
Key Points
- Micron's moat is a blend of scale economics (one of three scaled DRAM/NAND suppliers), process-technology parity-to-leadership (1γ DRAM and G9 NAND ramping as the highest-volume nodes in company history; next-gen nodes on track for 2H-CY27), and a first-mover position in differentiated form factors
- The 16 signed SCAs, take-or-pay, five-year, non-cancellable outside automotive, with price floors management states sit “well above peak quarterly margins in any past cycle” has structurally raised the earnings floor for the next downcycle.
- We reaffirm BUY on Micron with a price target of US$1,665, implying ~+40% upside to the indicative post-print level of ~USD 1,186.
- The quarter was a broad-based earnings beat. Results again blew past an already-high bar, while Q3 nearly doubled revenue to USD 41.5bn at 84.9% GM and USD 25.11 EPS, the fifth consecutive revenue record.
Executive Summary
Micron again delivered earnings ahead of both market consensus and our expectations (since our previous update on March 26).
We reaffirm BUY on Micron with a price target of US$1,665, implying ~+40% upside to the indicative post-print level of ~USD 1,186.
What has changed since our last update
Our March note framed Micron as an attractive tactical AI-memory trade with meaningful upside even as the stock sold off on capex and cycle-peak fears; one quarter on, the fundamentals are stronger and the valuation gap has closed. The share price rally was far beyond our initial expectations (our initial TP at USD 680 in March 26), and we reiterated our positive stance towards the counter in this report.
Read More: Micron Q2 26 Earnings: Stronger Than Ever, Yet Questioned
Figure 1: Micron’s share price and our previous update

Micron’s Q3 26 earnings summary
The quarter was a broad-based earnings beat.
Results again blew past an already-high bar, while Q3 nearly doubled revenue to USD 41.5bn at 84.9% GM and USD 25.11 EPS, the fifth consecutive revenue record. The FY27 capex concern that weighed on shares in March now appear increasingly consistent with a supply-constrained industry, rather than signalling excess capacity..
Management now guides capex above the mid-USD40 bil range, with over half allocated to construction, yet new greenfield capacity will only come online in CY28 and still falls short of demand, reinforcing expectations of prolonged supply tightness while free cash flow continues to improve.
Table 1: Micron’s Q3 26 earnings summary
|
Metric |
Q3 26 |
Q2 26 |
Q3 25 |
Q/Q |
Y/Y |
|
Revenue |
41.46 |
23.86 |
9.3 |
73.70% |
345.70% |
|
Gross Margin |
84.90% |
74.90% |
39.00% |
+1,000 bps |
+4,590 bps |
|
Operating Expenses |
1.52 |
1.42 |
1.13 |
6.80% |
34.00% |
|
Operating Income |
33.69 |
16.46 |
2.49 |
104.70% |
1252.70% |
|
Net Income |
28.86 |
14.02 |
2.18 |
105.80% |
1223.10% |
|
Diluted EPS |
$25.11 |
$12.20 |
$1.91 |
105.80% |
1214.70% |
Table 2: Revenue segmentation
|
Business unit (USD mil) |
FQ3-26 |
% of rev |
Q/Q |
Y/Y |
|
Cloud Memory (CMBU) |
13,769 |
33% |
78% |
307% |
|
Core Data Center (CDBU) |
11,524 |
28% |
103% |
653% |
|
Mobile & Client (MCBU) |
11,521 |
28% |
49% |
254% |
|
Auto & Embedded (AEBU) |
4,634 |
11% |
71% |
311% |
|
Total |
41,456 |
100% |
74% |
346% |
Source: Micron, Bloomberg, iFAST Compilation, Data as of 25 June 2026.
More importantly, Strategic Customer Agreements (SCAs) have evolved from concept to execution. Micron has signed 16 SCAs covering ~20% of DRAM and ~one-third of NAND bits, with 14 agreements securing ~USD100 bil of floor-price revenue and backed by USD 22 bil of customer commitments (~USD18 bil in cash). The non-cancellable, cash-backed structure meaningfully improves revenue visibility and underpins a higher through-cycle valuation multiple.
The newly announced Anthropic strategic agreement (22 Jun 2026) reinforces the thesis: it bundles memory/storage architecture design, a multi-year supply deal and a strategic investment in Anthropic's funding round, tying a frontier AI lab's infrastructure roadmap to Micron, providing another indication of memory's increasingly strategic role in AI infrastructure, though financially immaterial near-term. The tightness horizon has also extended, with guidance moving from "through CY27" to "beyond CY27" and the USD100 bil HBM TAM milestone pulled forward from CY28 to CY27.
Table 3: Micron’s Q4 26 guidance
|
Q4 26 guidance |
|
|
Revenue |
USD 49 bil-USD 51 bil |
|
Gross Margin |
~86% |
|
Operating Expenses |
~USD 1.65 bil |
|
Diluted EPS |
$30-$32 |
Source: Trendforce, iFAST Compilation, Data as of end-March 2026.
Management's central message changed in tone this quarter: tightness now persists “beyond calendar 2027,” with explicit acknowledgement that they “do not have line of sight” to when supply catches demand. The drivers are structural: slower bit growth per node, HBM's rising trade ratio (more wafers per bit), greenfield fabs that are slow and capital-heavy, and NAND cleanroom being redirected to DRAM. CY26 industry DRAM bit shipments are guided to grow low-to-mid-20s% and NAND ~20%.
Escape from brutal history
Memory economics have historically been highly cyclical. Fixed-cost-heavy fabs, commodity pricing, and a feast-or-famine margin profile that swung from -7.7% gross margin (FY23 trough) to 40.9% (FY25) to a record 84.9% now. The SCA framework is an attempt to break this pattern. By converting almost half of revenue to multi-year take-or-pay contracts with price floors and ceilings, Micron is trading some upside (ceiling at current CQ2 prices on the largest agreements) for a much higher floor, management states the floor enables gross margins “well above our peak quarterly margins in any past cycle.”
If adopted more broadly, SCAs could represent one of the more meaningful structural changes to memory industry economics in recent years.
Our 10.5x fair P/E sits above the ~6–8x that memory peak earnings have historically commanded, precisely to reflect this.
Bigger cake, bigger competition, but not now
Micron's moat is a blend of scale economics (one of three scaled DRAM/NAND suppliers), process-technology parity-to-leadership (1γ DRAM and G9 NAND ramping as the highest-volume nodes in company history; next-gen nodes on track for 2H-CY27), and a first-mover position in differentiated form factors: SOCAM/LPDRAM for the data centre, where Micron was sole-sourced for a period, and 245TB data-center SSDs / Gen-6 leadership in NAND. The HBM franchise is the swing factor: HBM4 12-high is ramping ~2x faster than HBM3E, with >US$1bn already shipped and a stated goal of holding HBM share consistent with overall DRAM share.
On the IP dimension, management cites ~65,000 patents and a multi-decade enforcement record. The competitive threat from China (such as CXMT in DRAM, YMTC in NAND) remains real but, per management, their output is overwhelmingly sold within China and has not yet shown up as meaningful competition in Micron's leading-edge, high-performance segments.
We treat Chinese supply as the primary structural bear risk for CY28+, not a CY26–27 issue.
Table 4: NAND market share in Q1 26
|
Supplier |
Share |
|
Samsung |
~29% |
|
SK Hynix (incl. Solidigm) |
~22% |
|
Kioxia |
~13–14% |
|
Micron |
~14% |
|
SanDisk |
~14% |
|
YMTC |
~13–14% |
Source: Trendforce, iFAST Compilation, Data as of end-March 2026.
Table 5: DRAM market share in Q1 26
|
Supplier |
Share |
|
Samsung |
~ 38.60% |
|
SK Hynix |
~ 28.80% |
|
Micron |
~ 22.40% |
|
Others (CXMT, Nanya, etc.) |
~ 10% |
Source: Trendforce, iFAST Compilation, Data as of end-March 2026.
Game changer – SCAs strengthen earnings visibility
The 16 signed SCAs, take-or-pay, five-year, non-cancellable outside automotive, with price floors management states sit “well above peak quarterly margins in any past cycle” has structurally raised the earnings floor for the next downcycle.
With USD 22 bil of deposits/commitments (~USD18 bil cash) already in hand, ~40% of revenue heading to fixed-or-ceiling pricing, and tightness now guided “beyond CY27,” we model FY28E revenue of USD 255 bil and EPS of USD 158.5, reflecting our expectations for sustained demand and tighter industry supply. On those numbers the stock trades at just ~7.5x FY28E EPS; our 10.5x fair multiple (a modest premium to memory’s historical peak band, reflecting the SCA floor) supports our target price of USD1,665, implying around 40% upside.
Key Caveats
Nevertheless, customer concentration is increasing as hyperscalers account for a growing share of AI memory demand. The 16 signed SCAs include four “very large” and three medium customers (the rest automotive), reflecting the industry's increasing reliance on a small number of large AI infrastructure buyers.
On the supply side, Micron continues to expand manufacturing capacity to support future demand. he company recently concluded a multi-year EUV agreement with ASML to support its 1δ DRAM node and beyond, while continuing to expand manufacturing capacity across Idaho, New York, Taiwan, Japan and Singapore.
Taken together, these investments should strengthen Micron's long-term production capacity. Even so, management expects demand to continue outpacing supply beyond CY27, suggesting capacity additions are unlikely to materially alter the current supply-demand balance over the medium term.
Figure 2: Our revenue forecast in FY28E

BUY, TP at USD 1,665 with ~40% upside potential
Memory has historically commanded only ~6–8x on peak earnings, reflecting the certainty of a downcycle. We assign a fair P/E of 10.5x, a premium to that historical band to capture the structural floor that SCAs and AI scarcity introduce.
Coupled with our forecast EPS in FY28 of 158.5, we have computed a target price of Micron at USD 1,665, implying a 40% upside potential compared to today’s price of USD 1,186.
We reaffirm our BUY recommendation and target price of USD1,665, supported by improving earnings visibility, structural AI demand and a more constructive industry backdrop.
Given the stock's strong rally and the inherently cyclical nature of the memory industry, we believe Micron is better suited as a tactical overweight in a diversified portfolio than as a standalone core holding.
Table 6: Micron’s Valuation
|
FY25A |
FY26E |
FY27E |
FY28E |
|
|
EPS |
8.29 |
70.1 |
135.87 |
158.51 |
|
y/y |
745.6% |
93.8% |
16.7% |
|
|
Implied PE |
143.1 |
16.9 |
8.7 |
7.5 |
|
Current Price |
1186 |
|||
|
Upside Potential (based on fair PE of 10.5x) |
40% |
|||
|
Target Price |
1665 |
Table 7: Our forecast
|
USD mil |
FY23A |
FY24A |
FY25A |
FY26E |
FY27E |
FY28E |
|
Revenue |
15,540 |
25,111 |
37,378 |
127,576 |
225,000 |
255,000 |
|
DRAM |
10,978 |
17,603 |
28,578 |
89,465 |
168,000 |
193,000 |
|
NAND |
4,206 |
7,227 |
8,503 |
24,952 |
56,500 |
61,500 |
|
Gross margin % |
-7.70% |
23.70% |
40.90% |
77.80% |
84.50% |
84.00% |
|
Op. income |
-4,819 |
1,935 |
10,846 |
~91,800 |
183,125 |
206,000 |
|
Op. margin % |
-31.00% |
7.70% |
29.00% |
~72% |
81.40% |
80.80% |
|
Non-GAAP EPS |
-4.45 |
1.3 |
8.29 |
70.1 |
135.87 |
158.51 |
Declaration:
For specific disclosure, at the time of publication of this report, IFPL (via its connected and associated entities) and the analyst who produced this report hold a NIL position in the abovementioned securities.
This research report was prepared with the assistance of artificial intelligence (AI) tools. iFAST Financial Pte Ltd does not rely exclusively on AI for content generation; the content of this report – including all investment theses, ratings, price targets and conclusions – has been independently reviewed and verified by the research analyst(s) to ensure accuracy and professional integrity.
