Using economic experiments in behavioural insights

9 August 2017Danielle Merrett and Sam Miller

Categoryeconomic experiment

Tagsbehaviour change, behavioural insights, economic experiment, trial

So you have a great idea about how you could change people’s behaviour. You want to go ahead and test your idea out, but you don’t know if people will respond to the intervention, or if it will have the impact you hope. Rather than going ahead and investing heavily in a field trial, you could use an economic experiment to test your idea first.


How it works

People are invited to participate in an experiment. They are seated in separated partitioned stations, so they can’t be influenced by others. Then they are asked to make decisions based on information they are given.

What makes economics experiments different from other psychological experiments is that the decisions in economic experiments involve explicit trade-offs, for example, if you do A, then you can’t do B, or if you do A it will cost you X dollars.

In economic experiments the outcome for each participant depends on the decisions they make and the decisions of the other participants they interact with. Experiments try to make decisions as reflective of real life scenarios as possible. People tend to make more realistic, or truthful, choices when their payment at the end reflect the trade-offs they had to navigate during the experiment.

Economic experiments and kidney donations

Economic lab experiments have been used to test a new system of organ donation in the United States. Kidneys are often donated by a recipients’ family and friends. In addition to have a willing donor, the kidney also needs to be a biological match to the recipient for it to work. Economist Al Roth had an idea of a new system of organ donation that would increase the number of kidneys available to recipients whose families were not a match. He did this by creating a kidney exchange where a match is made between people in need of a kidney and members of different families who are willing to donate a kidney, but who are incompatible to donate to their loved one.

Before Roth tried his idea out on people, he first tested it in an economic experiment to see if people would, given the scenario, participate in the exchange. He used the learnings from the lab to further refine the design of the intervention and to show others there was merit in testing the idea in the field. His idea was eventually implemented in hospitals across the USA, resulting in many more kidney donations and lives saved. Al Roth won The Nobel Prize in Economics for his work in improving kidney matches.

Pros and cons of economic experiments


Unlike surveys and interviews, where participants may feel as if they need to tell researchers what they think you want to hear or paint themselves in a favourable light, economics experiments rely on participants anonymously revealing their preferences when faced with real life decisions.

Additionally, because of how economic experiments are structured, we can gain insights into how people interact with each other. Through them we can explore behaviours like trust, generosity and honesty. We also use them to explore the effects behavioural insights techniques such as social norms, messenger effects, and cognitive biases in different contexts.

Another pro is that we can get quicker results. Because economics experiments are run in a more controlled environment than field research, results can be analysed much faster. Also, in comparison to field data, often the effects of the intervention can be seen more easily through an economic experiment. This is because what we are looking for isn’t hidden among any number of confounding variables as in field trials.


Economic experiments can often be criticised for relying on a WEIRD (Western, Educated, Industrialised, Rich, Democratic) population of university students. However developments in technology are making these experiments more common and easier to implement, allowing researchers to take their experiments to the people and places relevant to the study. That said the classic computer lab is still the most common place to build and refine these experiments.

Economic experiments are no substitute for well-grounded qualitative research that gives you a rich insight into the people whose behaviour you are trying to change. Economic experiments carried out over short time periods such as minutes and hours, cannot always give you information on long-term behaviours changes and habits, especially where the outcomes are qualitative. But well thought-out experiments, built in collaboration with an expert in experimental design, can often overcome these issues. Some experiments can last days, weeks, or even months!

Economics experiments offer the freedom to explore behaviour change ideas in a safe space, and then proceed with greater confidence into a field trial. They provide the robust evidence you need to stand behind your novel interventions as you break new ground in the BI-sphere.

An experienced behavioural economist can help you design your policy problem in an experimental setting. From there, they can run a series of experiments to test different behavioural interventions and begin your journey of scientific discovery. And you never know, like Roth, your big idea could lead to a Nobel Prize.