¹ SeeRansom, Roger L. and Sutch, Richard, One Kind of Freedom: The Economic Consequences of Emancipation (New York, 1977), chap. 1.Google Scholar

² The market wage rate, used to measure the value of leisure, reflects the value to individuals on the margin, so only the last increment of leisure taken is valued correctly. Given that blacks reduced their work effort significantly from the levels extracted by slaveowners, the wage rate underestimates the value of leisure to blacks. This measure also excludes the utility derived from nonpecuniary sources such as the ability to choose a consumption bundle, free speech, the freedom to choose marriage partners, the right to vote, and the pure joy of freedom. Measuring these items is difficult because there are no quantitative measures of the degree to which they were achieved by blacks. Valuing them is nearly impossible because they were not traded in markets. For both these reasons the measured value of freedom is a lower-bound estimate.Google Scholar

³ The sample of farms was taken from the manuscript schedules of the Census of Agriculture for 1880 and is described in Ransom and Sutch, One Kind of Freedom.Google Scholar

⁴ Robert Higgs estimates that 13 percent of blacks lived in urban areas in 1870. See Higgs, Robert, Competition and Coercion: Blacks in the American Economy, 1865–1914 (Cambridge, 1977), pp. 32–35. Black wage workers also made up a significant fraction of the agricultural labor force.CrossRefGoogle Scholar

⁵ See Ransom and Sutch, One Kind of Freedom, appendix G, for a description of the sample. Their procedure for estimating income is described in appendix A, where they report estimates of black income based on black tenants operating small farms. On pp. 6–7 they report only the income of black sharecroppers. They estimate family income earned by black tenants operating farms with 50 acres or less in crops and 26 weeks or less of hired labor in the subset of their sample that they refer to as the “Cotton South”.Google Scholar

⁶ See Virts, Nancy, “Estimating the Importance of the Plantation System to Southern Agriculture in 1880”, this JOURNAL, 47 (12 1987), pp. 984–88.Google Scholar

⁷ See Alston, Lee J. and Higgs, Robert, “Contractual Mix in Southern Agriculture since the Civil War: Facts, Hypotheses, and Tests”, this JOURNAL, 42 (06 1982), pp. 328–30;Google Scholar and Ransom and Sutch, One Kind of Freedom, p. 89.Google Scholar

⁸ The value added from pork production and the value of garden produce are estimated by the same method used by Ransom and Sutch, One Kind of Freedom, pp. 214–16.Google Scholar

⁹ The assumption that all share farms were sharecropped leads to an overestimate of rent on those farms where the operator supplied his own working stock and farming implements. These share tenants usually paid a rent of one-fourth of the cotton and one-third of the grain grown on the farm. The census made no distinction between the two forms of tenure in 1880, so the number of share tenants and sharecroppers is not known. Although it is thought that a majority of these tenants were sharecroppers, all contemporary descriptions of southern agriculture mention share tenants as a significant class of tenants. This bias in the rental estimates is offset in part because neither the value of corn fed to the animals nor the depreciation of farm implements is deducted from income as it should have been for those farms where the operator owned his own working stock and implements.Google Scholar

¹⁰ The census reported the number of 400-pound bales of cotton and bushels of corn grown on each farm. We assume a price of 9.5 cents a pound for cotton and 62.3 cents a bushel for corn. See Ransom and Sutch, One Kind of Freedom, p. 167.Google Scholar

¹¹ We assume a 15 percent depreciation rate and feed requirements of 30 bushels of corn per mule and 35 bushels of corn per ox or horse. See Ransom and Sutch, One Kind of Freedom, pp. 208, 248.Google Scholar

¹² We eliminate farms with inadequate wage data and those farms without information on the value of output.Google Scholar

¹³ Weighted averages were constructed by first estimating average county income. County income was computed by taking a weighted average of sharecropper and sharerenter income, where the weights were the proportion of croppers and renters in the Ransom and Sutch sample for each county. State income figures were computed as a weighted average of the county income estimates, where the weights were each county's share of the total population of all counties sampled in each state. Finally, average income was computed by taking the weighted average of the state figures, where the weights were the share of each state in total southern population.Google Scholar

¹⁴ With the inclusion of those counties with samples of fewer than 10 farms the average family income Was $358.Google Scholar

¹⁵ See Ransom and Sutch, One Kind of Freedom, appendix A, for their estimates.Google Scholar

¹⁶ The size of this decrease has generated some controversy. See Goldin, Claudia, “N Kinds of Freedom”, Explorations in Economic History, 16 (01 1979), p. 11;CrossRefGoogle Scholarand Wright, Gavin, “Freedom and Southern Economy”, Explorations in Economic History, 16 (01 1979), p. 95.CrossRefGoogle ScholarHowever, the average number of hours worked per capita for 1860 is significantly lower than the average number of hours worked on the slave plantations examined by Olson, John, cited in Robert Fogel and Stanley Engerman, “Explaining the Relative Efficiency of Agriculture in the Antebellum South”, American Economic Review, 67 (06 1977), p. 287. Most of the decrease in hours worked per capita in 1880 is due to the lower participation of women and children in the labor force. The labor force participation rates reported by Ransom and Sutch are based on the number of blacks reporting agricultural occupations to the census enumerators. While there is evidence that enumerators were less likely to report white females and children as employed on family farms, this problem was much less common for blacks. The decrease in participation rates that Ransom and Sutch found is consistent with anecdotal accounts from the period. There is little hard evidence to suggest that this estimate of labor hours in 1880 is biased downward.Google Scholar

¹⁷ The hours worked by women and children have been converted to male equivalent hours by multiplying by their relative productivity. To compute an average amount of labor per capita, the estimated labor hours for each type of worker was multiplied by the share of the black population in that age group. Although we use the same methodology as Ransom and Sutch, we get different results for two reasons. When calculating the average labor hours in 1880 we use the population weights for 1880, not 1860 as Ransom and Sutch did. The population weights we use for 1860 are also slightly different than those used by Ransom and Sutch, but the reason for the difference remains unclear.Google Scholar

¹⁸ See Wright, Gavin, Old South, New South: Revolutions in the Southern Economy since the Civil War (New York, 1986), chaps. 3 and 4. Wright suggests that the demand for seasonal labor was limited to harvest time, when tenants were so occupied with their own crops that they could not hire out.Google Scholar

¹⁹ References to the practice of tenants or their families hiring themselves out as day laborers include Hilgard, Eugene, Report on Cotton Production in the United States (Washington, DC, 1884), pp. 517, 521;Google ScholarBrannen, C. O., “Relation of Land Tenure to Plantation Organization”, USDA Bulletin No. 1269 (October 1924), pp. 22;Google Scholarand Langsford, E. L. and Thibodeaux, B. H., “Plantation Organization and Operation in the Yazoo Mississippi Delta Areas”, USDA Technical Bulletin No. 682 (May 1939), p. 23. Brannen's study of plantations in 1920 found that hiring extra wage labor from plantation tenants and their families was more common than hiring extra labor from other towns. Since the costs of transporting labor were higher in 1880 than in 1920, it seems likely that it was even more common to hire extra labor from tenants or their families in 1880. Langsford and Thibodeaux reported that in the Mississippi Delta area it was common practice to limit the size of tenant farms to allow the tenant and his family time to work on the landowner's farm for wages.Google Scholar

²⁰ Reid, Joseph, “Sharecropping as an Understandable Market Response: The Post-helium South”, this JOURNAL, 33 (03 1973), pp. 106–30, for a summary of the variation in terms of contracts and side payments.Google Scholar

²¹ For examples of contracts which make provision for extra work, see Reid, “Sharecropping”, pp. 116–19, 128–30.Google ScholarSee also Higgs, Competition and Coercion, p. 49.Google Scholar

²² We assumed workers received the harvest wage 30 percent of the time. The average wage for each state was weighted by its share of the South's black population to compute the southern average. Wage rates are from Report of the Commissioner of Agriculture 1882 (Washington, DC, 1883), p. 644. Population shares are computed from Compendium of the Tenth Census, table 23.Google Scholar

²³ See Ransom and Sutch, One Kind of Freedom, p. 233.Google Scholar

²⁴ This suggestion is due to an anonymous referee.Google Scholar

²⁵ It has been suggested that the move to sharecropping from wage labor was motivated in part by the planters' desire to have a dependable source of harvest labor. Since sharecroppers received their payment after the harvest, they would be less likely to leave. Given the fact, however, that sharecroppers often received advances and wage workers often were not paid all their earnings until the end of the year, the actual timing of payments was not that different between the two forms of payment. If sharecroppers were less likely to leave before the end of the contract, it must have been because they had more to lose.Google Scholar

²⁶ Ransom and Sutch, One Kind of Freedom, p. 259.Google Scholar

²⁷ As this discussion makes clear, our measures of the value of leisure have been based on the assumption that the only source of gain was from working fewer hours. It is also possible that slaves worked both more hours and more intensely than free blacks.Google Scholar

²⁸ Haskell, E. S., “A Farm Management Survey in Brooks County, Georgia”, USDA Bulletin No. 648 (May 1, 1918), p. 16.Google Scholar

²⁹ Reid, “Sharecropping”, p. 116.Google Scholar

³⁰ It is also possible that the difference between the estimated wage of tenants and the observed wage of agricultural laborers is that the wage rates reported to the commissioner of agriculture were unreasonably high. Although this may have been the case, there is no known reason why systematically throughout the South wages paid should have been overreported. The data from later years are comparable to those reported in 1879.Google Scholar

³¹ The present value is calculated using an interest rate of 5 percent. The number of individuals of given age and sex in an average family is from Historical Statistics of the United States, series B 84–91 and series A 71–85.Google Scholar

³² The payment is computed using black income in 1860 from Table 1.Google Scholar

³³ The higher estimate was computed by multiplying the 1984 average family income of $26,433 times the present value of emancipation given in the text. The lower estimate was computed using the 1984 average black family income of $15,432. Family income data are from Historical Statistics of the United States, series G 147 and 149.Google Scholar

³⁴ See Davis, Lance E. et al. , American Economic Growth: An Economist's History of the United States (New York, 1972), p. 40.Google Scholar