Adelaide Laboratory for Experimental Economics (Adlab)

During an experimental session you will be seated at a computer. The experimenter will provide you with detailed instructions, where everything you need to know is explained. In particular, these instructions explain how your choices, and those of other participants, affect the amount of money you receive.

Once you have read and understood the instructions you then make you decisions via the computer terminal. At the end of the experiment you might be asked to fill in a short questionnaire. Finally, you will be paid your earnings in cash and can leave. Duration of experiments vary. A typical session takes about an hour. Payments vary as well from experiment to experiment and are on average about AUD 20 per hour.

All experiments conducted at AdLab have Human Research Ethics Committee approval.

The other social science discipline that regularly uses experimental methodology is psychology. Economics experiments differ in three key aspects:

1. Experimental economics links real money to participant’s actions. By doing so, experimental economics can transport the material incentives from the real world into the laboratory. This increases the validity of experimental results.
2. In economics experiments you receive detailed scripted instructions that explain all aspects of the experiment and clearly describe how actions influence earnings.
3. Economics experiments never use deception. All information you receive is absolutely truthful.

One of the most difficult tasks in a social science like economics is establishing causality. Suppose someone claims that the more ice-creams are sold in South Australia, the more cars are stolen. It is important to know if there is a causal relationship. If ice-cream consumption increases car theft, then we probably should ban the sale of ice-cream. Experiments can provide the controlled environment necessary to find out if there is a causal link between the two. We could randomly choose a few days per year where no ice-cream can be sold and compare car thefts on these days with those on others. This experiment would establish that there is no causal relationship between ice-cream consumption and car thefts, and that another factor (e.g. weather) may be influencing both simultaneously.

In experimental economics we establish causality by comparing behaviour in different treatments that only differ in one instance. Differences in behaviour can then be causally attributed to this treatment variation. Ernst Fehr and Simon Gachter showed in a seminal paper that peer punishment improves cooperation. In some treatments they allowed participants to destroy money of others by paying some of their own money, and compared behaviour to another treatment where this punishment option was absent. This finding has applications to many real-world dilemmas where cooperation between people or organisations can increase overall welfare (e.g. teams in the workplace, State action on climate change, contributions to group assignments).

Experiments can be used to inform a wide range of issues in government, the private sector, and not-for-profit organisations. Where field data is unavailable or not easily collected, experimental data can support the formulation of evidence based policy and strategy. It is common for economics experiments to replicate real-world markets, for example the best way to auction off water rights, or how regulatory changes could improve competition. However, experimental methodology is not limited to markets, and can be applied to understanding how people make decisions more generally.

Other prominent examples of the clever use of experiments include:

• Vernon Smith’s experiments on trading institutions and asset bubbles;
• Elinor Ostrom’s experiments that show which factors help groups to avoid over-exploitation of a common resource; and
• Al Roth’s tests of the efficiency of different matching mechanisms that can be used to match schools and students, organ donors and recipients, or students and student housing.

All three authors have received Nobel Prizes for their work.

Researchers at AdLab study a broad range of topics related to the strategic behaviour of agents embedded in various social dilemmas, the welfare impact of regulatory policies and taxation, the design and comparison of market institutions, and the effect of psychological biases on economic decisions and market outcomes. Topics of particular interest include:  

1. The role incentives and information on economic decision-making, e.g., the design of information policies that pre-empt collusive behaviour in markets, the organisation of efficient contests (for work promotion or sport events);  
2. The effect of taxation regimes on agents’ behaviour and social welfare, e.g., the design of effective and efficient corporate tax policies;
3. The impact of market structure on behaviour, efficiency and profits, e.g., how should commodities be auctioned off to maximise the seller’s revenues or the efficiency of allocations;
4. The role of heuristics and psychological biases on economic decision-making, strategic behaviour and market performance, e.g., the design of marketing strategies or ‘nudges’.

The Adelaide Laboratory for Experimental Economics was provisionally established in late 2003. The School of Economics was still housed in the Napier building, and the provisional laboratory utilized a standard computer pool on level two. As privacy is important for economics experiments, we had to come up with some kind of removable dividers. After the 2004 general election, long-time lab manager Mickey Chan drove around Adelaide from polling station to polling station and collected discarded cardboard polling booths, which we then modified with gaffer tape and a carpet knife.

The somewhat wacky charm of these provisional dividers stayed with us for some time. In the meantime, other universities in Australia established purpose-built experimental laboratories. When the School relocated to 10 Pulteney street in 2009, it was finally time for AdLab to grow up. The University supported us with a large University initiative grant, and gave us a permanent home with purpose-built furniture in the Nexus 10 building. In April 2010, the new lab was officially opened with a public lecture. In 2014 the University honoured AdLab for its research by inviting us to present a public lecture in the prestigious Research Tuesday series.

We would like to thank all the current and previous members of AdLab and everybody who has helped to make AdLab a success.