Between 1500 and 1870, roughly 12.5 million people were embarked from Africa onto Atlantic slave ships. About 10.7 million survived the Middle Passage. That flow of human beings, sustained over nearly four centuries by European capital, African coastal supply networks, and American plantation demand, was the largest forced migration in recorded history and the labor foundation on which the Atlantic plantation complex was built. This chapter treats the system as economic history. It asks what the trade moved, how the plantations that received the labor performed as productivity systems, what the trade and the plantations contributed to industrial capitalism in Britain, how the Atlantic powers ended slavery and at whose expense, and what the institutional substitutes for slavery generated in the long run.
Named literature: Slave Voyages Database (Eltis, Behrendt, Florentino, Richardson eds.); Eltis & Richardson (2010); Fogel & Engerman, Time on the Cross (1974); Gutman (1975); David & Sutch (1976); Wright (1978); Olmstead & Rhode (2008, 2018); Hilt (2017); Williams, Capitalism and Slavery (1944); Engerman (1972); Beckert, Empire of Cotton (2014); Baptist, The Half Has Never Been Told (2014); Acemoglu, Johnson & Robinson (2002); Nunn (2008); Albouy (2012); Glaeser, La Porta, Lopez-de-Silanes & Shleifer (2004); Drescher (2009); Berlin & Rowland (1997); Gaffield (2015); Crosby, The Columbian Exchange (1972); Cook (1998); Richter, The Ordeal of the Longhouse (1992); Restall (2003); Foner (1988); Northrup (1995); HM Treasury (2018) disclosure on the British loan retired in 2015.
Twelve and a half million people embarked, ten and seven-tenths million arrived. The gap is the Middle Passage, the ocean crossing on which roughly 1.8 million people died, an average mortality of about 14 percent across the full arc, declining over time as ship management improved and voyage durations shortened. The arrived total is the figure that matters for plantation labor supply; the embarked total is the figure that measures the system’s African extraction. The chapter uses both, and distinguishes them on first reference because the difference is institutionally significant: the embarked-versus-arrived gap is a property of the trade itself, not of the destination economies. The numbers come from the Slave Voyages Database, the Trans-Atlantic Slave Trade Database hosted at slavevoyages.org and consolidated by David Eltis, Stephen Behrendt, Manolo Florentino, and David Richardson; Eltis and Richardson’s 2010 atlas provides the canonical published cross-check.
The trade resolves into five distinguishable phases. The Iberian opening (1500–1639) carried roughly a million people, most of them to Brazil, where Portuguese sugar planters in Pernambuco and Bahia built the prototype of the Atlantic plantation complex on captives drawn from Loango and the Kongo kingdom. Spanish demand for the Caribbean and the mainland silver economy was smaller and supplied through the asiento contract licensing trade to non-Spanish carriers. The Dutch and English entry phase (1640–1699) saw Dutch West India Company captures of Brazilian and African positions, the founding of the Royal African Company in 1672 with English crown monopoly, and the rise of Barbados and Jamaica as English Caribbean sugar economies. Roughly 1.3 million people were embarked in this phase, with the Caribbean share rising sharply and the carrying-nation mix diversifying away from Portuguese dominance.
The industrial-scale system phase (1700–1790) is where the trade reached its peak volumes. Roughly 6.2 million people, half the entire 370-year total, were embarked in these nine decades. British carriers dominated after the Royal African Company’s monopoly was opened to private traders in 1698; Liverpool overtook Bristol and London as the trade’s European capital. Annual flows reached their maximum of approximately 80,000 people per year in the 1780s. The destination mix tilted toward the Caribbean (Saint-Domingue, Jamaica, Cuba) but Brazil remained a major receiver, and mainland North America took its small but socially consequential share. The revolution and abolition phase (1791–1819) opens with the Saint-Domingue uprising in August 1791 and runs through the British 1807 trade abolition and the Anglo-American 1808 prohibitions. Volumes contracted but did not collapse: roughly 1.95 million people were embarked, mostly to Cuba and Brazil, as the legal-trade closure of British and US carriers shifted the residual demand onto carriers and routes that remained legal.
The illegal afterlife phase (1820–1870) carried roughly 2 million people in defiance of multilateral abolition treaties. British naval suppression by the Royal Navy’s West Africa Squadron intercepted approximately 10 percent of voyages; the rest reached Brazilian and Cuban plantations under concealment, false papers, and re-flagging strategies that left the records partial. The phase ended with Brazilian enforcement of the 1850 Eusébio de Queirós law and Cuban abolition in 1886. By destination across the full arc: Brazil received roughly 46 percent of arrivals; the Caribbean roughly 38 percent; the Spanish mainland (Mexico, Venezuela, the Andean ports) roughly 10 percent; mainland North America roughly 4 percent; Europe and elsewhere roughly 2 percent. The mainland North American figure is small in relative terms and structurally significant because of natural population growth among the enslaved: by the 1860 census, the United States held roughly four million enslaved people, far more than its 388,000 imports could account for.
The trade was an institution as well as a flow. The carrying-nation evolution and the destination splits described above tell only half the story; the other half is what made the system work as an organized commercial enterprise. The Middle Passage itself, the ocean crossing from African embarkation port to American arrival port, was the artifact in which the system’s commodification of human beings became most operationally explicit. Voyages took six to ten weeks. Captives were chained below decks in spaces averaging four to six square feet per person. Mortality came from dysentery, scurvy, smallpox, and suicide; the 14 percent average across the full arc concealed wide variation by route, season, ship size, and ship master. Resistance was continuous: shipboard revolts occurred on roughly one voyage in ten by current Slave Voyages Database estimates, and the threat of revolt shaped every operational decision from chaining protocols to crew size to feeding schedules.
The Zong massacre of 1781 makes the system’s commodification legible in a single legal document. The Zong, a British slave ship sailing from West Africa to Jamaica with 442 enslaved Africans aboard, ran low on water due to navigational error. Over three days the captain ordered 132 enslaved people thrown overboard alive. The ship’s owners then filed an insurance claim for the loss of cargo. The London insurance underwriters refused; the case went to court. The legal record turns on whether the people thrown overboard were “jettisoned” cargo (claimable) or murdered (not claimable). The court treated the question as a maritime-insurance dispute. The institutional fact the case surfaces is that the Atlantic system had constructed a category in which 132 human beings could be filed under the same legal heading as damaged sugar.
The trade’s African end ran through coastal supply networks operated by African states and merchants. The Kingdom of Dahomey on the Slave Coast (modern Bénin), the Asante empire on the Gold Coast (modern Ghana), and the kingdoms of Loango and Kongo on the West-Central African coast (modern Republic of Congo, Angola) were the largest suppliers across the trade’s peak phases. Captives reached the coast through wars of capture, raids, debt enslavement, and judicial enslavement. The European trade’s demand intensified all four mechanisms and shifted political incentives toward states organized around military capture. Nathan Nunn’s 2008 paper in the Quarterly Journal of Economics, “The Long-Term Effects of Africa’s Slave Trades,” documents that African regions from which more captives were drawn between 1400 and 1900 are systematically poorer today, with proposed mechanisms running through pre-colonial state-capacity collapse and post-colonial trust deficits. The trade’s African legacy is not a peripheral footnote to the Atlantic system; it is one of its constitutive economic outputs.
The mercantilist regulatory frame held the system together at the European end. The Navigation Acts in England (1651, 1660, 1663), the Royal African Company charter (1672), the Dutch West India Company charter (1621), the French Compagnie des Indes occidentales (1664), and the Spanish asiento contract structure all set rules under which Atlantic trade could be conducted, by whom, in whose ships, and at what duty. Whether mercantilist regulation was a coherent doctrine or a bundle of fiscal and strategic expedients is the subject of long debate; the free-trade walkthrough works the doctrine question end to end. The economic-historical fact this section needs is that the Atlantic slavery system operated inside a regulatory architecture that was protectionist by design and that the eighteenth-century critique of that architecture (Smith and the physiocrats) ran in parallel with the abolitionist critique without yet merging with it.
The figure’s five phases match the Atlantic tab’s five sub-pieces on the trade map by name. The map carries the spatial-temporal voyage-level view—origin ports, routes, destination ports, year-by-year voyage counts—that the chapter’s static figure necessarily flattens. The map’s Liverpool port popup in particular carries the slave-trade capital-flows narrative: Liverpool was the European capital of the trade across its industrial-scale phase, and the popup walks the merchant houses, the financial instruments, and the reinvestment of trade profits into the Lancashire industrial economy that §9.4 below will treat as the Williams thesis’s strongest empirical pillar. The plantation complex that the Middle Passage fed coordinated labor under force at scale across production, processing, and shipment. The next section walks its productivity record.
Sugar in sixteenth-century Pernambuco, tobacco in seventeenth-century Chesapeake, rice in the eighteenth-century Carolina low country, cotton in the nineteenth-century US South: across three centuries the plantation complex moved from one staple to the next while keeping the same labor-coordination logic. The plantation complex, as Philip Curtin used the term, was an integrated economic technology: large-scale monoculture of a tropical or subtropical commodity grown for Atlantic export, organized around coerced labor coordinated through direct supervision, and tied to processing and shipping infrastructure under unified ownership. Prose treatment in this chapter follows that definition. The complex was not a single agricultural enterprise; it was a coordination mechanism.
Brazilian sugar in Pernambuco and Bahia is the prototype. By the early seventeenth century, a Brazilian engenho integrated cane cultivation, on-site mill, refining, and shipment under a single planter or partnership, with enslaved labor performing every stage from planting to crating. The Caribbean sugar economy of the eighteenth century scaled the model. Saint-Domingue, the French western half of Hispaniola, was by 1789 the world’s most productive sugar economy: roughly 40 percent of global sugar output and 60 percent of global coffee output came from a colony of 30,000 white planters, 25,000 free people of color, and approximately 500,000 enslaved Africans. Jamaica was Britain’s largest sugar producer; Cuba scaled most aggressively in the nineteenth century after the Saint-Domingue collapse opened market share. The diversification into other staples ran in parallel: tobacco in the Chesapeake from the 1620s onward, rice in the Carolina low country from the 1690s, indigo in South Carolina and the French Caribbean, coffee in Saint-Domingue and Jamaica.
The labor-coordination forms inside the complex were not uniform. Gang labor, associated with sugar and with post-1793 cotton, organized workers into synchronized groups under continuous overseer supervision; the gang moved together through field operations on a pace set by the overseer and enforced by physical violence. Task labor, associated with rice cultivation in the Carolina low country, assigned each worker a daily quota (a task), after which the worker was free to manage their own time, including subsistence cultivation on small plots. The two systems produced different patterns of resistance, different demographic outcomes, and different post-emancipation institutional inheritances. Gang labor maximized output per acre at the cost of brutal supervision intensity; task labor produced lower per-acre output but sustained higher fertility rates and more autonomous Black community structures within the system.
Eli Whitney’s cotton gin (1793) was the structural inflection that built the nineteenth-century cotton South. Before the gin, separating short-staple cotton fiber from its sticky green seed required roughly ten labor-hours per pound of clean cotton; after the gin, the same operation took an hour or less. The labor cost per pound of clean cotton collapsed, and with it the geographic constraint that had limited cotton cultivation to the long-staple sea-island variety on a narrow Atlantic coastal strip. The upland short-staple zone (Georgia, Alabama, Mississippi, Louisiana, eastern Texas) became cotton country, and the internal slave trade from the Chesapeake to the Deep South moved roughly one million enslaved people across the next sixty years to staff the new plantations. The cotton South of 1860 was not the eighteenth-century plantation system continued; it was a new geography built on a single technology.
What was the plantation complex’s productivity record? Robert Fogel and Stanley Engerman’s Time on the Cross (1974) reported that Southern slave plantations were approximately 35 percent more productive than Northern free farms by total-factor-productivity measures, a finding contested almost immediately by Herbert Gutman (1975), Paul David and Richard Sutch (1976), and Gavin Wright (1978) on data construction, sample selection, and TFP-accounting choices, and one whose core empirical claim has nonetheless survived the methodological backlash in subsequent rounds of work. Total factor productivity measures output not explained by measured inputs of land, labor, and capital; the residual captures combined effects of organizational, technical, and selection factors. The formal model treatment lives in economics ch. 13; for the chapter’s purpose, what matters is that Fogel and Engerman were measuring the productivity of plantation output relative to its inputs, and the plantation complex tested as a high-output system by that metric.
Olmstead and Rhode’s 2008 Journal of Economic History paper and their 2018 follow-up identified the specific mechanism behind the cotton-yield gains that drove much of the antebellum productivity record: selective seed breeding. Cotton-yield-per-acre in the Deep South roughly quadrupled between 1800 and 1860, and Olmstead and Rhode show that the gains tracked the diffusion of improved seed varieties (Petit Gulf cotton in particular) through plantation networks. The mechanism is biological-technical, not coercive. The relevance for §9.4 below is that one strand of the “new history of capitalism” argument identifies intensifying labor coercion as the source of the cotton-productivity gains; Olmstead and Rhode’s evidence locates the gains elsewhere. The chapter takes this question up in §9.4 and lands a position in §9.7.
The plantation complex’s productivity record was real and is one of the chapter’s settled empirical claims. The historiographical context in which that claim was contested includes a much older intellectual line: Adam Smith’s argument in The Wealth of Nations (1776) that slave labor was inefficient relative to free labor because the slave had no incentive to work beyond what the lash extracted. Smith’s argument was empirically wrong about the eighteenth-century plantation as a productivity system, and the abolitionist movements of the nineteenth century carried it forward as a moral-and-economic case against slavery throughout the period of British and American debate. Where Smith sits on the intellectual-history timeline, and how the abolitionist political-economy lineage develops out of his framework, is visible on the history of economic thought timeline.
The plantation system the demographic collapse made room for connected to industrial Britain in ways the chapter’s signature historiographical question is now ready to face. But before that question, the plantation complex was not a uniform Atlantic geography. Outside the plantation zones, indigenous economies persisted under varying degrees of colonial constraint, and the rise of African slavery in the plantation zones was not unrelated to the demographic collapse that had created the labor shortage there in the first place.
The labor shortage that African slavery filled in the plantation zones was not a permanent geographic given. It was created by the sixteenth-century epidemic disease collapse that reduced indigenous American populations by 60 to 90 percent across most regions. Alfred Crosby’s The Columbian Exchange (1972) named the mechanism: smallpox, measles, typhus, and influenza arriving with European contact encountered populations with no acquired immunity, and the resulting mortality compounded across successive epidemic waves through the sixteenth and into the seventeenth century. Noble David Cook’s Born to Die (1998) consolidated the regional estimates. Central Mexican population fell from approximately 25 million in 1518 to roughly 1.3 million by 1623; the Andes saw comparable collapses on a similar timeline. Ch. 4 carries the substantive treatment of pre-contact indigenous economies. This section’s point is the precondition the collapse created for what followed.
Andean reciprocity persisted under Spanish colonial rule in repurposed form. The Inca mit’a, a labor-rotation system in which subject communities supplied periodic labor to the imperial state for public works, transport, and agriculture, was reorganized by the Spanish viceroy Francisco de Toledo in the 1570s into the colonial mit’a that supplied labor to the silver mines of Potosí. Communities within a designated hinterland sent rotating contingents to the mines for one year out of seven; those returning carried back the silver-economy currency that integrated highland village economies into the colonial commercial system. Cuzco, Potosí, and the mining-route towns operated functioning markets in coca, textiles, food staples, and silver-denominated goods through the seventeenth and eighteenth centuries. The reciprocity logic the Inca state had used for its own purposes was carried into colonial extraction; the indigenous-community institutions that hosted the rotation continued to operate as administrative units.
North American fur-trade economies organized indigenous societies as primary economic actors in long-distance commercial networks. The Iroquois Confederacy, comprising the Five Nations (Mohawk, Oneida, Onondaga, Cayuga, Seneca) joined by the Tuscarora in the eighteenth century, controlled the fur trade between the Great Lakes interior and the Dutch and English Atlantic ports through the seventeenth and eighteenth centuries, conducting the Beaver Wars against rival Algonquian-speaking nations to secure trapping territories. Daniel Richter’s The Ordeal of the Longhouse (1992) treats the Iroquois economy as a functioning regional system of contracts, exchange, and territorial control. Further north, the Hudson’s Bay Company chartered in 1670 ran a fur-trade economy with Cree, Ojibwa, Dene, and Inuit suppliers across the Canadian Subarctic. The company’s standardized exchange rates, posted credit, and seasonal supply schedules organized Indigenous-European commercial relations that persisted into the nineteenth century.
Mesoamerican market continuity was the third pattern. The pre-Columbian market system documented in the Aztec capital Tenochtitlan, anchored by the Tlatelolco market with its standardized weights, market-court enforcement, and specialized goods sectors, persisted under Spanish colonial rule through the seventeenth century. Tlatelolco continued to operate as a regional commercial center; smaller market towns in the Valley of Mexico maintained the pre-conquest weekly-market schedule that synchronized rural-urban exchange. Matthew Restall’s work on Yucatec Maya commercial culture documents comparable continuity in southeastern Mesoamerica.
The pattern these threads describe is geographic zoning. The plantation zones (the Brazilian coast, the Caribbean islands, the Carolina low country, the Chesapeake, the post-1793 cotton South) absorbed African labor because demographic collapse had hollowed out the indigenous labor base and because plantation production required labor at intensities that free workers under colonial labor-market conditions would not supply at acceptable cost. Outside the plantation zones, indigenous economies persisted under varying degrees of colonial constraint. Reorganized, taxed, partially Christianized, demographically pressed, they nonetheless operated as functioning systems with measurable continuity into the nineteenth century. Ch. 10 takes up the post-1815 indigenous experience under settler-colonial intensification. The plantation system the demographic collapse made room for connected to industrial Britain in ways the chapter’s signature historiographical question is now ready to face.
Did slave-trade and plantation profits finance British industrialization, or were the magnitudes too small to be load-bearing? Was slavery constitutive of industrial capitalism, or marginal to its emergence? The eighty-year debate that started with Eric Williams’s Capitalism and Slavery (1944) has accumulated four positions, all empirically informed, none reducible to the others. This section walks each at strongest form, then signals where the chapter’s house line lands. The full call is in §9.7.
Williams’s 1944 argument is a macro-financing thesis. Capitalism and Slavery, written from Williams’s Oxford D.Phil. and published while he was a young Trinidadian historian on the threshold of a political career that would make him the country’s first prime minister, advanced three claims that Williams himself treated as a single integrated case. First, the Atlantic slave trade and the West Indian plantation economy generated profits at scale: profits running from African captures through Middle Passage transport through Caribbean sugar production and back through London and Liverpool merchant houses to British investors. Second, those profits financed the early industrial revolution: the Lancashire cotton mills, the Boulton and Watt steam engines, the canal network, the iron foundries of the Midlands. Third, British abolition in 1807 and emancipation in 1833 came when the West Indian plantation system had ceased to be profitable and the British industrial economy had outgrown its dependence on it. The moral movement was real, but the economic foundation that made political abolition feasible was the prior decline of plantation profitability. Williams’s evidence ran through specific cases: West Indian planters as a political-financial bloc in the British Parliament; Liverpool merchant capital as a measurable input to Lancashire textile finance; the triangular trade’s profit accumulation; the timing correlation between abolition and British industrial maturity. The argument’s force depended on the macro magnitudes being load-bearing, not merely present.
The Engerman critique, advanced by Stanley Engerman in his 1972 Business History Review paper and developed across subsequent work, attacks the macro magnitudes directly. Engerman calculated that slave-trade profits represented approximately 5 percent of British total investment during the relevant decades, a non-trivial share but not the load-bearing share Williams’s argument requires. West Indian planter rates of return were comparable to other colonial ventures, not anomalously high. The merchant houses that funneled trade and plantation profits into British finance were one input source among several, and their relative weight in industrial-investment finance never exceeded what diversified British capital markets could have supplied from alternative accumulations. Engerman’s implicit counterfactual is that British industrialization without slave-trade profits would have proceeded at a slightly slower pace from a slightly smaller capital base, but not at a structurally different trajectory. The political weight of the West Indian planter bloc in Parliament, on Engerman’s reading, was institutional rather than financial: planters had political influence disproportionate to their share of British capital, and the abolition timing reflects the erosion of that political influence rather than the prior collapse of plantation profitability. The Williams claim is empirically wrong on the macro question, in this view, even if it remains historiographically generative on adjacent questions.
The new history of capitalism, which emerged in the early 2010s around Sven Beckert’s Empire of Cotton (2014) and Edward Baptist’s The Half Has Never Been Told (2014), reframes the question. The argument is not that slave-trade profits financed British industry at the macro magnitudes Williams claimed; the argument is that slavery was constitutive of industrial capitalism in operationally specific ways that the financing-flow framing misses. Beckert’s case is built on the cotton supply chain. Lancashire textile mills depended on Southern US cotton from the 1820s onward; by 1860, roughly 75 percent of British raw cotton came from the slave South. The supply chain ran through Northern US merchant houses (the Browns, the Lehmans), London commission agents (Baring Brothers, the Rothschilds’ correspondents), and credit instruments that linked Southern plantation finance to British industrial finance through working-capital advances against future cotton crops. Plantation managerialism, the use of standardized accounting, performance measurement, and labor-output records on Southern cotton plantations, anticipated industrial managerialism in measurable ways: the plantation account books that survive show productivity-tracking practices that the textile mill manager would later inherit. Baptist pushes further on the mechanism: slavery’s productivity gains in cotton came from intensifying labor coercion (“the whipping-machine” mechanism), and the modernity of plantation discipline was not a parallel development to industrial capitalism but its precursor. Industrial capitalism, on the new-history view, did not merely benefit from slavery; it grew out of it. The operational integration of the cotton supply chain is the load-bearing piece of this position; the C6 claim of this chapter (Lancashire-South operational integration) lands here.
Olmstead and Rhode (2008, 2018) and Eric Hilt (2017) push back on the new history’s specific empirical mechanisms while granting much of the broader picture. The cotton-productivity gains Baptist attributes to labor coercion intensification are, on Olmstead and Rhode’s evidence, attributable to selective seed breeding: cotton-yield-per-acre roughly quadrupled in the Deep South between 1800 and 1860, and the gains track the diffusion of Petit Gulf and other improved seed varieties through plantation networks rather than tracking measures of coercion intensity. The biological-technical mechanism dominates the productivity record. Hilt’s 2017 review essay in the Journal of Economic History identifies data-construction problems in several of the new history’s quantitative claims, including double-counting in the supply-chain accounting and inconsistent productivity-measure definitions. The productivity-historian critique does not deny that slavery and industrial capitalism were operationally integrated; it denies that the specific mechanisms the new history names—intensifying coercion as the productivity engine, supply-chain integration at the magnitudes claimed—survive close empirical scrutiny. The normative thrust may be right; the specific empirical mechanisms need correction.
The four positions, surfaced at strongest form, do not collapse to a binary. Williams and Engerman disagree about macro magnitudes. Beckert and Baptist disagree with both about whether the macro-magnitudes question is the right question, and locate the constitutive role at the operational supply-chain level. Olmstead and Rhode disagree with Baptist about the specific cotton-productivity mechanism while granting much of Beckert’s broader operational-integration argument. The chapter’s house line takes a position closer to the new history than to Engerman on the constitutive question, while granting Engerman magnitudes on financing-flow specifics and granting Olmstead and Rhode the cotton-mechanism specifics. The full house line is in §9.7; the next section turns to abolition’s mechanics. The British industrial revolution chapter walks the productivity numbers and the institutional preconditions of British industrialization at the depth this section necessarily compresses.
Smith’s anti-slavery efficiency argument and the abolitionist political-economy lineage that develops out of it sit on the classical-era cluster of the intellectual-history timeline; the historiographical context for both the Williams thesis and its successors runs through that lineage. The four-position debate concerns slavery’s role in industrial capitalism’s emergence. Slavery’s end (abolition) has its own economic mechanics, varying across the Atlantic powers in ways that shaped the post-emancipation distribution as much as the trade itself shaped the pre-1865 distribution.
British slaveowners were compensated £20 million when Britain emancipated in 1833. US slaveowners were compensated nothing when the Civil War ended slavery in 1865. Haitian ex-slaves and their descendants paid France 90 million gold francs through the 1880s for the right to be recognized as a sovereign nation. The three cases set the chapter’s asymmetric-compensation thread; the section walks the abolition sequence that produced them and works the Haitian fork in detail.
The abolition sequence is a half-century arc with distinct economic mechanics in each case. France abolished slavery first by decree of the National Convention in 1794, in the political wake of the Saint-Domingue uprising; Napoleon restored it in 1802; the July Monarchy and then the Second Republic re-abolished it definitively in 1848 with compensation to French slaveowners financed from the public treasury. Britain abolished the slave trade in 1807 under the Slave Trade Act, prohibiting the carrying of enslaved Africans on British ships, and emancipated enslaved people in the West Indian colonies in 1833 under the Slavery Abolition Act. The 1833 act paid £20 million in compensation to slaveowners, approximately 5 percent of British GDP at the time and 40 percent of the British government’s annual expenditure. The compensation was financed by a public loan; HM Treasury disclosed in 2018 that the loan was retired only in 2015, meaning that British taxpayers were servicing the debt that had compensated slaveowners through the early twenty-first century. The recipients of the compensation are documented in the Legacies of British Slavery database at University College London, which traces the £20 million across roughly 47,000 individual claims and shows the reinvestment of the proceeds into British industry, country estates, the railway boom, and overseas commercial ventures. The compensation was capital that financed the post-1833 British economy.
The United States abolished slavery in 1865 by the Thirteenth Amendment following the Civil War, with no compensation to slaveowners. The human-capital wealth held by Southern slaveowners, estimated at approximately $3 billion in 1860 dollars and the largest single asset class in the antebellum US economy, was wiped out without indemnification. The destruction of slaveowner wealth was achieved by military defeat rather than legislative buy-out, and the cost of the war (in lives, in physical destruction, in fiscal dislocation) exceeded any plausible compensation figure several times over. Cuba abolished in 1886 under Spanish administration with a graduated emancipation through the patronato system, in which slaveowners retained the labor of the formerly enslaved as “apprentices” for an eight-year transition period under wage-and-rations contracts that approximated continued slavery in practice. Brazil abolished in 1888 under the Lei Áurea signed by Princess Isabel, the last of the Atlantic slave economies to do so. Brazilian abolition came without compensation to slaveowners and without land redistribution to the freed, leaving the post-emancipation labor regime open for the immigration policy that §9.6 will treat. Drescher’s Abolition (2009) provides the synoptic comparative treatment of the sequence; the asymmetries across the cases are the section’s subject.
The Haitian Revolution of 1791–1804 is the abolition sequence’s most consequential single event and its sharpest fork. The revolution opened in August 1791 with a coordinated uprising in the Northern Plain of Saint-Domingue, the world’s most productive sugar economy, with roughly half a million enslaved people working under some of the trade’s most brutal labor regimes. The Bois Caïman ceremony, a Vodou gathering that has become the revolution’s symbolic point of origin, preceded the rising by days; the rising itself spread across the plantations of the north within weeks. Toussaint L’Ouverture emerged as the dominant military and political leader through the 1790s, building a Black-led army that defeated successively the French planter militias, a Spanish invasion, a British expeditionary force, and finally Napoleon’s 1802 expedition under General Charles Leclerc that arrived to restore slavery after the Brumaire coup. Toussaint was captured by deception and died in a French prison in 1803; his lieutenants Jean-Jacques Dessalines and Henri Christophe completed the war. Dessalines declared Haitian independence on January 1, 1804: the world’s first Black republic and the second independent nation in the Americas after the United States.
The 1825 Haitian indemnity reversed the asymmetry of every other abolition case. France refused to recognize Haitian sovereignty for two decades after independence, and trade embargoes and the threat of military reconquest constrained the new republic’s commercial relations with the Atlantic economy. In 1825, French king Charles X dispatched a naval squadron to Port-au-Prince with an ultimatum: recognition in exchange for an indemnity of 150 million gold francs, payable to compensate French planters for the loss of their property, including the people they had enslaved. Haitian president Jean-Pierre Boyer accepted under duress; the figure was reduced in 1838 to 90 million francs, and Haitian payments were stretched through the 1880s. To meet the schedule, Haiti borrowed from French banks at compound interest. The bonds were traded as a sovereign-debt category in the Paris and New York financial markets through the second half of the nineteenth century. Julia Gaffield’s archival work on Haitian state finance and recent reconstructions by economists at the University of California Davis place the present-value cost of the indemnity, including the compounding service of the loans Haiti took to make the payments, at somewhere between $20 billion and $115 billion in current dollars depending on assumptions. The fiscal cost shaped Haitian state capacity through the entire post-independence century: roughly 80 percent of Haitian government revenue went to indemnity service in some years, leaving residual fiscal space for infrastructure, education, and public health near zero.
The asymmetry the case names: French slaveowners were compensated for losing their slaves; Haitian ex-slaves and their descendants paid the bill for being free. The British case sits in the middle of the same logic: slaveowners compensated, the enslaved emancipated without compensation for the labor they had performed under bondage. The US case sits at the other end: slaveowners not compensated, the freed enslaved people not compensated either, the human-capital wealth wiped out and no transfer made in either direction. Across the abolition sequence, the assets that emancipation reorganized were either preserved on the slaveowner side (British, French, Cuban, Brazilian compensation), reversed against the enslaved (Haitian indemnity), or destroyed entirely (US Civil War). The post-emancipation distribution of wealth across the Atlantic economies was set in this redistributive moment, and the long-shadow consequences are what §9.6 takes up. The classical-era cluster of the intellectual-history timeline carries the abolitionist political-economy lineage that argued the moral case across these decades; the inequality walkthrough takes up the modern racial wealth gap as one of the long-shadow consequences. Abolition’s mechanics shaped the post-emancipation distribution. The labor regimes that replaced slavery came next.
Emancipation ended legal slavery in the Atlantic powers between 1794 and 1888. The labor regimes that followed (sharecropping in the US South, coolie indenture in the Caribbean and Indian Ocean, European immigration replacing Black labor in Brazilian coffee) were institutional substitutes that preserved much of the slavery-era distribution under different legal forms. The long shadow runs through the design of these substitutes, not through abstraction.
Sharecropping in the post-Civil-War US South is the canonical case of the institutional-substitute pattern. Land ownership remained concentrated with the planter class; the freed Black population (and many poor whites) farmed as sharecroppers in exchange for a share of the crop, typically half on cotton land. The credit-merchant trap layered on top of the share contract. Planters or local merchants advanced food, seed, tools, and household supplies to the sharecropper at the start of the season, against a lien on the year’s crop. The interest rates ran 25 to 60 percent. Cotton prices that fell during the season, or harvests that came in low, left the sharecropper deeper in debt at year’s end than at the start; the debt rolled into the next year’s contract. Contract-enforcement asymmetry made exit structurally difficult: vagrancy laws criminalized unemployment and authorized arrest of Black workers without proof of employment; convict leasing channeled the resulting prison labor back into agricultural and industrial use under conditions that approximated slavery in everything but the legal name. The Black Codes of 1865–66, passed by Southern state legislatures immediately after emancipation, codified the new labor coercion explicitly; congressional Reconstruction (1867–1877) blunted them temporarily through federal occupation and Freedmen’s Bureau enforcement; the post-Reconstruction redemption regime restored them in modified form. Jim Crow, consolidated in the 1890s through state constitutional revisions, electoral disenfranchisement, and the Supreme Court’s 1896 Plessy v. Ferguson decision, was the broader political-economic enforcement architecture under which sharecropping’s debt-and-coercion logic operated through the next half-century. Eric Foner’s Reconstruction (1988) is the canonical synthesis; the Berlin and Rowland edition of Families and Freedom (1997) carries the documentary record of the post-emancipation transition.
Coolie indenture supplied the Caribbean and Indian Ocean plantations with labor after the British 1833 emancipation cut off the prior labor source. Indian and Chinese workers signed multi-year contracts—typically five to ten years—under which they traveled to the destination plantation, worked for a fixed wage and provided housing, and earned a return passage at contract’s end. The institution sat halfway between slavery and free wage labor. The contracts were enforced with criminal sanctions for breach; wages were often consumed by deductions for housing, food, and tools that the contract had nominally provided; return passages were honored unevenly. David Northrup’s Indentured Labor in the Age of Imperialism (1995) documents approximately 1.5 million indentured Indian workers transported under British and other European auspices between 1834 and 1917, with destinations including British Guiana, Trinidad, Mauritius, Fiji, Natal, and the Indian Ocean island plantations. Chinese coolie labor moved through Cuban sugar estates (approximately 125,000 workers, 1847–1874) and Peruvian guano islands. The institution’s post-slavery character was systematic: the legal architecture changed, the labor-extraction structure largely did not.
Post-emancipation Brazil is the third case. After the 1888 abolition, the Brazilian state and the coffee planters of São Paulo state pursued a policy of embranquecimento (“whitening”), subsidizing European immigration to replace Black labor in the coffee economy. Italian, Portuguese, Spanish, and German immigrants received subsidized passage and access to coffee-region labor contracts that the freed Black population was systematically excluded from. The policy was explicit in its racial logic: Brazilian elites hoped that European immigration would dilute the Afro-Brazilian population and produce a whitening of the national demographic over generations. The economic consequence was that Afro-Brazilians, freed from slavery, were excluded from the post-emancipation labor market on which the country’s new agricultural-export expansion was built. The racial wealth gap that opened in the post-1888 decades has not closed; modern Afro-Brazilian household-wealth statistics show the long-run distributional consequence of the institutional design choice.
What does the long-shadow empirical literature say? Two pieces of evidence carry the chapter’s case. The first is Daron Acemoglu, Simon Johnson, and James Robinson’s 2002 American Economic Review paper “Reversal of Fortune: Geography and Institutions in the Making of the Modern World Income Distribution.” Within the colonized world, AJR document a pattern they call the reversal of fortune: regions that were rich in 1500, measured by population density and urbanization (indicators of pre-contact state capacity and economic development), are systematically poorer today; regions that were sparse and “settler-friendly” in 1500 are systematically richer. The Aztec Mexico, Mughal India, Ming China, and Yoruba kingdoms of 1500 sit poorly today; the temperate Americas, Australia, and New Zealand sit richly. AJR’s mechanism is institutional: in densely-populated regions, European colonizers implanted extractive institutions designed to channel surplus from indigenous populations into colonial export; in sparsely-populated regions, colonizers implanted inclusive institutions because they were the institutions colonizers themselves needed for their own production. The Atlantic plantation zones are paradigm cases on the extractive side. AJR’s settler-mortality instrument, the use of European mortality rates in colonial postings as an instrument for the type of institution implanted, has been contested by David Albouy (2012) on data-coding choices and by Edward Glaeser, Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei Shleifer (2004) on whether human capital rather than institutions is what the instrument identifies. The chapter cites the contestations alongside the original finding because the empirical claim is one whose interpretation depends on which empirical critique a reader weighs. The substantive pattern, the cross-sectional correlation of 1500 prosperity with present-day under-development across the colonized world, is robust to the instrument-validity debate; what the instrument-validity debate constrains is the causal-attribution claim about institutional channels specifically.
The second piece of evidence is Nathan Nunn’s 2008 Quarterly Journal of Economics paper “The Long-Term Effects of Africa’s Slave Trades.” Nunn assembles slave-trade-intensity data by African ethnic-group origin and shows that African regions from which more captives were extracted across the 1400–1900 period are systematically poorer today, with proposed mechanisms running through pre-colonial state-capacity damage and post-colonial trust deficits. The Nunn paper is the African counterpart to AJR’s Latin American and Asian results; together they describe a pattern in which the regions most directly implicated in the Atlantic system as labor sources or as plantation hosts carry measurable economic consequences into the present. The modern racial wealth gap in the United States, which the inequality walkthrough treats in detail, is one face of the same long-shadow process: the post-emancipation labor regime that sharecropping and Jim Crow designed compounded over generations into a present-day Black-white household-wealth ratio of approximately one to ten. The institutional persistence the rich-and-poor-countries walkthrough takes up at the cross-country level operates within the United States as the racial dimension of the elephant curve that ch. 18 takes up. The formal institutional-economics frame that the long-shadow argument operates in lives in economics ch. 18; the modern-end GDP-per-capita levels of the AJR reversal countries (Brazil, India, Mexico, the United States, and the rest) are visible on the GDP map.
The chapter’s argument is now ready: what the Atlantic system made was not just a forced-labor coordination mechanism for sugar and cotton but a wealth distribution whose post-emancipation institutional forms generated effects measurable into the present.
The Atlantic slavery system was the period’s largest forced-labor coordination mechanism, a high-output economic technology measurable in TFP terms, a load-bearing input to industrial capital, an economic event of measurable mechanics when abolished, and a system whose post-abolition labor regimes generated long-run distributional effects measurable today. The chapter’s argument is that the four-position debate over slavery’s role has typically been framed as a single binary; the right answer is found at a different decomposition of the question. The chapter takes the call here.
On the productivity-as-output question: Fogel and Engerman’s finding survives. The Southern plantation system tested as approximately 35 percent more productive than Northern free farms by total-factor-productivity measures, and the long methodological backlash conducted by Gutman, David, Sutch, Wright and others modified specific data-construction choices without overturning the core empirical claim. The plantation complex was a high-output coordination mechanism that produced sugar, tobacco, rice, indigo, and cotton at scale and at unit costs no free-labor agricultural system in the same period matched.
On the slavery-and-industrial-capital macro question: Williams and the new history (Beckert and Baptist) are closer to right than Engerman. Engerman’s magnitudes calculation, 5 percent of British total investment, is correct on its own terms but answers a question different from the one Williams was asking. The constitutive integration of cotton supply chains, credit instruments, and managerial practices between the slave South and industrial Lancashire ran deeper than financing-flow magnitudes capture. Lancashire textile mills depended on Southern cotton for roughly three-quarters of their raw material by 1860; the credit instruments that financed Southern planters ran through Northern US merchant houses and London commission agents into the same capital pool that financed British industry; the plantation account books and the textile mill account books document a continuous evolution of managerial practice, not parallel emergence. Slavery was not just a contributor to industrial capitalism; it was constitutive of its specific form.
On the specific cotton-productivity mechanism: Olmstead and Rhode are closer to right than Baptist. The cotton-yield-per-acre quadrupling between 1800 and 1860 tracks selective seed breeding (Petit Gulf and successor varieties) more closely than it tracks measures of coercion intensity. The biological-technical mechanism dominates the productivity record, and Baptist’s “whipping-machine” mechanism does not survive the productivity-historian critique on the specific question of what drove the cotton yield curve. The chapter is closer to Beckert and Baptist on the macro question and closer to Olmstead and Rhode on the specific cotton mechanism, at different decompositions of what looks superficially like a single argument. The decomposition is not a hedge; it is the answer the evidence supports.
On the moral question: the productivity reading does not dilute the moral claim, and the moral claim does not negate the productivity reading. The plantation complex was a high-output economic technology; it was also an institutionalized system of human bondage that filed 132 people thrown off the Zong under the same legal heading as damaged sugar. The two claims operate at different categorical levels. A high-output coordination mechanism for sugar and cotton can be both economically efficient by the metric the historian uses to measure it and morally indictable by the standard the moral philosopher applies. Treating the productivity claim as if it pre-empts the moral claim, or treating the moral claim as if it disqualifies the productivity claim, conflates categories that have different evidentiary standards and different decision relevance. The chapter’s position is to hold both at full strength.
The chapter’s house line, four-decomposed: the productivity numbers are real; the macro-constitutive role of slavery in industrial capitalism is closer to the new history than to Engerman; the specific cotton mechanism is selective seed breeding rather than coercion intensification; and the productivity claim and the moral claim operate at different categorical levels and do not dilute each other. The position is testable against §9.1–§9.6 evidence at every step. Where a future round of empirical work reverses one of the four pieces, the position revises on that piece without collapsing the rest. The decomposition is what makes the position robust.
Compensation went to the people who lost their slaves; the bill came to the people who had been enslaved and their descendants. The trade itself was not the chapter’s only legacy. The abolition’s mechanics shaped the post-1865 distribution as much as the trade shaped the pre-1865 distribution. The next chapter takes up the new extractive system that replaced the Atlantic slavery economy after the trade’s legal end: the late-nineteenth-century scramble for Africa, the formal colonial systems built across most of the Asian and African world by 1914, and the relationship between European industrial-finance capitalism and the colonial extraction that financed it. The modern faces of this chapter’s question (why some countries are rich and others poor; why some people are poor in rich countries) live in BQ02 and BQ09. The formal institutional-economics machinery that the long-shadow argument operates in lives in economics ch. 18. The within-country inequality with its US racial dimension that traces back through this chapter is the subject of ch. 18 of this book.