Hostname: page-component-586b7cd67f-r5fsc Total loading time: 0 Render date: 2024-11-28T15:56:35.204Z Has data issue: false hasContentIssue false
  • English
  • Français

Economic Causes of Late-Nineteenth Century Agrarian Unrest:New Evidence

Published online by Cambridge University Press:  03 March 2009

Robert A. McGuire
Robert A. McGuire
Affiliation:
Assistant Professor of Economics at Ball State University, Muncie, Indiana 47306.
Get access Rights & Permissions [Opens in a new window]

Abstract

Several economic historians have suggested that economic instability in American agriculture was a primary cause of agrarian discontent during the late nineteenth century. This paper, in providing a rigorous analysis of the issue, presents estimates of economic instability in agriculture and evidence on the location and intensity of agrarian unrest for 14 northern states from 1866 to 1909. I statistically test for a relationship between the two. The results strongly suggest that late-nineteenth century agrarian unrest was directly related to a state's degree ofeconomic instability. This conclusion holds for the entire last third of the century and for several subperiods.

Type
Articles
Information
The Journal of Economic History , Volume 41 , Issue 4 , December 1981 , pp. 835 - 852
Copyright
Copyright © The Economic History Association 1981

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 None of these scholars distinguishes between uncertainty and risk nor does the present paper. Both terms simply refer to the amount of variability in the economic conditions of agriculture. See the second section of this paper for further clarification.Google Scholar

2 Higgs, Robert, The Transformation of the American Economy, 1865–1914 (New York, 1971), pp. 102–03.Google Scholar

3 North, Douglass C., Growth and Welfare in the American past (Englewood Cliffs, N.J., 1974), pp. 134–36.Google Scholar

4 Parker, William N., “Agriculture,” in Lance E. Davis, et al., American Economic Growth: An Economist's History of the United States (New York, 1972), P. 407.Google Scholar

5 Neither Higgs, North, nor Parker Provides empirical support for his conjectures. Recently an economic historian analyzed the effect of income variability, among other factors, on farmer discontent but received rather poor results. However, as will be discussed later, his dependent variable—third-party votes—could be considered a poor measure of the location of farmer unrest and his explanatory variable—the variance of original time-series data—is an inappropriate measure of risk. See Klepper, Robert W., The Economic Bases for Agrarian Protest Movements in the United States, 1870–1900 (Unpublished Ph.D. Dissertation, University of Chicago, 1973).Google Scholar

6 A discussion of the characteristics of the data and the data sources are presented in detail in McGuire, Robert A., “U.S. Agricultural Statistics: State Estimates, 1866–1914”, Agricultural History, 54 (04 1980).Google Scholar

7 Gross income is simply the product of price and yield per acre. The impossibility of obtaining accurate state-level cost data for the crops used in this paper necessitated the use of gross income as a proxy for crop returns per acre. As has been argued elsewhere, this should not create any substantive problems. See McGuire, Robert A., “A Portfolio Analysis of Crop Diversification and Risk in the Cotton South,” Explorations in Economic History, 17 (10 1980), 369–70.CrossRefGoogle Scholar

8 Recently, scholars have challenged the importance of northern activity in the “agrarian revolt.” In separate studies, Lawrence Goodwyn and Robert McMath have argued that the true origin of the “revolt” was the South and that traditional historical accounts of agrarian discontent have misrepresented the historical experience of farmers in the North. A careful reading of their views, however, shows that they are interested in understanding a particular cultural phenomenon. Specifically, Goodwyn's interest is in explaining the Populist movement in terms of its democratic ethos—namely, a mass movement based on a cooperative attitude. McMath views the southern Alliance in terms of its fundamental source of popularity—its cooperative program and social settings. Nonetheless, while both argue that the “true” Populist culture was derived from the South, neither deny that there were many discontented, angry, and frustrated farmers in the North. See Goodwyn, Lawrence, Democratic Promise, the Populist Moment in America (New York, 1976)Google Scholar and McMath, Robert C., Populist Vanguard, A History of the Southern Farmer's Alliance (Chapel Hill, NC, 1975).Google Scholar

9 For an elaboration of the view that northern and southern farmer discontent should not be analyzed together, see Mayhew, Anne, “A Reappraisal of the Causes of Farm Protest Movements in the United States, 1870–1900,” Journal of Economic History, 32 (06 1972), 464–65.CrossRefGoogle Scholar

10 Higgs, Transformation of the American Economy, p. 102.Google Scholar

11 Mayhew, “Reappraisal of Farm Protest,” pp. 469–74.Google Scholar

12 North, Growth and Welfare, pp. 134–136; Parker, “Agriculture,” pp. 405–06; Higgs, Transformation of the American Economy, p. 102.Google Scholar

13 Parker, “Agriculture,” pp. 405–06.Google Scholar

14 Carstensen, Vernon, ed., Farmer Discontent, 1865–1900 (New York, 1974), pp. 53–58, 73–89;Google ScholarHicks, John D., The Populist Revolt (Minneapolis, 1931), pp. 85–95;Google ScholarMcMath, Populist Vanguard, pp. 67, 83–84, 156–57; Goodwyn, Democratic Promise, pp. 94–102.Google Scholar

15 One particular aspect of these early cooperative movements was related to economic uncertainty, the development of cooperative insurance plans. See Buck, Solon Justus, The Granger Movement, A Study of Agricultural Organization (Cambridge, 1913), pp. 211–89 andGoogle ScholarShannon, Fred A., The Farmer's Last Frontier (New York, 1945), pp. 300–26.Google Scholar For discussion of cooperative insurance plans during the Alliance Period, see McMath, Populist Vanguard, pp. 84–85, and Goodwyn, Democratic Promise, pp. 156–58.Google Scholar

16 For a detailed discussion of the view that fluctuating farm income produced “demands that ‘something be done’ by government or by voluntary association to stabilize farm incomes,” see Parker, “Agriculture,” pp. 409–16. Parker also includes a lengthy discussion of group organization without using government and group action employed to influence the government for the benefit of farmers.Google Scholar

17 The demands of agrarian organizations, which specifically concerned agricultural uncertainly, included the continual demand to prevent futures markets for crops, because farmers were convinced that crop speculation created price instability (see Buck, Granger Movement, pp. 211– 89; Carstensen, Farmer Discontent, pp. 74, 77, 79; Goodwyn, Democratic Promise, p. 81). Also, the demand for the subtreasury system directly concerned agricultural uncertainty because the “avowed intent of subtreasury commodity loans was… to stabilize short-run commodity prices.” (Goodwyn, Democratic Promise, pp. 572–73).Google Scholar

18 For discussion explaining that situations of risk depend on random outcomes, see Arrow, Kenneth J., Essays in the Theory of Risk Bearing (Chicago, 1971), Chapter 1.Google Scholar

19 Tintner, Gerhard, The Variate Difference Method, Cowles Commission for Research in Economics, Monograph No. 5 (Bloomington, Ind., 1940);Google ScholarTintner, Gerhard, Rao, J.N.K., and Strecker, Heinrich, New Results in the Variate Difference Method (Gottingen, West Germany, 1978).Google Scholar

20 The actual amount of differencing of the original data for the entire period was seldom carried beyond two times and often only first differences were used. For the subperiod estimates the original data usually showed no trend and, in those cases, were used without finite differencing.Google Scholar

21 This random variability coefficient has been used successfully and discussed in detail elsewhere. See McGuire, “A Portfolio Analysis of Crop Diversification,” pp.347–51.Google Scholar

22 The scholarly literature used for this purpose includes Buck, , Granger Movement, and The Agrarian Crusade, A Chronicle of the Farmer in Politics (New Haven, 1920);Google ScholarCarstensen, Farmer Discontent; Goodwyn, Democratic Promise; Hicks, Populist Revolt;Google ScholarHofstadter, Richard, The Age of Reform, From Bryan to F.D.R. (New York, 1955);Google ScholarMcMath, Populist Vanguard;Google ScholarSaloutos, Theodore, Farmer Movements in the South, 1865–1933 (Berkeley, 1960);Google ScholarShannon, Farmer's Frontier;Google ScholarUnger, Irwin, The Greenback Era, A Social and Political History (Princeton, 1964).Google Scholar

23 For examples of the type and availability of qualitative data, see the extensive bibliography in Buck, Granger Movement, pp. 315–51, and the extensive notes in Goodwyn, Democratic Promise, pp. 615–93.Google Scholar

24 While it is true that Goodwyn (Democratic Promise, pp. 82, 201–10, 600–14) is highly critical of most other scholars and that McMath (Populist Vanguard, pp. 77–89, 101–09) disputes their description of the origin of the “agrarian revolt,” it will be documented throughout the remainder of this section that neither Goodwyn nor McMath disagrees with other scholars concerning the location of angry and discontented farmers in the North. Both simply claim that in certain states—for example, Nebraska—discontented farmers did not have “true” Populist origins. However, the cultural heritage of certain farmers is not the subject matter of the present paper.Google Scholar

25 The ranking should be accurate at least as a first approximation. Problems also exist with using only quantitative data such as third-party votes. In particular, a protest voting index might be a poor measure of farmer unrest because differences in state election laws predict rather well differences in third-party votes. I am currently working on an index of agrarian unrest that uses quantitative data to alleviate this and other problems. For a study that used a protest-voting index and recognized that it might be a poor measure of farmer discontent, see Klepper, Economic Bases for Agrarian Protest.Google Scholar

26 Buck, Granger Movement, pp. 211–89; Carstensen, Farmer Discontent, pp. 17–68; Shannon, Farmer's Frontier, pp. 300–26.Google Scholar

27 Unger, Greenback Era, pp. 345–407; Goodwyn, Democratic Promise, pp. 16–24.Google Scholar

28 Unger, Greenback Era, pp. 345–407.Google Scholar

29 Ibid., pp. 255–70; Goodwyn, Democratic Promise, pp. 16–17.

30 Hicks, Populist Revolt, pp. 103–48; Goodwyn, Democratic Promise, pp. 98–102, 154–61, 201–10; McMath, Populist Vanguard, pp. 100–09.Google Scholar

31 Hicks, Populist Revolt, pp. 103–48; Hofstadter, Age of Reform, pp. 94–109; Carstensen, Farmer Discontent, pp. 69–89; Shannon, Farmer's Frontier, pp. 311–26; Goodwyn, Democratic Promise, pp. 154–61, 201–10; McMath, Populist Vanguard, pp. 106–09.Google Scholar

32 Hicks, Populist Revolt, pp. 225–65, 274–300, 328–29; Goodwyn, Democratic Promise, pp. 259–64, 316–23, 394; McMath, Populist Vanguard, p. 128; Shannon, Farmer's Frontier, pp. 317–26.Google Scholar

33 The use of an eight-year sample period also was based on statistical considerations. Much shorter sample periods would have led to meaningless estimates of the random variance.Google Scholar

34 As expected, the geographical pattern found in Appendix Table 1 indicates that prices were more variable in the less settled, newer agricultural regions of the country. Although the highly competitive nature of agriculture during this period might imply a common market price with similar fluctuations regionally, one should expect certain supply and demand conditions to differ among states. Farmers in a new area would have to familiarize themselves with climatic conditions, soil and terrain types, insect problems, and techniques of production best suited to the area. The initial lack of knowledge about such factors would lead to greater random fluctuations on the supply side than for older areas where farmers had that knowledge. On the demand side, some differences in random fluctuations can be accounted for by the existence of more established and steadier markets for crops in the older agricultural regions of the U.S. than in the newer areas. For a detailed discussion of regional differences in supply and demand conditions, see Parker, “Agriculture,” pp. 376–406.Google Scholar

35 Spearman's rank correlation coefficient was used. Because there were tied ranks in the ordering of agrarian unrest, an appropriate statistical adjustment was made. For a detailed discussion of the use of rank correlations with tied ranks, see Yule, G. Udny and Kendall, M.G., An Introduction to the Theory of Statistics (London, 1965), pp. 258–67.Google Scholar

36 Because of the subjective nature of the assignment of ranks to the location of agrarian discontent, correlations between the variability estimates and two orderings of discontent other than the ones presented in Table 1 were computed. The other orderings contained only minor modifications and the correlations did not fundamentally change the results reported here. The results of these sensitivity tests are available from the author on request.Google Scholar

37 The geographic pattern of yield variabilities found in Appendix Table 2 is similar to that of price variabilities found in Appendix Table 1. In addition to those factors already mentioned in footnote 34, unpredictable technological change throughout this period might well account for part of the random fluctuations in crop yields. See Parker, “Agriculture,” pp. 381–91.Google Scholar

38 Although much of the recent research into the economic bases of farmers' discontent has consisted of time-series studies, most of these studies contain implications for the location of farmer discontent. In general, the authors have argued that markets for land, money, and tenancy contracts were highly competitive and nonexploitative in the northern states. Similar conclusions have been made concerning output and other input markets. Although differential freight rates, indebtedness levels, and rates of return to farming among states may account for some of the geographical dispersion of discontent, scholars have argued that those differences were not generally significant enough to explain the entire pattern of discontent. In short, recent econometric research has not been able to discover a consistent economic determinant of the location of agrarian unrest. For discussion of these views, see Bogue, Allan G., Money at Interest: The Farm Mortgage on the Middle Border (Ithaca, NY, 1955) and From Prairie to Corn Belt: Farming on the Illinois and Iowa Prairies in the Nineteenth Century (Chicago, 1963);Google ScholarBowman, John D., “An Economic Analysis of Midwestern Farm Land Values and Farm Land Income, 1860–1900,” Yale Economic Essays, 5 (Fall 1965);Google ScholarBowman, John D. and Keehn, Richard H., “Agricultural Terms of Trade in Four Midwestern States, 1870–1900,” Journal of Economic History 34 (09 1974);Google ScholarCooley, Thomas F. and DeCanio, Stephen, “Rational Expectations in American Agriculture, 1867–1914,” Review of Economics and Statistics, 54 (02 1977);Google ScholarFogel, Robert and Rutner, Jack, “Efficiency Effects of Federal Land Policy, 1850–1900,” in Aydelotte, William et al. , Dimensions in Quantitative Research in Economic History (Princeton, 1972);Google ScholarHiggs, Transformation of the American Economy, pp. 86–101;Google ScholarLee, Susan Previant and Passell, Peter, A New Economic View of American History (New York, 1979), pp. 292–301;Google ScholarNorth, Growth and Welfare, pp. 130–39.Google Scholar

39 Carstensen, Farmer Discontent, p. 12.Google Scholar