¹ Annals of Congress, Fifth Session (Washington, D.C., 1881), 06 13, 1798, pp. 1917–1925.Google Scholar

² A Letter from George Nicholas of Kentucky to His Friend in Virginia (Lexington, Kentucky, 1798). Evans' American Bibliography Number 35973.

³ The Papers of Henry Clay, eds. Hopkins, James F. and Hargreaves, Mary W. (Lexington, Kentucky, 1959), Vol. 1, pp. 10–11, quoting a letter written in 02 1799.Google Scholar

⁴ There are numerous historical studies pertaining to the inequality in wealth and income. Many of these point to increased inequality in wealth and income in the seventeenth, eighteenth, and nineteenth centuries in various communities, states, and the whole of the United States, and are contrary to the findings of my Kentucky study. These include: Kearl, J. R.. Pope, Clayne L.. and Wimmer, Larry T., “Household Wealth in a Settlement Economy: Utah, 1850–1870,” this Journal, 40 (09 1980), 477–96;Google ScholarLemon, James and Nash, Gary, “The Distribution of Wealth in Eighteenth-Century America: A Century of Change in Chester County. Pennsylvania,” Journal of Social History, 2 (1968);CrossRefGoogle ScholarLockridge, Kenneth, “Land, Population and the Evolution of New England Society, 1630–1790; and an Afterthought,” in Colonial America. Essays in Politics and Social Development, ed. Katz, Stanley N. (Boston, 1971). Other specific studies includeGoogle Scholar: Curti, Merle, The Making of an American Community (Stanford, 1959);Google Scholar more authoritative is Jones, Alice Hanson, Wealth of a Nation to Be: The American Colonies on the Eve of the Revolution (New York. 1980);Google Scholar strong proponents of increased inequality are Williamson, Jeffrey G. and Lindert, Peter H., as summarized in American Inequality, A Macroeconomic History (New York, 1980). My views of these two studies are found in reviews in, respectivelyGoogle Scholar, Economic Development and Cultural Change (forthcoming), and Journal of Economic Literature, 20 (1982), 92–94; see also my articleGoogle Scholar, “Distribution of Income and Wealth,” in Porter, Glenn, ed., Encyclopedia of American Economic History, Vol. 3 (New York, 1980) pp. 1087–1119. The study of inequality in various areas obviously is a continuing process. Two examples areGoogle Scholar: Gallman, Robert, “Influences on the Distribution of Landholdings in Early Colonial North Carolina,” this Journal, 42 (09 1982), 549–75;Google Scholar and Soltow, Lee and May, Dean, “The Distribution of Mormon Wealth and Income in 1857,” Explorations in Economic History, 16 (04 1979), 151–62.CrossRefGoogle Scholar

⁵ In almost all other states, properties are listed by situs county or township only, and one must examine all pages of every county tax list to glean the number of properties owned by a single individual. This intractible problem is discussed in Soltow, Lee, “Wealth Distribution in England and Wales in 1798,” and in “English Land Tax Redemption Records, 1798–1963,” Economic History Review, 2nd ser., 34 (1981) and 35 (1982), respectively.CrossRefGoogle Scholar

⁶ The amount of household information divulged in the tax lists is very limited. The name of the taxpayer is given for each entry and this tells us that there very few women in the lists. We really do not know the taxpayer's age since there is only one figure—that stating the number of white males 21 and older—associated with the taxpayer, as well as the number of white males 16–20, and the numbers of slaves in various age categories. Entries reporting more than one adult white male indicate that farm laborers, sons, or other relatives are present, but the number of each is unknown. Some bounds are suggested from the following distribution of the 32,826 adult white males: Number of white males 21 and older reported tax book entry Total number of entries We will consider only 1,008 persons in the 21 and older category who might have inherited wealth.

⁷ I first arranged counties alphabetically and numbered all pages consecutively from first to last page. I then recorded all entries on every tenth page. The taxpayers within each county book are arranged alphabetically by the first letter of the last name only, and it follows that I have a random selection from all ethnic groups to the extent that they are revealed by this first letter. I next eliminated the one or two individuals in my sample with 10,000 or more acres and substituted in those places a complete enumeration of large holders. This supplementary list of large holders was compiled by recording all properties belonging to each of the 107 taxpayers with 10,000 acres or more. The final sample to be reported is then a disproportionate sample where an entry has a weight of 10 or 1. My sample for 1820 followed the same procedure except that I sampled only every hundredth page, and used weights of 100 and 1.

The tax lists exist in incomplete form for a few years prior to 1800, but are complete and in excellent form from 1800 until after the Civil War. I chose the 1800 records for special study because of their completeness and because it was a federal census year.

⁸ There are 80 different “aggregates of the summary tables” for the 41 counties in my Kentucky sample; I have found errors in addition in some of these. It is possible that the federal census count of June 1800 is consistent with the tax list count since immigration was so rampant.

⁹ The Federal and State Constitutions, Colonial Charters, and Other Organic Laws of the United States, Part I, comp. Ben Perley Poore (Washington, D.C., 1877), pp. 657–58.Google Scholar

Also see Coward, Joan Wells, Kentucky in the New Republic (Lexington, Kentucky, 1979), pp. 27–28.Google Scholar

¹⁰ Adams felt that, in a nation with 80 to 90 percent of its population propertyless, “if all were to be decided by a vote of the majority, … [the propertyless] would not think of usurping over the rights of [those] who have? Property is surely a right of mankind as really as liberty.” The Works of John Adams, Second President of the United States, ed. Adams, Charles Francis (Boston, 1856), pp. 9–10.Google Scholar

¹¹ The PHP determined from the sample in 1800 was not statistically different from a hypothesis that the population PHP was 0.5. The sample size, including the 93 large landholders, was 510.

¹² Soltow, Lee, Men and Wealth in the United States 1850–1870 (New Haven, 1975), pp. 41, 184. The above sample is of size 455 free males 21 and older, including 26 with estates of more than $100,000.Google Scholar

¹³ Toulmin, Harry, Secretary of the Commonwealth of Kentucky, A Collection of All the Public and Permanent Acts of the General Assembly of Kentucky, Part V, Law passed 12 21, 1799, p. 92.Google Scholar

¹⁴ The federal census for Kentucky in 1800 is quite explicit in showing that three percent of persons lived in towns. The remaining numbers who were landless undoubtedly included artisans, perhaps many, and those owning horses and cattle who had access to some acreage. Detailed studies of occupational groups in western Pennsylvania at the time are found in Soltow, Lee and Keller, Kenneth, “Tenancy and Asset-Holding in Late Eighteenth Century Washington County, Pennsylvania,” Western Pennsylvania HLctorical Magazine, 65 (1982), pp. 1–16;Google Scholar and Soltow, Lee and Keller, Kenneth, “Rural Pennsylvania in 1800: A Portrait from the Septennial Census,” Pennsylvania History, 49 (1982), pp. 25–47.Google Scholar

¹⁵ Property distributions compiled from county lists of the form f(X) are often assumed to represent wealth distribution h(NX) when they do not. The multiple properties of an individual may at best be collated for the given township where the properties lie, or, at best, his properties in a given county. Neglected are his properties in other counties, states, or countries. The main advantage of wealth distributions compiled from estate data is that they can include all properties of the deceased, wherever owned.

¹⁶ There may have been substantial tenancy in both Kentucky and Ohio, but I have been unable to find any substantive data to measure its degree. See Soltow and Keller, “Tenancy.”

¹⁷ The influence of slavery on various levels of inequality is a complex issue and difficult to unravel. It is perhaps sufficient to point out that among landholders, those with slaves had greater inequality and that those without slaves had the same distribution of land as did persons in Ohio. These facts are shown in the appendix table.

¹⁸ Jilison, Willard Rouse, Old Kentucky Entries and Deeds (Baltimore, 1969).Google Scholar

¹⁹ Mortality and Miscellaneous Statistics, 1860, table 3, pp. 296–319; Soltow, Men and Wealth, pp. 156–57.Google Scholar

The price index is from David, Paul and Solar, Peter, “A Bicentenary Contribution to the History of the Cost of Living in America,” in Research in Economic History, ed. Uselding, Paul, vol. 2 (1977), p. 16.Google Scholar Kentucky time series data from 1830 to 1863 are given in Kentucky Auditor's Report, Auditor's Office (Frankfort, Kentucky, 10 10 1831) fold-out table; see the same Report for later dates. The aggregate valuation, in millions of dollars (unadjusted for price index changes) is:Google Scholar Per capita real wealth (W) is obtained from these values, from 1831–1860, by using David's index (1800 = 100) and the population of males 20 and older (interpolated between census years). The trend equation implying 2.5 percent growth is: log w = 2.9320 + .010540(Year – 1800), N = 30, R² = .74 (.0553) (.00120) (Standard errors in parentheses). This equation is plotted as the lowest trend line on Figure I with an extension to 1800. The 1840 value of $272 million in the above series is the most significant. It translates into an average wealth (W) of $3,088, which is the peak point of the series, as presented in Figure I. This value is compared with W = $1,200 in 1800 in the form 3088 = 1200(l.0239)⁴⁰, shown in Figure I with a rate of 2.4 percent.

²⁰ The Statutes at Large of South Carolina (Columbia, 1874), pp. 250–395; the stated value was $150 from 1794–1798 and $200 in 1800;Google ScholarBlodgett, Samuel, Economica: A Statistical Manual for the United States (Washington, D.C., 1806), p. 196.Google Scholar

²¹ Films 2764, part I, and 35620, part 4, in LDS Genealogical Library, Salt Lake City. The price per acre for 21 recorded deeds in Mason County from 1796 to 1798 yields an unweighted average of $2.92 an acre. It is very time consuming to ferret out market prices from deed records. Prices in Jefferson County (Louisville) were closer to an average of $3.33.

²² The 2.4 percent rate exceeds slightly the 2.2 percent rate found by Raymond Goldsmith for reproducible wealth per capita in the United States in the period from 1805–1850, and the 2.0 percent rate from 1805–1950. It exceeds the per capita rate of factor productivity in the United States from 1800 to 1840 of .60 found by Robert Gallman, and Paul David's finding of less than 1.5 percent a year. Cf. Goldsmith, Raymond W., “The Growth of Reproducible Wealth of the United States of America from 1805 to 1950,” in Income and Wealth of the United States: Trends and Structure, International Association for Research in Income and Wealth, Series II (Cambridge, England, 1952), p. 269;Google ScholarGailman, Robert, “Changes in Total U.S. Agricultural Factor Productivity in the Nineteenth Century,” Agricultural History, 46 (1972), p. 208;Google ScholarGaliman, Robert, writing in American Growth Before 1840, eds. Davis, Lance, Easterlin, Richard, and Parker, William (New York, 1972), pp. 22, 24;Google ScholarDavid, Paul, “The Growth of Real Product in the United States Before 1840: New Evidence, Controlled Conjectures,” this JOURNAL, 27 (03 1967), pp. 154–5.Google Scholar

²³ Soltow, Lee, “Male Inheritance Expectations in the United States in 1870,” Review of Economics and Statistics, 64 (1982), pp. 252–60.CrossRefGoogle Scholar

²⁴ Slavery does not seem to account for any changes in relative inequality from 1800 to 1860. The proportion of slaveholders among all free adult males of .18–22 prevailed in the state in 1800, 1830, 1850, and 1860, and the Gini coefficient of the distribution of slaves among slaveholders was between .87 and .91 in each of these years. Finally, the distribution of real estate value among adult males 20 and older in Kentucky, a measure excluding personal estate and the value of slaves, was .88 in 1800, .90 in 1850, .86 in 1860, and .87 among whites in 1870. The proportion of men with real estate was .44 in 1800, .45 in 1850, .46 in 1860, and .49 for whites in 1870, as derived from my samples described earlier. I have made many computer runs of distributions in 1800 for slaveholders and nonslaveholders. They tend to show greater average values and greater dispersion for slaveholders than for nonslaveholders. An example can be given for the acreage of owners. The danger with this type of classification is that one might attribute Kentucky's inequality to slaveholding. Yet the rich might have held almost as many properties and almost as much acreage (including considerable amounts of unoccupied land) had they had no slaves.