Countries worldwide have dramatically ramped up their climate policies over the past two decades. The number of climate measures has quadrupled since 2000, with some datasets showing a fifteen-fold increase.
Governments now deploy dozens of different policies simultaneously – carbon taxes, renewable energy subsidies, building codes, emissions standards, research funding, and more. They all work together, influence each other, and jointly affect emissions.
But when emissions drop (or don’t), how do we tell which policies deserve the credit? Which ones actually make a difference in any circumstances? That’s the challenge facing climate policy researchers today – and we’ve found a way to solve it.
Our new research analysed 1,737 individual climate policies across 40 countries over 32 years, and we identified 28 policies that consistently reduce emissions across diverse contexts. More importantly, we developed a new approach that could transform how researchers evaluate policies in any field where complexity keeps growing.
Old approaches
Traditional approaches to evaluating climate policies struggle with this new reality.
Some researchers study individual policies in isolation. This kind of research can be helpful, but it’s like judging a football player only when they’re playing solo, ignoring their teammates. The results – whether positive or negative – might not apply when the player joins a different team with different tactics.
Other researchers look at entire policy “packages”. To extend our football metaphor, this is like judging a whole team’s performance without identifying which individual players contribute most. Here, you know whether the team is winning or losing, but not why.
A high-profile 2024 study applied a different strategy: looking for sudden drops in emissions, then checking what policies were introduced just before. But even this approach risks missing policies that work gradually over time rather than producing dramatic immediate results.
The fundamental problem with analysing all policies simultaneously to see which ones actually work is that you quickly run into a statistical wall. Too many variables, not enough data. It’s like trying to solve an equation with more unknowns than knows.
Read more:
Why some climate policies are more popular than others – a psychologist explains
A statistical filter
Our solution uses a statistical approach which is akin to a series of increasingly strict quality filters.
Think of it this way: when you have too many potential explanations for why something happened, you need filters to separate real effects from statistical noise. To do this, we use what statisticians call “Bayesian priors” – essentially, different sets of assumptions about how policies should behave.
Importantly, our approach doesn’t just pick one filter and stick with it. Instead, we look for policies that pass through multiple different filters. If a policy shows up as effective across different tests with different assumptions, we can be more genuinely confident that it works.
So which policies actually work?
Our conservative approach identified 28 climate policies with high certainty of emission reductions. These span a range of instrument types:
Carbon pricing and taxation (8 policies): Carbon taxes across sectors, emissions trading schemes, congestion charges and fossil fuel excise taxes all show robust effects, even when controlling for all other policies. This counters the claim that carbon pricing only seems effective because it’s usually accompanied by complementary measures.
Read more:
How Europe’s new carbon tax on imported goods will change global trade – and our shopping habits
Energy efficiency and standards (5 policies): Building energy codes, air emission standards, minimum energy performance standards and motorway speed limits consistently reduce emissions.
Renewable energy and research (11 policies): R&D expenditure on carbon capture, nuclear, hydrogen, energy efficiency and renewables, and planning for renewable expansion and auction schemes all reliably drive emission reductions.
Reporting and accountability (3 policies): Greenhouse gas emissions reporting requirements across sectors show significant effects.
Subsidy reduction (1 policy): Eliminating fossil fuel subsidies in transport reduces emissions.
Practical impact
To illustrate real-world implications, we modelled emissions in Portugal across four sectors: buildings, energy, industry and transport. If Portugal had implemented all 28 effective policies at maximum stringency since 2000, cumulative emission savings would total 538 Mt CO₂eq. This is equivalent to an entire emission-free year across these sectors for South Korea, whose economy is about six times larger than Portugal’s.
Country-specific analysis is vital, as it pinpoints clear, targeted opportunities. Germany, for instance, could enhance climate action through stricter motorway speed limits. Australia, Canada and Japan could significantly improve performance via higher fossil fuel excise taxes (taxes levied on manufacturers as opposed to consumers). Our analysis enables policymakers to identify the blind spots in otherwise ambitious climate strategies.
The star players
Our research shows that effective climate action doesn’t depend on finding one perfect solution. Multiple pathways exist, but some instruments prove more reliable than others – carbon pricing, taxation and investment in renewable energy research are the star players who will improve any team they join.
Countries like Sweden and Norway have successfully implemented all 28 effective policies (though with varying intensity), proving this approach is politically viable. But even climate leaders like Germany have blind spots, like the aforementioned motorway speed limits.
As climate policy continues expanding and pressure mounts to ensure it actually delivers results, this approach provides a powerful new tool. It helps policymakers avoid wasting resources on ineffective measures while identifying proven strategies that work across different contexts.
Beyond climate action
While our findings give policymakers a clear list of climate policies that actually work, the approach itself is just as significant.
Policy complexity isn’t unique to climate. Healthcare, education, financial regulation, social policy – in all these areas, governments keep adding new programs, new rules, and new incentives. Researchers everywhere struggle with the same question: which specific interventions actually work when everything’s tangled together?
Our filtering approach offers a template. When policies multiply, traditional evaluation methods struggle to keep up. This new approach lets researchers model everything simultaneously while maintaining statistical rigour: identifying what genuinely works in complex environments.
A weekly e-mail in English featuring expertise from scholars and researchers. It provides an introduction to the diversity of research coming out of the continent and considers some of the key issues facing European countries. Get the newsletter!
