Abstract: I believe that economics should focus more external validation of models. It's refusal leads to an economic religion. With implementation of economic theories local factors and alternatives should be more prominent. If you want to know why, please read the rest of the article.
For some time now, the attack on economics as a science has been made by the media, the politicians and even economists themselves. The main focus here is on mainstream macro-economics (aka neoclassical economics). People criticize the simplistic nature of the science and mainly the weird assumptions that are being made. Now it is true that perfect rationality, homo economus, perfect competition perfect information and such do not actually exist. In fact they don't exist anywhere in the real world. But in my opinion and that, this is not the real problem
To assess the actual problem it is useful to compare economics to the science it wants to be: physics. In physics some very weird assumptions are also made: the existence of holes with negative mass in electric currents is but one example. Also extreme simplifications of reality are present: no gravity, no friction, no air, infinite time to relax a system and many others. Yet physics is not in any distress at the moment. Weird underlying simplifications and assumptions could not possibly be the only explanation then.
In fact economists are not that protective of their assumptions either. When I discuss with a more traditional economist, I find that challenging assumptions and replacing them with more realistic ones can often be achieved quite simply. This often happens when I introduce evolutionary models to them. Assumptions and simplicity are simply not the main culprit here.
Also the mathematical way of structuring the models, another thing critiqued by some, seems to work out reasonably well. After all what better tools do we have that can provide such internal validity to any science?
The validation of the science is touching a critical part here. The external validation is where economics and physics diverge. In physics anything that can not be proven will remain a good theory, but not an absolute one, until proof is found. This is why we spend billions of euros on the CERN complex with a primary goal of proving the existence of the Higgs boson. Regardless of the field of physics this almost always holds true (fields in which this doesn't hold are often ridiculed by other for providing mathematical fantasies).
Economics is rather different. It is true that there are a lot of people, some of which great economists, working on external validation. This is not the problem. However, the main divide in economics seems to be occurring not between economists of different traditions (Classical, Keynesian, Austrian, Evolutionary etc.) but between those that make models and those that look at the real world. This leads to a situation in which great models are made but are never tested thoroughly. And great discoveries that are made in economic data and also within other sciences are not seriously being taken into economic models. Rather they are added as outside features. This leads to so called fads, Naim has a good paper on this from the asia crisis.
The sad thing what then happens, is that models that were introduced with a very good intent are being used in the wrong way. They are also never really validated or adapted to new insights. In fact they are even more simplified into one concept or idea. Some examples are 'the free market' (Neoliberals), 'stimulus' (Keynesians) and 'class warfare' (Marxists). This then leads to an orthodoxy and dogma in the economics science that borders on religion. This is most visible and most painful in the economic experiments from the 20th century.
It is therefore not a coincidence that I chose a paper about developing economies when it comes to fads. This is usually where the dissonance between economic macro theories and reality is often the most visible. Also this is where economic religions cause the most harm. A good example of this is the free market paradigm that is presented by neo-liberals. Look at Russia and the former Warsaw pact after the fall of the iron curtain. If you want to know how the 'stimulus' experiment is working out just look at the US or for more historic reference: Japan. Finally the Marxist experiment (communism) does not need further introduction.
The main issue with all of this is that the real world experiences are not used to update, alter or refine the existing theories. At the current time economists are still calling for free markets and stimulus, even though history has already shown that the economy does not behave according to these grand concepts.
In an additional note it is often astonishing how economic macro theory often fails to appreciate the different structures that are present in the economy. In a previous post I have addressed different institutions and their impact on economies. But not only institutions are different: culture and history are also of major influence of the economic workings in a country. Therefore a policy that succeeds in one country can fail completely in another even when economic factors (capital, labor etc.) are similar. I would still like to see a push into adapting global theories more into local contexts and also local experiences adapting the global theories.
In conclusion, the economic science needs to change. This change is not better models or better data collection but a better communication between the two. This will actually lead to better models and better data to work with. Local contexts are important and are often missing completely from the theories. They should at least add to them that a given theory is only valid under a certain underlying structure of the economic system.
A final note: I don't think it is actually possible to find a grand unified economic theory as we can find it in physics. People are not electrons and at times behave unpredictably, also when in groups. Therefore be very wary when an economic solution is offered that can be summarized in one term. The world is just not that simple. This should not stop us from finding increasingly better ways to model economics as by doing so we can improve many lives.
No comments:
Post a Comment