Everyone’s talking about this Paul Krugman essay on where economics, as a discipline, went wrong. Partly, “going wrong” means the failure to appreciate the risks in our financial system, and the corresponding failure to predict the crash we’re currently trying to deal with. But from an insider’s perspective, something else has happened: an uneasy consensus between two different approaches to economics has been shattered. One, which Krugman labels the “saltwater” approach, is associated (in the U.S.) with some variety of post-Keynesian analysis, generally identified with some willingness to have the government intervene in the economy over and above the Fed’s control of the money supply. The other, the “freshwater” approach favored by the Chicago School and other inland economists, is more purely free-market and non-interventionist. (Truth in advertising: I am not an economist.) These camps could more or less get along when everything was going fine, but have dramatically different reactions to a crisis. To the extent that you find freshwater economists claiming that unemployment is currently high, not because there aren’t many jobs, but because there are too many incentives for people not to work.
One of the reasons it’s a great essay is that it’s a wonderful example of popularizing science. You can debate all you like about whether economics counts as a science, but there’s little doubt that Krugman does an amazing job at explaining esoteric ideas in non-technical language, and is so smooth about it that you hardly realize difficult ideas are even being discussed. I wish I could write like that.
One part of the essay worth commenting on, or at least musing about, is the punchline. Krugman thinks that a major factor leading to the failures of economics to understand the mess we’re currently in was the temptation to think that beautiful models must be right.
As I see it, the economics profession went astray because economists, as a group, mistook beauty, clad in impressive-looking mathematics, for truth. Until the Great Depression, most economists clung to a vision of capitalism as a perfect or nearly perfect system. That vision wasn’t sustainable in the face of mass unemployment, but as memories of the Depression faded, economists fell back in love with the old, idealized vision of an economy in which rational individuals interact in perfect markets, this time gussied up with fancy equations. The renewed romance with the idealized market was, to be sure, partly a response to shifting political winds, partly a response to financial incentives. But while sabbaticals at the Hoover Institution and job opportunities on Wall Street are nothing to sneeze at, the central cause of the profession’s failure was the desire for an all-encompassing, intellectually elegant approach that also gave economists a chance to show off their mathematical prowess.
Without knowing much of anything about the relevant issues, I nevertheless suspect that this moral might be a bit too pat. Sure, people can fall in love with beautiful theories, to the extent that they overestimate their relationship to reality. But it seems likely to me that the correct way of understanding all this, once it’s properly understood, will look pretty beautiful as well. General relativity is widely held up as an example of a beautiful theory — and it is, when understood in its own language. But if you put the prediction of GR in the Solar System into the language of pre-existing Newtonian physics (which you could certainly do), it would look ugly and ad hoc. Likewise, Newton’s theory itself is quite elegant, when phrased in the language of potentials on a fixed spacetime background; but if you express the theory in terms of differential geometry (which you could certainly do), it looks like a mess. Sometimes the beauty/ugly distinction between theoretical conceptions is more a matter of how well we understand them, and less about their intrinsic qualities.
So my counter-hypothesis would be that it wasn’t beauty that was the problem, it was complacency. If you have a model that is beautiful and works well enough, you’re tempted to take pride in it rather than pushing it to extremes and looking for problems. I suspect that there is a very beautiful theory of economics out there waiting to be developed, one that understands perfectly well that individuals aren’t rational and markets aren’t perfect. One that has even more impressive-looking equations than the current favored models! Beauty isn’t always a cop-out.
Beauty isn’t always a cop-out.
Ah yes, but I suspect that it is more of a cop-out when it comes to economics, which works with ridiculously heterogeneous fundamental units of analysis (individual humans) in environments too complex for the methods of the (more truly scientific) natural sciences like physics to be practical tools. No wonder so many financial journalists blame the attitudes that led to the crash on physicists coming to Wall Street.
I suspect not.
There will never be a single mathematical theory of economic behavior. Why? Because an economy is the means by which humans, or groups of humans act on the perceived value of materials and actions. So a mathematical theory that explains economic behavior would have to, by definition, predict the human perception of value.
There is however hope for solid empirical theories based on the work of psychologists and sociologists, in experiments that have elucidated the statistical connection between the demographics of people and their perceptions.
Mechanics of the solar system is a simple system in the sense that there are only a few interacting components (it is still chaotic). I guess that economics is in some sense like statistical mechanics, in that there are a huge number of components to be averaged over. However, we are safe to assume particles in stat mech are identical and have predictable properties. I think here our analogy with physics breaks down. Has anyone ever been able to model human behaviour sufficiently precisely? Humans do crazy unpredictable things, and I think this will always be a problem for economics.
From my understanding of Mandelbrot’s The (Mis)behavior of Markets, all these models use Gaussians. But most economic data doesn’t fit Gaussians, but long-tail curves like 1/(1+x^2). So, why do they use Gaussians? Because they can do the calculations with Gaussians. Bad idea.
Also, with supercomputers these days, I wonder if they couldn’t just accurately model the entire economy. It only takes about 300 million individual fundamental elements, along with additional things like corporations, governments, markets, transporation, etc. Surely, one could find a mix of characteristics of greed, fear, entrepreneurial spirit, inertia, etc., let the thing run, and see if it matches reality at all. And if not, vary the mix.
I suspect not as well. The laws of physics seem to be invariant with respect to time: we can, for example, model the evolution of the universe from very early times to the present and the results look very much like what we see today. I suspect that if the laws of physics changed with time this would not be the case.
Markets, however, are not stationary: participants innovate, learn and evolve (at least over a business cycle or two). Markets also aren’t closed systems: their behavior is subject to external events which might have nothing to do with the assets being traded in the markets at all, whether that’s varying levels of government regulation, events like 9/11 or something else entirely.
The fundamental error of economics as a science was trying to emulate the hard sciences and find an “elemental” factors on the basis of which a theory could be constructed. The elemental factors on which economics settled were rational expectations of a representative rational actor having perfect knowledge and participating in markets that are inherently efficient. This looks suspiciously more like the world of physics rather than the real world of people interacting socially, politically, and economically. Of course, it blew up with these asinine assumptions that have been disproved by cognitive and behavioral science. That’s just not the way human beings are wired, and it isn’t how they behave. Even the so-called law of supply and demand are not real laws as Keynes showed by introducing the concept of “stickiness.” Stickiness is similar to friction in physics, but there is no coefficient of stickiness.
The idea that now people are unemployed because of perverse incentives or laziness would be ludicrous if it weren’t so insulting. These professors should be required to get out the ivory tower and work at Walmart for awhile to learn what the real world is like, or live in a tent city.
Go to CSPAN and watch a Senate or House Committee meeting. Economics has to deal, in part, with this type of very unpredictable and often very irrational behaviour.
If you’re looking for beauty here, and one can find a sort of odd form of beauty here, it definitely won’t look like science. Much of the time, it doesn’t even look like our Constitution.
I think Tom is on the right track. I wonder how much of the problem with mathematical economic models comes from the fact that they are really sociological at their root. Predicting humans and when mass hysteria will kick in is to predict an incredibly complex thing that involves speed of communication, the scariness of the communicator, and what other things are in the news. I wonder whether economists sometimes loose track of how much is science and how much is human behavior.
THE THEORY EXISTS. SORRY TO SHOUT BUT IT SEEMS TO ME THAT THIS IS THE ONLY WAY TO REACH OUT TO PEOPLE BECAUSE MAINSTREAM ECONOMICS AS IT EXISTS IS A BIG DISGRACE.
There are these bunch of people mainly associated with the Levy Institute http://www.levy.org You can find it in the work of Wynne Godley and Marc Lavoie. They have written a beautiful book called “Monetary Economics: An Integrated Approach to Credit, Money, Income, Production and Wealth ” http://www.amazon.com/Monetary-Economics-Integrated-Approach-Production/dp/0230500552 Unfortunately you cannot preview the book on google books or any other site. However the following may be helpful.
I swear on my knowledge of General Relativity and Quantum Field Theory that this is what you may be wanting to see for Economics.
The subject is called Post Keynesian Economics. The logic of this subject is as tight as any branch of science. It does NOT assume “agents” being rational or maximising any profits. It divides economies into households, firms, banks and the government. Instead of agents maximising, it uses something contrary – money as buffer and products as buffer instead of inventories.
There are two starting points in the approach – government does not need to finance itself and issues bonds for interest rate maintenance and secondly “loans create deposits”
The first one is totally opposite of what a “neoclassical” would tell you. But in fact it is perfectly logical. Deficits run an economy. Without deficits, an economy would collapse. Any transaction in the private sector creates both assets and liabilities. The government spends and takes the liability on itself so that the private sector grows. I wish I could write more here 🙁
The second one is hard to believe it is contrary to a Loanable funds market which is what economists think exists. It seems to show that banks create money. The approach to study is called the “Engoneous money approach” In fact it is true. Again, I can write more on this but space is an issue here and this is a huge subject.
Can I recommend you visit my favorite blog – Billy Blog – the index is available at http://bilbo.economicoutlook.net/blog/?page_id=1667
I also recommend
http://neweconomicperspectives.blogspot.com/2009/08/endogenous-money-approach.html
http://neweconomicperspectives.blogspot.com/2009/08/primer-on-government-surpluses.html
the person’s name is Randall Wray and he has really figured out stuff in the most elegant manner.
(Continuing .. )
Sean,
The approach of Godley and Lavoie is the most logical approach to Economics. Firstly, they are consistent right from the start. They use – what they call – Stock Flow consistent approach. Stock is like stock of money, stock of debt etc. Flow is whatever happens in one period. e.g., consumption, investment (on capital goods), government spending for the period, tax for the period etc. (“I found out what economics is; it is the science of confusing stocks with flows” – Michal Kalecki)
They divide an economy into households, firms, banks and the government. Now ‘everything comes from somewhere and everything goes somewhere’ is their elegant conservation law. Its different from money being conserved because it is not (Econophysicists don’t get it unfortunately). This is the transactions flow matrix whose rows and columns sum to zero. There is also a balance sheet matrix which is the full balance sheet at the end of every period. Here too all rows and columns, except one row sums to zero. Thats the net worth of the economy – the “tangible capital” This is their “social accounting matrix”
Now, they never assume anyone is rational or that anyone maximises something. Instead e.g., firms have a target and their expectations are adaptive. Households too have adaptive expectations and money deposits are like “buffer”.
They start from “sratch” – everything comes from somewhere. The economy kickstarts by the governments spending and reaches firms and then households. Households consume and give the money back to firms and save as well. There is no market for “loanable funds” – banks lend and then look for reserves at the central bank. they do not use the money “given” by households.
One may ask – if the government is not constrained why does it tax – the answer is that it reduces “aggregate demand” else prices will rise. Now coming to prices, in Post Keynesian Economics, there is no market clearing mechanism (demand-supply) – producers keep enough supply so that the demand is satisfied. (Ever bargained with Steve Jobs). However if consumption is high, producers’ inventories will clear fast and instead of producing more they increase prices. Hence governments cannot solve all the problem at one go.
This is a very short write-up. Sean – I will be very happy if you or someone reading this gets interested. I know what you are talking of precisely, since I had been in physics myself, I can relate to your post.
Another important result of this school is that the so called “NAIRU” is a hoax. It is simply possible to have full employment unlike what an economist tells you. This is because of the way government spends in a modern economy which is different from the way it could spend in the Gold-Standard and the Bretton-Woods eras.
Thanks. again – I know precisely what you are talking of – one doesnt need a jazzy theory for economics with path integral summed over different topologies and Lagrangians but at some level you may want a theory to be very logical and yet be “elegant”. Post Keynesian Economics is such a subject. (Not to be confused with Keynesian or New Keynesian)
I end my comment here pointing to the best prediction of the crisis http://www.levy.org/pubs/sevenproc.pdf – Seven Unsustainable process by Wynne Godley – everything in that paper came true. It was based on the CBO projections and the “crash-landing” got delayed by 2-3 years because the Bush government ran a bit of deficit for a while. But it is simply the best prediction. Find me a better one and I will go to the Himalayas. (I do not write like this – not my personality but have to resort to this because Economists do not seem to understand simply accounting and only others can save the subject)
Howard Stern has beautiful models on his show all the time, and judging from their responses, I find it hard to believe that anyone would ever think they were right about anything…
It’s curious that the most obvious parallel with physics has not been mentioned: string theory, which certain influential physicists have been telling us for years must be true because of it’s inherent mathematical beauty. And string theory is also similar to freshwater economic theory in that it is rather un-predictive, and that virtually any set of observed facts can be incorporated post-hoc.
The above comparison is a bit unfair to string theory, however, in that it is still possible string theory is correct.
Government intervention is more than an alternate idea. Its a huge part of the economy that is impossible to model, even though it is way more predictable then human economic behavior.
As an economist, I agree more with Paul Krugman’s argument than with Sean Carroll’s.
It may be true that in the “long-run”, economic models that incorporate market imperfections can have just as much if not more mathematical rigor than economic models that assume markets work perfectly. Of course, as Keynes famously stated, in the long-run we’re all dead.
In the short-run, in general it is much harder to rigorously explain many market imperfections in the context of a model. If such a market imperfection cannot with present tools be rigorously modeled, you have two options:
(1) Simply assume the market imperfection is there and continue with the rest of the model;
or
(2) Assume the market imperfection does not exist (because we cannot explain why it should be there in a rigorous fashion), and go on with the model.
The second approach is perceived by some economists as more “rigorous” and more “scientific”. Other economists would disagree, and would argue that you need to include in your model a market imperfection that is empirically observed to exist. Just because you cannot rigorously explain something does not leave you free to ignore it.
So, I think Paul Krugman has identified a problem with some economists’ philosophy of what is the most scientific form of economics. I have no idea of how this methodological debate would be perceived by a physicist. I’m not even sure if there are analogous issues in physics.
Sounds like a perfect description of string theory……
Pingback: Tales from the Tubes — 5/09/09 | Young Australian Skeptics
It’s worth noting that there were several economists (and just plain observers as well) who DID foresee this collapse coming, but their timing was often off (some had been predicting it for 4-6+ yrs. before it finally came), and their voices were drowned out by the din of the Pollyannas. They saw through the false “beauty” that mesmerized others, and that Krugman accurately describes. Sean makes an interesting point in speculating that an even more beautiful mathematics may yet account for it all, though I’m not sure that math is even much required to explain the results of deregulation + human greed.
Pingback: 4 September 09 (pm) « blueollie
Have you ever read _Debunking Economics_?
It’s not a matter of complacency with a beautiful mathematical theory. The math of economics is a sick joke. It’s more a matter of forgetting that a mathematical model, no matter its beauty, is *useless* if it does not actually model.
And economic models don’t (model, that is). So Krugman is exactly right. If anything, he’s understating the case. Economists* want economics to be a “hard” science so badly that they’ll accept patent absurdities in their models, just so long as they get their math, so as to separate themselves from the unwashed masses of the social sciences.
And I’m not exaggerating — the holes in modern economic theory are insane. It’s as if General Relativity started out by positing: “we shall take it as given that no objects have mass…” and then building a nice model from there.
* Of course, not all economists. But there is really no economic discourse, outside of the broken kind, that gets any real play outside high academia, from what I see. And the kind of economic thinking used in policy-making is 100% of the broken variety.
Maybe there should be a meta-mathematical model for how applied mathematics is used, in economics in particular. Maybe there already is. If current maths has a difficult route ajusting to unpredicatble random variatons in a classical system, of all of the systems used, economics is the best one to study. It seems that in Economics, as an interesting and very relevant subject, it includes a lot of human behaviour, as mentioned here, combined with concrete systems. What about breaking this up into two parts. 1) Regardless of the subject, what can be learned from how we combine predictive math theory with ones less so. Then 2) In Economics, as an example, how it can be used. The reasoning relationship between the two parts could be too great though. Maybe that’s part of the problem.
Claire
Economics is not a science it is ideology. When you hear of an economic theory ask “cui bono” ? The other Nobel prizewinning liberal economist Joseph Stiglitz is quite open about that:
“there is no theory backing the doctrine of unfettered markets. That’s just an ideology that serves special interests”
Krugmann touches on it when he says.
“The renewed romance with the idealized market was, to be sure, partly a response to shifting political winds, partly a response to financial incentives. But while sabbaticals at the Hoover Institution and job opportunities on Wall Street are nothing to sneeze at…”
The looney mathematical models are not beauty, they are just propaganda.
The problem with GR is that it is a low energy approximation of large scale systems, and as such is perfectly useless for almost every human endeavor of immediate consequence. It does provide a boundary for relevant physics, but for the most part it can be effectively ignored (and don’t confuse GR with Riemannian geometry). As such, it represents a perfect failure of science because it gives the appearance of utility without actually requiring understanding of the universe.
Economists don’t know simple bank reserve accounting.
Is there a selection effect occurring, where people who suspect the fundamental laws of physics to be ugly and stochastic choose to leave theoretical physics?
Neoclassical calculus and geometry is the CNBC of mathematics. There is nothing beautiful or elegant about it. It is noise.
Algebra is the future of economics.
It’s also the basis of Post-Keynesian logic.