agent based modeling

/ / , , , , /

Limits

I don’t understand why people get upset when I say that economics is a waste of time. I suppose it’s because I don’t make a clear enough difference between economics as a general topic and economics as a formal, mainstream, body of knowledge. It’s the latter that is a waste of time. The former is wonderfully interesting.

At its heart economics is a study of human behavior, where that behavior is specific to certain activities. It is thus deeply rooted in psychology, so it is more closely associated with biology than physics. This is not a new idea: some of the greatest economists of the past have argued as much. Trying to transfer in ideas from physics, even metaphorically, therefore tends to lead to dead ends.

Like the notion of efficiency. That’s something of great interest to engineers, but has little to do with economics. You can have an efficient physical system. You cannot have an efficient social system. There’s just too much we don’t know and can never know. Still economists all over the world are obsessed with efficiency. So what do they do? They start to abstract and simplify. They model and fine tune. They test and re-test. And still their ideas run afoul of reality: human beings are not efficiency seeking machines, and so any system filled with humans is likely to be darned near impossible to steer towards efficient outcomes. Nothing daunted economists press on. If humans are unlikely to be efficient the logical next step is to construct a theory to exclude actual humans. That’s what’s happened in economics: the faulty decision to root economics in a physics-like setting rather than in a biology like-setting forced subsequent generations of economists to “refine” their thinking and, eventually, to force real people out of their theoretical world. Voila! Modern economics ends up as a wonderful edifice with extravagant claims as to its ability to understand human behavior precisely by eliminating all contact with humanity. Weird.

Ergo, if your goal is to understand real economies replete with real humans, modern economics is a waste of time.

Go study something else. You can learn a great deal about real economies by reading psychology literature. Behavioral economics — which despite all the press it gets has had only a marginal impact on the mainstream and on textbook economics — is an attempt to do that. The behavioral economics project is in its infancy. Go get involved.

By the way: anything that refers to strategic behavior is also useful. Real humans are constantly trying to outwit each other. That’s when they’re not cooperating, which is another human characteristic economics determinedly overlooks.

Humans are complicated. Too complicated for an economics built on an exclusive belief in relentless rationality.

Furthermore, many psychologists are now challenging the entire premiss that humans are calculating machines in the manner of computers. Our ability to compute sits uneasily with a far more powerful and deeply rooted set of survival skills that are experiential and heuristic. We are not rational information processors. We cobble things together. We try things out. We are really good at pattern recognition. As calculators we, well, suck. Yet economics bases itself on the notion that we are computational wizards. There is nothing in modern psychology to support this idea, but it persists in economics. Indeed economics is proud of its ersatz psychology and exports its “insights” into other social sciences with imperial abandon. It ought not. Its insights are embarrassingly rotten.

Worse: so proud is economics of its [bad] understanding of human behavior that it insists all economic theorizing be based on it. Thus we have the insistence on building macro theory on micro-foundations. At the risk of upsetting thousands of dedicated and well-meaning economists: garbage-in garbage-out just about sums up the micro-founded enterprise.

There is in evolutionary biology, as I am sure you all know, a notion called a “fitness peak”. Roughly speaking, this is the idea that a species refines its survival strategy and succeeds in a certain context. As it does so it “climbs” higher up a successful strategic slope that defines it relative “fitness”. Problems arise when the context subsequently changes. The more superbly refined fitness becomes, there more difficult it is to go back down and climb a different and more relevant slope. Yesterday’s cleverness becomes todays’ ignorance. The advantage of the old peak becomes a hindrance in the newer landscape.

So it is with economics.

It began in the context of the mid and late-1800’s obsession with physics. It’s greatest critique is situated in mid-1800’s industrialization. Economics began its climb up its own fitness peak back then. But the context has changed. Science has moved on. Economics hasn’t kept up. Instead it keeps on refining its obsolescence. It keeps on getting more and more specialized at something that is less and less relevant.

It needs re-doing.

Which is what a lot of people are working on: socio-economics; behavioral economics; evolutionary economics; experimental economics; institutional economics and many other fields are veritable hives of activity, they just haven’t made it into the textbooks.

Meanwhile what we call mainstream economics is, indeed, a waste of time.

 

Addendum:

This article in Aeon is a good example of the kind of thinking we need to import into the new economics mainstream. 

/ / , , , , , , , , , , , , , , /

The State of Economics. Hint: Bad.

This next weekend Paul Krugman has an article in the New York Times that I think it is well worth your time reading. It is called: “How Did Economists Get it So Wrong?”.

I have inveighed here, in my own inadequate way, many times about how I see the failure of economics as a body of both theoretical and practical thought as one of the main causes of our recession. Krugman, naturally doesn’t go that far, but that’s because he’s knee deep in the intellectual mire.

“It wasn’t simply greed: it was intellectually sound and risk adjusted greed.”

The simple fact is that economics as a body of thought has let us down with a thump still resounding around the hallowed halls of the University of Chicago [and Minnesota] whose intellectual tradition in the subject has now collapsed. Or rather has self-immolated.

This may seem to most of you to be a peripheral and dull thing to talk about, but I would urge you to reconsider.

Most people who are engaged enough to discuss the state of the economy bring with them a set of unstated assumptions or ‘truths’ about it. I hear statements every day such as ‘markets self correct’ or ‘markets are more efficient than government’ and so on. These are ideas bedded in a particular economic theory. They are the result of hard fought intellectual battles that extend back through time all the way to the pre-capitalist years of the late 1600’s and early 1700’s. They are not infallible, yet they are kicked around carelessly as if they were eternal truths.

They are not.

And the thrust of the Krugman article is to describe the more recent aspects of the fight between two major competing schools of thought within economics: the ‘New Classicals’ and the ‘Keynesians’. Never has an academic bun fight been more important to the public. The outcome may well determine the next thirty years of public policy.

So get involved. At least read Krugman’s side of the argument.

There is no point in my hiding my allegiance in this battle: I am firmly on the side of Krugman, although I think he has only just begun to understand the trajectory needed: he is steeped in the old way of thinking and is only now emerging into the brighter light of progress. Nonetheless he gets it. I cannot name an instance at the University of Chicago where I would say the same. Indeed they now all appear to be astrologists rather than academics.

So what?

Well, the issue before us can be boiled down to simple terms: are markets efficient?

I say no. There is no practical way of meeting, in the real world, the necessary conditions that underpin an ‘efficient’ market. So none – absolutely none – of the properties so valued by efficient market economists can be discovered out here on planet earth. That is not to say that the mathematics of the efficient market folks isn’t the apogee of elegance and rigor. It’s just that is has no earthly application or relevance.

The heroes of this school of thought – Lucas et al – represent the ‘who’s who’ of economics over the past three or four decades. To them have gone the preponderance of Nobel Prizes and awards. To them have gone the best contracts and the bulk of the recognition.

So well established did they become that they came to dominate the formation of public policy. And it is here that you all ought to pay attention. For decades the guiding ‘worldview’ of our legislative process and of the elites in business and public policy has been driven by the New Classical paradigm. Within which the notion that markets are efficient reins supreme. Flowing logically from this are the practical ideas of deregulation and limited government intervention – indeed to a Lucas influenced central banker or policy maker any kind of government action is, at best, self defeating, and at worst damaging. This viewpoint dovetails nicely with libertarian style Republicanism of the Reaganite mold – indeed it provides that form of Republicanism with a gloss of intellectual virtue.

You can see the effects all around you: deregulated banking is a primary cause of the recent debacle; weakened government institutions fail to respond to Katrina effectively; and the current health care debate is largely forged in those terms.

So to argue that the current dispute within economics is simply an academic spat of no consequence to the wider world is just not true. The outcome is essential.

The good news, as Krugman indicates, is that the lost knowledge acquired at such cost in the Great Depression, has been resurrected and used. In the nick of time we re-learned the lessons of the 1930’s. Keynesian economics is back in full force.

That is good news.

But is also just a beginning.

Financial firms all across the world are replete with analysts brought up inside the New Classical tradition. Our colleges still use text books written by New Classical heroes. The flood of PhD graduates going from the top schools to Wall Street to model, wrongly, derivatives and other arcane aspects of our financial world is unabated. Every time one of these highly educated and well meaning people run a model based upon the efficient market hypothesis they propagate a falsehood. They bake in an error. Every time a talking head on TV speaks breathlessly about the ‘market’ they, perhaps unknowingly, spread the meme that markets are good things we should pay attention to and leave well alone. The error spreads in all these small ways. It will be very difficult to root out.

The truth, such as we now see it, is very different and much more nuanced. An economy has properties that exhibit regular behavior, but markets are ’emergent’ from a group of individuals – you cannot simply sum up the various individual behaviors and get a market. So the properties we must use in our models and in finance are far deeper than the simple mechanisms taught us by the likes of Samuelson, Friedman, and so on. Complexity intrudes at every stage. This is the great lesson of the financial crisis: the very mathematics upon which we based, in good faith, our understanding of market behavior, is wrong. The point is that even wrong theories can produce good results for a time: Ptolemaic astronomy served well in the limited seas of the Mediterranean for centuries. The notion that risk is always clustered at the center of a normal curve, and that those, now famous, ‘long tail’ events are so improbable as to be sensibly ignored , is such a theory. Wrong, but useful enough to dominate thinking for a while. And it was responsible in large part for the collapse of the derivative market and hence the big banks. Cleared headed, textbook economics, perfectly executed and properly implemented destroyed the financial system. It wasn’t simply greed: it was intellectually sound and risk adjusted greed.

And it sank like a stone when the real world’s complexity intruded.

So now it’s time to import fresh ideas to buttress the Keynesian vision.

Mandelbrot, Prigogine, Minsky, and others have all something to add.

This is why Krugman is at the start not the end. He asks where does the profession go from here? I respond by saying: into complexity theory, information theory, statistical dynamics, agent based modeling and evolution. All these great traditions are inter-twined. From their nexus we will create the new economics.

What fun we will have. For those of you who have a practical bent: agent based modeling is the tool that I see as having the most immediate impact on things like risk management and portfolio analysis.

Meanwhile, remember this: markets are flawed. They work well sometimes, but there are moments when they hit barriers and need help. The one thing we know is that they are not efficient. Learning that has cost 6.7 million people their livelihoods. Perhaps one or two professors should join them.