The Zhitong Finance App learned that after a three-day roadshow with Seagate (STX.US) Chief Financial Officer Gianluca Romano and Investor Relations Head Shanye Hudson, the team wrote the conclusion quite directly: in the past three months, Seagate's stock price fell 43%, retreating 59% from the 52-week high of $1,145, while the team's earnings forecast for FY27/FY28 per share increased 25% — 50% during the same period . The current price is only 6.1 times its FY28 EPS and 5.2 times the 2028 EPS in natural year 2028, which is less than the lowest point in the history of the hard drive industry (8 times). Damo maintains Seagate Technology's “gain” rating, with a target price of 1,148 US dollars.
The report classified the reasons into two categories: rising interest rates and capital expenditure financing issues underpinned valuation compression, and the market's doubts about “whether the hard drive industry is still disciplined” in the past month have widened the gap between Seagate and Western Digital and memory, semiconductor equipment, computing power, and optical modules — the hard drive stock outperformed the above four categories by as much as 5 percentage points in October, even though the performance itself was strong.
Last week, Seagate Technology's stock price plummeted. The trigger for the sell-off was that after the midterm elections, the market was worried about the tightening of data center approvals. The deeper level was doubt about how long the storage price increase would last. The report was released just after the second day of sharp decline, which is tantamount to giving “preferred stocks” against the market.
Five disputes, answered one by one
First, near-line shipments in the September quarter only increased 2% month-on-month. Has demand peaked? The report is due to two things: Seagate allocates more supply to enterprise OEM customers, and such customers have historically purchased less single-drive capacity than cloud service providers; at the same time, the company prepared 40TB HAMR hard drives in advance and waited for certification by large cloud customers to be completed after this quarter. Management clearly denied that there were technical barriers to land, electricity, plant or HAMR certification, and stated that the deceleration was entirely a matter of product structure and pace. The near-term EB will accelerate again in the December quarter — the company's $4.8 billion quarterly guidance implied a month-on-month increase of about $600 million, and this time the increase came more from 40TB product volume rather than simply price increases.
Second, is the industry returning to the old path of grabbing shares? The answer to the report is no. Seagate has no plans to increase single-plate production capacity, and capital expenditure has remained at 4% to 6% of revenue. This has been the case for many years; the increase in annual shipping capacity of about 25% was mainly achieved by increasing surface density (that is, migration to higher capacity HAMR products), and physical shipments remained roughly the same. Management believes that competitors are also focusing on technology-driven capacity growth. Low-end vendors are constrained by lack of vertical integration and production capacity from key suppliers. The technology gap is more than 3-4 years, and there have been no cases of price grabbing demand. Regarding the “Toshiba Production Expansion” report a month ago, the report called it a misleading headline.
Third, will component price increases eat up gross profit? The cost of in-house manufactured magnetic heads and media continues to drop, and increasing the capacity from 30TB to 40TB single disk can reduce the cost per TB by 25%; however, the prices of outsourced DRAM, NAND, SoC, controllers, and structural components are rising, and low-price inventory has already been used up, so this year's cost reduction curve will become narrower. The path given by management was “rise first, then stabilize, then fall back”, and no timetable was given. The point is that pricing has taken into account the cost inflection point: the near-term price increase per TB in the September quarter, and the incremental profit margin was higher than in the June quarter. The report therefore determined that the price increase in annual terms would accelerate throughout the 2027 calendar year and continue until 2028 — even if NAND prices fall back.
Fourth, are there any problems with HAMR certification? No. Mozaic 4 has already been deployed to the two largest cloud customers. The certification of other large cloud vendors and enterprise OEMs is progressing according to the regular 6-9 month cycle, and several will be completed this quarter. The company expects HAMR to account for 50% of near-line shipping capacity by the end of the quarter, with 40TB products accounting for more than half of the HAMR structure. Further, HAMR certification for 5TB/single disk will commence at the end of the 2027 calendar year, corresponding to the 50TB single disk being shipped in the second half of 2028; the laboratory has achieved 8TB/single disk, and it usually takes 3-4 years from laboratory to mass production.
Fifth, when will repurchases be accelerated and how many can I buy back? Over the past 12 months, Seagate reduced its total debt from $5 billion to $2.8 billion and settled all previously diluted convertible bonds. Management said the balance sheet repair phase is over, the focus is shifting to shareholder returns, and “opportunistic” buybacks can be initiated outside of regular plans. The report estimates that FY27 buybacks about 5.6 billion US dollars, FY28 up to 14 billion US dollars, and FY29 about 16 billion US dollars, which means that about 30% of the share capital can be repurchased within 2.5 years.
Three misinterpreted facts
According to the report, the valuation compression is due to the market's triple misunderstanding of the hard drive industry.
First, a hard drive is not an ordinary component, but a bottleneck infrastructure in a data center. Using historical data, the report shows that the correlation between Seagate's stock price and near-line shipping capacity is far higher than the correlation with price; when cloud demand, high-capacity product penetration, and data-intensive loads advance simultaneously, near-line capacity growth itself is a leading indicator of stock prices. AI adds a layer of data generation on top of traditional cloud workloads, strengthening the positioning of hard drives as “critical infrastructure” rather than “components that follow server shipments.”
Second, the periodicity of hard drives is weaker than memory. The report compared the 2023 to 2027 calendar year: Most of the increase in NAND and DRAM came from price, and the increase in hard drive shipping capacity was 4 times the price increase; once the price factor was removed, the fluctuation of hard drives was significantly less. Capital intensity can also explain the problem: Seagate and Western Digital's capital expenditure accounts for only 3.4% and 4.3% of revenue, and depreciation and amortization accounts for 5.3% and 8.2% of sales costs; for comparison, Micron is 36.2% and 38.9%, SK Hynix 25.4% and 38.0%, and Kioxia 21.9% and 27.2%. The report emphasizes that this is not a judgment on the memory industry itself; it only explains why hard drives should enjoy higher valuations than memory.
Third, surface density innovation changed the economics of this business. From 2012 to 2022, single-drive capacity growth basically stagnated, and PMR technology hit the physical limit; HAMR pulled this curve back up, while the revenue structure of hard drives quickly shifted to data centers, which currently account for 80% to 90% of the industry's revenue. Product economics improvements, structural upgrades, supply discipline, and lower capital intensity are compounded to support higher structural profit margins rather than a return to historical cyclical averages. The report underwent a stress test under pessimistic assumptions: even if the price of solid-state drives were reduced by 5% to 6% a year until 2030, and the price fell by another 80%, the operating profit margin for hard drives was still over 25%.
The valuation range, and another set of answers given by the market
Damo gave a target price of $1,148 for 15.0 times FY28 EPS of $76.64, which is 1 multiple point higher than the peak multiple of the previous industry upward cycle (2021-2022); $1,580 corresponds to 15.0 times EPS and $105.31 for a bull market; and $522 for a bear market, corresponds to 9.0 times and EPS of $58.00. According to the report, the long-term operating margin implied by the market's current pricing is less than 30%, and for this level to be established, solid-state drive prices need to drop more than 80% from the level of the third quarter of 2026 — according to the report, this is already close to “treating the bottom of the cycle as the norm.”
It is worth mentioning the implied probability of options included in the report (based on November 5 volatility data): the probability that the stock price will fall below the bear price of $522 in the next 12 months is about 65.1%, the probability of breaking through the target price of $1,148 is about 0%, and the probability of breaking through the bull market price of $1,580 is also about 0%. In other words, while giving +146% implied space, the options market hardly gives this target price any probabilistic weight. This data is tied with “increased holdings+preferred shares,” which explains more than any adjective how big the current differences are.
Sellers are not pessimistic as a whole: under Refinitiv's perspective, 89% of analysts covering the stock increased their holdings, 11% were neutral, and had no loss. The average target price was $708.21, up to $1,588. Morgan Stanley's target price is about 60% higher than the average.
Support and rebuttal: hard data on both sides
Let's take a look at the season the company just handed over. Revenue for the first quarter of fiscal year 2027 (ending October 2) was $4.233 billion, up 61% year over year and 17% month over month; non-GAAP gross margin of 57.8% (up 17.7 percentage points year over year), non-GAAP operating margin of 50.7%, and non-GAAP earnings per share of $7.68 billion, all record highs; operating cash flow of $1.7 billion and free cash flow of $1.48 billion also set records; debt was repaid during the quarter, and total debt fell to $2.35 billion, and a further $7.5 billion to $1.6 billion was returned after the quarter, and net leverage was reduced to 0.2 times. Data center revenue was $3.45 billion, up 63% year over year, accounting for 81% of total revenue; 220EB was shipped during the quarter, of which 198EB was near line, up 24% year over year. The second fiscal quarter led to revenue of $4.8 billion ± 100 million, non-GAAP earnings per share of $9.20 ± 0.20, and management expected an operating margin close to 55%. The company also said that the vast majority of near-line data center supply has been distributed through long-term agreements until the end of the 2028 calendar year, and some cloud customer agreements have been extended to 2030.
Let's look at the cards in the opposite hand. The first one is the price increase itself: According to TrendForce, contract prices for general DRAM rose 58% to 63% month-on-month in the second quarter of 2026, narrowed to 13% — 18% in the third quarter, and fell further to 10% to 15% in the fourth quarter; the judgment of Group Intelligence Consulting was that the price increase was highly concentrated on AI server-related products. The consumer level clearly weakened, and the contradiction between supply and demand shifted from an “overall shortage” to a “structural mismatch.” Although hard drives and memory are in different categories, they are the same assets that “make money by increasing prices”, and a slowdown in the price increase slope will directly weigh on valuations. The second one is supply: Toshiba is reportedly planning to invest about 60 billion yen (about 380 million US dollars) to double AI data center hard drive production capacity by the 2027 fiscal year. Seagate and Toshiba are also bidding for TDK's hard drive head business — the latter is related to the supply and cost of outsourced magnets. The third one is a demonstration effect of the stock price itself: from $805.63 on October 7 to $466.53 on November 5, it fell 42% in one month and collected $460.99 on November 6.
The report's own risk list is also very clear: the certification cycle for high-volume products is lengthening, the cyclical market is falling steeper, gross margins are no longer expanding, pricing normalization exceeds expectations, and geopolitical and tariff costs. Conversely, the upward trigger conditions are further acceleration of cloud demand, increased demand for hard drives brought by AI, gross margin rising to a 50% higher range, a smoother cloud cycle driving multiplier expansion, further strengthening pricing power, and HAMR seizing share.
What is the Dama report really betting on
Looking at the three levels together: the report is not about “good demand for hard drives,” but rather that the valuation framework for hard drives should be changed — from “cyclical stocks are priced at the bottom of the cycle” to “infrastructure is priced according to cash flow.” The evidence it gives is structural (area density, data center share, capital intensity), and the evidence given by the market is cyclical (price increase slope, production expansion news, stock price momentum). The difference is the time scale.