This is an excellently written piece in several ways, which should be
extended. From this stand point, a progressively more functional economics
can be defined. A side point is that communication is fully a multiple
process. and that the person (s) attempting to convey shares
responsibility in having a "text" or body(unit) of knowledge be
understood. Academic economists have as a profile been well insulated from
that responsibility, which abandons the process to dys-information and
persuasion/threats. An interesting side note is that where the smell of
economics and testosterone is strong, there are also several domains which
have been largely abandoned, monetary economics and economic history being
two examples. This is on top of the near black (worse than just dismal)
out on institutional economics/macroeconomics as anything other than the
fiction extrapolated from Milton Friedman's "positive" micro economics.
Georgist economics (cf Henry George 19th century commons focused),
Thorstein Veblen, and others being outside of that conformity.
as we go, Tadit
Economics Debunked: Chapter Two for Sixth Graders
Posted: 22 Sep 2011 01:35 AM PDT
Readers gave high marks to Andrew Dittmer’s summary of a dense but very
important paper by Claudio Borio and Piti Disyatat of the BIS and asked if
he could produce more of the same.
While Andrew, a recent PhD in mathematics, has assigned himself some truly
unpleasant tasks, like reading every bank lobbying document he could get
his hands on to see what their defenses of their privileged role amounted
to, he has yet to produce any output from these endeavors that are ready
for public consumption.
However, I thought readers might enjoy one of Andrew’s older works. He
e-mailed me right after I started working on ECONNED. Our conversation
went something like this:
>
> Andrew: I am a Ph.D. student in mathematics at Harvard. I have found
> your blog to be terribly interesting while trying to make sense of the
> way in which the global >economic architecture is evolving, or devolving
> as the case may be.
>
> I should finish my doctorate in a month, assuming nothing goes terribly
> wrong. I would be very happy to help you with your book in whatever ways
> are useful for >you. In particular, I would definitely be able to review
> and comment in a fairly minute and exacting way on chapters of your book.
>
> Me: That is a very kind offer, thanks! What is your dissertation on,
> BTW? And are you doing theoretical or applied math?
>
> Andrew: Theoretical math
> Me: [inwardly] Holy shit, this is one of the smartest people in the
> country!
It turns out that Andrew did indeed provide detailed commentary on the
logic of all the chapters, as well as the thread of the argument across
the book after it was drafted (something few editors do). Chapter 2 was
the most daunting chapter, in that it is a fundamental critique of
economics from a methodological standpoint. Every time I though I had a
draft nailed, I’d get extensive feedback from Steve Waldman and Andrew,
with the net result that I’d go back and start from close to scratch. This
happened eight times and I was clearly losing patience.
Andrew, recognizing that I was starting to lose it (writing an ambitious
book in six months was rather deranged, and all the revisions were adding
a lot of added pressure) suggested how to restructure the chapter in the
form of a lesson guide for sixth graders (as mentioned earlier, Andrew
taught six and seventh graders math in the Cambridge public schools on the
side). It was useful and also provided some badly needed comic relief.
Those of you who have read ECONNED may recognize the parallels (I adopted
his outline in large measure). I am confident readers who have not read
the book will still enjoy his rendition.
* * * * * (HISTORY)
Once upon a time, about a hundred years ago, economics was different from
how it is today. Many famous people had thought about economics, including
Adam Smith, Ricardo, Marx, and others. Economics was not a perfect
science, and lots of people who thought about economics disagreed. But in
the last century, economists started to agree about two things: economics
should become as mathematical as it possibly could, and politicians should
listen to economists.
Economics at the time was not a very mathematical science, at least
compared to the way it is today. Economists used to write articles with
very few equations, or even without any equations at all. When did
economists start to use math more? The beginnings might have been toward
the end of the 1800′s, with some economists named Jevons, Menger, and
Walras. But a person who was particularly important was Paul Samuelson. He
comes into the story later.
In the 1920′s and 1930′s, there lived a great economist named John Maynard
Keynes who came up with new ideas about economics. Many people thought
that they made a lot of sense because they seemed to explain how
governments had managed to get their countries out of the Great
Depression. Politicians began to listen to economists more and more.
Someone named Lorie Tarshis went to Keynes’ classes, took notes, and made
a book out of it. The name of the book was “Elements of Economics.” It was
published in 1947. Back then, there were people who thought that communist
spies were everywhere. Some of these people, including William F. Buckley,
decided that Tarshis’ book was also part of a communist plot, and started
saying so very loudly to as many people as they could find. Many colleges
became afraid to use “Elements of Economics” as a textbook.
Meanwhile, Paul Samuelson was writing his dissertation, and he saw how
angry the people had gotten with Tarshis. He decided to make it so his
dissertation could not be attacked in the same way. So he did three
things, writing very “carefully and lawyer like.” First, he changed
Keynes’ ideas a little bit so that in Samuelson’s version of them, they
said that corporations in the capitalist system would always give
everybody a job as long as the government and labor unions didn’t bother
the corporations. That made corporations happy with him, so that they
didn’t think people should call him a communist. Second, he called his
ideas “neoclassical synthesis Keynesianism.” That made other economists
think that he wasn’t changing their ideas very much, so they were more
happy to support him. Third, he wrote all of his arguments up in
mathematical form. Now why would he want to write up his ideas as
mathematical arguments?
Well, for one thing, a lot of people don’t understand complicated math, so
he could automatically win arguments with someone who didn’t understand
math. Even the people who went around calling people communists couldn’t
call him a communist if they didn’t understand the math he was doing.
There was another reason, too. Samuelson could turn problems that
economists used to argue about into equations. Then he would solve the
equations and the argument would be over. [Lucas quote]
Samuelson’s book was called “Foundations of Economic Analysis.” It was
published in 1947, and it became a big success. Nobody told Samuelson that
he was a communist. Other economists started to use mathematical models
more.
Later on, in the 1950′s, two economists named Arrow and Debreu made a
simplified mathematical model of the economy, and proved that in their
model, there would never be a time when people wanted to buy something and
nobody would sell it to them. Economists thought that this was very
exciting. Now they used math even more.
Nowadays, most papers on economics have equations in them. All economists
have to be able to talk about their ideas using mathematical models and
statistics, or other economists won’t respect them. This is part of what
is called being able to “think like an economist.” It is a special ability
that you can only learn by going to graduate school in economics. A
professor named David Colander surveyed economics graduate students to
find out which of seven factors were most important in order for them to
succeed in graduate school. The top three factors all had to do with being
good at math. The least important factor was having “a thorough knowledge
of the economy.”
So learning how to “think like an economist” is very important, because if
you don’t do it, then no matter how well you understand the actual
economy, economists still won’t take your arguments seriously.
Have these changes made economics a better science, or a worse science?
* * * * * (CRITICISM OF AXIOMATIC MODELS)
Samuelson, by using mathematics, was able to win arguments and make sure
people liked him, especially the more important people. Other economists
followed his example. How did Samuelson use mathematics to become
successful? The first thing he did, if you remember, was that he made sure
that his economics would say that corporations would make sure people had
jobs, and so corporations were good. So his first technique was to make
sure that his economics said things that people would like. But that
doesn’t seem to make sense. Isn’t the point of mathematics that in math,
things are either right or wrong?
Actually, the way that things work in math is that first, you make
ASSUMPTIONS, and then you figure out from the assumptions whether things
are right or wrong. So the way that you use a mathematical argument to say
things that people will like is that you change the assumptions until they
make it so the answers are answers that people like.
[At this point the kids become angry. "That's stupid, " they say. "It
sounds like they're going in circles." The responsible teacher at this
point tries to avoid being too political with his or her young audience,
and tries to make vague excuses for the economists. However, the teacher
can't help recognizing that the students sort of have a point.]
Even though economists were supposed to be using mathematics so that they
wouldn’t have to argue any more, it doesn’t seem to help much if they can
still get any answer they want by just changing their assumptions around.
For example, [insert McCloskey quote about the A'/C'-theorem].
That means one way to figure out whether mainstream economics makes sense
is to see what the assumptions are, and to try to decide whether those
assumptions make sense. For example, what were Samuelson’s assumptions?
What were Arrow and Debreu’s assumptions?
Samuelson had one big assumption, that economists call ergodicity.
[Teacher pauses to give kids time to stumble over the word.]
When they say ergodicity, they mean that no matter what happens in the
world, in the end, everything will reach a point whether things stop
changing. That point is called the “equilibrium.” At the equilibrium,
everyone will end up with a certain amount of money. The amount of money
that everybody gets at the equilibrium depends on how talented they are,
and not on anything that happened before. So if you rob a bank, it won’t
matter because when you get to the equilibrium, if you’re stupid, you will
still have the same amount of money you would have had if you didn’t rob
the bank.
[A kid with disciplinary issues mutters, "This is bullshit. Why do we have
to learn this?" Other kids ignore him and try to take notes.]
What’s more, at the equilibrium point, everyone will have a job, everyone
will have lots of stuff, and no one will feel like there is any way that
America could be a better country.
[Eyes glaze over.]
Actually, what’s kind of funny is that in physics, if there are three
stars or planets and gravity pulls them around, what do you think happens?
They end up going into orbits around each other. But their orbits will be
different if they start out in different places. So it would be kind of
weird if an economy with millions of people doing all sorts of complicated
things always ended up in the same way, if three planets can end up in all
sorts of different ways depending on where they start moving from. But who
knows? Maybe the economists are right about ergodicity.
It’s actually even weirder than that. Because in physics, Poincaré figured
out one hundred years ago that even if you know where the planets start
out, if you’re wrong about how far apart two planets are by even an inch,
then as time goes on, the orbits that you think the planets will go on
will start to get more and more wrong, until there comes a point when the
orbits you thought the planets were going to settle on are totally
different from the way that the planets are actually traveling. So it’s
really hard to predict what will happen even to three planets, if you try
to look far enough into the future. But who knows? Maybe economists are
right about ergodicity, and in an economy if you measure things carefully
enough and are clever enough, you can figure out exactly what will happen
to the economy for the next one hundred years.
Then there were Arrow and Debreu, who proved that in their model people
who wanted to buy something would always be able to find someone who would
sell it to them. They had assumptions, too. They assumed ergodicity, like
Samuelson, but they also assumed other things. They assumed that everybody
in the world knows everything about everything that is being sold all over
the world. Also that you know the odds of whether it will rain on a
Tuesday in a thousand years. This is called “perfect information.”
Some people who don’t think like economists have made fun of economists
for making unrealistic assumptions like these. Those people seem to think
that if economics is based on assumptions like these, that maybe aren’t
true, then economics must not be a useful science. But these people don’t
know that in 1953, Milton Friedman destroyed all of their arguments with a
magical “get-out-of-reality-free card.” When you play this card, it makes
it so you’re not allowed to make fun of economists for basing their
theories on assumptions that aren’t true.
Friedman said that if you could use a theory to describe the world
correctly, it didn’t matter if your assumptions weren’t true. Actually, it
was even better if they weren’t true, because that would mean that your
theory was very, very clever!
That means to decide if standard economics makes sense, we need to see
whether it says things that are true, and we shouldn’t pay attention to
whether or not the assumptions are true. For example, economists say it’s
bad to pay the people with the worst jobs more money
[Kids snap out of their stupor. "What?" a kid asks, dazed.]
because if you do, then the people who give the poor people jobs will fire
some of them. A couple of economists named Card and Krueger tried to test
the theory and they announced that the theory was wrong, and you could pay
poor people more money and have it be a good thing. Other economists saw
that if Card and Krueger were right, then standard economics had to be
wrong, and they went into shock. They were sure that Card and Krueger had
to be wrong somehow.
[One kid asks, "But wait. I thought that Friedman said that the
assumptions of the theory weren't important, just whether it was true in
real life. If it wasn't true in real life, then it was a bad theory,
right?" Teacher tells the kid to wait, there isn't much time left in class
and there are a lot of other things left to discuss.]
Some of them tested what happened when you gave poor people more money and
said that no, the theory was right after all. Other ones tested it too and
said that Card and Krueger were right and the other economists were wrong.
So even though Friedman’s magic get-out-of-reality-free card sounds like a
cool thing, it’s actually really hard for economists to use it in real
life.
But even though the card might not really work, mostly people don’t argue
with economists and so they still use their theories and “think like
economists.” “Thinking like economists” is kind of like looking at the
world with 3-D glasses. When you’re watching a 3-D movie, it makes it so
that you see really neat things. When you’re not watching a 3-D movie,
then everything looks red and blue and kind of weird. But economists still
like wearing their 3-D glasses.
When economists look at the world through their 3-D glasses, they see it
as having “ergodicity.” Remember what that means? It means that if you
just leave corporations alone and help them to do what they want faster,
the world will, all by itself, become a happy place. You don’t need to
stop them from doing anything they want to do, or try to make them wear
seatbelts. Just help them to drive as fast as possible. It also means that
economists can figure out what the economy is going to do, and you can
trust them.
For example, economists have invented computer programs called “DSGE
models” that they use to predict the future. Because of ergodicity, the
DSGE models say that nothing bad will happen to the economy unless
something crazy happens, like the people in the Middle East not wanting to
sell us any more oil or Martians attacking the earth.
Another example is that some of the big banks invented really complicated
things that were sort of like money, but sort of not like money. Those
things are called “derivatives.” The big banks liked the derivatives
because they were sure they could make a lot of real money from them. Some
other people who weren’t economists thought that derivatives might be
dangerous. Those people thought that if things in the economy went faster,
they might also break more easily. But because of ergodicity, economists
were sure that since derivatives helped corporations do things that they
want to do faster, the derivatives would be good. So the economists made
it so nobody paid attention to the people who said derivatives were
dangerous, and the big banks got to make all of the derivatives they
wanted to.
Later, the derivatives helped to make trouble in the economy. That’s why
some of your parents lost their jobs. The big banks who made all the money
from the derivatives had trouble too and were almost destroyed, but the
government gave them more money so that nothing bad would happen to them.
Meanwhile, the government won’t let people find out what it did with the
money it gave to the banks. It says that the details need to be kept
secret by a group of bankers and economists called the Federal Reserve.
The people on the Federal Reserve think like economists and so it’s okay
for them to know the secret.
* * * * * (OTHER APPROACHES TO ECONOMICS)
There have been some people who don’t like the economics that starts with
assumptions and then tries to do math with the assumptions. [you could
cite Blaug here] Some of these people have tried to make other kinds of
economics.
Some people tried to make a kind of economics that is called “systems
dynamics.” In this economics, you pretend like the economy is a really big
machine, and sometimes parts of it can go crazy or break. Some people
liked this kind of economics, including some of the people who said that
derivatives were dangerous. But economists mostly don’t like this kind of
economics, so they don’t use it.
One day, some economists noticed that if you’re playing cards, if you peek
at someone else’s hand, then you’ll probably win more than someone who
doesn’t peek. That’s because you know your cards and their cards and they
only know their own cards. The economists who noticed this called it
“asymmetric information.” Since economists before that assumed “perfect
information,” so everybody playing cards knows everybody else’s cards,
they were amazed at how smart these economists were and gave them Nobel
Prizes.
Another day, some economists noticed that sometimes people are stupid and
do things that waste money. They made a theory about this called
“behavioral economics.” Since economists before that assumed “perfect
information,” [and rational expectations] or in other words, people know
everything that is happening everywhere in the world and always do
whatever makes the most money, they were amazed at how smart these
economists were and gave them Nobel Prizes.
There were also some economists who noticed that if you give another kid
your lunch and tell him to hold it for you until lunch, he might eat your
chocolate bar and then tell you that someone stole it. Their theory is
called the “principal/agent dilemma.” This theory also seemed very new and
exciting to other economists.
When people started making fun of economics because economists hadn’t
realized that there was going to be an economic crisis, some economists
like Eichengreen and Rodrik told those people that they were wrong and
economists could have been able to know that there was going to be a
crisis. Eichengreen and Rodrik said that the only problem was that
economists hadn’t used the new theories like asymmetric information,
behavioral economics, and principal/agent theory, but economists would
remember and use them next time.
But since all of the new theories are different from the old economics,
what usually happens is that economists use only one of them at a time. If
they got rid of the old economics completely, then other economists who
like the old economics would be angry at them or not pay attention to
them. So they use the old economics and then add on a little bit of the
new economics and hope that it works. An economist named Peter Dorman said
that if the old economics is like a big giant, then each thing that is
wrong about the old economics is like a wound that blood is pouring out
of. Each new kind of economics is like a band-aid that economists try to
put on one of the wounds, but they can’t put band-aids on all of the
wounds at the same time. So the giant lumbers forward, blood spurts out of
him all over the place, and nothing changes. Gruesome, huh?
Some economists have tried to use a lot less theories and mostly just
figure out what is actually going on in the world. This kind of economics
is called “empirical research.” For example, when Card and Krueger tried
to figure out what would happen if you paid more money to people with
crappy jobs, that was empirical research.
But when they did it, a whole lot of people got angry with them and
disagreed. So it can end up being pretty hard to tell what is going on in
the world.
There are a few reasons why this is true. For one thing, the way
economists usually try to find out what is going on is by looking at some
numbers or graphs and trying to find a pattern. But the numbers might not
be right. And what happens if someone wants to study something and they
can’t find any numbers to study it with? For example, some economics
students decided they wanted to find out if the trouble with the economy
had to do with making it so companies that try to get people to buy houses
didn’t have to follow as many rules. But those companies wouldn’t give
them any of the numbers they needed to find out if their idea was right.
So the economics students gave up. They didn’t have to give up – they
could have talked to people who bought houses and interviewed them and
things like that. But it would have been more work and other economists
might have thought that they were weird to use interviews instead of
numbers. So instead they gave up.
This is kind of like the story about a drunk guy who loses his keys and
walks over to the street light and looks under it. Somebody asks him why
he’s looking under the light when he probably lost them some other place.
He says that it’s dark in the other places so it’s hard to look there. Do
you see the connection with the economics students who wanted to study
houses? They couldn’t find the numbers that made it easy to look at their
problem, so they stopped looking.
[In fact, the connection between the joke and the problems with the
mortgage lender deregulation research is the one idea here that is
abstract enough that it would be tricky to explain to sixth graders.]
Another problem is that if you look at enough graphs, you’ll eventually
find one with a pattern just by chance. This is bad, because it might not
be a real pattern. It might just be an accident. If you take a fake
pattern and make people think that it’s a real pattern, that’s called
“overfitting the data.” It’s kind of like cheating.
If you don’t want to cheat and overfit, there are some ways to make it
more likely that you’re finding a real pattern and not a fake pattern. If
you find a pattern one year, you can look at another year, or another
place, and see if there is the same pattern. This is called
“cross-validation.”
Even though it’s a good idea to do cross-validation, a lot of economists
don’t do it. A couple of people named Gerber and Malhotra did detective
work on economics papers, and they found out that lots of economists were
probably overfitting. They couldn’t tell who was doing it, just that a lot
of people were doing it. If you’re an economist, you want to have a good
job, and to get a good job, it helps to find patterns and write papers
describing the patterns to other people. So some economists maybe wanted
to get a better job and so they overfitted so they could find more
patterns.
* * * * * (STATUS AND FUTURE OF ECONOMICS)
Economics seems to be in a lot of trouble right now. The old economics has
problems, and the new kinds of economics have problems too. Some
economists have even given up studying the economy and now study things
like speed-dating and violence in movies.
A guy named Thomas Kuhn said that when people make a science, they keep
using it as long as they can. Sometimes they can tell the science isn’t
working very well. This is called the “late-paradigm” stage. It sort of
means that the old science has become sick. Even then, people will only
stop using the old science when someone invents a new better science AND
when all of the professors who liked the old science get old and die.
It looks like economics is in a “late-paradigm” stage. But people don’t
have a new better economics, so people still keep using the old economics.
What are some things that should happen?
>
> (1) Economists should be honest about when they don’t know what will
> happen in the future so that people don’t rely on them in ways that they
> shouldn’t.
> (2) Economists should admit that in economies some people want some
> things to happen and other people want other things to happen. They
> should be honest about >what kind of world they want to live in, and not
> pretend like they know how to find a world in which everybody will be
> overjoyed.
> (3) Economists should work less at trying to find reasons not to listen
> to people, and try harder to learn about the economy, even from theories
> that they don’t like, >methods like interviews that don’t involve
> numbers, and from the ideas of people who are not economists.
