Loaded words and modeling

Here is a nice review by Burton Malkiel of "Models Behaving Badly" by Emanuel Derman. The models of the title are from the world of finance: how are assets priced?

I am a layperson to the world of finance, so I find it difficult how to apportion "blame" for financial crises on faulty models or fraudulent inputs to them. Certainly history is littered with financial crises, so the influence of modern modeling alone cannot explain everything.

In any case, I just want to take this opportunity for a small lament that the beautiful act of modeling must be dragged through the mud by a financial crisis in this way. It would be fair to say that I am almost fanatical about the virtues of the concept and practice of modeling. I believe that modeling is inescapable. The world is complicated. Our senses deliver so much information, our mental apparatus must work so hard, that to process the world around us is to model. It is too much to ask that we understand everything; we have to understand a version of everything that is not so complex as the world.

This is also why economics works with models. We don't have a scale replica of the world that we can play with to see how this affects that. We have to build a scale replica from scratch, using our best judgment to push insistently at the boundary between complexity (so that we can understand our model) and usefulness (so that we can make something from it).

In a way we are much luckier in economics than in finance. Progress in economic theory comes as our models are improved upon and refined, but we are more able to iterate forward because our models are not embedded in a Leviathan global finance industry that depends on their continued function. Creative destruction of old models is hard when the house comes down with them.

With all this in mind I want to highlight this passage from the review:

He sums up his key points about how to keep models from going bad by quoting excerpts from his "Financial Modeler's Manifesto" (written with Paul Wilmott), a paper he published a couple of years ago. Among its admonitions: "I will always look over my shoulder and never forget that the model is not the world"; "I will not be overly impressed with mathematics"; "I will never sacrifice reality for elegance"; "I will not give the people who use my models false comfort about their accuracy"; "I understand that my work may have enormous effects on society and the economy, many beyond my apprehension."

How many of these will I accept for economics? Certainly the first; the model is not reality. Certainly the second; math is helpful in model-building but is not the point of model-building. The fourth and fifth are hard to argue with.

The third I don't like. Everything that we do must sacrifice reality. The test of a model is not its realism (a realistic model airplane would be no fun at all). All models are unrealistic because all models are wrong. Of course elegance is not the test of a model either, except that an elegant model is one that illuminates a relationship in a clear way by cutting to the heart of what matters.

Anyway, the point is that I think that "model" is not a dirty word. I feel possessive about "modeling" much the same way as I feel possessive about "rationality" - what they mean to me is important and wonderful and I hate to see them sullied by misrepresentation that stems from their overlap with the real-world ideas of modeling and rationality. I wish that all of the things like these could have their own words that are not borrowed from natural language.

Abstract revolutions or: if something is new, the old thing must be bad

There is a part of this small, otherwise enjoyable article about the advent of neuroeconomics that bothers me. The premise of the piece is that neuroeconomics is "seeking a physical basis for [economic theory] inside the brain". This is a field that is certainly sexy and possibly exciting; a while back I argued that it is a field that in some sense rediscovers the absolute primacy of "preferences" as the keystone of economic theory. I said:

The potentially exciting thing about neuroeconomics is that, even allowing for inexactness, it might tell us more about the actual hedonic motivators of people. Ambitious, yes, but not unimaginable. Of course, to an economist who wasn't under the mistaken impression that simplified preferences are supposed to be realistic, it might just amount to saying "your simplification is a simplification", which is slightly less exciting news. Or not news at all.

OK. In today's article, we learn that 

modern economic and financial theory is based on the assumption that people are rational, and thus that they systematically maximize their own happiness, or as economists call it, their “utility.”

Since it's hard to figure out what is going on in people's heads, the argument continues, we employed an idea called revealed preference, the reconstruction of unobserved objectives from observed choice. Neuroeconomics, the claim runs, may one day be able to identify brain structures that are associated with various components of choice, and so neatly sidestep the problem of unobservability.

But this is much too much:

While Glimcher and his colleagues have uncovered tantalizing evidence, they have yet to find most of the fundamental brain structures. Maybe that is because such structures simply do not exist, and the whole utility-maximization theory is wrong, or at least in need of fundamental revision.

Utility-maximization theory is wrong. It is wrong by construction, because it is a model, and models are wrong by construction. Why must we go through this? It might be easier to sell me an iPod if you first convince me that CDs are useless, but that's marketing. Why do we need to market neuroeconomics this way? That tiny little word "wrong" up there is a sin, because it belies the very essence of modeling as a means to make sense of things.

Is it perhaps that what we should understand by the quote is that looking into the brain will tell us that people are not, in some sense, maximizing their hedonic pleasure by the choices they make? This amounts to both an attempt to open the black box called "preferences" and to pin down the (biological?) process by which decisions are actually made. If this is the sense in which utility-maximization will be proved "wrong", then in the first place it is not clear to me that neuroeconomics can accomplish such a thing. Leaving aside the tricky questions of intent, free will and consciousness, can there be anything inside the black box but another, and another? The leap from the correlates of physical choices in brain activity to the content or existence of a utility function is huge.

But more importantly, even taking literally the notion that we will be able somehow to trap preferences or process in a cage, it is surely impossible that any breadth or depth of evidence on what these preferences are could preclude the need for us to model. What if the thankless treadmill of refining imperfect models of the world could at last be switched off, that there is an apple of knowledge that will free us from the need for models ever again? This is a seductive idea, but it cannot be. To do away with models would be to be as complex as reality, and that is a fight that reality will win every time. 

More rationality, and a million models

Here is an article on a conversation with Raquel Fernández that I found very interesting. The whole thing is worth reading, but I will quote at length this passage on rationality:

There is a beauty to the models in and of themselves. You assume, for example, that people are rational. I don’t think any really good economist thinks that people are perfectly rational, but, on the other hand, if you want to model people as not rational, all of a sudden it’s not clear what choice you should make. There are a million and one ways to be non-rational; there’s only one way to be rational within the confines of a model. Rationality means one thing: you’re maximizing your welfare subject to constraints. Now, if you say people don’t always maximize, and they’re beset by this and that, then all of a sudden you can have a million models. And that’s a little bit unsatisfactory too.

This is pretty close to my own view. My extra take is that any "irrational" behavior - and so anything that the "million models" generate - can be rationalized, either by revising what we assume that the decision-maker cares about, or by adding constraints on the decision-maker's ability to choose: the question of rationality is a red herring.

Daniel Kahneman and the rational choice model

Daniel Kahneman has a new book, "Thinking Fast and Slow", that is prompting a lot of excellent articles about his work. For example, Vanity Fair has a nice article by Michael Lewis. Back when I was an undergraduate, Kahneman and Amos Tversky's Prospect Theory paper was one of the first academic papers I read that motivated me to become an economist.

The article quotes Kahneman as being astonished to learn early in his career that:

The agent of economic theory is rational, selfish, and his tastes do not change.

I just want to point out that this model of decision-makers in economic models is more innocuous than it seems. It means that when we put a decision-maker in an economic model, we assume that she

•  has things that she cares about, and
• will do the best she can to achieve the things she cares about. 

This is what we mean by rationality. The great thing about this assumption is that it is completely flexible - it can accommodate any preferences at all.

For example, the decision-maker might care about the wellbeing of her neighbor, and so give up some of her own material wealth to help her neighbor be better off. This is both rational and selfish!

For any decision we observe, some preferences will rationalize it. As economic theorists, the crucial assumption is therefore not rationality, but on what it is that the decision-maker cares about and what constraints there are on her making a good decision.

The great contribution of Kahneman and Tversky was to present evidence on the psychological constraints on our decision-maker's ability to achieve the things she cares about. Making decisions is hard: even if I know what I want to achieve, "doing the best I can" is subject to how I process the options I have to choose from.

Rationality is not supposed to be a realistic model of the process by which people actually make decisions. Rather it attempts to capture the outcomes of decision-making in a plausible way, so that we can try bit by bit to analyze economic settings without simply saying "people are completely unpredictable, so let's give up". Kahneman and Tversky helped to show how to write models of rational choice that better reflect the decisions we observe people make.

Meta-ignorance

The "how badly people do on civics and general knowledge test" sub-category of Cheap Journalism is always good for some light entertainment. But this recent example from Newsweek is especially delightful for the pontificating it inspires.

For more than two centuries, Americans have gotten away with not knowing much about the world around them. But times have changed—and they’ve changed in ways that make civic ignorance a big problem going forward. While isolationism is fine in an isolated society, we can no longer afford to mind our own business. What happens in China and India (or at a Japanese nuclear plant) affects the autoworker in Detroit; what happens in the statehouse and the White House affects the competition in China and India. Before the Internet, brawn was enough; now the information economy demands brains instead.

Goodness, where to begin!

Let's start with the obvious. It's always puzzling to see "competition" used with blithe disregard for what it means; why is it that competition is considered a lovely value at the level of the firm but a horror at the level of the country? Does a similar principle not apply? Of course decisions are interrelated and it is worthwhile to understand that interrelation, but the latent assumption seems to be that if we ain't getting ahead we're falling behind. It is not a fundamentalist opinion to demonstrate that the global economy is not a zero-sum game.

But that's too close to real arguments for my taste. Let's focus instead on

Before the Internet, brawn was enough; now the information economy demands brains instead.

This is a remarkably peculiar statement. To achieve more with a given set of resources requires innovation: once upon a time the only "production" humans achieved was to transform time and stone tools into dead animals and dinner. "Brains" have, of course, are irreplaceable in the process of achieving more. This is not abstract: the printing press, electricity, the loom, agriculture... we could populate this list ad infinitum. One entry on the list would be the Internet. The article is instead in a peculiar corner of arguing that brawn was enough to invent the Internet which immediately made brawn redundant.

Which is trivially nonsense. The "information economy", whatever that may be, is not a revolution in the progress that pushes gently but tangibly at the constraint inherent in turning scarce resources into output that humans value. That process has never been the brute force of brawn.

Shared economics

Roger Goodell has joined the debate on the semantics of the word "economics". Welcome, Roger! From ESPN (warning: obnoxious auto-load video alongside article) today:

Commissioner Roger Goodell wrote NFL players Thursday, outlining the league's last proposal to the union and cautioning that "each passing day puts our game and our shared economics further at risk."

Now what might "shared economics" mean? Of the several points of disagreement between NFL owners and players in the current conflict, one is, of course, money. I'm going to regretfully assume that the resource to which Goodell's "economics" refers is only money. So what does he mean about that money?

Can "shared economics" mean a "shared allocation of resources"? It's possible; the two sides can take, in some sense, shared ownership of a specific or formulaic allocation of the pot o' gold from professional football. But this is odd. A problem (paraphrased) is that the owners want a larger cut of money off-the-top, which is different from the status quo. So the mutually agreed allocation of resources from the last CBA has thus far never been on the table.

The whole problem, of course, is in the sharing, in the allocation. It is nearly tautological that a disagreement about the allocation threatens the allocation. So that can't be it.

Can "shared economics" mean "shared money"? This is also possible. There's a work stoppage. In the worst case, the 2011 season is lost or curtailed and there will be a massive loss of revenue. Even in the best case, there is the fuzzier risk of alienating fans, contractual partners, and antitrust authorities, which could plausibly do the same. This seems a much likelier contender for what Goodell had in mind when he picked the word "economics" in his letter.

So to Goodell economics is a synonym for money, or revenue, or resources. This is depressing even for someone of my own view that the word "economics" is hopelessly lost to language forever. "Economics" is no more the resource than football is 22 people running around on some grass.