Economics and Ergodicity

 

Economics posits human behaviour and thus human actions on the premise of utility maximization. Therefore, the next step obviously was in calculating the expected utility. The expected utility theory posits the economic agent being rational will evaluate all possible outcomes of his or her action and the probability of their occurrence before determining their action. In other words, the probability of the expected outcome will determine the trajectory of behaviour on the part of the economic agent. If the agent feels the probability is skewed towards a positive payoff, they will go ahead with the action else they are likely to adopt a risk averse attitude and go with the contrary action. Yet in real life, experiments have indicated something to the contrary. While mathematically the actions do point towards a probable positive outcome, the agents however shirk from undertaking those actions. These have been explained through models like prospect theory in behavioural economics. While the conventional economics theory would posit these actions as irrational and thus going against the grain of economics, behavioural economics comes to the rescue in the analysis of how agents view differently the expected the positive outcomes and the expected negative outcomes. The concepts of framing and anchoring go in some ways trying to explain the puzzles of preference reversals or ownership effects or failure of transitivity in practice.

 

Economics apparently assumes ergodicity in decision making. Yet as Nicholas Nassem Taleb has pointed out the process are rarely ergodic. In other words, history does not have a bearing on the the current trajectory of the outcomes expected from economic agent’s actions. As Taleb would point out, lessons on how the past pandemics behaved would hardly give any insights on how the current Chinese pandemic has been behaving. In absence of ergodicity, the evaluation of actions have to be undertaken differently if the outcomes could result in prospective ruin events. For instance the current pandemic resulted out of the Chinese virus is one such ruin event which if happens frequently might destroy human life. Yet, economists tend to dismiss these concerns stating ergodicity does not factor in economic decision making.

 

Meanwhile, there is an interesting article in Bloomberg quoting an UK theoretical physicist Ole Peters that everything one has learnt about modern economic theory is wrong. The article is available here. The sacrilege in the current context happens to be that Ole Peters is not an economist but a physicist. This perhaps is sufficient for economists to dismiss him. Many journals have rejected his paper on grounds of the readers not being interested in the same. Yet Ole Peters raises certain points which cannot be dismissed. He might be a physicist with little training in economics, but his criticism does merit certain discussion.

 

The article posits that if Peters is proved right, it potentially upsets the entire fulcrum around which the modern economic thought has revolved around. As Peters suggests, modern economic thought centres on the presence of ergodicity which he claims is not present in real life. This thesis is something that finds support from Taleb who was one of the first to build his case for managing non-ergodic processes. In Peters’ analysis, the economic thought has to find itself reshaped having its impact on areas like risk management to income inequality to pandemic management. He expresses a feeling that something is not right in the way it should be. He argues the economic thoughts and concepts are based on flawed reasoning and thus need to be viewed with scepticism. However, he does concede in the article that his ideas have not found much traction among the economics fraternity. While many economists might dismiss him, his paper on ‘The Ergodicity Problem in Economics’ has attracted quite a bit of attention. In fact, a virtual conference planned on his works too has attracted significant number of registrations. In his view, expected utility is calculated as average of all possible outcomes of a given event. Therefore, mathematically, while it might convey some direction, the agents in real life behave differently when they have to place their gambles. In other words, the action is different when one does not have a skin in the game and when one does have a skin in the game. Yet these arguments have found solutions and economics has addressed it in the past. Yet, there does exist some merit in his argument on expected utility theory. It is important to note that while expected utility theory assumes absence of information asymmetry, there do exists costs of information processing and cognition. The cognitive constraints do manifest in decision making something behavioural theories have factored. What is warranted is a reconciliation of behavioural and classical and neo-classical schools of thoughts into arriving at insights on decision making. Yet as Peters thesis suggests, it is far from likely.

 

The methodology employed by Peters without doubt is highly mathematical something expected given his training. At the same time, a group of neuroscientists in Copenhagen are experimenting in their labs on subjects testing their attitude towards risk taking. The subjects did change their behaviour and thus actions willing to take higher risk when circumstances changed. Interestingly, the experiments seemed to validate the thought process and equations put forth by Peters. Yet, despite the apparent success in experimental economics, his ideas still do not have takers in mainstream economics. Apparently, it is because of his outsider image and also the fact that economics with certain adaptations can explain the anomalies that get highlighted by researchers like Peters. Economics has always assumed the presence of constraints in decision making. The objective of neo-classical school is to project utility maximization in presence of income constraints and projecting profit maximization in presence of cost constraints. Yet these constraints might not be the only one. When one factors in the other constraints like cognitive thinking or others, the economic actions of the agents under study could be explained. The gap between the experimental economists and the mainstream utility theory proponents arises in the failure to take into the additional constraints that exist. Behavioural economics gives us the framework to handle these non-income or non-cost constraints. Interestingly however, there has been a thriller made around the ideas of Peters. While the actions posited by economists might differ in practice, the solution lies in working out the range of constraints around which the utility maximisation arises rather than throwing the baby out with the bathwater.