Models

Geographers know the lay of the land but economists do it with models, as the witticism goes.

It takes quite a bit of training to appreciate the utility of models, and how and why they are indispensable in explaining the artificial world we live in.

Many of us are familiar with models from our childhood. As a kid, I used to build model airplanes from hobby kits. Not just as hobbies, model airplanes are extensively used and tested in wind tunnels to design the real thing even in this age of sophisticated computer modeling.

Very complex systems are simulated on computers as models. Weather forecasting relies on running huge models with billions of data points on supercomputers. Climate models attempt to predict what the climate may be like in coming years and decades. 

The models economists use are unlike those used in aerospace engineering and meteorology. Our models are essentially stories we make up. The  characters in our stories are caricatures of real life characters. We make up these characters and let them interact with each other using rules that we make up for them. We then let them go play out their roles. We are most interested in knowing how the stories end. We don’t have control over that; we only create the characters and the rules of interaction. Then we sit back and observe.

Real life is complex. To understand how real life works, we have to simplify it by making assumptions about how the characters in our models behave. These are behavioral assumptions. In one commonly told story about producers and consumers, we assume that consumers are utility maximizing: that they try to get the most bang for their bucks. We assume that producers are profit maximizing: that they produce as much as they can as long as they are making a profit. 

We assume that when the price goes up, consumers reduce the quantity they buy; we assume that when price goes up, producers produce more, etc. These are assumptions. Seems realistic. Then we ask ourselves this: if there are no barriers to entry or exit in a market (a free market), what will happen? We conclude that the interactions of buyers and sellers will lead to an equilibrium price — a so-called “market clearing price” at which the quantity demanded is equal to the quantity supplied. 

This is the model we call “perfect competition.” In the real world, there is no such thing as perfect competition. But the model is useful. It serves its didactic purpose very nicely and is taught in every introductory class on price theory. It also explains what happens in the real world such as when there’s a negative supply shock: the price goes up.

Models are theories. They are conjectures — which is a component of the scientific method. The scientific method has three steps: conjecture, criticism, and tentative acceptance. That is the method of the hard sciences but economics (which is at best a social science) also profitably uses it. We non-physicists struggle with physics theories (try to understand quantum physics) but non-economists can without too much effort understand economic models.

Good economists don’t take their models too seriously. They use them as tools, not as some infallible oracle. All models are wrong; some are useful. The relationship between models and the real world is the same as between maps and the territory. Maps are abstract and simple; the real world is concrete and complex. A map that attempted to show every feature of the real world would be totally useless. Without simplifying assumptions, economic models would be worthless too.

In most cases, we begin with the simplest model. For instance, we begin with a model of perfect competition. Then we introduce some complexity. We say what if the actors are not fully rational; or what if they don’t have all the information; or what if the market is monopolistic or is oligopolistic, etc. We follow that story and see where it leads us. If we have a good story, we learn the moral of the story—which we call a theorem.

One example of an elegant theorem is the Coase theorem, named after the British-American economist Ronald Coase (1910 – 2013). He was mentioned in the previous post, Negative Externalities. Coase was awarded the Nobel Memorial Prize in Economics in 1991 for his analysis of transaction costs and property rights in the functioning of the economy.

Fact is that most important theorems are somewhat hard to understand. They appear to be simple but even trained economists sometimes get it wrong. An example: the theory of comparative advantage. The Coase theorem is another.[1]

I’m no intellectual slouch and I’m not a newbie either since I’ve been in the business for over 30 years. Despite that I have to re-read the fundamentals of economics several times to really understand them. I believe that the most important meta-lesson I have learned in all these years is that some of the lessons are deceptively simple. At first, on learning of a model one would say, “Well, so what?” Only after a lot of thinking one comes to the “Wow! That’s awesome.”

The important lesson I learned was how to think slowly, patiently and methodically from the assumptions to a tentative conclusion instead of taking giant leaps of fancy. The discipline itself has progressed slowly and methodically over the last couple of centuries of its existence. 

One comment to that post on externalities (linked above) motivated this piece on models. The commenter was dismissive of the story of the railroad and the farmer, and wrote: 

Sounds great in theory, except in practice there’s a huge difference in bargaining power between an individual farmer and a multi million dollar railroad company.

Undoubtedly there’s a difference between theory and practice. But the theory helps us focus on the salient bits of the real world. What happens when the transaction costs are high or property rights are not clearly defined? What happens when there’s asymmetry of information or bargaining power? We continue to explore the real world guided by our model.

Our model suggests that property rights have to be clearly defined (by the government) and transaction costs have to be reduced (by technology, perhaps) for the result of the Coase theorem to hold.

The case of the railroad company and the farmer has been very cogently explored by Donald Boudreaux in his article linked to in that post. While it’s true that the railroad company is more powerful than the farmer, that dispute can be settled in a court of law. True, the railroad company has deeper pockets but the case hinges on who has the property rights. If the railroad company loses, the farmer can sue for damages.

That’s all for now. Be well, do good work and keep in touch. 

See also: Coase and His Theorem. Post from December 2010

The image at the top of this post is from the front image of a monograph “On the Economics of Ronald Coase.” You can download the PDF for free.

NOTES:

[1] About the Coase theorem, the University of Chicago Law School writes:

The influence of Ronald Coase’s 1960 paper, “The Problem of Social Cost,” cannot be overstated. . . . Coase’s theorem directs our attention to the real world—to the world of messy transactions and of choices constrained not just by individual budgets but by the design of the institutions in which those choices are made. Scholars are still grappling with the implications of “The Problem of Social Cost,” the most-cited law review article in history, precisely because it requires us to deal with the world as it is, not as we wish it were.

Download a PDF copy of the paper.