BOE A Deep-Dive Update: Citi Raises TP to RMB8.7; Repricing After a 2% LCD TV Panel Decline and Fading Glass-Substrate Premium
目录
Too Long; Didn’t Read
I. The Core of Citi’s Report Is Not the Downgrade, but a “Premium Audit”
II. Citi vs. Morgan Stanley: Both Assign a Glass-Substrate Premium, but with Different Price Discipline
III. Glass Substrate: The Technology Roadmap Is More Real, but Profit Timing Is Later
IV. LCD Remains the Base: Pricing Pressure Exists, Structural Upgrade Is Offsetting It
V. Citi Model Changes: Revenue Down, Net Profit Up; the Market Should Focus on Profit Quality
VI. Three Updated Scenarios: Even a Bull Case Must Define What Could Go Wrong
VII. Over the Next Four Quarters, the Real Thing to Watch Is Not News Heat
VIII. Conclusion: BOE A Has Not Lost Its Option, but the Share Price Has Lost Some Cheapness
本内容基于公开资料和研报数据整理，不构成任何投资建议，不代表任何个人观点，仅供学习参考，请理性阅读
The incremental message from Citi’s July 7 report is concentrated: the target price was raised from RMB5.0 to RMB8.7, while the rating was cut from Buy to Neutral. Our prior full deep dive already covered display cash flow, OLED repair, and the glass-substrate option. This update recalculates one question only: after the LCD base has become more stable and the glass-substrate indicators more specific, how much valuation margin of safety remains.
Too Long; Didn’t Read
The downgrade is not bearish. Citi raised its target price to RMB8.7 while cutting the rating to Neutral. It recognizes that BOE A’s asset quality and glass-substrate option deserve a higher P/B. The real change is that the trade has moved from value discovery into realization audit; near the current share price, the room to keep chasing higher has narrowed.
The glass substrate is more tangible, but also further out. BOE A has disclosed process progress including a 510mm x 515mm validation line, TGV aspect ratio of 20:1, sub-2-micron lines, and more than 20 build-up layers. However, the mass-production investment decision still needs to wait until mid-2027, and line construction is expected to take about another two years. Citi therefore pushes meaningful profit contribution to after 2029.
LCD remains the valuation base. Citi expects LCD TV panel prices to fall about 2% YoY in 2026, but industry concentration is close to complete and screen sizes continue to move up. BOE A’s two Gen-8.5 lines have net margins above 20%, and the depreciation peak likely appeared around 2025. The core business provides the valuation floor; the next thing to watch is whether price declines remain moderate.
Profit quality matters more than revenue. Citi raised its 2026/2027 net profit forecasts by 9%/6%, respectively, but cut revenue forecasts for the same periods by 5%/6%. This means the earnings improvement is coming more from product mix, innovative businesses, and easing depreciation pressure than from simple revenue expansion. Whether the valuation can hold depends on whether net margin and ROE can keep rising.
The trade needs validation. If glass-substrate customer validation, yield, and the mid-2027 investment decision continue to advance, BOE A still has room for asset revaluation. If LCD pricing is weaker than expected, OLED losses narrow slowly, and glass substrate remains only at the sample and sentiment stage, P/B above 2x will face pullback pressure. Future tracking should return to orders, yield, and ROE, not just sample-related news.
I. The Core of Citi’s Report Is Not the Downgrade, but a “Premium Audit”
The headline conclusion of Citi’s report looks contradictory: the target price was raised from RMB5.0 to RMB8.7, while the rating was cut from Buy to Neutral. The correct reading should not focus only on the word “downgrade,” nor only on the target-price increase. The real incremental point is that Citi has revalued BOE A from an “LCD cycle leader” into a framework of “LCD cash-flow base + glass-substrate option,” and then concluded that the current share price has already reflected both layers at a level that is not cheap.
This connects directly with the main thesis of the previous BOE A deep dive.
BOE A Deep Dive: Morgan Stanley Raises TP to RMB9.30; Revaluation Logic from Display-Panel Cycle Leader to Glass-Substrate Option
The previous article discussed why the market was willing to assign BOE A a higher valuation: first, after LCD industry concentration improved, the damage from panel-price volatility to profits declined; second, as depreciation and capex cycles moved down, cash-flow quality began to improve; third, glass substrate mapped the company from display assets to advanced-packaging assets, creating new valuation imagination. Citi’s report adds a colder price judgment to this framework: the logic can be valid without every price being worth chasing.
Citi’s new target price is RMB8.7, corresponding to 2.3x 2026E P/B. This multiple is already close to the high end of the company’s valuation range over the past five years. Citi’s rationale for the higher multiple is improved LCD earnings quality and potential market sentiment from glass-substrate applications starting in 2027. The issue is also here: once the sell side has already embedded glass-substrate sentiment into the target valuation, further upside depends more on “technical progress turning into orders and profits,” rather than retelling the story.
This is also the reasonable explanation for “raising the target price while downgrading the rating.” Citi recognizes improving asset quality, so the target price goes up; but with only a mid-teens total return from the current price to the target price, the rating can only be cut to Neutral. The investment implication is direct: BOE A remains one of the few A-share companies that can capture both display-cycle recovery and an advanced-packaging option, but it has moved from a low-valuation revaluation phase into a stage that needs quarter-by-quarter validation of realization quality.
II. Citi vs. Morgan Stanley: Both Assign a Glass-Substrate Premium, but with Different Price Discipline
The previous report mainly centered on Morgan Stanley’s RMB9.30 target price. Morgan Stanley’s focus was on repricing BOE A from a traditional LCD cycle stock into a composite asset of “display cash flow + glass-substrate option.” Citi’s new target price of RMB8.7 is below Morgan Stanley’s RMB9.30, but significantly higher than its previous RMB5.0. The two do not imply a directional conflict. The differences are mainly in three areas.
First, Citi places more emphasis on the fact that the current valuation has already priced in expectations in advance. Citi cuts the target multiple to 2.3x 2026E P/B, which already assigns BOE A a valuation above its historical center. This multiple is based on more stable LCD supply-demand, improved profitability, and potential glass-substrate applications beginning to ferment from 2027. In other words, Citi is not denying option value; it believes the option value has already entered the model.
Second, Citi is more cautious on the pace of profit realization. The most important sentence in the report is that the glass-substrate mass-production investment decision is expected by mid-2027, and line construction may take about two years, so meaningful net profit contribution is unlikely before 2029. This timeline cools down the market’s most excitable short-term area. Between technical samples, customer proof of concept, pilot-line validation, and large-scale net profit contribution, there are still yield, capex, customer adoption, packaging-architecture migration, and ecosystem coordination barriers.
Third, Citi separates the core business from the option more clearly. The LCD core business provides the cash-flow base and valuation floor; glass substrate provides the upper end of P/B. As long as LCD does not collapse, BOE A is unlikely to return easily to the old pure-cycle low valuation. But as long as glass substrate has no orders and profits, valuation also cannot move toward advanced-packaging companies without constraint.
Together, these three points form the updated core judgment: BOE A’s investment thesis has not changed, but the payoff structure has. In the previous stage, the market bought “from being undervalued to being re-recognized.” In the next stage, the market needs to see whether “after revaluation, enough new evidence can be delivered.”
III. Glass Substrate: The Technology Roadmap Is More Real, but Profit Timing Is Later
The most valuable incremental content in Citi’s report is its more detailed breakdown of BOE A’s glass-substrate technology progress. The market previously knew the company had exposure to glass substrate, TGV, and advanced packaging, but it was easy to jump directly from “having a technology roadmap” to “immediate profit contribution.” By laying out the intermediate steps, Citi’s conclusion is closer to this: the industrialization direction is clear, but commercialization realization cannot be rushed.
BOE A’s disclosed progress is not weak. The company has more than six years of TGV packaging-process accumulation. In 2022, it invested in a 510mm x 515mm large-size validation line, and completed the full process flow about six months after equipment move-in. It also has an 8-inch experimental line for material and formula validation, and completed a domestic glass-substrate innovation pilot line in 2024. Citi noted that the company has completed customer proof of concept and delivered second-generation samples in 2025.
The process metrics are more important. According to Citi’s compilation, BOE A has achieved a TGV aspect ratio of 20:1, with waist diameter adjustable between 65-100 microns, via roundness above 90%, and no via-corner or via-wall cracks. Plating uniformity is below 5%, surface and in-via copper thickness consistency is about 80%, with no via blockage or bubbles. Line width is below 2 microns, line spacing above 2 microns, alignment accuracy 1 micron, wiring uniformity below 5%, and the process can support more than 20 build-up layers and vertically stacked vias. In reliability, single-layer substrates have passed 1,000 temperature cycles, HAST, high/low-temperature storage, and high-temperature/high-humidity aging. A 75mm x 75mm sample had room-temperature warpage of about 40 microns and maintained low warpage after 260°C reflow.
These indicators show one thing: BOE A is not merely talking about a concept. It has indeed migrated capabilities in large-size glass, precision processing, automation, and yield management from the display-panel era into the packaging-substrate direction. Compared with typical new entrants, its advantages lie in large-size glass handling, panel-level processes, equipment automation, and mass-production management experience. For glass substrate to become a key carrier in AI computing packaging, the final competition is not only about materials, but whether vias, lines, layers, warpage, cutting, reliability, and customer design coordination all pass together.
But Citi also makes the timeline clear. Over the next two years, the company’s focus will remain on improving yield, stabilizing mass-production capability, advancing ecosystem cooperation, and working with customers to reduce layer count and optimize packaging architecture. The real mass-production investment decision is expected to be made only by mid-2027; if the decision lands, line construction may still take about two years. Therefore, from 2026 to 2028, glass substrate is more of a valuation catalyst and technical validation item, and too much net profit should not be booked into the income statement in advance.
This matters for investment judgment. What glass substrate brings BOE A is the possibility of “asset-attribute migration,” not a 2026 profit explosion. The market can assign a higher P/B to this roadmap, but that P/B needs sustained evidence. The most important things to watch later are not single sample-related news items, but four harder indicators: customer lists and validation stages, yield and defect rates, capex plans for mass-production lines, and whether the mid-2027 investment decision lands on schedule.
IV. LCD Remains the Base: Pricing Pressure Exists, Structural Upgrade Is Offsetting It
If one discusses only glass substrate, it is easy to portray BOE A as an advanced-packaging option company. If one looks only at LCD pricing, it is easy to underestimate the company’s cash-flow improvement and changes in industry concentration. The strength of Citi’s report is that it puts both in the same table: LCD still faces short-term pricing pressure, but industry structure and company efficiency are already better than before.
Citi expects LCD TV panel prices to fall about 2% YoY in 2026. This figure is not aggressive, but it is enough to remind the market that LCD is still a cyclical asset and will not escape supply-demand volatility because of the glass-substrate story. Consumer-electronics end demand is still being affected by memory price increases. Brand vendors may lower some specifications, or launch higher-spec products to offset price increases, ultimately creating disruptions to product mix and shipment cadence.
BOE A’s base advantage lies in industry concentration and larger screen sizes. Citi believes LCD TV panel industry consolidation is close to complete, with the top-five vendor concentration broadly stable in 2026. More importantly, size mix is moving up: the shipment share of 70-75-inch TVs rises from 8% in 2025 to 22% in 2026; shipments of 80-inch-plus TVs are expected to grow more than 20% YoY in 2026; average TV shipment size is expected to reach 55 inches in 2027. Management views 55 inches as the key threshold for supply-demand balance, because larger TVs consume more glass-substrate area and can absorb capacity faster at the same unit shipment volume.
Company-level improvement is also relatively clear. Citi noted that BOE A’s Chongqing and Fujian Gen-8.5 lines already have net margins above 20%. Its two Gen-10.5 lines are not relying on capacity expansion, but continue to increase shipments through quality and efficiency improvements. On depreciation, overall depreciation pressure likely peaks in 2025, and may rise slightly in 2027 because of B16, but the long-term trend remains downward. After the OLED Gen-8.6 line is completed, capex is also expected to decline.
