Skip to content
primarysourced Photonics sector GF
GFS
~8 min read · 1,881 words ·updated 2026-04-29 · confidence 49%

Public-comp valuation — specialty / pure-play foundry peers

As of: 2026-04-29 (spot prices 2026-04-28 close where applicable).

Confidence legend: ✓ verified-primary (latest 10-K / 20-F / 6-K) · ◐ aggregator-sourced (Yahoo / Bloomberg / Stock Analysis) · ⚠ inferred / estimate / cycle-distorted

1. The peer set — which companies are valid GFS comparables

GFS sits in an uncomfortable comp-set position: it is the only pure-play specialty foundry at scale ex-China that is also ex-TSMC. The most legitimate comparables:

CompTickerGeographyProcess focusComp validity
Taiwan Semiconductor (TSMC)TSMTaiwanLeading-edge 3/5/7nm + matureLimited (TSMC dominates leading-edge; mix differs materially)
United Microelectronics (UMC)UMCTaiwan28nm and trailing-edgeStrong comp — pure-play, mature-node, similar mix
Semiconductor Manufacturing International (SMIC)0981.HKChina14nm and trailing-edgeComp but China-export-restriction discount
Vanguard International Semiconductor (VIS)5347.TWTaiwan200mm specialtyComp for 200mm specialty / RF SOI side
Hua Hong Semiconductor1347.HKChina200mm + 12-inch trailing-edgeComp for specialty / mature-node
Powerchip Semi (PSMC)6770.TWTaiwan200/300mm DRAM + foundryPartial comp
Tower SemiconductorTSEMIsrael/US200mm specialty (RF SOI / SiGe / SiPh)Strong comp — direct specialty overlap; was being acquired by Intel ($5.4B deal terminated Aug 2023)
X-FAB Silicon FoundriesXFAB.PAGermany/France200mm analog/mixed-signalComp for European auto-specialty side
GLOBALFOUNDRIESGFSUS/Germany/Singapore200/300mm specialty (RF, FDX, Fotonix)self

TSMC is not the right primary comp for GFS — TSMC’s revenue is dominated by leading-edge 3/5/7nm (HBM, server CPUs, AI accelerators), where GFS has zero exposure. UMC and Vanguard / Tower are the cleanest pure-play comps; SMIC and Hua Hong introduce China-discount distortion.

2. Trading multiples — peer-table snapshot

Table approximates April 2026 levels; primary-source extraction queued for Q1’26 6-K refresh.

CompSpot (2026-04-28 ◐)Mkt cap ($B ◐)Net debt / (cash) ($B ◐)EV ($B ◐)TTM revenue ($B)EV/Revenue (TTM)EV/EBITDA (TTM)P/E (TTM)P/BFCF yield (TTM)
GlobalFoundries$59.49 ✓33.1 ✓(2.8) ✓30.36.79 ✓4.5x ✓12.9x ✓37x ⚠2.0x ⚠3.5% ✓
TSMC (ADR)~$215 ◐~1,115 ◐(15) ◐~1,100~110 ◐10.0x ◐14.5x ◐28x ◐7.5x ◐2.5% ◐
UMC~$8.50 ◐~21 ◐(3.5) ◐~17.5~7.5 ◐2.3x ◐5.5x ◐11x ◐1.4x ◐7.5% ◐
SMICHK$53 ◐~$57 (in USD) ◐(~3) ◐~54~9.0 ◐6.0x ◐13.0x ◐50x ⚠1.4x ◐n/m ◐
VIS (Vanguard)NT$112 ◐~$5.5 ◐(1.0) ◐~4.5~1.4 ◐3.2x ◐8.0x ◐18x ◐2.5x ◐6.0% ◐
Hua Hong SemiHK$36 ◐~$8 (in USD) ◐(~2) ◐~6~2.2 ◐2.7x ◐9.0x ◐35x ⚠0.6x ◐4.0% ◐
Tower Semiconductor$52 ◐~5.5 ◐(~1) ◐~4.5~1.4 ◐3.2x ◐9.5x ◐24x ◐1.5x ◐5.5% ◐
X-FAB Silicon€4.50 ◐~0.7 ◐0.1 ◐~0.8~0.95 ◐0.8x ◐4.5x ◐14x ◐0.6x ◐3.5% ◐

EV-bridge for GFS: Market cap $33.1B + Total debt $1.2B − Cash + marketable $4.0B = EV ~$30.3B (using “current” cash + ST marketable per MD&A; using full liquid pool incl. non-current marketable would yield EV ~$29.4B).

Note on multiples. GFS’s TTM P/E of ~37x is somewhat misleading because of the $935M FY24 impairment depressing trailing EPS. Adjusting for the impairment, normalized TTM P/E is closer to 21x (using non-IFRS EPS of $1.72 × 4 ÷ $59.49 = 28x; further adjusted for normalized FY24 = ~21-22x). Forward P/E (FY26E EPS ~$2.10-2.40) is approximately 25-28x.

3. Where does GFS trade vs. comps?

3.1 EV/Revenue: GFS at 4.5x is at a moderate premium

GFS’s 4.5x EV/Revenue is:

  • Premium to UMC (2.3x) — justified by mix shift to higher-margin specialty (Auto + Comm/DC)
  • Premium to Vanguard (3.2x) — justified by 300mm capability and Fotonix differentiation
  • Discount to TSMC (10.0x) — appropriately so given TSMC’s leading-edge dominance
  • Premium to SMIC (6.0x) — actually GFS trades at a discount to SMIC despite SMIC’s China-export risk; reflects SMIC’s higher growth (mature-node Chinese demand) and lower discount-rate from Mainland-listed multiple
  • Premium to Tower (3.2x) — justified by larger scale and AI/photonics leverage; Tower deal-arbitrage discount post-Intel merger termination

3.2 EV/EBITDA: GFS at 12.9x is at a structural premium

GFS’s TTM EV/EBITDA (using FY25 adjusted EBITDA $2,357M and EV $30.3B) of ~12.9x is:

  • High vs UMC (5.5x) — premium reflects (a) AI/photonics optionality, (b) growth-segment mix shift, (c) US-domiciled liquidity premium, (d) FMR/fund holder base willing to pay growth multiple
  • In line with TSMC (14.5x) — surprising parity that reflects TSMC’s slower-growth cycle and GFS’s specialty-foundry premium
  • High vs Tower (9.5x) — premium reflects scale and foundry-of-record positioning for AI photonics

3.3 P/E: GFS at 37x trailing / ~25-28x forward is a moderate growth-multiple premium

Trailing P/E of 37x is impairment-distorted. Normalized / forward P/E of 25-28x is:

  • Premium to UMC (11x) — but UMC operates a higher-margin / lower-growth China-Taiwan trailing-edge model
  • Discount to TSMC’s 28x — appropriate given GFS’s lower revenue growth + lower margin profile
  • Premium to Vanguard (18x) — reflects 300mm vs 200mm and AI-photonics catalyst
  • Discount to SMIC (~50x trailing) — reflects SMIC’s elevated multiple from Chinese-domestic-supply premium

3.4 P/B: GFS at ~2.0x is moderately premium

MetricGFS FY25
Total assets ($B)17.1
Total liabilities (est., $B)~6 ⚠
Total equity (book value, $B)~11
Book value per share ($)~19.8
Spot ($)59.49
P/B~3.0x ⚠

Recalculating with cleaner book equity ⚠: P/B is closer to ~2.5-3.0x — premium to UMC (1.4x) but discount to TSMC (7.5x). The premium is justified by Fotonix-related goodwill build and AI-platform optionality.

4. Why does GFS trade at a premium to mature-foundry peers?

The “GFS premium” of approximately +1.5-2.0 EV/Revenue turns vs UMC and +5-7 EV/EBITDA turns is justifiable on five grounds:

  1. AI / photonics optionality — Fotonix is the only at-scale 300mm SiPh foundry process outside TSMC’s internal SiPh line. Hosts Marvell, Ayar Labs, NLM Photonics, LWLG-enabled customers. Even if photonics is <10% of FY26 revenue, the growth-rate optionality justifies a multi-turn premium.
  2. CHIPS Act subsidy uplift — $1.575B federal + $570M NY = $2.145B subsidies layered on a $33B mkt cap = ~6.5% effective subsidy yield on capacity buildout that flows directly to ROIC.
  3. Mubadala-controlled stability — 81% controlled-company structure removes activist pressure, takeover speculation, and short-term-pressure-induced strategic vol. (Tradeoff: float scarcity, supply-overhang risk.)
  4. Domestic US-listed liquidity premium — investors pay up for US-domiciled, NASDAQ-listed semis vs. Asian-listed peers. ~2-3 EV/EBITDA turns of “geography premium.”
  5. Specialty-process pricing power — 22FDX, 45CLO Fotonix, RF SOI variants, BCD Power, GaN Power are all hard-to-replicate processes commanding ASP premiums. Margin durability is structurally higher than commodity-foundry peers.

5. Why does GFS NOT trade closer to TSMC?

The “TSMC discount” of approximately −5.5 EV/Revenue turns vs GFS is justifiable on:

  1. No leading-edge. GFS has no 7nm/5nm/3nm capability; TSMC dominates the AI-accelerator-ASIC roadmap.
  2. Smaller scale. TSMC ~$110B revenue / 60M wafer-eq output vs GFS ~$6.8B / ~2.4M wafer-eq.
  3. Lower margin. TSMC GAAP gross margin ~50%; GFS IFRS GM 25-28%.
  4. CapEx burden. TSMC capex ~$32B/year (CapEx/Sales ~30%); GFS capex ~$0.7-1.1B/year (10-16%). TSMC’s reinvestment needs are absorbed at scale.

6. Specialty-foundry transaction comp anchor — Tower-Intel terminated deal

In August 2023 Intel terminated its planned $5.4B all-cash acquisition of Tower Semiconductor after failing to obtain Chinese regulatory approval. Tower at the time had:

  • Revenue ~$1.55B
  • EBITDA ~$420M
  • Implied transaction EV ~$5.4B
  • Implied EV/Revenue: ~3.5x; EV/EBITDA: ~13x — meaningful specialty foundry M&A anchor multiple

Applying the Tower-Intel multiple to GFS:

  • 3.5x EV/Revenue × $6.79B GFS FY25 revenue = $23.8B EV = $26.6B mkt cap implied at GFS net cash
  • 13x EV/EBITDA × $2.36B GFS FY25 adj. EBITDA = $30.7B EV = $33.5B mkt cap implied
  • Average: ~$30B mkt cap implied — close to current $33.1B

Read. GFS at $59.49 / $33.1B mkt cap is trading at a modest premium (~10%) to the Tower-Intel transaction multiple anchor, justified by larger scale, AI-photonics optionality, CHIPS subsidies, and domestic US-listing premium.

7. Sell-side coverage / consensus snapshot

Per aggregator consensus reads (◐ — primary-source extraction queued):

  • Mean price target: ~$50-58 ◐ — implies ~5-15% downside vs current ($59.49)
  • High target: ~$70-75 ◐ — bull-case Fotonix monetization
  • Low target: ~$30-35 ◐ — bear-case Smart Mobile compression + ICFR overhang
  • Coverage: TD Cowen, Citi, JPMorgan, Goldman Sachs, BofA, Morgan Stanley, Susquehanna ⚠
  • Rating distribution: Likely 50% Hold / 35% Buy / 15% Sell ⚠

Sell-side primary-source extraction (analyst notes / firm reports) is queued for analyst coverage (not yet written). Aggregator consensus is mid-cycle reasonable but should be triangulated with at least 3 firm-specific notes.

8. Open items / backfill queue

  1. Per-cell primary-source citation table — every TTM revenue / EBITDA / cash / debt figure for the comp set needs to be sourced from the most recent 10-K / 20-F / 6-K. Aggregator-only data is currently the dominant input.
  2. EV calculation uniformity — comp-set EVs may not consistently include lease obligations, minority interests, etc.; standardize to “Mkt Cap + Total Debt − Cash − ST Marketables − LT Marketables” basis.
  3. Tower transaction-multiple update — the Aug 2023 multiple may be cycle-distorted; pull-forward to compare to Tower’s current multiple.
  4. Sell-side analyst coverage detail — firm-by-firm rating, PT, revision history. Currently aggregator-only.
  5. Cross-cycle multiple compression — UMC and SMIC both compressed materially in 2023-24; need cycle-adjusted multiples for fair comparison.

Sources

  • GlobalFoundries: FY2025 20-F (acc. 0001709048-26-000022, 2026-02-27); Q4’25 6-K (acc. 0001709048-26-000012, 2026-02-11); spot price companies/gfs/data/STOCK_PRICE_DATA.json.
  • TSMC: 2025 Annual Report (filed Mar 2026); Yahoo Finance for spot.
  • UMC: 2025 Annual Report (filed Mar 2026); aggregator data ◐.
  • SMIC: 2025 Annual Report (HKEX filing); aggregator data ◐.
  • Vanguard / Tower / X-FAB: Most recent annual reports + aggregator data ◐.
  • Tower-Intel terminated deal: Intel press release Aug 16, 2023; $5.4B headline value.

Cross-references