Skip to content
primarysourced Photonics sector Tower
TSEM
~13 min read · 2,995 words ·updated 2026-04-29 · confidence 35%

DCF assumptions — load-bearing inputs

Scope. This file enumerates the inputs that drive a Tower DCF — it does not run a complete model. The full triangulated valuation triangulation lives in the bull-case thesis and bear-case thesis. Use this file as a reference for any per-share fair-value scenario.

As of: 2026-04-29 (data through Q4 2025 6-K release; spot $191.54)

Confidence legend: ✓ verified-primary 20-F / 6-K · ◐ aggregator / market data · ⚠ inferred / estimate / scenario

1. Capital structure / WACC anchor

ComponentValueNotes / Source
Risk-free rate (10Y UST)4.30% ◐April 2026 yield curve; Fed-rate-cycle-adjusted
Equity risk premium5.5% ⚠Mid-cycle US ERP per Damodaran-style framework
Israeli equity risk premium adjustment+1.0-1.5% ⚠Tower-specific; Israel sovereign + geopolitical-risk premium per industry practice
Beta (5-yr levered)1.4 ✓Per STOCK_PRICE_DATA.json
Cost of equity = Rf + β × ERP + Israel premium~12.5-13.0%Computed
Pre-tax cost of debt6.0% ⚠Israeli bank facilities + structured debt
After-tax cost of debt (T = 16% Israeli Approved Enterprise effective)5.0%Computed
Debt / (Debt + Equity) at market values~0.8%$181M debt / $21.4B mkt cap
Equity / (Debt + Equity)~99.2%
WACC (base)~12.4% ⚠Near cost-of-equity (de minimis debt)
WACC range (sensitivity)9-12%Bear: 12% / Base: 11% / Bull: 9.5%

Read on Tower WACC. Tower is structurally near cost-of-equity because of its de-minimis debt load ($181M vs $21.4B mkt cap = 0.8% leverage). This is meaningfully higher than typical US specialty-foundry peers (GFS WACC ~12.3% but with 3.5% leverage; UMC ~10-11%; TSMC ~9-10%) because of the Israeli equity risk premium adjustment which adds ~100-150 bps to the cost of equity.

Sensitivity to Israel risk premium. If Tower’s Israel risk premium compressed from +1.0-1.5% to 0 (e.g., post-Israeli-geopolitical-risk de-escalation), WACC drops from ~12.4% to ~11%, and DCF fair value rises by approximately +15-20% at constant terminal growth. The Israel-risk-premium is the largest single discount-rate sensitivity in the Tower DCF.

2. Terminal growth assumptions

ScenarioTerminal growth (g)Implied terminal yearComment
Bear2.0%2032+Mature-foundry secular = global semis CAGR; specialty-cycle reversion
Base3.0%2032+Sector-mature with Tower-specific currency drift + AI-photonics structural lift
Bull4.0%2032+AI / Comm-DC mix shift sustains above-average growth + LWLG EO-polymer monetization

Implied EV/EBITDA exit multiples (using 11% WACC base):

  • Bear (g=2%): 1 / (12% − 2%) = 10.0× EV/Forward-period-EBITDA
  • Base (g=3%): 1 / (11% − 3%) = 12.5×
  • Bull (g=4%): 1 / (9.5% − 4%) = 18.2×

These exit multiples are consistent with the comp-set trading multiples (UMC ~5.5x, SMIC ~13×, VIS ~8×, GFS ~12.9×, TSMC ~14.5×). The bull-case 18× is aggressive but defensible if the FY 2028 model materializes and SiPh growth sustains into the late 2020s.

3. Revenue forecast — bull / base / bear scenarios

Build-up of FY 2026-FY 2030 revenue from the FY 2025 base of $1,566M:

Bear case — SiPh ramp underperforms; specialty cyclicality returns ⚠

SegmentFY 2025 ($M)FY 2026EFY 2027EFY 2028EFY 2029EFY 2030E5Y CAGR
RF Infrastructure423480580700820950+17.5%
RF Mobile360340340360380400+2.1%
Sensors and Displays251260280300320340+6.3%
Power Management251270300330360390+9.2%
Discretes / Other282250230210200190−7.6%
Total1,5661,6001,7301,9002,0802,270+7.7%

Base case — management trajectory ⚠

SegmentFY 2025 ($M)FY 2026EFY 2027EFY 2028EFY 2029EFY 2030E5Y CAGR
RF Infrastructure4236008501,1501,4001,650+31.3%
RF Mobile360380420480540600+10.8%
Sensors and Displays251280320370420470+13.4%
Power Management251290340400460520+15.7%
Discretes / Other282250220200190180−8.5%
Total1,5661,8002,1502,6003,0103,420+16.9%

Bull case — full FY 2028 model + extension ⚠

SegmentFY 2025 ($M)FY 2026EFY 2027EFY 2028EFY 2029EFY 2030E5Y CAGR
RF Infrastructure4237001,1001,5001,9502,450+42.0%
RF Mobile360420500600700800+17.3%
Sensors and Displays251300360430500580+18.2%
Power Management251320400500600700+22.8%
Discretes / Other282260240220210200−6.6%
Total1,5662,0002,6003,2503,9604,730+24.7%

The load-bearing segment is RF Infrastructure. In the base case, this segment goes from $423M (FY 2025) → $1,150M (FY 2028) at +31% CAGR — driven by SiPh + SiGe scaling on the December 2026 capacity > 5× Q4 2025 actual target with 70%+ reserved through 2028 customer-prepaid. In bull case, $1,500M FY 2028 = 46% of total revenue — pure-play AI-optical-interconnect monetization. This segment carries structurally higher gross margin (~35-45%) than corporate average (~30%), making it the dual-lever of growth + margin expansion.

Management-disclosed FY 2028 revenue target = $2.84B (TipRanks ✓). The base case ($2.60B) is 8% below management — appropriate analyst conservatism. The bull case ($3.25B) is +14% above management — would require LWLG-Tower commercialization + secondary CPO ramp + envelope-tracker Apple-PMIC scale all converging by FY 2028.

4. Margin trajectory — the FY 2025 trough → FY 2028 inflection

Per margins and pricing, Q4 2025 IFRS GM was 26.8% with Q1 2025 trough at 20.4%.

YearIFRS GM (analyst)Operating marginAdjusted EBITDA margin
FY 2023 actual ✓~24.9%~13% ⚠~30% ⚠
FY 2024 actual ✓23.5%13.3%~30% ⚠
FY 2025 actual ✓23.2%12.4%~32% ⚠
FY 2026E base ⚠27.0%15.0%34%
FY 2027E base ⚠32.0%22.0%38%
FY 2028E mgmt model ✓39-40%31.7%~45%
FY 2029E base ⚠40.0%32.0%46%
FY 2030E base ⚠40.5%32.0%46%

Drivers per management commentary:

  • Mix shift to RF Infrastructure — higher-ASP SiPh / SiGe replacing legacy RF Mobile / Discretes
  • Tower-ST Agrate utilization fill-out — startup-cost drag of FY 2024-2025 reverses as utilization climbs
  • Discontinuation of legacy 150mm Fab 1 flows transferring to higher-margin 200mm Fab 2
  • Customer-prepayment-funded capacity reservation reduces under-utilization risk during ramp
  • Cost discipline + opex leverage — SG&A scaling sub-linearly with revenue

The +1,600-1,700 bps GM expansion FY 2025 → FY 2028 is the steepest trajectory in the specialty-foundry comp set. Compare GFS expanding +400-500 bps from FY 2024 to FY 2027 in the GFS DCF base case. The Tower trajectory is uniquely steep because of the concentration of the SiPh ramp — a single segment compounding at +30%+ CAGR with structurally higher GM acts as a margin-engine multiplier.

5. Capital intensity (Capex / Sales)

YearCapex ($M)Revenue ($M)Capex / Sales
FY 2023 ✓4711,42233.1%
FY 2024 ✓4321,44030.0%
FY 2025 ✓4371,56627.9%
FY 2026E base ⚠8001,80044.4%
FY 2027E base ⚠6002,15027.9%
FY 2028E base ⚠3502,60013.5%
FY 2029E base ⚠3203,01010.6%
FY 2030E base ⚠2903,4208.5%

Maintenance steady-state ~10% capex/sales by FY 2030 — consistent with peer specialty-foundry maintenance levels (GFS at 10.6% FY 2025; VIS / DBHiTek similar).

Government grant offsets to capex are MINIMAL for Tower:

  • No CHIPS Act award (Tower is an Israeli FPI; CHIPS funds US-incorporated entities only)
  • No NY State / state-level US incentive
  • Israeli Investment Center / Israel Innovation Authority grants — historical receipt; quantum modest (~$10-30M annually ⚠); flow through deferred-grant-income line over asset life
  • Tower-Intel Fab 11X — Tower benefits from Intel’s existing CHIPS-funded facility infrastructure but does not receive direct CHIPS funding for the equipment investment

Tower’s capex deployment is essentially fully self-funded (plus customer prepayments). This is structurally different from GFS where ~50-70% of net capex is offset by CHIPS / NY State / AMITC credits. The lack of subsidy uplift is a structural valuation drag vs GFS — Tower’s incremental ROIC on capex is genuinely the company’s own balance-sheet generation, not subsidy-offset returns.

Customer-prepayment offset (analyst working assumption ⚠):

  • Estimated $200-400M of customer prepayments through FY 2026-2028 against the $920M envelope
  • Treated as deferred-revenue liability on balance sheet; recognized into P&L as wafers delivered
  • Effectively reduces Tower’s net capital deployment from ~$660M remaining envelope to ~$260-460M
  • The dollar quantum of customer prepayments is the most important undisclosed data point for forward FCF modeling

6. Tax rate

Tower benefits from Israeli “Approved Enterprise” / “Beneficiary Enterprise” tax incentive status, which provides reduced corporate tax rates (typically 7-16% effective, vs Israeli statutory 23%) for qualifying capacity-extension investments.

YearEffective tax rate ⚠Notes
FY 2023~10%Including Intel-fee tax-treatment
FY 2024~12%Approved Enterprise + US dual-jurisdiction allocation
FY 2025~14% ⚠Triangulated from net income / pre-tax income
FY 2026E base ⚠14%Approved Enterprise sustained
FY 2027E base ⚠15%
FY 2028E base ⚠16%Some normalization as approved-asset depreciation matures
Terminal17%Long-term Israeli effective rate (sub-statutory)

Read. Tower’s effective tax rate is structurally low (10-17% range) for the forecast period vs Israeli statutory 23% — providing meaningful EPS accretion vs naive statutory-rate models. Watch for FY 2025 20-F effective-rate disclosure to refine; this is one of the most material-but-poorly-quantified DCF inputs.

7. Working capital and cash conversion

Specialty-foundry typical:

  • DSO: ~50-65 days
  • DIO: ~75-95 days
  • DPO: ~60-75 days
  • Cash conversion cycle: ~65-85 days
  • Working capital / sales: ~12-18% sustained

These metrics are typical for a 200mm specialty foundry. Tower’s customer-prepayment-funded capacity model could reduce effective working-capital intensity by 200-400 bps (~2-4% of sales) — i.e., customer cash arrives ahead of wafer deliveries, partially offsetting the receivables build during the ramp years. This is a modest favorable swing in DCF cash-conversion modeling.

8. Per-share fair-value range — illustrative scenarios

Illustrative only — full triangulation lives in bull case / bear case.

Methodology: FY 2028 EBITDA × exit multiple → FY 2028 EV → discount back to today at WACC → add net cash → divide by share count.

ScenarioFY 2028E revenueFY 2028E adj. EBITDAExit EV/EBITDAImplied FY 2028 EV ($B)Discount factor (3yr at WACC)Implied current EV ($B)Net cash ($B)Implied mkt cap ($B)Per share ($)vs. $191.54 spot
Bear1,9000.559×4.950.71 (12% WACC)3.51+1.04.51$40−79%
Base2,6001.1014×15.40.73 (11% WACC)11.24+1.012.24$110−43%
Bull3,2501.5018×27.00.76 (9.5% WACC)20.52+1.021.52$192+0%
Stretch bull3,2501.5022×33.00.7625.08+1.026.08$233+22%

Probability-weighted (20% Bear / 50% Base / 25% Bull / 5% Stretch):

  • 0.20 × $40 + 0.50 × $110 + 0.25 × $192 + 0.05 × $233 = $122/share ⚠

vs spot $191.54: implied −36% downside under this methodology.

Read. Probability-weighted DCF is materially below spot under conservative assumptions. The current $191.54 share price requires the Bull case to be effectively the central scenario (60-70% probability weight) — which is more aggressive than typical analyst risk-weighting. If the FY 2028 model fails to materialize, the floor anchor returns to the Intel-deal-implied $50-70 range (per comps_valuation).

Alternative DCF construction: If the customer-prepayment-funded capacity reservation is treated as substantially de-risking the bull case (e.g., promoted to 50-60% probability), the probability-weighted fair value rises to $155-175 — closer to but still below spot. The bull case must be the dominant scenario to support the current price.

8.1 Bull-case explicit triggers

For Bull case to materialize, ALL of the following must come true:

  1. December 2026 SiPh capacity expansion completes on schedule with utilization >70% within 6 months
  2. >70% of capacity reserved through 2028 holds firm — no customer cancellations
  3. LWLG PH18 PDK customer tape-outs succeed in 2026 → first revenue 2027 H2 → material ($50M+) by 2028
  4. GFS Fotonix competitive pressure remains rational — pricing not collapsed by 2028
  5. No Israeli geopolitical disruption material enough to compress Tower’s WACC premium
  6. RF Mobile envelope-tracker ramp delivers $200M+ incremental revenue by FY 2028

8.2 Bear-case explicit triggers

For Bear case to materialize, any 2-3 of the following must come true:

  1. SiPh capacity ramp slips by 6+ months — December 2026 target missed
  2. Customer prepayments refund / reduce — hyperscaler demand softens; capacity-reservation cancellations
  3. GFS Fotonix wins material 1.6T volume share — merchant SiPh becomes price-competitive duopoly rather than two-source premium
  4. Israeli geopolitical disruption triggers a step-change in cost-of-equity (+200 bps WACC = ~−25% DCF impact)
  5. Specialty-analog cyclicality returns — RF Mobile / industrial / automotive analog demand contracts in 2026-2027
  6. Capex envelope expansion required — initial $920M proves insufficient; second tranche announced; FY 2026 capex breaches $1B

9. Sensitivity table — single-variable ±10%

SensitivityBase case fair value+10% favorable−10% adverse
FY 2028 revenue ±10%$110$135$85
FY 2028 GM ±300 bps$110$145$80
WACC ±100 bps$110$130$90
Terminal growth ±100 bps$110$125$95
Israel risk premium ±100 bps$110$130$90

Most sensitive variable: FY 2028 GM (±300 bps swing produces ~$30 of share-price sensitivity in either direction). GM expansion is the single most important DCF driver — more sensitive than revenue growth.

10. Open items / backfill queue

  1. Sell-side DCF reconstruction — extract analyst notes (Citi, Needham, Stifel, JPMorgan, Susquehanna, Wells Fargo) and reconcile WACC + terminal growth + GM trajectory across the desk. Aggregator-level mean PT is currently $190-220 ◐.
  2. Beta primary-source verification — STOCK_PRICE_DATA.json reports 1.4; reconcile vs Bloomberg / Yahoo / Refinitiv.
  3. Israeli effective tax rate — FY 2025 20-F Note disclosure required for precise calibration.
  4. Customer-prepayment dollar quantum — referenced qualitatively at Q4 2025 release; specific dollar figure undisclosed. FY 2025 20-F deferred-revenue note expected to disclose.
  5. Per-segment gross margin assumptions — currently inferred at corporate level; sub-segment GM extraction from Industry Analyst Day materials or sell-side disaggregations.
  6. Capex by site / by program — not disclosed publicly; private channel checks would refine.
  7. LWLG-Tower revenue contribution timeline — joint development agreement signed 2026-03-11; no revenue guidance issued; analyst working assumption is $0 in 2026, modest ($10-50M) in 2027, material ($50-150M) in 2028+ if PH18 + EO-polymer commercializes.

Sources

  • FY 2024 20-F (acc. 0001178913-25-001537, filed 2025-04-30) — SEC EDGAR ✓. Historical financials baseline.
  • Q4 2025 6-K release (filed 2026-02-11) — Tower release ✓ and GlobeNewswire ✓. FY 2025 actuals + FY 2028 model.
  • Q4 2025 earnings call transcript — Investing.com ✓. FY 2028 financial model verbal disclosure.
  • TipRanks Q4 2025 capacity targets — TipRanks ✓. December 2026 capacity > 5× target; 70%+ reserved through 2028.
  • STOCK_PRICE_DATA.json — spot $191.54, beta 1.4, mkt cap $21.4B.
  • Damodaran ERP framework — sector-applied US ERP; standard practitioner reference.
  • CBOE-derived options data (STOCK_OPTIONS_DATA.json) — option-implied vol input to DCF cross-check.

Cross-references