Hostname: page-component-78c5997874-4rdpn Total loading time: 0 Render date: 2024-11-19T06:26:51.175Z Has data issue: false hasContentIssue false
  • English
  • Français

The Economic Cost of the American Civil War: Estimates and Implications

Published online by Cambridge University Press:  11 May 2010

Claudia D. Goldin  and
Frank D. Lewis
Claudia D. Goldin
Affiliation:
Princeton University
Frank D. Lewis
Affiliation:
Queen's University, Ontario
Get access Rights & Permissions [Opens in a new window]

Extract

We are right to see power, prestige, and confidence as conditioned by the Civil War. But it is a very easy step to regard the War, therefore, as a jolly piece of luck only slightly disguised, part of our divinely instituted success story, and to think, in some shadowy corner of the mind, of the dead at Gettysburg as a small price to pay for the development of a really satisfactory and cheap compact car with decent pick-up and road-holding capability. It is to our credit that we survived the War and tempered our national fiber in the process, but human decency and the future security of our country demand that we look at the costs. What are some of the costs? Robert Penn Warren, The Legacy of the Civil War (New York: Random House, 1961), pp. 49–50.

Type
Articles
Information
The Journal of Economic History , Volume 35 , Issue 2 , June 1975 , pp. 299 - 326
Copyright
Copyright © The Economic History Association 1975

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 See Cochran, Thomas C., “Did the Civil War Retard Industrialization?” Mississippi Valley Historical Review, XLVIII (September 1961), 197–210CrossRefGoogle Scholar; Engerman, Stanley, “The Economic Impact of the Civil War,” Explorations in Entrepreneurial History, Second Series, III (Spring-Summer 1966), 176–199Google Scholar; Salsbury, Stephen, “The Effect of the Civil War on American Industrial Development,” in Andreano, Ralph, editor, The Economic Impact of the American Civil War (New York: Schenkman Publishing Co., 1967)Google Scholar. The Cochran and Engerman articles are also reprinted inthe Andreano volume. See also Scheiber, Harry N., “Economic Change in the CivilWar Era: An Analysis of Recent Studies,” Civil War History, II (December 1965)Google Scholar, for a summary of these and other papers.

2 Clark, John Maurice, The Costs of the World War to the American People (New Haven: Yale University Press, 1931), p. xiGoogle Scholar.

3 On the issue of commercial stoppage see R. A. Kessel and A. A. Alchian, “Real Wages in the North During the Civil War: Mitchell's Data Reinterpreted,” reprinted in Andreano, The Economic Impact of the American Civil War.

4 Charles and Mary Beard interpret the Civil War as enabling the North throughvictory to achieve greater economic progress. The Beards state, “The Second American Revolution, while destroying the economic foundation of the slaveowning aristocracy, assured the triumph of business enterprise…. In 1860, just a little more thana billion dollars was invested in manufacturing and only 1,500,000 industrial wage earners were employed in the United States. In less than fifty years the capital had risen to more than twelve billion and the number of wage earners to 5,500,000….” (See The Rise of American Civilization [New York: Macmillan Co., 1927], II, pp. 166–192Google Scholar.) Louis M. Hacker repeated the same theme. “From about 1843 on, this process [the conversion of mercantile capitalism into industrial capitalism] visibly began to take place, stopping short of complete fulfillment because the rising industrial-capitalist class was not in possession of the instrumentalities of political power. In short, industrial capitalism, it became apparent by the late 1850's, was incapable of achieving full maturity unless it had control of the state. In these terms we are to read the meaning of the Civil War; for that conflict was a revolution in the sense that it represented a desperate struggle for political power between two classes each of which required control of the state to underwrite its own economic and socialprograms.” (See The Triumph of American Capitalism [New York: Columbia University Press, 1940], p. 200Google Scholar.)

5 Andreano, The Economic Impact of the American Civil War.

6 For example, see Shannon, Fred, America's Economic Growth (3rd ed.; New York: Macmillan Co., 1951), pp. 325–326Google Scholar. Shannon complicates matters even further by not discounting, by not deflating, and by counting the emancipation of slaves asa real social loss.

7 We equate the Union with the North and the Confederacy with the South merely for literary purposes.

8 The direct cost of the war can be represented graphically using the familiarproduction possibilities frontier where the axes include both present and future consumption of the outputs guns and butter. The distance AB in the figure below represents all expenditures on war machinery. CE is the direct measure of the cost of

the war, but since it is evaluated at the current price ratio, the true cost is over-stated by DE. However, since annual expenditures were a minor fraction of total product (about 10 to 20 percent of GNP for both North and South) DE can be considered negligible. This follows because the marginal cost curve of guns probably had a very high elasticity. The direct cost estimate also may exceed actual foregone consumption because the war may have generated an increase in work effort shifting the production possibilities curve outward. We have not considered this point in our analysis. The destruction of human and physical capital shifts the production possibilities frontier inward, say to the dotted curve in the diagram. The distance FC represents additional foregone future consumption of butter due to the loss of productive capital. The total direct cost of the war is the sum of CD and FC.

9 Randall and Donald put total Union deaths at 360,222. See Randall, J. G. and Donald, D., The Civil War and Reconstruction (2nd ed. revised; Lexington, Mass.: D.C. Heath and Co., 1969), p. 531Google Scholar. Kriedberg, M. A. and Henry, M. G. (History of Military Mobilization in the U.S. Army: 1775–1945, Department of the Army Pamphlet No. 20–212 [Washington, D.C.: Dept. of the Army, June 1955], p. 97Google Scholar) report 359,528 Union deaths, and the U.S. Bureau of the Census (Historical Statistics of the U.S.: Colonial Times to 1957 [Washington, D.C.: U.S.G.P.O., 1960], p.735) puts this figure at 364,511. The Randall and Donald estimate is used in our analysisGoogle Scholar.

10 For many purposes the addition of monetary losses due to deaths and injuries would be clearly inappropriate. We have therefore presented these data separately.

11 For descriptions of the coordination of ordnance activities see Vandiver, Frank E., Ploughshares into Swords: Josiah Gorgas and Confederate Ordnance (Austin: University of Texas Press, 1952)Google Scholar. Ramsdell, Charles W., Behind the Lines in the Southern Confederacy (Baton Rouge: Louisiana State University Press, 1944)Google Scholar also recounts the well intentioned but ineffective southern efforts at maintaining both military troops and civilians in times of extreme scarcity. On troop pay, Scheiber, Harry N. (“The Pay of Confederate Troops and Problems of Demoralization: A Case of Administrative Failure,” Civil War History, XV [1969])Google Scholar states that confederate inability to meet payrolls was encountered early in the war. Wiley, Bell, The Life of Johnny Reb (New York: Bobbs-Merrill Co., 1943)Google Scholar also discusses this problem and the steps the rebels took to provide for their families in the absence of military pay. If and when the soldiers were finally paid, it was usually in far lower real terms than had been promised, due to rapid inflation. This partially accounts for the low reported expenditures given in Table 2.

12 This statistic can be calculated using Tables 1 and 2 and estimates of militarymanpower given in Table 2, fn. g.

13 “[O]n March 26, 1863… [the Confederate] Congress approved an act authorizingand regulating the impressment of private property for public use…. Before supplies could be impressed, officers had to try to buy them from the owner…. Originally, the list prices were only slightly less than the market price, but by the end of the war they had dropped far below.” Todd, Richard C., Confederate Finance (Athens: University of Georgia Press, 1954), pp. 165–171Google Scholar.

14 See Table 2, fn. e, for a justification of this assumption.

15 See Table XII of “Notes”, for Union army wage rates.

16 The basic assumption chosen is, of course, one of many which could be applied to such a hypothetical warless economy. Nonetheless, there is some evidence to supportour choice. George Rogers Taylor concluded in “The National Economy Before and After the Civil War,” that “the economy had developed a tremendous thrust during the 1840's and 1850's, a momentum the Civil War may have temporarily retarded or accelerated but could not, or at least did not, fundamentally affect.” See Gilchrist, D. T. and Lewis, W. D., editors, Economic Change in the Civil War Era (Greenville, Del.: Eleutherian Mills-Hagley Foundation, 1965), p. 22Google Scholar. In summarizing this conference volume Harold F. Williamson noted that with the exception of commercial banking and possibly government-business relations, “principal speakers and critics generally agreed mat the Civil War had relatively little or no effect on the particular institutions they were asked to discuss” (p. 172).

17 The indirect cost of the war can also be represented graphically using aproduction possibilities frontier. We again draw the axes in terms of present and future guns and butter. If the only effect of the war were to increase the output of guns, from OA to OB, and reduce that for butter, from OD to OC, then the direct and indirect measures would be almost identical. We could represent the indirect cost as CD on the butter axis, that is the foregone consumption due to the war. Butif the war involved factors such as political instability, commercial disruption, andcapital destruction, the production possibilities frontier would shift inward, say tothe dashed curve in the diagram, and the cost would increase by FC. The direct estimate picks up only part of this additional cost. Our indirect computation should incorporate all foregone consumption.

Another possibility to consider is an outward shift in the production. possibilities frontier due in some way to the outcome of the war. We depict the situation in the North if there are benefits which accrue to the Union, as implied by the Beards and Hacker. The war cost the Union CD in foregone butter, but if CE of this commodity is returned as the victor's prize, only ED remains as the net indirect cost. If the production possibilities frontier shifts out sufficiently, CE can be greater than CD, implying that net benefits could have accrued to the citizens of the triumphant North.

18 For a discussion of the change in post-bellum savings rates and its possiblesources see Williamson, J. G., “Watersheds and Turning Points: Conjectures on theLong Term Impact of Civil War Financing,” Journal of Economic History, XXXIV (September 1974), 636–661CrossRefGoogle Scholar.

19 This assumes a 6 percent discount rate, and we present results using alternativerates below. These additional computations indicate that our quantitative results arenot very sensitive to the discount value chosen.

20 See Fogel, R. W. and Engerman, S. L., Time on the Cross: The Economics ofAmerican Negro Slavery (Boston: Little Brown and Co., 1974)Google Scholar, Chapter 6 for a discussion of scale economies. Other war and emancipation related costs are also included. For example, the value of cotton burned during the war is implicitly incorporated in the indirect measure. Although there is no estimate of the quantity actually destroyed, most accounts report it was substantial. See Eaton, C., A History of the Southern Confederacy (New York: Macmillan Co., 1954), p. 241Google Scholar.

21 R. Ransom and R. Sutch (What Was Freedom Worth? [forthcoming], Chapter 3, “The Myth of the Devastated South”) estimate that total male equivalent workhours supplied to agriculture per capita declined by about 32 percent due to emancipation (p. 13). The average wage rate of farm labor in the states which seceded was $182.08 (current) in 1870. See Lebergott, S., Manpower in Economic Growth (NewYork: McGraw-Hill, 1964), p. 539Google Scholar for monthly earnings, and Eldridge, H. T. and Thomas, D. S., Demographic Analysis and Interrelations, Population Redistribution and Economic Growth: United States, 1870–1950, Vol. 1 (Philadelphia: American Philosophical Society, 1964), pp. 609–621Google Scholar for the farm labor force used as weights in each state. Wage rates are augmented by one third to allow for board. The 1860 slave population in the states which seceded (3, 521, 110) is assumed to increase between 1860 and 1870 at the rate given for all blacks in the South, implying a freedman population of 4,239,461 in 1870. Applying Ransom and Sutch's cohort shares (p. 13), multiplying by the fraction of each group engaged in agriculture, and adjusting for the lower productivity and shorter work year of women and children yields 1,704,800 male-equivalent workers in 1870. This group earned by our calculations 213,836 thousand 1860 dollars but would have received 47.3 percent more at the higher 1860 work levels. This addition represents 18.97 percent of actual 1870 southern consumption. Therefore, the higher pre-war work effort increases measuredconsumption in 1870 by 18.97 percent. Applying this same proportion to actual southern consumption in every year following the Civil War yields a total present value for the increased leisure of former slaves of $1.96 billions of 1860 dollars in 1861.

We believe this figure is an upper bound because the Ransom and Sutch estimate of the decline in work effort appears quite high. For example, in a recent paper R. Keith Aufhauser suggests that work effort in Louisiana by ex-slaves and poor whites together rose after the war. Therefore, even if the work effort of ex-slaves fell, it may have been offset by an increase in work by the whites. See R. Aufhauser, K., “The Effects of Emancipation in Louisiana and Jamaica,” unpublished manuscript (Queen's University, N.Y.)Google Scholar. In addition, although work effort may have dropped substantially just after emancipation, it probably rose in subsequent decades.

Wright, Gavin, “Cotton Competition and the Post-Bellum Recovery of the American South,” Journal of Economic History, XXXIV (September 1974), 610–635CrossRefGoogle Scholar, discounts the importance of the decline in work effort ana the loss of scale economies, relative to the issue of the growth in demand for cotton. He states that “productive efficiency per se may be less important for the study of southern income growth than the position of the South in the world economy” (p. 635). If the demand for cotton rose at a slower rate after than before the war, some of our indirect measure wouldbe capturing this change, which is probably not due to the war.

22 We have separated costs due to war deaths from other components because thisis an item which can be computed in several ways depending on one's point of view. In addition, there are many uses of these statistics for which an estimate including war deaths would be inappropriate.

23 Note that we have excluded children who would have been born to men whose deaths were due to the war.

24 Most of these arguments concern the detrimental effects of slavery on thesouthern economy. For an excellent summary of this literature see Fogel, R. W. and Engerman, S. L., “The Economics of Slavery,” in their The Reinterpretation of American Economic History (New York: Harper and Row, 1972)Google Scholar.

25 See Table XV of “Notes.”

26 Hacker, The Triumph of American Capitalism, p. 251.

27 See Cochran, “Did the Civil War Retard Industrialization?” and Engerman, “The Economic Impact of the Civil War,” in Andreano, The Economic Impact of the American Civil War.

28 Our reading of the Beards and Hacker is only one of several possible interpretations. For example, they could be implying that the distribution of income, notactual consumption, was changed. But even if this is the correct interpretation, we can test the proposition that the North alone gained at the expense of the South. In addition, although the Beards and Hacker allow for “spillover” effects to other sectors, they may be saying that only northern capitalists gained at the expense of labor. We have not attempted to test this proposition. See Engerman, “The Impact of the American Civil War,” on this point.

29 As noted above, this conclusion holds only if we assume that per capita consumptionin the North would have continued to grow at close to, or above, the prewar rate in the absence of a war. Using both alternative hypotheses (Case 3, Table 8) changes our results, and we get a, $1.09 billion ($3.366-$2.271) gross gain to the North from the war. But since the indirect cost is still positive this possible redistribution was outweighed by other costs resulting from the conflict.

30 We have netted out losses due to a decline in freedman's work effort by subtracting our upper bound figure of $1.96 billion.

31 $6.66 billion has been accounted for in our direct cost estimate although theindirect estimate yields ($14.70 – $1.96 = $12.74). We subtract from the indirectmeasure our estimate for the cost of greater leisure time in the post-bellum South.