Abstract
In 1711, British Parliament chartered the South Sea Company, a public–private corporation chartered to reduce the cost of government borrowing by swapping illiquid short-term government debt for tradeable shares of the South Sea Co. To attract subscribers, the government also awarded the South Sea Co. an international monopoly in the trade of African slaves to Spanish America—the Asiento de Negros. This paper considers the extent to which Asiento-related slave trading was profitable for South Sea Co. shareholders and beneficial to the British financial revolution between 1713 and 1743. First, we use historical financial data to estimate the parameters of a capital asset pricing model of excess returns for South Sea Co. shareholders. We find that the Asiento contract increased risk-adjusted excess returns on South Sea Co. stock between 18 and 24% per year. Second, we estimate profit margins in the South Sea Co. Asiento slave trade. These show a stark positive correlation with company share prices before and after the South Sea Bubble of 1720. Adding slave ship departures to the CAPM specifications confirms the direct contribution of slave trading to shareholder returns. We also find that the Asiento and Asiento-related slave trading increased central government fiscal surplus by 16%. This suggests that profitable slave trading by the South Sea Co. under the Asiento enhanced Great Britian’s fiscal capacity, which could be utilized to enhance a military capacity necessary for securing an empire.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
While much of the literature on the importance of the slave trade focuses on the profitability of slavery and slave trading, an alternative mechanism has been considered by Forestier (2005). To the extent that slavery and slave trading introduced significant principal–agent problems that one observes in modern market societies, exposure to slavery and slave trading by European nations could have created networks of mutual trust in which trading ideologies could emerge, which is complementary to the development of a materially prosperous market society.
Acemoglu et al. (2001) argue that the winner(s) of this race and the structures of their colonial institutions have had long-term and lasting effects on patterns of economic growth and inequality around the world.
We are only aware of a few attempts to do this. Inikori (2002) documents the relationship between the British slave trade and the development of British financial institutions, but does not consider the South Sea Co. or the British financial revolution. Sheridan (1958) looks at the impact of long-distance slave trading on increased use of bills of exchange and other commercial paper, but this study also does not consider the South Sea Co. or the British financial revolution.
See Du Bois (1896) for an overview of the Asiento.
The scheme to restore confidence began to take shape in August of 1710, following a Tory political victory in Parliament. Robert Harley was appointed Chancellor of the Exchequer and sought to centralize management of Parliament’s debt. In January of 1711, the House of Commons agreed to appoint a committee to investigate. The debt until then had been a hodgepodge of departmental budgets. The House Committee concluded that Parliament had a debt exposure of £9 million with no allocated income to pay it off. Harley then proposed that the exposed debt be converted into equity shares of a newly formed private monopoly called the South Sea Company. On September 11, 1711, government awarded the Asiento to the South Sea Company.
Spain was prohibited from trading in Africa. In 1494, Pope Alexander VI negotiated the Treaty of Tordesillas between Spain and Portugal. The treaty divided the newly discovered lands outside of Europe along a line 370 leagues west of the Cape Verde islands off the West coast of Africa. Lands east of the line went to Portugal. Lands west of the line went to Spain. Since Spain did not have a presence on the coast of Africa, the Spanish government, in 1518, introduced the Asiento, a license to deliver African slave to the Spanish Empire. The license was made available to merchants and nations for a fee paid to the King or Queen. It was held by France prior to being held by Britain.
Values come from the data annex to Hills et al (2010). The measure for discounted debt = (market value of the debt/nominal value of the debt). Nominal values are taken from the series in Mitchell (1971, p. 401) and are the nominal value of the unredeemed capital of the public debt at the end of the year. The market value comes from the estimates in Janssen et al (2002) which estimates the market value of central government debt at the end of the financial year.
The long-term government bond yield is the yield implied by Janssen (2002).
In his classic Memoirs of Extraordinary Popular Delusions and the Madness of Crowds, McKay (1869) quotes Robert Walpole, a staunch opponent of the scheme, as saying, “The great principle of the project was an evil of first rate magnitude; it was to raise artificially the value of the stock, by exciting and keeping up a general infatuation and by promising dividends out of funds which could never be adequate to the purpose." In a prophetic spirit he added, that if the plan succeeded, the directors would become masters of the government, form a new and absolute aristocracy in the kingdom and control the resolutions of the legislature. If it failed, which he was convinced it would, the result would bring “general discontent and ruin upon the country (page 50).” Temin and Voth (2013) call the 1720 South Sea scheme a “classic Ponzi scheme (p. 100).”
See Paul (2011) for a nuanced view that argues that the increase in share prices was a rational response to the set of investment opportunities afforded by the South Sea Co., rather than a kind of gambling mania that gripped early eighteenth-century Europe.
Stasavage (2003) argues that the formation and ascendancy of the Whig Party, with its changing political coalitions, improved government access to capital at least as much as the changes in the British constitution following the Glorious Revolution of 1688.
Neal (1990, pp. 111–112) argues that the first appreciation of South Sea Co. stock during the first subscription reflects the public’s willingness to pay for liquidity. If a person held the stock from January 1, 1720, to December 31, 1720, they would have realized a return of 56 percent.
Carlos, Fletcher, Neal and Wandschneider (2013) document the wider holdings of South Sea Co. stock among small investors, compared to the holdings of the stocks of the Bank of England and the East Indian Company.
The Whig interpretation emphasizes the lure of South Sea gold and silver and the competitiveness of British manufactures, if only British merchants could gain access to the South Sea market (Mackay 1869; pp. 46–53).
According to Frey and Frey (1995), “Although Robert Harley, earl of Oxford, negotiated secretly with the French from late 1710 to April 1711, it was with the idea that the allies would be included in the final agreement for a peace conference. After June 1711, however, he dropped the intent of including them. He was bent on creating a South Sea Company to fund unsecured government debt. For this he needed monopolistic concessions in the Spanish colonies. When Mesnager (the French negotiator) met with Oxford and Charles Tolbert, Duke of Shrewsbury, on 15 August 1711, he was told that the territorial concessions were necessary for peace. Five days later, however, Oxford dropped his demand for the four towns and told Mesnager that Britain would accept in their place an extension of the Asiento treaty from 10 to 30 years. Mesnager worked through September on the other points in Oxford’s demands (p. 285).”
The pieza de Indias was the unit of measure used in the Asiento slave trade after 1663. A slave was considered a full pieza (“piece”) if he or she was at least seven palmos (about five feet) tall and between the ages of 15 and 35. Children and adults with “defects” counted as less than one pieza.
Other evidence of monopoly practices is the branding of South Sea Co. slaves out of Barbados and the constant threat posed by free traders trying to enter the South Seas slave trade (Palmer 1981; chapter 5).
“To determine with finality whether the company was conducting its Negro sales at a profit or loss would require a much more extended study than is here possible. From cursory examination of scattered figures it seems obvious that the business was a losing one and that the question for examination is the size of the loss” (Donnan 1930, p. 447).
“… when company accountants in London began their final audit of factory books at the end of each five year period so as to submit a complete record of the Asiento trade they had insufficient information to determine the amount of duties owed from the introduction of slaves or the profit due to Philip V from the merchandise introduced on the license and annual ships… The company and Spanish officials agreed that for purposes of bookkeeping the sale price of each slave introduce would be entered as 200 pesos, a figure that probably favored the company (Sorsby 1975, p. 258).”
This follows from assuming that wealth maximizers are risk averse and prefer to receive a fixed payment \(\tau\) to a random payment of wealth \(W\) = \(\sum\)\(w_{i}\)\(\mu_{i}\), where \(0 \le w_{i} \le 1\) is the weight of asset \(i\) in the wealth portfolio and \(\mu_{i}\) is the expected return on asset \(i\). If an individual is indifferent between \(E\) [\(U\) (W)] and \(U\) [\(E\) (W) − \(\tau\) ], then these two payments must be equal or \(E\) [U(W)] = \(E\) (U(E[W] − \(\tau\)) = U(\(E\) [W] − \(\tau\)), where \(E\) is the expected value operator and \(U( \cdot )\) is a utility function. Let \(z\) = − \(U^{''}\) (\(E\) (W))/ \(U^{'}\) (\(E\) (W), where \(U^{n}\) (\(E\) [W]) is the \(n\) th derivative of \(U( \cdot )\) with respect to its argument evaluated at \(E\) (W), a first-order Taylor expansion—neglecting higher-order terms—on both sides of the indifference relationship with respect to \(W\) allows representing utility as \(U\) = \(U^{i}\) (\(E\) [U(W)]) = \(\mu_{p}\) − ½ (\(z\sigma_{p}^{2}\)), where \(\mu_{p}\) = \(\sum w_{i} \mu_{i}\) and \(\sum w_{i} = 1\). The maximization of \(U^{i}\)\(( \cdot )\) subject to \(\sum w_{i} = 1\) generates a solution for the relative return on equity \(i\) as \(\mu_{i} - r = \beta_{i}\) (\(\mu_{p} - r\)), where \(\beta_{i} = \frac{{\sigma_{ip} }}{{\sigma_{p}^{2} }}\).
ESFD data are available at www.esfdb.org/Database.aspx.
The data are available at www.bankofengland.co.uk/publications/Pages/other/monetary/mpreadinglistf.aspx.
Available at https://www.slavevoyages.org/voyage/database.
While earnings-based models are an alternative to rationalizing stock prices, Foerster and Sapp (2005) find that the dividend discount model performs better at explaining stock prices.
This 3-stock index appears to be a good approximation of the market portfolio for eighteenth-century investors, as Mirowski (1981) found that for an index of 8 stocks (which includes those of the South Sea Co., East India Co. and the Bank of England) the share prices are all highly correlated except for the period of the South Sea Bubble.
All parameters were estimated with STATA 15.0.
The Lagrangian multiplier test for heteroskedasticity is based on a specification of the variance of the error term as \(h_{t} = \gamma_{0} + \gamma_{1} \varepsilon_{t - 1}^{2}\) and testing \(H_{o}\): \(\gamma_{1}\) = 0 . The Jarque–Bera test statistic (Jarque and Bera 1987) is JB = N/6[S2 + ¼(K − 3)2], where N is sample size, S is skewness and K is kurtosis. For the null hypothesis that the residuals are normal distributed, the test statistic is asymptotically distributed as a Chi-squared distribution with two degrees of freedom.
The Ljung–Box test (Ljung and Box 1978) statistic is Q = N(N + 2)Σh(ρk/(N − k) for h lags with sample size N and k parameters. For the null hypothesis of zero autocorrelation, the test statistic is asymptotically distributed as a Chi-squared distribution with h degrees of freedom.
The 83-year period under consideration would result in 1008 monthly observations in the absence of missing data, instead of the 995 observations upon which the parameter estimates are based on.
Suppose for a specification \(Y^{*}\) = \(X\beta_{1}\) + \(\varepsilon_{1}\), and the rule determining whether we observe \(Y^{*}\) is given by Y = \(Y^{*}\) for \(H^{*} > 0\) where \(H^{*}\) = \(Z\beta_{2}\) + \(\varepsilon_{2}\). It follows that for H = 1 (actually observing Y):
\(E(Y|X,H = 1) = X\beta_{1} + E(\varepsilon_{1} |\varepsilon_{2} > - Z\beta_{2} )\)
Estimating \(E\) (\(Y|X\)) without accounting for the sample selection—the process determining whether observations on the dependent variable are non-missing—introduces an omitted variable bias and an additional source of heteroskedasticity in the standard error of the parameter estimates. A threshold specification accounts for the sample selection by allowing for H = 1 when \(d_{t - 1}\) = 1 for \(\varepsilon_{t - 1} < 0\) and zero otherwise to enable standard errors robust to the heteroskedasticity caused by possible omitted variable bias that results from failing to account for how Y is selected into the sample.
For the CAPM specification of excess returns:
\(\mu_{t} - r_{t} = \beta_{o} + \beta_{1} A_{t} + \beta_{2} (\mu_{pt} - r_{t} ) + \beta_{3} [A_{t} \times (\mu_{pt} - r_{t} )] + \varepsilon_{t}\)
we estimate as a marginal effect:
\(\partial (\mu_{t} - r_{t} )/\partial A_{t} = \beta_{1} + \beta_{3} [(\mu_{pt} - r_{t} )|A_{t} ]\)
In each instance, as the marginal effects are measure at the monthly level, we express the marginal effects as annualized monthly returns according to [(1 + R)12 − 1], where R = the monthly estimate of \(\mu_{t}^{s}\) − \(r_{t}\).
For the ARCHT/GARCHT specifications in Table 2, the minimum AIC GARCHT parameter estimates suggest that the Asiento increased the risk-adjusted excess return on South Sea Co, stock by approximately .017 monthly, which translates into an approximate 22% increase in the annual excess return. If a South Sea stock holder held shares for a year, this translates into an average twenty-two nominal pounds of capital gains per share, based on the average share price of approximately 101 nominal pounds over the period. This suggests that for South Sea Company shareholders, including the British central government, slave trading or at least the right to trade slaves under the Asiento, was profitable.
For a variety of reasons, the data on these ventures are scattered and piecemeal. Conflicts between Britain and Spain often prolonged ship voyages for years; ships and their merchandise were often confiscated in times of war; reparations were repaid in installments that stretched over many years; and disputes over contraband and duties were constant irritants that disrupted the legal trade.
As chronicled by Nelson (1945), the informants were Dr. John Burnet who had been a factor at Porto Bello and Cartagena; and Matthew Plowes who was secretary and principal accountant for the Company. They produced 42 documents which included detailed financial statements. Included in the documents were names of Spanish officials who had accepted bribes and allowed the import of contraband goods. The informant also affirmed that the chief Spanish representative to the company in London had also received bribes, one being £1000 and an annual pension of £800 in return for false measurements of permission ships and other frauds on behalf of the company.
Having secured solid evidence of non-compliance, Philip V of Spain was able to place an agent on the company’s board of directors, Sir Thomas Geraldino, the Spanish ambassador to Britain (Nelson 1945, p. 58). In 1730, the company responded by placing sub-governors and deputy governors in charge of all matters of importance. Only routine matters were placed before the court in the presence of the Spanish agent. Much of the evidence on contraband trade after 1730 is contained in the private correspondences of these sub- and deputy directors and is contained in the Shelburne Collection. They reveal rampant bribery and falsification of documents (Nelson 1945, p. 58; Palmer 1981, p. 27, 72; Reibman 2012).
In 1729, the Treaty of Seville established a commission to settle disputes between England and Spain. The first meeting was in 1732, but it disbanded without solving the competing claims. England claimed it was due compensation for confiscated ships and merchandise. Spain claimed compensation for the duty on slaves that it never received. Geraldino was the chief negotiator for Spain. A plan was devised that would have Spain pay England £140,000 and the South Sea Company pay the King of Spain £68,000. Neither side trusted the other and demanded that the other pay before they pay. On January 10, 1739, the King of Spain reserved to himself the right to suspend the Asiento in case the demands of Spain were not met. On June 11, at the time payment was overdue, the British government ordered its merchants to withdraw their ships and effects from Spain immediately. War was declared in October 1739, de facto ending the Asiento contract for the South Sea Company. It was company factors in the Spanish ports who were responsible for the clamor against Fandino, a guarda costa suspected of cutting off Jenkins’ ear (Hildner, 1938, p. 324).
This is consistent with the negative and significant coefficient on the contraband dummy reported in Tables 5, 6, 7 and 11. If a ship had a history of engaging in an illegal contraband trade that did not accrue to stockholders, then a stock investor would discount their expected return from the venture.
Of the 390 ships into Spanish America for which Palmer could identify port of origin, 59.2% came from Jamaica and 8.5% from Barbados.
To illustrate how the totals were allocated to the four branches, consider the allocations for 1714 and 1715. In 1714, transatlantic deliveries into Spanish America (0) were less than the total number of slaves delivered into Spanish America (324). The difference was made up from British Caribbean slave populations. Since transatlantic deliveries into the British Caribbean (1,265) exceeded the 324 British Caribbean deliveries into Spanish American, the 324 slaves coming from the British Caribbean are counted as transatlantic slaves who were “refreshed” in the Caribbean before being transferred to Spanish America. The remaining 941 transatlantic deliveries into the British Caribbean are counted as being sold there. In 1715, transatlantic deliveries into Spanish America (2297) exceed the transatlantic deliveries into Spanish America (1170), the difference (1127) being made up from British Caribbean supplies. In total, 246 of them are counted as coming from transatlantic deliveries that were “refreshed” in the British Caribbean, the remaining 881 being purchased from the existing slave population in the British Caribbean. Allocations for all years are arrived at in a similar manner.
The same relationship is seen between slave deliveries and share prices.
The War of Jenkins Ear (1739–1741) and the War of Austrian Succession (1741–1748) effectively put an end to the Asiento trade in October of 1739.
Five (5) South Sea Company slave ships found in the Transatlantic Slave Trade Database did not have information on month of departure and were excluded from the analysis.
Stasavage (2003, pp. 77–78) notes that “… for a brief period between 1710 and 1713 the British government actually found itself paying interest rates that were higher than those that had prevailed before the Glorious Revolution.” The company then receives the Asiento contract, commences slave trading and stock values begin to rise. Dincecco (2011, p. 67) consistently estimates the period 1711–1716 as a structural break in the government deficit ratios for Britain.
The -271 Spanish American delivery from the Caribbean in 1718 is a measurement error (more deliveries from Africa than total deliveries). I set this equal to zero. 1718–1721 had very few deliveries from the Caribbean anyway. The transatlantic ships are selling primarily in the Caribbean in 1718 and 1719. This is indirect evidence that the company was not just in the business of selling contraband to Spanish America. They carried a brisk slave trade in to the Caribbean in 1719.
There is evidence that the company underreported slave prices in this report because the report was being prepared for negotiations with the king of Spain who was due some of the profits. At the company meeting in 1734, shareholders complained that the company was operating at a loss, with 1.7 million in expenses and only 35,000 profit. The company pushed back on this complaint and continued to engage in the slave trade.
Palmer (1981) over-estimates Jamaica slave prices. He sets the Jamaica price at 107 pesos, or £24, for 1714–1721 (p. 162). He does not explain why, other than it is based on his “analysis of the selling prices of slaves in Jamaica and Barbados (p. 162).” The Eltis and Richardson (2004) data record an average Jamaican price of £16.4 for this period. Palmer’s is an over-estimation of 46% on the purchase price of the slaves purchased in Jamaica and sold in South America.
Palmer also over-estimates African slave prices. He sets African slave price at £10 lb (p. 32, 163). These are the value of English cargo divided by the number of slave the company expected the cargo to purchases on the African coast. The sample is 7 ships in 1723–1726. The Anglo-African Trade Statistics and the Transatlantic Slave Trade Database estimate an average price of £7.01 for 1723–1726. Palmer over-estimates by 43%. In his discussion, he basically acknowledges his mistake: “The average cost of a slave in West Africa during the 1730 s was about £6 (27 pesos), calculated on the basis of the prime cost of the commodities exchanged for the slave… (p. 163).” This is almost identical to the £5.9 price estimated by the Anglo-African/TSTD ratio for the 1730 s, but Palmer instead uses the £10 figure in his profit calculations (Appendix Tables 4-10, pp. 164–168). That is an over-estimate of 43% for 1723–1726 and 69% for the 1730s.
References
Acemoglu D, Johnson S, Robinson J (2001) The colonial origins of comparative development: an empirical investigation: reply. Am Econ Rev 102(6):3077–3110
Acemoglu D, Johnson S, Robinson J (2005) The rise of Europe: Atlantic trade, institutional change, and economic growth. Am Econ Rev 95(3):34
Angeles L (2013) On the causes of the African slave trade. Kyklos 66:1–26
Banz RW (1981) The relationship between return and market value of common stocks. J Financ Econ 9:3–18
Blackburn R (1997) The making of new world slavery: from the baroque to the modern, 1492–1800. Verso, London
Bollerslev T (1986) Generalized autoregressive conditional heteroscedasticity. J Econom 31:307–327
Brewer J (1988) The sinews of power: war, money and English state, 1688–1783. Harvard University Press, Cambridge
Brown VL (1926) The South Sea Company and contraband trade. Am Hist Rev 31:662–678
Cable J, Holland K (1999) Regression vs. non-regression models of normal returns: implications for event studies. Econ Lett 64:81–85
Carlos AM, Fletcher EK, Neal L, Wandschneider K (2013) Financing and refinancing the war of spanish succession, then refinancing the South Sea Company. In: Coffman DM, Leonard A, Neal L (eds) Questioning credible commitment: perspectives on the rise of financial capitalism. Cambridge University Press, New York
Carlos AM, Fletcher EK, Neal L (2015) Share portfolios in the early years of financial capitalism: London, 1690–1730. Econ Hist Rev 68:574–599
Carswell J (1961) The South Sea Bubble. Stanford University Press, Palo Alto
Clare AD, Priestley R (1998) Risk factors in the Malaysian stock market. Pac Basin Finance J 6:103–114
Coffman D, Leonard A, Neal L (eds) (2013) Questioning credible commitment: perspectives on the rise of financial capitalism. Cambridge University Press, Cambridge
Darity WA (1992) A model of original sin: rise of the west and lag of the rest. Am Econ Rev 82:162–167
Daudin G (2004) Profitability of slave and long-distance trading in context: the case of 18th century france. J Econ Hist 64:144–171
Dickson P (2017) The financial revolution in England: a study in the development of public credit, 1688–1756. Rutledge, New York
Dincecco M (2011) Political transformations and public finances: Europe, 1650–1913. Cambridge University Press, Cambridge
Dincecco M, Prado M (2012) Warfare, fiscal capacity, and performance. J Econ Growth 17:171–201
Donnan E (1930) The early days of the South Sea Company, 1711–1718. J Econ Bus Hist 2:419–450
Drelichman M, Voth H-J (2008) Debt sustainability in historical perspective: the role of fiscal repression. J Eur Econ Assoc 6:657–667
Du Bois WEB (1896) The suppression of the Atlantic slave trade to the United States of America, 1638–1870. Longmans, Green and Company, New York
Egrioglu E, Aladag CH, Yolcu U, Uslu VR, Basaran MA (2010) Finding an optimal interval length in high order fuzzy time series. Expert Syst Appl 37:5052–5055
Eltis D, Engerman SL (2000) The importance of slavery and the slave trade to industrializing Britain. J Econ Hist 60(1):123–144
Eltis D, Richardson D (2004) Prices of African slaves newly arrived in the Americas, 1673–1865: new evidence on long-run trends and regional differentials. In: Eltis D, Lewis FD, Sockoloff KL (eds) Slavery in the development of the Americas. Cambridge University Press, Cambridge, pp 181–218
Eltis D, Lewis FD, McIntyre K (2010) Accounting for the traffic in Africans: transport costs on slaving voyages. J Econ Hist 70(4):940–963
Elton EJ, Gruber MJ (1997) Modern portfolio theory, 1950 to date. J Bank Finance 21:1743–1759
Engle RF (1982) Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation. Econometrica 50:987–1007
Engle RF (2001) Garch 101: the use of Arch/Garch models in applied econometrics. J Econ Perspect 15:157–168
Fama EF, French KR (2004) The capital asset pricing model: theory and evidence. J Econ Perspect 3:25–46
Foerster SR, Sapp S (2005) The dividend discount model in the long-run: a clinical study. J Appl Finance 15:55–75
Forestier A (2005) Principle—agent problems in the French slave trade: the case of Rochelais Armateurs and their agents, 1763–1792. In: Working paper. D. O. E. H. London School of Economics and Political Science, London
Frey L, Frey M (1995) The treaties of the war of spanish succession: an historical and critical dictionary. Greenwood Press, Westport
Goetzmann WN, Ukhov AD (2006) British investment overseas 1870–1913: a modern portfolio theory approach. Rev Finance 10:261–300
Hildner EG Jr (1938) The role of the South Sea Company in the diplomacy leading to the war of Jenkins’ ear, 1729–1739. Hispan Am Hist Rev 18(3):322–341
Hills S, Thomas R, Dimsdale N (2010) The UK recession in context: What do three centuries of data tell us? Bank of England Q Bull (Q4)
Inikori J (1981) Market structure and the profits of the British African trade in the late 18th century. J Econ Hist 41:745–776
Inikori J (2002) Africans and the industrial revolution in England: a study in international trade and economic development. Cambridge University Press, Cambridge
Janssen N, Nolan C, Thomas R (2002) Money, debt and prices in the United Kingdom, 1705–1996. Economica 69:461–479
Jensen MC (1968) The performance of mutual funds in the period 1945–1964. J Finance 23:389–416
Johnson M, Thomas Lindblad J, Ross R (1990) Anglo-African trade in the eighteenth century: English statistics on African trade, 1699-1808. Centre for the History of European Expansion, Leiden
Koch-Medina P, Wenzelburger J (2018) Equilibria in the CAPM with non-tradeable endowment. J Math Econ 75:93–107
Lehnert T (2003) Explaining smiles: GARCH option pricing with conditional leptokurtosis and skewness. J Deriv 10:27–39
Linter J (1965) Security prices, risk and maximal gains from diversification. J Finance 20:587–615
Lo AW (2008) Where do alphas come from: a measure of the value of active management. J Invest Manag 6:1–29
Low C, Nayak S (2009) The non-relevance of the elusive holy grail of asset pricing tests: the “true” market portfolio does not alter CAPM validity conclusions. Q Rev Econ Finance 4:1460–1475
Mackay C (1869) Extraordinary popular delusions and the madness of crowds. Routledge and Sons, London
McCloskey DN (2010) Bourgeois dignity: why economics can’t explain the modern world. University of Chicago Press, Chicago
McWatters CS (2008) Investment returns and La Traite Negriere: evidence from 18th century France. Account Bus Financ Hist 18:161–185
Mirowski P (1981) The rise (and retreat) of a market: English joint stock shares in the 18th century. J Econ Hist 41:559–577
Mitchell BR (1971) Abstract of British historical statistics. Cambridge University Press, Cambridge
Mokyr J (2017) A culture of growth: the origins of the modern economy. Princeton University Press, Princeton
Neal L (1987) The integration and efficiency of the London and Amsterdam stock markets in the 18th century. J Econ Hist 47:967–980
Neal L (1990) The rise of financial capitalism: international capital markets in the age of reason. Cambridge University Press, New York
Nelson GH (1945) Contraband trade under the Asiento, 1730–1739. Am Hist Rev 51:55–67
North D, Weingast BR (1989) Constitutions and commitment: the evolution of institutions governing public finance in 17th century England. J Econ Hist 49:803–832
O’Brien P, Engerman SL (1993) Exports and the growth of the British economy from the glorious revolution to the piece of Amiens. In: Solow B (ed) Slavery and the rise of the American system. Cambridge University Press, Cambridge, pp 177–209
Palmer CA (1981) Human cargoes: the British slave trade to Spanish America, 1700–1739. University of Illinois Press, Chicago
Paul HJ (2011) The South Sea bubble: an economic history of the origins and consequences. Routledge, London
Pynnonen S (2005) On regression based event studies. Acta Washensia 143:327–354
Reibman H (2012) Pvt. Profits and the South Sea Company: illicit trafficking under the Asiento. Mich J Hist 4:49–78
Richardson D (1991) Prices of slaves in West and West Central Africa: towards an annual series, 1698–1807. Bull Econ Res 43(1):21–56
Rodney W (1972) How Europe underdeveloped Africa. Bogle-L’Ouverture Publications, London
Scott WR (1912) The constitution and finance of English, Scottish and Irish joint-stock companies to 1720. Cambridge University Press, London
Sharpe WF (1964) Capital asset prices: a theory of market equilibrium under conditions of risk. J Finance 19:425–442
Sheridan RB (1958) The commercial and financial organization of the British slave trade, 1750-1807. Econ Hist Rev 11(2):249–263
Smith A (1776) The wealth of nations. Methuen and Co., Ltd., London
Solow BL (1985) Caribbean slavery and British growth: the Eric Williams hypothesis. J Dev Econ 17:99–115
Song Q, Chissom BS (1993) Forecasting enrollments with fuzzy time series—part I. Fuzzy Sets Syst 54:1–9
Sorsby VG (1975) British trade with Spanish America under the Asiento, 1713–1740, U. O. London, Doctoral Dissertation. London
Stasavage D (2003) Public debt and the Birth of the democratic state: France and Great Britain, 1688–1789. Cambridge University Press, Cambridge
Temin P, Voth H-J (2013) Prometheus shackled: goldsmith banks and England’s financial revolution after 1700. Oxford University Press, Oxford
Wennerlind C (2011) Casualties of credit: the English financial revolution, 1620–1720. Harvard University Press, Cambridge
Whatley W (2018) The Gun-Slave hypothesis and the 18th century British slave trade. Explor Econ Hist 67(1):80–104
Williams E (1944) Capitalism and slavery. University of North Carolina Press, Chapel Hill
Wright G (2006) Slavery and American economic development. Lousiana State University Press, Baton Rouge
Zakoian J-M (1994) Threshold heteroscedastic models. J Econ Dyn Control 18:931–955
Zhang CY, Jacobsen B (2013) Are monthly seasonals real? a three century perspective. Rev Finance 17:1743–1785
Acknowledgements
For invaluable comments and critique on earlier versions of this paper, the authors thank participants of 17th World Economic History Congress, Kyoto, Japan, August 3–5, 2015, and referees of this Journal. Financial support from the Division of Business Administration and Economics at Morehouse College is also acknowledged and appreciated.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix: Estimating South Sea Company slave profit
Appendix: Estimating South Sea Company slave profit
1.1 The four branches of the South Sea Co. slave trade
Palmer (1981, chapter 6) reports the total number of slaves the South Sea Company delivered to Spanish America. These are reported in the first column of Table 7. The company’s slave trading enterprise consisted of four (4) distinct branches of trade. Some company ships departed London, purchased slaves on the coast of Africa and delivered then to the Americas. These ships are recorded in the Transatlantic Slave Trade Database (TSTD) as ships that list the South Sea Company as first or second owner. The Transatlantic Slave Trade Database captures 119 such ships. Sixty-one voyages carried slaves from Africa directly into Spanish American ports (Q_AFRICA), the vast majority into Rio de la Plate or Buenos Aires.
The remaining 58 transatlantic voyages disembarked their slaves in the British Caribbean, primarily in Jamaica or Barbados. There, the African supplies could be sold or “refreshed” before being forwarded to Spanish America. In addition, the company could purchase new slaves in the Caribbean and ship them to Spanish America. The Caribbean supply network was a large part of the company’s overall operation and was centered on Jamaica. Palmer identifies 451 British ships that delivered slaves to Asiento-designated ports in Spanish America. Of the 390 ships for which he could identify port of origin, 59.2% came from Jamaica and 8.5% from Barbados.
Our estimate of the size of the Caribbean supply is the total number of deliveries into Spanish America (from Palmer) minus the transatlantic deliveries into Spanish America (from TSTD).Footnote 45 The remaining three branches of the South Sea Company’s slave trade can be parsed out from here. If, in year t, the number of transatlantic disembarks into the Caribbean exceeds the number of Caribbean deliveries into Spanish America, then the Caribbean deliveries into Spanish America are transatlantic slaves that were refreshed (Q_REFRESHED) in the Caribbean before being forwarded to Spanish America. Any remaining transatlantic deliveries into the Caribbean were sold in the Caribbean (Q_CARIBBEAN SALES). If, in year t, the total number of transatlantic slaves disembarked in the Caribbean is less than the number of Caribbean deliveries into Spanish America, then the refreshed (Q_REFRESHED) slaves equal the number of transatlantic disembarks in the Caribbean, and the additional slaves delivered from the Caribbean to Spanish America were Caribbean purchases (Q_CARIBBEAN PURCHASES).
1.2 The prices of slave
South American prices (P_SA) Palmer (1981, pp. 123–124) reports average annual sale prices for South Sea Company slaves sold in Asiento-designated ports between 1715 and 1719. These prices are found in the Spanish records. They are average prices for 8893 contracts recorded in 21 different year-port samples. The majority of the samples capture more than 80% of total sales, with none of the capture rates falling below 50% (Palmer 1981, footnote 16, p. 127). Figure 3 presents the average annual prices for the 8893 contracts. They range from £43 in 1715 to £50.3 in 1718. Palmer (p. 125) also notes that “Unfortunately, information for the 1720s and 1730s is scant. Scattered evidence suggests, however, that during those years slave prices steadily increased. It appears that by the mid-1730 choice slaves were sold at prices ranging from 230 pesos (£51 15s.) to 270 pesos (£60 15s.) each.”
African, Jamaican and Spanish American slave prices, 1713–1743
Palmer then reports three estimates of South American slave prices found in South Sea Company corporate records (archived in the Shelburne Manuscripts Collection). The British records report prices for the 1730s. Two of the samples support the Spanish estimates, reporting prices in the mid-£50 range. The first is a 1733 report that includes company price expectations (Shelburne MSS, vol. 43, 144). Based on 19 years of experience, the South Sea Company expected average slave prices in Spanish America to be approximately £54.7 in 1733 (reported in Palmer, Table 29, p. 124). The Shelburne Collection also contains a 1736 report that contains sale prices for 5300 slaves sold in Spanish America (Vol. 44, p. 144; Palmer 1981 Appendix Table 11, pp. 168–169). Prices range from an average of 220 pesos (£49.5) in Cartagena to 250 pesos (£56.3) in Panama, with an average of £54.9. The third report is an outlier. It is a South Sea Company financial statement reporting average expenditures and receipts per slave for 1731–1736 (Shelburne MSS, vol. 43, pp. 428–429, reported in Palmer 1981, Table 37, p. 152). Palmer believes these prices to be too low. They range from an average of 97 pesos (£21.8) in Buenos Aires to 195 pesos (£44.1) in Cartagena.Footnote 46 I use the estimates of 54.7 for 1733 and 54.9 for 1736.
Caribbean prices (P_Caribbean) Caribbean prices are annual average prices for prime age male slaves in Jamaica, taken from Eltis and Richardson (2004). I thank David Eltis for making these data available.Footnote 47
African prices (P_Africa) The average prices that British traders paid for slaves on the coast of Africa are calculated from annual data found in the Anglo-African Trade Statistics (Johnson et al. 1990) and the Transatlantic Slave Trade Database. The Anglo-Africa Trade Statistics record the real value of annual British net exports to Africa used to purchase slaves. The annual net exports are then divided by the annual number of British slave purchases, as recorded in the Transatlantic Slave Trade Database. The result is an estimate of the average annual prices that British slave traders paid for slaves on the coast of Africa. See Whatley (2018) and Richardson (1991) for more detailed discussions.Footnote 48
Price trends The three price series are presented in Fig. 3. African prices are almost complete, except for two years in the early 1700 that have been averaged in. The Jamaica price series has a number of gaps and the Spanish America price series has longer gaps. To deal with this, the profit calculations use the trend prices predicted by linear regression lines through the three price series. The regression lines are also reported in Fig. 3. t = (year − 1713), so the intercept is the estimated 1713 price. African prices are relatively flat. Jamaican prices exhibit an upward trend and Spanish American prices exhibit a steeper upward trend. Since the Spanish series is so incomplete, I use the flatter Jamaica trend to estimate Spanish price increases: Spanish Price (t) = 44.0 + .2851* t. This underestimates Spanish prices (and company profits) for the later years, which would be approximately 10% higher in the 1730 had we used the Spanish slope.
1.3 Delivery costs
Delivery costs are the costs associated with purchasing slaves and delivering them to their destinations. From the Spanish and English records Palmer surmises that “…charges for provisions and wages remained fairly stable throughout the company’s trading years… (p. 163).” Annual delivery costs for each of the four trade flows = (avg. slave purchase price + delivery cost per slave) × (quantities sold).
Delivery Cost from Africa to Spanish America = (P_Africa + £11.9) × (Q_AFRICA) + (P_Africa + £3.72) × .15(Q_AFRICA).
These delivery costs are calculated from Palmer (1981, p. 163). For 1734 and 1736, the South Sea Company estimated 33 pesos /4.44 = £7.43 for delivery costs between Africa and Spanish America, and 20 more pesos/4.4 = £4.5 while at the Spanish American factory (commissions, duties, etc.). Total cost is £11.9 per slave sold. Palmer assumes a mortality rate of 15% who, on average, died halfway through the transatlantic voyage. On average, the company paid 50% of the £7.43 freight charges for each deceased slave = £3.72.
Delivery Cost from Africa to Spanish America via Caribbean “refreshment” = (Delivery Cost from Africa to Spanish America + £2.25) × (Q_REFRESHED).
The £2.25 of additional expense includes some additional maintenance and medical charges, but no additional freight charges. “…10 pesos, which would cover clothing, medical care, supervisors’ wages, and perhaps a commission for the agent (Palmer, p. 164).” 10/4.44 = £2.25.
Delivery Cost from Africa to Caribbean = (P_Africa + £7.31) × (Q_CARIBBEAN SALES) + (P_Africa + £3.66) × .15(Q_CARIBBEAN SALES).
Eltis et al. (2010, pp. 948–949, Table 3) estimate £5.28 for transport cost per slave for 22 Royal African Company (RAC) voyages between Africa and Jamaica/Barbados in the mid-1680s. The figure includes payments per slave to ship owners (£3.71), agents’ commissions (£.76), captain’s commissions (£.43) and slave provisions (£.38). They also report theoretical trends in transport cost per slave for the long-term (1675–1807), taking into account the impact of mortality and morbidity and assuming “normal” competitive profits. These theoretical transport costs increase from approximately £6.5 per slave in 1675–1700 to approximately £9.0 in 1700–1735, with no increase in trend over the later period (pp. 958–961). I add the 38% increase to the £5.28 figure from in the RAC records, generating the estimated 1.38 × £5.28 = £7.31 for transport cost between Africa and Jamaica/Barbados for the period 1713–1740. This branch of South Sea Co. trade represented a small fraction of total company trade. It applies to net slave sales in the Caribbean in 1714, 1718, 1719, 1726 and 1740.
Delivery Cost from Caribbean to Spanish America = (P_Caribbean + £7.88) × (Q_CARIBBEAN PURCHASES).
See Palmer (1981, Appendix Table 1, p. 162). The £7.88 transport cost figure is for 1714–1721 and includes average charges for maintenance, freight, medical, agents’ commissions, duty, etc. No allowance is made for mortality between the Caribbean and Spanish America ports because it was “statistically negligible (p. 163).”
1.4 Slave trade profits
The four profit equations are listed below. The resulting annual profit totals are reported in Table 12 and depicted in Fig. 2.
Deliveries from Africa: (P_SA × Q_AFRICA) − (P_Africa + £11.9) × (Q_AFRICA) − (P_Africa + £3.72) × .15(Q_AFRICA)].
Refreshed deliveries: (P_SA × Q_REFRESHED) − (P_Africa + £11.9) × (Q_REFRESHED) − (P_Africa + £3.72) × .15(Q_REFRESHED) − (£2.25 × Q_REFRESHED).
Sales in Caribbean: (P_Caribbean × Q_CARIBBEAN SALES) − (P_Africa + £7.31) × (Q_CARIBBEAN SALES) − (P_Africa + £3.66) × .15(Q_CARIBBEAN SALES).
Purchases in Caribbean: (P_SA × Q_CARIBBEAN PURCHASES] − (P_Caribbean + £7.88) × Q_CARIBBEAN PURCHASES.
Rights and permissions
About this article
Cite this article
Price, G., Whatley, W. Did profitable slave trading enable the expansion of empire?: The Asiento de Negros, the South Sea Company and the financial revolution in Great Britain. Cliometrica 15, 675–718 (2021). https://doi.org/10.1007/s11698-020-00219-w
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11698-020-00219-w