In this piece, by Sebastian Berger and Felix Creutzig, they challenge the assumption that our preferences are fixed, arguing instead that they are malleable, shaped by our context, experiences and social norms. This means that changing the environments in which we make choices could shift our preferences towards more sustainable or healthier ways of living, while still maintaining or even improving our wellbeing.
When decision makers in Ljubljana attempted to make the city centre car-free in 2007, the policy did not receive a warm welcome. Allegedly, the former mayor was being confronted at city hall and slapped by angry protesters. Objections were made by shop owners fearing revenue losses, by inhabitants who feared that their homes would become inaccessible, and by older people who feared being left out of the city’s life. Today, air pollution has fallen substantially, noise pollution has gone down as well, and the policy is widely approved by the public. Older and mobility-impaired people can use free electric carts to get around the city centre, access by vehicles to restock shops is restricted to certain hours, and many cafes have opened up on now car-free streets to create a novel lifestyle. The policy has been so successful that the tourist office now even markets the car-free city centre as one of the city’s defining characteristics.
What happened in Ljubljana is not an isolated case. Similar patterns have been observed with congestion pricing in Stockholm, where opposition shifted after people experienced the effects of the policy. In Gothenburg, attitudes towards a congestion tax became more positive following a trial, even though a subsequent referendum rejected the tax. Opposition to urban tolls also declined after their introduction in Bergen, Oslo and Trondheim. Comparable shifts have been documented for smoking bans and seat-belt legislation in different countries.
The phenomena described here show something puzzling, at least for those who accept the underlying ontology of neoclassical economics or rational choice theory. If people know what makes them better or worse off and if their preferences are stable and “given” rather than shaped by their environment, why should their assessment of a policy change so substantially simply because they have experienced the policy? And can this be applied to behavioural changes required for climate change mitigation?
Demand-side mitigation and its potential
Much of the climate debate focuses on producing cleaner energy and developing new technologies. But there is another side to the equation: changing how we use energy, transport, food and buildings. These are often called demand-side solutions. They include eating plants and less meat, reducing food waste, living in more energy-efficient homes, walking and cycling more, using public transport, and adopting low-carbon technologies such as heat pumps. According to assessments published during the most recent IPCC assessment cycle, demand-side measures could reduce greenhouse-gas emissions by 30–70% relative to counterfactual emissions, while maintaining or even improving the services people receive.
Despite their potential, many demand-side solutions are often dismissed in public debates and by citizens because they appear to interfere with people’s freedom of choice. A vegetarian day in a university cafeteria removes the opportunity to eat a meat dish on that day. A cycle lane may take space previously used by cars. A congestion charge makes driving more expensive. A ban on short flights removes the option altogether. From a conventional economic perspective, this creates a straightforward problem. People have preferences. They choose according to those preferences. If a policy prevents them from making the choices they would otherwise make, the policy risks making them worse off. But there is a surprisingly strong assumption hidden in this argument. It assumes that preferences come first and that our current way of living and our choices are largely a consequence of those preferences, rather than the other way around.
The idea of fixed preferences
A cornerstone of neoclassical economics and rational choice theory is the idea that individuals possess preferences that are sufficiently stable and well-defined for their choices to reveal what they want. This assumption is enormously useful. It makes human behaviour mathematically tractable and allows economists to infer welfare from people’s choices. But what if our preferences are not fixed? Our recent paper reviews a large body of research suggesting that they are not. First, preferences can be context-dependent: what we choose changes depending on defaults, framing, reference points, and the environment in which a choice is made. Second, preferences can be constructed: rather than simply revealing an existing preference, making a choice can help create one. Third, preferences are often endogenous: social environments, institutions, markets, and habits shape the preferences people subsequently bring to their choices. And, fourth, there are cognitive limits to how completely people can specify and compare all the alternatives available to them as well as the degree to which they believe themselves to be malleable
Consider something as mundane as food: It is impossible to specify, in advance, our preferences over every possible meal and condiment. There are effectively infinitely many combinations of ingredients, recipes, restaurants, and social situations. Instead, we encounter foods, develop habits, observe what other people eat, and gradually learn what we like. In other words, preferences emerge through lived experience in a given environment, and our choices reflect our accumulated experience of the environments in which we live. So, preferences can be the cause of behaviour as well as its consequence.
Experience and sustainable societal transformation
Imagine asking a resident of a car-centric city: “How much would we need to pay you in order to make our city car-free?” It is tempting to interpret the answer as a fixed preference that should guide policy. But there is an epistemic problem. Cycling in a city without safe infrastructure is very different from cycling in a city where cycling is easy, safe and socially normal. Likewise, walking through a noisy traffic-filled street is not the same experience as walking through a quiet, pedestrian-friendly neighbourhood. It is therefore quite plausible that people do not take up cycling as an individual strategy in such a city – and that they are consequently unable to reliably state how they would value a different urban environment.
Our paper therefore argues that many demand-side transformations involve such an experience gap and that preferences are largely malleable. Right now, people are asked to evaluate a future environment using preferences formed under the existing one. Yet the policy itself may change the context in which our preferences are formed. This creates a fascinating possibility: some policies that appear to constrain choice may simultaneously create opportunities for people to discover alternatives they did not previously know they would value. That is precisely what makes the Ljubljana example and other transformations so interesting. The question is not simply whether people initially wanted a car-free city. They may genuinely have disliked the idea. The deeper question is whether their initial preferences were a reliable prediction of how they would value the city after experiencing it.
The evidence suggests that, in at least some cases, preferences were not reliable. This does not mean that every unpopular climate policy is automatically good, or that governments should simply ignore what people want. There are legitimate conflicts between freedom of choice, human needs, well-being, ecological integrity and justice between generations. Our argument is not that preference satisfaction should be discarded, but that it should not automatically be treated as the sole measure of welfare. Especially, when it comes to forming structures that shape behaviour, such as urban form or digital environments, structures should be formed towards sustainable outcomes, we argue. We need to take seriously the possibility that preferences themselves are part of the transformation and that the policy toolbox includes methods that allow experiential learning through pilots, experiments and tests. What kinds of lives and preferences could people discover if we changed the world around them? That is a much more difficult question for economics. But it may also be a much more useful one for a society trying to transform itself. It requires policy makers to be courageous and visionary, in a sense to be truly open to transformative change.
You can read the full paper here:
Berger, S. and Creutzig, F. (2026) ‘Malleable Preferences and the Normative Desirability of Demand‐Side Solutions to Climate Change’, WIREs Climate Change, 17(4), p. e70079
Sebastian Berger is Assistant Professor of sustainable development at the University of Bern, Switzerland. His research focuses on behavioural scientific support of climate policies as well as environmental psychology. He received his PhD in economic and social psychology from the University of Cologne. He is associate editor at the Journal of Environmental Psychology. Drawing on psychology, sociology, economics, and environmental science, his research aims to identify the key factors shaping individual and collective decisions in the realm of sustainability.
Felix Creutzig is Professor of Innovation and Policy Acceleration & Bennett Institute Chair and head of the working group on Cities at the Potsdam Institute for Climate Impact Research. He is member of enterprise, public district, municipal and national climate advisory boards. Previously he had been visiting fellow at Princeton University, the University of California, Berkeley, and the China Energy Foundation in Bejing. He studied physics and medical studies and holds a PhD in computational neuroscience/machine learning. His groundbreaking research on climate change mitigation on urban scale has been recognised by the Piers Sellers Prize of the University of Leeds in 2017. Felix Creutzig coordinated a chapter on demand, services and social aspects in the recent assessment report on climate solutions by the Intergovernmental Panel on Climate Change. He is also leading a crosscutting effort introducing digitalisation and AI in the context of climate change mitigation, both as method and as governance dimension.