By Frank Shostak*
While in the natural sciences a laboratory experiment can isolate various elements and their movements can be followed through, there is no equivalent in the economic discipline. The introduction of econometrics and model building is an attempt to produce a laboratory where controlled experiments can be conducted.
The idea of having such a laboratory is very appealing to economists and politicians, since once the model is built and endorsed as a good replica of the economy, politicians can evaluate the outcomes of various policies.
This, it is argued, will enhance the efficiency of government policies and thus lead to a better and more prosperous economy. It is also suggested that the model can serve as a referee in validating various economic ideas.
Apart from assessing the impact of various policies, the other purpose of a model is to provide an indication regarding the future.
By means of mathematical and statistical methods, a model builder establishes relationships between various economic variables.
For example, personal consumer outlays are related to personal disposable income and interest rates, while fixed capital spending is explained by the past stock of capital, interest rates, and economic activity. A collection of such various estimated relations—i.e., equations—constitutes an econometric model.
A comparison of the goodness of the fit between the dynamic simulation and the actual data is an important in assessing the reliability of a model. (In a static simulation, the model is solved using known lagged variables. In a dynamic simulation, the model is solved by employing its own generated lagged variables).
The final test of the model is its response to a policy variable change, such as an increase in taxes or a rise in government outlays. By means of a qualitative assessment, a model builder decides whether the response is reasonable or not. Once the model is successfully constructed, it is ready to be used.
Are Mathematical Models Valid in Economics?
By applying mathematics, mainstream economics is attempting to follow in the footsteps of the natural sciences. In the natural sciences, the employment of mathematics enables scientists to formulate the essential nature of objects.
By means of a mathematical formula, the response of objects to a particular stimulus in a given condition is captured. Consequently, within these given conditions, the same response will be obtained time and again.
The same approach, however, is not valid in economics, for economics is supposed to deal with human beings and not objects. According to Mises in Human Action,
The experience with which the sciences of human action have to deal is always an experience of complex phenomena. No laboratory experiments can be performed with regard to human action.
The main characteristic or nature of human beings is that they are rational animals. They use their minds to sustain their lives and well-being. The usage of the mind, however, is not an automatic procedure, but rather every individual employs his mind in accordance with his own circumstances. This makes it impossible to capture human nature by means of mathematical formulae, as is done in the natural sciences.
To pursue quantitative analysis implies the possibility of the assignment of numbers, which can be subjected to all of the operations of arithmetic. To accomplish this, it is necessary to define an objective fixed unit.
Such an objective unit, however, does not exist in the realm of human valuations. On this Mises wrote in Human Action, “There are, in the field of economics, no constant relations, and consequently no measurement is possible.” There are no constant standards for measuring the minds, the values, and the ideas of men.
People have the freedom of choice to change their minds and pursue actions that are contrary to what was observed in the past. Because of the unique nature of human beings, analyses in economics can only be qualitative.
Individual goals or ends set the standard for valuing the facts of reality. For instance, if the goal of an individual is to improve his health, then he will establish which goods will benefit his health and which will not.
Among those that will benefit him, some will be more effective than others. There is no way, however, to quantify this effectiveness. All that one could do is rank these goods in accordance with perceived effectiveness.
The use of mathematics in economics poses another serious problem. The employment of mathematical functions implies that human actions are set in motion by various factors.
For instance, contrary to the mathematical way of thinking, individual outlays on goods are not “caused” by real income as such. In his own context, every individual decides how much of a given amount of income will be used for consumption and how much for savings.
While it is true that people respond to changes in their incomes, the response is not automatic, as depicted by a mathematical formula.
An increase in an individual’s income does not automatically imply that his consumption expenditure will follow suit. Every individual assesses the increase in income against the goals he wants to achieve. Thus, he might decide that it is more beneficial for him to raise his savings rather than raise his consumption. From this perspective an econometric model, which is a collection of various equations, is a misleading description of the real world of human beings. In the world of econometric models, individuals are reduced to robots that respond mechanically to a change in various driving variables.
Why Probability Distribution Is Not Relevant in Economics
The econometric model building in addition to mathematics also employs probability. What is probability? The probability of an event is the proportion of times the event happens out of a large number of trials. For instance, the probability of obtaining heads when a coin is tossed is 0.5. This does not mean that when a coin is tossed ten times, five heads are always obtained.
However, if the experiment is repeated a large number of times then it is likely that heads will be obtained 50 percent of the time. The greater the number of throws, the nearer the approximation is likely to be.
In economics, we do not deal with homogeneous cases. Each observation is unique. Consequently, no probability distribution can be established. (Again, probability distribution rests on the assumption that we are dealing with homogeneous cases.)
Let us take, for instance, entrepreneurial activities. If these activities were homogeneous, with known probability distributions, then we would not need entrepreneurs.
An entrepreneur is an individual who arranges his activities toward finding out consumers’ future requirements. People’s requirements however, are never constant with respect to a particular good.
Since entrepreneurial activities are not homogeneous, this means that probability distribution for entrepreneurial returns cannot be formed.
The assumption that mainstream economics makes, that probability distribution is valid in economics, leads to absurd results, for it describes not a world of human beings who exercise their minds in making choices, but machines.
The employment of probabilities implies that a random process generated the various pieces of economic data, similarly to tossing a coin.
Note that random means arbitrary, i.e., without method or conscious decision. However, if this were the case, human beings would not be able to survive for too long. In order to maintain their lives and well-being, human beings must act consciously and purposefully. They must plan their actions and employ suitable means.
Other Issues in Using Econometric Models
Given that, human beings are governed by freedom of choice, various policy analyses by means of models, known as “what if” or multiplier analyses, are likely to generate questionable results.
In conducting the “what if” experiment, as a rule, a model builder utilizes a given model whose equations’ parameters remain intact. This is however, questionable. For instance, say the model builder wants to evaluate the effect of a change in government outlays on various markets. It is quite likely that a change in government outlays will affect the parameters of various equations. If the model builder were to ignore this and leave the structure of equations intact, this would mean that individuals in the economy ceased to be alive and were, in fact, frozen. On this Mises writes in The Ultimate Foundation of Economic Science:
As a method of economic analysis econometrics is a childish play with figures that does not contribute anything to the elucidation of the problems of economic reality.
Another major problem with most large-scale econometric models is that they are designed along the lines of Keynesian thinking. The main variable in these models is gross domestic product (GDP), which is explained within the model framework by the interactions between various lumped data, known as aggregates.
The interaction between various aggregates in the model framework gives the impression that the economy is about gross domestic product, not about human beings and human life. Obviously, this runs contrary to the fact that everything in the human world is caused by man’s purposeful conduct.
To improve an econometric model’s capability as a forecasting tool, the predictive capability of each equation in the model is tested against the actual data. The difference between the actual data and the data obtained from the equations, i.e., the error term (also known as the add factor), is extrapolated forward and incorporated into the model’s equations.
In many instances the forecast produced by an econometric model is heavily influenced by the add factor, which allows the model builder to force the outcome of the forecast in line with his “gut feelings.” All this casts doubts on the scientific procedures employed by econometric modeling.
Source: This article was published by the MISES Institute