Build a grid

Annual demand (TWh)

coal 35%oil 3%gas 23%biomass 2%hydro 15%nuclear 9%wind 8%solar 5%

Deaths

297,000–2,470,000 per year

≈ one death for every 2,000–15,000 people's annual electricity, at this mix.

This mix blends to about 10.1 deaths/TWh. For scale, an all-coal grid is ~24.6, all-gas ~2.8, all-wind ~0.02.

counted, about 22,100/yr · modeled, about 281,000/yr

CO₂

11,300–16,800 Mt/yr

377–559 gCO₂eq/kWh

Land

194,000–1,620,000 km²

At the high end, roughly the area of Iran — about 1.1% of global land area.

Wind uses a dual land figure: direct occupation to total wind-farm area.

Cost

Include Lazard 2026 firming-cost adder for wind and solar

53.5–116 USD/MWh

1,610–3,490 USD bn/yr

Oil, hydro, and biomass render as no comparable cost data and are omitted rather than mixed with another methodology.

The true cost

The bill above, plus the two harms the market never charges for — carbon and mortality — each priced at a value you choose. Mortality and carbon.

Value of a statistical life

$14.1M

Social cost of carbon

$190

bill $85 · carbon $82 · mortality $142 /MWh

True cost ≈ $310/MWh (3.6× the bill)

$9.29T per year, all-in

The bill covers only sources with a Lazard cost figure; carbon and mortality are counted for the whole mix. A coal-heavy grid's true cost runs well above its bill; a clean grid's barely moves.

Hydro dominated by one 1975 event — Hydro's global death rate is dominated by the 1975 Banqiao Dam failure; excluding it, the rate is about 0.04 deaths/TWh.

The safest sources cannot be ranked apart — Nuclear, wind, and solar cannot be reliably ranked against one another; their uncertainty ranges overlap.

Fossil rates come from an older European study — Fossil death rates derive from a 2007 study using European pollution controls; newer global air-pollution research suggests higher impacts.

Nuclear deaths are modeled, not counted — Nuclear cancer deaths use the linear no-threshold model, a contested assumption.

Wind land is a dual figure — Wind land use is shown as a dual figure: total wind-farm area includes turbine spacing, while direct land occupation is much smaller.

One source drives the death total — Coal accounts for about 85% of the deaths in this mix — the total mostly reflects that one source, not the grid as a whole.

Some coefficients are missing, not zero — Missing coefficients omitted rather than zeroed: oil cost, biomass cost, hydro cost.