Yes maybe you’ve read the Caplan paper.
Here is the rebuttal:
Yes tell me more.
Reminds me of a joke: “I don’t have a girlfriend, but I know a girl who’d be upset to hear that.”
There are economists who’d be surprised that all their macro models have microfoundations, such as Krugman.
I’m tried to edit this so that I wasn’t suggesting that the macro models don’t have microfoundations, just that macro economists don’t bother going through the derivation each time, and basically just work with the macro models, but i’m ran out of time…
Under the axiomatic approach, the utility function is differentiable even if the utils themselves are ordinal–that’s just math. Basically we chose our axioms such that they will result in “nicely behaved” utility functions.
But I have a more basic question. Your quote says
- "The crucial fact is that one cannot divide preference ranks by one another and then compare the result to a ratio of prices. It is obvious that equality between the ratio of marginal utilities (preference ranks) and the ratio of the prices could only exist under two conditions. One, if preference ranks and prices had the same dimension (that is, if they were the same kind of thing), then their ratios could undoubtedly be equal. "*
Quick question for you: what is the unit attached to a ratio of prices?
My world involves slavery? I’m pretty sure my world outlawed slavery. You’re the one endorsing the guy who explicitly said parents should be allowed to sell their children to factory owners. Which I think also qualifies as terrorizing.
Part (not all) of this difficulty can be addressed with the “Kelly Criterion” (named after a Doper? 
), making Utility the logarithm of total Value. This is used by professional gamblers and, I think, some investment fund managers. (Some fund failures can be attributed to ignoring this Criterion, I think, though, in their “defense,” those managers were attempting to maximize their own utility, not their clients’.)
Do economists ever use Utility ≈ logarithm(Total_Value) explicitly?
This heuristic criterion was first discovered not by Doper Kelly nor any other Kelly, but by Daniel Bernoulli 280 years ago. His short paper doesn’t mention the word ‘logarithm’ but …
… The integral of an inverse proportion is a logarithm.
Oh definitely. The skills they possess that translate to politics are off the charts. If more of them lived to be 60 we might have a WWE President. Dwayne Johnson looks on track to do that.
Yeah, that’s a pretty common utility function (the natural log is much more common than, say, log base 10). It’s so common that it has a name: the Cobb Douglas utility function (technically the Cobb Douglas utility function is x^a*y^(1-a), but since the log is a monotonic transformation it’s equivalent)
I’ll just leave this here cause it cracks me up:
I said all “academic” models.
By this, I mean exactly the same thing Simon Wren-Lewis says in the post Krugman cites: we are in “an age where journal papers in macro theory are nearly always microfounded DSGE models”.
This is the way the academic journals are today. That’s what I was referencing. With extremely few exceptions, academic macro models just aren’t going to get published unless they have all the micro stuff built in
Yes.
As Evil Economist says, this is Cobb-Douglas utility.
One of the weird things with log utility in an intertemporal model is that it can collapse the dynamic problem into a static one. Log utility is sometimes called “myopic” because of this property.
OK, but your quotes don’t support “all”.
That’s why I included the link:
“The same problem appears on the side of price ratios. The common view that sees no difficulty in the comparison of price ratios is unwarranted. The problem becomes obvious once we recall that prices are themselves ratios. A price is not just “3 dollars” but rather “3 dollars / 1 hamburger.” Now consider the ratios of this price with two other prices, say, “1 dollar / 1 banana” and “2 dollars / 1 coke.” The ratio of the hamburger and the banana prices would be “3 bananas / 1 hamburg- ers,” and the ratio of the hamburger and the coke prices would be “3 cokes / 2 hamburgers.”
It is clear that we encounter here exactly the same problems as above in the case of ratios of preference ranks (see Hülsmann 1996, chap. 6). The first problem is to interpret the meaning of the different units. What does (banana / hamburger) and (coke / hamburger) actually mean? But the most important problem is that all these ratios are incommensurable. They are idiosyncratic just like the ratios of preference ranks. It is impossible to tell whether any number of the dimension (banana / hamburger) is equal to another number of the dimension (coke / hamburger).
Thus the central proposition of neoclassical price theory, that in equilibrium the ratio of the preference ranks of the various goods equals the ratio of their prices, is fallacious in all its parts.”
http://austrianeconomics.org/sites/default/files/qjae2_4_1.pdf
I don’t see that Cobb-Douglas, nor its inverse, is a simple logarithm. And all utility functions, if sensical at all, will be monotonic.
(The base of logarithm is irrelevant is such a discussion — it’s just like the choice whether to measure in dollars or kilo-dollars.)
x[sup]a[/sup]y[sup]1-a[/sup] isn’t, itself, a logarithm.
But if you have an ordinal preference relationship for which u[x,y] = x[sup]a[/sup]y[sup]b[/sup] accurately expresses the ordinal preferences, then that same preference relation can also be expressed fully accurately with logarithms.
And of course, the reverse is true. If you have an ordinal preference relation that can be expressed accurately with logarithms, then it can also be expressed with the Cobb-Douglas formulation.
Some of the most famous proofs in economics don’t actually rely on monotonic utility, but the weaker criterion of locally non-satiated preferences. With this property, it’s possible for people to have more of a good in certain cases and be “less happy” having more of that good, all other goods held equal. Not commonly used in practice. A little weird, obviously. Violates some basic common sense. But because local non-satiation is a weaker condition than monotonicity, any proof made with this property is slightly more general.
Econ theory peeps like more general proofs, yo.
But the point Evil Economist was making is that a monotonic transformation of Cobb-Douglas utility will preserve the ordinal preference relation. Because obviously. It’s ordinal. The distance between “utilities” simply does not matter for the model. Well, a log is a monotonic function, which means multiplying Cobb-Douglas utility by a log will preserve the same relationship. This means that the utility function x[sup]a[/sup]y[sup]b[/sup] can be equivalently expressed as log[x] + β log[y], with the appropriately chosen beta. Using this form of the function in a model often can emphasize that the beta is a discount, but of course, there are other equivalent formulations. More generally, any monotonic transformation should preserve the ordinal preference relation.
Linking to people just as ignorant as you are does not help anything.
If you’re interested in learning about an ordinal preference relation, expressed as a utility function, you can ask in GQ, or even ask here. The math can be explained. But a link to someone else who doesn’t understand the topic will never help you understand the topic.
Why should I believe you over someone who can describe the flaws in econometrics with logic? What have econometricians done that is noteworthy? Please don’t embarrass yourself by pointing to academic credentials.
Econometricians have no use in the real world. They are an insular group speaking to themselves. It is people who can explain the logic of economics that have an impact on society. Milton Friedman is not popular because of his models. They are irrelevant now. He is popular because he could explain economics in a logical way. Keynes was trying logic. He failed, but his logic is influential.
So far, there has been a lot of chortling from people who have devoted their lives to an irrelevant study. They do not want to speak to the logic of economics. Econometric models are discarded nearly immediately after even a small change in human action. There is a reason for that. They do not describe reality, and merely attempt to imitate it without reckoning with the action axiom.
The emergence of behavioral economics is due to phenomena that neoclassicals can’t grapple with. People are unsatisfied by the repeatedly failing models that clog up economics journals, and thus behavioral economics has to compensate for these deficiencies.
I wrote out a response to this, but think it might be more helpful just to ask: what do you think the ratio 3 banana/1 hamburger means?
Okay.
First off.
I have not written one word about econometrics in this thread. Not a single word.
I have been explaining standard academic economic theory. Academic economic theory is deductive. It proceeds, with pure logic, from axioms to theorems by means of proof. You can disagree with the axioms. You almost certainly do disagree with the axioms. But you can’t disagree with the logic of the proof, as deductively derived from the axioms. You can say “I think that axiom is bullshit”, and that’s fine. Lots of people do. But you can’t say, “That theorem does not derive from those axioms.” It does. It’s been proven. That’s what proof is.
Econometrics is not the same thing as economic theory. Econometrics is essentially statistics, applied to economic problems. To put it as simply as I can: economists think that statistics (econometrics) is important, because there is disagreement about which (axiom-based) theory we should use. (Basically.) But there is no room for disagreement about the deductive implications of theories. Logic is what it is. Standard consumer theory uses ordinal preferences, and there is no disagreement about that – among people who actually know the logic – just as there is no room for disagreement that 2+2=4 in the real number system. You can claim, if you want, that 2+2=5. But you would be wrong.
If you think that standard consumer theory uses cardinal utility, then you are wrong.
If you cite people who believe that standard consumer theory uses cardinal utility, then they are wrong, too, just as they would be wrong if they thought 2+2=5.
You don’t have to believe me.
You are free to believe that 2+2=5. You are free to believe any random thing you’d like to believe.
But ordinal preference relations in standard consumer theory work by pure logic. You are not going to learn the steps of that logic from someone who does not understand the steps. I would hope that this was obvious. It is unfortunate that this does not seem obvious to you.
I could, personally, work through a few steps of this concretely proven logical structure. But would you listen? You seem to have already made up your mind. You are relying on people who have made mistakes, and you don’t see their mistakes. Their “logic” seems fine to you, and this gives you an easy excuse not to pay attention. This is your choice. Again: you can believe whatever you want. If you want to believe falsehood, that’s your choice.
But if you actually wanted to learn something new, the SDMB is a good place to do that.
I could quite easily walk you through a few steps of the logic, and recommend books that have this logic. But the question becomes: Are you going to listen to the real logic of academic economics, as explained by someone trained in academic economics? This is not the same as believing academic economics. You can continue to disbelieve it at your discretion. But if you’re going to disbelieve it, then at minimum you should disbelieve it for the right reasons. Are you going to hold on to your current understanding of the subject, as explained by people outside that subject who don’t know what they’re talking about, just because the “logic” of their arguments seems attractive to you, despite your own lack of training in this subject? Again: None of this requires that you believe the axioms, or to believe academic economics. I’m fairly certain you will reject them, even after you understand them. I have a feeling about that.
But the quite indisputable point here is that you do not yet understand what this stuff is about, or you would not be making false statements or citing people who make false staements. You would be better off disbelieving academic economics for true reasons, rather than false ones. To believe that utility is necessarily cardinal is a false reason. Just a fact.
The explanation of the logic would go easier if you knew some calculus. But even if you don’t have any calculus, you can still learn the outlines of the logic.
Do you know what a mathematical relation is? Do you know what a preference relation is? That’s the place to start, if you actually want to learn. The “preference scale” Rothbard uses in Man, Economy, and State is a kind of preference relation. Do you know what a preference relation is? More generally, do you know what a mathematical relation is?
That’s the first step to understand the logic of utility functions which express ordinal preference relations.
I have no problem with any of the mathematical concepts as described thus far, and I have a good understanding of calculus. So I will continue to wait for exposition of content, as I have waited in many threads before these. I require only one answer from you: Is the Samuelson text adequate for the understanding of “academic economics” as you understand it?
There is no problem with the understanding of mathematics. Rothbard was a math undergrad. All of the Austrian economists I have learned from, and continue to learn from, studied neoclassical mathematics and received their PhD from neoclassical institutions. This makes them “academic economists” whether you like it or not. Indeed there are economists who reject Austrian economics, yet still come to the basic policy conclusions that I would favor, albeit for different reasons.