industry has averaged an operating margin of less than 20% over the past five years, but now leading companies can achieve over 80%. The Bank of America Securities analyst Simon Woo's team proposed a bold judgment in the Global Storage Technology Weekly on June 26: this is structural profit growth, and the industry supercycle may last until 2027, even extending to 2030.

Five Structural Changes Supporting Long Cycle Judgments

The first is that the chip shortage may continue until 2027 or even 2030. The demand for computing power driven by AI has fundamentally changed the demand curve for storage chips. The explosive demand for AI-related products such as HBM (High Bandwidth Memory), SOCAMM, LPDDR5, and GDDR7 has severely squeezed traditional DRAM capacity. This reflects a permanent shift in demand structure.

The second is that long-term agreements (LTA) are changing the industry’s rules of the game. Major manufacturers like Micron are increasingly signing long-term pricing agreements with tech giants and OEMs. This means that ASP (Average Selling Price) is no longer experiencing wild fluctuations as in the past but is locked at a higher level. Bank of America believes that storage chips are becoming like TSMC's logic foundry, characterized by high profits and low volatility. This shift is significant for investors, as the severe cyclical volatility is being smoothed out.

The third is that the threshold for building new fabs is extremely high. Construction costs have skyrocketed, local government regulations have tightened, and there is a strain on water and power supply. Even in 2028, capacity expansion will be very limited. Micron is building factories in various locations around the world (Boise/New York/Virginia in the U.S., Tongluo in Taiwan, Singapore, and Japan), but new capacity will still take years to materialize. This is a matter of physical constraints.

The fourth is the structural contraction of capacity. Upgrading existing old factories requires a large amount of clean room space, manufacturing cycles are extended, and the front-end and back-end equipment have larger sizes, with HBM's capacity squeeze on traditional DRAM being extremely high. The result is that even if capital expenditure doubles or more, the increase in usable capacity remains limited. This is a hard constraint on the supply side.

The fifth is that free cash flow is experiencing explosive growth. Despite capital expenditures being more than twice the levels in normal cycles, Micron and its peers’ free cash flow is still significantly increasing. This industry was previously known for "burning cash," but is now starting to generate its own cash. This transition provides long-term support for stock prices.

Data Validates Cycle Extension

Korean semiconductor exports (mainly storage chips) reached $25.5 billion in the first 20 days of June, a sequential increase of 16% and a year-on-year increase of 188%, maintaining triple-digit growth for five consecutive months. The DRAM spot price has risen for five consecutive weeks with a cumulative increase of 20%.

Bank of America raised its global DRAM/NAND sales forecast for 2026-2028 by 2-4%, but emphasized that the adjustment range is limited. The most dramatic price increases (in the first half of 2026) have basically been reflected in the data. The growth rate of ASP in the second half will slow down due to increased LTA, but the overall volume-price structure is healthier. Forecasts indicate that global DRAM+NAND revenue will reach $87.68 billion in 2026, soar to $121.22 billion in 2027, and $128.51 billion in 2028. This is derived from existing agreements and capacity constraints.

Target Differentiation: Not All Manufacturers Are Equal

Bank of America's industry coverage layout shows significant differentiation among the leaders. Samsung Electronics (with a market share of 41%) is expected to generate $254.2 billion in DRAM+NAND revenue in 2026, with an operating margin of 64%. SK Hynix has the strongest historical data, but Bank of America did not provide a clear forecast for 2026-2028. Micron (25% share) is expected to generate $130.6 billion in revenue in 2026, with an operating margin of 81%, and Bank of America has also raised Micron's ASP assumptions. Nanya Technology (2-3% share) mainly produces traditional/old DRAM and has almost no exposure to LTA, and Bank of America did not provide a positive forecast.

A warning signal is that NAND spot prices recently fell by 5%. While DRAM spots increased, NAND weakened, indicating that the AI benefits are not evenly distributed in the storage sector. The expected growth rate of NAND's ASP is forecasted to decline from +234% in 2026 to +8% in 2027, and then turn negative (-13.3%) in 2028. If AI demand's structural migration is more skewed towards HBM/DRAM rather than NAND, NAND manufacturers may face an earlier cyclical turning point.

Trend Perspective

Bank of America's most noteworthy assumption in this report is that "the chip shortage may continue until 2030." This is rarely seen in history. The previous strongest supercycle from 2017-2018 lasted only about 18 months. Bank of America's logic chain is that the storage demand increment brought by AI continues to expand, while the supply side is constrained by physical bottlenecks (factory approvals, electricity, clean rooms, equipment delivery).

However, there is an obvious weakness here: Bank of America assumes that LTAs will "lock" ASPs to prevent sharp declines, which holds only if major AI clients do not "change their mind." If large cloud providers shift to self-developed storage chips or reduce HBM procurement ratios in 2027-2028, the entire LTA framework will be shaken. The "structural change" narrative in the storage industry mostly ends with a return to cycles. Whether this time is really different remains to be seen. For investors focusing on U.S. and Hong Kong stocks related to storage, Bank of America’s perspective provides a highly confident mid-to-short-term framework, but the certainty of predictions beyond 2027 is rapidly declining.

Disclaimer

This article is a compilation and interpretation of third-party brokerage research reports by Trend Guide. The ratings, target prices, earnings forecasts, and related judgments quoted in the text are the views of the respective brokerage analysts and only represent the position of their institutions, not the views of Trend Guide, and do not constitute any investment advice.

The market has risks, and decisions need to be made independently. This article should not be used as a basis for buying or selling any securities.