For useful comments, he thanks Mary Alston, Daniel K. Benjamin, Ralph Bradburd, Henry Bruton, Joseph Ferrie, Price V. Fishback, Robert Higgs, Randy R. Rucker, Andrew Rutten, Morton Shapiro, Mark Schmitz, Gary Shea, Richard Sylla, Eugene White, and anonymous referees. The results presented in this paper would still be forthcoming were it not for the patient programming assistance of Daniel Klepinger. Sharie Olson assisted in the data collection. The initial research of this paper was done while the author was on leave at the University of Washington. He is grateful for their generous support.

¹ Legally, a foreclosure is the exercised right of a creditor to redeem his property on failure of the mortgagee to discharge his obligations of mortgage debt.Google Scholar

² For a discussion of the causes and consequences of the state farm foreclosure moratoria see Alston, Lee J., “Farm Foreclosure Moratoria Legislation: A Lesson From the Past”, American Economic Review (forthcoming, 06 1984).Google Scholar

³ For discussions of the interplay between ideology and the growth of government see Higgs, Robert, Crisis and Leviathan (Cambridge, forthcoming), andGoogle ScholarNorth, Douglass C., Structure and Change in Economic History (New York, 1981).Google Scholar

⁴ Throughout this paper my use of the term foreclosure rate includes legal foreclosures, loss of farm as a result of bankruptcy, loss of title by default of contract, sales to avoid foreclosures, and surrender of title or other transfers to avoid foreclosure.Google Scholar

⁵ This number was estimated by multiplying the average foreclosure rate for 1921–1940 by the number of farms in 1930 after first deducting the number of cropper farms (in order more accurately to measure ownership units). See Sources in Table 1.Google Scholar

⁶ The same estimation procedure was employed as in Footnote 5.Google Scholar

⁷ Farm foreclosure data for 485 countries for 1935 and earlier years are available in United States Department of Agriculture, Bureau of Agricultural Economics, Transfers of Farm Real Estate (Washington, D.C., 08 1939). This report, a WPA project during 1936 and 1937, is a collection of farm transfers from official country records. The data are not readily comparable across states because of differencs in dates for which the data are reported. Also, there is no indication of the representativeness of these counties in each state. These data are best used in a time series context.Google Scholar

⁸ Genung, A. B., The Agricultural Depression Following World War I and its Political Consequences (Ithaca, 1954), p. 14.Google Scholar

⁹ For accounts of the variation in distress across regions, see the sources in Table I and Genung, Agricultural Depression; Jones, Lawrence A. and Durand, David, Mortgage Lending Experience in Agriculture (Princeton, 1954);Google ScholarMurchison, Claudius, “Depression and the Future of Business,” in Culture in the South, ed. by Couch, W. T. (Chapel Hill, 1934);Google ScholarShideler, James H., Farm Crisis 1919–1923 (Berkeley and Los Angeles, 1957);Google ScholarSaloutos, Theodore and Hicks, John D., Agricultural Discontent in the Middle West 1900–1939 (Madison, 1951); andCrossRefGoogle ScholarWoodruff, Archibald M. JrFarm Mortgage Loans of Life Insurance Companies (New Haven, 1937).Google Scholar

¹⁰ Iowa Agricultural Experiment Station, “The Agricultural Emergency in Iowa,” IX. “Farm Mortgage Foreclosures,” by Murray, William G. and Bently, Ronald C., Circular No. 147 (Ames, Mar. 1933), p. 159. This circular is part of a series of 10 bulletins issued by the Iowa Agricultural Experiment Station on the crisis in agriculture. In addition to state issues, national issues were analyzed by a talented group of economists including Theodore W. Schultz.Google Scholar

¹¹ See Table 1 for the estimation procedure.Google Scholar

¹² The average number of foreclosures was highest in Texas at 5,660 farms per year and lowest in Rhode Island. These polar cases are more a reflection of the number of farms in Texas and Rhode Island rather than indicative of relative distress. The figures in Table 1 on the number of farm foreclosures indicate relative distress in states where the number of farms is approximately equal.Google Scholar

¹³ Genung, Agricultural Depression, pp. 10–11.Google Scholar

¹⁴ The data on prices received by farmers discussed in this paragraph are found in U.S. Department of Agriculture, Agricultural Statistics 1943 (Washington, D.C., 1943), p. 395.Google Scholar

¹⁵ Jones and Durand, Mortgage Lending Experience.Google Scholar

¹⁶ Quoted in Genung, Agricultural Depression, pp. 90–91.Google Scholar

¹⁷ Calculated from data in U.S. Department of Agriculture, Agricultural Statistics 1942 (Washington, D.C., 1942), p. 660.Google Scholar

¹⁸ Calculated from data in U.S. Department of Agriculture, Agricultural Statistics, 1943, p. 403.Google Scholar

¹⁹ Jones and Durand, Mortgage Lending Experience, pp. 110–12 and 126.Google Scholar

²⁰ Johnson, H. Thomas, “Postwar Optimism and the Rural Financial Crisis of the 1920s,” Explorations in Economic History, Vol. II, No. 2 (Winter 1973–1974).Google Scholar

²¹ See Table 2.Google Scholar

²² U.S. Department of Agriculture, Bureau of Agricultural Economics, “Farm-Mortgage Credit Facilities in the United States,” by Horton, Donald C., Larsen, Harald C., and Wall, Norman J., Miscellaneous Publication No. 478 (Washington, D.C., 1942).Google Scholar

²³ For regional and selected state data on size of farm mortgages from 1917 to 1935, see ibid., pp.235–38.

²⁴ See Table 2.Google Scholar

²⁵ The figures in this paragraph were calculated from data cited in Table 2.Google Scholar

²⁶ U.S. Department of Agriculture, Bureau of Agricultural Economics, “Farm-Mortgage Credit Facilities,” p. 4.Google Scholar

²⁷ U.S. Department of Agriculture, “The Farm Real Estate Situation, 1939–40, 1940–41 and 1941–42,” by Regan, M. M. and Johnson, A. R., Circular No. 662 (Washington, D.C., Nov. 1942), pp. 4–5.Google Scholar

²⁸ See Table 2.Google Scholar

²⁹ See Table 2.Google Scholar

³⁰ A model designed to explain a non-pooled cross-section of farm foreclosures should include a variable measuring the promptness in instituting a foreclosure across creditors. This variable is irrelevant in my model because few foreclosures were averted in the depressed period under consideration by delaying foreclosure proceedings. Contrary to some expectations, earnings did not substantially improve until the arrival of World War II. For a discussion and test of the importance of who held the farm debt in explaining cross-state farm foreclosure moratorium legislation, see Alston, Lee J., “Farm Foreclosure Moratorium Legislation”.Google Scholar

³¹ Since only farms that are mortgaged are foreclosed, the number of foreclosures does not increase the number of mortgages. The holder of the mortgage simply changes when the creditor sells the farm after it is foreclosed. Therefore no problems of simultaneity arise.Google Scholar

³² By relative earnings, I mean actual farm income in each state relative to expected farm income where expected income is measured by earnings in the recent past. Ideally the appropriate measure of the ability to service debt is net operating income. Data limitations preclude us from using this measure, yet I do not believe the omission is serious since all farmers' costs were subject to the same temporal influence of relatively rising non-farm prices. The bias, in my estimate, depends on the variance across farms in the percentage of their inputs purchased in the market. I am grateful to an anonymous referee for bringing this issue to my attention.Google Scholar

³³ Johnson, “Postwar Optimism”.Google Scholar

³⁴ Ibid., pp. 184–85.

³⁵ The estimates do not differ substantively when the log-log functional form is specified. The coefficient of Land Value becomes significant at the 95 percent confidence level. In three states— Arizona, New Mexico, and Wyoming—improved acreage increased by over 1,000 percent. To test whether the estimates in Table 3 are sensitive to these outliers, I estimated the parameters of equation (1) with the above states deleted from the sample. R² goes up to .69. The coefficient on Land Value becomes more significant and t-ratio on Improved Acreage falls to 1.95. None of the other estimates is affected significantly.Google Scholar

³⁶ When the log-log functional form is estimated, the coefficient of Debt/Value becomes significant while the t-ratio of Percent Farms Mortgaged falls to t = 1.44.Google Scholar

³⁷ When either Percent Farms Mortgaged or Debt/Value is dropped from the regression equation the other variable is highly significant with the correct sign.Google Scholar

³⁸ The estimates of the coefficients on Average Earnings/Base Earnings and Urbanization are not substantially sensitive to functional form.Google Scholar

³⁹ As discussed earlier, urbanization most likely captures other influences as well as the opportunity for off-farm earnings. Considering the results presented here, further research is needed to determine more precisely the stabilizing effect of urban areas.Google Scholar

⁴⁰ Johnson, “Postwar Optimism,” p. 184.Google Scholar

⁴¹ This test and the following tests cannot be strictly interpreted as testing Johnson's hypothesis because my data set only begins in 1926, whereas Johnson focuses his discussion on the entire decade of the 1920s. Nevertheless, the results could be interpreted as loosely consistent or inconsistent with his hypothesis, depending on the point one might be arguing.Google Scholar

⁴² The F-statistic, when the linear functional form is specified, is 1.87. When the log-log functional form is specified, the F-statistic falls to 1.45.Google Scholar