Methodology of Modern Economics

My paper is a survey of the huge amount of solid empirical evidence against the utility maximization hypothesis that is at the core of all microeconomics currently being taught today in Economics textbooks at universities all over the world. It is obviously important, because if what it says is true, the entire field of microeconomics needs to be re-constructed from scratch. Nonetheless, it was summarily rejected by a large number of top journals, before being eventually published by Jack Reardon as: ” The Empirical Evidence Against Neoclassical Utility Theory: A Review of the Literature,”  in International Journal of Pluralism and Economics Education, Vol. 3, No. 4, 2012, pp. 366-414.   Speaking metaphorically, my paper documents the solid evidence that the earth is a round sphere in world where educational institutions teach the widely held belief that the earth is flat. Readers of RWER blog will recall that when challenged on the failure of macroeconomics after the Global Financial Crisis, economists retreated to the position that while macro theory may be in a bad shape, at least Microeconomics is solidly grounded. My paper blows this claim out of the water. As a result, nothing is left of Micro and Micro, and of economics as whole. This supports my earlier claim that a Radical Paradigm Shift is required to make progress — patching up existing theories cannot work.

None of the several leading journals that I sent the paper to made any comments about any mistakes in my arguments. There were two main reasons which were stated for rejections. One was that the paper was “not appropriate” for the journal. This seems ridiculous; how can a paper which challenges the foundations of the subject be “inappropriate” (however, an explanation will be provided later). Two was that the results of the paper were well-known. I also sent the paper for comments to many leading economists. Kenneth Arrow, who was once a teacher of mine at Stanford, responded as follows “Thank you for the very complete and well-argued critique of the utility-maximization theory. Of course, the remaining question is, what should take the place of that theory?”.  This is just to document that the paper itself provides solid and irrefutable evidence against modern microeconomic theory. Given that the vast majority of economists continue to teach microeconomics based on utility maximization, and that this theory is widely believed by students of economics throughout the world, it would seem of great importance to document the empirical evidence against it and use it as a springboard for developing alternative theories, more consistent with empirical observations of human behavior. However, none of the mainstream journals displayed any interest in publishing this paper.

This leads to a puzzle, which requires some thought. As famous theoretical physicist Richard Feynman put it: “It doesn’t matter how beautiful your theory is, it doesn’t matter how smart you are. If it doesn’t agree with the experiment, it’s wrong.”  The experimental result — corresponding to the empirical evidence — is all important in physics, which is why there has been tremendous progress in physics. On the other hand, the economists display no interest in empirical evidence at all. This indifference of economists to empirical evidence which contradicts their theories has been noted by many; see previous post on “Quotes Critical of Economics” for the full quote briefly cited here:

1. Keynes: Economists are unmoved by lack of correspondence between their theories and facts.
2. Solow: DSGE models are so crazy that their founders are like lunatics, and, policy making using these models is designed for some alien planet.
3. Sitglitz: Economists frequently make claims in conflict with easily observable facts, because economics is a religion, not a science.
4. Olivier Blanchard: DSGE models makes assumptions profoundly at odds with what we know about consumers and firms.
5. Paul Romer: Macroeconomic theorists ignore mere facts by feigning an obtuse ignorance.

So there is massive amount of evidence that economists ignore plain facts of observation in conflict with their theories. This is like someone who says that “let us assume it is midnight,” and who finds it irritating and irrelevant if someone points out that the sun is shining. This is exactly what happened to my paper, which provides strong, solid, robust evidence against the core assumption of utility maximization, which is the foundation of modern microeconomic theory. Most major journals desk-rejected the paper, but the Journal of Economic Surveys actually provided a referee report. I cite one paragraph from this report:

This is one example of what I believe to be a theme running through the paper

which is “economists are blinkered idiots who are willfully ignoring the way

in which people behave when they build their models, and therefore say all

sorts of stupid things” This is a view that may (or may not) have been valid 30

years ago, but I don’t think it is today. (Almost) every major economics

department has someone who works on behavioral issues, and almost all

important economic issues have been considered in the light of the

phenomena that the authors discuss (social preferences, temptation and self

control, etc). A look at the February edition of the AER shows an article on

the origins of co-operation, one on the effect of confusion on asset prices, one

of the effect of conflict on social preferences and one on the behavioral

foundations of mircrocredit. It is therefore in my opinion unfair and unhelpful

to label all economists as ignorant (or dismissive) of behavioral issues.

Of course, this is not to say that all economists spend all their time using or

designing behavioral models. Some of them think that the more traditional

models of behavior capture what is important for the question they are

studying. Whether they are correct in this belief or not is an empirical

question, and an extremely important one, but not one this paper does a good

job of dealing with. In fact, I would argue that it presents an extremely one

sided view of the data: while it is true that the factors that the author presents

in this paper are important for behavior, it is also true that the forces

considered by traditional economics are also extremely important. In fact, if

you open any major economics journal at random, you will find articles that

describe how a model that boils economic behavior down to an (obviously

untrue) set of primitives can actually do a surprisingly good job of explaining

many phenomena. A crucial question for behavioral economists is identifying

which behavioral phenomena are quantitatively important for understanding

which questions. It would be great if this survey could try to do this in a more

evenhanded way

 

This referee report is full of contradictions, created because, astonishingly, the author of the report is himself a behavioral economist, and actually likes the paper — early in the report, he writes that: “As a behavioral economist, I think that many of the issues that are raised in the paper are extremely interesting and worthwhile.” Why then does he eventually end up suggesting a major revise and re-submit? Note that the major complaint with the paper is about the “tone”, which is too insulting to economists. The referee does not have any complaints about the substance, which he agrees is extremely interesting and worthwhile. After assuming it is midnight, if the economists are having a conversation about how it is difficult to see things in the dark, it is rude to say “excuse me, but it is mid-day, and there is a bright sun shining above”.

There are many puzzles created by the facts listed above, and pondering them can lead to deep insights, as I will now discuss. First let us consider how it is possible for economists to ignore, or consider as irrelevant, facts which directly contradict their theories. One explanation is that economists are “blithering idiots” as suggested by the referee. Similarly,  Lars Syll has a post on Idiot-Savants, because economists can perform very advanced mathematical operations, but are clueless about their relevance to the economy. One could argue, with Stiglitz, the economics is a religion, and one adheres to the faith, regardless of any empirical evidence to the contrary. Along similar lines,  I have earlier argued that economics is an ideology.

However, having been trained as an economist, I would like to reject all of these simplistic explanations. Closer to the mark is the accusation of “Innocent Fraud” leveled by JK Galbraith. The economics training given at the top departments is highly mathematical, and no one with low IQ would be able to survive in such departments. There is no ideological component to the training. Although we all end up affirming and proselytizing free markets and capitalism, this is not taught as an ideology. It is built into the foundations and frameworks of the mathematics and the modeling style. Thus economists are unaware of any ideological underpinnings to their models, and would strongly deny any ideological bias — hence the “innocence”.

The key question I would like to focus on here is: Is it possible for a rational human being to systematically reject empirical evidence which flies in face of his theories? It was only after thinking about this for a while that i realized that the answer is YES, IT IS POSSIBLE. This can be done if we derive scientific theories based on the axiomatic-deductive methodology which is used in geometry. After we put down the axioms about points and lines — which are intuitive and “self-evident” in some sense — then all further deduction is done on the basis of solid logic which cannot be challenged. We have a closed system. The axioms are CERTAIN, and the logical process leads to CERTAIN conclusions. At no point do we need to consult the facts of observation. In fact, if we are confident enough about our methodology, we can confidently reject facts of observation, and assume that they must be contaminated by some (unknown) errors. It is very important to understand this, so let me explain this further.

Suppose that we follow the standard axiomatic/deductive procedures to prove the Pythagorean Theorem. There is no possibility of error. Suppose someone comes and draws a right angled triangle and measures it sides and shows that this triangle does not satisfy the theorem. Should I, as a mathematician, be interested in this triangle, this apparent exception to my theorem? I would say no, I would not have any interest in this. My theorem is an impeccable exercise in logic, and is beyond the reach of empirics. I can be certain that there is a mistake somewhere — perhaps the right angle is not right, or some other mistake has been made in the measurement, or the surface of the paper is not flat or  a thousand other possibilities. If a parlor magician makes it appear that a ball suspended in the air defies the laws of gravity, I would not start re-examining the foundations of physics. The laws that we derive from axiomatic-deductive method have a universality and certainty that goes beyond the realm of mere facts.

It is exactly this type of certainty that economists have in their methods, which they feel to be beyond the reach of mere facts. Let us examine the statement of Lionel Robbins, who laid down the scarcity-based foundations of modern economics:

“The propositions of economic theory, like all scientific theory, are obviously deductions from a series of postulates. And the chief of these postulates are all assumptions involving in some way simple and indisputable facts of experience…”

We start with simple and indisputable facts of experience, and we build upon them using logical deductions. There is no possibility of error, and never any need to check against facts of observation. The certainty available via the axiomatic-deductive method is far beyond anything achievable using mere facts of experience, which are always fallible.  Of course there is one important catch to this: your assumptions must indeed be beyond the shadow of doubts. If you start with the wrong assumptions, then you are done for. In my previous post on “Quotes Critical of Economics“, there are many quotes which show that this is precisely what happened. For example, Keynes says that “It is an extraordinary example of how, starting with a mistake, a remorseless logician can end up in bedlam.” Similarly, Solow says that if someone (Lucas & Sargent) start with the premise that they are Napoleon Bonaparte, you don’t want to accept these assumptions and continue on to the discussion of cavalry tactics at Austerlitz. Once you accept ridiculous assumptions, the entire carefully and precisely constructed mathematical superstructure will only lead to equally ridiculous conclusions. This is why Krugman said that the profession as a whole went astray because they mistook the beauty of mathematics for truth.

There is a lot more to this than meets the eye. We must appreciate the power, beauty, and elegance, of modern economic theory — it is no mean trick to persuade a generation of intelligent youth that it is midnight, when the sun is shining brightly. We are all witness that economics successfully performs this magic, and it is of utmost importance for us to LEARN how this trick is done. If we can learn this, perhaps we can learn to undo its effects, and remove the blindfolds which prevent the economists from realizing that sun is shining, when their theories proclaim that it is midnight.

One important clue to this mystery is given in the statement of Lionel Robbins above. As he states, if we start with indisputable facts of experience, and build upon them using rigorous logic, then we will undoubtedly arrive at the truth. But what happens if we start from FALSE assumptions and build upon them with with equally rigorous logic. It should be clear that we will get to false conclusions.   The concept of the Axiomatic-Deductive method is the following. We use our experience to arrive at the fundamental certainties, the axioms on which we will build our entire system of knowledge. These must be absolutely certain, since entire structure will be built on these foundations. This is actually a description of the Greek Axiomatic Deductive methodology, which relies on experience only to the extent of using it for the formulation of certain (indisputable) axioms. In my earlier post on “Economists’ Confuse Greek Method for Science“, I have explained how current methodology of economics is like the pre-scientific Greek methodology used for creating Euclidean geometry. Once the axioms have been chosen, this methodology is blind to facts of experience, and rightly so – experience can only lead to lower level temporary truths, which may be invalidated by the next experience. The axiomatic-deductive method leads to grand universal truths, beyond the reach of experience. HOWEVER, this is dramatically different from Scientific methodology which is characterized by the Feynman quote cited earlier. Every theory is tentative, and remains so even if confirmed by facts. Because the next conflicting fact would lead to re-consideration of the theory. An unappetizing conclusion is the science can only lead to uncertain truths, and never to grand universal truths. This is exactly the conclusion of Karl Popper, that one of the key characteristics of scientific truths is falsifiability — the possibility that an empirical experience may invalidate or refute the law.

However, as I have come to realize more recently, it is not so that Economists are using the Greek methodology espoused by Lionel Robbins. They have moved on. In the early post war era, a large amount of research on actual firm behavior led to fairly solid conclusions firmly rejecting the marginal product theories. Followers of scientific methodology would have revised their theories, in line with Feynman’s principle quoted earlier. Even followers of the Greek methodology would have had the option of changing the axioms they used, to produce a better match to the observations. This would correspond to the (Keynesian) realization that we live in a non-Euclidean world, so that the parallel postulate must be dropped, and replaced by a suitable alternative. However, neither of these options was taken by the economists.

Instead, Friedman produced his famous F-twist in the his essays on positive economics, stating that:  “Truly important and significant hypotheses will be found to have “assumptions” that are wildly inaccurate descriptive representations of reality, and, in general, the more significant the theory, the more unrealistic the assumptions.”  Huge rivers of ink have been spent on unraveling the meanings of this message., which I do not intend to re-trace. However, the F-twist had a deadly effect on the methodology of modern economics — see “Friedman’s Methodology: A Stake Through the Heart of Reason“.  As a result of this unwillingness to abandon even false axioms, modern economic methodology can be summarized as follows.

Capsule Description of Current Methodology: The only acceptable models are ones where all decision makers must maximize, and we must look for equilibria that result from this process of maximization by all. Thus, the firmly established behavioral insight that people use heuristics, and cannot maximize in face of uncertainty, cannot be incorporated into mainstream economic theory. Also, the firmly established insight that the behavior of dynamical systems in not governed by the equilibria but by the disequilibrium dynamics, is totally out of scope of economics. To make matters worse, these methodological foundations are axioms of modern economic methodology, and not subject to question.   Since wildly unrealistic assumptions are perfectly fine, indeed desirable, economics cannot be questioned by showing that the assumptions are wildly unrealistic. More surprisingly, if the results/outcomes/predictions of the models are wildly unrealistic, that also is not a relevant consideration. This is because of two reasons. One is that it is the intrinsic mathematical elegance of optimization and equilibria that is prized — whether or not it has an application is only of secondary importance. The second is the theory/practice distinction. The theoretician produces beautiful models. It is the job of the practitioner to see whether or not the model can be applied to any given real world context. If a model fails to apply, that is not a failing of the model; it is up to the APPLIED economist to find a suitable model for use for his real world context. It is not the job of the theoretical economists to assess whether his model produces results which work in the real world. I have not been able to find the quote, but I recall reading that some economists were bewildered when asked about the Global Financial Crisis — they felt that this real world event had no bearing on their work, perhaps due to specialization.  For example, if an economist works on refinement of Nash Equilibria, he will generally have only a dim concept of how his research relates to game theory in general, and even less of how game theory is used by economists within the context of the general project of using theory to solve real world economic problems. Thus specialization has led to a situation where people thinking on their own have thought up astonishing creations, never once checked by reality — neither in formulation of the assumption, nor in terms of matching predictions to real world observations. Neither of these is required by current economic methodology.

The central question remains of what is to be done. A complex response which requires coordination on multiple fronts is required. The first step has to be the development of a coherent alternative. I have developed a two semester micro course which starts with the Anti-Textbook of Hill & Myatt, and follows up by using two main tools OPPOSED to maximization and equilbrium. Maximization is replaced by behavioral heuristics, which allow operation in environments with genuine uncertainty. Maximization is actually IMPOSSIBLE in uncertain environments. If I have to choose between actions a and b but the outcome to me depends on unknown state of natures and unknown acts of other agents then maximization is not possible. I cannot calculate which of the two actions will yield greater utility because the outcome depends on many thing I do not know, and cannot learn.  SIMILARLY, if we equip all agents with heuristic behavior and try to compute what will happen, this can only be done within an Agent Based Model using computer simulations — which will generally lead to disequilibrium outcomes. Using these two tools to replace the central micro tools, one can go quite far in replacing standard micro with sensible alternatives.