The Epiwafer Monopolist Nobody Wanted to Own
Taiwan sale unlocks the re-rate. Margin recovery on top of a photonics supercycle.
A first look at IQE plc: the compound-semiconductor foundry at the center of the AI photonics supply chain, and why the market is still pricing it like a distressed industrial
Position disclosure: I am long IQEPF. This is my largest position at 35.1% of the portfolio. I entered near cycle lows at under 1x revenue. I reclassified IQEPF off my avoid list on 2026-04-18. This write-up reflects a position I already hold. Read accordingly.
IQE plc trades at $0.74 per share as of April 20, 2026, with a market cap of $724.64M and trailing revenue of $97.27M. That is an EV/Revenue multiple of 8.1x on a business with negative gross margins and $69.83M in net debt. On the surface, that looks expensive for a broken foundry. It is not. The market is pricing a distressed epiwafer manufacturer. What it is actually pricing is the world's largest compound-semiconductor epitaxial wafer foundry at the precise moment when every AI data-center laser, every 5G power amplifier, every solid-state LiDAR system, and every next-generation AR display needs exactly what IQE grows.
The stock was below $0.06 at its 52-week low. It has traded as high as $0.85 in the past year. It 10x'd in roughly four months as the photonics narrative broke into the mainstream. Shareholders who held for five years are still down 88%. That is the setup. A company that was left for dead, now re-rated by a theme, still carrying the balance-sheet scars of a multi-year downcycle, with a strategic review underway, a new permanent CEO installed, a Taiwan facility sale in process, and a May 20 earnings date that will either validate the recovery or reset expectations again.
Let's go.
I. Executive Scorecard: Bear, Base, Bull, Moonshot
Before the deep-dive, here is where the scenarios land. Full math in Section VI.
Probability-weighted expected value:
weighted_EV = 0.20 × $0.18 + 0.45 × $0.62 + 0.25 × $1.55 + 0.10 × $3.10
= $0.036 + $0.279 + $0.3875 + $0.31
= $1.01
At the current price of $0.74, the probability-weighted EV implies 37% upside to the weighted mean. The bull case implies 110% upside. The moonshot implies 319% upside. The bear case implies 76% downside.
Annualized IRR at base case over three years: approximately 11%. Annualized IRR at bull case over three years: approximately 28%. Over five years at bull case: approximately 16% annualized, assuming the stock reaches $1.55 by 2028 and holds.
The math is the math.
II. What Just Changed: The Setup Nobody Was Pricing
Twelve months ago, IQE was a distressed UK AIM-listed foundry with a collapsing revenue line, negative gross margins, and a balance sheet that looked like it needed emergency surgery. Revenue fell 31.5% year-over-year to $97.27M TTM. Gross margin went to zero. Operating margin hit negative 31.7%. Net margin was negative 50.5%. The stock was priced for liquidation.
Three things changed.
First, the photonics super-cycle arrived with force. AI data-center operators are deploying co-packaged optics (CPO) and silicon photonics transceivers at a pace that requires compound-semiconductor epiwafers at scale. There is no silicon substitute for the InP and GaAs laser structures that drive 800G and 1.6T optical interconnects. IQE grows those structures. Nobody else at IQE's scale does it in the Western world.
Second, IQE began a strategic review that included exploring a full sale of the company. That review surfaced deal interest and drove a 25% single-day surge in the shares. The review is ongoing. A Taiwan facility sale is actively in process. The combination of balance-sheet repair and strategic optionality changed the risk profile.
Third, short interest collapsed. Short interest in IQEPF dropped 85.2% in a recent measurement period. A month earlier it had already dropped 73.1%. The bears who were positioned for liquidation have largely exited. The stock is no longer a crowded short.
None of this means the operational problems are solved. They are not. But the market was pricing permanent impairment. The actual path is a cyclical trough with structural tailwinds arriving at the same time as a balance-sheet catalyst. That is the asymmetry.
III. The Market: Why the Photonics TAM Is Real and Why IQE Sits at Its Center
The compound-semiconductor epiwafer market is not a niche. It is the substrate layer of every photonic and RF system that matters in the next decade.
Start with AI data-center interconnects. Hyperscalers are deploying co-packaged optics to move data between GPU clusters at speeds that copper cannot sustain. The transceivers in those systems use vertical-cavity surface-emitting lasers (VCSELs) and edge-emitting lasers built on InP and GaAs epiwafers. IQE supplies those wafers. Lumentum, one of the largest laser component makers in the world, extended a multiyear epiwafer supply agreement with IQE. Coherent (formerly II-VI) is another named customer in the photonics supply chain. These are not speculative relationships. They are contracted revenue.
Then there is 5G. Every high-band 5G smartphone power amplifier uses a GaAs pHEMT or GaN-on-SiC structure. Qorvo, Skyworks, and Broadcom are the dominant RF front-end module makers. They source epiwafers from IQE. As 5G penetration deepens globally and mmWave deployments accelerate in the US and Asia, the RF front-end wafer volume grows.
Defense is a third vector. GaN-on-SiC is the material of choice for phased-array radar, electronic warfare systems, and mmWave communications. IQE supplies export-controlled GaN epiwafers to US and UK defense customers. This revenue is sticky, high-margin, and largely invisible to the market because it does not get disclosed in detail.
Power electronics is the fourth. AI data centers run 48V power rails that need to be converted to 12V at the rack level. GaN power transistors are the most efficient solution. Navitas Semiconductor, Efficient Power Conversion, Infineon, and Power Integrations are all deploying GaN power devices. The epiwafer for those devices comes from foundries like IQE.
Finally, automotive LiDAR. Solid-state LiDAR systems for autonomous vehicles use VCSEL arrays. IQE has qualified its VCSEL epiwafers with automotive-grade customers. The LiDAR market is early but the qualification cycles are long, which means IQE's position today translates to revenue in 2027 and beyond.
The photonics epiwafer market is not one TAM. It is five overlapping TAMs, all growing, all requiring compound semiconductors, all requiring a foundry that can grow the epitaxial layers to specification. IQE is the largest Western foundry doing this at scale.
That is the structural context. The market is pricing none of it.
IV. What IQE Actually Sells: The Epiwafer Foundry Model
IQE does not make chips. It does not make lasers. It does not make power modules. It grows the epitaxial layers, atom by atom, on semiconductor substrates, and ships those wafers to the companies that fabricate the final devices.
The process is called metalorganic chemical vapor deposition (MOCVD). IQE operates MOCVD reactors at facilities in Cardiff (Wales), Bethlehem (Pennsylvania), and Taiwan. Each reactor grows a specific epitaxial structure to customer specification. The customer then takes that wafer and processes it into devices.
This is a foundry model with high switching costs. Once a customer qualifies an IQE wafer for a specific device, they do not switch suppliers. The qualification process takes 12 to 24 months and involves extensive device characterization. The wafer spec is embedded in the customer's device design. Switching means re-qualifying the device, which means re-testing, re-certifying, and potentially re-designing. Nobody does that unless they have to.
IQE's revenue breaks into three broad segments: wireless (GaAs and GaN for RF), photonics (VCSELs, edge emitters, InP), and power/other (GaN power, infrared). The photonics segment is the highest-margin and fastest-growing. The wireless segment is the largest by historical revenue but has been the most cyclical. The power segment is early-stage but strategically important.
The Taiwan facility is a separate manufacturing site that IQE has been operating for wireless wafer production. The strategic review has identified this asset as non-core. A sale would raise cash, reduce complexity, and allow management to focus on the higher-margin photonics and defense segments. The proceeds would go toward debt reduction. The net debt position is $69.83M against cash of $16.99M and gross debt of $86.83M. A Taiwan sale that generates $30M to $50M in proceeds would materially change the balance-sheet picture.
IQE is also expanding its IP position. The company recently acquired the Luxtaltek IP portfolio covering VCSEL and Micro LED technologies. This is not a defensive move. It is an offensive one. Micro LED displays are the next-generation display technology for AR/VR headsets. Apple's Vision Pro uses compound-semiconductor components for eye tracking and depth sensing. The next generation of Vision Pro and competing AR headsets will need Micro LED illumination. IQE just bought the IP to supply that market.
The epiwafer foundry model, at scale, with a diversified customer base across photonics, RF, defense, and power, is a high-value position in the compound-semiconductor supply chain. The current financials do not reflect that value. They reflect a cyclical trough.
V. Technology Deep-Dive: What Goes Into an Epiwafer and Why It Is Hard to Replicate
An epitaxial wafer is a semiconductor substrate with one or more thin crystalline layers grown on top. The layers are grown atom by atom in a MOCVD reactor at temperatures above 600 degrees Celsius, using organometallic precursor gases. The thickness of each layer is controlled to within a few atomic monolayers. The composition of each layer determines the electronic and optical properties of the final device.
For a VCSEL, the epitaxial structure includes a distributed Bragg reflector (DBR) stack, an active quantum-well region, and a current-confinement layer. The DBR stack alone can have 40 or more alternating layers of AlGaAs and GaAs. Each layer must be grown to a specific thickness and composition. A deviation of a few percent in aluminum content changes the reflectivity of the mirror and degrades device performance.
For a GaN-on-SiC RF transistor, the epitaxial structure includes a GaN buffer layer, an AlGaN barrier layer, and a two-dimensional electron gas (2DEG) at the interface. The 2DEG is what gives GaN its high electron mobility and high breakdown voltage. Growing a uniform, defect-free GaN layer on a SiC substrate requires precise control of temperature, pressure, and gas flow across the entire reactor.
IQE has been doing this for over 30 years. The process knowledge is embedded in the people, the equipment configurations, and the proprietary recipes. A new entrant cannot replicate this by buying MOCVD reactors. They need the recipes, the process engineers, and the years of customer qualification data.
IQE's patent portfolio covers key aspects of its epitaxial processes, including GaN-on-SiC structures, its proprietary cREO (crystalline rare-earth oxide) technology for next-generation substrates, and VCSEL structures. The Luxtaltek IP acquisition adds VCSEL and Micro LED patents to that portfolio. Chinese foundries have been attempting to build compound-semiconductor capability for years. IQE's IP position, combined with export controls on compound-semiconductor technology, creates a meaningful barrier.
The cREO technology deserves specific mention. IQE developed a process for growing crystalline rare-earth oxide layers on silicon substrates. This allows compound-semiconductor devices to be integrated with standard silicon CMOS processes. If this technology reaches commercial scale, it could allow IQE to supply epiwafers for silicon-integrated photonics at a cost structure that is not achievable with traditional III-V substrates. This is a long-duration option that the market is not pricing at all.
The technology moat is real. It is not a marketing claim. It is 30 years of process recipes, customer qualifications, and patent filings.
VI. The Math: Four Scenarios, Explicit Revenue Ramps, Implied Prices
The current financials are the trough. Revenue TTM is $97.27M, down 31.5% year-over-year. Gross margin is approximately zero. Operating margin is negative 31.7%. EBITDA is negative $10.59M. Free cash flow is positive $5.12M, which is the one number that tells you the business is not burning cash at the operating level despite the accounting losses.
The recovery path depends on three variables: revenue recovery (capacity utilization), mix shift toward photonics (higher margin), and the Taiwan sale (balance-sheet repair).
Note: FY2026E figures are Shawarma Capital model projections based on publicly available data. They are not guidance.
The chart above shows the trough in FY2025 and the recovery path. The gross margin line is the critical variable. At roughly $130M in revenue, IQE crosses back into positive gross margin territory based on its fixed-cost structure. At $155M, gross margin reaches approximately 18%. At $200M, it approaches 22%. These are not heroic assumptions. They reflect a return to 2022-era revenue levels with a better mix toward photonics.
Bear Case (20% probability): Revenue stalls at $90M. The Taiwan sale falls through or generates minimal proceeds. Gross margin recovers to only 2%. The photonics cycle disappoints. IQE trades at 1.5x EV/Revenue. Implied share price: $0.18.
Base Case (45% probability): Revenue recovers to $155M by FY2028, driven by photonics growth and 5G RF stabilization. Gross margin reaches 14%. Taiwan sale closes, reducing net debt by $35M. IQE trades at 3.0x EV/Revenue, a discount to US photonics peers but a premium to its current AIM-listed multiple. Implied share price: $0.62.
Bull Case (25% probability): Revenue reaches $220M by FY2028. Photonics demand from AI data centers accelerates faster than expected. Apple Vision Pro successor drives VCSEL volume. GaN defense revenue grows on increased NATO spending. Gross margin reaches 22%. IQE re-rates to 5.0x EV/Revenue as a US listing or M&A premium is applied. Implied share price: $1.55.
Moonshot Case (10% probability): Revenue reaches $320M by FY2028. A strategic acquirer (Coherent, a defense prime, or a sovereign wealth fund) takes IQE out at a premium. cREO technology reaches commercial scale and opens a new silicon-integrated photonics market. IQE trades at 7.0x EV/Revenue in an acquisition. Implied share price: $3.10.
Weighted EV: $1.01. Current price: $0.74. The asymmetry is clear.
VII. Customer Roster: Who Is Buying IQE Wafers
The customer base is the thesis in concrete form. IQE does not disclose individual customer revenue in detail, but the confirmed and indicated relationships are substantial.
Customer vs Status
The Apple relationship deserves specific attention. IQE has been a VCSEL epiwafer supplier for iPhone FaceID since the technology launched. The Vision Pro uses compound-semiconductor components for its eye-tracking and depth-sensing systems. Apple has explored in-housing compound-semiconductor production, but the capital intensity and process complexity of MOCVD epitaxy make full vertical integration unlikely in the near term. IQE's qualification history with Apple is a durable competitive position.
The Lumentum multiyear agreement is the most important confirmed relationship for the photonics thesis. Lumentum is one of the largest suppliers of VCSELs and edge-emitting lasers for data-center transceivers. A multiyear supply agreement means IQE has contracted revenue from the photonics super-cycle, not just exposure to it.
Customer concentration is a real risk. The top five customers likely represent a substantial majority of revenue. If any one of them reduces orders, the revenue impact is material. This is acknowledged in the risk matrix below.
The customer roster is not speculative. These are real relationships with real contracted volumes. The market is pricing IQE as if the customers do not exist.
VIII. Peer Multiples: Why the Comp Set Is Wrong
The market compares IQE to other AIM-listed UK industrials. That is the wrong comp set. IQE is a compound-semiconductor foundry with photonics, defense, and power electronics exposure. The right comp set is US-listed compound-semiconductor and photonics companies.
The table reveals the gap. MACOM, a compound-semiconductor component company, trades at approximately 8x EV/Revenue with 55% gross margins. IQE, at the same EV/Revenue multiple, has zero gross margins. That sounds like IQE is expensive. It is not. IQE is at the trough of its margin cycle. MACOM is at steady state. The question is not where IQE trades today. The question is where it trades when gross margins recover to 20%.
At $155M revenue and 20% gross margin, IQE is a different business. At that point, the right multiple is 3x to 5x EV/Revenue, not 8x. The current 8x multiple on trough revenue is actually a lower absolute EV than a 3x multiple on recovered revenue. The market is confusing a high multiple on a small number with an expensive stock.
The UK AIM listing is a structural discount. AIM-listed companies trade at a discount to US-listed peers because of lower liquidity, lower institutional coverage, and lower index inclusion. IQE's primary listing is on AIM. IQEPF is the OTC ADR. A US listing or a strategic acquisition by a US company would remove this discount immediately. The strategic review is the mechanism by which this discount could be eliminated.
The comp set is wrong. The multiple will compress as revenue recovers. That compression is the re-rate.
The gap between IQE's gross margin and every peer in the table is the entire thesis. Close that gap and the multiple compresses to something rational. The market is not pricing the closure.
IX. Catalyst Map: The Next 12 Months
The next 12 months have more binary events than most stocks carry in three years. Each one is a potential re-rate trigger.
The May 20 earnings date is the nearest binary. The market will be looking for three things: revenue trajectory (is the recovery real), gross margin direction (is it turning positive), and any update on the Taiwan sale or strategic review. A strong print on all three could drive a 20% to 30% move. A miss on any one of them, particularly gross margin, could give back a similar amount.
The Taiwan facility sale is the balance-sheet catalyst. Net debt of $69.83M against a revenue base of $97.27M is a meaningful constraint. If the Taiwan sale generates $30M to $50M in proceeds and those proceeds go to debt reduction, the net debt falls to $20M to $40M. That changes the covenant picture, reduces interest expense, and removes the balance-sheet overhang that has been suppressing the multiple.
The strategic review is the wildcard. IQE is a Western compound-semiconductor foundry with export-controlled technology, a 30-year customer qualification history, and the largest epiwafer capacity in the non-Chinese world. In a geopolitical environment where the US and UK are actively trying to build domestic compound-semiconductor supply chains, IQE is a strategic asset. A defense prime, a photonics integrator, or a sovereign wealth fund could pay a significant premium for that asset. The 30% probability assigned to a full sale outcome reflects the uncertainty of timing, not the strategic logic.
The Newport Fab litigation is a risk, not a catalyst, but its resolution removes an overhang. The details of the federal court case are not fully public, but the outcome could affect IQE's competitive positioning in certain wafer types.
The next catalyst is May 20. Everything else follows from that print.
X. Risk Matrix: What Could Go Wrong
Every real bear point deserves acknowledgment. Here they are.
The dilution risk is real and recent. IQE proposed a placing of ordinary shares that caused the stock to slide. The shares outstanding are 979.25M with a float of 795.12M. Further placings are possible if the balance sheet requires additional repair. Each placing is dilutive to existing shareholders.
The management transition is also real. Jutta Meier was appointed as permanent CEO. She is new to the role. The strategic review, the Taiwan sale, and the margin recovery all require execution. A new CEO executing multiple simultaneous initiatives is a risk. It is also an opportunity: a new CEO with a clean mandate can make decisions that a legacy team could not.
The insider selling reported by Simply Wall St three days ago is worth noting. PDMR (person discharging managerial responsibility) share sales are a yellow flag, not a red one. They could reflect personal liquidity needs rather than a negative view on the company. But they deserve monitoring.
The risks are real. They are also known. The market has been pricing these risks for years. The question is whether the risks are already in the price at $0.74.
They are.
XI. Capital Structure: The Dilution Math and the Debt Path
IQE has 979.25M shares outstanding and a float of 795.12M. The difference (approximately 184M shares) is held by insiders and strategic holders. Cash is $16.99M. Gross debt is $86.83M. Net debt is $69.83M.
The recent share placing raised capital but diluted existing shareholders. The block admission of shares for the employee incentive plan added further shares. The total voting rights update reflects these issuances. The share count has been growing. This is a headwind to per-share value.
The Taiwan facility sale is the primary mechanism for debt reduction without further dilution. If the sale generates $40M in proceeds and $35M goes to debt repayment, net debt falls to approximately $35M. At that level, the debt-to-revenue ratio (on $115M FY2026E revenue) is approximately 0.3x. That is manageable. The covenant pressure largely disappears.
The operating leverage math is straightforward. IQE's fixed cost base is approximately $45M to $50M per year (based on the negative operating margin at $97M revenue). At $130M revenue with a 10% gross margin, gross profit is $13M. That does not cover fixed costs. At $155M revenue with a 15% gross margin, gross profit is $23M. Still short of fixed costs but closing. At $180M revenue with a 20% gross margin, gross profit is $36M. The business approaches operating breakeven. At $220M revenue with a 22% gross margin, gross profit is $48M. The business is profitable.
The breakeven revenue level is approximately $175M to $185M, depending on the margin mix. That is roughly 80% above the current TTM revenue of $97.27M. It is not an easy climb. But it is achievable over two to three years if the photonics and defense segments grow as the structural thesis suggests.
XII. The IP Moat: Patents, cREO, and the Chinese Threat
IQE's patent portfolio is a genuine competitive asset. The company has filed patents on epitaxial growth processes for GaN-on-SiC, InP-based laser structures, and its proprietary cREO technology. The Luxtaltek IP acquisition added VCSEL and Micro LED patents to this portfolio.
The cREO technology is the longest-duration option in the IQE story. cREO allows the growth of crystalline rare-earth oxide layers on silicon substrates, which in turn allows compound-semiconductor device structures to be integrated with silicon CMOS. If this works at commercial scale, it means compound-semiconductor photonics can be manufactured on standard silicon wafer lines, dramatically reducing cost. IQE would be the IP holder for that process.
Chinese foundries have been building compound-semiconductor capacity for years. The Chinese government has invested heavily in domestic III-V semiconductor production. But export controls on MOCVD equipment, compound-semiconductor process technology, and related IP have slowed Chinese progress in the highest-performance segments. IQE's GaN-on-SiC and InP processes are in the export-controlled category. Chinese foundries cannot legally acquire IQE's process technology or equipment configurations.
This is not a permanent moat. Chinese foundries will eventually develop their own processes. But the timeline is measured in years, not months. And in the defense and high-reliability photonics segments, Western customers will not qualify Chinese-sourced epiwafers regardless of technical capability. The geopolitical moat is as real as the technical one.
The Luxtaltek acquisition is particularly interesting for the Micro LED angle. Micro LED displays are the next-generation technology for AR/VR headsets. Apple, Meta, and Samsung are all developing Micro LED display programs. The display panels require compound-semiconductor LED structures grown on epiwafers. IQE just acquired the IP to supply that market. The revenue from Micro LED is not in the current financials. It is a 2027 to 2030 option.
The IP portfolio is not priced. The cREO option is not priced. The Micro LED option is not priced. The market is pricing the current revenue and the current margins. That is the gap.
XIII. The Photonics Super-Cycle: Why This Is Not a Trade
The photonics super-cycle is not a meme. It is a capital expenditure reality.
Microsoft, Google, Amazon, and Meta are spending hundreds of billions of dollars on AI data-center infrastructure. The interconnects inside those data centers are moving from copper to optical. The optical transceivers in those interconnects use VCSELs and edge-emitting lasers built on compound-semiconductor epiwafers. The volume of epiwafers required to supply that interconnect market is growing faster than any single foundry can currently supply.
Co-packaged optics (CPO) is the next step. CPO integrates the optical transceiver directly into the switch or GPU package, eliminating the pluggable module. CPO requires even tighter integration of compound-semiconductor laser sources with silicon photonics waveguides. IQE's epiwafers are the laser source in that architecture.
Lumentum's multiyear supply agreement with IQE is a direct expression of this demand. Lumentum is one of the largest suppliers of VCSELs and edge emitters for data-center transceivers. They locked in a multiyear supply agreement with IQE because they needed supply certainty for a demand environment they could see coming.
The 5G RF cycle is a separate but parallel tailwind. High-band 5G (mmWave) requires GaAs pHEMT and GaN power amplifiers in every handset. The penetration of mmWave 5G is still in early innings globally. As it deepens, the RF front-end wafer volume grows. IQE supplies GaAs and GaN epiwafers to Qorvo, Skyworks, and Broadcom, the three dominant RF front-end module makers.
Defense is the third leg. NATO members are increasing defense spending following the Russia-Ukraine conflict and rising geopolitical tensions. Phased-array radar, electronic warfare, and satellite communications all use GaN-on-SiC power amplifiers. IQE supplies export-controlled GaN epiwafers to US and UK defense customers. This revenue is growing and is largely invisible to the market.
Three super-cycles, all requiring compound-semiconductor epiwafers, all arriving simultaneously. IQE is the largest Western foundry. The market is pricing a distressed industrial.
You don't need to believe me. Look at what Lumentum signed.
XIV. Bottom Line: The Asymmetry in One Paragraph
IQE plc is the world's largest compound-semiconductor epiwafer foundry. It is trading at $0.74 per share with a market cap of $724.64M and trailing revenue of $97.27M. Gross margins are at zero. Net debt is $69.83M. The stock is down 88% over five years and up 825% year-to-date. A strategic review is underway. A Taiwan facility sale is in process. A permanent CEO has been installed. Short interest dropped 85.2%. The Luxtaltek IP acquisition added VCSEL and Micro LED patents. Lumentum extended a multiyear supply agreement. The May 20 earnings date is the next binary. The probability-weighted expected value across four scenarios is $1.01 against a current price of $0.74, with a bull case of $1.55 and a moonshot of $3.10. The bear case is $0.18. The asymmetry is 37% to the weighted mean, 110% to the bull case, 319% to the moonshot, against 76% downside in the bear. I entered near cycle lows at under 1x revenue. This is 35.1% of my portfolio. The next catalyst is May 20.
Position accordingly.
What is Coming in Parts 2 and 3
Parts 2 and 3 drop on the paid tier. Part 2 builds the valuation matrix: bear, base, bull, and moonshot scenarios with probability weights, peer-multiple re-rating math against Qorvo and Skyworks, implied price targets through 2028, and the EPS trajectory that bridges current losses to 14% EBITDA margins. Part 3 maps the catalyst calendar across the next twelve months: May 20 earnings, Taiwan facility close timing, gross-margin inflection milestones, customer wins in AI photonics, and the strategic review outcome, each with dated probability weights and position-sizing rules tied to realized volatility. The risk matrix closes Part 3 with balance-sheet stress scenarios, competitive pressure from TSMC's GaAs entry, and the customer concentration risk that sits beneath the recovery thesis.
This is research synthesis, not investment advice. You should not buy or sell securities based on anything I write. I am not a registered investment advisor. I do not owe you a fiduciary duty. My conclusions could be wrong in ways I have not anticipated. Financial projections are model outputs based on publicly available data. They are not guarantees. Do your own due diligence.*
More in this series
The Epiwafer Monopoly: The Complete Series