Which school of economic thought do you prefer?

[ag]

Let's be clear: you have beasts of very different type listed.

The "Chicago school" and the (Neo)-Keynsians (it's hard to find the ones without a "neo" these days) are among the many currently existing approaches to the study of macroeconomics (though, these days we'd, probably, be speaking more often of the Minnesota school, when discussing the freshwater thinking ) ). Neither term is particularly applicable to any other branch of economics: as a microeconomist, for instance, I have no opinion whatsoever on the relative virtues of these (nor is anyone expecting me to have such an opinion). Still, I am aware of the broad division of that world into the freshwater and the saltwater types - macroeconomists are a bit like crocs ) The saltwaters concentrate on all sorts of somewhat undermodelled imperfections, which they tend to accept exogenously, as descriptions of the empirical fact. Whereas the freshwaters want to model everything from the first principles, so if any imperfections are there, they have to be generated endogenously. Still, methodologically the two "schools" are rather similar: you, probably, need graduate training in economics to figure out which paper in a learned journal is representing what.

Both the Marxism and the Austrian School have at one point been important schools of economic thought. They played enormous role in formulating modern economics. Marxism, though much more traditionalist of the two (in terms of his approach Marx was much more faithful to the classics like Adam Smith or Ricardo than most of his contemporaries), turned out to be somewhat a dead end, considering the development in economics that began already in Marx's own lifetime. To the extent that modern Marxists still exist in the economic profession, they tend to concentrate on issues of planning and such - things that, frankly, bore most modern economists to death. The bulk of self-identified Marxists wouldn't be identified as economists by the rest of us. Still, people like Duncan Foley or John Roemer have made some very important contributions to the modern economic profession at large fairly recently (and are still active).  So, at least, there are people who are both Marxists and economists out there today )

Austrian school was, in a sense, much more important until fairly recently. A lot of the modern views on basic micro where either formed by the Austrians or in conversation with them. Unfortunately, "Austrianism" is dead as a scientific school today. Its last two representatives of note were Hayek and Mises, who, unquestionably, were great economists. Unfortunately, what right now mascarades as the Austrian school is a Misesian religious cult dedicated to the eternal interpretation of the words of the Prophet. Their rejection of the scientific method as applied to social sciences has put them completely outside of not merely economics, but of any science. Their rejection of the common language of modern research has made them unintelligible to anyone but themselves. There are some decent people trying to bridge the gap between the modern Austrians and the rest of the humanity, but they are few and I don't envy them ) For the most part (with very few exceptions), it's a cult and a theology, not economics at all.

[ag]

BTW, who here can define each "school" ) It's a funny thing to vote on something without a clear idea of what is what.

[ag]

isn't the Chicago school pretty much a variety of the Austrian?

To the extent that they have about exactly nothing in common (other than both are somewhat associated w/ political conservatism) this is rather funny ))  For the Austrian cultists Chicagoans are worse than the Devil reincarnate, much worse than the Marxists. For Chicagoans the Austrians don't exist.

[ag]

also, opebo would probably call himself a Keynesian.

He is anything but

[ag]

However I'm skeptical that there is any difference between 'economic thought' and political preference.

That's because you don't really bother to learn economics ) In terms of political preference, an average Chicago econ guy is, probably, not that much different from an average self-identified Austrian: in Chicago they do tend to be very much pro free-market. However, in terms of their views on economics, they wouldn't be even able to agree on what IS economics ) There is really not much of a possibility of a reasonable conversation between them. And, of course, a fairly recent student of Lucas (can't be more Chicago than that) went on to be a member of the Venezuelan cabinet under Chavez (and one of the early ideologues of that regime) )

[ag]

However I'm skeptical that there is any difference between 'economic thought' and political preference.

That's because you don't really bother to learn economics

Actually no, the skepticism is the reason I didn't bother to learn economics.  I don't know much about religion either.

It's what we, economists, call the autarky equilibrium If you believe life is sh**t and can't be enjoyed, you'll wind up confirming your belief ))) Frankly, I find your intellectual autarky rather amusing )))

BTW, though a lifelong atheist, I am extremely fond of reading about religion )

[ag]

Econometrics

Calculus. English. Econometrics is not a school of thought in any identifiable sense of the word

[ag]

Anyway, though I have my disagreements, the Austrian school certainly makes the most sense among those listed here.

How so?

The Austrian school approaches the issue of economics as it should; by actually taking into account human behavior. Building upon human behavior, it keeps things simple and axiomatic. Economics is not a natural science.

Your description is admirable in having exactly nothing to do w/ modern Austrianism The key feature of Austrianism is that it rejects empirical observation of human behavior. The "mainstream" economics, in contrast, is precisely based on modeling and analyzing human behavior. In fact, all of modern economics is precisely the science of human behavior.  Likewise, the formal axiomatic approach is anathema to the Austrians but is quite common within the "mainstream".

[ag]

Human behavior cannot be predicted by empirical observation.

If that were the case, then opebo would have been absolutely right in his view of economics As somebody who's actually run lab experiments, I can assure you that human behavior is highly predictable )) There are extremely strong empirical regularities, replicated from lab to lab all over the world, believe it or not ))))))

What modern Austrians call "axioms" is, mostly, tautologies and logical tirivialities. That's actually the reason they hate math: when forced to use non-ambiguous language they are unable to hide that fact behind the meaningless verbiage, as they normally do. It's a religious cult, not a school of scientific thought.

[ag]

Human behavior cannot be predicted by empirical observation.

If that were the case, then opebo would have been absolutely right in his view of economics

What's with the baseless, feeble insults, ag?  And all the smilies?  I suppose they were funny for a while, but are you going to continue it forever?

I don't deny: your posts amuse me ) I don't believe I was insulting you. I guess, there is no way you might suppose that I agree with you that economics is meaningless blabber. Whereas, were I to accept Libertas's point of view, I'd have to agree w/ you on that without reservation. If anything, that was almost an appreciation of your point of view )

[ag]

...Whereas, were I to accept Libertas's point of view, I'd have to agree w/ you on that without reservation. If anything, that was almost an appreciation of your point of view

My views aren't based on human behavior being unpredictable, ag, though of course I am skeptical of such prediction.

You seem strangely cheerful - what's up?  

I just happen to have some sense of humor ) You should have noticed, I am very frequently smiling on these boards

All I meant to say is that if human behavior were unpredictable, I'd be in full agreement with you that economics is a piece of crap ))) That is, we'd be in agreement, if not on the reasoning, than on conclusions.

[ag]

Human behavior cannot be predicted by empirical observation.

If that were the case, then opebo would have been absolutely right in his view of economics As somebody who's actually run lab experiments, I can assure you that human behavior is highly predictable )) There are extremely strong empirical regularities, replicated from lab to lab all over the world, believe it or not ))))))

Sounds like bollocks to me. Evidence?

I'd refer you to any amount of published experimental work. But let me tell you about some first-hand experience.

I recall when I started running pilots for a certain auctions experiment before reading up on the relevant literature (auctions isn't really my field, this was my first foray into them). I discovered what to me seemed like an incredible behavior (people were bidding above value in circumstances where I wouldn't think it reasonable at all). To my surprise, when I actually did read about what others have done, it turned out to be something well-known for over 15 years: in this whole class of auctions people were known to overbid. My bid scatterplots were indistinguishable from what has long been published (literally: if the same software were to be used to produce the graphs, I wouldn't be able to tell which is which by eyeballing). Now, given my previous ignorance, there is no way I could, consciously or unconsciously, make my results conform to what others (many others) have been observing. This, by the way, was the first experimental run of this class of auctions in a Latin American lab, the experimental instructions were produced independently of anybody else, etc., etc., etc.  Oh, yeah - human behavior is predictable. Very predictable.

[ag]

...To my surprise, when I actually did read about what others have done, it turned out to be something well-known for over 15 years: in this whole class of auctions people were known to overbid. ...

 Oh, yeah - human behavior is predictable. Very predictable.

Ok, firstly, everyone knows people tend to overbid at auctions.  It is literally common knowledge - it is in hollywood films, little children know about it.  How could you not know this?

And secondly, no one has disputed the notion that human behavior is predictable.

Your first point might be true about the movies, but not true in practice ) One change in the rules (going, for instance, from the second-price to the first-price auction) and that particular behavior disappears as if by charm ) Same, demographics, same everything )

Your second point is, likewise, not true: Libertas did argue that human behavior is not predictable ) Now, you were brought into that discussion only to make the point that if Libertas were right in his claim, I would have fully agreed w/ you in your conclusion ))

[ag]

Human behavior cannot be predicted by empirical observation.

If that were the case, then opebo would have been absolutely right in his view of economics As somebody who's actually run lab experiments, I can assure you that human behavior is highly predictable )) There are extremely strong empirical regularities, replicated from lab to lab all over the world, believe it or not ))))))

Sounds like bollocks to me. Evidence?

I'd refer you to any amount of published experimental work. But let me tell you about some first-hand experience.

I recall when I started running pilots for a certain auctions experiment before reading up on the relevant literature (auctions isn't really my field, this was my first foray into them). I discovered what to me seemed like an incredible behavior (people were bidding above value in circumstances where I wouldn't think it reasonable at all). To my surprise, when I actually did read about what others have done, it turned out to be something well-known for over 15 years: in this whole class of auctions people were known to overbid. My bid scatterplots were indistinguishable from what has long been published (literally: if the same software were to be used to produce the graphs, I wouldn't be able to tell which is which by eyeballing). Now, given my previous ignorance, there is no way I could, consciously or unconsciously, make my results conform to what others (many others) have been observing. This, by the way, was the first experimental run of this class of auctions in a Latin American lab, the experimental instructions were produced independently of anybody else, etc., etc., etc.  Oh, yeah - human behavior is predictable. Very predictable.

Auctions - with their regularized practices and rules - hardly strike me as a good place to study 'human behaviour' (Oh, and what were the demographics of the people you were studying). Though in saying that I desire details - what were reasons for this overbidding - circumstances?

To make the long story short, overbidding, defined as bidding above the resale value of the object with no sentimental value (e.g., for a dollar bill), occurs nearly always when subjects participate in sealed-bid second-price auctions (and their analogs). It does not occur either in first-price auctions nor in the ascending-bid English auctions. The result is extremely robust . My subjects have been first-year undergraduates, but others have done it in different pools (though, I am sure, if I were to run it among the Natives in a jungle, it would change - that would be a job for an anthropologist, not for an economist). I've almost literally been selling 5 dollar bills for 5 dollars and a quarter, on average (though, having tried various things I also know what to do to "teach" people to stop doing it). I have a theory why that happens, but the issue is, actually, a subject of debate among the researchers.

If you don't like auctions, I've run voting experiments, I've run bargaining experiments - you want first-hand accounts from those? I nearly always run, as a control, and experiment others had run in a different lab. Whenever there's been substantial body of evidence accumulated on such experiments, I tend to get results that are fairly similar to others. Admittedly, lab experiments are a structured environment by definition - like, say, physics lab experiments, where they take pains to eliminate effects that would confound their observation, even though those effects are present in real life. I don't see what's the difference here.

[ag]

quote]

No, my point was that people know that it is commonplace to overbid at auctions.   I never mentioned anything about 'rules changes'.  I was talking about actual auctions as they exist in practice.

And I was talking about the actual auctions that exist in practice There are just many types of these. In fact, I believe that the auctions that you have in mind (most likely you are thinking about the ascending bid English auction) do not result in the particular kind of overbidding I've been talking about )

[ag]

Oh I see.  I was considering my conversation with you, not your conversation with Libertas.  My conclusion regarding economics was based more on a skepticism about the underlying motivations behind it than the predictability of behavior.   

I know ) On the other hand, I sense I should be getting offended by you clearly impugning my motivations )) As far as I am concerned, I only care about understanding human behavior (and that's where, for once, I am serious).

[ag]

Human behavior cannot be predicted by empirical observation.

If that were the case, then opebo would have been absolutely right in his view of economics As somebody who's actually run lab experiments, I can assure you that human behavior is highly predictable )) There are extremely strong empirical regularities, replicated from lab to lab all over the world, believe it or not ))))))

Sounds like bollocks to me. Evidence?

I'd refer you to any amount of published experimental work. But let me tell you about some first-hand experience.

I recall when I started running pilots for a certain auctions experiment before reading up on the relevant literature (auctions isn't really my field, this was my first foray into them). I discovered what to me seemed like an incredible behavior (people were bidding above value in circumstances where I wouldn't think it reasonable at all). To my surprise, when I actually did read about what others have done, it turned out to be something well-known for over 15 years: in this whole class of auctions people were known to overbid. My bid scatterplots were indistinguishable from what has long been published (literally: if the same software were to be used to produce the graphs, I wouldn't be able to tell which is which by eyeballing). Now, given my previous ignorance, there is no way I could, consciously or unconsciously, make my results conform to what others (many others) have been observing. This, by the way, was the first experimental run of this class of auctions in a Latin American lab, the experimental instructions were produced independently of anybody else, etc., etc., etc.  Oh, yeah - human behavior is predictable. Very predictable.

Auctions - with their regularized practices and rules - hardly strike me as a good place to study 'human behaviour' (Oh, and what were the demographics of the people you were studying). Though in saying that I desire details - what were reasons for this overbidding - circumstances?

To make the long story short, overbidding, defined as bidding above the resale value of the object with no sentimental value (e.g., for a dollar bill), occurs nearly always when subjects participate in sealed-bid second-price auctions (and their analogs). It does not occur either in first-price auctions nor in the ascending-bid English auctions. The result is extremely robust . My subjects have been first-year undergraduates, but others have done it in different pools (though, I am sure, if I were to run it among the Natives in a jungle, it would change - that would be a job for an anthropologist, not for an economist). I've almost literally been selling 5 dollar bills for 5 dollars and a quarter, on average (though, having tried various things I also know what to do to "teach" people to stop doing it). I have a theory why that happens, but the issue is, actually, a subject of debate among the researchers.

If you don't like auctions, I've run voting experiments, I've run bargaining experiments - you want first-hand accounts from those? I nearly always run, as a control, and experiment others had run in a different lab. Whenever there's been substantial body of evidence accumulated on such experiments, I tend to get results that are fairly similar to others. Admittedly, lab experiments are a structured environment by definition - like, say, physics lab experiments, where they take pains to eliminate effects that would confound their observation, even though those effects are present in real life. I don't see what's the difference here.

So therefore not 'human behaviour' then. Glad we cleared that up.

Your confession that you are not human puzzles me: I thought that most dogs try to avoid confessing to that fact on the internet ))

Seriously, though, it would be ridiculous to argue that culture doesn't matter for human choices. There are also many, many other things that do matter and matter a lot. Change the environment and you will change the outcomes. Only an Austrian can be so uninterested in human beings to think otherwise )

[ag]

Human behavior cannot be predicted by empirical observation.

If that were the case, then opebo would have been absolutely right in his view of economics As somebody who's actually run lab experiments, I can assure you that human behavior is highly predictable )) There are extremely strong empirical regularities, replicated from lab to lab all over the world, believe it or not ))))))

Sounds like bollocks to me. Evidence?

I'd refer you to any amount of published experimental work. But let me tell you about some first-hand experience.

I recall when I started running pilots for a certain auctions experiment before reading up on the relevant literature (auctions isn't really my field, this was my first foray into them). I discovered what to me seemed like an incredible behavior (people were bidding above value in circumstances where I wouldn't think it reasonable at all). To my surprise, when I actually did read about what others have done, it turned out to be something well-known for over 15 years: in this whole class of auctions people were known to overbid. My bid scatterplots were indistinguishable from what has long been published (literally: if the same software were to be used to produce the graphs, I wouldn't be able to tell which is which by eyeballing). Now, given my previous ignorance, there is no way I could, consciously or unconsciously, make my results conform to what others (many others) have been observing. This, by the way, was the first experimental run of this class of auctions in a Latin American lab, the experimental instructions were produced independently of anybody else, etc., etc., etc.  Oh, yeah - human behavior is predictable. Very predictable.

Auctions - with their regularized practices and rules - hardly strike me as a good place to study 'human behaviour' (Oh, and what were the demographics of the people you were studying). Though in saying that I desire details - what were reasons for this overbidding - circumstances?

To make the long story short, overbidding, defined as bidding above the resale value of the object with no sentimental value (e.g., for a dollar bill), occurs nearly always when subjects participate in sealed-bid second-price auctions (and their analogs). It does not occur either in first-price auctions nor in the ascending-bid English auctions. The result is extremely robust . My subjects have been first-year undergraduates, but others have done it in different pools (though, I am sure, if I were to run it among the Natives in a jungle, it would change - that would be a job for an anthropologist, not for an economist). I've almost literally been selling 5 dollar bills for 5 dollars and a quarter, on average (though, having tried various things I also know what to do to "teach" people to stop doing it). I have a theory why that happens, but the issue is, actually, a subject of debate among the researchers.

If you don't like auctions, I've run voting experiments, I've run bargaining experiments - you want first-hand accounts from those? I nearly always run, as a control, and experiment others had run in a different lab. Whenever there's been substantial body of evidence accumulated on such experiments, I tend to get results that are fairly similar to others. Admittedly, lab experiments are a structured environment by definition - like, say, physics lab experiments, where they take pains to eliminate effects that would confound their observation, even though those effects are present in real life. I don't see what's the difference here.

So therefore not 'human behaviour' then. Glad we cleared that up.

Human behaviour that isn't natives in a jungle isn't human behaviour?

Errr. .. No. If you are claiming that something is 'human behaviour' then you might want to make sure it applies to all humans.

No. I am not saying anything about behaviors characteristic of all humans - in fact, establishing such behavioral patterns, if they exist, has little to do w/ economics at all, this is biology, probably. All I am saying is, I can predict behavior of humans if I get relevant information. Their background is relevant information, as is their behavior in other experiments. If I were to make a point of studying natives in the jungle, within a few years I'd get pretty well predicting that. It so happens, this study is not very interesting for an economist, but is done by antropologists. They run their experiments in the rainforest and are pretty good at that. Normally, cultural issues are not very interesting to economists - we look at variations in different variables, trying to keep culture constant.

[ag]

Human behavior is usually used in a generic sense; like 'human nature' and so on. As something that is typical and applicable to all of us at least in theory, rather than 'behavior that some humans do'.

Where did you take such "usual" usage from? Would you mind a citation? It surely isn't a common usage among economists: such study wouldn't be of interest to almost any of them.

[ag]

Human behavior is usually used in a generic sense; like 'human nature' and so on. As something that is typical and applicable to all of us at least in theory, rather than 'behavior that some humans do'.

So, studying human behaviour would mean studying only acts committed by all humans? That strikes me as pretty odd. Most human behaviour isn't the same for every individual so it would seem a bit pointless.

Most economic studies in these fields tend to be of the type "People with attribute X tends to do Y in situation Z" which is basically what I would expect.

Yes, but it makes useless any claim that the models of economics represent 'human nature'.

Note, that this claim has been made by nobody but yourself here ) In fact, having spent many years in this profession, I think I've never heard it made (I am not going to argue that there is no economist that would make it - I just never met one). So, it can hardly be at the foundation of modern economics )

It is always easy to destroy the paper tiger of your imagination. Having made a ridiculous argument, it is easy to laugh at it. It is a bit harder if you have to argue with the real, not an imaginary, opponent. I this case, for instance, you'd have to learn something about economics: something that you, obviously, never bothered to do ))

[ag]

Human behavior is usually used in a generic sense; like 'human nature' and so on. As something that is typical and applicable to all of us at least in theory, rather than 'behavior that some humans do'.

Where did you take such "usual" usage from? Would you mind a citation? It surely isn't a common usage among economists: such study wouldn't be of interest to almost any of them.

I take it from the fact that I've never really seen it used in a different way before; and because that is what I take that term to mean when I see it used. The behavior of humans at an extreme basic level (the level at which we are all 'the same' rather than at the social) in other words. As it happens, I also think that it is a completely useless term as you can't separate humans from the societies in which they live in.

Wow! You've never understood human behavior to simply mean "behavior of humans", not "behavior of all humans"? Yeah, we have to work harder ))

Now, all economics is about behavior. Nevertheless, I've never seen an economics text that would make any claim about any behavior characterizing all humans. Nobody is even thinking in these categories. Even if we assume rationality (which economists do frequently, but not, by any means, always), rationality means only completeness and transitivity of preferences. It says exactly nothing about what preferences should anyone have. That we can only get from empirical observation. Clearly, cultural (and not only cultural) determinants would be important here. Those who are interested in these things do all sorts of interdisciplinary research. The bulk of economists study something else, but nearly everybody would view this point as obvious - not even worth discussing.

I wonder, where all these idiotic claims about what modern economics is about come from?

[ag]

Human behavior is usually used in a generic sense; like 'human nature' and so on. As something that is typical and applicable to all of us at least in theory, rather than 'behavior that some humans do'.

Where did you take such "usual" usage from? Would you mind a citation? It surely isn't a common usage among economists: such study wouldn't be of interest to almost any of them.

I take it from the fact that I've never really seen it used in a different way before; and because that is what I take that term to mean when I see it used. The behavior of humans at an extreme basic level (the level at which we are all 'the same' rather than at the social) in other words. As it happens, I also think that it is a completely useless term as you can't separate humans from the societies in which they live in.

Wow! You've never understood human behavior to simply mean "behavior of humans, not "behavior of all humans"? Yeah, we have to work harder ))

Now, all economics is about behavior. Nevertheless, I've never seen an economics text that would make any claim about any behavior characterizing all humans. Nobody is even thinking in these categories. Even if we assume rationality (which economists do frequently, but not, by any means, always), rationality means only completeness and transitivity of preferences. It says exactly nothing about what preferences should anyone have. That we can only get from empirical observation.

But a cornerstone of all modern economics as far as I understand it is that humans behave rationally (whatever that means) so that is claiming an empirically testable idea about 'humans'. After all, what else are your experiments meant to prove in the first place?

Even that is an exaggeration: these days there are zillions of models that are not based on rationality. However, rationality itself is not really implying anything at all about behavior. It's rationality together with the model of what is important for decisions (i.e., what economists call the "consumption space") that gives testable restrictions on behavior.

In any case, rationality is just completeness and transitivity of prefernces. Its only testable restriction is what is known as the Strong Axiom of Revealed Preference. It says nothing whatsoever about what an individual should do in any situation, but merely imposes a restriction on which observations would be mutually consistent (assuming we know the consumption space, that is). The fact that a Macheguenga Indian in Peruvian jungle isn't, say, reacting to monetary payoffs in the same manner as a Michigan State undergrad is entirely consistent with modern economists' views of the world. In fact, it would have been an extraordinary phenomenon if there were no difference in actual choices in this example, that would have been inexplicable within economics as such (in this case we'd have to go to biologists and ask them to tell us what the hell is going on).

The hypotheses that we actually test (in the lab or with real life data) are, 99% of the time, not rationality per se, but something with a lot more structure. This structure, of course, we, in turn, get from observation: we build models  that explain existing facts and then subject them to tests w/ new data.  That's the normal process of scientific research. Yes, most of the models use the language of rationality, but rationality per se is not generating the predictions we are normally testing.

[ag]

You raise many issues, let me touch on just some of them.

1. There are zillions of papers on things other than rationality. For recent examples scholar google, say, the rational shortlist (recently discussed by Manzini and Mariotti). Or anything on choice w/ frames or from lists (see Rubinstein). Or, say, on sequential rationalizability, or on choice with status quo. Or, in general, on bounded rationality (I could give many other buzzwords, but these should be enough). If you can't find a 100 papers that appeared in the last few years, I'd conclude that you can't use scholar google.

2. Good economics is not about moral judgements at all, not about human actions, not about anything else.  I make many moral judgements in my life, and not one of them is based on anything I ever learned in economics. In my view, anyone, who tries to make one's morals based on economics, is an idiot.

3. Math is used for simplicity sake, as a language. It is forcing us to state assumptions and make clear logical steps involved in getting to the conclusions. It is true, it is possible to do this without math - it's just much harder and very few people are capable of it. The shoddy logic and poorly defined assumptions of those members of the modern Austrian cult, who claim to reject math, provide ample (though, by no means exclusive) evidence for that

4. Defining consumption space forces me to take a stand on what I think is relevant for individual choice in a particular setting. It's just another clear assumption I have to state explicitly every time I start working out a model. It is specific to the model I am considering - no universal claims are made. I am quite aware, that I might be wrong in my definition in the sense that individual choices I am studying were made w/ a different consumption space in mind.  If the choices studied don't satisfy, say, the above-mentioned Strong Axiom of Revealed preference given the consumption space I can conclude that something in the model was wrong: either the consumption space, or the rational model of decisions out of it.  Frankly, I'd be quite happy about it: falsifying (proving wrong) theories is bread and butter of every researcher. What to relax (rationality or the consumption space) would depend on the particulars of the problem at hand.

5. Rationality itself is used as a sort of a professional Occam's razor: again, it makes reasoning simple and clear and assumptions easy to state and understand. As it is nearly impossible to falsify on most sets of data economists deal with, there is no advantage of using other models of decision-making in those cases, but there would be a great disadvantage: one would have to define zillions of things clearly and make sure readers understand them. Using rationality in an "as if" sense is, mostly, fairly harmless. When need arises, other models are, in fact, used.

6. I am, actually, fairly agnostic on the reasoning that people use to make their choices. So, no, I don't make a claim you imply I make.  No, I don't think Macheguengas make decision same way I do. Nor do I really take a stand that you make decisions same way I do either.

7. Obviously, behavior is fairly predictable only in a statistical sense. I can fairly well predict distributions in many cases - based on empirical observations. I don't claim to be able to predict behavior of any one person in the room, unless I really know that person well (and, in that case, I am likely to be doing this not as an economist but, say, as a spouse ) )

Did I skip anything else of importance?

[ag]

Btw.

Rationality is simply defined as:

1.) a > b > c

and

2.)

a > b
b > c
a > c

Perhaps consistency may be a better word.

Not quite. It is

1) either a is subjectively at least as good as b for individual i or b is subjectively at least as good as a for individual i (or both)

and

2) a is subjectively at least as good as b for individual i  and b is subjectively at least as good as c for individual i implies a is subjectively at least as good as c for individual i

It is usually followed by an assumption that an individual w/ such rational preferences always chooses the best alternative according to them. It's a neat model and a great disciplining device, but it ain't worth dying for BTW, Austrians, to the best of my knowledge, have exactly no problem w/ this assumption )) (except for some ridiculously irrelevant mumbo-jumbo about the nature of indifference to which any sane economist is entirely indifferent ) )

PS: BTW, notice, I deliberately avoided using math notation. If they insist, I am willing to abjure it completely. Of course, translating a 10-page notation-dense article into notationless English might make it into a 200-page volume, but it's doable )

[ag]

I presume that he means in the sense that you can look at collective behavior, social behavior, whatever you want to call it; the actions of a large group of people. Which is perfectly legitimate, though the dangers of determinism and essentialisation are always something to be wary of.

Oh I knew what he meant - my point was what you just expressed in the second sentence.

That's because you a) never read any empirical research in economics (or, for that matter, any theoretical research either, I am afraid) and b) maintain an utterly ridiculous assumption that economics has anything to say about morals )