We need to look at how economics, as portrayed in political discourse by way of simple narratives (i), is a fairy tale much like religion is. To be fair, the field of economics is a science (ii) – assumptions are stated and predictions are made – but once ideas become repackaged as absolutes for use as rhetoric, as the “invisible hand” (v), then we must put them in their place, right next to religion.
Admittedly there is an aesthetic to economics – in its coherence and logic – and I get why many economists are attracted to it. But after looking at how deductive and assumption-laden neoclassical economics is, especially when presented formally, it is indeed a miracle that homo economicus (iii) survives at all. Here, we will look at how its axioms create the rational man.
Types of Reasoning
It will be helpful to review the different types of reasoning we engage in before going any further. When we make observations from our natural world to form more general conclusions, then we are reasoning inductively, and when we base our conclusions on reasoning alone without observation, then we are reasoning deductively.
In other words, deductive reasoning is to a philosopher as inductive reasoning is to a scientist. In reality, we use both types of reasoning (iv) but a distinction is that deduction can work with claims that are not true but still be valid, as illustrated below.
- All people are rational. [premise]
- Jon is a person. [premise]
- Therefore Jon is rational. [conclusion]
People can be rational, but if we know Jon, who can be irrational, then this conclusion can’t be true. But it is still logically valid – that is, the conclusion follows from the premises. This is how neoclassical economics works because they start with assumptions and then deduce models from it. In fact, its appeal was because it claimed to be “completely axiomatized” akin to mathematics (see comment #2).
Axioms of Economics
Below are the three assumptions that the entire field is built upon, often referred to as axioms. The first assumption says that “we have rational preferences” which means we can assign value to specific items or services. More specifically, rational choice theory says that we pick outcomes that provide us with the greatest benefits and satisfaction given the choices available.
- we have rational preferences
- we maximize our utility
- we all have perfect information
The second assumption is that we “maximize our utility” or self-interest, which equates to rational choice theory defined above. Note that as long as we consistently rank what choices are important, then the goals, which are based on our preferences and desires, can be anything per rational choice theory.
“Perfect information” means everyone has access to the same information on pricing and that we know our utility. In conclusion, we have explained how the model for a rational actor works, known as homo economicus, as it’s just a summary of the axioms. To be succinct, it is a man that is consistently rational, self-interested, and who pursues his subjectively-defined ends optimally [6].
Notes:
i) I am referring to the simplified version used for political discourse purposes, which are a series of myths and narratives; for example, “government intervention is always bad” and the market is “infinitely wise”.
ii) If assumptions are explicit and neoclassical economics attempts to make predictions, then it at least deserves to be in the category of science. The question should always be how well does it do its job and not framed in absolute terms.
iiI) Homo economicus is a model for how man makes decisions regarding his needs and wants. The exact definition is a human agent who is consistently rational, narrowly self-interested, and who pursues their subjectively-defined ends optimally. [6]
iv) Actually, we reason neither deductively nor inductively but through inference. In fact, deduction and induction are human inventions that just so happen to have utility. There is no such thing as “universal logic.”
v) I have argued that the “invisible hand” is more of a rhetorical device to promote neoliberalism policies than it is an economic insight. I don’t want to, however, give the impression that it has no value because economists have given it utility.
It can refer to the increased or created utility for both the buyer and seller after a voluntary exchange of goods or services as well as driving competition amongst firms to meet our needs for lower prices, which results in commoditization [2].
References:
[1] Barrett, Lisa Feldman. How Emotions Are Made: The Secret Life of the Brain. HMH Books.
[2] Frank, Robert H.. The Darwin Economy (p. 27). Princeton University Press.
[3] Kennedy, Gavin. Adam Smith and the Invisible Hand: From Metaphor to Myth. Econ Journal Watch 6(2): 239–263.
[4] Lakoff, George. Moral Politics . University of Chicago Press.
[5] Lakoff, George. Philosophy In The Flesh.
[6] Lakoff, George. The Political Mind. Penguin Publishing Group.
[7] https://en.wikipedia.org/wiki/Rational_choice_theory
Wouldn’t axiom 2 imply axiom 1? At least, I’m having trouble imagining how you could irrationally maximize your utility.
Secondly, there is some ambiguity to “maximize utility”. A traveling salesman considering two routes which cover all his necessary stops would obviously choose the shortest one (this is one interpretation of maximizing utility) or he would have to solve the traveling salesman problem, which theoretically has an optimal solution, but in practice finding it is hard.
I am far from being an expert on rational choice theory and finding optimal solutions. And agree with you that it’s redundant. I think axiom 1 needs to be “rational preferences” instead such that each preference has a weight or value to it allowing for a rational decision, which is axiom 2. Thank you. As far as the optimization problem, I would like to see you explain it because I am just bogged for time and am no expert.
No. Axiom 1 is not implied by Axiom 2.
Axiom 2 states that we pursue our goals rationally, but it doesn’t say anything about how we pick our goals.
Axiom 1 states that we choose which goals to pursue rationally.
A person with no ability to do math at all will have a very hard time becoming an architect. A person with a tremendous artistic ability might dream of becoming an architect, and might choose to attend an architecture school rather than an art school (Axiom 2: pursuing the goal rationally), but when you have no math skills and the artistic talent of Picasso, the rational goal would have been to pursue a career in art and thus go to art school.
If you have axiom 2 but not axiom 1, then that person goes to school to become an Architect (and fails).
If you add axiom 1, that person goes to art school & becomes famous and at least moderately well off.
If you retain axiom 1 but lose axiom 2, then that person pursues a career in art, but might do so by visiting the Australian outback, studying the rock paintings, then attempting to reproduce them using original techniques like chewing dry dirt naturally rich in rare mineral pigments into mud, then spewing it onto the rock. You might learn a lot about ancient art techniques that way, but no one can buy the art that you’ve painted on the sides of huge boulders or deep caves. You have a rational goal, given your talents (to become an exceptionally skilled artist), but you have pursued them in an irrational way. Without economic support from selling your masterpieces of reverse-engineered art, you can’t even continue your desert jaunts, much less secure the material needs (shelter, food, etc.) that you would need to guarantee in order to focus on producing your next piece of art.
Thanks for the comment and clarification with examples. Those axioms were a quick attempt at dumbing down something that I have little familiarity with. It’s even more clear to me, however, if presented as function(preferences, risks) = Optimal choice, where optimal choice is in line with our weighted preferences. No, I don’t know the math notation as it should be, but my point in the article was to express that deductive reasoning can yield results that are inconsistent with reality, which is no surprise, I know, and that neoclassical economics is built upon these axioms. This sets the stage to the next post which is titled Destruction of Homo Economicus and then the finale titled Resurrection of Homo Economicus. I know how original and sounds like an Alien Trilogy. I plan to show that albeit rational choice theory has value in many fields, it is very problematic as a model for how we actually make choices as we are far from rational.
Economics axioms aren’t really comparable to mathematics axioms, though. Mathematical axioms aren’t statements about the world; they’re statements about mathematics. Any use of math to describe the world isn’t strictly mathematical; the question, “Of the mathematical structures and expressions that we have, which will be the most accurate and/or useful in describing the situation at hand?”, is a science question rather than strictly a math question. Sciences shouldn’t be “completely axiometized”, because anything axiomatic describing the real world is a set of assumptions for a model, not a math-style axiom. Physics, for instance, doesn’t have “axioms” in this sense, just working assumptions for a particular model (and those working assumptions are always open for revision depending on the descriptive needs of the situation). If economics is “completely axiometized”, that makes it less a science, not more.
I agree with you, and you highlight what I will hit on for the next post. I was perplexed to learn myself that neoclassical economics was perceived to be more “scientific” as a consequence of being “axiomatized”. Also a lot of people are leery of relying too heavily on math to make predictions in general, but Milton Friedman quips, in attempt to defend what some see as silly assumptions, by saying “Theories should be judged by their ability to predict events rather than by the realism of their assumptions.” My personal experiences with math are that it is very useful at modeling physical phenomena, that which can easily be verified through physical measurement, but I am far from intimate with the models used to describe human behavior – say in political science and evolutionary theory – that are laden with math.
“As far as the optimization problem, I would like to see you explain it because I am just bogged for time and am no expert.”
Consider a salesman who has to visit three cities, A, B and C. He wants to find a route that minimises his travel time (or distance, or fuel cost, minimising this maximises his utility.) Now there are 6 possible routes he can take: namely ABC, ACB, BAC, BCA, CAB or CBA so it is fairly easy to find which route is optimal, you calculate the travel time of each and choose the one which is minimal.
However, when the salesman has to find a route which visits 20 cities, things get more complicated. Now there are 20 factorial = 2432902008176640000 different routes. Even with a very fast computer checking a billion routes per second, it’ll take about 1800 years before you find the optimal route.
The solution is that you don’t ‘maximise’ your utility, but choose a route that is ‘good enough’. There are many heuristic algorithms which will not find the optimal route, but will find a route that is nearly optimal. For example, you could always head towards the nearest city you haven’t visited yet. This is not optimal, sometimes a small early detour prevents a later long detour, but in many cases it works well enough, or gives a starting point from which you can look for better routes (for example: try small random modifications of this route by switching around two stops, if the travel time decreases keep the new route and repeat.)
The point I’m trying to make is that maximising your utility is not always practical. (Unless you include the computational cost of optimising utility into your definition of utility… which becomes circular…)
And if you think this is a theoretical exercise, and that such large problems rarely turn up in practice… A few years ago I helped with some route optimisation for garbage trucks in a large city. Every day, a fleet of garbage trucks has to go by 1000s of garbage containers. Shaving a few % off fuel costs quickly turns into of millions saved per year. Problem is, you’re never done… since it’s practically impossible to find the best route, if you spend enough time you can always find a better route.