QUANTIFYING GREED AND GRIEVANCE IN CIVIL WAR: THE AMERICAN WAR OF INDEPENDENCE

Abstract

‘Greed’ vs. ‘grievance’ is weighed using a generally applicable methodology as motivations in the American War of Independence. Greed is quantified as the expected economic benefit of Independence – escaping colonial trade burdens and expected increased economic growth rates. Grievance is measured as willingness to pay to escape perceived political burdens. Quantification of the relative contributions is made possible by using estimates of expected war-costs. To the extent that the economic burden was insufficient to explain the War, the residual is ascribed to the grievance motivation. Both motives are shown to have contributed to the War, but grievance dominates.

Keywords:

• American revolution
• Civil war
• Causes of war
• American War of Independence
• Secession

JEL Classification:

• F54
• N41

Notes

A ‘thank you’ is due to Stephen Sacks for computational aid.

¹The Declaration of Independence complains about both economic and political burdens of colonial status, with political grievances against the Crown being the more prominent: ‘the history of the present King of Great Britain is a history of repeated injuries and usurpations, all having in direct object the establishment of an absolute tyranny over these states’. An economic concern was the ‘cutting off our trade with all parts of the world’.

²There is also a literature on the optimum size of states, or, more generally, ‘political units’ such as empires - see Alesina, Spolaore, and Wacziarg (2000), Alesina and Spolare (2005), Bolton, Roland, and Spolaore (1996), Bolton and Roland (1997), and Wittman (1991). However, this literature overlooks the costs of war in the formation of new political units. Thus, in this literature, large size has benefits – such as the advantages of specialization and trade according to regional comparative advantage as well as economies of scale in the provision of public goods. However, large size has costs too, largely due to differences in preferences across populations over the choice of public goods. Optimum size alters as the arguments in these benefit and cost functions change over time. Using these ideas, it is easy to describe some events that led to the American Revolution. Thus, defeat of the French in North America in the French and India Wars (1756–1763) reduced the need as perceived by some colonists for low-cost provision of defense ‘services’ (a public good) through the British Empire; George III and the British Parliament in the dozen years prior to 1775 tried to impose a system of government (another public good) that did not match the preferences of the American colonists; and restrictions on trade (through the Navigation Acts), far from promoting beneficial specialization and exchange reduced colonial income per head. However, what is absent from this ‘optimal size’ literature is a strong theoretical basis explaining why secessions are often preceded by civil war. Change in optimum size is not the whole story; somebody has to make it happen, often by force of arms.

³See Rapoport (1971, p. 13).

⁴According to John Adams about one-third of Americans were patriots, one third loyalists and one third were neutral.

⁵Here we ignore the collective action problem of gathering a revolutionary group together. The fact is that such a group did form and was, demonstrably, large enough for the probability of victory, pr_t, to be greater than zero. The gathering together of the revolutionaries is of course the subject of many historical studies.

⁶Gross gives examples of war-cost shifting in the town of Concord (1976, pp. 147–153); for example, voluntary enlistment in the Revolutionary army was replaced by the draft in the summer of 1776, and for two years after that by pulling names from a hat. But if rich enough somebody who was drafted could pay for somebody else to go in their place – as Gross points out, the Revolution became a ‘poor man’s fight’, Gross (1976, p. 147).

⁷They also estimate the per capita burden of the civil war as ranging from a low in 1861 of 6.9% of per capita income to a high of 30.5% in 1865.

⁸The net burden subtracts the benefits of colonial status, such as access to British markets, from the gross burden caused by the trade restrictions.

⁹Quoted New York Times, April 18th, 1909, http://query.nytimes.com/mem/archive-free/pdf?res=9E06EFD9113EE033A2575BC1A9629C946897D6CF.

¹⁰Revolutionary soldiers certainly incurred a nonzero probability of death, and even Revolutionaries who were not soldiers may have faced death – especially, had the British won.

¹¹Egnal (1988) examines, on a person by person basis, the political activities of ‘Expansionists’ – those very likely to become patriots in the Revolutionary War. For example, in Massachusetts some were ‘merchant leaders’ of nonimportation in 1768 or are identified as being engaged in activities at earlier dates. In New York colony, many Expansionists are identified as being politically active in the 1740–1759 period. In Pennsylvania, many Expansionists saw military activity in 1748 or 1755.

¹²The test payback period acts rather as do confidence intervals in statistical analysis – they allow a researcher to choose between acceptable and nonacceptable hypotheses.

¹³According to Deane and Cole (1967, p. 80), the rate of growth of British real gross output and real per capita output accelerated from an average rate of 0.3% p.a. from 1695 to 1765, to 0.9% p.a. to 1805.

¹⁴Adjusting the growth rate after 1799 upward from 1.2% per annum to 1.5% makes hardly any difference, as it is not until 1822 that net present value turns positive. In fact, raising the discount rate from 3 to 7.5% undoes even that small improvement in payback period – NPV does not turn positive until after 1831.

¹⁵Even with a 0.0% discount rate, the minimum total burden need to meet the 10-year test payback period under scenario (2) that falls only to 11.3%. Given an economic burden of 2%, WTP to remove the political burden would still be 9.3/11.3 = 82% of the total burden.

¹⁶Perkins (1994) table 5.4.

¹⁷It is apparent, therefore, that there was scope for some war-cost shifting, especially away from the southern states to the northern states and from the wealthy plantation and merchant classes onto the general population whether engaged in, or, even not supporting in the war.

¹⁸To this total Manning (1956) adds $84.2 million in interest cost and $70.0 million in veterans’ benefits, giving grand total cost of the American Revolutionary war in specie of $255.3 million. However, as the latter two items were incident on tax-payers after 1783, rather than citizens during the war period, they will not be considered here as relevant to the calculation of war-cost; the meaning of ‘war-cost’ is the cost incident on decision-makers at the time of the war. Similarly, Resch (1999, p. 142) indicates the considerable strain on Federal government finances caused by paying veterans’ benefits – so great that in 1819 that they ‘almost bankrupted the government’.

¹⁹According to Chambers (1999, p. 849), 6,824 revolutionaries were killed, 8,445 wounded, and 18,500 noncombatant civilians were also killed.

²⁰Goldin and Lewis (1975), Tables 1, 2 and 8.

²¹They also estimate the per capita burden of the Civil War as ranging from a low in 1861 of 6.9% of per capita income to a high of 30.5% in 1865.