Moore, Henry L.
Moore, Henry L.
Henry Ludwell Moore (1869–1958), American economist, made the first major attempts to combine economic theory and statistical techniques in the empirical estimation of theoretical economic relationships. Quantitative estimates of elasticities of demand and supply, of productivity changes and of the nature of strikes, of cost curves and of determinants of wage rates are so prominent in— even characteristic of—modern economics that it is difficult to remember that they were initiated only in the present century, and by Moore more than by any other economist.
Moore’s life was that of a scholar wholehearted in his devotion to research. After receiving his PH.D. from Johns Hopkins University in 1896, he began teaching at Smith College, and in 1902 he went to Columbia University, where he remained until 1929; aside from two terms in Karl Pearson’s statistical laboratory (in 1909 and 1913), he had no important association with any other institution or type of activity. His premature retirement was due to poor health.
His first professional interest was the history of economic thought. Moore’s dissertation was a competent survey of the vast literature on von Thünen’s celebrated natural rate of wages, , where a is the subsistence of a workingman’s family and p the total product per workingman (1895). A second study was devoted to Cournot, the great pioneer of the mathematical method in economics (1902). Soon, however, his interests shifted to what he called statistical economics and is now more commonly called econometrics.
His first publication in statistical economics was a set of essays, all connected with labor, Laws of Wages (1911). Several of the essays were devoted to a verification of the marginal productivity theory, as applied to the pattern of average wages in coal mining over time and space and to the differences between wages of individuals. In general these investigations displayed careful and (for that time) sophisticated statistical methodology, but the hypotheses were very loose in their theoretical formulation. Moore’s contemporaries properly applauded the purpose and criticized the execution of these studies.
A second set of essays in this first volume proposed empirical uniformities, for which theoretical explanations might then be sought. One was a demonstration that within an industry or occupation, the larger the establishment, the higher the wage rates—a finding confirmed by later research. A second was an attempt to measure the influence of unions on the outcome of strikes, with much more ambiguous results.
Three years later Moore’s Economic Cycles (1914) launched the most important of all his work, the empirical estimation of theoretical relationships. Yet these estimates were essentially only by-products of Moore’s search for a truly fundamental explanation of fluctuations in the level of economic activity. The main theme, as he saw it, is as follows:
The principal contribution of this Essay is the discovery of the law and cause of Economic Cycles. The rhythm in the activity of economic life, the alternation of buoyant, purposeful expansion with aimless depression, is caused by the rhythm in the yield per acre of the crops; while the rhythm in the production of the crops is, in turn, caused by the rhythm of changing weather which is represented by the cyclical changes in the amount of rainfall. The law of the cycles of rainfall is the law of the cycles of the crops and the law of Economic Cycles, (p. 135)
The support for this bold claim consisted of four steps:
(1) The discovery of several cycles—the chief being of eight years’ duration—in rainfall in Ohio.
(2) The argument that the rainfall cycles lead to cycles of equal duration in yields per acre (but lagged by half a cycle).
(3) The demonstration that yields per acre are inversely related to the prices of the grain products.
(4) The argument that demand curves for agricultural products shift upward in periods of rising industrial prices.
If, as Moore for a time believed, the demand curve for pig iron (a typical industrial good) is positively sloped and the volume of pig iron falls when crops decline, then the price of pig iron falls when crops are small, thus lowering the demand for the crops. The cycle in rain has thus been carried through to the cycles in outputs and prices of industrial goods.
Only the third step in this argument, in which statistical demand curves are estimated, was well founded, and it was this part of Moore’s work which had the major impact on economics. Moore’s first demand curve, that for corn from 1866 to 1911, illustrates his procedures. The historical series of prices and outputs are influenced by increasing population and fluctuations of price levels, so the annual price and quantity are expressed as ratios to the previous years’ prices and quantities (link relatives). A linear demand equation then yields an elasticity of -1.12 (with r = -.789); a cubic equation yields an elasticity of —.92. Moore examined demand functions for periods of rising and of falling general prices—they differed little— but never introduced prices of substitutes or income (for which no satisfactory data existed).
The work on demand curves was extended in Forecasting the Yield and the Price of Cotton (1917). Here he developed predictions of the size of the cotton crop on the basis of early-season rainfall which were more reliable than the elaborate crop forecasting system of the U.S. Department of Agriculture. Subsequently Moore introduced the concept of the flexibility of prices (the relative change in price divided by the relative change in quantity—the reciprocal of the elasticity of demand in a two-variable relationship) and the partial elasticity of demand.
Two years later Moore extended this type of analysis to supply curves by correlating percentage changes in acreage with percentage changes in prices a year earlier (1919). This analysis assumes that farmers predict that this year’s price (or price change) will continue next year, and this approach led Henry Schultz (who was Moore’s chief disciple) to formulate the cobweb analysis.
Until 1923, however, Moore continued to consider his research on the extraterrestrial theory of cycles of primary importance. He found eight-year cycles almost everywhere and ultimately attributed them to the transits of Venus. Eventually he accepted the futility of this work, and he stopped working on the subject shortly after the book Generating Economic Cycles (1923) appeared.
Moore’s final book, Synthetic Economics (1929), proposed the boldest of programs: the statistical estimation of Walras’ equations of general equilibrium. But although Moore’s vision continued to be superlatively farsighted, he had not the power to translate this vision into a workable research program.
The increasing rigor of economic theory, the expanding arsenal of statistical techniques, and the increasing intervention of the state in economic life all contributed to the cordial reception of Moore’s work. The 1920s saw an extensive application of his techniques to agricultural products, and from this base empirical estimation of theoretical functions has spread over the entire corpus of economics.
[For the historical context of Moore’s work, see the biographies ofCournotandThÜnen. For discussion of the subsequent development of Moore’s ideas, seeDemand And Supply, article on Econometric Studies; Econometrics; Time Series, article on Cycles; and the biography ofSchultz.]
WORKS BY MOORE
1895 Von Thünen’s Theory of Natural Wages. Quarterly Journal of Economics 9:291–304, 388–408.
1902 Antoine-Augustin Cournot. Revue de métaphysique et de morale 13:521–543.
1911 Laws of Wages: An Essay in Statistical Economics. New York: Macmillan.
1914 Economic Cycles: Their Law and Cause. New York: Macmillan.
1917 Forecasting the Yield and the Price of Cotton. New York: Macmillan.
1919 Empirical Laws of Demand and Supply and the Flexibility of Prices. Political Science Quarterly 34: 546–567.
1923 Generating Economic Cycles. New York: Macmillan.
1929 Synthetic Economics. New York: Macmillan.
SUPPLEMENTARY BIBLIOGRAPHY
Stigler, George J. 1962 Henry L. Moore and Statistical Economics. Econometrica 30:1–21. → Contains a complete bibliography of Moore’s work on pages 19–21, and references both to the leading reviews by his contemporaries and to the work on statistical demand curves in the 1920s.