I spent two years at UChicago trying to learn some economics and I came away 
with the strong impression that:
 Economics is a profession of people who weren't good enough in math to 
 become physicists who go on to spend their careers trying to prove that 
 they were.  It is the ultimate pseudo-science.

So now I just started looking at the recent article by Paul Krugman in the NYT
and he seems to be saying something similar, more polite, but the same idea.
Funny, this comming from a person who has just got the nobel prize in economics.

My experiences trying to learn some economics shocked and amazed me.
Initially approaching the subject one has a tremendous curiosity about
how to explain the huge economic disparties that exist, why some societies
or periods in history are so materially successful and others not.
Courses in economics are very abstract and limited in scope.  One thinks about 
simplified mathematical models, but is hopeful that they will yield results.
Sometimes the things you seem to prove are very suprising, maybe there is
something important to understand here, something new.
But my experience was that the more you think about it what you understand
is that the result that suprises is actually wrong, it doesn't supprise
becuase there is something that you never noticed about reality 
(like in physics) but because it isn't reality.

Here is an example:
One of the most important topics covered in basic graduate courses in economics 
is Pareto efficiency.  One is taught the important "theorem" that a free market 
will generate a situation which maximizes overall satisfaction.
Of course a basic assumption here is that everyone in this system has
all the information, called "perfect information".

This theorem has a kind of comic genius about it. 
I would go as far as to say that when one thinks about it a bit one sees that
the defining characteristic of "real life" is the lack of information and 
very unequal distribution of the information that exists.
In the real world that is pretty much all you see, lots of people
with very different information.  
The knowledge barrier is the main barrier when it comes to almost every
human endevour.  
Why do people start bussiness of the type that they have had experience
working in? 
Why are all forms of education so expensive and important?
(even internships these days).
Why does no one have good experiences with home contractors?

Assume away the most potent ingredient and your going to get a theory that
tells you nothing.  Build all your understanding on top of this theory and,
I don't even want to imagine... 
unfortunately, I don't have to.

Someone got the nobel prize for this brilliant theorem.
People get Nobel prizes in economics for the most sensationally wrong theories.
Then 30 years latter other people get Nobel prizes for essentially explaining in the 
same general context of these wrong theories spots where they obviously do not work.
The truth is that after a few months thinking about these things I could see why.
Its easy to get drawn into this imaginary universe of models and theories,
there is a logical consistency to it and comfort in knowing the rules of the game.
Courses and curricula are orginized within this universe, it's how people write papers 
and how the people around you think.
And then small deviations from standard models, new theories that take in
some point that no one ( no one in econmics at least) was looking at 
start to seem brilliant. 

But when you take a fresh look at these things there really isn't much there.
To put it bluntly, you can get a Nobel prize in economics for writing down mathematically 
things that everyone who is not retarded and who hasn't studied economics already knows.

These theories aren't adding understanding, they are just recovering some of what
has been lost after the years of thinking about economic theory.

One recent example of Nobel prize winning material is essentially an explanation 
of how used car salesmen could possibly succeed at cheating people being that
markets are so great.  It turns out that if people only buy cars every once 
in ten years and not once per week the market doesn't work so perfectly.

Krugman himself got the nobel prize for explaining how it could be that
the conventional wisdom in economics - that if you are a country that is good
at growing mangos you are wasting time even thinking about making cars or
computers - sometimes isn't right.  The explanation being that maybe eventually 
you can produce enough cars or computers to do it efficiently.  

One gets that feeling that there was a time when the more formulas you could
write the more respect you got and the more true/real people considered 
whatever it was you had to say.  (The 1950's ?? If you're fuzzy you're a communist?)
And so in a bid for power and success practitioners of economics turned
to mathematicizing their research at the expense of intellectual integrity.
People who were critical or resisted this move, as evolution predicts, failed
to reproduce as much and whole generations were educated who didn't know 
anything else, generations of students who chose this field of work becuase they 
felt comfortable with this sort of logic, and were good at producing more or less, 
the same kinds of results they were taught.  

One wonders how teaching theories like these can be considered education
and how this education both selects for character and shapes the character
of those who recieve it.

In his recent NYT article Krugman draws lines between (neo)Keynsians and (neo)Cons,
he blames the (neo)Cons for being (neo)Cons and blames the (neo)Keynsians for
falling so far in line with the (neo)Cons.  He asks why this happend. 
How could all the theories have deviated so far from the economic reality?
But what he doesn't seem to see is that its not clear that there ever was any 
connection.

Reading "The wealth of nations" I was amazed at the delicacy of 
observation and acknowledgement of human folly, the implicit 
understanding that economics is part of the human sphere of action
in which absolutes don't exist. 

In contrast, I haven't seen a single example of the sort of mathematical "economic model" 
that modern economists invent which actually adds anything to ones' understanding 
of how reality works.
More often than not, they seem to obscure reality or even outright lie.
Whats worse this whole edifice of economic theory seems only to create a
kind of mental cage in which people can neither see nor understanding reality 
and certainly can't think about in the recent article in the recent article it critically.
Mainstream economics has evolved into a sort of subfield of mathematics with its own
insights and triumphs, and holy grails, but with no connection to reality and to 
the questions and problems which originally motivated its development.
Its practitioners are neither trained nor inclined to see things any other way,
their goal is to advance the mathematical sophistication and prestige of their subject.
And so one wonders how they got where they are (see the opening lines above...)