Offshore wind bankability: Japanese industry-cost sensitivity (November 2026)#
News trigger and business question#
On 5 November 2026, Reuters reported Tokyo Gas’s withdrawal from the Kashima Bay offshore wind project, citing rising costs. The report disclosed no project-level CAPEX, OPEX, financing or generation; no financial conclusions about that project can be drawn from this model.
Question: how do construction-cost escalation, real financing costs and the net capacity factor affect the levelized cost and flat real electricity price needed to break even?
Evidence and provenance#
Japan METI, 111th Procurement Price Calculation Committee, January 8, 2026 lists the Japan Wind Power Association (JWPA) presentation on offshore-wind investment and cost reduction.
An independent summary of that presentation reports planning-model inputs of JPY 900,000/kW CAPEX, JPY 12,300/kW-year OPEX and 36.6% net capacity factor for fixed-bottom offshore wind.
The offline JSON input and provenance classify these figures as
published_industry_assumption, not measured site performance or invoices. The source PDF could not be independently retrieved in this run; numerical values were checked only against the secondary synthesis. Verify the original presentation before investment use.Full-source licensing is unverified; only three attributed numerical facts are transcribed, not original text, graphics or bulk data. All assumptions beyond those three quantities are expressly illustrative.
Input → method → analysis#
The versioned input loader energy_data_retrieval.cost_inputs.load_offshore_cost_snapshot validates evidence classification, values and source metadata. No live API, secret or data download is needed.
The reusable energy_methods.offshore_wind_economics.evaluate_offshore_wind computes
E_year_kWh = capacity_kW * 8760 * net_CF * (1 - annual_degradation) ** (year - 1)
LCOE_real = (CAPEX_year_0 + sum(OPEX_year / (1+r)**year)) /
sum(E_year_kWh / (1+r)**year)
NPV_real_tariff = real_tariff_per_kWh * discounted_energy_kWh - discounted_cost
CAPEX occurs at year zero; generation and fixed OPEX occur at year end. A real discount rate is applied to constant real JPY costs. The capacity factor includes wake, electrical and availability losses; degradation is an additional scenario assumption. This is not a price-capture or debt-financing optimizer.
The reproducible gallery example plots CAPEX multipliers (0.7–1.5) under three illustrative real discount rates (3%, 6%, 9%) and capacity-factor (26%–46%) sensitivity under different construction cost multipliers. Ranges are scenarios, not confidence intervals. It also checks that NPV is approximately zero at its computed flat-tariff break-even price.
Illustrative example only: 100 MW, 25 operating years, real 6% discount rate, 36.6% net capacity factor and 0% degradation yield approximately JPY 25.8/kWh simplified LCOE; a purely hypothetical JPY 30/kWh fixed tariff implies approximately JPY 17.2 billion NPV. These outputs do not reproduce JWPA’s reported LCOE and do not describe Kashima Bay.
Reproduce offline#
uv sync --locked --all-packages
uv run --no-sync python examples/analysis/plot_japan_offshore_wind_cost.py
uv run --no-sync pytest methods/tests/test_offshore_wind_economics.py data-retrieval/tests/test_jwpa_cost.py
The example appears under Analysis → Offshore wind economics in the two-level gallery.
Decision limits and next research#
This screening model omits measured hourly wind/generation, turbine power curves, installation delays, location-specific EPC/O&M quotes, volatile or inflation-linked contracts, replacement CAPEX, degradation calibration, curtailment, taxes, subsidies, dynamic prices, debt structures, transmission, and project-specific site risks. Do not use it to determine whether any named project should proceed or withdraw.
Next: connect licensable hourly Japanese offshore resource and project bid data, model net production with turbine-specific power curves, and compare calibrated P50/P90 scenarios to published actual costs and contract terms.