US Waste-to-Energy Leaves Its District Heating Revenue in the Condenser

Every waste-to-energy plant in the United States makes two products, and it sells the cheaper one. The first is electricity, which drops into a wholesale market that solar has spent a decade dragging down toward $30 a megawatt-hour. The second is heat, two to three times the raw energy of the electricity, and most US plants shove it through a condenser and book it as a cooling load. The Energy Information Administration is blunt about the split: as of 2022, about 90% of the energy US waste-to-energy plants deliver reaches the grid as electricity, and only around 10% leaves the fence as steam. My position, after a decade putting prices on these assets, is simple. Waste-to-energy district heating isn't a technology the US lacks. It's a revenue line American developers never wrote onto the term sheet, because the kilowatt-hour they optimized the whole plant around is the byproduct.

That reads like an engineering failure. It isn't one. Whether the combustor is a Martin reverse-acting grate or a plain mass-burn line, the hardware on a US plant is the same hardware a Danish plant runs. The only real difference is a pipe leaving the building with hot water in it, and a signature on the far end of that pipe.

The 90/10 Split Isn't Physics

Burn municipal waste for electricity alone and you convert maybe a fifth to a third of the fuel energy into power; the balance leaves as low-grade heat you actively dump. Capture that heat instead and total energy utilization climbs past 80%. The EPA puts combined heat and power at 65 to 80 percent efficient against roughly 36 percent for the average fossil-fired US plant, which is the same fuel doing close to double the work. The turbine hall isn't where this gets won or lost. I've argued before that the steam turbine is the easy part of a waste-to-energy plant, and I'd say it again: the boiler and the genset are mature, boring, and solved. Finding someone to buy the heat is the hard part, and it always has been.

Put a number on what gets discarded. Those US plants push out roughly 14,000 gigawatt-hours of electricity a year, and that's the smaller half of what they actually generate; the larger, thermal half mostly leaves as warm water nobody buys. That discarded thermal half is the whole case for waste-to-energy district heating. It's not a rounding error on the energy balance; it's most of the energy, thrown away because it never had a contract.

The rest of the world settled this decades ago. Denmark pipes district heating to nearly all of Copenhagen and about two-thirds of Danish homes (decades of trenching, not one ribbon-cutting), and it runs its incinerators as heat plants that happen to spin a turbine. Sweden sends more than 97% of its incinerated waste into district heating rather than power, according to Swedish district-heating data. Walk the European waste-to-energy projects commissioned this century and the heat main is usually in the trench before the turbine is even ordered. There, the renewable energy from waste that ends up as electricity really is a leftover, and the financing was built to say so.

Heat Has No PPA

So why build a plant that makes heat and then decline to sell most of it? One asymmetry decides it, and it lives on the balance sheet. Electricity arrives with a grid interconnect, a market clearing price, and a power purchase agreement a lender has financed a thousand times over. Heat arrives with none of that. To sell heat you need a physical network in the ground, an anchor load close enough that pumping and thermal loss don't swallow the margin, and a bilateral heat-supply contract with a counterparty a project-finance desk has never seen a template for.

Debt gets sized against the contracted line, not the one you hope to build later. A signed PPA is collateral a bank recognizes on sight. A heat main you might construct if the county hospital signs up is not. So the developer optimizes for the kilowatt-hour even though it's the low-value product, because the kilowatt-hour is the one the capital structure can lend against. That isn't stupidity. It's the debt doing exactly what it was designed to do, and equity stays in the room only as long as the offtake stays in writing.

I learned how fragile the phrase "revenue line" is the expensive way. In 2024 my RDF export book lost a 40,000-tonne position to one chloride-spec rejection: the cargo failed the buyer's spec at the load port and flipped from an asset into a demurrage bill almost overnight. The tonnes on the water hadn't changed. The contract had. Heat is the mirror image of that trade: you can't ship it across an ocean, but nobody can bounce it back at the port either, provided the pipe and the buyer exist. The value was never in the commodity. It sat entirely in whether someone was obligated, in writing, to take it.

The Heat Load Is Already There

The US isn't Scandinavia. We don't have the winters or the compact cities. District heating only pencils where it's freezing and everyone lives in an apartment block.

I hear some version of that on nearly every diligence call, and it's about a third right. Cold, dense cities do make the arithmetic easier. But the excuse quietly assumes the US has no heat customers, and that assumption is wrong. Hospitals, universities, prisons, military bases, food processors, and paper mills all buy large, steady, year-round heat. Several US cities already run district steam loops: New York, Boston, Baltimore, Philadelphia, Minneapolis. The Hennepin Energy Recovery Center sells steam into downtown Minneapolis; Baltimore's plant has fed the city's steam system for years. None of those are Nordic. They work because a heat host was under contract before the concrete was poured.

That's the actual tell. The plants that sell heat signed a buyer first. The ones that don't were sited for a cheap grid tie and a convenient water source, and by the time anyone thought to ask about heat the nearest load was three miles and a freeway away. It's a sequencing problem wearing a geography costume, and the urban waste recovery planners on the municipal side are usually the ones who spot it first, because they can see the hospital sitting on the same block as the transfer station.

What Bankable Heat Looks Like

The structures that actually close share one feature: whoever controls the waste also has real skin in the heat. A municipality that owns the tonnage and also owns the hospitals, the civic campus, and the university it wants warmed can underwrite both sides of the trade at once. An industrial host that needs process steam can sign a take-or-pay a lender will finally accept as offtake. Federal tools help at the edges, PURPA's cogeneration qualifying-facility rules give a cogeneration plant certain rights to sell into the grid, but a statute doesn't sign a heat contract for you. This is the deal shape waste-to-energy district heating has always needed, and the order of operations is the whole game: you lock the heat offtake before you build the plant, not after.

What does the contract itself have to carry? A tenor that matches the debt, fifteen to twenty years, so the heat revenue lands across the same horizon the lender is underwriting. A take-or-pay floor, so a warm winter doesn't gut the year. Price indexation that tracks the host's alternative, usually natural gas (let gas prices crater against a fixed heat price and that line quietly bleeds), so the plant isn't stuck selling heat below the cost of the fuel it displaces. Land those three terms and a heat contract starts to look like the PPA the electricity side has always had. Miss them and you've signed a handshake, not an offtake. There's an ESG dividend on top: every therm of gas the network displaces is a documented emissions cut, and ESG-compliant projects that can show displaced fossil heat, not just clean electrons, screen better with the funds now writing the largest checks in this sector.

Get the order wrong and no retrofit saves it. The proven waste-to-energy technology for pulling usable heat out of a boiler has existed for forty years; an extraction or back-pressure turbine and a couple of heat exchangers are not exotic. What's scarce, and what actually gates the project, is a counterparty willing to commit to buying that heat for twenty winters. Equipment you can order in an afternoon. A twenty-year heat obligation you have to negotiate, and most US developers reach financial close without seriously trying.

Where It Doesn't Pencil

None of this makes heat the right call everywhere. The limits are worth naming rather than buried, and not every plant should chase a heat contract; I'd distrust anyone who said it penciled on every site. Past roughly two kilometers to the nearest steady load, pumping and thermal losses start eating the premium, and a plant stranded out by the interstate genuinely isn't worth re-plumbing. A facility in the US South with no winter heating demand needs a year-round industrial steam host or a district-cooling loop driving absorption chillers, or the heat sits idle two-thirds of the year. The smallest units, burning a single county's waste, rarely sit near a load big enough to matter. And where a grid genuinely pays up, through capacity payments, renewable energy credits, or a data-center load next door, merchant power can out-earn heat on its own. Actually, let me walk back the 80%-plus number I threw out earlier: that utilization only holds when the heat load runs most of the year. A plant selling heat four months out of twelve looks, on an annual basis, a lot like the electricity-only plant it was trying not to become. On any honest read of waste-to-energy services, heat is a revenue line only where a real load shows up, and a rounding error everywhere else.

So when a US developer tells me district heating "doesn't work here," I hear something narrower and more honest underneath it: we never lined up a heat buyer, and no lender was going to fund a pipe to nobody. Fair enough. That's a real constraint, and it's the one worth fixing rather than narrating. But don't tell me the heat is worthless. The plant is already engineered to make it, in volumes that dwarf the electricity, and every January it goes out the condenser and up into the sky while a hospital two miles down the road burns gas to keep its wards warm.

The heat was never the problem. The counterparty was.

Sources & Notes

The 90/10 split, the plant count, and the capacity figures come from the U.S. Energy Information Administration's 2022 waste-to-energy briefing, Waste-to-energy plants are a small but stable source of electricity.

The efficiency comparison, combined heat and power at 65 to 80 percent against about 36 percent for the average fossil plant, is straight off the EPA's page on CHP benefits.

The Danish and Swedish district-heating shares are widely reported national figures from those countries' heating sectors; read them as directional rather than audited. The 40,000-tonne RDF position and the chloride rejection are from my own 2024 export book, not a public dataset.