Alternative Economic Theory

[wphelan]

Interesting that you bring this up. The majority of the reading I've done in the last several months has been about economics. Specifically, I've concentrated on books written by people associated with the Austrian school of economics, which, for the most part, rejects the use of mathematics and statistics in economic analysis.

In fact, much of the criticism directed at Austrian economic thought focuses on its lack of "scientific" analysis and its subjective nature. However, I find it difficult to argue with Austrian analysis of the current economic situation, especially since it warned against much of the economic trouble we find ourselves in now.

Some of the details may be too laissez faire for some too accept, but I think the core theory, especially as it relates to the theory of the business cycle is dead on. Friedrich Hayek won the Nobel prize in economics in 1974 for his work on the theory, and I find it amazing that he is completely ignored today.

[wphelan]

Wow...it's like I'm on Lost or something. My post appears to reply to a post that was created after my reply. I guess the time issue I noticed earlier has been fixed. :cool:

[Michael]

This thread is for Dominick to lay out his comprehensive critique of classical economic theory and to describe for us a comprehensive alternative theory that is superior to it.

[Michael]

As far as the critique of existing economical theory concerns, I've laid it out several times already. The mathematical tools that are used are entirely deficient for the goal. And that fact is not a squibbling in the margin but the very essence of the issue. Any theory that is scientifically sound constitutes of nothing but mathematics. In fact that theory would be mathematics.

There is a fundamental difference between linear and non-linear systems. Linear systems are e.g. the acceleration of a car or the interaction of gravity between two bodies (and just two -it doesn't work for three bodies because such a system is already non linear). Those are intuitive and straightforward and when the actual mathematical function of the system is not known, perturbative (i.e. approximative) methods can and should be used.

Non-linear systems are fundamentally different. They have what is called a 'sensitive dependency on initial conditions' (ie. the butterfly-effect). Because of this sensitivity perturbative methods are useless. One might as well us a dartboard. Such systems can only be described by describing the whole. Needless to say it is far more difficult to achieve this when one has no intermediate -perturbative- steps. That's why the study of this is far more recent (a few decades) than that of linear systems (starting with Newton or even Euclid).

There can be no question that the global economy is a non linear system. It is riddled to the brim with feedback mechanisms which define non-linearity. For that reason a sound economical theory must be non-linear in toto. Current, classical economical theory is not.

When applying linear methods to non-linear systems, one will of course still get results. And as you say, they may 'more often than not' hit the mark. But that would also be the case with a set of random genetic algorithms that are left to compete. And that is essentially what economists are doing. They keep inventing new formulas until one hits the mark of today. And when the prediction for the next day is way off, they invent still another one. These are the methods that have been used to evaluate debt positions in the recent past. With the well known result. *

Only a theory that always hits the mark is meaningful (albeit with an error margin that all theories have) . It would have had to predict the credit crunch to be a serious theory.

Thus, as far as an alternative, I have none, and never pretended to. I have none because there is none and can't be one. We, mankind, at the moment lack the knowledge and tools to describe such a massively complex system. The ones in use today are purely ideological in nature and are scientifically invalid.

I know you won't like this explanation and would prefer one in terms of interpretation rather than fundamentals but interpretation doesn't matter when the fundamentals are wrong.

*Anecdote :
Other critics of economical theory have given a pack of chimpansees a big board with stock indicators on. The ones they hit or walked on the most were interpreted as 'buy', while the ones they never or rarely hit or walked on were interpreted as 'sell'. The apes had better results than the stock exchange advisors.
I see no critique of classical macroeconomic theory here.

I see a critique of the statistical applications of microeconomic theory (which you've made before). This doesn't seem to have anything to do with macroeconomics. Classical macroeconomic theory is all about 'gross' levels of employment, fiscal and monetary policy and GDP. That's it.

Predicting stocks on the stock market is pure microecomics (which has a very poor track record). Predicting any singular activity or the act of singular actors is pure microeconomics. Macroeconomics is the really big picture. Critiquing one does not impinge upon the other since they are not directly related. Macroeconomic theory is just that - macro (only).

Please identify an example of the failure of macroeconomics. Seems that macroeconomic theory provides excellent explanations of the meta-economic trends. Knowing macro-economic trends allows one to make general predictions about meta-economic trends (like recessions). Macroeconomic theory predicted that the US negative savings rate and the US current account deficit could not continue indefinitely and that it would cause a contraction. That is exactly what is happening. No funky math required to see this.

Microeconomics is perhaps a field worthy of your critique. They try to apply the principles of macro theory to the micro level - and use all kinds of funky math and statistics to do this. The results are highly questionable.

Classical macroeconomic theory supplies highly functional explanations of macroeconomic trends. I'm not aware of any significant failures of the general theory here.

[Michael]

Interesting that you bring this up. The majority of the reading I've done in the last several months has been about economics. Specifically, I've concentrated on books written by people associated with the Austrian school of economics, which, for the most part, rejects the use of mathematics and statistics in economic analysis.

In fact, much of the criticism directed at Austrian economic thought focuses on its lack of "scientific" analysis and its subjective nature. However, I find it difficult to argue with Austrian analysis of the current economic situation, especially since it warned against much of the economic trouble we find ourselves in now.

Some of the details may be too laissez faire for some too accept, but I think the core theory, especially as it relates to the theory of the business cycle is dead on. Friedrich Hayek won the Nobel prize in economics in 1974 for his work on the theory, and I find it amazing that he is completely ignored today.
I'm much more inclined to the Austrian school rather than the Chicago school. I find Hayek to be the most interesting economic scholar of the last half century. :)

And yes, I do believe that a greater appreciation of the Austrian school might have prevented the wild excesses of US economic management over the last twenty years, but that's giving too much credit to 'theories' and not enough to 'human motivations'.

Fact is, the US bad economic mismanagement was extremely profitable for many people for a long time. They had a vested interest in pretending that their theories were working fine.

[Dominick]

As far as the critique of existing economical theory concerns, I've laid it out several times already. The mathematical tools that are used are entirely deficient for the goal. And that fact is not a squibbling in the margin but the very essence of the issue. Any theory that is scientifically sound constitutes of nothing but mathematics. In fact that theory would be mathematics.

There is a fundamental difference between linear and non-linear systems. Linear systems are e.g. the acceleration of a car or the interaction of gravity between two bodies (and just two -it doesn't work for three bodies because such a system is already non linear). Those are intuitive and straightforward and when the actual mathematical function of the system is not known, perturbative (i.e. approximative) methods can and should be used.

Non-linear systems are fundamentally different. They have what is called a 'sensitive dependency on initial conditions' (ie. the butterfly-effect). Because of this sensitivity perturbative methods are useless. One might as well us a dartboard. Such systems can only be described by describing the whole. Needless to say it is far more difficult to achieve this when one has no intermediate -perturbative- steps. That's why the study of this is far more recent (a few decades) than that of linear systems (starting with Newton or even Euclid).

There can be no question that the global economy is a non linear system. It is riddled to the brim with feedback mechanisms which define non-linearity. For that reason a sound economical theory must be non-linear in toto. Current, classical economical theory is not.

When applying linear methods to non-linear systems, one will of course still get results. And as you say, they may 'more often than not' hit the mark. But that would also be the case with a set of random genetic algorithms that are left to compete. And that is essentially what economists are doing. They keep inventing new formulas until one hits the mark of today. And when the prediction for the next day is way off, they invent still another one. These are the methods that have been used to evaluate debt positions in the recent past. With the well known result. *

Only a theory that always hits the mark is meaningful (albeit with an error margin that all theories have) . It would have had to predict the credit crunch to be a serious theory.

Thus, as far as an alternative, I have none, and never pretended to. I have none because there is none and can't be one. We, mankind, at the moment lack the knowledge and tools to describe such a massively complex system. The ones in use today are purely ideological in nature and are scientifically invalid.

I know you won't like this explanation and would prefer one in terms of interpretation rather than fundamentals but interpretation doesn't matter when the fundamentals are wrong.

*Anecdote :
Other critics of economical theory have given a pack of chimpansees a big board with stock indicators on. The ones they hit or walked on the most were interpreted as 'buy', while the ones they never or rarely hit or walked on were interpreted as 'sell'. The apes had better results than the stock exchange advisors.

[wphelan]

I'm much more inclined to the Austrian school rather than the Chicago school. I find Hayek to be the most interesting economic scholar of the last half century. :)

And yes, I do believe that a greater appreciation of the Austrian school might have prevented the wild excesses of US economic management over the last twenty years, but that's giving too much credit to 'theories' and not enough to 'human motivations'.

Fact is, the US bad economic mismanagement was extremely profitable for many people for a long time. They had a vested interest in pretending that their theories were working fine.

Indeed, the more Hayek I read, the more I like.

And I completely agree that there has been too strong a vested interest in pretending the current system we have is ok. That's why I always cringe a little when I hear influential people say we only need to "restore faith" in our financial system. It seems to me there are problems at its core.

[Dominick]

A bit of delving into macro-economics and the work of Mr. Hayek doesn't alleviate the problem, rather the contrary :

First of all, I'd like to say that the very existence of the concepts micro- and macroeconomics points to a theoretical weakness. There is no such thing as macro- and microphysics for the simple reason that there is no discontinuity in the field of study. There is a continuous range over which the effects of physics are active. That is of course the same with economics. Any line drawn across the spectrum is necessarily arbitrary and artificial. Such arbitrary classifications are symptomatic of sciences that are in their infancy.

And that's precisely what I see in this field. When I read about it, I keep wondering "Where is the theory?" What I see is a rationalisation and justification of an ideology. There is no objective theory in sight.

Let's look at the list of his works (http://en.wikipedia.org/wiki/List_of_books_by_Friedrich_Hayek) :

Titles such as Monetary Theory and the Trade Cycle (1929) or Prices and Production (1931) are what one would expect in this context, but not titles such as The Political Ideal of the Rule of Law (1955) or Law, Legislation and Liberty.

Those are a clear indication that the work of Mr. Hayek is not independent of his political views. And that's a general problem with economical theory. There is one for each ideology. This critique would read no different if the subject were Marxist economical theory.

The obvious result of such a pick of theories that correspond to ideologies is that those that are fond of a certain ideology will support the accomodating economical theory. It's then no accident that e.g. Mrs. Thatcher subscribes to Mr. Hayek's views.

But a sound theory should of course be entirely free of politico-ideological contamination. After all there is no classical-liberal relativity or Marxist string theory. I could also give examples from less clear cut fields, such as geology, biology, etc.

A theory has to explain what happens, formalize it, quantify it in such a way that the theory can be tested by predictions, by quantified predictions. I see none of that in this work. I want facts, figures, numbers, formulas, not rhetoric.

On the contrary, when one dives into the particulars of this model, one of the things one encounters is the Kiyotaki-Moore (http://en.wikipedia.org/wiki/Kiyotaki-Moore_model)model.

More specifically, in the model of Kiyotaki and Moore (1997), two types of households (and firms) with different time preference rates are assumed: "patient" and "impatient."
This appears to be heralded as a major improvement. The completely diverse reactions of 6,5 billion people are now represented by a variable with two values, "patient" and "impatient". Having a binary variable representing a continuous, possibly multi-dimensional aspect of a system is hilarious at best.
I couldn't make up a better example of underachieving tools.

And let's have a case study to evaluate the explaining and forecasting strength of the model:
http://newsimg.bbc.co.uk/media/images/45146000/gif/_45146024_light_crude_226.gif

Could anyone quantitatively explain this graph ? Not with mere words but with figures and formulas ? A link to such an exhaustive explanation will also do.

[wphelan]

A bit of delving into macro-economics and the work of Mr. Hayek doesn't alleviate the problem, rather the contrary :

First of all, I'd like to say that the very existence of the concepts micro- and macroeconomics points to a theoretical weakness. There is no such thing as macro- and microphysics for the simple reason that there is no discontinuity in the field of study.

Physics isn't my best area, but isn't there a disconnect between the general theory of relativity and quantum mechanics? It seems like the micro/macro analogy is actually quite fitting.

As far as the rest is concerned, are you saying that no theory is meaningful unless it can be expressed in purely mathematical form? I agree that we don't have the capabilities to adequately express macro-economics in a purely quantitative manner.

However, I will compare macro-economic theory to evolutionary theory. There doesn't seem to be any quantitative expression to adequately describe either one, but that doesn't mean they can't be useful theories. Evolutionary theory, like macro-economic theory, gives us meaningful and useful explanations of overall trends. I don't think we need to quantify these theories to make them useful.

[Dominick]

Physics isn't my best area, but isn't there a disconnect between the general theory of relativity and quantum mechanics? It seems like the micro/macro analogy is actually quite fitting.
There is a problem yes, but that's not one of scale technically. Both theories still describe everything, literally, on any scale. Quantum mechanics for instance is often viewed as the realm of the real small but it has many implications on an astronomic level.


As far as the rest is concerned, are you saying that no theory is meaningful unless it can be expressed in purely mathematical form? I agree that we don't have the capabilities to adequately express macro-economics in a purely quantitative manner.
Not meaningful is not a claim I would make. Not objective and not scientifically sound on the contrary are some of my claims, yes. And that is sufficient to categorize macro-economics or economics tout court amongst some real bad company.


However, I will compare macro-economic theory to evolutionary theory. There doesn't seem to be any quantitative expression to adequately describe either one, but that doesn't mean they can't be useful theories. Evolutionary theory, like macro-economic theory, gives us meaningful and useful explanations of overall trends. I don't think we need to quantify these theories to make them useful.
Evolutionary theory has a much better track record than that. It has many quantitative aspects and ditto subdomains or is based on those of other, related domains (geology, chemistry, physics, etc.). C14 dating e.g., genetic laws, animal morphology studies, etc.

The explanations that economics provide are all hineininterpretierung. A decent theory, of any nature, needs to make accurate predictions. For instance "What will be the price of crude oil on 2009/7/1". The answer needs to be given in the format x ± y. That error margin (y) may be large, that doesn't matter, but the actual result will need to fall in the range and there has to be a theoretical -mathematical- basis to the prediction.

[partofme]

A bit of delving into macro-economics and the work of Mr. Hayek doesn't alleviate the problem, rather the contrary :

First of all, I'd like to say that the very existence of the concepts micro- and macroeconomics points to a theoretical weakness. There is no such thing as macro- and microphysics for the simple reason that there is no discontinuity in the field of study. There is a continuous range over which the effects of physics are active. That is of course the same with economics. Any line drawn across the spectrum is necessarily arbitrary and artificial. Such arbitrary classifications are symptomatic of sciences that are in their infancy.

And that's precisely what I see in this field. When I read about it, I keep wondering "Where is the theory?" What I see is a rationalisation and justification of an ideology. There is no objective theory in sight.

Let's look at the list of his works (http://en.wikipedia.org/wiki/List_of_books_by_Friedrich_Hayek) :

Titles such as Monetary Theory and the Trade Cycle (1929) or Prices and Production (1931) are what one would expect in this context, but not titles such as The Political Ideal of the Rule of Law (1955) or Law, Legislation and Liberty.

Those are a clear indication that the work of Mr. Hayek is not independent of his political views. And that's a general problem with economical theory. There is one for each ideology. This critique would read no different if the subject were Marxist economical theory.

The obvious result of such a pick of theories that correspond to ideologies is that those that are fond of a certain ideology will support the accomodating economical theory. It's then no accident that e.g. Mrs. Thatcher subscribes to Mr. Hayek's views.

But a sound theory should of course be entirely free of politico-ideological contamination. After all there is no classical-liberal relativity or Marxist string theory. I could also give examples from less clear cut fields, such as geology, biology, etc.

A theory has to explain what happens, formalize it, quantify it in such a way that the theory can be tested by predictions, by quantified predictions. I see none of that in this work. I want facts, figures, numbers, formulas, not rhetoric.

On the contrary, when one dives into the particulars of this model, one of the things one encounters is the Kiyotaki-Moore (http://en.wikipedia.org/wiki/Kiyotaki-Moore_model)model.


This appears to be heralded as a major improvement. The completely diverse reactions of 6,5 billion people are now represented by a variable with two values, "patient" and "impatient". Having a binary variable representing a continuous, possibly multi-dimensional aspect of a system is hilarious at best.
I couldn't make up a better example of underachieving tools.

And let's have a case study to evaluate the explaining and forecasting strength of the model:
http://newsimg.bbc.co.uk/media/images/45146000/gif/_45146024_light_crude_226.gif

Could anyone quantitatively explain this graph ? Not with mere words but with figures and formulas ? A link to such an exhaustive explanation will also do.

I think Charles Darwin said it best:

“A scientific man ought to have no wishes, no affections, - a mere heart of stone.”

[Michael]

:fence:

A bit of delving into macro-economics and the work of Mr. Hayek doesn't alleviate the problem, rather the contrary :
Actually, I think micro/macro marks an important theoretical distinction here.

First of all, I'd like to say that the very existence of the concepts micro- and macroeconomics points to a theoretical weakness. There is no such thing as macro- and microphysics for the simple reason that there is no discontinuity in the field of study. There is a continuous range over which the effects of physics are active. That is of course the same with economics. Any line drawn across the spectrum is necessarily arbitrary and artificial. Such arbitrary classifications are symptomatic of sciences that are in their infancy.

And that's precisely what I see in this field. When I read about it, I keep wondering "Where is the theory?" What I see is a rationalisation and justification of an ideology. There is no objective theory in sight.
Of course there is no actual or specfic "objective scientific theory" of economics, any more than there is one for history, psychology, politics, sociology, religion, anthropology, archeology or any other human science - and likely never will be.

That's why 'science' has always been academically distinguished between the "arts" and the "sciences". One is concerned with the study of human beings, the other with inanimate and/or non-human animate objects (generally speaking).

It is logically proper to draw this distinction. Theories about inanimate objects (such as the physics of spatial bodies) are of one kind. Theories about human activities are of another.

Let it also be noted that there are many in the 'arts' who seek to claim that economics, politics, religion, anthropology and sociology are "social sciences" and are the same thing as "hard sciences". I consider this nonsense. For me, there are the "arts" and there are "sciences". There is some crossover, but there will always be a distinction here - it is all about that unpredictable human subjective 'freewill' thing.

Those are a clear indication that the work of Mr. Hayek is not independent of his political views. And that's a general problem with economical theory. There is one for each ideology. This critique would read no different if the subject were Marxist economical theory.
Absolutely. All economic theories are political - or human - by definition. I doubt if they will ever get effectively functional equations when modeling human behavior. The error margins will always be just too large.

The obvious result of such a pick of theories that correspond to ideologies is that those that are fond of a certain ideology will support the accomodating economical theory. It's then no accident that e.g. Mrs. Thatcher subscribes to Mr. Hayek's views.
Yes, humans are often inclined to act like that. That's always the 'spoiler' in every theory involving human society in all of its forms. Humans have subjectivity and/or free will and an annoying habit of using them (usually only in rare and unexpected ways). :)

But a sound theory should of course be entirely free of politico-ideological contamination. After all there is no classical-liberal relativity or Marxist string theory. I could also give examples from less clear cut fields, such as geology, biology, etc.
There are no rocks, stars, planets or moons that claim to be Marxists or act like idiots.

Now if planets could subjectively choose their own orbits on a whim, I suspect quite a few laws of physics might have to be adjusted.

A theory has to explain what happens, formalize it, quantify it in such a way that the theory can be tested by predictions, by quantified predictions. I see none of that in this work. I want facts, figures, numbers, formulas, not rhetoric.
As you noted above in your earlier post here, there just are too many subjective variables when humans are involved for that stuff to work with anything above modest 75-85% probabilities (I'm just guessing this, I have no stats or data on it). That kind of prediction rate is pretty good for a human science, utterly useless for physics or chemistry.

On the contrary, when one dives into the particulars of this model, one of the things one encounters is the Kiyotaki-Moore (http://en.wikipedia.org/wiki/Kiyotaki-Moore_model)model.

This appears to be heralded as a major improvement. The completely diverse reactions of 6,5 billion people are now represented by a variable with two values, "patient" and "impatient". Having a binary variable representing a continuous, possibly multi-dimensional aspect of a system is hilarious at best.
I couldn't make up a better example of underachieving tools.
This game has been going for a long time and will continue long into the future. Indeed, it seems to be driven by exactly the same drive as the argument you are making here.

You apparently want to reject the whole of all the learning in the arts, humanities and social sciences because none of those theories will ever fit the strict definition of a 'theory' under the scientific method?

Well, guess what? There are some in the social sciences who seek to elevate the social sciences to exactly the same status as the hard sciences and to do this they use fancy mathematical models and speak of equations. And they produce gibberish nonsense.

I certainly have never denied that. Mathematical mumbo-jumbo is for the physics geeks and chemistry nerds. Let them have it. For the study of the human arts, mathematical theories and statistics are at best, only a supporting tool.

Could anyone quantitatively explain this graph ? Not with mere words but with figures and formulas ? A link to such an exhaustive explanation will also do.
I'm sure some whizbang physics or math-dude from Princeton might with some really nasty looking equation...

But in plain English, it is a perfect description of the benchmark market price of a barrel of (West Texas / Sweet Brent Crude) oil on the world spot markets over the last 12 months. That makes sense in a way that no math equation will.

And all that is why the distinction between micro and macro is significant. The funky math is most often being used on the micro level in economics. On the macro level, there are just 'broad general principles' that form the essential theory ('law of supply and demand' being one of them) - which is of course, political.

[Michael]

“We have seen the triumph of sensible ideas and have reaped the rewards in terms of macroeconomic performance,” Romer concluded. “The costly wrong turn in ideas and macropolicy of the 1960s and 1970s has been righted and the future of stabilization looks bright.”

She was, as we now know, wrong about the happy ending. In retrospect, what is striking about Romer’s lecture is not the chastened tone of its opening but the celebratory nature of its conclusions. “Better policy, particularly on the part of the Federal Reserve, is directly responsible for the low inflation and the virtual disappearance of the business cycle in the last 25 years,” she said (emphasis added). “In this area, the policy mistakes of the 1960s were a painful, but not permanent, detour on the road to excellent economic performance.”
Source (http://www.theatlantic.com/doc/200904/economic-policy)

What we have here is an admission by a leading US economist that for the last 25 years, the US Federal Reserve and US regulators have believed that they had transcended classical macroeconomic theory.

The bottom line here is that the failure of Greenspan and US free market fanatics was at the expense of classical macroeconomic theory, not because of it. What drove US deregulation policies was not classical macroeconomic theory.

Indeed, classical economic theory was the one that has been predicting a major financial crisis for quite a while, based on the unsound market fundamentals in the US economy. Classical macroeconomic theory correctly called the dot-com bubble and classical macroeconomic theory correctly called the housing bubble - and the extreme danger posed by the negative US savings rate and the negative US balance of payments deficit.