Experimental Economics: What Does it Offer?

The research output of experimental economics has been showing a marked increase in the recent past. This post provides a few insights regarding the works of the experimental economists. As Shyam Sunder writes, there are three streams within experimental economics: (1) macro stream to examine the properties of social structures, (2) micro stream to examine the behavior of individuals, and (3) agent stream to explore the links between the micro and macro phenomena using computer simulations.[Sunder 2007] At the first look, experimental economics seems capable of handling the micro stream as well as the agent stream.

The following image shows how laboratory data is presented for a double auction. (Retrieved from http://veconlab.econ.virginia.edu/htm_small/da_06_small.png )
An Illustration

It is believed that Edward Chamberlin conducted the first experiment in economics. Chamberlin examined market conditions in his classroom in a controlled atmosphere, whose results deviated from the prevalent Walrasian conclusions. The rationale of experimental economics is that theory/models suggest general tendencies among the variables of interest. The strength of the general tendency is not provided by theory. In other words, theory cannot tell us about its explanatory power. This is where experimental economics makes its contribution. Also, this is the reason why one carries out field surveys and why econometricians look for a high R-square.

Initially, experimental economics seemed to be aping the sciences to a large extent. However, after reading Sunder’s account, I am begining to get convinced about its potential in understanding aggregate economic behaviour. Sunder wonders how we can study social beings by completely negating free will – that is, by treating it as a science. The other option is to highlight the heterogeneity of human behaviour and refuse to come up with general tendencies. As he writes:

“Hence we see the dilemma of social sciences. Do we abandon free will, personal responsibility, and special human identity; and treat humans like other objects of science? That is, drop the “social” qualifier, and become a plain
vanilla science? Or, do we drop the “science,” abandon the search for universal laws, embrace human free will and unending variation of behavior, and join the humanities? Either way, there will be no social science left. Is there a place where we can keep the “social” and the “science” together?”

Advances in computer technology assists economists working on aggregate economic variables. This is done by conducting computer simulations. The model can be taken from any theory – Marxian, Walrasian, Ricardian, Keynesian, etc – and specific behavioural rules can be assigned so as to arrive at some conclusions. But what is to be remembered is that, these computer simulations are tools which assist the policy maker and are not substitutes for theory. It must be noted that when assigning behavioural restrictions, we are negating the free will of the individual economic agents. As economists, we are used to making such an assumption – utility maximization, bounded rationality, tit-for-tat strategy, satisficing, reciprocative behaviour are all aspects of an individual’s behaviour.

To conclude, experimental economics offers economics the following benefits. Firstly, it provides data for verifying/falsifying a particular hypothesis. Secondly, through computer simulations, aggregate economic behaviour can be replicated to a limited extent. And lastly, experimental economics provides a strong flavour of science to economics.


Sunder, Shyam (2007) ‘Determinants of economic interaction: behavior or structure’, Journal of Economic Interaction and Coordination.