Models are the common analytical instruments in economics, but they often seem to be misused.
One of the silliest mistakes I have ever seen is provided by the empirical analysis of real convergence hypothesis based on the simplest form of neoclassical growth theory. According to its core idea (i.e. beta-convergence), poorer national economies should produce higher growth rates in term of per capita income. All this is traced back to the age-old theory of diminishing returns, so capital-rich countries are expected to grow at lower levels. This is the theory, while the empirical side is something different.
It is very typical how the editors of page Convergence on Wikipedia try to sort out the problem that the „fact that a country is poor does not guarantee that catch-up growth will be achieved.” The entry describes this phenomenon as a limitation to the theory. This is a wrong approach to the problem, similarly when the editors say that there „are many examples of countries which have converged with developed countries which validate the catch-up theory.” So what’s up now…? If one finds some poor countries growing at low rates then he can validate the theory – while, on the contrary, if the statements of the theory are not true on their face value, then we should think that the theory itself is refuted. I do not thinks so.
No wonder if we come across contradicting statements about the value of such theories, since the big picture is admittedly confusing. Some empirical facts support a given theory (or, more precisely, they seem to support the theory), while others do not. However, this problem can easily be solved on the basis of the Weberian methodology. Theoretical economics is nothing more than a base line case, a system built from abstract-idealized components, from which actual economies should always be supposed to be different. And it is the difference that should be understood and explained. Why actual economies are different from our theories? The answers are available in economic policy, institutional economics, transitology or even sociology, and the like – while pure theory is only responsible for generating the base line cases, that is, the basis for comparisons.
So what if we made philosophy of economics, or methodology if you like, the theoretical ground for our everyday practice…? This is my idea.