* This research is dedicated to two of my teachers: Douglass C. North, who taught me that differences in institutions were important and needed to be explained by economists, and Gary S. Becker, who taught me that seemingly non-economic issues could be analyzed economically by turning them into cost minimization problems. Earlier versions of this paper were presented at the Conference on Production and Coercion: Unfree Labour in Comparative Perspective, University of Pennsylvania, November 1992; at the Eleventh International Economic History Congress, Milan, September 1994; at the University of Delaware History Seminar, February 1994; and at the Purdue University Economics Seminar, April 1995. The author thanks Anne Boylan, Pieter Emmer, Jon Haveman, Lynn Lees, Ken Matheny, Doug Munro, Janet Netz, and the other participants of these seminars for helpful comments, and thanks Anne Pfaelzer de Ortiz for her editorial assistance.

¹ For a discussion of self-enforcing contracts see Carmichael, H. Lorne, ‘Self-Enforcing Contracts, Shirking, and Life Cycle Incentives’, Journal of Economic Perspectives 3 (1989) 65–83.CrossRefGoogle Scholar

² See Morris, Richard B., Government and Labor in Early America (Boston 1981) 390–531Google Scholar; Smith, Abbot Emerson, Colonists in Bondage: While Servitude and Convict Labor in America 1607–1776 (New York 1947) 226–284Google Scholar; Steinfeld, Robert J., The Invention of Free Labor: The Employment Relations in English and American Law and Culture. 1350–1870 (Chapel Hill 1991) 3–121.Google Scholar

³ See Galenson, David W., White Servitude in Colonial America (New York 1981) 5–15Google Scholar; Morgan, Edmund S., American Slavery, American Freedom: The Ordeal of Colonial Virginia (New York 1975) 123–130; Smith, Colonists in Bondage, 8–16.Google Scholar

⁴ See Dunn, Richard S., Sugar and Slaves: The Rise of the Planter Class in the English West Indies, 1624–1713 (New York 1972) 72Google Scholar; Calenson, White Servitude, 10–15, 171–172; Morgan, American Slavery, American Freedom, 71–130; Morris, Government and Labor, 391–531; Nash, Gary B., The Urban Crucible (Cambridge, MA, 1979) 15CrossRefGoogle Scholar; Salinger, Sharon V., ‘To Serve Well and Faithfully’: Labor and Indentured Servitude in Pennsylvania, 1682–1800 (New York 1987); Smith, Colonists in Bondage, 226–284; Steinfeld, The Invention of Free Labor.Google Scholar

⁵ Dunn, Sugar and Slaves, 72.

⁶ Morgan, American Slavery, American Freedom, 126–127.

⁷ The same prohibition on resale was generally true for craft apprentices in England and America, see Morris, Government and Labor, 364.

⁸ Morgan, American Slavery, American Freedom, 128–129.

⁹ See Grubb, Farley, ‘Servant Auction Records and Immigration into the Delaware Valley, 1745–1831: The Proportion of Females Among Immigrant Servants’, Proceedings of the American Philosophical Society 133 (1989) 157Google Scholar; Grubb, Farley, ‘The End of European Immigrant Servitude in the United States: An Economic Analysis of Market Collapse, 1772–1835’, Journal of Economic History 54 (1994) 810; Morris, Government and Labor, 364.CrossRefGoogle Scholar

¹⁰ See Farley Grubb, ‘Differential Punishments of Runaway Indentured Servants in Seventeenth Century Maryland’ (working paper 90–95, Economics Dept, University of Delaware); Morris, Government and Labor, 391–531; Smith, Colonists in Bondage, 226–284.

¹¹ See Grubb, Farley, Runaway Servants, Convicts, and Apprentices Advertised in the Pennsylvania Gazette, 1728–1796 (Baltimore 1992)Google Scholar; Morris, , Government and Labor, 391–461; Pennsylvania Gazette, 1729–1786 (Philadelphia 1968); Salinger, ‘To Serve Well and Faithfully’, 103–114.Google Scholar

¹² Smith, Colonists in Bondage, 264, 277.

¹³ Grubb, ‘Servant Auction Records’, 159; Grubb, F. W., ‘The Disappearance of Organized Markets for European Immigrant Servants in the United States: Five Popular Explanations Reexamined’, Social Science History 18 (1994) 9–12.CrossRefGoogle Scholar

¹⁴ Rappaport, Steve, Worlds Within Worlds: Structures of Life in Sixteenth-Century London (New York 1989) 232–233, 311–322.CrossRefGoogle Scholar

¹⁵ Nash, The Urban Crucible, 15; Salinger, ‘To Serve Well and Faithfully’, 46.

¹⁶ See Dunn, Sugar and Slaves, 72; Morgan, American Slavery, American Freedom, 71–130.

¹⁷ See Dunn, Sugar and Slaves, 46–83, 224–262; Grubb, Farley and Stitt, Tony, ‘The Liverpool Emigrant Servant Trade and the Transition to Slave Labor in the Chesapeake, 1697–1707: Market Adjustments to War’, Explorations in Economic History 31 (1994) 376–405CrossRefGoogle Scholar; Menard, Russell R., ‘From Servants to Slaves: The Transformation of the Chesapeake Labor System’, Southern Studies 16 (1977) 355–390.Google Scholar

¹⁸ See Morgan, American Slavery, American Freedom, 71–130; Smith, Colonists in Bondage, 226–284.

¹⁹ For a contract to be broken, the terms specified in the contract must have been initiated. The contract is broken when some terms are fulfilled but others are not by the end of the contract. By contrast, an agreement to contract at some future date that is not subsequently honoured and whose execution was never initiated is not considered here to be a broken contract, because none of the contracted terms were ever executed.

²⁰ The party paid first wants everyone else to be moral so that opportunities to commit an immoral act, e.g. breaking the contract, are present. This is ‘free riding’ or taking advantage of the moral behavior of others.

²¹ This is an example of the ‘invisible hand’, Smith, Adam, Wealth of Nations (New York 1937) 423Google Scholar. For an application of the same idea to the explanation of the rise of factory discipline during the industrial revolution, see Clark, Gregory, ‘Factory Discipline’, Journal of Economic History 54 (1994) 128–163CrossRefGoogle Scholar. The same idea can be seen at work in contemporary society in the example of speed limits for automobiles. As an individual I want a twenty-five mph speed limit on the street in front of my house to stop fast drivers from whizzing by and risking my family's life, and I want the limit enforced, but when I'm in my car, if I can violate the law in front of other people's houses, I will. I don't want to be caught and don't like being caught but I still support the law and its enforcement. Without it, I know my life would be far worse. So I've knowingly and voluntarily submitted myself to being coerced and stopped from doing something that I personally would benefit from (speeding in my car) in order to stop others from doing something that causes me harm (the speeding by other people).

²² In equilibrium, the relative amounts of laws, customs, and economic sanctions are chosen such that the value of increasing the probability of achieving a completed contract through increasing the use of an input, divided by the cost of increasing that input, is equalized for all inputs. This is the standard cost-minimizing decision rule for the choice of inputs used by the firm to produce a given level of output: choose the levels of x and y inputs such that VMP_x/P_x = VMP_y/P_y Here the problem is to choose the least costly combination of l, c, and e to produce a given level of contract compliance, which is the same as choosing l, c, and e such that VMP_l/C_l = VMP_c/C_c = VMP_e/C_e, where VMP = value of marginal product, l = legal coercion, C = marginal opportunity cost, c = customary coercion, e = economic reputation sanction, P = price.

²³ Kussmaul, Ann, Servants in Husbandry in Early Modern England (New York 1984) 31–93Google Scholar. See also, Beier, A.L., Masterless Men: The Vagrancy Problem in England, 1560–1640 (London 1985)Google Scholar; Laslett, Peter, The World We Have Lost: England Before the Industrial Age (second ed., London 1971)Google Scholar; Macfarlane, Alan, The Origins of English Individualism: Family, Property and Social Transition (New York 1978); Morgan, American Slavery, American Freedom, 58–70.Google Scholar

²⁴ The few contracts that were less than a year long were designed to end at the same time as the typical year-long contract, Kussmaul, Servants in Husbandry, 32.

²⁵ Borrowing to pay wages continuously during the years based on the security of the crop just shifts the contract problem from farmer/labour to farmer/banker. It is the same contracting problem because non-synchronized payments are involved. An interesting line of inquiry would be to investigate whether the ability to create self-enforcing contracts became less costly for farmer/banker trades versus farmer/servant trades and so explains the disappearance of servitude-in-husbandry in English agriculture.

²⁶ Kussmaul, Servants in Husbandly, 62.

²⁷ On servant mobility see Beier, Masterless Men; Kussmaul, Servants in Husbandry, 31–93; Laslett, The World We Have Lost, Macfarlane, The Origins of English Individualism.

²⁸ See Galenson, White Servitude, 41–43, 200–203; Kussmaul, Servants in Husbandry, 31; Smith, Colonists in Bondage, 14–15.

²⁹ Kussmaul, Servants in Husbandry, 31–32; Steinfeld, The Invention of Free Labor, 74.

³⁰ Kussmaul, Servants in Husbandry, 4, 33, 46–47, 66–67. See also Beier, Masterless Men; Laslett, The World We Have Lost; Macfarlane, The Origins of English Individualism.

³¹ The skills learned by the apprentice are general human capital, saleable to any master or customer in the market. Being the ultimate beneficiary, workers will have to pay to acquire these skills. Masters will not pay to have workers trained, unless they have long-term property rights to the workers’ labour, such as under slavery. For a discussion of general human capital and who pays for the training see Becker, Gary S., Human Capital (second ed., Chicago 1980) 15–44.Google Scholar

³² This is a common assumption in the literature, for example see Elbaum, Bernard, ‘Why Apprenticeship Persisted in Britain but not in the United States’, Journal of Economic History 49 (1989) 337–349CrossRefGoogle Scholar; Gillian Hamilton, ‘Enforcement in Apprenticeship Contracts: Were Runaways a Serious Problem? Evidence from Montreal’, Journal of Economic History (forthcoming); Quimby, Ian M. G., Apprenticeship in Colonial Philadelphia (New York 1985) 22.Google Scholar

³³ See also Rorabaugh, W. J., The Craft Apprentice (New York 1986) 11.Google Scholar

³⁴ See Davies, Margaret Gay, The Enforcement of English Apprenticeship (Cambridge, MA, 1956);Google Scholar Elbaum, ‘Why Apprenticeship Persisted’, 337–349; Jacoby, Daniel, ‘The Transformation of Industrial Apprenticeship in the United States’, Journal of Economic History 51 (1991) 887–910CrossRefGoogle Scholar; Jacoby, Daniel, ‘The Legal Foundations of Human Capital Markets’, Industrial Relations 30 (1991) 229–250; Morris, Government and Labor, 379; Quimby, Apprenticeship in Colonial Philadelphia, 76–92; Rappaport, Worlds Within Worlds, 234–235.Google Scholar

³⁵ Conversely, if the training were provided early in the contract, the expense incurred thereafter would be counted as a sunk cost Thus, even if apprentices performed so poorly after training that their training-enhanced productivity would not cover the training expenses, masters would not dismiss them early (as long as their productivity continued to exceed the subsistence income of room and board). If the apprentice's productivity were not expected to exceed this subsistence income before being trained, the master would not have contracted with the apprentice in the first place.

³⁶ For example, one type of apprentice contract form had pre-printed clauses which stated that the master could annul the contract within three months if he were dissatisfied with the apprentice for any reason, but that apprentices over age ten thus returned to their parents ‘shall be entitled to two dollars per month’. Another type of apprentice contract form had printed clauses stating, ‘Provided, nevertheless, that, in case the said Minor [apprentice] shall, at any time during his minority, obstinately refuse and continue to refuse whatever the said party of the first part [the master] may lawfully require of him then, in such case, the said party of the second part [the parent of the apprentice] covenants and agrees with the said party of the first part to take back the said Minor, upon due notice being given him by or from the said party of the first part so to do, without making or requiring any Charge for the Service which the said Minor may have performed for the said party of the first part, or for not keeping the said Minor during his minority’. Taken from Shaker apprenticeship contracts made in upstate New York between 1851 and 1866, held in the Edward Deming Andrews Memorial Shaker Collection, folder 751, Winterthur Museum and Library, Winterthur, Delaware, USA. On the use of annulment clauses in Canadian apprentice contracts see Hamilton, ‘Enforcement in Apprenticeship Contracts’.

³⁷ See Hamilton, ‘Enforcement in Apprenticeship Contracts’; Quimby, Apprenticeship in Colonial Philadelphia, 51; Rorabaugh, The Craft Apprentice, 3–15; Yarbrough, Anne, ‘Bristol's Apprentices in the Sixteenth Century: The Cultural and Regional Mobility of an Age Group’, (unpublished PhD thesis. Catholic University of America 1977) 86.Google Scholar

³⁸ See Hamilton, ‘Enforcement in Apprenticeship Contracts’.

³⁹ See Elbaum, ‘Why Apprenticeship Persisted’, 337–349; Grubb, Farley, ‘Fatherless and Friendless: Factors Influencing the Flow of English Emigrant Servants’, Journal of Economic History 52 (1992) 96–97; Jacoby, ‘The Transformation of Industrial Apprenticeship’, 887–910; Jacoby, ‘The Legal Foundations of Human Capital Markets’, 229–250; Morris, Government and Labor, 369; Rappaport, Worlds Within Worlds, 232–238, 291–322. The ability to substitute parental sponsorships for up-front cash fees may have depended on the legality of enforcing parental liability as well as the parents’ level of wealth which could be targeted as compensation in a suit brought by the master for apprentice misconduct. Legal differences between the French civil code and English common law may explain why apprentices in Quebec made more use of parental sponsorship than apprentices in English America. See Hamilton, ‘Enforcement in Apprenticeship Contracts’.CrossRefGoogle Scholar

⁴⁰ Hamilton, ‘Enforcement in Apprenticeship Contracts’.

⁴¹ Rappaport, Worlds Within Worlds, 233–234, 311–322.

⁴² For example see Quimby, Apprenticeship in Colonial Philadelphia, 74; Rorabaugh, The Craft Apprentice, 3–15.

⁴³ See Galenson, While Servitude; Morgan, American Slavery, American Freedom; Morris, Government and Labor, 391–531; Smith, Colonists in Bondage; Steinfeld, The Invention of Free Labor.

⁴⁴ Grubb, Farley, ‘The Incidence of Servitude in Trans-Atlantic Migration, 1771–1804’, Explorations in Economic History 22 (1985) 316–317CrossRefGoogle Scholar. For the typical labourer there was little income left after paying for food, shelter, clothing, and other living expenses which could be accumulated as savings to pay for passage expenses. For German immigrant servants in Pennsylvania it has been estimated that eighty per cent of each year's work effort went to cover food, shelter, and other living expenses, seven per cent went to cover freedom dues expenses, leaving only thirteen per cent of each year's work effort to repay their passage expenses. Grubb, Farley, ‘The Auction of Redemptioner Servants, Philadelphia, 1771–1804: An Economic Analysis’, Journal of Economic History 48 (1988) 588.CrossRefGoogle Scholar

⁴⁵ See Grubb, , ‘Differential Punishment’; Grubb, Runaway Servants; Klepp, Susan L. and Smith, Billy G. eds, The Voyage and Adventures of William Moraley, An Indentured Servant: Or the Infortunate (University Park, PA, 1992Google Scholar; orig. 1743); Narrative of Johann Carl Buettner in the American Revolution (New York 1915).Google Scholar

⁴⁶ See Grubb, ‘The Auction of Redemptioner Servants’, 588; Grubb, Farley, ‘The Long-Run Trend in the Value of European Immigrant Servants, 1654–1831’, Research in Economic History 14 (1992) 185Google Scholar; Grubb, Farley, German Immigrant Servant Contracts Registered at the Port of Philadelphia, 1817–1831 (Baltimore 1994).Google Scholar

⁴⁷ For example see Galenson, White Servitude, 101, 253–254; Galenson, David W., ‘Labor Market Behavior in Colonial America: Servitude Slavery, and Free Labor’ in: Galenson, David W. ed., Markets in History: Economic Studies of the Past (New York 1989) 57Google Scholar; Grubb, ‘The Disappearance of Organized Markets for European Immigrant Servants’, 10; Heavner, Robert Owen, Economic Aspects of Indentured Servitude in Colonial Pennsylvania (New York 1978) 50–51.Google Scholar

⁴⁸ For example, Morris, Government and Labor. 398, suggests that freedom dues were enforced savings so that the servant would not become a public charge when dismissed. Immigrant servitude in other parts of the world made use of return passage tickets as freedom dues. For example see Cloud, Patricia and Galenson, David W., ‘Chinese Immigration and Contract Labor in the Late Nineteenth Century’, Explorations in Economic Histoty’ 24 (1987) 22–42CrossRefGoogle Scholar; Emmer, P.C. ed., Colonialism and Migration: Indentured Labour before and after Slavery (Boston 1986)CrossRefGoogle Scholar; S., Marks and P., Richardson eds, International Labour Migration: Historical Perspectives (Hounslow 1984)Google Scholar; Shlomowitz, Ralph, ‘Melanesian Labor and the Development of the Queensland Sugar Industry, 1863–1906’, Research in Economic History 7 (1982) 327–361. The problem of the role of legally required freedom dues in the institution of immigrant servitude is currently being investigated by the author. Preliminary findings are available upon request.Google Scholar

⁴⁹ Bailyn, Bernard, Voyagers to the West (New York 1986) 85–240; Galenson, White Servitude, 102–113, 220–227; Grubb, ‘The Incidence of Servitude’, 332–337; Smith, Colonists in Bondage.Google Scholar

⁵⁰ See Dunn, Sugar and Slaves. 46–116. Kidnapping came to be called ‘Barbadosized’.

⁵¹ Archives of Maryland (Baltimore, Maryland Historical Society, 1883)Google Scholar; Grubb, ‘Differential Punishment’; Main, Gloria L., Tobacco Colony: Life in Early Maryland, 1650–1720 (Princeton 1982) 116.Google Scholar

⁵² Morris, Government and Labor, 478.

⁵³ Smith, Colonists in Bondage, 56.

⁵⁴ See Adams, Donald R. Jr, ‘Some Evidence on English and American Wage Rates, 1790–1830’, Journal of Economic History 30 (1970) 499–520CrossRefGoogle Scholar on relative wages, and Grubb, ‘The Auction of Redemptioner Servants’, 588 on immigrant servant forced saving rates. It would be hard for labourers to have as high a saving rate in Europe because food, shelter, and other living expenses were relatively higher in Europe, so this estimate may understate the labour time difference. In addition, the wage differential may have been larger earlier in the colonial period, which would also imply that this estimate may understate the labour time difference.