Lamppost Afterthoughts

from Peter Radford

Notes from the cutting room floor:

Lest anyone be mistaken I am aware that we live in a world of models. Everything we do can be described at some level of abstraction as a model. Our entire ability to live is based upon modeling. Our various systems and bodily functions are models. That is these things are solutions to problems, and those solutions are encoded some way so as to allow that code to cause behavior that, within our environment, is efficacious,

Models are simply a way of testing and examining a specific solution to a problem. Sometimes they begin life in a rather ad hoc fashion, as guesswork, or heuristic. Other times they are the result of prior thought, in which case they represent a point along a line of modeling stretching back a long time. Perhaps so far back that their origins are forgotten. Or perhaps only so far back as the last test of an older model.

In any case models are small packages of our ideas spliced together towards some end. That end being a solution. A problem solved.

Evolution, of course, is the greatest problem solver of all, so the processes of evolution give us insight into modeling. We create models. We test them. We select, as successful, those that survive when let loose into our environment.

That environment might be the economy. That model might be some theory of how the economy behaves.

When you and I enter the marketplace to participate in exchange we expect certain things to be there and that certain things will occur. Those expectations exist because we have a model of the economy in mind. For most of us that model is opaque. It is hidden from our conscious expression of what will happen. All economists have done is to explore such models and eliminate the opaqueness. In so doing they have convinced themselves that they understand, at least partially, how an economy works.

My criticism of the use of models in economics is then not one of a denial of the value of models. It is directed at the selection of them. More to the point it is directed at the severity of the exclusion of zones of our experience that fall outside of our theoretical study because those zones are less tractable to our modeling technique. We have become monotheistic in our beliefs. Polytheism is frowned upon.

This is despite one of the great lessons of evolution: that diversity of approach gives the greatest chance of the location of solution. Economists have limited their options by reducing their subject to that set of issues most easily modeled according to the discipline’s self-referential code of modeling. And, further, they describe the problems they set out to solve in such a way as to produce results from their tests that conform to a prior code of acceptability.

So the code of modeling limits the usable technique, and the code of acceptability limits the range of problem. This duality then combines to define the discipline. Economics thus becomes defined as the subject that theorizes about problems tractable to the preferred technique.

Since the preference for techniques has moved through time, becoming steadily more formal as it goes, the subject itself has shifted about a bit. Nowadays it is easily mistaken for being applied mathematics or the mathematics of maximization. In order to arrive at this point a great deal of what was economics has had to be jettisoned or left to one side as being no longer of interest. Other topics have simply been ignored because they didn’t fit easily within the technical purview of economics thus described.

Interestingly economists have become so enamored of their maximization techniques and their mathematical mastery of things such as efficiency that they have begun to describe a whole host of topics as being capable of economical modeling. They have thus muddled themselves up. Instead of arguing that applied mathematics may shed light on some problems, they have come to believe that economics has something to say about those problems. Economists themselves have conflated subject and technique.

By so doing they have spread their error far and wide.

In finance, for example, much of modern financial theory traces its roots back to standard microeconomics. The problem is that the conditions needed to make microeconomics work the way standard theorists would like it to work aren’t very sensible in the closer-to-reality world of finance. In fact they rarely apply at all. Which means much of modern finance is little better than guesswork. Much like much of modern microeconomics.

In any case I don’t mean to go on too much. I am simply saying that I am aware of the universal presence of models. As Karl Popper once said: “All life is problem solving”. So economics is hardly unique. Where economics fails is in the conflation of subject and technique I noted above and in its monotheism. It thus commits a grave sin in the eyes of evolution. It has committed so much to its method and its beliefs that further movement in the economy, new problems or new topics of interest, are more likely to sit outside its self-selected purview than within. It is always tempted, therefore, to bang such problems into bizarre shapes in order to attempt to redefine them for analysis, hence the blind spots and contortions over the recent crisis and over inequality.

It has become too specialized. Like the dinosaurs.

And just for reference: I sit on the side of disequilibrium. The cultural, social, economic, and political worlds are complex. Why ignore that when we model?