Subsidies for direct air capture must exceed $200–330 per ton of CO2 and be sustained for decades. That’s not an activist estimate — it’s the central finding of the first rigorous uncertainty analysis of DACCS scaling, published by researchers at Politecnico di Milano and ETH Zurich.
The total public bill? Between $900 billion and $3 trillion.
Before you close this tab: the investment pays back by mid-century. But only if — and this “if” is load-bearing — we simultaneously cut emissions hard.
What the paper actually did
Chiani et al. ran a global sensitivity analysis on DACCS deployment using the WITCH integrated assessment model. They tested three technology pathways: liquid solvent systems, solid sorbent systems, and CaO ambient weathering. Then they threw uncertainty ranges at everything — technology costs, learning rates, energy prices, policy trajectories — and asked: what does the probability distribution of DAC deployment actually look like?
The answer is humbling. Most scenarios show modest DAC uptake. The distribution has fat tails: there’s only a 4–6% probability of reaching gigaton-scale removal by mid-century. That’s not zero, but it’s a long way from the confident gigaton projections you see in climate roadmaps.
The subsidy floor is non-negotiable
Across every scenario, every technology pathway, every assumption set, subsidies below $200/tCO2 produced essentially no deployment. The range of $200–330/tCO2 isn’t a ceiling to aim for — it’s a floor below which nothing happens.
And these subsidies can’t be a five-year experiment. The researchers found that stop-and-start funding destroys learning curves. DAC technology costs decline through deployment, and deployment requires sustained financial support measured in decades, not election cycles.
The giant asterisk
Here’s where it gets uncomfortable for DAC maximalists: without aggressive emission reductions, DAC subsidies become a money pit. The economics only work when DAC supplements deep decarbonization, not when it substitutes for it.
Pair DAC with strong climate policy? The $900B–$3T investment generates net positive returns by 2050. Use DAC as an excuse to keep emitting? The same investment becomes a bottomless hole.
This is a policy paper dressed as a technology paper. The researchers aren’t saying DAC doesn’t work. They’re saying DAC works inside a system where emissions are falling — and falls apart outside that system.
What this means
For policymakers: commit to sustained subsidies above $200/tCO2. The US 45Q tax credit maxes out at $180/tCO2. That’s below the floor. Europe’s subsidies are fragmented. Nobody is currently meeting the bar this paper sets.
For investors: the technology risk is manageable. The policy risk is enormous. Your DAC bet is really a bet on whether governments can maintain climate policy for 20+ years.
For everyone else: DAC is expensive, uncertain, and worth funding — in a world that’s also cutting emissions. All three of those things are true at once.
Source: Chiani et al. (2026), arXiv