Economic growth serves as the prominent standard for measuring the performance of an economy. However, what is published as the gross domestic product (GDP) does not represent production but reports overall spending. The calculation of economic growth is based on the nominal gross domestic product deflated by a price index.
Thus, the figure for real economic growth is subject to two distortions: the indicator does not measure production but reports expenditures, and, secondly, the obtained number is dependent on the techniques that are applied to the calculation of the respective price indices.
Economic growth figures can be determined in a fairly accurate way for an economy, which is in a primitive state and when only a few, easily identifiable and compoundable items are being produced, as it is the case with basic agricultural products. In the 1950s and 1960s it was thought that tons of steel could be used as a proxy for an objective estimate of economic performance. Nowadays, the figure for GDP gets all the attention, although the basis for its calculation its actually weaker than ever before.
Economic growth had its heyday with the spread of the social gospel that it is up to the State to guarantee general welfare by managing the economy and to actively redistribute income. In this context, economic growth was conceptualized as an increase in standardized goods production, and the increase—”growth”—of the output served as the criteria for the standard of living. It was for such aims that the modern system of national income accounting with the concept of economic growth at its heart was developed, and this measurement device has never lost its link to mass production.
National income statistics and macroeconomic models use as a premise the identity between spending and production based on the tautology that sold production equals expenditures. What is being calculated here is exactly this: income and – as its tautological counterpart – spending. However, production itself can only be measured in goods or in units of simple services. When heterogeneous goods and complex services get produced, overall aggregation is not possible in a non-monetary form.
The calculation of economic growth in terms of “real GDP” requires deflating the nominal values of expenditures. In order to do that, the statistical offices create a basket of goods and compare the prices of the goods in this basket to the respective reference periods. But there is no objective representative basket of GDP other than as a statistical construct based on many disputable assumptions, and there is no common standard (as a tertium comparationis) which would allow the comparison of one period’s production to the other when in fact current output in terms of new, obsolete and modified goods and services is quite different from that of the past.
One does not need to resort to more extreme examples like how to measure today’s musical output and compare it in a quality-adjusted form to that of the past. The measurement problem appears also when trying to give a percentage change for the output of software programs or administrative and engineering activities, not to speak of health, legal services, and education. The statisticians may answer that the “measurement” of output is derived from expenditures. However, money prices do not measure anything. Prices only have a meaning as relative prices as they reflect the exchange ratios on the market.
As Ludwig von Mises explained, “(t)he money equivalents as used in acting and in economic calculation are money prices, i.e., exchange ratios between money and other goods and services. The prices are not measured in money; they consist in money. Prices are either prices of the past or expected prices of the future. A price is necessarily a historical fact either of the past of the future. There is nothing in prices which permits one to liken them to the measurement of physical and chemical phenomena.”1
Adding up all sales or compounding all assets in an economy eliminates the meaning of prices. This kind of aggregation is different from what a company or a person does when calculating profits or the relative wealth position. When a person adds up the prices of his various assets, he gets a number about his current wealth relative to the price universe that he selects as his point of reference. For a company, it is sales, costs and profits that matter, and for that sound business accounting is required. Neither for personal matters nor for business decisions GDP figures are necessary.2
Few are aware that measuring the economy as a whole as it is intended by the GDP-concept owes its popularity to the cold war, and that its origins lie in the management of the war economies of the first half of the 20th century.3 Before World War I, economists worked in a tradition that was mainly for peace, free trade and for limited government. Thereafter, the outlook changed. With the experience of the industrialized warfare machinery and the expansion of the welfare state, economists found their new expanding field of activity in government, and consequently the dominant philosophy of the discipline changed from laissez faire to interventionism. It was in this context that the statistical and aggregate approach to economic issues gained its momentum.4
The managers of a war economy want to measure output and its growth, because the economy is put in the service of the war aims. The central planning authorities are presumed to be aware what goods and services are needed, at which proportions the factors of production should be allocated and to whom the results of production are to be distributed. Under such conditions the increase in output of the items determined by the planners can be ranked accordingly, and economic growth, as it is measured as an increase in output, serves as the indicator of economic performance.
In a private market economy the aims of economic activity are highly diverse and represent individual and subjective valuations. For an economy that is to serve multiple private needs, the calculation of economic growth makes little sense, if any at all. One may add up nationwide the various monetary prices of the goods and services that were sold, but besides the aggregation of the monetary values of diverse items—what is the true and reliable informational value of this exercise?5
Each good and service has a different value for each user, and there is no common standard of value available. This is even more so the case, when new products and new kinds of services come to the market. Valuations are not only heterogeneous among persons, but also differ for the same person according to the specific circumstances. Human beings have different needs and wants in different situations, and they experience changes of taste over time. Preferences themselves are experimental devices.
Quality is not an attribute inherent to the things, but it is a valuation, which is imputed to the goods and services by the economic actor. Economic action is directed at improvement, but what constitutes improvement is subject to continuous change. Therefore, there is no objective way to measure overall wealth in aggregate form without coarse distortions and without violating the basic principles of economic valuation.6
It is the prerequisite of measurement that there must be identifiable objects in the measuring space and that a corresponding fixed standard of measuring unit has to be applied. Barrels of oil can be measured at the well and it can be determined how much the production has grown or not. Measurement is per definitionem quantitative. In technical terms one may measure “quality” such as that of crude oil, for example, based on its sulfur content, but this measurement is also quantitative. In this case, the measurement indicates the usefulness of that good in terms of a criteria that is derived from an industrial process.
One can determine the weight of the overall output of a certain types of steel, but one cannot in the same way come to a reasonable result by measuring in one number the aggregate production of automobiles, of refrigerators, or of personal computers—not to speak about the problems one confronts when one tries to add up the output of teachers, nurses, songwriters or software programmers together with the production of apples and oranges.
A company can count its production in terms of units of model X or T. If the company wants a figure for the total, it must resort to sales. Before sales, one can only enumerate how many units of each specific item category are on stock, and only by assuming that the company’s products will catch certain prices, is it possible to calculate the expected monetary amount – but not the “value” of production.
Mises explained it quite clearly this way: “Prices are always money prices, and costs cannot be taken into account in economic calculation if not expressed in terms of money. If one does not resort to terms of money, costs are expressed in complex quantities of diverse goods and services to be expended for the procurement of a product.” Likewise one cannot add up values or valuations. “One can add up prices expressed in terms of money, but not scales of preference.”7
The more we move away from very basic goods, and have a more advanced and a dynamic, non-stationary economy with many heterogeneous goods and services, attempts to measure “the economy” become ever more complicated and finally these calculations lose even a rudimentary economic meaning. The concept of total output and its measurement and thus of economic growth is a statistical construct that loses its informational value for an economy characterized by a wide variety of goods and services and in which the production of new types of goods and services occurs, while many other items become obsolete.
The economy is not like one gigantic pumpkin that grows to maturity and whose size can be determined at each stage and compared from one season to the next. Also, the economy is not a cake that we all bake and then collectively consume. It is this pumpkin-like and cake-like understanding of economic activity that has provided the basis for most of the popular fallacies regarding production, distribution, and economic policy-making.8
For governments, using the figure for GDP as an indicator of economic performance has contributed to some of the most severe illusions of fiscal and monetary policy such as when spending for consumption is said to produce wealth or when government spending is said to boost economic growth as it happens—among others – with military expenditures.9
Times of war and the preparation for it come along with high economic growth rates. Another high economic growth period was certainly the time after a pharaoh had died in ancient Egypt and the economy was put under the command to erect a new pyramid. The fascist economy of Germany in the 1930s up to the end of World War II had terrific rates of economic growth. These periods are obviously quite different from those that were experienced in Britain during the industrial revolution, or in the United States in the late 19th century, or during West Germany’s “economic miracle” after World War II.
Currently, all eyes are on China’s magical economic growth rates and by that it is put in the first league of economic performance. However, China’s economic transformation is the result of a development dictatorship. What is being measured as high economic growth rates in China is quite different from that which takes place in the periods of economic transformation when economic development is guided by free markets and based on limited government—such as it is currently the case, for example, in Ireland, and what may happen more so in the future in Eastern Europe or India.
Economic growth as it is calculated as changes of real GDP is a very crude figure. Taken at the face value its informational value is highly deceiving. While there is a wide agreement among economists that economic growth figures do not indicate well-being, their use as a measurement of economic performance is fully en vogue. Economic growth as a performance figure seduces governments and many an investor when they do not differentiate between the causes of this growth and its consequences.10
Modern national income accounting is the outgrowth of industrialized warfare and of the interventionist welfare state. The theoretical counterpart was delivered by modern collectivist macroeconomics with its averages and aggregates. Even as of today, economic policy is still widely guided by the propositions of these theories with their statistical constructs that are said to interact mechanically with each other in a relationship of cause and effect.
The problem with economic growth goes beyond statistics. Approaching the economic problem in terms of “growth” and “stability” is probably the most severe obstacle against understanding the true nature of economic activity as an exchange-oriented action directed at the improvement of personal conditions. Economic growth as measured by GDP directs the policy maker to the lump sum of an imaginary output instead of allowing a market driven adaptation to the diverse wants of the individuals
In the context of a non-collectivist economic theory, economic growth, as it is measured by real GDP, has no place. Likewise, in a non-collectivist economic system, the focus would not be on “stable high economic growth”, but on the conditions of market exchange as the way to economic amelioration. Given that the criteria for assessing economic improvement are individual and subject to change, no guideline is adequate other than that there is an unhampered market and the protection of property rights.
In a non-collectivist economic system, the focus would not be on “stable high economic growth” as the oxymoronic expression says for the “common good” in economic policy. The individualistic economic theory focuses on the prevalent conditions of market exchange as the way to economic amelioration. In this view, what brings about improvement comes not by economic growth or by stability, but through economic transformation that is guided by the freedom of private initiative within an open market system.
The grand-scale interventions that are performed by monetary and fiscal policy in the name of growth and stability disrupt and misguide the plans for the individual, and they distort the decisions at the business level. The application of macroeconomic growth models has caused havoc when economic leaders naively adopted the interventionist creed and believe that it just takes the handling of a few economic policy instruments—like easy money or government expenditures – to achieve the blissful state of economic plenty.
Instead of its fixation on economic growth and stability, a non-interventionist system would favor the space that is given for the individual to demonstrate and actively pursue his preferences.11 The interventionist system, in contrast, puts the individual under a modern kind of serfdom where “output” or rather “expenditure” becomes the criteria. Economic growth puts a criterion of performance upon the individual that is detrimental to change and adaptation and to what was once called the “pursuit of happiness”. Not unlike the slave masters of the past, the modern interventionist state uses its levers to push the individual by incentives and constraints towards an obscure output that is called “economic growth”.
This article was originally published August 10, 2005.
1. Ludwig von Mises, Human Action . The Ludwig von Mises Institute. Auburn 1998, p. 218
2. Also for international comparisons the GDP figures are either deceiving or unnecessary. They were deceiving when the seemingly high growth figures of the Soviet Union were observed and they were unnecessary for the immigrants that came to the United States as the land of their choice in search of personal and economic freedom.
3. Both “fathers” of modern national income accounting, John Richard Nicholas Stone (Nobel Prize in economics 1984) and Simon Kuznets (Nobel Prize in economics in 1971) served in war planning offices where they developed and refined the concepts. Stone worked in the British government’s War Cabinet Secretariat and Kuznets was Associate Director of the Bureau of Planning and Statistics and Director of Research at the Planning Committee of the War Production Board in the U.S.
4. Concise overviews of this intellectual transformation can be found in the remarks by Robert W. Fogel at the Symposium on “The Role of Research Universities in Innovation, Social Mobility, and Quality of Life in the 20th Century” at the Association of American Universities () and by the article of James Galbraith, Chair of “Economist Allied for Arms Reduction” (EAAR): Notes on the Economics of War and Empire ( www.epsusa.org/publications/newsletter/nov2003/galbraith.htm). See also Murray N. Rothbard: The Politics of Political Economists: Comment, in The Quarterly Journal of Economics. February 1960, pp. 659-666, online at: mises.org/rothbard/polofpol.pdf
5. Kuznets himself was well aware of the shortcomings of national income accounting, as his intention was to obtain a measure for overall well being that would also include housework and leisure, a project that was doomed from the beginning in the eyes of the U.S. Department of Commerce when he assisted with the design of the national income statistics.
6. Market prices “are not expressive of equivalence, but of a divergence in the valuation of the two exchanging partners”, and the value attached to the unit of supply is subject to the law of diminishing marginal utility. See Ludwig von Mises: Human Action , op. cit., p. 699
7. Ludwig von Mises: Human Action , op. cit., p. 39 and p. 332
8. And for economic and financial analysis, too, one may add, considering that many financial ratios in common use apply the GDP number as a denominator for indicators, thereby neglecting that the registered value for GDP as the denominator of the ratio is not independent from the value of the enumerator. For example, in the phase of a credit expansion, the ratio of debt to GDP may not signal risk as both numbers in the fraction increase and make the ratio more or less constant.
9. As reported by Richard Vedder ( Statistical Malfeasance and Interpreting Economic Phenomena, in: The Review of Austrian Economics . Vol. 10, No. 2, 1997, pp. 77-89) the calculated output decline for the U.S. economy in 1946 was 20.6 percent as later statistical revisions suggested, but this reflected the “statistical fiction” that GDP was declining when in fact private employment and personal income were rising. The end of the wage-and-price controls meant a higher inflation rate, which in turn increased the recorded GDP price deflator (p. 82). One may only wonder what the government at that time would have done if the statistical information had already been available then.
10. n the 1970s, it was Paul Samuelson, who never got tired to present a graph in the various editions of his popular economics textbook that indicated that it would be just a matter of a few decades until the Soviet Union would take over the United States in “production”. Of course, Soviet production figures were humbug, like that shown by countries that were not Communist, but nevertheless were also managed by governments that pursued authoritarian State-guided development strategies at that time, like Brazil, for example, and a host of countries in Africa.
11. See Murray Rothbard: Toward a Reconstruction of Utility and Welfare Economics. (Toward a Reconstruction of Utility and Welfare Economics )