NC-METH-001 v1.1 — Part D: Cross-Cutting Modules¶

Yield Modeling · CAPEX Build · OPEX Build · Tariff/PPA · Carbon · BESS Economics · Financing · Tax · Y10 Exit Reanchor¶

12 May 2026 · Internal · Author: AI-assisted, M. Forni review

D.0 Overview¶

Part D documents the cross-cutting modules that convert Part B pipeline outputs (per-segment kWp, CAPEX, yield, offtaker, confidence, tier) and Part C per-typology specifications into the financial-model inputs that drive returns analysis. Where Parts B and C describe what the segments and typologies are, Part D describes how they're combined into a coherent financial model.

The nine cross-cutting modules in Part D:

Module	What it does	Primary feed
D.1 Yield modeling	Per-segment P50/P90 yield with typology and degradation adjustments	Revenue
D.2 CAPEX build	EPC unit costs + soft cost markups → total project cost	Sources/uses; debt sizing
D.3 OPEX build	Per-typology cleaning/monitoring/insurance/IEAT JV → annual OPEX	EBITDA margin
D.4 Tariff/PPA	MEA TOU, BTM contractual discount, flat vs TOU, contract tenor	Revenue rate
D.5 Carbon	GS PoA $15/t base, T-VER fall-back, grid EF, LC no-carbon	Revenue (BP and others; LC zero)
D.6 BESS economics	Revenue stack (curtailment + demand-charge + time-shift), dispatch, mid-life	Revenue (where applicable)
D.7 Financing	Debt sizing, cost, tenor, DSCR/LLCR covenants, equity sources	Capital structure; sensitivities
D.8 Tax	CIT 20%, BOI 8yr standard, depreciation 70/30 split	Cash flow after tax
D.9 Y10 exit reanchor	13.5× EBITDA terminal multiple; partial exit modeling	Returns uplift case

These modules are not strictly sequential — they interact. Yield (D.1) feeds revenue, which feeds DSCR (D.7), which constrains debt sizing, which feeds total project cost (D.2). Carbon (D.5) is mostly LC-zero but materially matters at BP. BESS (D.6) cuts across yield, tariff, and load-side economics.

The modules are written with explicit canonical positions (per memory, LC v1.0 model, methodology RES-001 / FIN-003 series), gaps flagged where conventions disagree (AUDIT-001 yield, AUDIT-014 debt cost, AUDIT-016 OPEX), and downstream impact on returns noted where the gap is material.

D.1 Yield modeling¶

D.1.1 Yield convention reconciliation (AUDIT-001)¶

The single largest open methodology gap in v1.1 is the yield convention disagreement between the LC financial model and the methodology RES-001 number for LC:

Source	LC P50 yield
LC model (NC-FM-LC-001 v1.0)	1,380 kWh/kWp/yr
Methodology RES-001 canonical	1,485 kWh/kWp/yr
Gap	7.6% revenue gap
IRR impact	±220 bps depending on resolution

The two numbers must use the same convention before being directly comparable. Conventions that affect the comparison:

AC delivery vs DC generation — DC nameplate yield is higher than AC-delivered yield (after inverter losses ~2–3%)
Pre-soiling vs post-soiling — gross yield is higher than soiling-allowance-applied yield (typically 2–4%)
Pre-degradation vs Y1-degradation — Y0 (commissioning) yield is higher than Y1 (first full operational year) yield (1.0%/yr TOPCon)
Annual average vs first-year nominal — methodology canonicals are often Y1; multi-year averages are lower due to degradation

WP1 (Solcast / PVGIS / Solargis cross-check + bankable PV yield study) is scoped to resolve which convention applies to each number. Until reconciliation lands, the LC model's 1,380 is the canonical for LC financial modeling, and the 1,485 is the canonical for methodology-level cross-estate ranking — the divergence is flagged.

[INCONSISTENT] — AUDIT-001 is the gating WP1 closure for v1.1.0 lock.

D.1.2 P90 derivation¶

P90 = P50 × 0.90 (industry-standard lender convention).

The 0.90 factor reflects ~1-sigma worse-than-expected weather across the 25-year operational lifetime. Some lender conventions use 0.92 or 0.85 — methodology uses 0.90 as the canonical.

P90 is the basis for DSCR-constraining lender base case sizing (per D.7.4).

D.1.3 Per-typology yield adjustments¶

Per Part C, each typology applies adjustments to the estate-canonical P50:

Typology	Adjustment	Cumulative Y1 effective P50 at LC
T1 rooftop	−3% to −5% (low-tilt tropical) + 0% bifacial	1,330–1,410 kWh/kWp/yr
T2 ground-mount	0% (baseline) + 5–7% bifacial if light ground	1,380–1,475 kWh/kWp/yr
T4A carport	−1% to −2% (light pole shading) + 4–6% bifacial (asphalt albedo)	1,400–1,510 kWh/kWp/yr
T4B-DC arterial	+1% to +3% (dual N/S sky capture) + 4–6% bifacial (road albedo)	1,420–1,540 kWh/kWp/yr
T6W canal canopy	−1% to −2% (humidity, biofouling) + 1–2% bifacial (water albedo)	1,370–1,470 kWh/kWp/yr

[INCONSISTENT — AUDIT-041] The LC v1.0 model applies estate-canonical 1,380 uniformly across all typologies; the per-typology adjustments above are methodology-specified but not applied. WP4 closure brings the model into alignment with the methodology per-typology framework.

D.1.4 Year-by-year degradation¶

Y0: construction (no generation)
Y1: nameplate × (1 − degradation)^0.5 ≈ 99.5% (mid-Y1 average); methodology simplification: Y1 = nameplate
Yk for k ≥ 2: nameplate × (1 − 0.010)^(k−1)
Y25: nameplate × 0.99^24 = 78.5% (Y25 cumulative)
Y25 average annual: integrates degradation curve to ~88% of nameplate

Module-level degradation 1.0%/yr is TOPCon standard. Earlier mono-PERC modules used 0.7%/yr; if module choice changes, this rate revises.

D.1.5 Soiling and cleaning frequency¶

Soiling allowance is baked into the estate-canonical P50 (typically 2–3% gross yield deduction). Cleaning frequency by typology:

Typology	Cleaning frequency	Driver
T1 rooftop	Quarterly	Standard rainfall-driven
T2 ground	Quarterly	Dust from surrounding ground
T4A carport	Quarterly	Mix of asphalt and vehicle dust
T4B-DC arterial	Quarterly	Road traffic dust slightly elevated
T6W canal canopy	Monthly (3× quarterly frequency)	Bird droppings, water spray, biofouling

The monthly T6W cleaning is the methodology's principal OPEX differentiator (see D.3). It is reflected in OPEX but not in P50 yield assumptions (the higher cleaning frequency keeps T6W yield at parity with other typologies despite the biofouling exposure).

D.1.6 Yield data source hierarchy¶

For new estates or yield revalidation:

Solcast Time Series API + Bankable PV Yield Study (commissioned; $5–7K per estate; WP1 procurement). Lender-acceptable.
Solargis (alternative commissioned study; $4–8K). Lender-acceptable as cross-check.
PVGIS (free, EU JRC). Adequate for desktop screening; not lender-grade.
Existing on-estate solar systems (where present): ground-truth validation against commissioned model.
TGO grid emission factor (locked at 0.4091 tCO2/MWh CM per memory edit 16): orthogonal to yield but used downstream in carbon.

D.1.7 Yield gaps¶

Three unresolved yield gaps in v1.1:

Convention reconciliation 1,380 vs 1,485 (AUDIT-001) — WP1 gating
Per-typology adjustments not applied in model (AUDIT-041) — WP4 closure
Per-segment yield modeling (AUDIT-017) — Solcast + Digital Surface Model via WP1; methodology currently estate-canonical

D.2 CAPEX build¶

D.2.1 EPC unit cost layer¶

Per Part C, EPC unit cost is per-typology with LC v1.0 register canonical and methodology canonical pending Annex H. Summing across typologies gives EPC for the active envelope. For LC: $25.04M EPC across 47 segments.

Premium adjustments (per Part C C.X.4 sections) apply on top of base unit cost: C4 corrosion +10–15% (coastal); drilled-pier geotechnical fail +10–25%; phased deployment +25% (T4B-DC); marine-grade steel +10–15% (T6W coastal); etc.

D.2.2 Soft-cost markup (development)¶

8% of EPC standard
Covers: project development, engineering studies, ESHS / EIA addendum, IEAT JV setup costs, permitting and approval, contingency for site-specific adjustments during development
LC: $25.04M × 8% = $2.00M

D.2.3 Financing fees¶

2% of debt amount standard (not 2% of EPC)
Covers: lender arrangement fee, legal documentation, independent engineer (IE), security trustee, debt service reserve account (DSRA) setup
LC at 52% LTV ($13.02M debt): $13.02M × 2% = $0.26M
Note: financing fees are calculated against the final committed debt amount, not against EPC

D.2.4 Contingency¶

5% of EPC standard
Drawable during construction for unanticipated cost overruns
Methodology treats contingency as part of total project cost (TPC), not as separate buffer
LC: $25.04M × 5% = $1.25M

D.2.5 Interest During Construction (IDC)¶

IDC accrues during the 12-month construction period
Calculated from debt drawdown schedule (typically 25% Q1 / 25% Q2 / 30% Q3 / 20% Q4 per LC model)
At 6.0% debt cost / 12-month average drawdown: IDC ≈ debt amount × 6.0% × 0.5 = 3% of debt
LC: $13.02M × 3% = $0.39M (approximation; LC model has precise period-by-period calculation)

D.2.6 Total project cost (TPC) build¶

For LC at 52% LTV (lender-sized canonical case):

Component	Amount	Basis
EPC	$25.04M	Per Part C
Soft costs (8% of EPC)	$2.00M	D.2.2
Contingency (5% of EPC)	$1.25M	D.2.4
Financing fees (2% of debt)	$0.26M	D.2.3
IDC	$0.39M	D.2.5 (approximation)
Other (site prep, IEAT JV admin)	$0.22M	LC model specifics
Total project cost (TPC)	$29.16M	LC IC paper §3

The TPC-to-EPC ratio for LC is 1.165 (i.e., EPC × 1.165 ≈ TPC). This ratio is consistent across IEAT estates with similar financing structures and is the working approximation for desktop estimation.

D.2.7 Equipment vs civil split (depreciation feed)¶

70% equipment / 30% civil works of EPC
Equipment: modules, inverters, BOS, monitoring, transformers, MV switchgear, structural steel for canopies, BESS containers
Civil: foundations, drainage, fencing, access roads, security infrastructure, perimeter walls
Depreciation: equipment 20% straight-line over 5 years; civil 5% straight-line over 20 years (see D.8)

D.2.8 CAPEX gaps¶

Per-typology methodology canonical CAPEX (AUDIT-040) — Annex H via WP2 + WP4 engineering review; gating for v1.1.0 lock
C4 corrosion premium application (AUDIT-002) — verify presence in unit costs for coastal estates (LC, MTP, MTP Port, Songkhla)
Bottom-up cost stack (AUDIT-039 retired) — deferred to v1.2 pending engineering review; not a v1.1 gap

D.3 OPEX build¶

D.3.1 Per-typology O&M frequency¶

Cleaning frequency drives the cost differential (per D.1.5):

Typology	Cleaning frequency	Cost factor
T1, T2, T4A, T4B-DC	Quarterly (4× /yr)	1.0× base
T6W canal canopy	Monthly (12× /yr)	~2.0× base for cleaning layer

D.3.2 OPEX cost layers¶

The methodology decomposes OPEX into seven layers:

Layer	Driver	LC weighted estimate (USD/kWp/yr)
1. Module cleaning	Cleaning frequency × labor + materials	$7.50 (quarterly) / $15.00 (monthly T6W)
2. Inverter & BOS monitoring	Remote monitoring + on-site response	$4.00
3. Vegetation management	Ground/buffer trimming, herbicide	$2.50
4. Security & site maintenance	Fencing, lighting, perimeter response	$3.50
5. Insurance	Property + business interruption + liability	$5.00
6. IEAT JV management fee	24.5% IEAT share admin/coordination	$4.50
7. Asset management fee	Sponsor administrative + reporting	$4.00
Subtotal (quarterly cleaning typologies)		$31.00
Subtotal (monthly T6W typology)		$38.50
Weighted average across LC envelope	(T6W = 43% of MWp, others = 57%)	~$33–34/kWp/yr

[INCONSISTENT — AUDIT-016] The LC v1.0 model applies a flat $20/kWp/yr across all typologies; this is below the methodology weighted average. The gap is **~$13/kWp/yr × 34.59 MWp = $450K/yr × 25 years**. Discounted at 10% real and net of escalation: NPV impact ≈ $3.5M. Material to IRR by ~80 bps.

D.3.3 OPEX escalation¶

2.5%/yr nominal canonical (flat across all cost layers)
[INCONSISTENT — pending WP2] Per-cost-layer refinement: modules deflation (−2%/yr); labor CPI+ (+3%/yr); services CPI (+2%/yr); insurance CPI+ (+3.5%/yr); IEAT management fee escalation (TBD per JV terms)
Per-layer escalation would marginally lower aggregate OPEX growth over 25 years vs flat 2.5%

D.3.4 OPEX in $/kWp/yr canonical¶

LC IC paper §4 cites $38.57/kWp/yr loaded — this is the methodology weighted target. LC v1.0 model uses $20/kWp/yr — the AUDIT-016 gap.

The methodology canonical for new-estate analysis: weighted average per typology per D.3.2, escalated 2.5%/yr.

D.3.5 BESS O&M¶

BESS O&M is separate from solar: - Per-MWh basis: ~$5/MWh throughput (annual cycling-driven) - Per-MW basis: $8K/MW/yr fixed (HVAC, monitoring, fire suppression maintenance) - Mid-life replacement (Y12): $80–120/kWh CAPEX (separate from O&M)

LC's no-BESS design means BESS O&M is zero for LC. BP's typical 5–6 MWh deployment yields BESS O&M of ~$30–50K/yr.

D.3.6 OPEX gaps¶

AUDIT-016 closure — model flat $20/kWp/yr vs methodology weighted ~$33/kWp/yr; WP4 closure brings model into methodology alignment
Per-cost-layer escalation refinement — pending WP2
IEAT JV management fee structure — methodology assumes flat 24.5% admin fee; actual JV terms may refine

D.4 Tariff and PPA structure¶

D.4.1 MEA TOU reference rates¶

The Metropolitan Electricity Authority (MEA) Time-of-Use (TOU) rate structure for medium-voltage industrial customers is the reference benchmark:

Peak: THB 4.28/kWh (09:00–22:00 Mon–Fri, business hours)
Off-peak: THB 2.34/kWh (22:00–09:00 weekday + all weekend)

These are 2026 rates per ERC (Energy Regulatory Commission) approved tariff schedule. Subject to annual revision; methodology should refresh annually.

D.4.2 BTM contractual discount factor¶

For BTM (behind-the-meter) PPA structures, the methodology canonical is to apply a 0.90 discount factor against the MEA TOU peak rate to derive the contractual tariff. This delivers a guaranteed-cheaper-than-grid economics that makes the BTM offering compelling to the offtaker.

LC contractual tariff: 4.28 × 0.90 = THB 3.85/kWh flat (user-locked per memory edit 16)
Note: the 0.90 factor is the BTM contractual discount, NOT a TOU-to-flat conversion. LC's tariff is contractually flat (not TOU-structured) at the discounted level.

D.4.3 Flat vs TOU tariff structures¶

The methodology recognises three tariff structures:

Structure	Description	Where applied
Flat	Single rate regardless of time	LC (THB 3.85); BMA (THB 4.20 user-locked); methodology default for new BTM deals
TOU-aligned	Peak + off-peak with PV-weighted average	Sometimes negotiated for tenant-direct ESAs
Hybrid	Flat + peak premium	Rarely used; only where BESS time-shifts

Most IEAT BTM deployments use flat tariffs. TOU-aligned and hybrid are reserved for special cases.

D.4.4 PPA escalation¶

0%/yr nominal standard (flat throughout 25-year tenor)
Methodology choice: flat tariff de-risks revenue assumption (no escalation-dependent revenue assumption) and aligns with the BTM customer's preference for predictable bills
Alternative: CPI-linked escalation (typically 1.5–2.5%/yr) — used in some IEAT-direct ESAs where IEAT prefers inflation hedge
LC v1.0 uses 0%/yr flat throughout

D.4.5 Contract tenor¶

25 years solar: from Y1 first full revenue year (per A.4.6 convention)
20 years BESS: with mid-life Y12 replacement (modules of cells, not container/HVAC)
IEAT-direct contracts align with the 25-year MSA concession
Tenant-direct contracts may be shorter (e.g., 15–20 years) to align with tenant lease terms

D.4.6 ESA structures¶

The methodology supports four ESA structures:

IEAT-direct: IEAT is offtaker; pays tariff per kWh delivered; canonical for IEAT-owned admin/factory + utility-grade ground-mount/carport
Tenant-direct: Tenant is offtaker; per-tenant credit considered (per Annex G when complete)
Power-producer ESA sub-class: Offtaker is a power-producer (e.g., LC-T6W-03 LC2 Power Plant) selling to grid; methodology should formalise this sub-class (currently treated as edge case per Part C C.5.6)
Mixed segment: Some segments IEAT-direct, others tenant-attributed (typical for T4B-DC dual-side arterial; per Part C C.4.6)

D.5 Carbon economics¶

D.5.1 ICC Aug 2025 — solar excluded from ITMO/CORSIA¶

Per the Thai International Carbon Credit (ICC) Guideline of August 2025, solar PV is excluded from international carbon transfer mechanisms — both ITMOs (Article 6.2 of Paris Agreement) and CORSIA (Carbon Offsetting and Reduction Scheme for International Aviation).

The exclusion means: - ITMO/CORSIA pricing ($30–50/t historically projected for solar) is NOT achievable for Thai solar - T-VER domestic-only mechanism remains available (~$5/t) - Gold Standard Programme of Activities (PoA) is the primary international-pricing channel at ~$15/t

This locks the methodology's carbon base at GS PoA $15/t, not the more aspirational ITMO/CORSIA tier.

D.5.2 Gold Standard PoA $15/t base¶

Gold Standard PoA: international voluntary carbon standard, deeper market depth than CDM
Pricing: $15/t methodology base (canonical per memory edit 16)
Volume: 0.4091 tCO2/MWh × estate generation × 75% (haircut for monitoring uncertainty + buyer discount)
Tenor: 10-year crediting period (renewable to total 30 years)

D.5.3 T-VER fall-back¶

T-VER (Thailand Voluntary Emission Reduction): Thai domestic-only mechanism, lower pricing
Pricing: ~$5/t methodology base
Used as fall-back if GS PoA registration fails or if buyer market doesn't materialise
Carbon revenue at T-VER is one-third of GS PoA; this drives the methodology's preference for GS PoA where achievable

D.5.4 LC no-carbon by design¶

Per LC IC paper §1, LC has no carbon revenue in the v1.0 financial model. Rationale:

BTM-only deployment; no grid-export means standard CDM-equivalent additionality argument is weak
Conservative IC posture: methodology preserves carbon revenue as upside, not base case
GS PoA registration timeline (~18–24 months from FID) means Y1–Y2 may have zero carbon revenue regardless

If LC carbon revenue were applied (illustrative): 34.59 MWp × 1,380 kWh/kWp/yr × 0.4091 tCO2/MWh × 75% × $15/t = $0.22M/yr — ~3% of revenue. Material but not transformative.

[INFERRED] The decision to forgo carbon at LC is a base-case posture; methodology should formalise the carbon-as-upside vs carbon-as-base decision per estate.

D.5.5 Grid emission factor 0.4091 tCO2/MWh (TGO Combined Margin)¶

Locked per memory edit 16 (replaces older 0.475 / Scope-2 only). The Combined Margin methodology (50% Build Margin + 50% Operating Margin) is the TGO-endorsed standard for grid-displacement carbon accounting in Thailand.

D.5.6 Carbon revenue attribution¶

Carbon revenue accrues to whoever owns the avoided-emissions claim: - BTM self-consumption → offtaker (IEAT-direct or tenant) by default - Grid-export → grid operator; typically not claimable by PV developer - ESCO structure → can be retained by PV developer if contract specifies

D.5.7 KTB ERPA structures (cross-reference)¶

The methodology supports two ERPA (Emission Reduction Purchase Agreement) structures developed for the BP/LCC carbon channels:

Broker Model: KTB as disclosed agent for PSO (Public Sector Organisation) buyers; carbon credits transferred at PSO market price; KTB earns commission
Underwriting Model: KTB as principal buyer; Advance Payment with No Clawback structure; PV developer earns fixed $/t regardless of subsequent KTB resale price

LC, with no carbon revenue, doesn't engage either model. BP and other BESS-equipped estates can use either depending on KTB's appetite.

[Cross-reference: NC-MN-001-R3 work plan WP5 includes Baker McKenzie review of ERPA structures]

D.6 BESS economics¶

D.6.1 Revenue stack composition¶

BESS revenue derives from three sources (Part C C.6.1):

Source	Description	Typical % of BESS revenue
Curtailment recovery	Store excess PV that would otherwise be curtailed	30–50%
Demand charge management	Shave evening demand peaks (16:00–19:00 Thai industrial)	30–50%
Time-shift	Discharge during higher-tariff TOU peak hours	10–20%

Estate-specific composition: BP demand-charge-heavy (load profile drives peak), Songkhla curtailment-heavy (export constraints), etc.

D.6.2 Curtailment recovery¶

Captures top 10–20% of PV generation peaks that would otherwise breach interconnection limits
BESS round-trip efficiency 88–92% means ~90% of stored energy is recovered; remainder is loss
Revenue: recovered energy × tariff (typically the contractual flat rate)
LC: zero (BTM, no curtailment design)

D.6.3 Demand charge management¶

Thai industrial demand charges: typically THB 175–220/kW-month for medium-voltage
Peak demand window: 09:00–22:00 weekday TOU peak; the "evening shave" target is the 17:00–20:00 sub-peak
BESS discharge shaves peak demand × 2 hours (typical sizing)
Revenue: avoided demand charge × 12 months
LC: zero (no peak-load profile justifying BESS)

D.6.4 Time-shift¶

Discharge during TOU peak hours when import tariff > stored generation cost
LC TOU spread limited (peak 4.28 vs off-peak 2.34, factor ~1.83): time-shift economics marginal
Time-shift is the smallest revenue contributor in most cases

D.6.5 Cycle dispatch profile¶

Typical industrial BESS: 250–350 full-equivalent cycles per year
Curtailment-dominant: cycle count higher (closer to 350) but cycle depth lower
Demand-charge-dominant: cycle count lower (closer to 250) but cycle depth higher (deeper SoC swings)
Annual throughput = container MWh × cycles × DOD × round-trip efficiency

For a 5 MWh container at 280 cycles × 80% DOD × 90% RTE: ~1,000 MWh/yr throughput.

D.6.6 Mid-life replacement Y12¶

$80–120/kWh for cell replacement at Y12 (per BNEF cost-curve trajectory: $40–80/kWh by 2035 is the projected forward; methodology uses 2026-published Y12 estimate)
Replaces battery cells; retains container, HVAC, inverters, BMS
Capital event in the 25-year operational period — accounted for in equity cash flow

D.6.7 LC no-BESS by design¶

Per Part C C.6.7, LC has no BESS. Rationale: - BTM-only deployment with self-consumption load profile that matches PV generation (industrial daytime peak) - No curtailment risk (export limits not binding) - Modest demand-charge profile (24/7 industrial; not evening-peaky) - BESS economics negative on standalone basis at LC

BP and other EEC estates have different load profiles and curtailment exposures that justify BESS economics.

D.7 Financing structure¶

D.7.1 Debt sizing¶

The methodology uses % LTV measured against EPC (not total project cost):

70% LTV canonical program structure — methodology baseline
60% LTV sponsor base case — LC IC paper canonical case (12.8% sponsor IRR)
52% LTV lender-sized execution case — where DSCR constrains; LC produces 12.2% sponsor IRR at 52% LTV
70% LTV breaches DSCR at LC — produces 15.1% IRR but fails 1.05× covenant; not financeable

The lender-sizing process: solve for max debt such that DSCR P90 covenant (1.05× default / 1.15× lockup) is met across all 12 debt-service years.

D.7.2 Debt cost¶

[INCONSISTENT — AUDIT-014] Methodology FIN-003 series specifies EXIM-equivalent 5.75%/yr; LC model uses 6.0%/yr. The 25 bps gap reflects either (a) prudent lender-pricing buffer in the model, or (b) gap from methodology canonical that should be closed.

WP4 closure: align model and methodology at 5.75% if EXIM commits at that rate; otherwise revise methodology
Impact: ±15 bps IRR; ~$0.1M NPV

D.7.3 Debt tenor¶

12 years from COD with 1-year grace (interest-only during year 1; principal repayment Y2–Y12)
Tenor matches typical Thai project-finance senior debt; EXIM and KTB both within this range
Methodology assumes tenor < contract tenor (12 vs 25) — debt is fully repaid by Y12 with 13 years of residual equity-only operations

D.7.4 DSCR profile and covenants¶

DSCR (Debt Service Coverage Ratio) = CFADS / Debt Service
CFADS (Cash Flow Available for Debt Service) = EBITDA − tax − maintenance capex
P50 DSCR target: > 1.30× (model average for LC: 1.42×)
P90 DSCR: > 1.15× lockup; > 1.05× default
LLCR (Loan Life Coverage Ratio) at COD: > 1.35× threshold (LC IC paper: 1.48×)

DSCR profile across the 12-year debt service period:

Year	LC P50 DSCR (illustrative)
Y1 (grace)	n/a (interest-only)
Y2–Y5 (BOI holiday)	~1.50× (no tax)
Y6 (BOI ends; degradation increasing)	~1.35×
Y10 (mid-degradation)	~1.30×
Y12 (final year)	~1.20×

The Y12 trough is the binding constraint; lender-sized debt is back-solved to keep Y12 P90 > 1.05×.

D.7.5 Equity sources¶

LC envelope: $25.04M EPC × 1.165 = $29.16M TPC; 52% LTV against EPC = $13.02M debt; $16.14M equity.

Source	Share	$M	Notes
IEAT (JV partner)	24.5%	$3.95M	Per 25-year MSA JV structure
Sponsor (NewCo + Fund I)	75.5%	$12.19M	Of which Fund I LP commits the bulk
Total equity	100%	$16.14M

[INFERRED] The 75.5/24.5 split is per the methodology's standard IEAT JV structure; per-estate variations may exist but are not canonical for v1.1.

D.7.6 Equity payback¶

LC payback: ~7.5 years (sponsor equity recovered by Y8 from operations)
Payback varies inversely with leverage and directly with IRR; the methodology tracks but doesn't constrain on payback

D.8 Tax¶

D.8.1 CIT (Corporate Income Tax)¶

Thailand standard CIT: 20% of taxable profit
Applies to JV SPV after BOI holiday expires

D.8.2 BOI (Board of Investment) tax holiday¶

Activity 5.2.1 Solar Power Generation: 8-year CIT exemption standard
EEC enhancement: up to 13-year CIT exemption under Phase 0 evaluation (legal/tax opinion pending)
BOI period starts in Y1 (first revenue year, not COD year — per typical Thai BOI convention)
LC base case: 8-year BOI (Y1–Y8); EEC 13-year enhancement is upside not in base

D.8.3 Depreciation¶

Asset class	Share of EPC	Depreciation rate	Period
Equipment (modules, inverters, BOS, structural steel)	70%	20% straight-line	5 years
Civil works (foundations, drainage, fencing, roads)	30%	5% straight-line	20 years

Tax shield from depreciation: ~$2.5M/yr equipment shield × 20% CIT = $0.5M/yr CIT savings during Y6–Y10 (post-BOI). 5-year equipment shield ends Y6 first batch; longer civil shield runs Y1–Y20.

D.8.4 VAT¶

Thai VAT: 7% on import of equipment + 7% on services
Methodology assumes VAT is recoverable for the SPV (input VAT credits)
Net VAT cost to project: ~0% over full lifecycle (timing impact only)

D.9 Y10 exit reanchor¶

D.9.1 Terminal multiple convention¶

13.5× EBITDA terminal multiple at Y10 (per BP v6.5 methodology baseline)
Source: comparable transaction analysis of operating Thai solar SPVs sold to infrastructure investors / yieldco entities; 13.5× reflects mid-cycle market multiples
Methodology updates this multiple annually based on Thai infrastructure transaction data

D.9.2 LC Y10 exit case¶

Per LC IC paper §3: - Y10 EBITDA (LC): ~$3.30M (estimated) - Y10 enterprise value at 13.5× = $44.55M - Less Y10 outstanding debt: $4.36M (Y10 debt balance after annuity) - Y10 equity value: $40.19M - Equity invested (recovered + remaining): $12.19M / 3.50× MOIC over Y0–Y10 - Y10 IRR: 14.3% (vs 12.2% full-tenor)

D.9.3 Reanchor mechanics¶

The Y10 exit "reanchor" is not necessarily a full sale; methodology supports three Y10 outcomes:

Full exit: 100% equity sold at 13.5× EBITDA to infrastructure buyer; methodology uplift case
Partial exit (refinancing): 60% equity sold to refinancing partner; remaining 40% continues to contract end Y25
Continue to maturity: no exit; full 25-year operations; IRR reflects full-tenor 12.2% rather than Y10 reanchor 14.3%

Per the methodology, the Y10 exit case is presented as upside scenario in IC papers (not base case). Base case assumes continue-to-maturity unless a specific transaction opportunity is identified.

D.9.4 Exit gaps¶

Multiple update — 13.5× is per BP v6.5 (2025 vintage); refresh against 2026 Thai infrastructure transaction comps
Per-estate exit scenarios — methodology applies one multiple across all estates; some estates (high-quality long-tenor) may command premium; others (concentration-risk) may discount
Tax treatment of exit — depends on SPV structure and capital gains taxation; methodology assumes infrastructure-fund-typical (low effective rate)

D.10 Acknowledged Part D gaps¶

Tracked in audit register R3 v0.3:

AUDIT-001 yield convention reconciliation (D.1.1) — WP1 gating; ±220 bps IRR impact
AUDIT-014 debt cost (D.7.2) — 5.75% vs 6.0% gap; ±15 bps IRR
AUDIT-016 OPEX gap (D.3.4) — model flat $20/kWp/yr vs methodology weighted ~$33; ~80 bps IRR
AUDIT-041 per-typology yield adjustments (D.1.3) — not applied in model; per-typology bifacial uplift differentiation
AUDIT-002 C4 corrosion premium (D.2 + Part C) — verify in unit cost build-up; pending
AUDIT-040 per-typology methodology canonical CAPEX (D.2) — Annex H gating event for v1.1.0 lock
OPEX escalation per-cost-layer (D.3.3) — refinement pending WP2
Carbon as base vs upside (D.5.4) — methodology should formalise the per-estate base-case carbon decision
Y10 terminal multiple refresh (D.9.4) — annual update against Thai infrastructure transaction comps
Power-producer ESA sub-class formalisation (D.4.6) — LC-T6W-03 LC2 Power Plant special case

Items 1, 2, 3, 4, 6 are the high-materiality gaps (>50 bps IRR or financial-model impact). The rest are completeness gaps.

D.11 References¶

NC-IC-LC-001 IC Paper v1.1 — financial model integration and returns analysis
NC-FM-LC-001 Laem Chabang Financial Model v1.0 — Inputs, Construction, Operations, Debt_Service, Returns tabs
NC-METH-001 v1.1.0 Part A — framing conventions (A.4)
NC-METH-001 v1.1.0 Part B — pipeline outputs (B.4 yield, B.5 CAPEX, B.6 offtaker)
NC-METH-001 v1.1.0-revA Part C — per-typology specifications (C.X.4 CAPEX, C.X.5 yield)
NC-MN-001-R3_audit_v0_3 — live audit register
Memory edits 1–30 (per memory_user_edits view 12 May 2026); particularly:
Edit 16 (FX 35; BESS $175/kWh; grid EF 0.4091; LC tariff 3.85 = 0.90 × MEA TOU peak 4.28)
Edit 13 (FX boundary main vs IET-only)
Edit 30 (American English language locked)
TGO (Thai Greenhouse Gas Management Organization) — grid emission factor reference
ERC (Energy Regulatory Commission) — MEA TOU tariff schedule
ICC Thailand August 2025 Guideline — international carbon transfer mechanism scope
BNEF — LFP cost curve and PV module pricing benchmarks

End of Part D v1.1.0.