The poverty of rationality

Note that even the very aggressive definition of rationality as per Arrow-Debreu and the construct of general equilibrium often cannot actually derive the strong claims that macroeconomists want to draw. Agent optimization is simply not enough to make interesting claims about the shape of excess demand functions.

Thus there is very little space between “vacuous” and “false”. Ultimately, Keynesian models and their intellectual descendants rely not on derivation from axiomatic assumptions but on the plausibility of their stylized narratives; their claim is of getting the relationship between aggregates right, not of getting all the micro in order. In other words, drop the illusion of axiomatic superiority and simply make a claim about the excess demand function of interest itself.

I find the “two-step of terrific triviality” to be a much better way to describe the problem here than “falsification.” The former is a clear criticism that anyone, regardless of their approach to science, could agree on. The latter is a now largely rejected approach within the philosophy of science that never was a very apt description of what scientists actually did.

That a point is made as a factual claim about the world and yet cannot ever be rejected as a result of evidence makes it a bad factual claim. To the extent that rationality is not a claim about the world, but rather an assumption that allows for theorizing about the world, it cannot really be true or false, but rather useful or not. In that case, we reject it because models built on that assumption have not proved useful.

Shorter shah8:

Derp, derp, derp, derp.

Williamson is engaged in something very sneaky indeed: reappropriating the term “rational” to mean whatever the outcome is, in spite of whether or not empirical evidence suggests optimisation.

This is a tautology equivalent to saying “humans are rational actors because they are rational actors.” If the definition of rational actors is shown not to fit the facts the response is “yes but humans are rational actors.”

This is quite an insidious example of the confirmation bias demonstrated by Kahneman and Tvsersky.

As Quiggin rightly points out, Williamson is oscillating between two entirely unrelated definitions of rationality, one of which is false, the other vacuous. Williamson wants to believe the one that is actually false (which is the one used by say, Becker ’76), and when challenged on it veracity, shifts to the vacuous one.

SW:
“If the phenomenon can be described, and we can find some regularity in it, then it can also be described as the outcome of rational behavior”

This would suggest that macroeconomists like Williamson actually observe carefully the behaviour of people and then try and interpret it in a Rational Choice framework. But it is manifestly not so. The assumptions about Utility functions and information sets and optimization contexts made in DSGE models have little to do with actually observed behaviour of people . On the contrary, they are ad-hoc and lacking in empirical discipline.

And I second what David says in the first comment and also Robert Vinneau at 8.20 AM (#5).

I thought the notion that people maximise anything had been falsified by behavioural economics.

I expect Williamson’s commenters to storm in on this one. He seems to have amassed an internet army that rivals that of the Mises Institute’s.

Williamson is oscillating between two entirely unrelated definitions of rationality, one of which is false, the other vacuous.

The oscillating economist is a frustrating being but pointing to the oscillating economist is not enough to debunk rational choice theories. Both the oscillator and the pointer seem to think that the models of preference must be simplistic-self-interest or arbitrary, but there are whole disciplines out there that spend their time studying what makes people tick. It’s just that many economists don’t seem to want to take them seriously.

The one I am most familiar with is Akerlof and Kranton’s use of identity – it takes the psychological and social studies of identity seriously (ie, they don’t just make stuff up about motivations) and constructs rational choice models based on it. It’s neither simplistic nor arbitrary.

In his review, Stephen Williamson, within section 3, devoted to the Efficient Markets Hypothesis, gives a narrative analysis of the origins of the housing bubble. He gives it as an example of the usefulness of “rationality” in building an explanation, and as an illustration of how the insights into arbrtrage pricing associated with the EMH, might apply in analyzing such a case.

What do we know that can make sense of the “bubble” in the price of real estate in the US, say over the period 2000-2006? First we need a definition for “bubble.” To get at this, define an asset’s fundamental price as what standard arbitrage pricing would deliver, i.e. the fundamental price is the expected discounted valued of the future payoffs associated with the asset, . . . Then, a bubble exists if the actual price of the asset exceeds its fundamental price. . . . where would a real estate bubble come from? MBS [mortgage-backed securities] were used extensively in financial market asset trading prior to the financial crisis, and . . . there was an MBS bubble, and this fed back into the market in mortgage originations. . . . and prices of real estate shot up. Of course, the underlying mortgages, particularly in the subprime market, did not generate the promised payoffs, there were a series of fatal incentive problems in the chain of financial transactions that created some MBS, and the MBS ultimately were no longer perceived as safe, liquid assets. A huge quantity of financial exchange went away, and the bubble “burst,” so to speak. Now, nowhere in my story did I invoke irrationality, nor did I violate any basic principle of arbitrage pricing, though certainly there are frictions that play a key role in the story. . . . As well, the basic financial market incentive problems associated with the financial crisis were moral hazard and adverse selection, which are the key private information frictions that economists know a lot about.

I have to say I like this use of “rationality” and the application of basic insights from economic reasoning. I do not find it vacuous or trivial. And, I am inclined to regard it as superior to an unfounded assertion that an inexplicable mania overtook housing markets, or the idea that it is all just so darn complex that no one could understand it, or should be blamed.
I see shortcomings in Williamson’s explanation of how problems in the market for MBS drove problems in mortgage origination; I think the feedback loop ran in both directions — corruption problems in mortgage origination drove problems in MBS, with Fannie and Freddie, and the related derivatives markets. I’d make more out of the importance of administrative control (a la Tanta of Calculated Risk) and be more explicit about control fraud (a la Professor Black). But, perhaps these are just more of the “frictions” Williamson waves his hand at.
It is not lost on me, that Williamson is wryly citing himself as the exemplar of the Neoclassical-Economist-of-Great-Understanding, rather than undertaking the impossible task of defending the recent public diagnostics of Lucas, Barro, Prescott, Fama or Cochrane. It might not be Popper’s idea of a test, but the inability of these gentlemen to interpret recent events intelligently or even to make proportionate claims regarding causality (e.g., reliance on just-so stories featuring “the confidence fairy”), goes a long way to convincing me that Quiggin is right about Zombie ideas rotting the brains of academic economics.
But, I’m left wondering if “rationality” is being asked to do too much work here, to stand in for a religious devotion to a deductive system of thought, immune to facts and evidence, in which all the truly hard problems are dismissed as “frictions” or simply ignored.

Let’s look at it from another angle: “does it sell nylons”, to paraphrase the famouse “Mad Man” quote?

And no, the neo-classical rationality assumption literally doesn’t sell nylons. There is a whole science about consumer behavior (I suggest as a textbook: Olson and Paul Peter, ‘Consumer behavior and marketing strategy’) which, unlike neo-classical micro economics, actually investigates what real people buying real stuff do in the real world do. And businesses al over the globe do use this science, and don’t use the rational choice model, when making decisions. Indeed, the neo-classical rational choice models do not sell, while the science of consumer behavior (sometimes) does. Welcome to the real world.

Williamson only shows his ignorance when he states that the ‘rational choice’ approach is the dominant approach in economics. It isn’t. One can look at accounting, or economic history, or organisational science or whatever kinds of economics there are – not much wel behaved utility functions overthere. Rational choice is a very limited branch of economics. And let nobody teaching neo-classical micro economics teach this ever again without having read a textbook on the actual science of consumer behavior.

I think that’s another example of the oscillation in definitions, Cahal: the strong (A) “farsighted, and purely egoistic, agents maximizing the expected utility of stochastic consumption streams over time” holds up poorly experimentally. But the weak (B) “Any consistent pattern of choice among objects (of any kind) that we can observe, can be represented as optimization, that is, as the maximization of a function” point is quite right. And observed behavior *is* pretty regular and patterned … including ill-advised behavior like overinvestment in housing. (Though it still does not follow that the individual is always the relevant unit for theorizing.)

The mild irony here is that the Keynesian/Kaleckian tradition has tended to work from highly regular, patterned behavior across large groups, and been criticized by orthodoxy on grounds that A-type models don’t generate that.

Conflating A and B might be the key orthodox shibboleth.

I find the “two-step of terrific triviality” to be a much better way to describe the problem here than “falsification.” The former is a clear criticism that anyone, regardless of their approach to science, could agree on. The latter is a now largely rejected approach within the philosophy of science that never was a very apt description of what scientists actually did.

I’ll point out yet again that it seems that most scientists believe they’re doing the falsification thing. If some philosophy guys want to say that what these scientists are doing isn’t “really” falsification, they’re perfectly free to do so. But if they want to insist that one explanation takes precedence over another, well, I’m going to go with what the scientists say.

the strong (A) “farsighted, and purely egoistic, agents maximizing the expected utility of stochastic consumption streams over time” holds up poorly experimentally. But the weak (B) “Any consistent pattern of choice among objects (of any kind) that we can observe, can be represented as optimization, that is, as the maximization of a function” point is quite right. … Conflating A and B might be the key orthodox shibboleth.

Yes. This, and what David said at 1.

There’s something fundamentally dishonest in the microfoundations argument which is so central to mainstream macro. (Pick up any graduate macro textbook, and the first chapter will have some long section on why the models in this book are better than the old Keynesian one because they are properly microfounded.) It’s a bait and switch where they insist that your aggregate relationships have to be derived from individual maximization, but then actually use a representative agent model even tho the behavior of the representative agent is never derived from optimization by heterogenuous individuals and there is no reason to believe that it could be. So their models are really no more microfounded than ours are.

The difference is that Keynesians say, “In the real world, the behavior of aggregates can’t be derived from the behavior of individuals, so let’s look for stable relationships we can observe at the aggregate level.” Whereas the Williamsons who dominate the profession say, “In the real world, the behavior of aggregates can’t be derived from the behavior of individuals, so let’s create a toy world where individual behavior is so radically simplified that we can derive the behavior of aggregates from it.”

And they do dominate the profession. This, I think, is what John Q. doesn’t like to acknowledge, and what his book, for all the good stuff in it, is basically an exercise in avoiding. He wants to marginalize the Williamson approach as just a few zombies, when really it’s right at the heart of mainstream neoclassical economics.

@23, if what people say they do doesn’t match what they do (when you substitute our definitions for theirs and don’t otherwise alter their explanation), it’s not clear why we ought to take it as a useful explanation of what they do. It’s not just philosophers, who have argued that the idea of falsification is incoherent (although they are well positioned to make that claim), it’s also historians and sociologists, who say it is not what scientists do.

There. Fixed that for ya. And since you seem to not be aware of what scientists actually do or how they operate, how about this latest bit about an experimental result that apparently falsifies a key implication[1] of the Special Theory of Relativity, namely that the speed of light is an upper limit.

[1]For various values of implication.

but how do you communicate that lack of interest to anyone but yourself?

Come on, show me guys, show me you ain’t nuttin’ but dino-brains insecure about tiny wees…

ScentOfViolets – if that claim about neutrino velocities is supposed to falsify relativity, then you seem to be working with a different notion of falsification than philosophers do. There are a large number of theoretical moves needed to get from the observation to the theoretical claim that some object moved faster than the speed of light, and it’s only that theoretical claim that would falsify relativity. The observation by itself falsifies only the conjunction of relativity with a lot of other theoretical material. Of course, you yourself seem to point out (in your addition to the blockquote) that you mean something different by “falsification” than philosophers do, but it would be useful to know what you do mean.

The claim that philosophers make is just that falsification as they talk about it can’t (and doesn’t, according to historians) play the role that scientists often claim. Since physicists seem to use the same words in describing falsification, philosophers naturally assume that they mean the same thing, but the fact that they point to instances like this one suggests that they mean something else that they’ve never really described.

The claim that philosophers make is just that falsification as they talk about it can’t (and doesn’t, according to historians) play the role that scientists often claim. Since physicists seem to use the same words in describing falsification, philosophers naturally assume that they mean the same thing, but the fact that they point to instances like this one suggests that they mean something else that they’ve never really described.

Sigh. Rather than go into a pointless digression of what going ftl means for SR, I’ll simplify and say that this would seem to mean that notion that nothing can go faster than light has falsified.

And really, what you mean to say is that what physicists mean by falsification is what pretty much everyone else means (I wouldn’t bother to point this out except for the fact that this seems like a rather obvious rhetorical ploy on your part).

But I think you’ve acknowledged that. Now – as I said in my very first post – what this means is that some philosophers have a different terminology using the same words. Which they are perfectly free to do. But when it comes to the usage of those words, I’m going to go with what the scientist guys say, and not you.

I’m sorry if that ruffles your feathers, but that’s just the way it is.

shah8, Williamson wrote that essay as a review for the Journal of Economic Literature so he’s entitled to assume his readers know some formal economics and can chase down papers; in any case many of them he refers to are classics that would be known to most economists with an interest in his subjects. Mind you, judging by his blog and having seen him trying to do a presentation in person I don’t think popularisation could ever be his forte anyway. Williamson may be (indeed, I think he is) wrong, but I note that he is not silly enough to defend something he hasn’t even bothered to read.

I’d criticise him far more for his inability to quality-assess positions outside his own speciality (as I am not a monetary economist I can’t speak of those WITHIN his speciality). He lumps Lucas’ quite profound critique of astructural modelling – a profundity that extends way beyond macroeconomics, BTW – and hence of policy based on it,*** alongside the empirically rubbish posturings of Prescott on European labour supply. Which shows that he is far too prone to accept the conventional wisdom of the freshwater circles he moves in.

It’s not that DSGE is so bad in principle – just that the modelling choices usually made in that framework tend to be made by people with – shall we say – less concern that their models is tested against observable fact than with ensuring their model points in an ideologically congenial direction (BTW John, I do think Williamson scores a hit when he points out that DSGE can be used to yield impeccably “lefty” conclusions, as well as the RBC rubbish).

*** Yes, the Lucas critique does set limits to macreconomic (and a great deal of other) policy, but that doesn’t justify the other extreme; “Limits to policy” is not the same as “policy never works”.

Anti-falsificationism is a core cultural belief of analytic philosophy. I think they must take a blood oath to abjure Popper and all his works somewhere in the third year. They hold the concept to a higher standard than any other philosophical notion in order to dismiss it

I find this one of life’s little mysteries, since falsificationism is one of the few philosophical ideas of the last century to influence actual practice. Scientists would like to say “thank you, philosophers of science,” but they never get that “You’re welcome” in response. Did Popper sleep with Quine’s wife, or something?

The “two-step of terrific triviality” has a name – the shell game.

i think i agree with you on the problem of rationality and behaviour, but i have some problems about other related issues. the first is about vernon smith and the so called “ecological rationality”. i’m just now reading his “rationality in economics” where he makes the disctintion between two forms of rationality: the constructivist one, where we find the traditional way to understand it in economics, and the ecological one, where the agent is not necessary a human being, but a agent that develop a heuristic that is adapted to the structure of an environment. The concept have its origin on Hayek and it is also developed by guys like gigerenzer. I was not aware about the cantor theorem but in this kind of rationality –ecological- i believe too that the problem is just the one pointed by you, that ever can be a context in reference to which rationality can be defended ex post.
a second point, the ticklish problem to be defined, from my point of view, is that of normativity. If rationality is relative what about normativity? For V. Smith the final word must to be on freedom of choice and success institutional results. I don’t like to apply the label “rationality” to what is on stage when market rules; but the concept of ecological rationality have interesting points: in it rationality in some way is defined by some kind of “juncture” (sorry, i have problems with my English), we can think on facebook, its “rationality” can be defined by its success, and it is strictly relative to it. But we can think too, in the pharma industry and prozac, and in its success… and on malaria, and so on…
to conclude i find that market is a key point for a productive society –with material copiousness – but i think that there is a complex and problematic and ticklish relationship between commodities, welfare and labour. Maybe one of the more important tasks today for economics is the problematization of those relationships within normativity.

Walt #37 has it exactly right. My son has just finished 2nd year undergraduate philosophy and came back completely flummoxed by this.

Kant can say all kinds of bizarre things and get treated with reverence and respect, as can any number of other historical figures (I can’t bring the list to mind myself, but he could). Meanwhile Popper, who did something that actually makes sense to my son, is the only one beyond the pale, and must be spoken of only in dismissive tones.

It really is very strange.

Bertrand Russell once pointed out the propensity of academic philosophers to pass over in silence the obviously silly things that famous philosophers say at one time or another. Kant is far from alone in regularly mixing profundities with risible absurdities that a six year old could see through – maybe it comes with the territory of being an original thinker.

Popper may be wrong or right in the end, but never in my reading of him did I come across anything that was OBVIOUSLY reality-challenged. Perhaps it is his air of common sense that makes him unfashionable – not esoteric enough for the priesthood of academic philosophers.

Some comments:

1. What he says about Black-Scholes (top of page 6) is wrong. He confuses the “real world measure”, which is where data come from, with the “risk neutral measure” that is used for pricing. Observed stock prices are not supposed to be martingales — there would be no point in buying them.

2. One way to guess whether someone is data driven is to look for data in their writing. Williamson has about 23 pages of text and the closest we get to actual numbers is some ballparking on page 16.

3. More evidence Williamson is not data driven. He attributes motives and opinions to Quiggin he has no evidence for, starting at the beginning: “What does John Quiggin want? He wants …”. I wish I had a shrink that good.

Business—like other, similar and important subcultures—has a kind of folk history of itself: a romanticized mythology, say, of how it is structured and how it functions. In some respects this mythology is accurate; in others, very inaccurate. It’s not entirely accurate because accuracy is not its primary function—rather, these kinds of mythologies function as aspirational ideals, as public-relations, and much else besides.

Furthermore, it’s not entirely accurate because being a businessperson is no more an adequate preperation alone for competency at comprehending the structure and functions of business than is being a native speaker of a language at comprehending linguistics.

Say what you will about the (in)dequate development of economics as a science, but from its infancy it demonstrated that many of the stories that business tells about itself are untrue; economics showed that the intuition of businesspeople about what they do and how they do it is often mistaken. Despite this, even today business often jealously guards its notions of itself against the data and analysis presented by economists.

In contrast, scientists, having been steeped for much of their lives in the notions of rigor, objectivity, intellectual specialization, and with a minimal distrust of commonsense intuition, have reacted with an admirable welcoming openmindedness and self-criticism to the advent of academic disciplines which have science itself as their object of study.

You really should use google wrt “selfish gene”.

http://www.science20.com/gadfly/extended_phenotype_how_richard_dawkins_got_it_wrong_twice

http://www.science20.com/gerhard_adam/problem_selfish_gene_theory

Okay, with the basic issue covered (and seriously, I’m not going to argue with anyone about the science involved. There are anti-“selfish gene” Orbital Death Laser Platform articles out there.), the relevance to the discussion about rationality derives from what I think are common tactics by guys like Williamson and Dawkins (and religious officials) which is the imputation (and yes, tie this to the original christian theological concept) of virture (or not) and the agency it implies onto an emergent property of complex systems. If pushed on it, of course, they’ll concede a cigar is just a cigar, but they behave this way because they profit off of people who assume wrongly that something has agency and can be motivated and/or deterred.

People are much more intimidated by forces that must propitiated (or they seek out theories that evade public personal responsibility), and both good and bad theories tend to be used this way as people negotiate power among their relations. Theories in certain fields, like anything to do with money (tax shelters), or evolution (evo-psych!), among others are rife with agencies that switch around depending on the needs of the scammer.

From my perspective, the issue here is that Quiggin talks (or intended to talk about the abuse) about micro-economics not intended for real world use, still being used for prescriptive economics media for real world (especially macro-E) problems. However, the section in the paper which the blog post referred to is basically gone turtle, yeah that’s a cigar, mode. I didn’t see anything particularly wrong with it, and had I been defending my paper/book, I would have focused much more on intent and intent to deceive, and got out the badger kit. Then I would have poked and prodded with the long “why is this bog-standard explanation appropriate” spear. Gotta realize, Williamson doesn’t really attack Quiggin so much as say that Quiggin is incompetent and rehash theory. That is a classic defensive move that inspired people to make up that ridiculous “The Courtesian’s Reply”. Don’t put up with that, bum rush the argument back into real world implications. Don’t get caught up in the Cantor, or you’ll wind up talking Gods and Devils like the old philosopher…

“By contrast, I normally use‘rational’ to refer to the kind of behavior found in the simplest form of the DSGE models: farsighted, and purely egoistic, agents maximizing the expected utility of stochastic consumption streams over time. Most of the time, at least when no-one is challenging them on it, this is the way neoclassical economists use the term themselves.”

I think a few commenters may have been alluding to what makes me feel left out of conversations like this. Not the endless discussion of things like “falsification” (which I can indeed follow, though they just bore me — reminds me of proofing my son’s analytic philosophy essays) or “rationality” (boy, what a dead-end; let’s not confuse either reason or logic with “rationality”). Rather, the premise of “purely egoistic agents” (I’m assuming the comma between “purely” and “agents” is a typo.) Do such things exist outside the realm of grammar, the first person singular? Where does “I” begin and end? (In particular, what is it’s relationship to “thou?”)
Physical and biological scientists seem to me more sophisticated about such problems of definition than economists, in my experience.

Kenny 50:

“Instead what we need is a notion of what it would take to reject a theory. There may be no observations that would strictly falsify a theory, but if there are situations that would make us bend over backwards too far to save it, then they may count as ones in which we would reject it.”

It seems to me that this notion of “falsification” is entirely consistent with Quiggin’s use of the term. (And probably also with SoV’s and Walt’s). Would you agree with that?

Have I shared Bruce’s Thumbnail Epistemologyâ„¢? It might help in sorting these things. BTHâ„¢ asserts that what could be, what is and what should be belong to three separate domains of knowledge, each with its own criteria of evaluation for “truth”, and there are no bridges connecting them. Speculative a priori analysis can be logically valid, without that validity entailing any synthetic, a posteriori assertion about what is. Observation needs logical reasoning to make useful inference about what is, but statements about what is, don’t carry the force of logical validity; generalizations seem to correspond to concepts, but generalizations, by their nature, can have exceptions, while a valid concepts, deriving logically from a definition, cannot. By extension, assertions about value or beauty can make use of logic and evidence, but neither logical validity nor factual evidence is sufficient to establish goodness. (BTHâ„¢ is consistent with the observation of Quine and the pragmatists, that no actual proposition is ever entirely in a single domain; this is an abstract distinction that I am making.)

A syllogism can be valid without being a factually true proposition. Aristotle’s discovery of the compelling nature of logical inference, based on the form, not the content, of a statement, was a great discovery.

1. All men are bald.
2.Socrates is a man
3.Therefore, Socrates is bald.

is a logically valid syllogism, but a factually untrue statement. I think a common idea of Popperian falsification (though not Popper’s own understanding, perhaps) is that an analytical theory is just an elaborate, systematic, elaborate version of a syllogism, and empiricism is testing the premises or the conclusion, to see if the logic and/or conclusion qua generalization holds. Find one man with a full head of hair — the black swan — and voila! falsification.
An alternative view, more in line with the BTHâ„¢, is that analytic theories are completely separate and apart from any operational models of the world, yielding tentative, but pragmatically useful “predictions” about how observable (aka measurable) variables relate in the operating mechanisms of the world-machine. An analytical theory, such as Pigou’s classical economics, or Samuelson’s neoclassical economics, can be elaborate, systematic and powerful, with the felt force of valid logic compelling those, who learn it.
Euclid’s Geometry, plus the calculus, was sufficient to Newton’s physics, but the point of Einstein’s Special Relativity, was that it was not sufficient to the “extreme” case of the speed of light. The axiom of parallel lines was fine up to a point, but resulted in absurdity in the “extreme” case. And, the act of careful measurement revealed that absurdity. That’s the kind of limits to explanation, revealed in measurement, which results in the fabled paradigm shift. I’m not sure “falsification” quite covers the nature or manner of such discoveries.
Any theory has its explanatory limits. The act of careful measurement “objectively” reveals those limits. But, great theories elaborate wonderfully, and therefore, seem very, very powerful.
Compressing a theory into a single, elegant sytematic elaboration is really about overcoming human psychological limits, limits of cognitive, computing capacity. Knowing how to elaborate E=Mc2, or “survival of the fittest” or “supply and demand” or “The American Civil War was about slavery” into detailed, objectively measurable statements about the world, is what knowledge is about.
Economists have a great, systematic, logically-vetting theory in Samuelson’s Foundations and related work. It is not accidental that it looks, in form, like Euclid’s Geometry. It is a powerful instrument. Zombie Zero is Dr. Pangloss, who clings to the power of that theory to generate explanations, against any and all measurement.
Economics is seriously confused about its theory. It thinks its “geometry” is a map. It is a more fundamental confusion that that exemplified by, “The Map is Not the Territory”; it is “your geometry is not a map”. What Economics lacks is a framework for organizing knowledge from a collection of operational models and measurements. There are countless examples of operational models applied to measure the economy, which come back with a result that demonstrates the shortcomings of the “geometry”.
It may well be that a revised geometry, an equivalent of Einstein applying a non-Euclidean geometry to the problem of relativity, is simply not possible for Economics. It seems to me that the problem of polar Rationalities exemplifies that possibility. In an operational model, rationality with a small “r” can be a very weak assumption to give observations a functional relationship; in a purely deductive analysis, though, it is really hard to come up with anything solveable, that doesn’t involve superrationality.
Money and finance, central, core institutions of the economy, are clearly all about practical solutions to problems of incomplete information and bounded rationality, but there’s no obvious candidate for the equivalent of a non-Euclidean geometry. It is hard to give up the “power” of a systematic, logical structure — the apt-named 1st best economics, even if you know its shortcomings are legion, when there’s no good replacement.

there are whole disciplines out there that spend their time studying what makes people tick. It’s just that many economists don’t seem to want to take them seriously.

ISTM that if they aren’t going to take seriously the study of what makes people (i.e. humans) tick, they shouldn’t describe what they are doing as “economics”, because it is clearly not the study of economies containing human beings.

If they want to study the behavior of certain kinds of abstract systems of idealized actors, fine, but using anything that comes out of that kind of toy universe to attempt to explain or describe any property of an actual economy containing human beings is akin to using a Picasso painting as an anatomy textbook. It was never intended to be representational and, to the extent that you evaluate it as a representation anyway, it is a very bad one.

Also, OT, but anybody else here reading David Graeber’s Debt: The First 5,000 Years? Awe. Some. If ever a book lived up to its hype, that one’s it. I bring it up because of the very smart connection he makes between the zombie legend, which after all began in Haiti, and the experience of slavery. A person is effectively dead, and yet still functions as a body, but entirely subject to the will of another. Hmm…