Introduction/background statement
The elementary economic logic of slavery in a setting of land abundance has been outlined many times. If land is available to all comers, and if cultivation may be practiced at any scale without major loss of efficiency, then there will be no way for an entrepreneur to achieve a large absolute profit except with un-free labor. Most of the seventeenth-century development of tobacco culture in Virginia and Maryland was based on white indentured labor, in a system with many of the economic features of slavery men and women could be bought and sold at any time, families broken up, harsh punishments inflicted for running away or misbehavior, and some planters were able to expand their operations well beyond family scale.
Indentured servitude was inherently transitional, however: as terms of service were completed, as free-labor wages rose and the cost of transport fell, only the full-fledged system of involuntary slavery-for-life survived. If the profitability of slavery is accepted, explanations based specifically on the notion that slave owners were losing money can be ruled out. But this agreement does not take us far, and the precise meaning of the statement “slavery was profitable” makes a great deal of difference.
Land-labor conditions were initially similar in all of the colonies, as were attitudes toward profit-making and slavery. What is missing from this story is the fact that in the seventeenth and eighteenth centuries, North American colonies had to buy African slaves on a world market at prices which reflected the high profitability of slavery in the sugar colonies of the West Indies. For this reason, despite the underlying similarity in factor proportions, slavery expanded only in areas where profitable export staples were available.
Though cotton was of no commercial importance before the 1780s, the growth of North American slavery revolved primarily around the fortunes of tobacco and rice, with indigo playing a briefer and lesser part in the mid-eighteenth century. In the North, according to a leading student of antislavery thought, the original and essential grievance of the colonists who cared about the matter was that they could not buy enough slaves at a reasonable price.
Objective
The objective of this paper is to analyze historical interpretations based on concept of profitability of slavery. To analyze some of the important assumptions of conventional economic analysis, basic to most of the existing research in econometric history. To analyze that the principle of opportunity costs implies that there is no essential difference between the production and consumption of nonmarket goods and services on the one hand, and the production, sale and purchase of commodities through the market, except for the costs of transportation and other costs directly associated with transactions.
Survey of Literature
A major flourishing of historical economic analysis of slavery dates from the 1958 Conrad and Meyer (1961) study on the profitability of slavery. The distinguishing feature of this work has been the application of basic economic principles of competitive market economies (principles commonly referred to as “price theory”) to the analysis of slavery. Generally, economists have found the slave economy readily amenable to their approach; its main features are well explained by price theory, and slavery seems to have functioned flexibly and efficiently in response to economic incentives, even generating a rapid rate of economic progress until the Civil War.
The path-breaking essay by Conrad and Meyer was more general in scope, based on an economist’s as opposed to an accountant’s approach to the problem. Rather than confining their attention to measuring the rates of return of particular plantations, they asked whether, on the average, a planter who purchased slaves and land at an ‘average’ price for the period ( 1830-60) could have expected to make as high a rate of return as if he had invested in some alternative asset. While not arguing that every plantation made money, their important conclusion was that, on the average, profits were to be made from slave ownership.
In their analysis, Conrad and Meyer (1961) separated the slave economy into two sectors, and estimated the profitability of slave ownership in each. Male slaves were regarded as capital goods used in the production of marketable output of agricultural staples. Female slaves, however, not only were used to produce staples but also were the source of additional slaves. Thus the female slave could be considered a capital good who produced the capital goods used to produce final output.
By regarding both male and female slaves as capital goods, Conrad and Meyer were able to test for the profitability of slavery by computing the rate of return on the total investment in slaves, including the land and other assets which the slave used. The basic computation involved solving to find that rate of return which equated the cost of obtaining slaves with the net stream of earnings derived from using the slaves. Separate rates of return were computed for male and female slaves.
There is some tendency for historians to take a “bottom line” approach to the economic literature, concluding, for example, that because almost all the economists agree that slavery was profitable, there is little reason for the layman to try to master the fine points of the debate. This kind of reading is not safe, because the “profitability” which Conrad and Meyer support is of a particular kind, having to do with the appropriateness of slave prices in relationship to returns.
This is very different from the kind of profitability analyzed by Yasuba, which has to do with the more fundamental territory, which was separated from Indiana over the slavery issue, had a proslavery majority as late as 1818. Only the threat of congressional rejection deterred Illinois from attempting to enter the Union as a slave state. Thus, nineteenth-century history down to 1860 offers a case study of comparative regional development under alternative institutional systems, in which the legal regime was the effective obstacle to slavery in the North.
Not every economist is committed to every one of these assumptions, and no one has proclaimed that these are eternal truths for all of history; but these are the working presuppositions that economists typically bring to the subject. Anyone can see that these are simplifying operational assumptions, and any analytical history of manageable proportions is bound to involve a rather drastically simplified characterization of behavior and the workings of markets.
Following Conrad and Meyer, a number of studies have attempted to estimate the rate of return on investment in slave property by comparing the market price of slaves to a hypothetical stream of returns over a hypothetical future. As Yasuba and Sutch argued, this is no test of the viability of slavery, but only a measure of the appropriateness of the capitalization of the income streamâa test that can never be conclusive, once we recall that the income stream is a hypothetical expected one.
Yasuba’s analysis is depicted in Figure 5.1: in each year the price of slaves is determined by the interaction of a demand curve with an inelastic supply curve, because, after the closing of the African slave trade, the aggregate slave labor supply could not be increased in response to higher prices, except over time. The observed slave price was in fact well
The point is not just that the real proof of profitability is the high slave prices themselves, but that the rising profitability is embodied in the higher prices. In the abstract, there is little point in sharply differentiating between the slaveholders’ interest in annual earnings on his crops and in the value of his slave property, because slave prices will reflect the expected stream of future earnings from the use of slave labor. For similar reasons, economists frequently use
Ramsdell held that economic factors would have made slavery unprofitable to the planting class and that slavery would have ended in the late nineteenth century without the Civil War. In his view, the decline of slavery was likely because the planters had reached the end of the land upon which cotton could be grown profitably. Ramsdell suggested that by 1860 the western limits of slavery had been reached in Texas, and no room existed for expansion northward.
Geographic containment would have caused a rise in the labor/land ratio and, as a consequence, slave prices would have fallen until it became too expensive for the owners to maintain their slaves. The end result would have been manumission.
The Ramsdell (1929) positionâfrequently called the natural limits hypothesisâis theoretically plausible, but Ramsdell argued as if there were little room for downward adjustment of slave prices before freedom of the slaves would become an attractive alternative to owners. He also exaggerated the potential pressure on slave prices by ruling out the possibilities that soil in the older areas might be refurbished or that slaves might be profitably employed in other agricultural or non-agricultural pursuits. Nevertheless the natural limits hypothesis has attracted a large number of adherents.
Genovese (1962) posited that the skewed income distribution which slavery created made for low demand for domestically manufactured goods within the South; wealthy planters preferred to import goods, slaves had no purchasing power, and low income whites had little market impact. This situation contrasted with that of the North and Midwest where a large ‘middle-class’ market was held to have led to the internal development of industry and a more diversified economy. Genovese (1965) argued that this restriction of the internal market deprived the South of economies of scale upon which to create an efficient industrial base.
The result was a rural, agricultural southern economy dependent upon outsiders for modern industrial goods. An additional effect of the skewed income distribution was that it caused, or at least permitted, conspicuous consumption and lavish living on the part of the rich planters. Consequently savings and capital formation were reduced.
Theoretical background of the topic
The whole discussion of agricultural progress would be much less important if the region had developed other strong economic sectors into which resources could have shifted during periods of slack cotton demand.
That the South was behind the rest of the nation in nonagricultural economic activity has never been in dispute. Several economists have interpreted this evidence to mean only that the South’s lag behind the North in industrialization is fully consistent with the proposition that during the antebellum era the South’s comparative advantage was in agriculture rather than in manufacturing. This perspective views each region as responding to existing factor endowments and specializing according to these differences” in the manner classically described by international trade theory. This simple argument has been powerful and difficult to refute.
Critics have generally been forced to argue in somewhat vague terms that slave labor was not well-suited to nonâ agricultural work, that patterns of demand generated by slavery did not mesh well with antebellum manufacturing technology, or that the distinct politics, ideology, and pattern of social behavior” of plantation slavery was hostile to industrial development arguments that are at best difficult to define precisely and verify, at worst easily refuted.
The secret of getting the right answer lies in posing the question properly. The fallacy in the analysis just presented is not in the claim that Southern resources were allocated according to the principle of comparative advantage, but in the implicit assumption that the North developed a thriving manufacturing sector because the North had a comparative advantage in manufacturing. The problem is once again the tyranny of words, in this case one of the most misused terms in the economic vocabulary.
Profitability to the planter
As noted above, Phillips’ (1905) contention that slavery was unprofitable to the planters found support in several studies of plantation records. Recent attacks on these studies have come from two directions. First, the records were found to be incomplete and to contain a crucial conceptual error. Second, economists have applied the traditional tools of economic analysis to test for the profitability of slavery in a manner which removed the reliance upon the fortunes of those particular plantations whose complete records survived. The results of this analysis contradicted the Phillips hypothesis.
Until at least the 1820s and probably later, many contemporary observers believed that the factor endowment of the United States could never support a large manufacturing sector. Their reasoning was not very different from a simple comparative advantage argument that, because of the abundance of land, the high productivity of labor in American agriculture would keep wages too high for manufacturing to compete successfully internationally.
One of the major tasks of nineteenth-century American economic history is explaining why these straightforward predictions, consistent with basic economic principles, did not prove correct. A full discussion of the topic would take us far afield, but even a partial list of relevant considerations would have to include the creative entrepreneurial and technological energies developed a uniquely American manufacturing technology and organization, which substituted for scarce factors and made use of abundant natural resources like timber.
The use of plantation records by Phillips and his followers contained several errors, the importance of which were first stressed by Govan, and later reiterated by Stampp. A conceptual error in accounting technique led to an erroneous conclusion. Phillips included interest on the capital invested in slaves and land as costs in his discussion of the expenses of plantation owners. The later accounting studies similarly included the imputed interest on invested capital as a cost to be deducted from revenues in calculating net profits. This net profit figure was then divided by the capital value of the plantation to measure the average rate of return on capital.
Since the rate of return computed in this manner was below the rate of return on alternative investments, slavery was considered unprofitable. Govan (1942) both pointed out that this was an illegitimate calculation, since it resulted in double-counting the cost of capital. Net profits are computed by deducting all expenses (including depreciation) from gross revenues. A positive residual includes the imputed return on the capital of the owner, as well as the wages of management and ‘pure profits’. When imputations for capital cost and the wages of management are deducted any positive sum remaining would be ‘pure profits’.
The existence of such ‘pure profits’ would indicate that the investment was profitable. There is no need to then compare this with the alternative rate of return, since deducting the imputed capital cost allows for this. The same test could be made by deducting from net profits the imputed wages of management, computing the rate of return upon this investment, and then comparing it with the rate of return upon alternative assets.
If the rate of return on the investment exceeds that upon alternative assets, the investment is considered to be profitable. Both methods described provide the same answer; what is shown to be profitable by one computation is profitable under the other. (Accounting and economic approaches differ in the way costs are measuredâthe accounting calculations using historical costs and the economic current opportunity costs ( Yasuba, 1961, pp. 60-67 ).)
Conclusion
In the South, slavery was never in serious danger, but what doubts there were dissipated with the appearance of cotton in the South in the 1790s. The significance of the rise of cotton in the 1790s was that it opened up a much broader area of the South to commercial agriculture and the profitable use of slaves. The emergence of cotton is typically dated from Whitney’s invention of the cotton gin in 1793, but it now seems clear that the appearance of the gin was only a dramatic episode in a process which had its origins in demand-side pressure. The high profitability of sea-island cotton during 1785-95 stimulated many attempts to grow this long-staple variety farther inland; however, only the short-staple green-seed cotton would grow in the upland areas.
This development, in turn, created the bottleneck that called forth Whitney’s gin, because the fuzzy fibers of upland cotton clung to the seeds and could not be separated with the old roller gin. In the year preceding Whitney’s arrival in Georgia, it is estimated that two to three million pounds of short-staple had been harvested, only a small fraction of which had been ultimately cleaned and sold.
It was never clear how much reliance should be placed upon conclusions drawn from the analysis of those few plantations whose records survived. Not all plantations kept records and even fewer were preserved. Special factors could have affected each plantation, but no adjustment can be made without prior knowledge of the biases. It is therefore not certain that the existing sample of plantations is representative.
There were long and short cycles in prices and output, regional variation was pronounced, and there were apparently differences based upon size of plantation. Conrad and Meyer did use these records, but they used only certain of the data they contained as sample observations in determining estimates of particular variables. They did not generalize from the profit position of a small number of plantations.
Saraydar was interested in testing a more restricted hypothesis about slave profitability than were Conrad and Meyer. He was concerned only with the profitability of owning male prime field hands; thus ignoring the gains from slave offspring which Conrad and Meyer accounted for in computing the rate of return upon females.
Saraydar (1964) understated the profitability of slave ownership to the individual plantation owner (as well as to the southern economy) as long as female slaves were a profitable investment. The specific formulation of the test for the profitability of male slaves used by Saraydar was the same as that of Conrad and Meyer. He computed the rate of return from the use of males in cotton production using averages of the period 1830-60.
His one major change was to provide an alternative measure of physical productivity in cotton operations. Rather than an average of estimates presented in various ‘contemporary journals’, Saraydar used a sample of counties taken from the 1849 census to estimate slave yields in various parts of the South.
The particular year chosen, 1849, was justified as being one of average crop size for the 1830-60 period. Saraydar selected counties in which little was produced besides cotton, and then adjusted the total slave population to obtain an estimate of the average number of field hands. Dividing total cotton output in these counties by the estimated number of field hands gave cotton productivity per field hand.
References
Conrad and Meyer, (1961) The Economics of Slavery, 82; Richard A. Easterlin, ‘Regional Income Trends, 1840-1950,” in Seymour Harris (ed.), American Economic History New York.
Gates, Henry Louis and Nellie McKay (1999) The Norton Anthology of African-American Literature.
Genovese, E. D. ( 1962), “The significance of the slave plantation for southern economic development”, J. of S. Hist., vol. 28, pp. 422-37.
Genovese, E. D. ( 1965), The Political Economy of Slavery in the Economy and Society of the Slave South, Pantheon.
Govan, T. P. ( 1942), “Was plantation slavery profitable?” J. of S. Hist., vol. 8, pp. 513-35.
Hilary beckles, (1988) “Caribbean Anti-Slavery: The Self-Liberation Ethos of Enslaved Blacks,” Journal of Caribbean History.
Phillips, U. B. ( 1905), “The economic cost of slave-holding in the cotton belt”, Pol. Sci. Q., vol. 20, pp. 257-75.
Ramsdell, C. W. ( 1929), “The natural limits of slavery expansion”, Mississippi Valley Hist. Rev., vol. 16, pp. 151-71.
Saraydar, E. ( 1964), “A note on the profitability of antebellum slavery”, Southern Econ. J., vol. 30, pp. 325-32.
Selwyn carrington, (2002) The Sugar Industry and the Abolition of the Slave Trade (Tallahassee, FL.
Sutch, R. ( 1965), “The profitability of antebellum slavery – revisited”, S. Econ. J., vol. 31, pp. 365-77.
Yasuba, Y. ( 1961), “The profitability and viability of plantation slavery in the U S”, Econ. Studies Q., vol. 12, pp. 60-67.