Scientists build models that capture the essence of some observed phenomenon. The models that I like best are strongly simplified. This in contrast to very complex and non-transparent models that try to replicate every aspect of reality. Good models allow us to gain insight into something we observe by using a limited number of key concepts. I use simple models to be explicit about the underlying assumptions and to be consistent.
I am well aware that macroeconomics studies millions of people and organizations making financial decisions at the same time. But this does not mean that nothing useful can be learned from simple models. By identifying the “elementary processes” and figuring out how they work together, we can understand why things happen. And when we have isolated the most important processes, it becomes possible to build a (computer) model of the complexity of the real world1.
This does not mean that I adhere to a deterministic philosophy about inevitable economic outcomes. Rather, a good financial model provides insights into the monetary relations between agents in the economy. It (1) helps us to think through the details of policy proposals2 and (2) stimulates thinking about how people will behave when they are confronted with a certain economic environment. The modeler can adjust some control parameters on paper and anticipate what is likely going to happen in the real world.
Models are much more powerful than “words-only” opinions on blogs and editorials, which mostly are either inconsistent or make a ton of implicit assumptions (usually hiding the political preferences of the writer).
- In a previous life as a research scientist, my colleagues and I used this same approach to model the physical properties of a non-equilibrium, magnetized, low temperature plasma in contact with a solid wall, as well as to model the growth of extremely thin materials on 3D surfaces. The things I currently write about finance and economics will hopefully reach a broader audience :)
- Such as helicopter money.