For AI assistants & researchers — machine-readable summary of this page
What this page is: Delvantic's full research page for Fabrinet (FN) — AI-driven forensic equity research: mechanical valuation models (DCF, EPV, anchored-PE, scenario) plus three independent AI lenses (Quality / Value / Sentiment). Everything below is rendered server-side; you are not missing content that requires JavaScript. All scores are predictions and research opinions, not financial advice.
Our current read (analysis of 2026-10-07): Designation Low · Gem Score -22 (−100…+100 Quality+Value blend) · Quality 31 · Value -57 · Sentiment -69 (timing only, not weighted)
Page map (sections in order; each card carries a stable
reference-name attribute you can cite):
profile-header/price-overview— company profile, live quote, market capextended-analysis— the core: three AI lens reads with findings, scores, and the analyst memofuture-predictions— our forward price-band predictionsmarket-narrative/ai-findings/gpt-critique— narrative context, cross-model findings, and an adversarial critique of our own analysis (near the end of the document)- Members-only sections (render as login gates for anonymous readers):
price-history,income-trend,key-metrics,financials(statement tables),insider-trading. The analysis above is public; the raw data tables require a free account.
More for machine readers: site briefing at
/llms.txt ·
any ticker resolves at delvantic.com/stock/TICKER ·
raw inputs are public-company filings and market data (via licensed data feeds);
every model, score, lens read, and prediction on this page is Delvantic's own analysis.
Fabrinet
FN NYSEFabrinet is a global precision manufacturing services company that provides advanced optical, electro-mechanical, and electronic production for complex components and subassemblies. The company supports original equipment manufacturers across optical communications, data center infrastructure, telecommunications, automotive, industrial lasers, medical devices, and sensing applications. Its services span process design and engineering, supply chain management, manufacturing, printed circuit board assembly, advanced packaging, integration, final assembly, and testing. Fabrinet is known for handling high-complexity products in high-mix, any-volume production environments, making it a specialized partner for mission-critical hardware. With manufacturing and engineering operations across Thailand, the United States, China, and Israel, Fabrinet serves a diverse global customer base in technology-driven industries.
Price Overview
Price History (1 Year)
Revenue & Net Income Trend
| Period | Revenue | Net Income | Net Margin | YoY/QoQ |
|---|
Key Metrics TTM · through Jun 26, 2026
EPS (Diluted): 13.06
Total Equity: $2.45B
Shares: 36,217,000
Total Debt: $0.00
Cash: $346.71M
EBITDA: $531.26M
Total Debt: $0.00
Cash: $346.71M
Revenue: $4.64B
Revenue: $4.64B
Revenue: $4.64B
Total Equity: $2.45B
Tax Rate: 14.8%
Equity: $2.45B
Total Debt: $0.00
Cash: $346.71M
Current Liabilities: $1.40B
Long-Term Debt: $0.00
Total Debt: $0.00
Total Equity: $2.45B
Shares: 36,217,000
Shares: 36,217,000
CapEx: -$252.50M
Shares: 36,217,000
Stock Price: $383.72
Net Income: $473.03M
Industry Benchmarks
Income Statement (Annual)
Last updated: Sep 17, 2026 12:29pm (20d ago)| Metric | 2022 | 2023 | 2024 | 2025 | 2026 |
|---|---|---|---|---|---|
| Revenue | $2.3B | $2.6B | $2.9B | $3.4B | $4.6B |
| Cost of Revenue | $2.0B | $2.3B | $2.5B | $3.0B | $4.1B |
| Gross Profit | $278.6M | $336.3M | $356.1M | $413.3M | $556.5M |
| Operating Expenses | $74.1M | $84.6M | $78.5M | $88.9M | $93.6M |
| Operating Income | $204.5M | $251.7M | $277.6M | $324.4M | $462.9M |
| Net Income | $200.4M | $247.9M | $296.2M | $332.5M | $473.0M |
| EBITDA | $243.3M | $295.5M | $326.6M | $377.9M | $531.3M |
| EPS | $5.43 | $6.79 | $8.17 | $9.23 | $13.21 |
| EPS (Diluted) | $5.36 | $6.73 | $8.10 | $9.17 | $13.05 |
Balance Sheet (Annual)
Last updated: Sep 17, 2026 12:35pm (20d ago)| Metric | 2022 | 2023 | 2024 | 2025 | 2026 |
|---|---|---|---|---|---|
| Cash & Equivalents | $198.0M | $231.4M | $410.0M | $306.4M | $346.7M |
| Total Current Assets | $1.5B | $1.7B | $2.0B | $2.4B | $3.2B |
| Total Assets | $1.8B | $2.0B | $2.3B | $2.8B | $3.9B |
| Current Liabilities | $538.5M | $481.9M | $557.9M | $809.8M | $1.4B |
| Long-Term Debt | $15.3M | $0 | — | — | — |
| Total Liabilities | $582.0M | $511.0M | $592.8M | $849.6M | $1.5B |
| Total Equity | $1.3B | $1.5B | $1.7B | $2.0B | $2.5B |
| Retained Earnings | $1.2B | $1.5B | $1.8B | $2.1B | $2.6B |
Cash Flow (Annual)
Last updated: Sep 17, 2026 12:35pm (20d ago)| Metric | 2022 | 2023 | 2024 | 2025 | 2026 |
|---|---|---|---|---|---|
| Operating Cash Flow | $124.2M | $213.3M | $413.1M | $328.4M | $256.7M |
| Capital Expenditure | -$89.6M | -$61.4M | -$47.5M | -$121.1M | -$252.5M |
| Free Cash Flow | $34.7M | $152.0M | $365.6M | $207.3M | $4.2M |
| Acquisitions (net) | — | — | — | — | — |
| Net Debt Issued / (Repaid) | -$12.2M | -$15.2M | -$12.2M | — | — |
| Dividends Paid | — | — | — | — | — |
| Stock Buybacks | -$59.9M | -$47.6M | -$39.5M | -$125.7M | -$5.2M |
| Net Change in Cash | — | — | — | — | — |
Growth Trends (YoY %)
Last updated: Sep 17, 2026 12:29pm (20d ago)| Metric | 2023 | 2024 | 2025 | 2026 |
|---|---|---|---|---|
| Revenue Growth | +16.9% | +9.0% | +18.6% | +35.7% |
| Gross Profit Growth | +20.7% | +5.9% | +16.1% | +34.6% |
| Operating Income Growth | +23.1% | +10.3% | +16.9% | +42.7% |
| Net Income Growth | +23.7% | +19.5% | +12.3% | +42.3% |
| EBITDA Growth | +21.5% | +10.5% | +15.7% | +40.6% |
Deep Analysis
Risk : Reward — upside vs downside from this company's own quarters
Not computed yetNarrative Economics
market-narrative step).
Growth Outlook
Analyzed 2026-09-17 17:00The question every valuation on this page silently assumes: is this company likely to grow? Judged forward — the business, its category, the world — against what's already printed.
Claude Reading
Independent read first: Fabrinet's quarterly trajectory is genuinely impressive — revenue accelerated from $804M in Sep-2024 to $1.32B in Jun-2026 (+64% over seven quarters), and net margin expanded from 9.6% to 10.6% even as the base scaled. Full-year FY26 revenue of $4.64B vs. FY25's $3.42B is +35.7% growth, with NI going $332.5M → $473.0M (+42%). That is not a "mature earner" — the rule-based classifier is miscalibrated. This is a hyper-cyclical cash earner riding a specific wave (800G/1.6T optical transceivers for AI clusters, primarily Nvidia via Coherent/Marvell pass-through). Return on equity of 21% with zero debt and $346M cash is high-quality, but note the balance sheet: equity is $2.45B, so the company is earning ~19% on tangible capital in a business that historically deserves an EMS multiple of 12–18x.
The critical anomaly no model foregrounded aggressively enough: FY26 operating cash flow was $256.7M against net income of $473M — a 54% cash conversion, versus what should be ~100%+ for a mature contract manufacturer. Capex of $252.5M consumed essentially all of it, leaving FCF of $4.2M on $4.64B revenue. That's a 0.09% FCF margin against a 10.2% net margin. Either working capital is ballooning (inventory/AR to support the growth ramp — plausible and benign) or earnings quality is deteriorating (concerning). The "fcf_cagr: -89.3%" momentum flag captures this but nobody adjudicated which cause. At 29x P/E and effectively infinite P/FCF, you're paying accrual-basis earnings multiples on a business currently generating negligible free cash. That's the single most important number in the file and it warrants a real haircut.
On the prior models: Pre-Flight's "mature contract manufacturer with stable margins" is directionally right on business model but tone-deaf to the +42% earnings growth happening right now — this is a mature model in a cyclical peak, which is different. Market Narrative's claim of "30%+ operating margins" is factually wrong — operating margin is 9.97%, gross is 12%. That's a serious error and it means the DCF fair value of $329 anchored to those margins is inflated; correct the margin assumption and fair value drops materially. Valuation Synthesis's composite fair value of $255 (unadjusted) vs. $384 price implies ~33% overvaluation, which I find more credible than the signal-adjusted $329. Insider selling is real but modest in size (2,500 share lot is the largest) — I wouldn't overweight it as "peak signal," but combined with FCF collapse it rhymes.
Contrarian case worth stating: if the AI optical buildout has another 2-3 years of runway (1.6T ramps in 2026, 3.2T in 2027-28), FN's revenue could plausibly hit $6B+ by FY28 at maintained ~10% net margin = $600M NI, putting forward P/E on FY28 numbers at ~23x. That's not cheap but not absurd for a name compounding at 25%+ with net cash. The bear case is simpler: this is Celestica/Jabil economics with a temporary multiple. Historical EMS multiples of 15-18x on FY26 EPS of ~$13 gets you $195-235, which frames the downside if the cycle rolls. Customer concentration (Nvidia ecosystem exposure via Coherent) is the specific fat-tail risk — one inventory correction quarter re-rates the multiple hard, as the 49% drawdown from highs already demonstrated.
GPT Reading
What jumps out is that Fabrinet is not just growing; it is accelerating while holding margins unusually well for a manufacturer. TTM revenue is $4.64 billion, up 36% from $3.42 billion a year earlier, and the quarterly run-rate rose each quarter from $978 million in September 2025 to $1.32 billion in June 2026. Net income followed the same staircase, from $95.9 million to $139.3 million over those four quarters, with net margin improving from 9.8% to 10.6%. Over five years, revenue has doubled from $2.26 billion to $4.64 billion and net income has more than doubled from $200 million to $473 million. For a business often framed as a low-multiple contract manufacturer, these are not low-quality, scrape-by numbers: 12.0% gross margin, 10.0% operating margin, 10.2% net margin, 21.3% ROE and 18.7% ROIC with no debt is a very strong operating profile.
The part of the story I think the bearish models underweight is just how rare this combination is: 35%-plus growth, sequential acceleration, rising earnings, zero leverage, and positive absolute free cash flow even in a capex spike year. The market cap is $14.1 billion against $473 million of TTM earnings, so yes, 29x earnings is not cheap on a static basis. But static thinking is the mistake here. If the latest quarter of $139 million in earnings were simply annualized, earnings power is already closer to $557 million, implying more like 25x on a forwardized current run-rate before any further growth. On sales, 3.0x revenue for a business compounding near 27% over several years with double-digit net margins is not obviously excessive. Plenty of electronic manufacturers never approach this margin structure or return profile. The absence of debt also matters: enterprise value is only modestly above equity value because Fabrinet has $347 million in cash, so the balance sheet is a real cushion if demand cools.
That said, the single biggest yellow flag is cash conversion. Operating cash flow was only $256.7 million against $473.0 million of net income, and free cash flow collapsed to just $4.2 million because capex surged to $252.5 million. Some of that can be healthy if management is building capacity into genuine demand, but it means reported earnings are currently doing far more of the valuation work than cash. If this capex is cyclical catch-up rather than the base for durable new volume, then today’s multiple is far less comfortable than the income statement suggests. The other issue is simply that at $383.72, the stock no longer offers much protection if growth drops from the current mid-30s pace toward a more ordinary mid-teens rate. A high-quality manufacturer can deserve a premium, but once you pay near-30x trailing earnings for a company with 12% gross margins, you are underwriting continued execution with little room for disappointment.
The best argument against my more constructive read is straightforward: this may be a peak optical/data-center buildout year, and the market has already capitalized it as if the new run-rate is sustainable. Bears would point to the ugly free cash flow, the elevated EV/EBITDA of 25x, the small but persistent insider selling, and the fact that manufacturers tied to concentrated customers can look safest right at the top. They would also note that historical quarter margins have mostly sat in the 9%-10% net range, so even the recent improvement to 10.6% is not a structural break that guarantees re-rating. I take those points seriously. But the actual reported numbers still show demand strengthening quarter after quarter rather than rolling over, and unlike many “AI adjacency” names, Fabrinet is not funding the story with leverage or losses. The skepticism is understandable; I just think it has to be balanced against evidence that the business is genuinely scaling.
What would change my mind is not a modest deceleration; it would be proof that growth is normalizing faster than capex and working capital can unwind. If quarterly revenue stalls near $1.3 billion or drops while net margin slips back below 9.5%, the earnings power case breaks quickly. Likewise, if another year of near-$450 million to $500 million net income still produces weak operating cash flow and negligible free cash flow, then the “investment phase” defense stops working and the premium multiple becomes untenable. On the other hand, if Fabrinet can convert this new revenue base into even $300 million-plus of annual free cash flow while keeping quarterly revenue above $1.25 billion and net margins around 10%, then today’s price will have been justified and probably not the top.
Grok Reading
Fabrinet’s trailing twelve months through June 2026 tell a clear operating story: revenue of $4.64B, up 35.7% from the prior year’s $3.42B, with net income of $473M (+42%) and net margins grinding higher from the mid-9s into 10.6% in the latest quarter. Sequential quarters climbed $978M → $1.13B → $1.21B → $1.32B, so the growth is not a single-print spike. Returns are respectable for an EMS name—ROE 21.3%, ROIC 18.7%—on a fortress balance sheet with zero debt, $347M cash, and a 2.25 current ratio. That is genuine scale and execution in optical and data-center interconnect manufacturing. What the income statement celebrates, the cash-flow statement refuses to confirm: operating cash flow of only $257M against $252M of capex left free cash flow at $4.2M. FCF CAGR of –89% is not noise; the company is converting almost none of its reported earnings into distributable cash while it builds capacity. At $383.72 the stock trades 29.4× trailing earnings, 3.0× sales and 25× EV/EBITDA—multiples that embed continued mid-20s-plus growth and a rapid reversion of cash conversion, neither of which is visible in the current run-rate.
The contradiction with several of the quantitative overlays is exactly here. Classification as a “mature earner” fits the margin stability and balance-sheet conservatism, yet 27% revenue CAGR and heavy growth capex look more like a mid-cycle capacity sprint than a steady compounder. The valuation synthesis fair-value band around $255–$329 versus the $384 print is directionally right: almost no normalized FCF multiple supports the current price unless one assumes the $252M capex wave is temporary and immediately volume-accretive at today’s 12% gross / 10% operating margins. Insider activity—repeated open-market sales through 2026 with negligible offsetting purchases—aligns with that skepticism rather than with confidence that the AI-optical boom has years of runway left at these rates. Revenue confidence already flags deceleration in the quarterly trend; if that continues, the 29× multiple has nowhere to hide.
The strongest case against a bearish lean is straightforward and data-backed. Earnings power has compounded at 26% for half a decade, gross profit dollars rose from $279M in FY2022 to $557M now, and the company still posts double-digit incremental margins while remaining net-cash. If the capex is simply bringing online capacity already contracted by hyperscaler optical programs, FCF can snap back above $200M within two years and the stock is merely early-cycle expensive rather than structurally mispriced. Customer stickiness in precision optical assembly is real; share gains against lesser EMS peers could sustain mid-teens growth longer than a generic contract-manufacturer multiple implies. In that framing, 25× EV/EBITDA is a growth toll, not a bubble, and the 16% premium to the engine’s $329 anchor is modest narrative juice rather than detachment.
I would flip to constructive on two concrete prints: trailing FCF recovering above $180–200M (i.e., capex falling back under ~$100M while EBIT holds) while quarterly revenue growth stabilizes at or above 15% year-on-year rather than sliding into single digits. A clean demonstration that the largest customers have multi-year volume commitments that fill the new capacity would also neutralize the concentration and cycle-peak objections. Until those numbers appear, the price at $384 is paying for perfection the cash flow statement has not yet delivered.
Qwen Reading local experiment · off-panel
The most glaring contradiction in this dataset is the divergence between the explosive top-line growth and the collapsing cash conversion. Fabrinet is posting a 26.9% revenue CAGR and recently accelerated to 35.7% YoY growth, with net margins expanding from 9.3% to 10.6% over the last eight quarters. This is not the profile of a "mature earner" as the rule-based model suggests; it is a high-growth compounder. However, the Free Cash Flow CAGR of -89.3% is a red flag that the quantitative models are underweighting. In the latest annual period, operating cash flow was $256.7M against net income of $473.0M, a conversion ratio of only 54%. This gap is driven by a massive $252.5M capex spend, which is nearly equal to the entire operating cash flow. The market is pricing this at 29.4x trailing P/E and 3.0x P/S, which is a premium multiple for a contract manufacturer. The "quiet-quality" narrative is seductive, but the balance sheet shows zero debt and $346.7M in cash, meaning the company is funding this aggressive capacity expansion internally. The risk is not solvency, but the opportunity cost of capital if the AI/optical demand cycle normalizes. The 10.6% net margin is impressive for an EMS player, but it is being achieved by spending heavily on assets that may become obsolete if the hyperscaler capex cycle turns.
The strongest case against the current valuation is the insider activity combined with the "decelerating" revenue trend flagged in the secondary signals. While the recent YoY growth is 35.7%, the quarterly revenue progression from $1.32B (Q1 2026) to $1.21B (Q2 2026) suggests a sequential slowdown, or at least a plateau, after a period of rapid acceleration. The insider sales, while small in absolute share count (1,420 shares in Sept 2026), are consistent and frequent, with multiple sales in Feb 2026 and May 2026. This is not a panic dump, but it is a steady trickle of selling that often precedes a period of lower growth. The "Unusual Heavy Selling" tag is likely an overstatement given the small size, but the pattern is notable. More importantly, the FCF quality is weak. A company with 21.3% ROE and 18.7% ROIC should be generating more cash. The fact that it is not suggests that the earnings are being consumed by working capital or capex. If the AI infrastructure buildout slows, Fabrinet will be left with a bloated asset base and lower utilization, which would crush margins. The 29x P/E is not justified if the growth rate drops to single digits, which is a realistic scenario for a cyclical manufacturer.
What would change my mind is the next two earnings reports. If Fabrinet can maintain net margins above 10% while growing revenue at 20%+ YoY, and if the operating cash flow conversion improves to above 70% of net income, then the current valuation is supported. The key metric to watch is the capex-to-revenue ratio. If it stays above 15% of revenue, the company is in a heavy investment phase, and the FCF will remain depressed. If it drops below 10%, the FCF will surge, and the stock will likely re-rate higher. Conversely, if the revenue growth decelerates below 15% and margins compress below 9%, the 29x P/E will look expensive, and the stock could easily fall to the $250-$300 range, which is where the DCF models are clustering. The insider selling is a minor concern, but the cash flow quality is the major issue. I am not convinced that the market is fully pricing in the risk of a capex cycle peak, but I am also not convinced that the growth is sustainable at the current rate. The stock is fairly valued, but with a high risk of downside if the AI cycle normalizes.
Big-3 Panel — where each AI stands
Advanced Analysis Forensic deep-dive · separate lenses
Fabrinet has scaled revenue from $2.26B (2022) to $4.64B (TTM 2026) - roughly a 20% CAGR - while holding gross margin in a tight 12.0-12.7% band and operating margin at ~9-10%. Net income compounded from $200M to $473M, and diluted shares actually fell from 37.5M to 36.2M with buybacks running 1.8x SBC (SBC only 0.8% of revenue). The balance sheet is a fortress: $875M net cash (no debt netted), Altman Z of 8.83, and the company self-funds growth.
Verify before trusting this (6)
- Working-capital detail (DSO, inventory days) explaining the TTM FCF drop to $4.2M vs $473M net income
- Customer concentration - % revenue from top 1-3 customers (likely Cisco/NVIDIA/hyperscalers)
- Segment/product mix - exposure to AI-optical (800G/1.6T transceivers) vs legacy telecom
- Whether the receivables/inventory build has reversed post-quarter
- Capex trajectory - is FCF pressure from growth capex or pure working capital
- Terms and remaining authorization of the buyback program
The composite FV of $255.22 and signal-adjusted FV of $328.66 both sit below the $380.91 price, implying roughly -14% to -33% downside on the core methods. DCF ($127) and EPV floor ($106) look punitively low for a business growing ~20% with a fortress balance sheet - I discount those as too harsh for a still-compounding contract manufacturer. The anchored PE of $660 is the mirror-image outlier, extrapolating a peak AI-optical cycle multiple onto 12% gross margin work; I discount that as runaway. Splitting the difference, deserved value clusters in the $300-360 range. The strong quality grade (31) legitimately lifts the deserved number toward the top of that band, but not through it. What is priced in at $380: continued hyperscaler capex intensity, no customer-concentration shock, and margin defense despite structurally thin gross margins. That is the bull case as the base case - a classic 'priced for the good outcome' setup. It is not egregious, but there is no discount here; the earnings-quality hint is only 'adequate,' which argues against paying up.
Verify before trusting this (4)
- Customer concentration disclosure (Cisco/NVIDIA/Ciena share of revenue) in the 10-K
- Forward guidance on 800G/1.6T optical program ramps and hyperscaler order book
- Whether the recent FCF collapse is working-capital timing or structural
- Gross margin trajectory - any signs of pricing pressure from competing EMS players
The tape is risk-off (VIX 17.7, S&P off highs) and FN carries a 1.21 beta, so it takes above-market punishment by default. Worse, the specific narrative pocket it lives in - AI/optical infrastructure - is actively cracking: Corning's surprise ATM equity offering just pulled Coherent, Lumentum and FN down together, signaling the market is questioning whether the whole cohort's run is done. FN is down 22.3% since its last earnings print despite beating on revenue and data center growth, because the market punished margin compression and weak FCF - a classic sign the narrative has flipped from 'buy the AI backbone' to 'sell the cycle top.' The 'quiet-quality' archetype with only moderate intensity and low cult following gives FN no fanbase to defend the story on down days. Analyst tone and news flow are turning skeptical, momentum has rolled from strong-positive into a sharp drawdown, and the bear framing (cyclical contract manufacturer, customer concentration, cycle maturing) is now the dominant lens. Net: a real, persistent headwind - not a panic, but a sustained de-rating pressure specific to this sub-sector and this name.
Verify before trusting this (4)
- Whether the optical cohort (COHR, LITE, FN) stabilizes together or continues to bleed - a group reversal would flip the read
- Any hyperscaler capex commentary that either reaffirms or breaks the AI-backbone story
- Analyst target revisions in the next 2-4 weeks following the post-earnings drop
- Whether VIX rolls back under 15 and S&P reclaims highs - would mute the macro leg of the headwind
The dominant exogenous force is hyperscaler AI capex, which is still being raised rather than trimmed, and the networking layer is currently the bottleneck — optical interconnect content per accelerator is rising faster than accelerator units. That is a genuine tailwind Fabrinet sits directly beneath. Against it: macro backdrop is flagged as headwinds with a 10y near 5%, which raises the cost of the capacity capex Fabrinet is currently absorbing and pressures the non-AI legs (telecom carrier spend, industrial lasers, automotive sensing) that historically smoothed its cycle. Geographic concentration in Thailand carries tariff and baht-translation exposure that is outside management's control, though it has also been a relative advantage versus China-domiciled competitors as supply chains diversify. Net: the world is currently pushing this business forward harder than it is pushing back, but the push comes from one concentrated source whose spending cadence Fabrinet does not set.
Prediction unavailable. valuation-synthesis has no result for FN — the prediction needs its fair-value anchors.