All Commentary
Tuesday, June 1, 2004

Fortune-Cookie Economics

The Laws of Human Action Form the True Basis of Economics

It is little wonder that some economists want to be perceived as having fortune-telling abilities. After all, history and literature are full of examples of revered and exalted prophets and oracles. So we can understand why many economists sit up straight, clear their throats, look us right in the eye, and foretell next year’s change in GDP or holiday-season spending.

The ability to foretell the future accurately appears for some to be the essence of true economic science. None less than Milton Friedman, one of the most influential economists of our time, famously stated that the “only relevant test” of a theory is to compare its predicted results with actual experience.1 Meanwhile, the trendy “econometric” school has flatly stated, “science is prediction.”

But science is really a quest for knowledge, and the method used varies, or ought to vary, according to the particular subject. When studying human economizing action, the natural scientific method of hypothesis-building, data-gathering, and hypothesis-testing is inappropriate. Human beings are fundamentally different from the matter studied by the physical sciences, and far less predictable.

Yet many economists today use this approach almost exclusively. First they create models that often include assumptions such as everyone has perfect knowledge, or everyone is seeking maximum material satisfaction. Then they set some arbitrary “parameters” for things like “consumer confidence.” Then they gather data, compute z-scores and other statistical analyses, and generate their “positive” predictions, which can be later tested against reality. And this is called “economics.”

Clearly the use of past data, statistics, and mathematics is important for anyone attempting to make budgetary or business decisions. For example, an investor considering buying a hotdog stand will want to know recent income and expenditure data for the stand, as well as recent trends in hotdog prices, seasonal sales, the present prevailing rate of return on other investments, and any other data deemed useful to making the decision.

Yet data from past economic experience, such as last year’s wheat prices or December’s unemployment figures, are always snapshots of economic history, nonrepeatable data distilled from highly complex chains of events. They can be used in economic calculation, indeed they are critical to it, but they are not the source of economic truth forever and for all time. Only deductive conclusions starting from some fundamental truth about human beings can provide such knowledge. That is what economics is all about.

Economics begins with the understanding that to act is to seek goals, and these goals are based on human valuations, which involve ranking a over b, b over c, and c over d. Indeed, valuations of this sort drive all human action. And while we might be able to measure, calculate, and predict with near certainty the responses to various stimuli of things we study in the natural sciences, we cannot do the same with people. Human valuations are purely subjective and ever subject to change. We might know with near-apodictic certainty that water will freeze at zero degrees Celsius; we cannot know with the same degree of certainty how my wife will respond to a gift of a dozen roses or how consumers will respond to some new product. (While I might hope my wife would love a dozen roses, that is not necessarily so, especially if I send them one week after our anniversary.)

Ludwig von Mises put the differences between the natural sciences and the sciences of human action—such as economics—this way: “What distinguishes the sciences of human action [from the natural sciences] is the fact that there is no . . . foreknowledge of the individuals’ value judgments, of the ends they will aim at under the impact of these value judgments, of the means they will resort to in order to attain the ends sought.. . . We know something about these things, but our knowledge of them and about them is categorically different from the kind of knowledge the experimental natural sciences provide about natural events.”2

Cause and Effect

Of course, in a sense, it can be said we use economics to “predict” the future. For example, we can say with certainty that increasing the minimum wage to $10 an hour will, other things remaining equal, result in new involuntary unemployment, or that rent controls will result in deterioration of existing apartment units, as well as a drop in rentable space. But these “predictions” are not based on empirical observation; in fact they are not really “predictions” at all. They are simply statements of cause and effect linked by logical reasoning to the fundamental and incontestable truth that humans act purposefully to achieve their goals.

Wages and rental rates are simply reflections of consumer valuations. They are not really expressed in dollars and cents but rather in other goods and services available (and therefore competing with them) in the market. When artificial barriers are erected around certain goods or services, consumers and producers respond to the new data. We cannot know exactly, in any quantitative sense, how they will respond, but we know that when a good is overpriced (other things remaining equal) a surplus will result; when another good is underpriced there will be a shortage.

Certainly, anticipating future events or conditions is fundamental to all human action. Indeed, if we did not believe we could change the course of future events—if we did not believe in the law of cause and effect—we would not act at all. But the key to understanding human action (past, present, or future) does not lie in historical experience or econometric modeling. It lies in uncovering the essential laws of human action, that is, in understanding true economics.


  1. Milton Friedman, “The Methodology of Positive Economics,” in Essays in Positive Economics (Chicago: University of Chicago Press, 1953), pp. 8–9.
  2. Ludwig von Mises, Theory and History (Auburn, Ala.: Ludwig von Mises Institute, 1985 [1957]), pp. 306–07.