Imperial Measurement: A Cost–Benefit Analysis of Western Colonialism


  • In recent years, we have seen a renewed interest in Britain’s imperialist past: the British Empire, the slave trade and the Caribbean slave labour plantations. More precisely, we have seen a revival of the idea that the wealth of the Western world – and Britain’s in particular – was originally built on slavery and colonial exploitation.

  • There is a lot to be said for a ‘warts-and-all’ approach to history, which does not gloss over or relativise the darker chapters of a country’s past. But the problem with the above narrative is that it is bad economics. While imperialism was undoubtedly extremely lucrative for some people, it is not at all clear whether Britain as a whole benefited economically from it. If such overall gains existed at all, they must have been very modest, and it is quite possible that the empire was a net lossmaker for Britain.

  • Before modern container shipping, transport logistics, telecommunication technologies, etc., made high volumes of trade possible, trade and overseas investment accounted for much smaller proportions of the British economy than they do today. In the eighteenth and nineteenth centuries, the great bulk of Britain’s economic activity was domestic.

  • Even then, Britain’s most important trading partners in the eighteenth and nineteenth centuries were not its colonies but other industrialising powers, such as Britain’s Western European neighbours.

  • Colonial empires do not come cheap. The acquisition, defence and administration of overseas territories require huge upfront investments and ongoing maintenance costs. This is why, in the eighteenth and nineteenth centuries, Britain and other colonial empires had higher levels of military expenditure than their less imperialist neighbours and, consequently, a substantially higher tax burden.

  • The economic benefits of empires are often overstated. Empires boost trade between their constituent parts, but they are far from the only determinant of trade volumes. At least some trade between Britain and India, for example, would have occurred anyway, even if India had never been colonised, or even if it had been colonised by some other European power.

  • The cost–benefit analysis for other European colonial empires is similar. The only major counterexample, i.e. a colony that was almost certainly profitable for the coloniser, is the Belgian Congo. But this is also a highly unusual example: a colony that was run like a private for-profit company, which the Belgian parliament stubbornly refused to subsidise. It was also a region that was exceptionally rich in sought-after natural resources.

  • The transatlantic slave trade was no more important for the British economy than brewing or sheep farming, but we do not usually hear the claim that ‘brewing financed the Industrial Revolution’ or ‘sheep farming financed the Industrial Revolution’.

  • Not all Western countries were major colonial powers. Some had only minor colonial possessions, some had only short-lived colonial empires, some only acquired colonies very late in the day, and some never had any colonies. Those minor players in the colonial arms race industrialised at roughly the same speed as the major colonial empires, so if there was an ‘empire bonus’, it is not visible in the macro data.

  • The claim that colonialism and slavery made the Western world rich is often accompanied by the claim that colonialism and slavery made the non-Western world poor. This companion thesis stands on stronger ground. There is indeed evidence for the long-term scarring effects of colonialism and slavery, since these corrupted the institutional development of the affected regions.

“[E]mpires do not come cheap. Burdensome expenditures are needed for military repression and prolonged occupation, for colonial administration, for bribes and arms to native collaborators […]

“But empires are not losing propositions for everyone. The governments of imperial nations may spend more than they take in, but the people who reap the benefits are not the same ones who foot the bill.”

— Michael Parenti, Marxist-Leninist political scientist and activist, 1995

“[T]he doctrine that imperialism made the West rich at the expense of the East and South is held passionately by the left in the West […] But understand: the counterargument does not praise imperialism, or excuse it. The counterargument claims that it was economically stupid.”

— Deirdre McCloskey, libertarian economist, 2009


In May 2020, in the US state of Minnesota, a police officer killed an unarmed suspect during an arrest. The officer was later found guilty of second-degree murder and various other charges, and sentenced to 22½ years in prison.

The police officer, Derek Chauvin, was white; the victim, George Floyd, was black. The eventual case against Chauvin did not indicate a racist motive (US Department of Justice 2022). Nonetheless, within hours of Floyd’s death, protests erupted all over the city, where slogans like ‘Jail killer KKKops’ and ‘Stop killin’ black people’ featured prominently.1

To say that these protests then ‘snowballed’ would be an understatement. What was initially a local policing matter for the city of Minneapolis soon spiralled, both thematically and geographically, and turned into a soul-searching exercise around the issue of race and racism, past and present, across the Western world. Protests organised or inspired by the Black Lives Matter (BLM) movement erupted all over the US. It took less than a week for them to spill over to the UK. Thousands gather in Britain to support US George Floyd protests. 2

The initial UK BLM protests had a copy-and-paste character, imitating the outward form, the slogans and the rhetoric of the American original. But they quickly developed a country-specific element, namely, a renewed focus on the UK’s imperialist past.

In June 2020, a group of BLM-inspired protesters in Bristol pulled down a statue of Edward Colston – a local merchant who, in the seventeenth century, had made a fortune from the slave trade – and threw it into a river. Shortly afterwards, a statue of Robert Milligan, an eighteenth-­century merchant and slave trader, was pre-emptively removed after having been targeted by protesters.3 London Mayor Sadiq Khan announced plans to set up a Commission for Diversity in the Public Realm, which would be tasked with a review of London’s statues and street names. Meanwhile in Oxford, thousands of people gathered to demand the removal of a statue of Cecil Rhodes, the nineteenth-century colonial governor.4 In September 2020, the University of Edinburgh announced that its David Hume Tower would be renamed;

“…because of the sensitivities around asking students to use a building named after the 18th century philosopher whose comments on matters of race, though not uncommon at the time, rightly cause distress today. This is ahead of the more detailed review of the University’s links to the past.”5

A few months later, Liverpool University renamed its Gladstone Halls, in response to a student campaign, on the grounds that William Gladstone’s father (though not William Gladstone himself) had been involved in the slave trade.6 In a perfect encapsulation of the zeitgeist, the halls were subsequently renamed after a female black communist Community Relations Officer.7

There were many more such examples up and down the country. In January 2021, the Guardian reported triumphantly:

“Scores of tributes to slave traders, colonialists and racists have been taken down or will be removed across the UK, […] with hundreds of others under review by local authorities and institutions. In what was described by historians as an ‘unprecedented’ public reckoning with Britain’s slavery and colonial past, an estimated 39 names – including streets, buildings and schools – and 30 statues, plaques and other memorials have been or are undergoing changes or removal since last summer’s Black Lives Matter protests.”8

There is, of course, nothing wrong with a public reckoning. But this is not an accurate description. What we have seen since the outbreak of ‘BLM-mania’ in 2020 was not simply a ‘public reckoning with Britain’s slavery and colonial past’, but rather the promotion of one very specific historical narrative about that past. It is a narrative in which colonialism and slavery are not just lamentable aberrations from an otherwise positive national story but, on the contrary, the very foundations on which the Western world’s wealth – and Britain’s in particular – was originally built.

One of the ‘Colston Four’, the organisers of the Bristol protests who pulled down the Colston statue, puts it succinctly when he says: ‘[S]o much of the prosperity enjoyed today in the UK […] comes off the back of historical atrocities.’9

This idea is endorsed by London Mayor Sadiq Khan: ‘It is an uncomfortable truth that our nation and city owes a large part of its wealth to its role in the slave trade’.10

The writer and broadcaster Afua Hirsch makes a similar case:

“The proceeds from […] enslavement […] provided the profits with which Britain modernised its economy. […] The case for reparations is becoming a global conversation to which every nation that systematically enriched itself by stealing black people’s very humanity […] now finds itself exposed.”11

For the writer and journalist Owen Jones, the slave trade and imperialism are capitalism’s original sin, which it is still tainted with today:

“Capitalism was built on the bodies of millions from the very start. […] [T]he transatlantic slave trade became a pillar of emergent capitalism. Much of the wealth […] was made from the enslaved labour of Africans. The capital accumulated from slavery […] drove the industrial revolution […]

“[T]he blood money of colonialism enriched western capitalism. India […] was a cash cow for British capitalism […]
The west is built on wealth stolen from the subjugated, at immense human cost.”12

Zarah Sultana, the MP for Coventry South and a social media influencer, believes that:

“[t]he wealth that enriched the British Empire and established it as a global superpower meant the murder, destruction, and brutalisation of people across the world.”13

Columnist and writer Nesrine Malik claims:

“If a country has […] profited from slavery and colonialism, it cannot escape or outrun the legacies of these foundational exploitations. […] Britain’s involvement in the slave trade […] explains much of why Britain […] looks how it does today.”14

BBC Bitesize, which provides educational materials for students and teachers, tells the same story:

“The British economy was transformed by the Atlantic slave trade. […] The slave trade was important in the development of the wider economy – financial, commercial, legal and insurance institutions all emerged to support the activities of the slave trade. Some merchants became bankers and many new businesses were financed by profits made from slave trading.”15

The idea is not new. In the academic literature, it is known as the ‘Williams Thesis’, after the historian and first post-­independence prime minister of Trinidad and ­Tobago, Eric Williams. Williams wrote in 1944 (cited in Engerman 1972: 432):

“[T]he profits obtained [from colonial trade] provided one of the main streams of that accumulation of capital in England which financed the Industrial Revolution.”

According to Williams, these profits ‘supplied part of the huge outlay for the construction of the vast plants to meet the needs of the new productive process and the new markets’.

We can find even earlier expressions of it. Karl Marx (cited in Heblich et al. 2023) wrote in the 1860s that:

“…the veiled slavery of the wage workers in Europe needed […] slavery pure and simple in the new world. […] [C]apital comes dripping from head to foot, from every pore, with blood and dirt.”

The Marx–Williams thesis was broadened and popularised in the 1960s, when the theories of ‘Maoism–Third Worldism’ became fashionable with the student protest movement. Disappointed with the domestic working class, which showed little inclination to fulfil its ‘historic mission’ to overthrow capitalism, student radicals looked for an ­ersatz-proletariat elsewhere and found it in the peasantry of the developing world. Jean-Paul Sartre, one of the leading intellectuals that inspired the student movements, did not see much revolutionary potential on ‘[t]his fat, pale continent […] [a]nd that super-European monstrosity, North America’ (Sartre 1961: 25–26). He located it elsewhere (ibid.: 11):

“[I]n those countries where colonialism has deliberately held up development, the peasantry […] stands out as the revolutionary class. […] [I]t demands no less than a complete demolishing of all existing structures. In order to triumph, the national revolution must be socialist.”

Sartre coupled this with a version of the Marx–Williams thesis (ibid.: 24–25):

“[W]e are exploiters. […] [W]e have laid hands on first the gold and metals, then the petroleum of the ‘new continents’, and that we have brought them back to the old countries. This was not without excellent results, as witness […] our great industrial cities […]

“With us, to be a man is to be an accomplice of colonialism, since all of us without exception have profited by colonial exploitation.”

The popularised version went beyond the original thesis in several ways. While Eric Williams had focused, more narrowly, on the transatlantic slave trade and the sugar plantations in the Caribbean, the popularised version broadened this to ‘Western imperialism’ in general. While Williams did not claim that slavery was the only factor or that the Industrial Revolution could not have happened otherwise, the popularised version simplified this to ‘the wealth of the West is built on slavery and colonialism’.

What we have seen since 2020 is an extension of that: the ‘BLM-isation’ or ‘Wokification’ of the Marx–Williams thesis. ‘Woke’, to borrow Noah Carl’s definition, means a view of the world;

“…which sees identity groups like sex and race as the primary units of society; which attributes to some groups the status of victims and to others the status of oppressors; and which posits that various ‘structural’ and ‘systemic’ forces stymie members of the former groups while conferring ‘privilege’ on members of the latter.”16

The ‘Great Awokening’, meanwhile, is the spread of that worldview in the media, the education system, cultural institutions and the corporate world (Rozado and Goodwin 2022; Kaufmann 2022a,b). The current version of the Marx–Williams thesis can be seen as part of that ‘Great Awokening’. It integrates it into a wider story about how capitalism is a system that was built on racism, slavery and imperialism, and that continues to foster racism and perpetuate racist oppression today.

For example, Arun Kundnani, a professor at New York University, writes:

“The liberal tradition sees racism as essentially a matter of irrational beliefs and attitudes. […] The radical tradition, on the other hand, sees racism as a matter of how economic resources are distributed differently across racial groups. […] Radical antiracists argue that the only way to fight this oppression is […] to dismantle existing social systems and build new ones. To them, racism is closely connected to capitalism. […] [R]acism weakens class struggle by dividing […] working people.”17

Kundnani’s book therefore comes with the self-explanatory title, What Is Antiracism? And Why It Means Anticapitalism (Kundnani 2023).
Similarly, in an article entitled ‘Why anti-racism must be anti-capitalist’ , the MP for Coventry South, Zarah Sultana, claims:

“[R]acism isn’t incidental. It’s central to capitalism […] It’s central to capitalism […] Racism in society isn’t a glitch, it’s a feature. It’s functional to […] the accumulation of capital. This […] is why racism is embedded in its social relations. […]

“The Conservatives and their billionaire press allies stir up hatred and fear in order to divide and rule. They use racism to project structural problems onto minority groups, to distract from social crises by demonising vulnerable people, and to make exploitation easier and more palatable. […]

“[O]nly a socialist analysis explains a system that breeds racism. This analysis tells us that alienation, exploitation, and falling living standards aren’t the fault of any religious or ethnic group, they are the nature of capitalism itself.”

Black Lives Matter UK also say that they are ‘guided by a commitment to dismantle imperialism, capitalism, white-supremacy, patriarchy and the state structures that disproportionately harm black people […]’.18 And elsewhere: ‘We are […] all anti-capitalists, and are committed to dismantle class as well as gender and racial domination.’19

While the ‘Culture War’ – that is, the Great Awokening and the backlash against it – gives these issues greater prominence, this book is nonetheless not a ‘Culture War publication’. It is not about ‘woke-bashing’, and it is not about who the ‘good guys’ and who the ‘bad guys’ in history are. For the purposes of this book, it is immaterial whether the reader takes a ‘Corbynite’ view of British history, in which Britain is always the aggressor and oppressor, a national conservative view, in which Britain is a benign force that sometimes fails to live up to its own ideals, a view that falls somewhere in between, or a different view altogether. The extent to which Britain benefited from colonialism is an empirical question, not a moral one. It can only be answered by looking at the costs and benefits of colonialism, to the extent that this is possible given the gaps in historical data, and comparing these to plausible counterfactuals.

The short summary of this book is that the Marx–­Williams view of colonialism is not borne out by the empirical evidence. Colonialism and the slave trade were, at best, minor factors in Britain’s and the West’s economic breakthrough, and quite possibly net lossmakers.

The Economics of Colonialism

The question of whether imperialism makes economic sense is older than Eric Williams’s or even Karl Marx’s writings. It is about as old as imperialism itself. The claim that the Empire is essential for Britain’s economy was initially made by proponents of imperialism, not anti-imperialists using it to portray Britain as an exploitative oppressor. It was the opponents of imperialism who believed that it would not pass a cost–benefit analysis. Adam Smith (2005: 499–500) wrote in 1776:

“The pretended purpose of it was to encourage the manufactures, and to increase the commerce of Great Britain. But its real effect has been to raise the rate of mercantile profit, and to enable our merchants to turn into a branch of trade, of which the returns are more slow and distant than those of the greater part of other trades, a greater proportion of their capital than they otherwise would have done […]

“Under the present system of management, therefore, Great Britain derives nothing but loss from the dominion which she assumes over her colonies.”

He thought the proposal ‘that Great Britain should voluntarily give up all authority over her colonies, and leave them to elect their own magistrates, to enact their own laws, and to make peace and war, as they might think proper’ was politically unrealistic for various reasons, but maintained (ibid.: 500):

“If it was adopted, however, Great Britain would not only be immediately freed from the whole annual expense of the peace establishment of the colonies, but might settle with them such a treaty of commerce as would effectually secure to her a free trade, more advantageous to the great body of the people, though less so to the merchants, than the monopoly which she at present enjoys.”

About sixty years later, Richard Cobden, one of the founders of the Anti-Corn Law League, described them as ‘a severe burden to the people of these realms’ (Cobden 1835: 111), and as ‘the costly appendage of an aristocratic government’ (ibid.: 150). Cobden wrote (ibid.: 241):

“[O]ur naval force, on the West India station […] amounted to 29 vessels, carrying 474 guns, to protect a commerce just exceeding two millions per annum. This is not all. A considerable military force is kept up in those islands […]

“Add to which, our civil expenditure, and the charges at the Colonial Office […]; and we find […] that our whole expenditure, in governing and protecting the trade of those islands, exceeds, considerably, the total amount of their imports of our produce and manufactures.”

Thus, Cobden believed that if Britain’s colonies fell into enemy hands, that ‘enemy’ would, in fact, be doing Britain a favour (ibid.: 243): ‘Where is the enemy (?) that would be so good as to steal such property?’

If imperialism was a loss-making activity, why did people engage in it for centuries?

Economists of the Public Choice School argue that governments frequently enact and sustain policies that make the country poorer overall, if the benefits of such a policy are concentrated, while its costs are dispersed. In this case, the beneficiaries will have a much greater interest in the policy than the people who bear the cost, and they will be much easier to organise politically. Most protectionist measures fall into this category, as do many industry subsidies and regulatory barriers to market entry.

Modern Public Choice Theory was a creation of the mid twentieth century, but the basic logic was already there – Smith’s (2005: 498) description of why imperialism exists:

“To found a great empire […] is […] a project altogether unfit for a nation of shopkeepers, but extremely fit for a nation whose government is influenced by shopkeepers. Such statesmen, and such statesmen only, are capable of fancying that they will find some advantage in employing the blood and treasure of their fellow-citizens, to found and maintain such an empire.”

For Smith, then, the Empire was what we would now call a rent-seeking scheme. The people who benefited from it were not the ones who had to bear the full cost of it, which was dispersed among all taxpayers and consumers.

This is also implicit in Cobden’s (1835: 24–25) writings:

“[I]f it could be made manifest to the trading and industrious portions of this nation, who have no honours or interested ambition of any kind at stake in the matter, that […] our dependencies are supported at an expense to them […] without improving our balance of trade – ­surely under such circumstances, it would become at least a question for anxious inquiry […] whether those colonies should not […] support […] themselves as separate and independent existences.”

Thus, for Cobden, the only people who can be relied upon to defend the Empire are those who have a direct commercial (‘interested ambition’) or a political (‘honours’) stake in it. The general public only does so because it has not been ‘made manifest’ to them they are the ones footing the bill.

Some liberals (most notably John Stuart Mill) were more ambivalent about the Empire, or even in favour of it, but there is a strong tradition of liberal anti-imperialism. The quasi–Public Choice case against imperialism, however, is not limited to liberals. We can also find examples of socialists, conservatives and people of other political persuasions, making similar arguments.

Otto von Bismarck was very much not a Cobdenite liberal, but he rejected colonialism in terms that Cobdenites – and Public Choice critics of pork barrel politics – would recognise (cited in BPB 2015, translation mine):

“The supposed benefits of colonies for the trade and industry of the mother country are, for the most part, illusory. For the costs involved in founding, supporting and especially maintaining colonies […] very often exceed the benefits that the mother country derives from them, quite apart from the fact that it is difficult to justify imposing a considerable tax burden on the whole nation for the benefit of individual branches of trade and industry.”

In his writing about the economics of imperialism, even Michael Parenti (1995), a Marxist-Leninist political scientist, sounds almost like a Public Choice economist:

“[E]mpires are not losing propositions for everyone. […] [T]he people who reap the benefits are not the same ones who foot the bill. […] [T]he gains of empire flow into the hands of the privileged business class […] The transnationals monopolize the private returns of empire while carrying little, if any, of the public cost. The expenditures needed […] are paid […] by the taxpayers.

“So it was with the British empire in India, the costs of which […] far exceeded what came back into the British treasury. […]

“[T]here is nothing irrational about spending three dollars of public money to protect one dollar of private investment – at least not from the perspective of the investors. To protect one dollar of their money they will spend three, four, and five dollars of our money. In fact, when it comes to protecting their money, our money is no object.”

This leads us to a curious situation. Today’s progressives would agree with old-school imperialists on one thing, namely, that the Empire was essential for Britain’s prosperity – but they would disagree on that question with Marxist anti-imperialists in the mould of Michael Parenti.

Another explanation for imperialism is that colonies are best understood as vanity projects, the purpose of which was not to enrich the economy, but to enhance the international status and prestige of the governing elites. As McCloskey (2009: 14–15) puts it:

“…the average person in Britain got little or nothing out of the British Empire. Yet […] Queen Victoria loved becoming an Empress and Disraeli loved making her one, and so imperial India was born.”

Adam Smith (2005: 500) also talked about how giving up colonies, although ‘agreeable to the interest’ of the nation as a whole, would be;

“…contrary to the private interest of the governing part of it, who would thereby be deprived of the disposal of many places of trust and profit, of many opportunities of acquiring wealth and distinction, which the possession of the most turbulent, and, to the great body of the people, the most unprofitable province, seldom fails to afford.”

A third explanation is that colonialism is an expression of jingoism and, as such, need not be economically profitable in order to be popular. Adam Smith thought that giving up colonies was ‘mortifying to the pride of every nation’ (ibid.).

These three explanations are not mutually exclusive. As we have seen, Adam Smith believed in all three of them. What they show, whether taken together or in isolation, is that the mere fact that empires existed does not, in any way, prove their profitability.

But were they profitable?

There would be no perfectly straightforward answer to this question, even if colonies still existed today. There would be disagreement about which costs and which benefits should be specifically attributed to colonialism and about what the exact alternative would be. We can see this today in the difficulties associated with modelling the costs and benefits of Brexit – even now that Brexit has already happened (and far more so while it was still a hypothetical). This problem is many times greater when we are talking about centuries for which data are much patchier, and when there is a much broader range of counterfactuals.

But there have been attempts to remedy this problem, which get us at least closer to an answer. They cannot give us definitive numbers, but they can at least narrow things down to a range of possible outcomes. The following chapters will show this for the most important colonial empires of the nineteenth and early twentieth centuries.

The British Empire

In a paper published in the Economic History Review, Patrick O’Brien (1982), an economic historian, compiled flows of trade, investment and profits between the Western world and what he calls ‘the periphery’ in the late eighteenth and early nineteenth centuries. ‘The periphery’, in his paper, is defined as Asia, Africa, Latin America and the Caribbean, but also the plantation states of the southern US.

The first thing that stands out is that, even though, by the standards of the time, Britain was an exceptionally open and trade-oriented economy, by today’s standards, trade only accounted for a relatively small proportion of Britain’s economic output. In the late eighteenth century, Britain’s trade-to-GDP ratio (exports plus imports divided by overall economic output) approached 25 per cent (ibid.: 4–5). Back then, this was a lot: more than twice the Western European average. Today, however, even heavily protectionist economies, such as Iran or Argentina, trade considerably more than that. In 2019, the global average trade-to-GDP ratio stood at over 50 per cent, and the British figure at over 60 per cent (Our World in Data 2023). In relative terms, Britannia may have ‘ruled the waves’, but in absolute terms, the bulk of its economic activity was domestic.

An exact breakdown is not available, but most of Britain’s trade, at the time, was not with the periphery, but with other industrialised or industrialising economies, particularly in Western Europe and North America (O’Brien 1982: 4–5). Even then, ‘the periphery’ was a lot more than the British Empire. It was the entire non-Western world, broadly defined.

O’Brien also adds up all the profits earned through commercial activities in the periphery in the late eighteenth and early nineteenth centuries. These were, indeed, substantial; if 100 per cent of them had been invested in the British economy, it would have been enough to finance about half of Britain’s total investment. Unfortunately, we do not know what proportion was really used for investment purposes. But O’Brien shows that the overall savings rate in Britain was about 12–14 per cent at the time. If we assume that colonial entrepreneurs were exceptionally frugal and far-sighted people who invested 30 per cent of their total profits every year (i.e. more than double the average savings rate), it would have been enough to fund around 15 per cent of Britain’s investment expenditure (ibid.: 7). If we assumed more typical savings rates, it would have been more like 7 per cent.

In short, the bulk of Britain’s economic activity was domestic, the bulk of Britain’s trade was with other Western nations rather than its colonies, and the bulk of Britain’s investment spending was financed from domestic savings and profits from intra-Western trade.

This still leaves a sizeable role for the colonies. The next step would be to disentangle what proportion of this constitutes a benefit of colonialism, and what proportion of it simply constitutes a benefit of trade. Not every economic transaction that took place in a colony took place because it was a colony. To what extent were economic interactions with the colonies dependent on their political status as colonies? If a British overseas merchant made a £100 profit from various activities in British India – what proportion of that profit is owed to the fact that British India was, indeed, British India, as opposed to Dutch India, French India, or an independent India? £90? £50? £10? This is one of the main points of disagreement between proponents of the Marx–Williams view and proponents of the Smith–Cobden view.

Richard Cobden believed that if Britain had dissolved its empire and granted independence to its colonies, colonial trade would simply have been replaced by free trade between sovereign, independent countries (Cobden 1835: 243).

Similarly, Deirdre McCloskey (2009: 10) believes that;

“…trade could have been achieved on more or less the same terms if India had been independent. It would have likewise if India had become a French rather than a British colony.”

O’Brien (1988: 165) also argues:

“[C]ommercial relations with France, Spain, the United States, even with Brazil, were not predicated upon similar degrees of political intrusion by the British government. Might not British businessmen, with or without the raj, have been engaged in similar kinds and levels of commerce with India? With or without the colonial office might not trade with Jamaica have been much the same as it was over the period 1850–1914?”

There is also a third position somewhere in between, which holds that some, but far from all of the gains from trade with the colonies depended on them being colonies. The historian Paul Kennedy (1989: 187) argues:

“[I]t remains difficult to envisage that the political circumstances – the non-existence of the raj – could be so drastically different without the economic circumstances […] being affected at all.”

He believes that this would require ‘a world in which Cobdenite rationality completely prevailed’.

More recent empirical evidence supports this intermediate position. It finds that empires boosted trade relative to alternative arrangements that were common at the time, but they were far from the sole driver.

In a paper published in The Economic Journal, Mitch­ener and Weidenmier (2008) investigate the determinants of international trade flows in the period of High Imperialism (i.e. 1870–1913). They show that, controlled for other factors such as geographical distance or access to the sea, trade flows between countries were higher when they were part of a common empire.

The size of the effect varies between different model specifications, but according to their main estimate, being part of an empire increased trade flows by 115 per cent, i.e. it more than doubled them (ibid.: 1818). This is because empires led to reductions in trade barriers and transaction costs, and while countries could, in principle, also have done all that without empires, in practice, they were not especially likely to.

If the empire increased trade to such an extent, it clearly had some economic benefits for Britain. But as a flipside, it also follows that nearly half of Britain’s trade with its colonies would still have taken place if they had never been colonies. As mentioned, the authors’ estimates vary widely, and the true proportion may have been higher or lower than that. But there is no estimate that says that there could have been no trade with Asia, Africa, Latin America or the Caribbean without the Empire.

Unfortunately, what is true for the benefits of empire is true for the cost of empire as well: we cannot boil it down to a single figure; we can only find a spectrum of possible outcomes. Some historians believe that without the empire, Britain could have drastically reduced its military and administrative expenditure, while others believe that the potential savings would have been much smaller. O’Brien (1988) shows that throughout the second half of the nineteenth and in the early twentieth centuries, Britain’s military expenditure was far above the European average in per capita terms, amounting to more than twice the French or the German levels. He goes on to show that if Britain’s military expenditure had been reduced to French or German levels, the British tax burden could have been reduced by almost a quarter (ibid.: 189). One need not be a believer in ‘Laffer Curve’ economics to see that a tax cut of such magnitude must have had some positive effect on the British economy. The historian Paul Kennedy (1989), on the other hand, believes that this is not a like-for-like comparison, that British military expenditure was not excessive, and that even without the Empire, there would only have been limited scope for cutting it.

We do not have the expertise to pick a side in this debate. Suffice it to say that, whether it was ‘excessive’ or not, the Empire came at a substantial fiscal cost (a point which Kennedy does not dispute). That cost needs to be subtracted from whatever gains we may want to attribute to the Empire.
To sum it up, we cannot say that the British Empire was definitely profitable or definitely unprofitable. What we can say is:

  • most of Britain’s economic activity was domestic, with overseas activities being, by today’s standards, a small proportion of the total;

  • most of Britain’s trade was with other Western economies, not its colonies;

  • most of Britain’s investment was financed via domestic savings;

  • at least some of the gains from ‘the Empire’ were really gains from trade, which would have occurred with or without the Empire; and

  • the Empire had at least some fiscal cost, in the form of higher military and administrative expenditure, which, without it, could have been used for other purposes.

In order to find a positive net contribution of the Empire, we have to make some combination of the following assumptions:

  • the bulk of the profits earned with the colonies were earned because they were colonies, and could not have been earned otherwise;

  • colonial entrepreneurs were very frugal people who invested large proportions of their profits; and

  • the cost of the Empire was modest.

These assumptions are not wholly implausible, so we cannot rule out the possibility that the Empire made a positive net contribution. What we can say is that if a positive net contribution existed, it could not have been anywhere near large enough to justify the conclusion that Britain’s wealth was ‘built on’ colonial profits.

The British Empire after 1850

In a follow-up paper, O’Brien (1988) asks a related but slightly different question. Rather than looking at the total cost and benefit of the British Empire over the whole period of its existence, he looks at the incremental costs and benefits of keeping the British Empire after the mid nineteenth century, and enlarging it further. Thus, his comparison is not with an alternative timeline in which Britain never had an empire. It is an alternative timeline in which the British Empire existed, but was dissolved over the course of the 1850s or 1860s. Framing the question in this way should bias the study in a Marx–Williams direction. The initial upfront investment, that is, the Empire’s acquisition cost, has already been paid at this point and is therefore disregarded. Only the operating cost of maintaining the Empire beyond this point matters.

O’Brien first shows the relative importance of the Empire to the British economy in the second half of the nineteenth and the beginning of the twentieth centuries. In that period, the Empire accounted for between a fifth and a quarter of Britain’s imports (ibid.: 167), a little over a third of Britain’s overseas investment, and about a quarter of Britain’s total investment (ibid.: 173–74). These are not tiny numbers. If all of that had suddenly disappeared, the British economy would clearly have taken a hit.

O’Brien goes on to cite a related study, which models what would have happened to British trade after a hypothetical loss of the Empire, under the assumption that the former colonies would have imposed tariffs on British goods at the same rate as the US did at the time. The model also makes the rather pessimistic assumption that the British economy would not be able to adapt to this new situation by redeploying its resources. Thus, in this model, the loss of the Empire does not just mean a one-off hit followed by a rebound, but a permanent loss that grows larger over time. Of course, if we just assume away an economy’s capacity to adapt, any change must seem like a bad thing, so by design, this model leads us to the conclusion that the loss of the Empire would have made Britain poorer. But even in this model, the magnitude of the effect is not dramatic. The author estimates that by 1913, this alternative, non-imperialist Britain would have been 3.3 per cent poorer than the actual Britain (ibid.: 168); 3.3 per cent is not trivial, but to put it in perspective, it is less than the estimated cost of Brexit.

O’Brien does not believe that the economic hit of dissolving the Empire would have been anything like as bad as this model implies. But he also believes that, even if it had been, it probably still would have been worth it because of the high fiscal cost of the Empire (ibid.: 189). Without the Empire, military spending could have been cut back to levels more in line with Britain’s continental peers, and the savings could have been used for some combination of tax cuts, deficit reduction and public investment. O’Brien (ibid.: 199) concludes:

“[T]he notion that the empire made any positive long-term contribution to the health of the domestic economy is unlikely to survive systematic economic analysis and statistical testing. Modern research in economic history now lends rather strong empirical support to Cobdenite views.”

The Transatlantic Slave Trade

Before it was popularised in the late 1960s, the original Marx–Williams thesis was less about colonial empires per se; it was more narrowly about the transatlantic slave trade and the plantation economies powered by slave labour. Between the seventeenth and the early nineteenth centuries, some families in Britain got very rich through their involvement in either the slave trade itself or in the Caribbean plantations, where slave labourers cultivated sugar, tobacco, coffee and other goods. This is completely undisputed, and it has left visible legacies across the country. Therefore, when the National Trust decided to publish its Interim Report on the Connections between Colonialism and Properties now in the Care of the National Trust, Including Links with Historic Slavery (Huxtable et al. 2020), this was not an exercise in performative ‘wokery’ or self-flagellation. It is part of the history of those estates, and there is nothing wrong with acknowledging that.

But did Britain, as a whole, benefit from the slave trade? As for the Empire as a whole, the debate around this is not new; it was already raging when the slave plantations were still going strong. Adam Smith (2005: 345) believed that slavery was inherently unproductive:

“[G]reat improvements are […] least of all to be expected when [proprietors] employ slaves for their workmen. The experience of all ages and nations, I believe, demonstrates that the work done by slaves, though it appears to cost only their maintenance, is in the end the dearest of any.”

His reasoning was simple: a slave labourer will do what they are forced to do, but they have no incentive whatsoever to go even minimally beyond or to develop any initiative of their own.

Smith saw slavery as a wasteful indulgence, which was pursued for non-economic reasons: ‘The pride of man makes him love to domineer’ (ibid.). As with all wasteful indulgences, only some people can afford to pursue them, and only under some circumstances: ‘The planting of sugar and tobacco can afford the expense of slave-cultivation. The raising of corn, it seems, in the present times, cannot’ (ibid.).

This turns the Marx–Williams logic on its head. According to the latter, sugar and tobacco plantations were profitable because they employed slave labour. According to Smith, it was the other way around: since sugar and tobacco plantations were profitable, they could afford to use an inferior method of production, which people in more competitive sectors would not have been able to get away with. Slavery, in other words, was not the cause of the plantation sector’s profitability, but a consequence of it.
Even Eric Williams did not claim that slavery was per se profitable. Williams’s argument was that when the British Empire abolished slavery in the early nineteenth century, this had less to do with humanitarian concerns and more with the fact that slavery had outlived its usefulness to the British economy by then.

Which raises the question: does this distinction between an early phase, when slavery was instrumental for Britain’s economic development, and a later phase, when it no longer was, actually exist?

Analogous to the argument about the profitability of the Empire as a whole, in order to reach a verdict, we would have to answer three questions:

  • How big were the profits derived from slavery, compared to the overall size of the British economy, or British investment?

  • Did those private profits exceed the cost to the taxpayer?

  • To what extent did the plantations rely on slavery? Could they have existed without it, on a smaller scale? Could they, as Smith believed, have been even more profitable without it?

In a paper in the Journal of Economic History, the economic historians David Eltis and Stanley Engerman (2000: 135) show that profits from the slave trade were about equivalent to just under 8 per cent of Britain’s total investment. Even if we assume that slave traders and plantation owners were very frugal and forward-looking people, with savings and investment ratios far above the national average, it would have been impossible for them to contribute more than a few per cent to Britain’s total investment. The authors’ verdict (ibid.: 129):

“Historical interest in the slave trade rests on its obvious immorality, not its economic importance. The business formed a relatively small share of the Atlantic trade of any European power. Its direct contribution to the economic growth of any nation was trivial. […] The slave trade […] accounted for less than 1.5 percent of British ships, and less than 3 percent of British shipping tonnage. Shares of cargoes carried and earnings from freight were in the same low range.”

They also show that the sugar plantations, around the peak, added just under 2.5 per cent to the value of the British economy. This was less than the share of sheep farming (ibid.: 134), but we do not usually hear the claim that ‘the profits from sheep farming kickstarted the Industrial Revolution’, or that ‘capitalism was built on sheep farming’.

The authors explore the possibility that the slave trade and sugar might have been ‘strategic’ industries in some way, which, even if their direct contributions to the British economy were small, might have had major multiplier effects, stimulating other branches of industry. But they find no plausible reason why they would have had larger multiplier effects than other industries.

The figures above refer to the gross contribution of the slave trade and the plantation sector to the British economy. To arrive at the net contribution, we would have to subtract the additional cost to the taxpayer associated with the slave trade and the plantation system, that is, the cost that was not borne by the slave traders and plantation owners themselves. Eltis and Engerman have no precise estimate for this, but they point out (ibid.: 128):

“[P]rivate profits were offset, at least in part, by the public defence of this new English territory, a cost which was many times higher than would have been the case if it had been part of the British Isles.”

We then need to consider the counterfactual in order to work out to what extent those gains (assuming there were positive net gains) were made because of the slave trade and to what extent they would have been made anyway. Eltis and Engerman believe that a plantation economy of some sort would probably have existed with or without the slave trade (ibid.: 136–37):

“[T]o assume that in its absence there would have been nothing but subsistence agriculture in the Americas seems unrealistic. […] [O]nly some rather strong assumptions would support an argument that there would have been almost no production of sugar in the Americas and no trading links with Africa.”

Eltis and Engerman represent one side in a debate, not a definitive conclusion. What is perhaps more interesting is that even scholars who see themselves much more on the Marx–Williams side of the debate than on the Smith–Cobden side struggle to come up with numbers that would match their rhetoric.

The authors of a recent study on the subject, Heblich et al. (2023), summarise their findings with the bold claim that ‘our results strongly suggest that Marx was right: slavery wealth accelerated Britain’s industrial revolution’. But then, their actual study says that ‘[a]t the aggregate level, we find an increase in national income of 3.5 percent’ (ibid.: 42).

A one-off increase of 3.5 per cent is nothing to be sneezed at. If a government could permanently make Britain 3.5 per cent richer, there can be no doubt that government representatives would be endlessly bragging about it. However, when Karl Marx said that ‘capital comes dripping from head to foot, from every pore, with blood and dirt’, it is safe to say that he was not thinking of a contribution of 3.5 per cent.

Even then, 3.5 per cent is the gross gain. What the authors do not explore is the question of whether that gain exceeded the additional cost to the taxpayer or the question of what part of it would have been made anyway, with or without the slave trade.

The most substantial recent addition to the literature, on the Marx–Williams side of the debate, is the book Slavery, Capitalism and the Industrial Revolution by Maxine Berg and Pat Hudson (2023). Although the title of the book is clearly a homage to Eric Williams’s (2022 [1944]) book Capitalism and Slavery, which is referenced very favourably throughout, the authors phrase their central thesis remarkably cautiously (ibid.: 7):

“We do not argue that slavery caused the industrial revolution. Neither do we suggest that slavery was necessary for the development of industrial capitalism in Britain. Even less does our study attempt to estimate that the gains from slavery contributed a particular percentage to Britain’s economic growth, GDP or capital formation […] as earlier studies have attempted. That is not our purpose, partly because many aspects of the impact of slavery are not measurable in quantitative terms. What we do say is that the role of slavery in the process of industrialization and economic transformation […] has been generally underestimated by historians.”

However, while it is true that they never literally say ‘slavery caused the industrial revolution’ or ‘slavery was necessary for the development of industrial capitalism’, the rest of the book strongly implies that it made a major contribution. The authors do not dispute that slavery was propped up by government subsidies (ibid.: 21):

“The state supported overseas trading activity with military force throughout the eighteenth century, especially in the Americas. Between 1660 and 1815, Britain was at war for seventy of the 155 years. Most wars […] involved trading rights and colonial possessions in the Atlantic. […]
Wars in defence of the American colonies and naval action to enforce the Navigation Acts relied on increased government spending and taxation. Tax revenues rose […], making the British second only to the Dutch as the most heavily taxed population in Europe.”

But rather than treating this military spending as a cost to the economy, they treat it as a stimulus package: ‘Spending on the navy, from shipbuilding and munitions to the provisioning of voyages, stimulated the economy’ (ibid.: 21). And elsewhere: ‘The ‘cost’ of colonial defence was […] largely offset by the […] stimulus it created for the economy, in the demand for munitions, ships, ships’ provisions and uniforms’ (ibid.: 44).

Thus, their argument is not necessarily that the gains made from slavery outweighed the cost (although they may well think that they did). Rather, they simply redefine the costs as disguised benefits.

Framed in those terms, the book can only come to the conclusion that slavery was profitable for Britain. This is, strictly speaking, not even a conclusion; it simply follows as per assumption. If an activity has no real costs, but only benefits, it must logically be profitable. But this represents a kind of über-Keynesianism, in which the state cannot waste resources.

A lot hinges on this assumption, because if the net gains from slavery were small or even negative, they could not have financed the Industrial Revolution. Berg and Hudson dispute this point (ibid.: 44):

“Were the returns from the Caribbean colonies worth the high costs of their defence and administration? Adam Smith thought not, as did a number of economic historians of the 1960s and 1970s. But this misses the point because as long as high net private returns were made, the potential was there for the proceeds to flow into the industrializing economy.”

It does not ‘miss the point’ at all. If those private returns were offset by losses elsewhere in the economy, then yes, the former may well have flown into the industrialising economy – but at the same time, the latter must have flown out of the industrialising economy.

Berg and Hudson even apply this über-Keynesianism to the slave ownership compensation programme of the 1830s. When slavery was made illegal across the British Empire, the former slave owners were entitled to sizeable compensation payments. This was a pure zero-sum redistribution from non-slaveowners to slaveowners. The authors themselves describe it as a ‘subsidy from British wage earners and consumers of basic commodities to […] ex-slave owners’ (ibid.: 195). But they then go on to claim that the compensation money ‘aided the mid-Victorian investment boom in British and overseas railways and public utilities. Some was invested in industry’ (ibid.: 197).

Macroeconomics is not the main focus of the book. As mentioned, Eltis and Engerman (2000) dispute the idea that slavery was a ‘strategic industry’ that created positive spillover benefits for other industries. Berg and Hudson, on the other hand, believe that it very much did. They argue that it led to innovations in finance, corporate governance and accounting, agronomics and other sectors. Again, we do not have the expertise to pick a side in this debate. But even if it turned out that Berg and Hudson are completely right about this and Eltis and Engerman completely wrong, it would still not confirm the Marx–Williams thesis.

Berg and Hudson criticise conventional accounts of the Industrial Revolution for overemphasising domestic institutional factors in Britain, and neglecting what happened elsewhere in the British Empire. However, the indirect benefits of slavery they focus on – the impact on finance, corporate governance, agronomics, etc. – are also, ultimately, domestic institutional factors, even if they had an external stimulus. At that stage, the argument is no longer that slave money financed the Industrial Revolution. It is that slavery indirectly triggered domestic institutional changes in Britain, which later made Britain more productive – a very different argument.

So, on the whole, the economics of slavery are analogous to the economics of the Empire. The gains were small relative to the size of the British economy, and they cannot have explained more than a small share of total investment. Once we subtract the fiscal cost, the net gains may well have been negative.

There is only one major difference between the slave trade and the colonial trade. As shown above, empirical studies show that empires boosted trade, but they were far from the only factor. Around half of the trade with the colonies – perhaps more, perhaps less – would still have happened if they had never been colonies. That argument cannot be made with any degree of confidence about the Caribbean plantations. There was already a rudimentary plantation economy in the Caribbean before the slave trade fully took off, but its growth prospects were constrained because plantation owners struggled to recruit workers who were willing to move there and fill those posts voluntarily (Wright 2020). So, the most likely alternative to the slave plantations would not necessarily have been a thriving plantation economy in which free labourers, both black and white, would have voluntarily sold their labour and earned market wages. Rather, these plantations might never have taken off, or not until much later.

Western Europe as a Whole

Data for Western Europe as a whole are even less kind to the Marx–Williams perspective than data for Britain alone. The aforementioned paper by O’Brien (1982) shows that in the second half of the eighteenth century, Western Europe’s trade-to-GDP ratio was less than half of Britain’s, and three quarters of that trade was with other industrialising economies, not with the global periphery. We can make a series of assumptions that are heavily biased in a Marx–Williams direction, such as:

  • those overseas activities were sensationally profitable;

  • the people engaged in those activities were exceptionally frugal and future-oriented, using a huge proportion of their profits for investment purposes;

  • colonial conquest and administration were sensationally cheap; and

  • that trade with the periphery would not have happened otherwise.

Even then, this would still not take us anywhere near the conclusion that it was colonial exploitation that financed Western Europe’s industrialisation. The numbers are simply not big enough.

O’Brien’s argument is not that an alternative, ‘anti-­imperialist’ timeline, in which no European nation ever acquired any overseas territory, would have been indistinguishable from our actual timeline, in economic terms. He believes that imperialism made a difference – but it did so mostly in terms of consumption, not production (ibid.: 10–12):

“[I]f Europeans had been compelled to pay a ‘free market price’ for their tropical imports (that is to say a price which reflected the real cost of attracting and maintaining ‘free’ labour in the New World) then the prices of sugar, tobacco, spices, cotton, indigo, coffee, and other produce would have been far higher. […] [T]heir real incomes would have diminished. […] But […] the fall in real income […] could not have made that much difference to the levels of wealth and income achieved in Western Europe by 1807. […]

“[L]ong-run gains from specialization, the division of labour, and the forces of competition […] originated overwhelmingly in exchanges between and within European countries and far less from trade with other continents. […]

“[T]here is no reason to claim that if Western Europe had been forced to manage without imported sugar, coffee, tea, tobacco, and cotton, its industrial output could have fallen by a large percentage. A decline of not more than 3 or 4 per cent in the industrial output of the core would seem to be the likely short-run effect from a total cut-off of imports. Over time that impact could be mitigated by patterns of substitution for tropical foodstuffs and raw materials.”

So, there may well have been some positive impact on living standards, at least in gross terms (before we subtract military and administrative spending), but this is still not confirmation of the Marx–Williams thesis. The latter implies that the slave trade and colonialism did not just enrich those Europeans who were alive at the time, but that they laid the foundations of the wealth we still enjoy today. However, if colonialism and the slave trade mainly affected consumption rather than production, the gains would have been limited to the generations experiencing them.

The French Empire

The French colonial empire was the second-largest of the European colonial empires of the nineteenth and early twentieth centuries. At its height, it covered most of North Africa but also large parts of Asia, the Middle East and the Caribbean.

The economic historian Jacques Marseille has compiled a range of macroeconomic figures of French colonialism in his book Empire coloniale et Capitalisme française. The most important ones of these have been reproduced by the historian David Fieldhouse (1986) in his (English-­language) review of the book for the Journal of African History.

Until the late nineteenth century, the French colonial empire was relatively unimportant for France in economic terms, but from then on, it rapidly increased in both relative and absolute importance. By the late 1920s, the colonies accounted for about 13 per cent of French imports, 17 per cent of French exports, and a third of French overseas investment (ibid.: 170). Those figures reached their peak in the late 1950s, just before the end of the French Empire, when the colonies accounted for 27 per cent of French imports and 37 per cent of French exports.

These are unusually high figures for a colonial empire, which has to be a reflection of the fact that, unlike most other empires, France integrated its colonies into a customs union–style trade arrangement (Mitchener and ­Weidenmier 2008: 1818–21). Customs unions facilitate trade among their members, but they can hinder external trade, i.e. trade with non-members. In such a case, a high trade volume between members does not necessarily indicate that that trade is especially lucrative, just that it is heavily encouraged by the tariff and quota regime.

Indeed, Fieldhouse (1986: 170) points out:

“[T]he great majority of imports from the colonies were things such as wine, cereals, rice, coffee, cocoa, oil seeds and sugar which (except in time of war or post-war international shortage) were in ample supply in the world market […]

“In no sense […] did the empire provide raw materials which were essential to France and which could not otherwise have been obtained; nor did France get any price advantage by importing from the colonies; in fact, to the contrary.”

Fieldhouse does not explicitly say whether France would, on the whole, have been economically better off or worse off, had it never had a colonial empire. Instead, he argues that the French Empire made economic sense during some periods but not others (ibid.: 171):

“[U]ntil about 1930 the ‘autarkic’ system of protected colonial markets was, on balance, probably good for the French economy because the domestic market was too limited to provide economies of scale and protected Western markets were too difficult for France to enter. […]

“But from c. 1930 […] [t]he effect […] was almost certainly adverse for the French economy. Obsolescent industries were kept alive; capital and labour, which might have flowed to newer, more dynamic industries, was penned back; both metropolitan and colonial price levels were raised above world levels.”

He sees the quarter century before 1914 as the golden age, but also points out that in that period, ‘the colonial share of metropolitan trade and investment was relatively small’ (ibid.: 172).

One could thus paraphrase this assessment by saying that when the economic effect of the colonies was positive, it was not large, and when it was large, it was not positive.

In a paper in the Journal of Economic History, Elise ­Huillery from the Economics Research Center of Paris Dauphine University looks at the fiscal impact of French colonialism between 1844 and 1957. She specifically seeks to challenge the notion that the colonies were a heavy burden for the French taxpayer. She shows that, while there were indeed substantial fiscal transfers from France into its colonies, there were also various direct and indirect transfers in the opposite direction. Once these are subtracted, annual net transfers from France to its empire amounted, on average, to a mere 0.29 per cent of French public spending (Huillery 2014: 25–27).

If these estimates are correct, it would show that French colonialism was broadly self-funding. What it would not show – and what Huillery does not claim – is that the colonies generated any fiscal surplus for the French state.

Overall, the French case remains somewhat ambiguous. We cannot definitively say whether an alternative France without colonies would have been better off, worse off, or about the same, as the actual France. It does not help that a lot of the economic effects that Fieldhouse describes – both good and bad – may really be effects of protectionism rather than colonialism. If we take a protectionist trade policy as given, having colonies may be beneficial simply because it enlarges the home market. Colonies then become a ­clumsy way to break out of protectionist constraints. But, of course, protectionism is not a given. It is a political choice. A better alternative would have been to pursue a policy of economic openness and free trade instead.

The German Empire

Germany did not become a major colonial power until the mid 1880s, partly for the obvious reason that it did not become a nation state until 1871 (some individual German states had smaller colonies of their own before), but also because, as mentioned, Chancellor von Bismarck did not believe in the benefits of colonialism. But once Germany entered the colonial arms race in earnest, it rapidly caught up with the established imperial powers: for a short period, the German Empire became the world’s third largest colonial empire, after the British and French ones.

In the German case, the colonies cannot have been a causal factor in the country’s industrialisation, even if they had been profitable, for the simple reason that industrialisation came first and the acquisition of colonies came later.

But they were not profitable. German colonial administrations kept profit-and-loss accounts, much like a private company. We can see from those accounts that, while a few of them made minor surpluses, the majority were major lossmakers. The colonial empire as a whole could only cover about three-quarters of its costs; the rest had to be met by subsidies from Berlin.

The colonies accounted for less than 1 per cent of German exports, less than 0.5 per cent of German imports, and no more than 2 per cent of German overseas investment (Baumgart 1992: 145). They also failed in their (flawed) ‘Malthusian’ aims of creating a living space for the ‘surplus population’ to emigrate to; the white population in the colonies only increased to a little over 20,000 people (ibid.: 144). German colonialism was, in short, an unambiguous failure in every respect, powerfully vindicating von Bismarck’s initial empire scepticism.

The Belgian Empire

Belgium’s colonial rule of Congo is often singled out as a particularly atrocious form of colonialism. It may have claimed up to 10 million lives (Darwin 2008: 312), and it attracted widespread condemnation even at the time: in the early twentieth century, due to international pressure, Belgium had to make changes to the way the colony was run.

The Belgian Congo is the closest thing we have found to a colony that fits the currently fashionable view of colonies as cash cows for the coloniser. It was, in all likelihood, profitable for Belgium (even if the gains were highly concentrated).

The Belgian Congo was also an unusual example of colonialism in various ways.

Belgium entered the colonial arms race around the same time as Germany, and thus very late in the day. While King Leopold II was a keen advocate of colonialism, Belgium’s federal parliament was opposed, and managed to block the monarch’s colonial ambitions for a long time. They eventually settled for the following compromise: Leopold II would be able to pursue his colonial project in ­Congo, but he had to do so in a private capacity, not in his role as the head of state of Belgium. This meant that he would have no recourse to the national budget.

As the historian John Darwin puts it (ibid.: 310):

“[H]is [King Leopold’s] ‘Congo Free State’ […] was a private empire that belonged to him, not the Belgian state.

“It was only in 1908 that the colony was ‘nationalised’, and became a Belgian colony in the conventional sense, but even then, parliament retained its reluctance to commit any public funds to it.

Related to that, the Congo venture was more about profit-making than ‘national greatness’. In its first incarnation (1885–1908), the colony was not even officially called the ‘Belgian Congo’, but simply the ‘Congo Free State’. As Darwin explains (ibid.: 312):

“[T]he real object of Leopold’s shambolic ‘government’ was to coerce the population into collecting ivory and rubber (both highly profitable crops).”

Thirdly, Congo was unusually rich in minerals and raw materials that had valuable industrial uses. Economic historians have compared that natural wealth to the oil wealth of a Middle Eastern petrodollar state (Buelens and Marysse 2009: 158).

In a paper published in the Economic History Review, Frans Buelens and Stefaan Marysse from the University of Antwerp have compiled the rates of return on investment in Congo, compared to domestic investment in Belgium.

They show that despite heavy losses during the first decade or so, profitability later soared, and for most of the first half of the twentieth century, investment in Congo was considerably more profitable (if more volatile) than domestic investment in Belgium. Congolese mining returns, in particular, were among the highest returns an investor could earn anywhere in the world. It was only in the 1950s, when colonial rule became unstable, that the value of Congolese assets dropped.

Nor was Congo’s contribution to the Belgian economy insubstantial. In the 1920s, companies that were active in Congo accounted for a quarter of the total market capitalisation value of Belgian stocks, a share that peaked at over 40 per cent in 1955 (ibid.: 150–51). Four large companies accounted for the bulk of this.

The authors are not trying to estimate whether Belgium was richer as a result of its colonial exploitation of ­Congo. But if four large companies (and several smaller ones) made enormous profits from the venture, and if the Belgian state made no net contribution to the upkeep of the colony, then in aggregate terms, the Belgian Congo must have been profitable, unless there were some substantial unaccounted-for indirect costs.

Thus, the example of Congo shows that it is possible to run a colony profitably. The Smithian–Cobdenite suspicion that colonies are expensive lossmakers is not always and everywhere true.

But it also suggests that profitable colonies require unusual circumstances that would be difficult to replicate. It does not show that ‘colonialism’ is profitable. It shows that a colony can be profitable if it is run like a profit-­maximising private business, if parliament obstinately refuses to subsidise it, and if it is extremely resource rich.

Even so, the Belgian Congo does not confirm the Marx–Williams thesis. Belgium was an early industrialiser (ERIH 2023), but a late coloniser. It was already a major industrial power long before the colonisation of Congo started, and it would have been one even if no Belgian had ever set foot on Congolese territory.

Cross-Country Variation and Legacy Effects

Proponents of the popularised version of the Marx–­Williams thesis usually treat Britain as the worst imperialist offender and the rest of the Western world as less extreme versions of the same tendency. But this understates the amount of variation that really existed across the West in terms of colonial practices. What we really get is a spectrum of imperialism, which ranges from vast colonial empires that lasted for centuries on the one hand, to countries that never possessed a single colony on the other. In between, we get countries with colonial acquisitions that were either short-lived or small in scale.

One way to get an idea of whether empires made economic sense or not is to add up the gains and subtract the costs (to the extent that this is possible given data constraints and the difficulty of disentangling the different factors at play). This is what we have done so far. Another way would be to compare countries that are far apart on the imperialism spectrum, but reasonably otherwise similar. Do the former enjoy a visible ‘empire bonus’?

If yes, how durable is it?

The short answer is that an empire bonus is hard to discern, and if it exists at all, it wears off quickly.

The UK and Switzerland were at opposite ends of the imperialism spectrum. The UK industrialised earlier, but Switzerland caught up towards the end of the nineteenth century and then overtook the UK in the early twentieth century. Today, in terms of GDP per capita (PPP), the UK is about in line with the Western European average, while Switzerland is about one-and-a-half times richer than that (Our World in Data/Maddison Project Database 2023).

France and Belgium were about on a par when France began to acquire colonies, but over the next half-century, the Belgian economy industrialised faster, so that by the time King Leopold II began his colonial project, Belgium was significantly richer than France. But they have not extended their lead any further since then. Today, both are about in line with the Western European average, and have been for decades.
Germany was on a par with the Western European average when it became a colonial empire and grew at roughly the same rate as the rest of Western Europe throughout its colonial period, neither visibly advancing nor falling behind in relative terms. Today, Germany is richer than the Western European average, but that is a much more recent, post–World War II phenomenon.

Spain and Portugal, the original, pre-industrial colonial empires, have been below the European average for as long as we have data.

Sweden, Denmark and Austria, which only had minor colonial possessions, are consistently above the Western European average today. While Sweden industrialised relatively recently, this was due to clearly identifiable domestic constraints (Norberg 2023: 3–8), not the lack of colonies or slavery profits. Once those constraints had been removed, Sweden industrialised as rapidly as any of its neighbours (ibid.: 8–20).

Japan used to be the major colonial power in Asia, but by Western standards, they were not especially rich during that period, and did not become so while they were a colonial empire. The association of Japan with high-tech industries is a much more recent development. Japan continues to be very rich by East Asian standards, but not uniquely so; Singapore, Hong Kong and Taiwan have surpassed them, and South Korea is on a par.

The best predictors of how rich or poor a country is today are economic policy and governance indicators such as the Economic Freedom Index and the Ease of ­Doing Business Index. This tells us a lot more than whether or not a country was involved in the slave trade, how many colonies it once possessed, or how long it held on to them.

The Impact on Colonies

The claim that the slave trade and colonialism made the Western world rich usually goes hand in hand with the claim that the slave trade and colonialism made the non-Western world poor. These two claims are not logically connected: one can believe one of them while rejecting the other. But in practice, they are companions.

This companion to the Marx–Williams thesis stands on much stronger ground than the thesis itself. There is empirical evidence of long-term scarring effects from imperialism and slavery, even though there is disagreement about the magnitude of the effect and the precise mechanism that explains it.

In a seminal paper published in the American Economic Review, Acemoglu et al. (2001) explore the idea of colonialism leading to ‘extractivist institutions’. They explain that (ibid.: 1375):

“…there were few constraints on state power in the nonsettler colonies. The colonial powers set up authoritarian and absolutist states with the purpose of solidifying their control and facilitating the extraction of resources.”

This still mattered after independence, because (ibid.: 1376):

“…the extractive institutions set up by the colonialists persisted long after the colonial regime ended. […] Setting up institutions that place restrictions on government power and enforce property rights is costly […] [W]hen the new elites inherit extractive institutions, they may not want to incur the costs of introducing better institutions, and may instead prefer to exploit the existing extractive institutions for their own benefits.”

They test this position empirically. Their basic idea is this: Europeans generally faced inhospitable conditions in the colonies because they were not used to a tropical or subtropical climate and disease environment. But this problem was much more severe in some places than others, as can be seen from the variation in mortality rates among European colonisers. Where the problem was least severe, colonisers were more willing to treat the place as a long-term investment opportunity and set up better institutions. Where it was most pronounced, colonisers acted in a short-termist way and set up institutions designed to make a ‘quick buck’.

Acemoglu et al.’s results show that, controlled for other factors, places that were once subject to short-termist colonialist extraction continue to have worse institutions today and are poorer as a result. The broad outlines of this result have been confirmed by similar, more recent studies (Kodila-Tedika et al. 2018).

In a paper in the Quarterly Journal of Economics, Nunn (2008) performs an equivalent empirical investigation on the long-term scarring effects of the slave trade. He explains (ibid.: 142):

“Africa’s slave trades […] had […] detrimental consequences, including social and ethnic fragmentation, political instability and a weakening of states, and the corruption of judicial institutions. The most common manner in which slaves were taken was through villages or states raiding one another […] As a result, ties between villages were weakened, which in turn impeded the formation of larger communities and broader ethnic identities. […] Because of this process, the slave trades may be an important factor explaining Africa’s high level of ethnic fractionalization today. This is significant for economic development.”

Drawing on shipping records and slave registries, he works out which parts of Africa were most affected by the slave trade and which were least affected. He finds that, controlled for other factors, the former tend to be poorer than the latter even today. The most important transmission mechanism seems to be that these places tend to be more socially fragmented, leading to worse political institutions.

None of this means that there is a deterministic relationship between a country’s colonial past and its present-day economic performance. An extractivist post-colonial institutional legacy is not a straitjacket that a country cannot break out of. But the legacy exists. A history of colonial extractivism, or a period of heavy involvement in the slave trade, made the subsequent development of good institutions less likely. The implication is that colonialism and slavery were not zero-sum games that benefited the colonisers at the expense of the colonised. It was more like a negative-sum game, which hurt the latter without really benefiting the former.


The ‘Great Awokening’ of recent years has led to a renewed focus on Britain’s imperial past. More specifically, it has revived a popularised version of the Marx–Williams thesis: the idea that the wealth of the Western world – and Britain in particular – was built on slavery and colonialism. This is presented as the original sin of capitalism, which it is still tainted with today and which still shapes the world we live in today.

Proponents of this view tend to simply assert this and treat it as ‘obviously’ true, rather than trying to substantiate it empirically. What we have tried to show in this book is that the empirical evidence is not kind to it.

The numbers are simply not big enough, and it is not even clear whether they have the correct sign in front of them. Colonialism and the slave trade made, at best, minor contributions to the West’s economic development, and they may well have been net lossmakers.

We cannot come up with exact numbers. We cannot say that the total combined contribution of the slave trade, the slave plantations and the Empire to Britain’s GDP was +3.4, +0.2, –4.1 or –5.9 per cent. But we can use the figures we have, and specify scenarios around them.

We know that the great bulk of Britain’s economic activity was domestic and that Britain traded more with other Western economies than with Asia, Africa, Latin America and the Caribbean. Neither the slave trade nor the plantations nor the colonies made a huge contribution to the British economy. Profits earned from overseas engagement were large enough to make some individuals very rich, but they were not large enough to seriously affect macroeconomic aggregates like Britain’s investment rate and capital formation.

There is evidence that empires boosted trade. But the same evidence also shows that they were far from the only factor. We cannot ascribe every economic transaction that took place within the Empire to the fact that there was an empire. At least some of them would have taken place anyway. To reach this conclusion, we do not need to assume that a world without the British Empire would have been a world of Cobdenite free trade and free enterprise (although that, in the author’s view, would have been the best of all worlds). We merely need to compare the Empire to other political arrangements that were common at the time.

It is only in the case of the slave trade that we can say with certainty that the political arrangement determined the economic arrangement: there can be no slave trade without slavery. Without slave labour, there would probably have been no large-scale sugar and tobacco plantations in the Caribbean.

We also know that the Empire, the slave trade and the plantations were propped up by the government, which is to say, they were implicitly subsidised, at a substantial fiscal cost, in the form of higher military and administrative spending. How high exactly that cost was, we cannot know, because we do not know what Britain’s military and administrative budget would have been without state-sanctioned overseas entanglements. But we do know that Britain’s tax burden was one of the highest in Europe, and that Britain’s military spending exceeded that of its European peers. It is safe to say that the Empire did not come cheap.

To find even modest positive effects, we have to make a series of debatable assumptions that are biased in favour of the Marx–Williams hypothesis. We have to assume that Britain’s administrative and defence expenditure was largely fixed, and that there is no huge cost difference between governing an island in the North Sea and governing a globe-spanning empire. We have to assume that the vast majority of the economic transactions that happened under the political structure of colonial rule happened because of that political structure, and could not have happened otherwise. And even then, we have to assume that the British economy was rigid and inflexible, unable to find substitutes for colonial imports or alternative uses for the resources it deployed in connection with the colonies. We have to assume that slave traders, plantation owners and colonial entrepreneurs were exceptionally frugal people who invested unusually high proportions of their profits.

If we do all that, we would arrive at a heavily watered-­down version of the Marx–Williams thesis. But we would still not come close to the magnitudes that would justify the rhetoric of a typical Guardian article on the subject.

This is one end of a spectrum of possible scenarios: we could call it the Marx–Williams end of the spectrum. At the opposite end is what we could call the Smith–Cobden end of the spectrum. It is possible that most of the ‘gains’ from colonialism were really just gains from overseas trade and investment, which could also have been realised under a different political structure. It is possible that, due to Britain’s imperial entanglements, military and administrative spending were much higher than they would otherwise have been. It is possible that a non-imperialist Britain could have enjoyed most of the gains from the Empire while avoiding most of its costs. If so, non-imperialist Britain would have been richer than the one we live in.

What is true for Britain applies a fortiori to the West as a whole. Other Western nations industrialised before they became colonial empires, or in some cases, industrialised without ever becoming so.

The only clear-cut exception we have found was the Belgian Congo, a highly unusual colony that was set up and run like a private for-profit company, which had to be self-funding because the Belgian state refused to subsidise it, and which was extremely resource rich.

If imperialism was, from a purely economic perspective, a bad investment – what is the implication for how we should view that legacy today, and what historic lessons should we draw from that period? What follows from that cost–benefit analysis?

The short answer is ‘not very much’.

The reader will have noticed that we have avoided promoting any specific narrative about Britain’s (or any other country’s) history or expressing a view of how that history should be collectively remembered today. A cost–benefit analysis cannot tell us any of that and is not supposed to.

The results we have presented here are compatible with a wide variety of interpretations of British history. It is perhaps inevitable that the way we remember important historic periods and events will, to some extent, reflect the political mood and the political priorities of the present. In a more optimistic age, people may have presented the Victorian era as a pioneering age characterised by breakthroughs in engineering, science and medicine. In a more patriotic age, people may have emphasised Britain’s positive achievements, presenting it as a force for good in the world. In the same vein, the resurgence of the Marx–­Williams thesis is the historic emphasis we would expect in an age of woke anti-capitalism.

Up to a point, this is all perfectly legitimate. Historic evidence offers scope for a huge variety of interpretations. But it also sets certain boundaries. At least in its stronger versions, the claim that slavery and imperialism made Britain rich is outside of those boundaries.

PDF Viewer

Fullscreen Mode

About the Author

Dr Kristian Niemietz is the IEA’s Editorial Director and Head of Political Economy. He studied economics at the Humboldt University Berlin and the University of Salamanca, graduating in 2007 as Diplom-Volkswirt (equivalent to an MSc in economics). In 2013, he completed a PhD in political economy at King’s College London.

He is the author of A New Understanding of Poverty (2011), Redefining the Poverty Debate (2012), Universal Healthcare Without the NHS (2016) and Socialism: The Failed Idea That Never Dies (2019).


Acemoglu, D., Johnson, S. and Robinson, J. (2001) The colonial origins of comparative development: an empirical investigation. American Economic Review 91(5): 1369–401.
Baumgart, W. (1992) Bismarcks Kolonialpolitik. In Bismarck und seine Zeit (ed. J. Kunish). Mainz: Johannes-Gutenberg-
Berg, M. and Hudson, P. (2023) Slavery, Capitalism and the Industrial Revolution. Cambridge: Polity Press.
BMA (2021) A missed opportunity. BMA response to the Race Report. British Medical Association (
BPB (2005) Deutschland in Afrika – Der Kolonialismus und seine Nachwirkungen. Bundeszentrale für Politische Bildung (
BPB (2015) Bismarck und der Kolonialismus. Bundeszentrale für politische Bildung (
Buelens, F. and Marysse, S. (2009) Returns on investments during the colonial era: the case of the Belgian Congo. Economic History Review (New Series) 62(S1): 135–66.
Cobden, R. (1835 [1903]) The Political Writings of Richard Cobden, with a preface by Lord Welby, introductions by Sir Louis Mallet, CB, and William Cullen Bryant, notes by F. W. Chesson and a bibliography, 2 vols. London: T. Fisher Unwin (https://oll
Darwin, J. (2008) After Tamerlane: The Rise and Fall of Global Empires, 1400–2000. London: Penguin Books.
Eltis, D. and Engerman, S. L. (2000) The importance of slavery and the slave trade to industrializing Britain. Journal of Economic History 60(1): 123–44.
Engerman, S. L. (1972) The slave trade and British capital formation in the eighteenth century: a comment on the Williams Thesis. Business History Review 46(4): 430–43.
ERIH (2023) On the industrial history of Belgium. European Route of Industrial Heritage (
Fieldhouse, D. K. (1986) Review: the economics of French Empire. Reviewed Work: Empire coloniale et Capitalisme française. Histoire d’un Divorce by Jacques Marseille. Journal of African History 27(1): 169–72.
Heblich, S., Redding, S. and Voth, H. (2022) Slavery and the British Industrial Revolution. NBER Working Paper 30451. Cambridge, MA: National Bureau of Economic Research (
Heblich, S., Redding, S. and Voth, H. (2023) Slavery and the British Industrial Revolution. CEPR/VoxEU column. Centre for Economic Policy Research (
Huillery, E. (2014) The black man’s burden: the cost of colonization of French West Africa. Journal of Economic History 74(1): 1–38.
Huxtable, S., Fowler, C., Kefalas, C. and Slocombe, E. (2020) Interim Report on the Connections between Colonialism and Properties now in the Care of the National Trust, Including Links with Historic Slavery. Swindon, Wiltshire: National Trust.
Kaufmann, E. (2022a) The Politics of the Culture Wars in Contemporary Britain. London: Policy Exchange.
Kaufmann, E. (2022b) The Political Culture of Young Britain. London: Policy Exchange.
Kennedy, P. (1989) Debate: the costs and benefits of British imperialism 1846–1914. Past & Present 1989(125): 186–92.
Kodila-Tedika, O., Tcheta-Bampa, A. and Lusenge-Ndungo, P. (2018) Colonial extraction and institution in Africa: empirical test of the hypothesis of Acemoglu–Johnson–Robinson. Social Science Research Network (
Kundnani, A. (2023) What Is Antiracism? And Why It Means Anticapitalism. Verso Books.
McCloskey, D. (2009) Slavery and Imperialism Did Not Enrich Europe. Munich Personal RePEc Archive. Draft chapter for McCloskey, D. (2011) Bourgeois Dignity: Why Economics Can’t Explain the Modern World. University of Chicago Press.
Mitchener, K. and Weidenmier, M. (2008) Trade and empire. Economic Journal 118(533): 1805–834.
Norberg, J. (2023) The Mirage of Swedish Socialism. The Economic History of a Welfare State. Vancouver: Fraser Institute.
Nunn, N. (2008) The long-term effects of Africa’s slave trades. Quarterly Journal of Economics 123(1): 139–76.
O’Brien, P. (1982) European economic development: the contribution of the periphery. Economic History Review 35(1): 1–18.
O’Brien, P. (1988) The costs and benefits of British imperialism 1846–1914. Past & Present 1988(120): 163–200. Oxford University Press.
Our World in data (2023) Trade as share of GDP, 1960 to 2020 (
Our World in Data/Maddison Project Database (2023) GDP per capita, 1815 to 2018 (
Parenti, M. (1995) Against Empire. San Francisco, CA: City Lights Books.
Rozado, D. and Goodwin, M. (2022) The increasing prominence of prejudice and social justice rhetoric in UK news media (
Sartre, J. P. (1961) Preface for The Wretched of the Earth by F. Fanon. New York: Grove Press.
Smith, A. (2005 [1776]) An Inquiry into the Nature and Causes of the Wealth of Nations (an electronic classic series publication). Pennsylvania State University.
Statista (2023) Einnahmen und Ausgaben deutscher Kolonien im Jahr 1912 (in Millionen Mark) (
US Department of Justice (2022) Former Minneapolis police officer Derek Chauvin sentenced to more than 20 years in prison for depriving George Floyd and a minor victim of their Constitutional Rights. Press Release, 7 July 2022. Department of Justice: Office of Public Affairs.
Williams, E. (2022 [1944]) Capitalism and Slavery. London: Penguin Modern Classics.
World Population Review (2023) Police Killings by Country 2023. Interactive data map (
Wright, G. (2020) Slavery and Anglo-American capitalism revisited. Economic History Review 73(2): 353–83.


  1. Hundreds demand justice in Minneapolis after police killing of George Floyd. The Guardian, 27 May 2020 (
  2. The Guardian, 31 May 2020 (
  3. Robert Milligan: Slave trader statue removed from outside London museum. BBC News, 9 June 2020 (
  4. Cecil Rhodes: Protesters demand Oxford statue removal. BBC News, 9 June 2020 (
  5. Equality, Diversity and Inclusion – an update. An update on the work of the University’s Equality & Diversity Committee and its Race Equality and ­Anti-Racist Sub-committee. University of Edinburgh press release, 15 September 2020 (
  6. William Gladstone: Liverpool students rename hall after anti-racism activist. BBC, 28 April 2021 ( side-56915021).
  7. Communist Party of Britain (CPB): Dorothy Kuya, n.d. (
  8. Tributes to slave traders and colonialists removed across UK. The Guardian, 29 January 2021 (
  9. I’m one of the Colston Four. Our victory confirms the power and value of protest. The Guardian, 6 January 2022 (
  10. Robert Milligan: Slave trader statue removed from outside London museum. BBC News, 9 June 2020 (
  11. The case for British slavery reparations can no longer be brushed aside. The Guardian, 9 July 2020 (
  12. Condemn communists’ cruelties, but capitalism has its own terrible record. The Guardian, 26 July 2018 (
  13. Why anti-racism must be anti-capitalist. Tribune, 23 January 2021 (
  14. If we can’t speak honestly about Britain’s links to slavery, we turn our backs on change. The Guardian, 30 March 2023 (
  15. Slave trade and the British economy. BBC Bitesize (n.d.) (
  16. Why woke is a useful word. Noah’s Newsletter, 3 January 2021 (
  17. There are two kinds of antiracism. Only one works, and it has nothing to do with ‘diversity training’. The Guardian, 29 June 2023 (https://www.the
  18. Who we are. GoFundMe (
  19. UK BLM: General FAQs (