Is the Austrian School of Economics Pseudoscience?
Posted: Fri Jan 12, 2018 3:50 pm
This was one of the questions we were discussing in the Logical Fallacies thread. However, that thread went so far into politics that I think it's better left alone. People feel very certain about their political opinions, yet, by the very nature of politics, they can't be based on conclusive evidence. So, as would be predicted by the Bayes'es theorem, people are never going to agree on politics, regardless of how much evidence they are faced with. As far as I can see, some form of statism (belief that the government is necessary for a civilized society) is a part of the vegan philosophy itself, regardless of it pretending to advocate freedom. Also, the way people discuss politics here is quite dishonest: they presuppose that great knowledge of history is somehow necessary for understanding social sciences, and then they use what appear to be extremely biased analyses of minor historical events to support their point (like pointing to some rare occasions where the government managed to "stop" terrorism, when the government is obviously what causes terrorism in the first place by encouraging discrimination). One of the ironic things in epistemology is that people feel certain about their political opinions, but they feel uncertainty when reasoning about math, yet they are many times more likely to be wrong about politics than about math.
So, is the Austrian School of Economics pseudoscience?
For those who are unaware, the Austrian School of Economics is a heterodox school of economics that mostly rejects econometrics and attempts to predict the behaviour of large groups of people using an axiomatic system (like the Euclid's Five Axioms) called praxeology. Mainstream schools of economics, like the Keynesian School, rely mostly on the empirical data, and they also often attempt to do experiments, to come up with new theories. But, despite the Austrian School mostly rejecting econometrics, its theories appear to align with empirical data even better than the mainstream economic theories do. Most notably, its theories correctly predict that inequality is negatively correlated with the economic freedom index, while the Keynesian economic theories predict that the correlation is positive. That's why many libertarians, right-wingers and anarchists subscribe to the Austrian school of economics (because it predicts that less government intervention in the economy will cause less inequality).
BrimstoneSalad thinks that the Austrian school of economics, because of it being based on axioms rather than experiments, is not a real science, but a religion and an ideology.
I would argue that the Austrian school of economics is, in fact, even more reliable than the mainstream economics is. Here is why:
1. Science doesn't have to be based on experiments.. It's, for example, impossible to do an experiment in astronomy or in historical linguistics. We can't change the mass of the Moon or the velocity of the Moon to see what effects it would have on its orbit. Likewise, we can't go back in time to insert a new root into Proto-Indo-European to see how it will evolve. Yet, we can be much more certain that the Newton's Law of Gravity and the Grimm's Law are correct (or at least that they are excellent approximations to what actually happens) than that anything that's derived from the experiments in the nutritional science or in the psycholinguistics is correct.
2. Background assumptions (which are essentially axioms) are often necessary in science. To understand why, suppose you wanted to explain the observation that the words for "two" and "ten" start with the same phoneme. There are basically two types of explanations you can propose, each with different set of axioms:
a)The words "two" and "ten" start with the same letter because they are related semantically. "Ten" means "two hands". The phoneme 't' means "two" in English. The same is seen, for instance, the word "tide". There are two states of tide: the sea level can be either lower than usual or higher than usual. Also, the word "try" can imply two states: you can succeed or you can happen not to succeed in something you try. The same is seen in other languages. The words for "two" and "ten" start with the same sound in German, Russian, Latin, Greek, and so on. That's because they mean similar things.
b)One phoneme in one language usually corresponds to another phoneme in a related language. The words for "two" and "ten" start with the same sound in all Indo-European languages except in Tocharian and Armenian and they don't start with the same sound in any non-Indo-European language except the Nayi language, the Balinese language, the Thai language and the Evenki language (I'm not sure that's factually correct, but you see the point). That's because the Proto-Indo-European language happened to have the words for "two" and "ten" that started with the same sound. The same is true for words that have semantically nothing to do with "two". English 't' corresponds to Greek 'd' in the pairs of the words "two"-"duo", "ten"-"deka", "tree"-"dendron", "tame"-"damnemi" and many more.
The explanation 'a' bases itself on a background assumption that's contrary to almost all of the background assumptions of linguistics (that the symbols are arbitrary). The explanation 'b' bases itself on the same background assumptions the mainstream linguistics is based on. If you start exploring languages using the background assumptions the explanation 'a' uses, well, let's just say you are bound to get the wrong answers and have a very hard time figuring out what you did wrong using the empirical data. If you don't start with the right background assumptions, it doesn't matter much how empirical you try to be, you will get the wrong answers and you will probably be able to shoehorn every single apparent counter-example into your theory. The same is true in natural sciences, at least when they are young. Aristarchus started with the background assumption that the cosine of the angle between the Half Moon and the Sun would be inversely proportional to the distance to the Sun, that the laws of geometry we see on Earth apply to the celestial bodies, and that by orders of magnitude more massive celestial bodies can't orbit around the less massive ones. Ptolemy discarded that because of his background assumptions that our intuitions about the star parallax being visible if the Earth were rotating around the Sun and that we would be able to feel the Earth's motion. Having access to hundreds of years more of empirical observations didn't help him. The same appears to be going on in economics today. When economists observed that the Great Depression occurred, they had three basic ways of explaining it, each with its own background assumptions:
a)The prices in the free market with no regulation don't provide accurate information because people are bad at making long-term decisions. Adding some regulation and increasing the government spending will fix the problem.
b)Such economic recessions didn't happen before the government started regulating the banking system. If we don't have privatized banking system, there is no incentive to keep the value of money constant, which is necessary for the economic development. We need to get rid of the nonsensical regulations of the banks.
c)The harvests of rare metals were record low for years before the Great Depression. The government can't do much here.
If you pick the incorrect background assumptions, you will probably never find out why your predictions come to be wrong. Trying to do economics without believing that a large group of people would behave as if all people in it are rational may be very much like trying to do historical linguistics without believing in regular sound correspondences between related languages. You will get everything wrong potentially without realizing it.
3.Mainstream economics doesn't appear to follow the scientific method either, despite claiming to. In science, a theory that makes incorrect predictions again and again will be discarded. That doesn't appear to be true in economics. Keynesian theory predicts that increasing the government spending during the period of recession would speed the recovery up, that recessions are caused by a decrease in overall spending. This has been tried over and over again. Almost every time, the government spending increases to a level well above the level before the recession, and the recession continues for years. I'd guess that what the politicians and the economists think when they subscribe to the Keynesian theory is similar to the Pascal's Wager. If the government regulation is really necessary for efficient markets to exist, and we remove the regulation that seems nonsensical, we lose everything. If it's necessary and we add some new regulation, we gain a lot. If, on the other hand, the Keynesian theory is incorrect, and we still add some new regulation, we lose little or nothing. If the Keynesian theory is incorrect, and we remove some regulation, we gain little or nothing. And, as with the religion, there is no way to estimate what you lose if you are wrong nor to estimate the probability that you are wrong. And, as with the religion, there is an infinite number of conceivable sets of government regulations and very large number of ways to enforce them, so the probability that you choose the right ones is basically zero.
The most common argument against the Austrian School of Economics is that it supposedly ignores the paradoxes of the game theory, such as the Braess Paradox (that many drivers choosing what's the fastest path for them can in fact lead to the traffic getting slower than if some of them had chosen suboptimal paths), that individuals rationally pursuing their self-interests can lead to sub-optimal results. However, such paradoxes are, just like the Simpson's paradox, very unlikely to occur at a large scale. Also, in the case of the Braess Paradox, the government is unlikely to solve the problem, because the government quite often blindly builds new roads in an attempt to fix the traffic.
So, I hope I've explained why I think that Brimstonesalad's epistemological philosophy is wrong, and that I correctly inferred what his epistemological philosophy is (that I correctly understood what he thinks is the difference between science and pseudoscience). I haven't posted for quite a while and I have reexamined my beliefs thoroughly. I have come up with some arguments for my position and thought about some counter-arguments I might encounter. The post is quite long, but I don't think I am doing the Gish-Gallop fallacy. It's just a few arguments, but explained in detail. I'd like to hear opinions from some other people from this forum, I guess there are some more people here who are at least as educated about the topic as I am or as Brimstonesalad is.
So, is the Austrian School of Economics pseudoscience?
For those who are unaware, the Austrian School of Economics is a heterodox school of economics that mostly rejects econometrics and attempts to predict the behaviour of large groups of people using an axiomatic system (like the Euclid's Five Axioms) called praxeology. Mainstream schools of economics, like the Keynesian School, rely mostly on the empirical data, and they also often attempt to do experiments, to come up with new theories. But, despite the Austrian School mostly rejecting econometrics, its theories appear to align with empirical data even better than the mainstream economic theories do. Most notably, its theories correctly predict that inequality is negatively correlated with the economic freedom index, while the Keynesian economic theories predict that the correlation is positive. That's why many libertarians, right-wingers and anarchists subscribe to the Austrian school of economics (because it predicts that less government intervention in the economy will cause less inequality).
BrimstoneSalad thinks that the Austrian school of economics, because of it being based on axioms rather than experiments, is not a real science, but a religion and an ideology.
I would argue that the Austrian school of economics is, in fact, even more reliable than the mainstream economics is. Here is why:
1. Science doesn't have to be based on experiments.. It's, for example, impossible to do an experiment in astronomy or in historical linguistics. We can't change the mass of the Moon or the velocity of the Moon to see what effects it would have on its orbit. Likewise, we can't go back in time to insert a new root into Proto-Indo-European to see how it will evolve. Yet, we can be much more certain that the Newton's Law of Gravity and the Grimm's Law are correct (or at least that they are excellent approximations to what actually happens) than that anything that's derived from the experiments in the nutritional science or in the psycholinguistics is correct.
2. Background assumptions (which are essentially axioms) are often necessary in science. To understand why, suppose you wanted to explain the observation that the words for "two" and "ten" start with the same phoneme. There are basically two types of explanations you can propose, each with different set of axioms:
a)The words "two" and "ten" start with the same letter because they are related semantically. "Ten" means "two hands". The phoneme 't' means "two" in English. The same is seen, for instance, the word "tide". There are two states of tide: the sea level can be either lower than usual or higher than usual. Also, the word "try" can imply two states: you can succeed or you can happen not to succeed in something you try. The same is seen in other languages. The words for "two" and "ten" start with the same sound in German, Russian, Latin, Greek, and so on. That's because they mean similar things.
b)One phoneme in one language usually corresponds to another phoneme in a related language. The words for "two" and "ten" start with the same sound in all Indo-European languages except in Tocharian and Armenian and they don't start with the same sound in any non-Indo-European language except the Nayi language, the Balinese language, the Thai language and the Evenki language (I'm not sure that's factually correct, but you see the point). That's because the Proto-Indo-European language happened to have the words for "two" and "ten" that started with the same sound. The same is true for words that have semantically nothing to do with "two". English 't' corresponds to Greek 'd' in the pairs of the words "two"-"duo", "ten"-"deka", "tree"-"dendron", "tame"-"damnemi" and many more.
The explanation 'a' bases itself on a background assumption that's contrary to almost all of the background assumptions of linguistics (that the symbols are arbitrary). The explanation 'b' bases itself on the same background assumptions the mainstream linguistics is based on. If you start exploring languages using the background assumptions the explanation 'a' uses, well, let's just say you are bound to get the wrong answers and have a very hard time figuring out what you did wrong using the empirical data. If you don't start with the right background assumptions, it doesn't matter much how empirical you try to be, you will get the wrong answers and you will probably be able to shoehorn every single apparent counter-example into your theory. The same is true in natural sciences, at least when they are young. Aristarchus started with the background assumption that the cosine of the angle between the Half Moon and the Sun would be inversely proportional to the distance to the Sun, that the laws of geometry we see on Earth apply to the celestial bodies, and that by orders of magnitude more massive celestial bodies can't orbit around the less massive ones. Ptolemy discarded that because of his background assumptions that our intuitions about the star parallax being visible if the Earth were rotating around the Sun and that we would be able to feel the Earth's motion. Having access to hundreds of years more of empirical observations didn't help him. The same appears to be going on in economics today. When economists observed that the Great Depression occurred, they had three basic ways of explaining it, each with its own background assumptions:
a)The prices in the free market with no regulation don't provide accurate information because people are bad at making long-term decisions. Adding some regulation and increasing the government spending will fix the problem.
b)Such economic recessions didn't happen before the government started regulating the banking system. If we don't have privatized banking system, there is no incentive to keep the value of money constant, which is necessary for the economic development. We need to get rid of the nonsensical regulations of the banks.
c)The harvests of rare metals were record low for years before the Great Depression. The government can't do much here.
If you pick the incorrect background assumptions, you will probably never find out why your predictions come to be wrong. Trying to do economics without believing that a large group of people would behave as if all people in it are rational may be very much like trying to do historical linguistics without believing in regular sound correspondences between related languages. You will get everything wrong potentially without realizing it.
3.Mainstream economics doesn't appear to follow the scientific method either, despite claiming to. In science, a theory that makes incorrect predictions again and again will be discarded. That doesn't appear to be true in economics. Keynesian theory predicts that increasing the government spending during the period of recession would speed the recovery up, that recessions are caused by a decrease in overall spending. This has been tried over and over again. Almost every time, the government spending increases to a level well above the level before the recession, and the recession continues for years. I'd guess that what the politicians and the economists think when they subscribe to the Keynesian theory is similar to the Pascal's Wager. If the government regulation is really necessary for efficient markets to exist, and we remove the regulation that seems nonsensical, we lose everything. If it's necessary and we add some new regulation, we gain a lot. If, on the other hand, the Keynesian theory is incorrect, and we still add some new regulation, we lose little or nothing. If the Keynesian theory is incorrect, and we remove some regulation, we gain little or nothing. And, as with the religion, there is no way to estimate what you lose if you are wrong nor to estimate the probability that you are wrong. And, as with the religion, there is an infinite number of conceivable sets of government regulations and very large number of ways to enforce them, so the probability that you choose the right ones is basically zero.
The most common argument against the Austrian School of Economics is that it supposedly ignores the paradoxes of the game theory, such as the Braess Paradox (that many drivers choosing what's the fastest path for them can in fact lead to the traffic getting slower than if some of them had chosen suboptimal paths), that individuals rationally pursuing their self-interests can lead to sub-optimal results. However, such paradoxes are, just like the Simpson's paradox, very unlikely to occur at a large scale. Also, in the case of the Braess Paradox, the government is unlikely to solve the problem, because the government quite often blindly builds new roads in an attempt to fix the traffic.
So, I hope I've explained why I think that Brimstonesalad's epistemological philosophy is wrong, and that I correctly inferred what his epistemological philosophy is (that I correctly understood what he thinks is the difference between science and pseudoscience). I haven't posted for quite a while and I have reexamined my beliefs thoroughly. I have come up with some arguments for my position and thought about some counter-arguments I might encounter. The post is quite long, but I don't think I am doing the Gish-Gallop fallacy. It's just a few arguments, but explained in detail. I'd like to hear opinions from some other people from this forum, I guess there are some more people here who are at least as educated about the topic as I am or as Brimstonesalad is.