Rationality and system thinking

When you tackle a maximization problem, you may need to consider an observer and his objective. System thinking tells you that you also need to define a system for the stakeholder. When we think this together with rationality, things become very interesting.

We know that humans are not rational agents. We do not use one measurement and freely substitute one thing for another. Due to limited information processing capabilities, we process information with heuristics which enables speedy and adaptive decisions. In other words, we sacrifice some immediate gain for other benefits such as cognitive speed and flexibility. Decision made by using heuristics are considered non-rational. However, I want to argue that in a system of long-term real life, it can be rational. Because we live in an uncertain and unstable environment, it is very important and beneficial that we can make quick response. Moreover, in the long run, only if we make flexible decision, that we can adapt to this unpredictable world. That is, we may maximize our long-term gains with heuristics although it seems irrational that some immediate benefits are forgone. Therefore, when we talk about rationality, it is wise to think about a well-defined system.

Here is another example about rationality and system thinking. I extract an example from Hardin’s article The Tragedy of the Commons and display it in the context of system thinking.

Picture a pasture open to all. Assume there are 1,000 herdsmen living in the day when the long-desired goal of social stability becomes a reality. The carrying capacity of the land is 10,000 cattle. Each herdsman also has a capacity of 100 cattle because of the limit of his energy. The utility of one herdsman who keeps Q cattle can be described as a function: U = Q – 1/1000 Q. This utility has one negative and one positive component. The positive component refers to the gain received by the herdsman from sale of animals. The negative component represents the effect of overgrazing shared by all the herdsmen. Now we can focus on each herdsman. How could each of them maximize his own utility? This can be translated into a mathematical problem. For each herdsman as a rational being, we can identify the objective function, decision variables, and constraints as following:

Objective function: max U = Q – 1/1000 Q

Decision variable: Q

Constraints: 0 (less than or equal to) Q (less than or equal to) 100

The solution is that Q = 100.

From above, we can see that for each herdsman, if he wants to maximize his utility, he will keep 100 cattle. Everyone on this pasture wants to maximize their own utility.  As a result, each herdsman owns 100 cattle and in total there will be 100 *1000 = 100,000 animals relying on the pasture which has the carrying capacity of just 10,000. The pasture ecosystem would collapse rapidly. At this point, we can finally understand the inherent logic of commons remorselessly generates tragedy. 

The problem here is that everyone just focuses on a very limited system, in this case, themselves. If they consider a bigger system namely the pasture, they may make a very different decision. Therefore, if we want to change human behaviors to more sustainable ones, we can always think about how to describe the system for them so that they can make relatively favorable decisions.