Behavioral Economics research – and later, Social Preferences research – started with lab experiments in the 1970s, where experimental economists found that their subjects’ behavior systematically deviated from self-interest behavior when making economic decisions.
In the following decades, these findings inspired various economic models to characterize investor’s social or environmental concerns (reciprocity, fairness, empathy, kin selection, cooperation…) based on psychological, biological, or economic models.
More recently, studies have focused on the applications of social preference theories in the business world, including sustainable investments.
Utility Function and Social Preferences
Experimental studies have shown that traditional economic theories have limited predictability because they fail to incorporate human cognitive biases and social preferences.
These traditional theories, as developed in the 1940s, often model investor’s preferences through a utility function. This function transforms the investor’s objective payoffs into the subjective investor’s perception of these payoffs. Traditionally, only the notion of risk is considered as a parameter for this utility function. In other words, the only element that determines the choice of an investor between Investment A and Investment B is their level of risk/payoff and the investor’s risk aversion.
Gary Charness and Matthew Rabin were amongst the first behavioral economists to incorporate social preferences into this traditional utility function model.
They developed a model where the investor’s utility is a combination of their material payoff and social welfare.
Other behavioral economists have later enriched this promising approach by offering various utility function models incorporating social preferences.
Parameters of this social preference utility function
In order to determine the parameters of this social preference utility function, investors are usually asked to choose a series of investments (or distributions) out of several options where there is a tradeoff between their own material benefit and the benefit to particular environmental or social issues.
Studies have proved that this approach gives more accurate results than asking directly for the willingness of investors to pay in absolute terms since responses can not easily be evaluated in terms of decision quality.
An example of such an approach to measure social preferences is the Silver Measure developed by Murphy.
A solution integrating this methodology
In line with this methodology, and in order to capture social preferences effectively, the InvestProfiler has developed an adaptive (each question depends on the answer to the previous question) gamified preference-based questionnaire of binary investment choices that measures the investor’s ESG preferences through their utility function.
Our model estimates, in a short number of questions the parameters of an isoelastic utility model. Our model has as many parameters as the number of candidate values that the investor may be interested in. Below is an example of such utility function, where two illustrative topics are presented : water preservations and CO2 reduction. With this model, we can see that the investor pays much more attention to water preservation, as an increase in the score of this topic increases the utility of the agent much more than the same increase on the score of the other topic.
As an indicator of the performance of our model, with 5 questions and 4 different topics proposed to the client, we achieve an accuracy of 94% when recovering the estimated parameters of artificial agents, compared to their true preferences.
Conclusion
Neuroprofiler InvestProfiler methodology is based on the latest advances in behavioral economics, and more specifically in social preference theories.
Its objective is to measure, through an adaptive preference-based questionnaire, the investor’s social preference utility function, in order to rank the ESG impacts the investor would like to have in his investments.