The Subsidy That Is Priced Per Kilo

Klimatklivet is Sweden's national investment support for local climate measures, run by Naturvardsverket under Klimatklivsförordning 2015:517. The form does not ask what the project costs. It asks what the krona buys. The ranking metric is the klimatnyttokvot: kilograms of carbon dioxide equivalent removed per krona invested. Every application in a round is sorted on that one number and a breakpoint is drawn through the list. Above the line goes forward. Below the line goes back. Cost is not the score. Size is not the score. Ambition is not the score. The arithmetic is the score. Across 2015 to 2024 the programme received 42,326 applications and approved 27,114 of them. The grant awarded was SEK 19.7 billion against SEK 46.0 billion of investment. About 86 per cent of that grant went to companies. The average approved project carried 1.14 kilograms of carbon dioxide equivalent per krona of investment, or 0.38 kilograms per krona of grant. That second figure is the price of the subsidy, and it is published.

Riksrevisionen's review of the programme published the breakpoints, and they are the most useful numbers in any file. 2015, rounds one and two: 1.00 kilograms per krona. 2016, round one: 0.75. 2016, rounds two and three: 1.00. 2017, rounds one to four: 0.75 to 1.00. The current estimated band is 0.5 to 0.75. That is the pass mark, and it moves with the queue rather than with the applicant. A project that cleared at 0.9 in 2015 does not clear at 0.6 in 2026. Nothing about the project changed. Naturvardsverket can approve below the breakpoint where a project carries technology diffusion or another environmental benefit. The route exists. It is not the route to plan on, because the probability of being picked out of the queue is not something the applicant controls. The single most common reason for rejection, at 22 per cent of rejections, is insufficient emission reduction per krona invested. The file that loses is rarely the weakest technology. It is the one that never did the division.

Emission reductions are not argued. They are calculated, and Naturvardsverket publishes the values to calculate them with. The current version of that document is dated 25 March 2026. Heating oil, Eo1: 288 kilograms of carbon dioxide equivalent per megawatt hour. Fossil gas: 249. Household waste: 144. Coal: 385. Peat: 425. Those are the expensive baselines. The number that replaces them is usually zero. Wood industry waste sits at 16.9. Forest chips at 20.5. Pellets and briquettes at 45. Diesel MK1 at 324 and petrol MK1 at 314, against HVO at 58. Then the trap. Swedish grid electricity carries 59 kilograms per megawatt hour and average district heating 55. Waste heat carries zero. So displacing the grid earns almost nothing, because the grid is already one of the cleanest in Europe. The kilos are not in the electricity. They are in the oil boiler, the landfill, and the carbon that stays in the ground, in that order.

Three pathways carry real weight, and one installation can claim more than one of them. The first is waste diversion. A tonne of household waste sent to landfill carries roughly 1,000 kilograms of carbon dioxide equivalent in methane and transport. The same tonne burned for heat and power is valued at 144 kilograms per megawatt hour. The gap between those two numbers is the grant. The second is fossil heat. A megawatt hour of thermal output from heating oil costs 288 kilograms; from fossil gas, 249. A megawatt hour recovered from waste heat costs nothing. Scale the difference by the hours the plant runs and the arithmetic writes itself. The third is biochar, added as an explicit priority measure in February 2024 for production and use as a heat source, a soil amendment or a carbon sink. The values document names biochar as eligible and then declines to give a standard factor, because sequestration is calculated per project. The applicant supplies the number. Nobody supplies it for them. One installation drawing on all three pathways is not three times the paperwork. It is one denominator divided into a much larger numerator.

Cost is not the score. The arithmetic is the score.

Six rounds run in 2026. Two are open to most project types and four are targeted. Fuel conversion ran 2 to 13 February. Open round one ran 2 to 20 March. Public charging, first tender, 13 to 30 April. On-farm biogas to electricity and heat, 25 May to 5 June. Open round two runs 24 August to 4 September. Public charging, second tender, 19 October to 6 November. The portal opens at 08:00 and closes at 16:30. Not midnight. A 16:30 close is a hard close, and Naturvardsverket does not accept supplements after the deadline. An incomplete application does not get a second pass. The open round excludes public charging infrastructure, on-farm biogas for electricity and heat, and hydrogen refuelling stations, each of which has its own round or tender. Everything else sits inside it: energy conversion, biochar, biogas, waste heat recovery, circular flows, industrial efficiency, fuel switching, biofuel production. Processing takes four to six months after the round closes. The decision arrives in the following calendar year's planning, not this one's.

Three requirements decide whether the arithmetic reaches a queue at all. The applicant needs a Swedish organisation number. A foreign company applies through a registered subsidiary, a branch, or a Swedish partner that owns the installation and buys the technology. Registration of a private limited company takes five to ten working days and SEK 25,000 of share capital. Every document is in Swedish. The project description, the emission calculation, the investment case, the financing plan. And the project must be anchored in the county's climate and energy strategy. All 21 counties publish one, and the application has to show how the installation sits inside it. Additionality is the fourth gate and the quietest one. The investment must not be profitable on market terms, must go beyond a legal requirement, and must not have started. Procurement before the decision revokes the grant. The applicant also files the latest annual report, balance sheet and income statement, a certificate of registration, an F-tax certificate and a VAT registration. All of it is published. All of it is recomputable by a stranger with the values document open in another window. That is the whole point of the programme, and it is why the file that shows its division wins.