We would like to acknowledge the helpful comments of Stanley Engerman, Robert Fogel and Morgan Reynolds on an earlier draft of this paper. We would also like to thank Arden Hall and Lynnae Wolin for their assistance. The National Science Foundation and the Computer Center at the University of California, Berkeley provided financial support.

¹ The data are based on a sample of farms discussed in the Appendix. The characterization of a farm as either a “Small Family Farm” or a “Plantation” is intentionally designed to reflect two stereotypes which emerge both from historians’ treatment and the 1880 Census of Agriculture. The, specific definitions are included in the notes to Table 1.

² The economic analysis sketched below draws upon “market signalling” models proposed by George Akerlof, Kenneth Arrow and Michael Spence. See Akerlof, George A.. “The Market for ‘Lemons’: Quality, Uncertainty and the Market Mechanism,” Quarterly Journal of Economics, LXXIV (August 1970)Google Scholar; Arrow, Kenneth J., “Some Models of Racial Discrimination in the Labor Market,” Rand Memorandum, RM-6253-RC (Santa Monica: Rand Corporation, 1971)Google Scholar; Spence, Michael “Market Signalling,” Discussion Paper, Public Policy Program, Number 4, (Cambridge: Kennedy School of Government, Harvard University, February 1972).Google Scholar

³ Reid, Whitelaw, After the War A Tour of the Southern States, 1865–1866 (New York: Harper and Row, 1965, originally published in 1866), pp; 564–65Google Scholar: Reid was referring to “portions of the Mississippi Valley”; however, there is ample evidence in the many commentaries on the period that similar views were held throughout the South.

⁴ The undeveloped state of the southern credit market is illustrated by the fact that, as late as 1890, only 5.3 percent of all farms in the South were mortgaged, compared to 36.4 percent of all farms outside of the South. These figures are computed from United States Census Office, Eleventh Census, Report on Real Estate Mortgages in the United States at the Eleventh Census 1890 (Washington: G.P.O., 1895), pp. 175–77.Google Scholar

⁵ Literacy data is available from United States Census Office, Ninth Census,The Statistics of Population of the United States … Compiled from the Original Returns of the Ninth Census (June 1, 1870), (Washington: G.P.O., 1872).Google Scholar

⁶ The figures are computed from United States Census Office, Twelfth Census, Census Reports: Twelfth Census of the United States Taken in 1900; Volume V: “Agriculture,” Part I (Washington, G.P.O., 1902).Google Scholar

⁷ The landlord could avoid this risk by insisting on prepayment of rent, but few Negroes would have sufficient capital to meet such a demand. As we noted above, they would be unlikely to obtain credit from a third party to finance a prepayment of rent.

⁸ This is not to deny that other costless signals of productivity, such as sex and age, were also used to classify workers. The fact remains that the employer was not likely to discriminate between workers within these broad categories; This behavior by a landlord is not unusual—indeed it is commonplace among employers today, Production workers on a modern assembly line typically receive a standardized wage.

⁹ Management control of labor inputs is universally observed to be necessary in sharecropping arrangements. Without such control, the laborer would seek alternative employment for a portion of his time, since he receives only a fraction of his marginal product from the sharecropped farm. Our analysis suggests an additional incentive for control when the tenant is black. The landlord would believe that control would substantially increase output, and therefore would be prepared to offer significantly better terms under such an arrangement. The laborer will accept the offer so long as the larger share of output he receives more than compensates for the controls involved.

¹⁰ It was common practice in the South for landlords to endorse the notes of their tenants. See Sisk, Glenn N., “Rural Merchandising in the Alabama Black Belt, 1875–1917,” Journal of Farm Economics, XXVII (November 1955), p. 706.Google Scholar

¹¹ We have presented a detailed analysis of the problems of the South in developing a credit structure to meet the demands of agricultural credit in another article, “Debt Peonage in the Cotton South After the Civil War,” Journal of Economic History, XXXII (September 1972).Google Scholar

¹² In addition to their relative shortage of untilled land, blade farmers reported fewer animals which could provide manure. We estimate that small family farms operated by whites had on average 1.9 working animals compared to only 1.3 on the black farms. Similarly, whites had more swing than blacks (9.9 compared with 5.2 respectively).

¹³ Truman, Benjamin C., “Report... Relative to the Conditions of the Southern People and the States in which the Rebellion Existed,” contained in Message of the President of the United States (May 7, 1866), United States Senate, Thirty-Ninth Congress, First Session, Executive Document, Number 43 (Washington: G.P.O., 1866), p. 10.Google Scholar

¹⁴ Davis, Lance E. and North, Douglass C., Institutional Change and American Economic Growth, (New York: Cambridge University Press, 1971).CrossRefGoogle Scholar