The most vertically integrated photonics company on earth. From raw materials (InP, GaAs, SiC) to lasers to transceivers to systems. Leopold's smaller position ($89M). Founded 1971 as II-VI, acquired Coherent Inc. for ~$5.7B in 2022. | AGI Score: 8 | Analysis date: 2026-03-12
Leopold holds ~$89M in Coherent Corp, a smaller position alongside his $479M in Lumentum. Both are optical networking plays, but Coherent is a fundamentally different animal: it is 3.5x larger by revenue ($5.8B vs $1.6B), has three diverse business segments, and is the most vertically integrated photonics company in the world -- from growing compound semiconductor crystals to shipping completed laser systems. The II-VI + Coherent Inc. merger in 2022 created a photonics superpower, but the integration is still playing out. New CEO Jim Anderson (ex-AMD/Lattice) joined June 2024 and is aggressively restructuring. The question: is Coherent a better risk/reward than Lumentum at similar market caps ($45B vs $44B)?
Coherent is an industrial conglomerate built around photonics -- the science of controlling light. Unlike Lumentum, which focuses primarily on datacom transceivers, Coherent operates across the full photonics stack and serves diverse end markets.
| Segment | Revenue | % Total | YoY Growth | Segment Profit | Margin |
|---|---|---|---|---|---|
| Networking | $3,421M | 59% | +49% | $644M | 18.8% |
| Materials | $954M | 16% | -6% | $355M | 37.2% |
| Lasers | $1,435M | 25% | +3% | $317M | 22.1% |
| Total | $5,810M | 100% | +23% | $1,316M | 22.7% |
Coherent's key differentiator is depth of vertical integration. The company doesn't just make transceivers -- it grows the compound semiconductor wafers (InP, GaAs, SiC), fabricates the laser chips, makes the passive optics and isolators, designs the driver ICs, manufactures the thermoelectric coolers, and assembles the final modules. No other company in the world has this breadth. This creates several advantages: (1) cost control -- making components internally is cheaper than buying them, (2) supply chain resilience -- less dependence on external suppliers, (3) quality control -- can optimize across the full stack, (4) innovation speed -- can iterate on materials and components simultaneously. The downside is complexity and capital intensity.
Coherent's quarterly revenue shows steady acceleration, though less dramatic than Lumentum's because the non-networking segments are flat/declining:
| Quarter | Revenue | Q/Q Growth | Gross Profit | Gross Margin | Op Income |
|---|---|---|---|---|---|
| Q2 FY2025 (Dec 2024) | $1,435M | -- | $509M | 35.5% | $145M |
| Q3 FY2025 (Mar 2025) | $1,498M | +4.4% | $528M | 35.2% | $146M |
| Q4 FY2025 (Jun 2025) | $1,529M | +2.1% | $546M | 35.7% | $145M |
| Q1 FY2026 (Sep 2025) | $1,581M | +3.4% | $579M | 36.6% | $172M |
| Q2 FY2026 (Dec 2025) | $1,686M | +6.6% | $623M | 36.9% | $199M |
Revenue up 17% from $1,435M to $1,686M over 4 quarters. Annualized run rate of $6.7B. The growth is entirely driven by Networking (+49% YoY) while Materials (-6%) and Lasers (+3%) are flattish. Gross margin steadily expanding from 31% (FY2024) to 37% (Q2 FY2026).
| Year | Revenue | Gross Profit | Gross Margin | Op Income | Net Income | EPS |
|---|---|---|---|---|---|---|
| FY2022 (Jun 2022)* | $3,317M | $1,265M | 38.1% | $414M | $235M | $1.45 |
| FY2023 (Jun 2023) | $5,160M | $1,618M | 31.4% | $82M | -$259M | -$2.93 |
| FY2024 (Jun 2024) | $4,708M | $1,456M | 30.9% | $123M | -$156M | -$1.84 |
| FY2025 (Jun 2025) | $5,810M | $2,043M | 35.2% | $535M | $49M | -$0.52 |
*FY2022 was the first full year post-Coherent Inc. acquisition (closed Jul 2022). Revenue jumped from ~$3.3B (II-VI standalone) to $5.2B including Coherent Inc.
FY2023-2024 were digestion years: integrating the massive Coherent Inc. acquisition, dealing with restructuring charges ($119M in FY2023), and navigating a weak telecom/industrial cycle. FY2025 shows the real earnings power emerging: revenue up 23%, operating income up 335% from $123M to $535M. But net income is only $49M due to $196M in interest expense, $160M in restructuring charges, and $85M in impairment charges. The reported EPS is negative because of preferred stock dividends to noncontrolling interest in Silicon Carbide LLC.
| Moat Source | Strength | Details |
|---|---|---|
| Vertical Integration | VERY STRONG | The deepest in the industry. Materials (InP, GaAs, SiC wafers) -> laser chips -> passive optics -> ICs -> thermoelectric coolers -> modules -> systems. No competitor matches this breadth. |
| IP Portfolio | VERY STRONG | ~3,100 patents globally. Combined IP from II-VI, Coherent Inc., Finisar heritage. |
| Manufacturing Diversity | STRONG | 30,000 employees across USA, China, Malaysia, Philippines, Singapore, Vietnam, Europe. Multiple 6-inch GaAs and InP fabs. Geographic diversity provides tariff/sanctions resilience. |
| Customer Qualification | STRONG | Two customers >10% of revenue. Protocol-agnostic transceivers (Ethernet + InfiniBand + NVLink) = qualified across all major AI platforms. |
| Material Science | VERY STRONG | Compound semiconductor expertise is a genuine barrier. Growing InP and GaAs crystals at scale is extremely difficult. Very few companies globally can do this. |
| Revenue Diversification | MIXED | Three segments provide revenue stability but dilute the AI narrative. Non-networking segments growing slowly or declining. |
| Metric | COHR | LITE | Advantage |
|---|---|---|---|
| Market Cap | $45.2B | $44.0B | Similar |
| Revenue | $5.81B | $1.65B | COHR (3.5x) |
| EV/Revenue | 8.0x | 21.9x | COHR (much cheaper) |
| Forward P/E | 32.8x | 42.8x | COHR |
| Gross Margin | 36.4% | 37.1% | Similar |
| Free Cash Flow | $193M | -$105M | COHR |
| Net Tangible Assets | $0.45B | -$0.39B | COHR |
| Networking Rev Growth | +49% | +30%* | COHR |
| Vertical Integration | Deepest in industry | Strong but less deep | COHR |
| AI Revenue Purity | 59% networking | 86% cloud & net | LITE |
| Debt/Equity | 39.9% | 392% | COHR |
| Beta (Volatility) | 1.91 | 1.41 | LITE |
| Employees | 30,216 | 10,562 | COHR has more scale |
*LITE Cloud & Networking grew 30% annually but quarterly sequential growth rate is faster (~65% on trailing 4 quarters)
On pure fundamentals, COHR looks significantly cheaper than LITE. Nearly identical market caps but 3.5x more revenue, positive FCF, positive tangible book, lower leverage, faster networking growth, deeper vertical integration, and a lower forward P/E. The market gives LITE a premium because of its higher AI revenue purity (86% vs 59%), but that premium seems excessive -- COHR's networking segment alone does $3.4B vs LITE's $1.4B in cloud & networking.
| Item | FY2025 | FY2024 | FY2023 | FY2022 |
|---|---|---|---|---|
| Total Assets | $14.91B | $14.49B | $13.71B | $7.84B |
| Cash | $909M | $930M | $820M | $2,580M |
| PP&E | $1.88B | -- | -- | -- |
| Goodwill | $4.47B | -- | -- | -- |
| Intangible Assets | $3.20B | -- | -- | -- |
| Total Debt | $3.89B | $4.30B | $4.49B | $2.44B |
| Stockholders' Equity | $8.13B | $7.57B | $7.23B | $4.38B |
| Net Tangible Assets | $0.45B | -$0.39B | -$1.10B | $2.46B |
$7.68B in goodwill + intangibles ($4.47B goodwill + $3.20B intangibles) -- 51% of total assets. This is almost entirely from the Coherent Inc. acquisition in July 2022 (~$5.7B deal). The old II-VI had only $1.9B in goodwill/intangibles.
Is this goodwill real? The Coherent Inc. acquisition brought genuine strategic assets -- world-class laser technology (excimer, ultrafast, CO2), a leading position in display capital equipment, and aerospace/defense capabilities. But $4.5B of goodwill on a $5.7B acquisition suggests a huge premium was paid. If any segment underperforms, impairment risk is real (they already took $85M in impairment charges in FY2025 on assets held for sale).
Key risk: At $45B market cap and $7.7B in goodwill/intangibles, a goodwill impairment wouldn't destroy equity value, but it would signal that the merger thesis is not working. The fact that Coherent is actively divesting non-core businesses (Newton Aycliffe sale, assets held for sale) suggests management is being realistic about what works and what doesn't.
| Component | Amount | Details |
|---|---|---|
| Term B Loans | ~$3.5B | Floating rate, being paid down. $45M lower interest expense YoY. Active deleveraging. |
| Other Debt | ~$0.4B | Various facilities |
| Total Debt | $3.89B | Down from $4.49B in FY2023. Paying down ~$400-500M/year. |
| Net Debt | $2.98B | Debt minus cash |
| Net Debt / EBITDA | 2.5x | Manageable but not comfortable. Targeting <2x. |
| Year | Operating CF | CapEx | Free Cash Flow |
|---|---|---|---|
| FY2022 | $413M | -$314M | $99M |
| FY2023 | $634M | -$436M | $198M |
| FY2024 | $546M | -$347M | $199M |
| FY2025 | $634M | -$441M | $193M |
Consistently FCF positive. ~$200M/year in free cash flow even during the integration/restructuring period. This is a key advantage over Lumentum (FCF negative). The company is using FCF to pay down debt.
Anderson's background is notable: he ran AMD's Computing and Graphics group during its turnaround under Lisa Su, then turned around Lattice Semiconductor (stock went from ~$5 to ~$90 under his leadership). He brought his CFO from Lattice (Sherri Luther) to Coherent. His playbook: simplify the organization, focus on high-growth segments, divest non-core assets, drive operational excellence.
The pattern: Anderson is running the AMD/Lattice playbook -- simplify, focus, execute. If he can get Coherent's operating margins from 9% (current GAAP) to 20%+ (which his segment profit margins suggest is achievable by stripping out restructuring/integration noise), the earnings power would be transformative.
| Optical Transceiver Market (2026) | $15.4B |
| Optical Transceiver Market (2031E) | $29.3B |
| Optical Interconnect Market (2026) | $21.9B |
| Optical Interconnect Market (2031E) | $40.0B |
| Silicon Photonics Market (2031E) | $13.2B |
| Five largest vendors share | ~60% |
| Networking Revenue (FY2025) | $3.42B |
| Datacom Revenue (est.) | ~$2.0-2.5B |
| Est. Share of Transceiver Market | ~15-18% |
| Top 5 vendor by revenue | Yes |
| NVLink-qualified | Yes (rare) |
Coherent is already a top-3 optical transceiver vendor globally. Its protocol-agnostic transceivers (supporting NVLink) give it access to the highest-growth segment -- NVIDIA GPU clusters.
The correct approach: start with the bull case terminal value, divide by 10, and compare to today's price. What entry price gives a 10x?
| Total Revenue | $10-12B | Networking doubles to $7B+, Industrial grows modestly, divested businesses exited |
| Networking Revenue | $7-8B | ~22% CAGR from $3.4B, driven by 800G/1.6T ramp + share gains |
| Gross Margin | 40-45% | Mix shift to higher-margin networking, vertical integration cost advantages |
| Operating Margin | 22-28% | Anderson's operational improvements, restructuring charges behind us |
| Net Income | $1.5-2.5B | After interest expense declines (debt paydown) and tax normalization |
| Shares Outstanding | ~190-195M | Modest dilution from SBC |
| EPS | $8-13 | vs forward EPS of ~$7.35 for FY2026 |
| P/E Multiple | 25-35x | High-growth photonics premium |
| Implied Stock Price | $200-455 | At trough-to-normal P/E on bull case earnings |
For a 10x over 5 years from entry, with FY2030 bull case EPS of $10-13 at 30x P/E:
| Bull Case Stock Price (2030) | $300-390 |
| 10x Entry Price | $30-39 |
| Last Seen at This Price | Jan-Oct 2024 ($28-45 range) |
Like Lumentum, the 10x window was in early-to-mid 2024. The stock was $28-45 for most of the first three quarters of 2024. It has already done a ~6-8x from its 2024 lows. From $241, the realistic upside is 1.5-2x over 3-5 years in the bull case.
At nearly identical market caps (~$45B), COHR has: 3.5x more revenue, positive FCF, lower leverage, faster networking growth, and a lower forward P/E (33x vs 43x). If you believe optical interconnects are the play, COHR gives you more revenue and more diversification per dollar of market cap. The main reason LITE trades at a premium is perceived "AI purity" -- but COHR's networking segment alone ($3.4B) is 2.4x the size of LITE's cloud & networking ($1.4B). The premium seems unjustified.
Unlike Lumentum, Coherent has positive net tangible assets ($0.45B) and real physical assets: $1.88B in PP&E (fabs, cleanrooms, manufacturing equipment), ~$900M in cash, plus the operating businesses generating $634M in annual operating cash flow.
Tangible Book Value Calculation:
| Stockholders' Equity | $8.13B |
| Less: Goodwill | -$4.47B |
| Less: Other Intangibles | -$3.20B |
| Tangible Book Value | $0.45B |
| Tangible BV / Share (187.5M shares) | $2.40 |
Tangible book of $2.40/share is not meaningful as a floor -- the company's value is in its operating businesses, not its balance sheet. But the operating cash flow ($634M/year) provides a floor on earning power:
| Trough Operating CF | $400-500M | Assume 25-30% decline from current |
| Less: Maintenance CapEx | -$250M | Rough estimate |
| Trough FCF | $150-250M | |
| Trough FCF Yield (at 3%) | $5-8.3B market cap | |
| Trough Stock Price (187.5M shares) | $27-44 | |
| With 50% Margin of Safety | $40-65 | Above trough but reasonable floor |
Confidence: Low-Moderate. Better than Lumentum because there are real assets and consistent FCF, but the goodwill-heavy balance sheet and dependence on AI demand cycle make any floor estimate uncertain. The floor exists but is 75-85% below current price, making this not a floor-price opportunity.
Coherent is the more interesting of the two optical plays. At nearly identical market caps, you get 3.5x more revenue, the deepest vertical integration in photonics, positive FCF, a proven new CEO running a transformation playbook, and a lower valuation multiple on every metric. The protocol-agnostic transceiver capability (including NVLink) positions Coherent at the heart of AI infrastructure.
Compared to Lumentum: If forced to pick one, Coherent is the better fundamental value. LITE is priced as a pure-play AI darling at 22x EV/Revenue; COHR at 8x gives you more business for less money. The market is paying a massive "AI purity" premium for LITE that seems hard to justify when COHR's networking segment alone is 2.4x larger.
The big risk is complexity. This is a company still integrating a $5.7B acquisition, restructuring aggressively, divesting businesses, and navigating tariff headwinds -- all while trying to capitalize on the AI demand wave. Jim Anderson's track record suggests he can execute, but the margin for error is thin.
| Current Price | $241 |
| Very Safe Entry | $40-65 |
| Fair Value Range | $200-280 |
| 5-Year Bull Case | $350-450 |
| Upside from Here (Bull) | 45-87% |
| Downside if Cycle Turns | -50% to -70% |
| 10x Entry Would Have Been | $28-39 (early-mid 2024) |
For Leopold's portfolio context: His $89M position in COHR is much smaller than his $479M in LITE. This could mean he views COHR as a smaller-conviction secondary play, or it could mean he views COHR as more expensive relative to his cost basis. Given both stocks were at similar prices in early 2024 ($30-45 range), the position sizing likely reflects a stronger conviction in LITE's AI purity and faster growth trajectory.
Our stance: More interesting than LITE at current prices due to better fundamental value. But still not cheap enough for our framework -- we need much lower prices to have genuine downside protection. If COHR pulled back to $100-130 (a realistic scenario in a market correction), the risk/reward would be compelling: you would be paying ~15x forward earnings for a top-3 global photonics company with AI tailwinds, consistent FCF, and an improving operating trajectory under a proven CEO.