Bank of America projects sustained AI memory demand, forecasting Micron’s earnings to grow at a 34% annual rate driven by HBM and advanced DRAM scaling.

Bank of America projects Micron could reach $236 in earnings per share (EPS) by fiscal 2030, representing a 34% annual earnings growth rate that the market is currently ignoring.

Micron’s DRAM revenue surged 343% year over year, driven by high-bandwidth memory (HBM), which requires three times the wafer capacity per bit and continues to constrain supply.

Trading at roughly six times forward earnings, Micron appears undervalued if AI demand extends the memory boom beyond what skeptical investors anticipate.

The most widely read finance newsletter on Substack is not published by a bank; it is Doomberg, where over 383,000 readers receive energy and macro analysis overlooked by the mainstream press. Readers of 24/7 Wall St. can save 17% on their first year.

The semiconductor market has spent decades teaching investors a consistent lesson: memory booms inevitably turn into busts. As supply catches up with demand, pricing declines, margins collapse, and previously record earnings quickly resemble a mirage.

This historical precedent explains why investors continue to treat Micron Technology (NASDAQ: MU) as a cyclical stock, despite AI pushing its financial results into uncharted territory. In fiscal Q3 2026, Micron reported $41.46 billion in revenue and an 85% gross margin, compared to just 38% a year earlier.

Bank of America is now urging investors to consider whether artificial intelligence has fundamentally broken that historical cycle.

BofA Global Research outlines a dramatically different trajectory for the company. Under its “SanDisk-like” scenario, Micron’s sales would reach $377.3 billion by fiscal 2030, significantly above the consensus estimate of $280.5 billion.

This projection implies a 30.7% compound annual growth rate (CAGR) for sales and a 34.1% CAGR for EPS through fiscal 2030. Yet the market currently values Micron at roughly six times forward earnings based on the consensus fiscal 2027 EPS estimate of $151.37, effectively pricing in the end of the memory boom long before Micron’s growth trajectory stabilizes. Bank of America considers this outlook misplaced.

Written by industry insiders who answer to no one, Doomberg’s editorial team brings extensive backgrounds in heavy industry, private equity, and the hard sciences. Operating without advertisers or institutional affiliations, the publication delivers lateral-thinking analysis on energy, finance, and geopolitics that diverges sharply from consensus financial media. A discounted subscription rate has been reserved exclusively for 24/7 Wall St. readers and is not available on Doomberg’s public website.

Bank of America is placing a significant bet that the AI revolution has fundamentally rewritten the rules of the semiconductor industry, directly challenging decades of established market history.

There is substantial merit to the bullish thesis. Micron’s fiscal Q3 DRAM revenue surged 343% year over year to $31.3 billion, while NAND revenue climbed 361% to $9.9 billion. Consolidated gross margins reached 84.9%, with fiscal Q4 guidance projected near 86%.

High-bandwidth memory is central to this structural shift. Micron notes that HBM requires more than three times the wafer capacity per bit compared to conventional DRAM, and supply remains tightly allocated. While hyperscalers demand greater volumes, manufacturers cannot rapidly flood the market. HBM4 is already shipping in volume, with HBM4E production slated for 2027.

Additional tailwinds include enterprise SSDs capturing a growing share of the NAND market as AI data centers generate expanding data volumes, alongside next-generation AI inference potentially requiring entirely new memory architectures. Earlier this month, SK hynix (NASDAQ: SKHY) and SanDisk (NASDAQ: SNDK) unveiled specifications for High Bandwidth Flash (HBF)—a category positioned between HBM and SSDs, capable of delivering up to 512 GB of capacity and 3 TB/s of bandwidth. While this development does not necessarily disadvantage Micron, it underscores the rapid evolution of memory architecture and serves as a reminder that current market leaders cannot be assumed to dominate every emerging category.

The broader AI thesis is compelling, but the specific Micron forecast demands a significant leap of faith. While an 80% gross margin is plausible in the current environment, projecting it as a structural baseline through 2030 stretches credibility. Historically, memory has been one of the most cyclical segments within semiconductors, with normalized-cycle margins typically ranging between 30% and 40%. Micron’s current 85% margin reflects an exceptionally tight supply-demand dynamic. The critical question remains how profitability will adjust once competitors expand capacity.

SK hynix remains a formidable competitor, while Samsung and potential Chinese suppliers are simultaneously ramping up production. If competitive pressures enforce pricing discipline rather than allowing Micron to sustain 80% margins indefinitely, earnings may converge closer to consensus estimates. This is precisely why Bank of America’s $236.16 EPS projection warrants skepticism, even if the overarching AI-driven thesis proves correct. Importantly, investors do not need to accept BofA’s most aggressive assumptions to find the investment case compelling: Micron is already generating record revenues, sustaining 85% margins, and producing $25.39 billion in quarterly operating cash flow.

Bank of America correctly identifies the directional shift but appears overly aggressive regarding the endpoint. AI is fundamentally altering memory economics by redirecting demand toward HBM, advanced DRAM, enterprise SSDs, and potentially HBF, while extended manufacturing lead times may slow supply responses relative to historical cycles. However, achieving $236 in EPS by 2030 would require Micron to transform structurally, breaking away from the cyclical profile investors have recognized for decades.

Nevertheless, the investment case does not hinge on assuming an 80% gross margin will persist indefinitely. Trading at approximately six times forward earnings, the market appears to be pricing in a swift return to historically weaker economics, potentially underestimating the duration of this expansion. Bank of America’s analysis should therefore be viewed as the bull case rather than the base case.

Through a dedicated reader link, 24/7 Wall St. subscribers can secure their first year of Doomberg for $332—a 17% discount off the standard $400 price, equating to approximately $28 per month. The subscription includes six to eight deeply researched monthly articles covering energy, finance, and geopolitics, access to the complete archive, and entry to one of the most substantive comment sections in financial media, where engineers, executives, and fund managers provide expert context beneath each publication. Renewal occurs at the standard rate after the initial term, with cancellation permitted at any time. If today’s article touched on energy or the macro picture, odds are the green chicken has already gone three layers deeper—accessible via the discounted rate.

For questions or corrections, contact editorial@247wallst.com.