Lately, a lot of our discussions have revolved around the idea of modelling the real economy. I also wrote a bit about how modelling is connected to other scientific pursuits like machine learning. What I left out however is an answer to the question “So what? Does it matter?”. I believe this is a critical point that economic departments from many universities around the world have been neglecting. Frankly speaking, during my 6-7 years formal education, they barely even talked about economic modelling until I reached my last year of study… which is surprising considering this is the hot sexy stuff at the forefront of the academic world now, especially if you want to dip your hands into macroeconomics.
Naturally, the failure to properly include economic modelling into the mainstream academic curriculum has led to a freak tonne of confusion outside of the little scholars’ bubble. Having talked to a bunch of people in and out of the academic including my students, one of the recurring themes is the relentless attack on mathematical economics and economic modelling in general. Frequently, people resort to citing the recent major flops like the global financial crisis of 2008 as being the signal of the incompetence of economic models. I take my stance in the middle as I do not have any strong feeling towards mathematical economics in particular (in fact, I have grown to hate it due to its sometimes unnecessary complexity), but I do recognize its importance in the evolving world of knowledge.
Of course, many results/forecasts generated by economic models were derived from some very unrealistic assumptions, so the conclusions do not represent very well what is going on in the real world. Still, it is worth pondering. Remember, the study of any model, and by that I meant ‘academic model’, is all about simplifying and extracting the most essential elements out of the complex world. This is to build a rigorous framework with explicit assumptions for thinking, upon which scientific analyses can be conducted. Each model will without doubt have its own flaws that cause the generated conclusions to depart from reality to a certain degree. However, the fact that the assumptions are explicit are what allows for concrete criticism and re-adjustment on the model at a later stage. Think about it. Before Einstein, Isaac Newton was the king. What allowed Einstein to create a superior model of the universe (and the reason why we know it is superior) is because Newton clearly laid out all the assumptions and other elements of his model. Only when our weaknesses are made known, can we improve upon them. And my friend, this is what modelling is all about. Sure, the first one might be basic, unintuitive and inaccurate. However, the most crucial role of primitive models is not to correctly inform or predict but to lay sufficient groundwork for later development. Each subsequent improvement of any model will take you closer to reality and generate more meaningful interpretation. As you progress further in your academic pursuit or maybe in your career, you will encounter with new models that may be based on the old model you have studied. While this departure from the old is important for practical purpose, you would not be able to understand it if you do not grasp the basic framework. That is why comprehending even the very basic and unrealistic approximate of the real-world setting is pivotal.
In fact, while economists at large failed miserably to foretell the crisis of 2008, if you can just take a few steps back to see the big picture, you will certainly notice the greatly improved economic performance over the past century hugely thanks to better informed economists and policy makers. Our economy has never been more stable and it will get better on the assumption that nuclear war does not arise (though nuclear war itself tends to be overstated). Recession, let alone depression, hits less frequently compared to the past. The one feature though that makes everything seems worse is the speed of information that allows for more public reports to be made; thus, it creates an illusion that the world economy is losing its balance (which I strongly disagree). Now, ask yourself, where did all the useful information conducive to economic stability come from? Obviously, from the expanding data pool and the more efficient and effective utilization of the available data as the economic models (that output those information) are continually updated from one generation to the next. Surely, these more sophisticated updates are built upon very simple initial models; hence, the significance of economic modelling at all stages.