Is the Economist’s Account of Rationality a Good One?
Image by Ayana Shirai
As someone who studies business economics and philosophy, I always had a mixture of the two as the domain of my interest. For example, this spring 2023 semester, I went to an upper level metaethics class in the morning just to find me learning about global economic problems from a macroeconomic perspective later, and close my date with a class on markets, morality, and capitalism.
But that was a while back, so let me take you to the evolution of my academic autistic interest. When I was a freshman, I knew I loved philosophy but also wanted to discuss some applied issues. It was an interesting time of reflection to solidify my academic interests.
Then in the sophomore and junior year, I had an opportunity to participate in the philosophy of finance fellowship with Dr. Kane. This workshop allowed me to get excited about grasping rationality better especially in the context of economics. After all, granting the assumption that agents are rational in economic models, we can mathematically go from there to the calculation of general social “welfare”, and make the purportedly empirical science of economics become entangled with specific normative commitments.
After thinking about this issue for a while, I decided to tackle the question, whether the economists’ account of rationality indeed is a good account of (practical rationality), in my thesis. The thesis has three arguments for why economists do not give the right account of rationality. Then, it covers a positive argument for a practical rationality, roughly understood and extrapolated from philosophy of action.
Overall, I say that the economists’ accounts are too broad in one sense and too narrow in another sense, and their being highly context dependent adds more complexity to the picture. I think a philosophical account from action theory does a better job explaining what we go through when we are rational vs. when we are not.
The economic account I used in my thesis is generally the one that can be found in undergraduate macro-economic classes. Even though the standard with which economists judge whether some preferences satisfy a rationality varies, the economic accounts do share a commonality that they build rationality mainly by specifying constraints on preferences that can be counted as rational, so that those agents can be treated as utility maximizing (coming from Neumann and Morgenstein) and can be used.
The conditions on preferences typically are stability, transitivity, completeness, monotonicity, and irreflexivity. Because economists also tend to assume that the preferences are direct mapping from a given object to the evaluator’s internal “push and pull”. In this understanding, the view on preferences is dramatically simplified as those which take the object of action to also be the object of our “preferring”, and such preferences having one dimension; in other words, only the first-order desires/ preferences are the important ones in this account.
Now, let me give a summary of the negative argument against claiming that the rational agents’ preferences need to satisfy these mathematical axioms, and calling the fulfillment of these axioms (condition) makes an act at stake rational. The following are specific “problems” of the economic account, and in the background, I generally assume uncontroversial premises that if an account is overinclusive/ underinclusive/ context-dependent, then they cannot be good accounts.
Firstly, it is overinclusive in a sense that no matter how much regret awaits the agent after the act, the patient can be called rational as far as they exhibit the fulfillment of assumed preference conditions/ axioms. Cases around the weakness of will are good examples of such.
Secondly, it is also underinclusive in other cases, meaning that economists end up calling rational people irrational. Think about a minimalist Buddhist monk. Their values are structured in such a way that they value getting the minimum requirement, and are satisfied with needs being met. In economics’ jargon, their marginal “utility” from the goods or services goes to zero beyond a certain point, and possibly negative. They need to call a Buddhist monk irrational because the monotonicity requirement says just the opposite — that more of a “good thing” is always better than less of it..
Finally, the economic account is highly context dependent in a sense that the preference axioms tend to pick out the nature of the market, and or the nature of rational agents within the market, and they do not pick out the essence of rationality per se. Explaining this takes longer than a typical blog-post length. so I will omit it, but the most important point here is that if we change the context, the rational agents’ actions and how their preferences appear to be probably differs. For example, we tend to deem family and friends both as important but they are incommensurable, or “on a par”.
These three critiques of the economic account naturally lead me to look for alternatives. I used a broadly understood reason-responsiveness based view of rational actions; what it is to be rational is to respond to all-things-considered reasons well. These agents can be understood depending on how they relate to all-things-considered reasons and judgments following from reasons. Here, I am again assuming that if an alternative account (philosophical) can overcome the problem of the original account (economic), then the former is better than the latter.
Philosophers fortunately already have a skill set to explain the weakness of will, by separating out causal/ motivational/ action-prompting reason from all-things-considered/normative reasons. Philosophers generally avoid the second problem, too, because they allow for reasons behind actions that are not tied directly to the object of pursuit; Finally, not only do philosophical accounts avoid the problem of having to weigh families against friends as if they were m&m vs KitKats, but they also can explain why rational or reason-responsive individuals behave in ways that support economic understanding of rationality at least prima facie by showing which kinds of features of the contexts allow rational agents to act as if they satisfied all the preference maxims, and thereby “utility-maximizing” at least in the mathematical sense.
That was a gist of my paper, and I hope that it twisted your brain in a good way to think deeply about these issues. Together with suppressed premises I briefly mentioned above, we should be able to conclude in a logical statement that the economic account of rationality is not really capturing the essence of rationality, and a philosophical alternative may be better. One thing to note is that I am not objecting to the use of mathematical models of utility maximization or demand curves; I am merely pointing out its limitations, and why they should not be taken as rationality in its essence.
If there is one thing I were to suggest to you, it would be to call the economists’ understanding of preference axioms/ conditions not the condition of “rationality”, but that of “economic rationality” or home economicus, following Mill. This should help clarify its limitation, and more clearly separate the all-things-considered rational, which is often taken to be normative, from economics so that they can really focus on their empirical research. Alongside, I think capturing the complexity of rationality by looking at how different reasons relate to one another allows us to have better conversations that go beyond economic efficiency vs other purported social values such as distributive justice.
