Is There a Place for Econometrics?
“It seems to me that this failure of the economists to guide policy more successfully is closely connected with their propensity to imitate as closely as possible the procedures of the brilliantly successful physical sciences – an attempt which in our field may lead to outright error. It is an approach which has come to be described as the ‘scientistic’ attitude – an attitude which, as I defined it some thirty years ago, ‘is decidedly unscientific in the true sense of the word, since it involves a mechanical and uncritical application of habits of thought to fields different from those in which they have been formed.’” – F.A. Hayek (1974)
The Pretence of Knowledge, Hayek’s Nobel Prize lecture and one of his most famous speeches, is known to many classical liberals. In it, he criticizes economists for misusing statistical methods and confusing economics as a natural instead of a social science. While Hayek supported how simple algebra can be used quite well to explain some general patterns in economics, he detested using statistical methods to numerically describe those patterns.
After all, since economists cannot know all variables influencing a specific issue they test, they actually never know if their result is correct. The results may even be wrong because they cannot account for all the specific variables. Or, as Hayek puts it, “On this standard there may thus well exist better ‘scientific’ evidence for a false theory, which will be accepted because it is more ‘scientific’, than for a valid explanation, which is rejected because there is no sufficient quantitative evidence for it.”
For example, while it is useful to algebraically explain that a minimum wage negatively affects unemployment, putting a distinct value on it – like a minimum wage of 15 euros an hour creates additional unemployment of one percent, could easily go wrong. There are two main issues: first, these numbers vary naturally depending on how you look at the issue. It depends, for instance, on what region you look at (are you testing data from South Africa or Germany?) and what time period. And second, and most importantly, you will never be able to find all necessary variables that influence unemployment. An economist will never know everything that influences the unemployment rate in that specific region during that specific time period. And even though there are many sophisticated statistical methods, using different variables one month later in that specific region could lead to completely different results.
One of many real-world examples can be found in a recent study by the European Central Bank, which tests income distribution effects. The conclusion of the paper is that the Cantillon effect, which states that monetary expansion by central banks leads to an uneven income distribution from the poor to the rich, has indeed not come into effect in Europe in the last years – something which is rather doubtful considering the massive expansion that has occurred. Thus, while the Cantillon effect is one of the biggest issues concerning central banks, a quantitative study can very well come to a different conclusion.
This is the crucial difference between economics as a social science and natural sciences: natural sciences can find specific effects which will always stay constant under specific circumstances. Economics is ever-changing.
Meanwhile, the study of human action, one of the focal points in the work of Ludwig von Mises and followed by your next-door Austrian economist, uses deduction of economic principles rather than testing them empirically. Praxeology is based on the axiom that ‘all men act purposefully’ and uses it to deduct all economic principles – from how minimum wages affect unemployment to how central banks are the reason for the booms and busts in the economy. Praxeology versus today’s mainstream economics is basically philosophy versus statistics.
But unfortunately, praxeology is not enough. In order to penetrate economic discussions more often – which, as should be clearly evident, is not the case right now, libertarians need to incorporate some of the methods of econometrics in their toolbox. In fact, Hayek mentioned himself how mathematics can indeed be used rather well to explain the “general character of a pattern even where we are ignorant of the numerical values which will determine its particular manifestation” to understand “the mutual interdependencies of the different events in a market”.
While limitations of quantitative methods in economics obviously exist, there are more ways these days than there were in 1974 to collect data, analyze and test it – and, indeed, there are ones that are actually useful. By ignoring these approaches, we seal our own fate to remain in irrelevance. Meanwhile, if we start conducting studies using econometric approaches and modelling, too, rival economists will start to listen. And since we are more aware of the limitations of these “scientific” methods than most mainstream economists, I am convinced we would produce even better results. Studies that go beyond those limitations would then need to be debunked – not only because they are most likely wrong, but also to stop the tiresome “New Study Shows …” headlines by the Left.
While this is especially important in fields like financial regulation, where libertarians, being too caught up in axioms to actually look at the real world, have not reached any clout in discussions yet, embracing econometrics in a reasonable amount can go a long way in economics in general. Finding the right balance by using the deductive approach in a non-dogmatic way, but adding proven econometric models to our repertoire, would produce even more arguments on why free markets and less regulation are the way to go.
This piece solely expresses the opinion of the author and not necessarily the organisation as a whole. Students For Liberty is committed to facilitating a broad dialogue for liberty, representing a variety of opinions. If you’re a student interested in presenting your perspective on this blog, click here to submit a guest post!