Looking Back at the Austrian Revival
By Adam Martin
- A Liberty Classic Book Review of The Foundations of Modern Austrian Economics, edited by Edwin Dolan.1
What’s so Austrian about “Austrian economics?” The label was originally a pejorative, coined by Gustav Schmoller, a harsh critic of Carl Menger’s work. It was an attempt to attach Menger’s ideas to a “provincial” or unsophisticated area. If he were writing today, Schmoller might have called Menger’s theory “Cleveland economics.” But by the time of Ludwig von Mises and Friedrich Hayek, the label was a historical one. Mises writes in 1969:
- But after some years all the essential ideas of the Austrian School were by and large accepted as an integral part of economic theory. About the time of Menger’s demise (1921), one no longer distinguished between an Austrian School and other economics. The appellation “Austrian School” became the name given to an important chapter of the history of economic thought; it was no longer the name of a specific sect with doctrines different from those held by other economists. (Mises 1969, p. 19)2
This view from 1921 proved to be overly optimistic. In the coming decades, the profession at large sided with Oskar Lange and Abba Lerner over Mises and Hayek in the socialist calculation debate, sided with John Maynard Keynes over Hayek about the causes and remedies of the Great Depression, and came to view the work of Austrian economists as an ideological carryover from an older age. If mainstream economists saw something of value in the work of Menger and his intellectual heirs, they thought it had been incorporated or surpassed in the further development of economic theory.
By the early 1950s, the reigning paradigm was Paul Samuelson’s neoclassical synthesis, which combined neoclassical price theory on the micro-level with Keynesian aggregate expenditure theory on the macro-level. Economics was heavily mathematized, not only as a lens for understanding but also as a tool of social engineering. This approach to economics dominated both academics and public policy for the next several decades. Deviations from perfectly competitive outcomes were to be corrected by regulation, and deviations from full employment were to be “fine-tuned” out of existence through fiscal and monetary policy.
This is the intellectual context in which the Institute for Humane Studies, in 1974, hosted a conference at South Royalton, Vermont on Austrian economics. The primary lectures at the program were given by Israel Kirzner, Ludwig Lachmann, and Murray Rothbard. The Foundations of Modern Austrian Economics, edited by Edwin Dolan, collects these lectures in written form along with an introduction by Dolan and a chapter by Gerald O’Driscoll and Sudha Shenoy on stagflation.
Foundations aims to introduce readers to what makes the modern Austrian school distinctive from Samuelson’s neoclassical synthesis (Dolan, p. viii). Though the “Austrian” moniker had been used informally by younger economists inspired by the work of Menger, Mises, and Hayek, around this time it started appearing as a label for a distinctive school of thought. Combined with Hayek’s 1974 Nobel prize win, this period is sometimes marked as the “Austrian Revival.”3 At the conference, participants debated whether to consciously adopt the label to distinguish their approach from the Samuelsonian mainstream. Famously, Milton Friedman stopped by the conference and counseled the participants, “there is no Austrian economics—only good economics, and bad economics” (quoted by Dolan, p. 4).
Let’s set aside the question of whether the “Austrian” label is useful. Like all labels, it is good for some purposes and bad for others. Rather, I would like to explore whether the comparisons between Austrian and mainstream economics scattered throughout this volume hold up. Foundations is very much a product of its time. After decades of government aggregate demand management, the 1970s saw both high inflation and high unemployment (stagflation). The conventional models said this combination should be impossible. The younger Austrians saw an opportunity to challenge Samuelson’s reign. But the mainstream economics of the 21st century is not the same as it was in 1974. While many elements remain similar, especially in undergraduate education, how relevant are the comparisons in Foundations today?
Kirzner, Lachmann, and Rothbard all argue that the distinctive task of Austrian economics is to render social phenomena intelligible as the result of individual plans and purposes. They each have a different focus, but all of them are concerned with this basic idea. Rothbard is concerned with distinguishing economic reasoning, which proceeds from the “fundamental axiom of action,” from mathematical and econometric approaches to understanding social phenomena. The axioms on which economics is built are “true and meaningful” from the outset of analysis (p. 22, emphasis added), in contrast to physics for which the axioms only become meaningful via prediction and control. Lachmann distinguishes between two tasks of economics (pp. 215-216): making the world intelligible and tracing the unintended consequences of action.4 He argues that Austrian school has always had more affinity with the former task.
What does it mean to make the social world intelligible? How is this different from the way most economists understand their task? Kirzner offers a helpful parable of a man from Mars (p. 45). The observer, Marvin-the-Martian-style, looks down in his telescope and sees bodies entering and exiting boxes at certain times of day. This activity is repeated with certain regularity, with pronounced bursts of activity twice a day. The observer develops complex statistical models that effectively predict the flow of bodies into and out of boxes. But what the observer does not know is that these are people getting onto and off of buses headed to and from work in the morning and evening.
Kirzner’s example highlights a distinction that James Buchanan would later make between explanation and prediction.5 Prediction as most economists understand it is usually “retrodiction.” It means accounting for historical relationships between observable phenomena and usually does not involve forecasting the future. But as the parable of the Martian observer illustrates, successful prediction is not the only form of knowledge we can have about the social world. We can also understand the why of social phenomena, their explanations. This, the Austrians argue, is what is missing from a mathematical approach to economics.
A related point, mentioned by Lachmann (p. 219) and elaborated on later by Buchanan, is that explanation involves appeal to purposeful human action while prediction reduces action to reaction. To predict social behavior, the theorist must make assumptions about how individuals will choose. The mere fact that they will choose purposefully is not enough. Individual choosers in a predictive, mathematical model respond to changes in circumstances according to a script that is already written by their utility functions or scales of values. According to Mises and Buchanan, these scripts are fictions—but they are necessary fictions for making predictions or grasping certain principles.
With this in mind, it is easy to see what the Austrian critiques of mainstream economics throughout this book get right and what they get wrong. What they get right is that explanation is a legitimate goal for social science, and that it is not reducible to making tractable predictions. Attention to the human concepts of meaning and purpose pays dividends when we try to understand the social world around us. In the words of Fritz Machlup, “if matter could talk” physicists would be foolish not to listen. Economists study subjects that can think, act, and speak purposefully, so we should listen as well.
Paying attention to human purposefulness is, moreover, a necessary task when explaining the origins of social phenomena. Rothbard touches on this in his lecture on the Austrian theory of money when he references Mises’s “regression theorem” (pp. 168-9). Mises’s attempt to explain the origins of money confronts a dilemma: the demand to hold money depends on its purchasing power, but its purchasing power depends on existing money prices. Prior to the existence of money, those prices do not exist, so how can a monetary economy get off the ground? Mathematical models will not solve this problem, because such models rely on the timeless logic of mutual determination.6 The price of one good, including money, depends on the prices of all others. The regression theorem solves this problem by recognizing that whatever good becomes money would have had a set of barter prices prior to it becoming money, thus giving it an initial level of purchasing power.
The substantive contributions of Austrian economics to social science have typically focused on these sorts of origin stories: the origins of money, price formation (as opposed to price determination), the causes of business cycles, the dynamics of interventionism, etc. Some of these origin stories not only offer distinct claims about how the world works but are also policy relevant. Kirzner’s theory of the market process (pp. 115-125) has strong implications for antitrust policy. Rather than assuming that producers know the least-cost method of production, Kirzner argues that these methods are only discovered when entrepreneurs compete for profits. (Note that this requires a conception of humans as creative, not merely reactive.) If Kirzner is right, we cannot determine whether a market is competitive simply by counting the number of firms. Rather, we need to focus on potential barriers to entrepreneurial entry into the market.7
Perhaps some or all of these Austrian theories are mistaken. But they are concretely distinct from mainstream economics. They offer substantive contributions or challenges to the way that economics is typically done. But there are a number of complaints about mainstream economics in Foundations that fail to meet this bar by misunderstanding the claims that most economists would make about their own arguments.
Rothbard (pp. 32-33) argues against the use of econometrics primarily on the grounds that historical events are heterogenous. Statistics apply to homogenous entities that can be grouped and counted.
- Econometrics not only attempts to ape the natural sciences by using complex heterogeneous historical facts as if they were repeatable homogeneous laboratory facts; it also squeezes the qualitative complexity of each event into a quantitative number and then compounds the fallacy by acting as if these quantitative relations remain constant in human history. (p. 33)
My colleagues in the Department of Agricultural and Applied Economics at Texas Tech would be surprised to learn that they treat the quantities they measure—such as the elasticity of demand for cotton—as historical constants. If this were the case, it is not obvious why the department would continue to exist after the first estimates were produced.
Applied economists know perfectly well that they are not measuring historical constants. They understand that they are either making measurements regarding particular historical episodes or that the results of their work are rough generalities. They are sensitive to the heterogeneity of the objects of their studies. They know that they rely on imperfect proxy variables. And many, if not most, understand that econometrics should not be divorced from economic theory.
Dolan, in his introduction, echoes Rothbard’s sentiments, arguing that “[a]cceptance of the Austrian paradigm entails a radical rejection of econometrics as a tool of economic theory” (p. 7). But he allows, later on, that this view “leaves open the possibility that econometrics could be a legitimate tool of technology and history” (p. 15). This concession brings his view closer to that of Mises, which Rothbard quotes in his lecture on methodology (p. 34). So what we really have here is a semantic issue. Econometricians know perfectly well that they are not doing economic theory. But does “economics proper” only include theory, or does it also include economic history, business administration, or the other tasks that economists perform? As long as the distinctiveness of the economic way of thinking is maintained, it is not clear what is at stake in answering this question.
The basic error that several of these essays make is to treat the explanation and prediction as substitute tasks rather than complementary ones. Some social scientific questions are about relationships between observable variables. Some are about the origins of social phenomena. Neither task can be reduced to the other, but a sound approach to social science uses each to reinforce the other. Austrians are perfectly comfortable using empirical generalities as part of the basis of theorizing (for example, the idea that more leisure is ceteris paribus preferred to less). These empirical generalities are typically said to come from common sense or everyday observation, but there is no reason they cannot be supplied by statistical studies. And while understanding human purposiveness underpins explanations of historical episodes, oftentimes there are multiple theoretically plausible explanations of those episodes. Econometrics can help figure out which explanation best accounts for a particular event.
The Austrians in Foundations were right to raise the alarm that the concepts that allow economists to analyze the origins of complex phenomena were (and probably still are) underemphasized by most economists. But they were also wrong to try to force a once-and-for-all choice between explanation and prediction. The same can be said for their rejection of mathematical economics. Yes, math has limitations and is not always necessary. No, individuals do not have utility functions that are differentiable (p. 24). But nor do they have an ordinal scale of values of the sort that Austrians appeal to. Mises makes clear that the scale of values is a mental tool of theorizing and does not exist in reality. While there is great value in the verbal and strict approach that Austrians take to discussing diminishing marginal utility (it offers an explanation for why utility diminishes on the margin), it is not clear why it is only acceptable to use a verbal imaginary construct rather than a mathematical one. Supply and demand is, after all, a mathematical construct.
One concern that Austrians might have with taking a more conciliatory attitude toward math and metrics is that these are the tools of social engineering. That was certainly part of the intellectual climate of the early 1970s. Gerald O’Driscoll and Sudha Shenoy document a number of famous economists (including Milton Friedman) declaring that aggregate demand management had made recessions and/or depressions a thing of the past (p. 203). Clearly the prediction did not hold up well. But while some modern economists still hold onto the view that the job of economists is to manage or control the economy, this attitude appears far less prevalent today. And it would be foolish to reject the usefulness of a tool because it can be misused. Verbal logic can also underwrite terrible policies, but few call for getting rid of it.
I have only touched on one of the main themes running through this volume: the relationship between Austrian and mainstream economics. There are several other fascinating threads for those either interested in Austrian ideas or the history of economic thought more generally. Rothbard and Kirzner each have lectures exploring the relationship between moral philosophy and Austrian economics. The volume presages internal debates that Austrian would have over the coming decades, such as the Kirzner-Lachmann debate over the tendency towards equilibrium and the fight between free bankers and those hostile to fractional reserve banking. Not all of these debates were particularly productive, but it is interesting to see the seeds being sown here. In particular, Kirzner’s chapters on methodology and the market process remain an excellent introduction to some key Austrian ideas.
Paul Samuelson’s specter still looms large in the economics profession. But most economists are not committed philosophical Samuelsonians. This is progress that should be celebrated. The relationship between Austrian and mainstream economics has changed over the past five decades because economics has changed. While I am entirely on board with the broad contours of the intellectual project laid out here, not all of the arguments have held up well.
The more important point of contention in the present day is whether economic theory will continue to underwrite academic research at all or whether economics will devolve into data science on top of a theoretical blank slate. In this, committed neoclassical economists and Austrians probably have more in common than they do in dispute. Austrians should follow their own advice: neoclassical economists can talk. We should listen.
 Ludwig von Mises, The Historical Setting of the Austrian School of Economics. Available online at mises.org.
 It may surprise contemporary readers that “spontaneous order” appears nowhere in this volume. Hayek’s Law, Legislation, and Liberty was still hot off the press.
 James Buchanan (1982), “The Domain of Subjective Economics: Between Predictive Science and Moral Philosophy,” in Method, Process, and Austrian Economics, ed. Israel Kirzner. Lexington, MA: Heath.
 This is not necessarily true of modern agent-based models.
 These implications are not spelled out in the Dolan text, which focuses on advertising. They are discussed in Kirzner’s 1973 Competition and Entrepreneurship (Chicago: University of Chicago Press.)