Being a former academic myself, I have many unpleasant memories of endless debates over methodology. Among these topics is the question of how quantitative methods ought to be used in economics, and when it is appropriate (if ever) for use in discovering and refining economic laws and axioms.
But academics aren’t the only ones who discuss these things. For example, I continue to hear from ordinary nonacademics who have apparently encountered this idea—a stereotype, really—that advocates for Austrian school economics “hate math.”
Indeed, one still occasionally encounters this question among some people attending Mises Institute events or among those who participate in the comment section at mises.org.
So, let’s address this issue from the point of view of nonacademics: Why do people have the idea that Austrians “hate math”?
Well, like many legends and stereotypes, this notion is based on some truth.
Ludwig von Mises and later economists in the Misesian tradition were indeed critical of the use of quantitative methods when it came to “doing economics.”
But for Mises and many Austrians, “economics” is a very specific thing, and it doesn’t include economic history or similar fields that involve describing trends in the economy. This skepticism in regards to quantitative methods was not intended to be interpreted as general opposition to the use of data in examining and illustrating the larger world.
So, what are these specific ways that “math” is a problem? And when is math useful?
Economic Theory—Narrowly Understood—Doesn’t Require Quantitative Tools
For Mises, many Austrians, and other practitioners of the “causal-realist approach,” the purpose of economics is to explain how the economy works by identifying the causal relationships between economic actions and events.
For example, when contemplating the effects of a minimum wage hike, economics should help us understand what happens as a result of the hike. That is, what is the cause and effect relationship between the wage hike and what comes next? Mises’s methods, it turns out, tells us that all else being equal, if the government forces the price of labor to go up, the demand for labor will go down.
For Mises, the particulars of this explanation could be deduced from basic axioms and laws around the fact human beings demand less of a good or service when the price goes up.
More generally, economics is largely the process of taking what few things we know for sure about people (such as the fact that when human beings act, they act in the hope of improving their lives in some way) and using that knowledge to deduce the effects of more complex phenomena such as rising wages.
This does not require complex mathematical calculations because deduction does not require a large number of quantified observations as is the case with an inductive approach. When induction is used—as is the case in physics, for example—a large number of specific observations are quantified and used to draw a general conclusion. Some people who call themselves economists try this method, of course. They look at many cases of minimum wage hikes, and attempt to identify correlations between the wage hike and the events that come after it. Sometimes, the data shows employment goes up in a region where a wage hike was mandated. Sometimes it goes down. These economists then debate endlessly about it. In no case, however, does identifying a correlation in this manner explain why employment goes up or down, and these methods—in spite of efforts to do so—don’t eliminate all the noise of countless other events that are going on in the economy at the same time and which affect wages.
Mises rejected this approach for the social sciences—among these reasons was the fact experiments are not repeatable in the social sciences, and economic phenomena cannot be isolated. To understand the connections between events such as wage mandates and employment levels, what is needed is a theory, not simply an observation that two things occurred near each other in time.) Mises writes in Human Action:
The insight that, ceteris paribus, an increase in demand must result in an increase in prices is not derived from experience. Nobody ever was or ever will be in a position to observe a change in one of the market data ceteris paribus. … No reasonable man can contend that the relations between price and supply is in general, or in respect of certain commodities, constant. We know, on the contrary, that external phenomena affect different people in different ways, that the reactions of the same people to the same external events vary, and that it is not possible to assign individuals to classes of men reacting in the same way.
That is, if we’re in the business of identifying basic axioms and laws, observing a bunch of transactions in the marketplace won’t, by itself, tell us what we need to know. After all, none of these observations tell us to what degree various external events are affecting the outcome. A employer’s decision to hire a worker cannot be isolated to just changes in the minimum wage. Other factors are always in play. Nothing ever happens with all else held equal (ceteris paribus).
So, deductive reasoning from foundational axioms becomes necessary.
The Fact that Economics Doesn’t Require Math Doesn’t Mean Math Can’t Enlighten Us
But here’s a key point: Mises absolutely did not suggest that economics tells us everything we need to know about the human experience, or that some grand theory of human civilization could be obtained through economics or deductive reasoning. Yet, some critics of the Austrians who haven’t done their homework actually suggest that many Austrians think everything we need to know outside the physical sciences can be obtained without empirical investigation. Mises never said this. No reasonable person actually thinks this.
When exploring the relationship between Mises’s Austrian methods and quantitative methods, it’s important to note that the scope of economics is quite limited. Indeed, Mises in his writings repeatedly explains how many different types of human thoughts and experiences are the domain of other fields such as psychology.
However, many people today confuse economics with many fields that are not actually economics as understood by Mises. For example, Mises writes that “Statistics is a method for the presentation of historical facts concerning prices and other relevant data of human action. It is not economics and cannot produce economic theorems and theories. The statistics of prices is economic history.”
He also writes: “Economic history, descriptive economics, and economic statistics are, of course, history.”
In other words, much of what people generally regard as economics is, in fact, history. This doesn’t mean that history in contemptible. Mises was no enemy of historians. It’s just that for Mises, a study on bank reserves in, say, 1935, is not an economic study. It’s an historical study. It’s economic history. It’s descriptive economics—as opposed to Mises’s economics which is explanatory economics.
It’s also worth noting that many Austrians, Mises included, were fond us consulting historical information and even writing about it.
As Per Bylund has noted:
Empirical studies (”history”) are important in Austrian economics and have larger scope than in mainstream economics. Mises worked with applied research in the Vienna Chamber of Commerce and founded the Austrian Institute for Business Cycle Research, for which he appointed Hayek as the first director. This is where Hayek did much of the business cycle research that later won him the Nobel Prize. What critics fail to understand is Austrians’ narrower definition of theory, which is not a collection of hypotheses but true, general statements. Austrian economic theory cannot be developed using incomplete and imprecise measurements of observations. But this does not mean Austrians cannot or will not do empirical research.
Murray Rothbard, of course, wrote a large number of books and articles that were history books and history articles.
When Quantitative Methods Are Very Helpful
Economic history and statistical studies on topics related to economics are good things. Here’s an example of how these studies are essential:
The anticapitalist left often likes to claim that the world has been dominated by free market economics for decades and that policymakers around the world have put in place a large amount of libertarian policies that favor so-called “neoliberal” economies virtually free from government intervention.
Is this true?
Well, we can’t come to a conclusion using economic theorems or deductive reasoning. The question is an empirical one: have political institutions been laissez-faire in recent decades? To answer the question, we must examine tax revenues, tax rates, government regulations, and the total number of government employees. We might have to compile large amounts of data.
If the data does indeed show that governments have been collecting more tax revenue and regulating markets more now than in the past (as is in fact the case), then we can say, “The empirical evidence shows that the anticapitalists are wrong. The world is not run by free market libertarians.”
[Read More: “Six Graphs Showing Just How Much the Government Has Grown“ by Ryan McMaken]
Clearly, we could not have answered this question without employing some quantitative methods.
Moreover, empirical data might be key in illustrating how certain economic laws and axioms are true.
For example, Mises’s work shows us that economic booms and busts are an artifact of artificial credit expansion. This is an economic law explained and discovered through Mises’s deductive methods.
But this by itself doesn’t do us much good in discussing political economy when an anticapitalist says, “The 2009 financial crisis was caused by the fact financial markets are virtually unregulated! Also, the crash happened because of greed!”
Now without economic history and empirical information, its very difficult to offer a rebuttal to this anticapitalist claim except in a very general theoretical way. On the other hand, if we do in fact show that financial markets are anything but “virtually unregulated” and by pointing out that there is no evidence that “greed” suddenly became more widespread in 2007, then it’s much easier to illustrate that Mises’s theory about credit expansion is actually a better theory that fits the facts. The empirical question of greed and regulation does not “prove” or disprove Mises’s theory to be correct. The empirical information does help illustrate, however, that’s Mises’s theory fits the facts.1
So, let’s abandon forever this idea that Austrians “hate math” or that quantitative methods are useless or somehow contemptible. They’re not. They’re just not key in understanding and formulating good economic theory. And that’s okay.
- 1Of course, sometimes experience doesn’t illustrate that a theory is a good fit. Maybe the opposite is true and experience suggests again and again that maybe we need to rethink a theory. Again, this empirical experience doesn’t in itself prove or disprove a theory—because there are too many variables in social science to draw precise conclusions. But the sort of expirical information gleaned through economic history can indeed suggest refinements to a theory are necessary.
