Latvia SW Power BESS financing — merchant margin and loan screening#
News: Byhmgard AB, 9 November 2026, original regulated-company announcement and issuer summary. The static source-fact JSON records source links, date, reported units, approximation qualifiers, and assumptions separately. It is not a live API connector or a power-price dataset. Checked on 10 November 2026. Numerical factual extracts only; the original article is not redistributed.
Business question: What merchant trading margin, utilization and battery efficiency/degradation must be established before relying on a company’s multi-year revenue forecast in a project-financing review? How would debt repayment compare with assumed annual operating cash? This is a first-pass due-diligence screen, not a lender model or investment recommendation.
Evidence boundaries#
Quantity |
Disclosed status |
Value |
|---|---|---|
Grid connection |
Project specification, not operating data |
14.99 MW |
Battery capacity |
Approximate planned rating |
~30 MWh |
Total investment |
Approximate financing requirement |
~€6.1m |
Signet Bank loan |
Project loan reported as just over €3m |
>€3.0m |
Arbitrage revenue 2027–2047 |
Issuer forecast, not earned revenue |
~€20m |
Additional balancing and capacity services |
Separate potential, not bank-base revenue |
~€33m |
O&M |
Reported annual contracted expense once operational |
~€70k/year |
Operation |
Forecast, not a commission certificate |
Q2 2027 |
The disclosed financing remainder is partly payment-plan equipment financing contributed through an affiliate; it is not established as debt-free equity. Monthly bank repayment is expected from autumn 2027, but the rate, tenor, drawdown schedule, reserve requirements, amortization profile and actual merchant revenue definition are not disclosed. The issuer’s estimates are forward-looking, and no independent price-series audit is possible from the announcement alone.
Reusable method and explicitly illustrative numerical assumptions#
energy_methods.bess_merchant_finance implements an intentionally transparent, 365-day/year algebraic baseline:
Delivered AC energy in operating year
y:365 * grid_charge_MWh_per_full_cycle * equivalent_cycles_per_day * round_trip_efficiency * availability * annual_capacity_retention**y. 30 MWh is treated as purchased grid energy per assumed full cycle solely for this screen; the announcement does not verify this measurement boundary. No intraday schedule, state-of-charge feasibility, spread chronology, degradation mechanism or revenue stacking is demonstrated.Implied constant net captured margin:
target_contribution_EUR / sum(annual_delivered_AC_MWh). A net captured margin excludes charging purchases and market fees. The issuer describes its €20m estimate as arbitrage revenue; its accounting boundary may differ. Therefore equating the disclosed forecast to this method’s contribution is conditional, not a validation.Monthly fully amortizing nominal debt service for assumed principal
D, monthly interestrandNmonths:D*r / (1 - (1+r)**(-N)), times 12 for an approximate full operating-year screen, with an exact zero-interest branch. It is not the unknown Signet Bank schedule and does not reproduce the 2027 partial year.Simplified annual cash cover:
(net_trading_contribution - annual_fixed_O&M) / hypothetical_annual_debt_service. No tax, inflation, merchant transaction detail, insurance, finance fees, maintenance capex, warranty replacement, debt service reserve, working capital, power-market constraints, or affiliate equipment obligations. This is not a verified contractual DSCR or EBITDA.
Scenario, not observed data: one equivalent full cycle/day, 88% round-trip efficiency, 95% availability, 98% retained energy capacity per annum, illustrative €3.05m senior debt, 7% annual interest, 10 years of level monthly payments. The scenario excludes potential balancing/capacity revenue and holds fixed O&M at the disclosed approximate €70k/year. Assumptions are not assigned confidence intervals. Rates and outcomes in plots are sensitivity tests, not probability distributions. The capital financing amount is used as context, not to infer equity returns.
Reproduction and review#
uv sync --locked --all-packages --group docs --python 3.13
uv run --no-sync pytest methods/tests/test_bess_merchant_finance.py tests/test_gallery_navigation.py
uv run --no-sync python examples/analysis/plot_latvia_bess_finance.py
uv run --no-sync ruff check methods/src/energy_methods/bess_merchant_finance.py methods/tests/test_bess_merchant_finance.py examples/analysis/plot_latvia_bess_finance.py
uv run --no-sync ty check
uv run --no-sync sphinx-build -W --keep-going -b html docs docs/_build/html
The gallery loads only the checked-in dated issuer-fact JSON and deterministic user assumptions; no network calls or synthetic market prices are made. Figure 1 compares the hypothetical contribution and fixed-cost cash cover against assumed debt service. Figure 2 varies the hypothetical loan interest rate. The necessary next step is independently retrieving time-stamped Nord Pool Latvia day-ahead and intraday buy/sell prices with a valid license, optimizing constrained charging/discharging with fees, reconciling the issuer’s revenue accounting and obtaining the signed lender term sheet. Until then, no case-specific bankability, payback period, or credit capacity is established.