Where Cotton's King and Men are Chattel

Was slavery necessary for Western economic development?

“fun” fact but this is the only statue of a slave in Buenos Aires, Argentina

… I will state the case as I laid it out in my mind. 'I am,' I thought, 'not only the slave of Master Thomas, but I am the slave of society at large. Society at large has bound itself, in form and in fact, to assist Master Thomas in robbing me of my rightful liberty, and of the just reward of my labor…

Frederick Douglass, “Life and Times of Frederick Douglass”, 1892

Yesterday, a tweet started discourse about slavery - in particular, that it wasn’t necessary for the Industrial Revolution. I weighed in, and it started even more discourse, so now instead of writing about South Korean gender politics, I get to write (“get” and not “have to” because we’re using self-help language here) about slavery.

I will take it as a given that slavery was bad - I mean the slaves didn’t like it, and that’s enough for me. But how, exactly, was it bad, and what is its relationship with the Industrial Revolution and economic development writ large?

How science got dismal

The field of economics has a long and complicated relationship with economics, but also one that’s weirdly proud: the reason the discipline is known as “the dismal science” isn’t because economists are notorious party poopers (which we are), or because of Thomas Malthus like those slightly more enlightened assume, but rather due to a dispute between early economists and the philosopher Thomas Carlyle. In particular, Carlyle objected to the economists considering all humans “the same” and thus objecting to slavery, based on the principles of supply and demand. Carlyle wrote that political economy (as economics was called back then) was:

… the social science -- not a "gay science," but a rueful -- which finds the secret of this universe in "supply and demand," and reduces the duty of human governors to that of letting men alone, is also wonderful. Not a "gay science," I should say, like some we have heard of; no, a dreary, desolate and, indeed, quite abject and distressing one; what we might call, by way of eminence, the dismal science.

Thomas Caryle, “Occasional Discourse on the Negro Question“ (1849)

Besides offense at the title of the essay and chuckles at calling economics “a gay science”, this more or less sets the stage for the line of thinking that econ 101 brain would get you: if all people are similar enough on ability, on average, then enslaving some of them to do barely paid work is bad economic praxis, since they are not allowed to (like free men) specialize and maximize the marginal value of their labor. Thus, slavery is bad economically because it is unprofitable for the slaves (who get paid less than they’re due in basically all cases), for society at large (since the loss of slaves’ potential contributions harms individuals), and even for the slavers themselves -since they aren’t getting the most they can get from their slaves, basically, and would receive higher earnings by giving them proper training, education, and pay.

Well of course, this rests on the assumption that all human beings are equal in ability and value, which should be a lot less controversial than it currently is, and which wasn’t back at the time. Using the tools of “scientific racism” (fake pseudoscience), thinkers aligned with Carlyle’s position argued that Black people were, in fact, inferior in ability to Whites, and thus would not be economically beneficial - in fact, by giving them roof and work (and also the tenets of Christianity), the slavers were doing their slaves a favor, since slavery was not profitable for them.

So basically, at the heart of the “economics of slavery” distinction, there is the central question of profit: if slavery was unprofitable, then this “paternalistic case” could be true, but slavery was a dead weight on economic performance. However, if slavery was profitable, then slavers were just being rational - but, to quote The Good Place, “that’s worse. You do get how that’s worse, right?”.

Well, in a seminal paper from 1958, Alfred Conrad and John B Mayer found that slavery was, in fact, very profitable for the slavers. The problem facing economists at the time was, basically, the lack of data, and the available data was mostly related to slave prices, so most research on that question used those (particularly, the meaning of “depreciation”, and other such morbidities). Basically, the authors decide to approach the question from the slavers’ point of view: slaves are both intermediate goods for producing cotton and tobacco, and capital goods for producing more slaves. The question, then, is whether the return of capital for Southern plantation owners in the United States was higher, lower, or the same, for slaves or for other types of investments. Basically, efficiency concerns about the cost of keeping slaves doesn’t consider that a high cost of owning a slave also means a low price for selling a slave, which is another possible use of slave labor. The impressive part of the paper isn’t really its conclusion, but that it tallied comprehensive data sources and built a coherent mode, but the conclusion is itself striking: slavery was very profitable, was a sustainably profitable system, but it did need a continued expansion of slavery into the American Southwest.

After this paper, an extremely protracted argument about basically everything in its contents ensued¹, and the paper also was linked with some economists’ political commitments to the Civil Rights Movement - in large part, Conrad and Meyer were responding to economist Ulrich Phillips, a man so racist he changed his name from Ulysses out of Grant Derangement Syndrome. Regardless, the contribution is that, with slaves being a capital good, slavery itself being less profitable would mean that slave prices went down, rather than fewer slaves be used. As a form of capital, slave prices responded efficiently to broader market considerations; the price of slaves at the New Orleans market closely tracked how well the Civil War was going for the Confederacy. Overall, further studies find similar results, particularly that slavery could have been moderately unproductive as an agricultural technique, but that it was also extremely profitable because of the amount of coercion it could extract from labor.

A large part of the controversy relied on whether plantations produced their own food and at what cost, but basically, if the slave is (again) understood as capital and not labor, then it would make sense that in the low season for cotton, slaves produced other things, which they consumed themselves later. But given that the average female slave had at least seven children, it doesn’t make sense for pressures to restrain population growth. In fact, separating slave families (which could have been economically efficient if there was a “surplus”, aka overcapitalization) seemed to be deleterious for the slaves and thus reflected in their prices at auction. And while we’re on a gender note, we can also report that apparently women made a large share of slaveowners, contrary to popular belief and particularly for enslaved women, and that they had ample decisionmaking abilities over their slaves. #Feminism #Representation

So, in sum, slavery was, indeed, capitalist: without getting into a semantic discussion of what “capitalism” means, the economic system of slavery was characterized by the search for profits and by profit-maximization writ large. Per the 1958 paper, if slaves are understood not as workers but as capital, the capitalist nature of slavery becomes clear. However, Conrad and Meyer finished off on this note:

The available productive surplus from slavery might have been used for economic development (…). In spite of this good omen for development, Southern investment and industrialization lagged. It is hard to explain this except on the social ground that entrepreneurship could not take root in the South (…). Furthermore, the American experience clearly suggests that slavery is not, from the strict economic standpoint, a deterrent to industrial development…

To the extent, moreover, that profitability is a necessary condition for the continuation of a private business institution in a free-enterprise society, slavery was not untenable in the ante bellum American South. Indeed ,economic forces often may work toward the continuation of a slave system, so that the elimination of slavery may depend upon the adoption of harsh political measures

Away down South in the land of cotton…

Well, why was the South economically behind the rest of the United States in industrialization? Not that this undercuts the argument about the (British) Industrial Revolution, which we’ll get to in a bit. It is part of that massive argument about “cliometrics” that I mentioned earlier, but roughly speaking, it was because of slavery.

Because slavery was so profitable, and because slaves were capital goods and not labor, then it stands to reason that investment decisions would have to track the relative profitability of the plantation economy versus the industrial economy. Well, like in most modern middle income countries, there was extremely limited transfer of capital from the agroexporting to the manufacturing sector. In particular, since slaves were capital for their owners, a large share of Southern capital was in the slave sector - which, because slave plantations were highly profitable and slavery as an institution was safeguarded by political institutions, then there was no need to save in non-slave wealth, which would have been open for investing in manufacturing. But contrary to what this entails, the Southern economy grew quickly: partially because of technological innovations such as the cotton gin (which improved the productivity of cotton plantation substantially), although such innovations were relatively uncommon in the South, for the previously mentioned capital-related reasons. Southern growth, thus, was largely driven by extensive increases in the land used for plantation agriculture, which raised agricultural yields by shifting cotton production to more productive lands.This would have entailed (over the long run) an expansion of slavery to more and more land in search for higher returns - meaning that curtailing the expansion of slavery was, in fact, a thing that the South would have militantly fought against and rendering the Civil War a likely event.

But this does mean that it’s, in fact, plausible that slavery-originated capital would have in fact financed industrialization - the claim being made about Britain that prompted all of this! The case is also made here (i.e. last Friday’s post), and is known as the “Eric Williams Thesis”, based on his book Capitalism and Slavery (1944)². The books case is, mainly, that profits from the slave trade were a large and irreplaceable part of industrial finance, ditto for the wealth of Caribbean planters, and that the Atlantic Economy was crucial for Britain’s development. This is an extremely long-running debate in economic history (even longer than the slavery one, in fact); the role of slave wealth and slave income is the “most wrong” part, and was revised nearly immediately, but the part about the Atlantic Trade, which included the slave and plantation trade, is mostly right with a lot of room for research and interpretation. A recent paper tackles this issue as well, finding that exogenous increases to slave wealth and slavery-linked income, caused by weather shocks in the New World, might have reduced agricultural employment and increased manufacturing employment back in Britain, particularly in cotton mills - meaning that, to some degree, we can find evidence that profits from slavery at least indirectly financed the Industrial Revolution in Britain, especially in the cotton trade and the textile sector.

However, there is another part of the puzzle: the recent Nobel Laureates. One of their papers tackles precisely this issue - the impact of the Atlantic trade on European development - though their preferred lens, institutions. To sum this lens up, Acemoglu, Johnson, and Robinson believe that institutions (the rules of play of the economy and society) determine political and economic structures, particularly by an elite making concessions (or not) on either in order to secure their power. If institutions are conducive to private economic activity, they’re called inclusive, and largely consist of secure property rights and pluralistic democracy. Meanwhile, bad institutions, where the government can take your stuff and you have no recourse, are called extractive institutions. Medieval Europe had extractive institutions, but the story goes, the growing economic power of European merchant classes (from intra-European and later transatlantic) trade gave them political and economic power in order to demand political rights, which they gained in events such as the Glorious Revolution. The Acemoglu, Johnson, and Robinson study finds (using bad data and flawed methods) that the rise in the Atlantic Trade was stronly correlated with urban development, and thus had a causal impact on European growth. So, in a perverse way, the supremely extractive institution of slavery caused the rise of inclusive institutions in Europe, particularly England.

Of course, this opens the question of whether slavery was an extractive institution or not. Engerman & Sokoloff (2002) thesis: that slavery promoted unrepresentative political institutions, which caused underdevelopment. In places with appropriate natural resources or weather, wealth extraction in plantations and mines using slavery and the slavery adjacent mita and encomienda systems was incentivized. Since (as stated before) labor was cheap and coercion was profitable, then the widespread use of forced labor led to significant wealth inequality within societies, resulting in an economically dominant elite holding all of the power over post-colonial institutions - and, therefore, lower investment in public goods and services. Looking at the Colombian gold mining economy, we find a clear example of this dynamic: the Spanish colonial elite maintained its power, and refused to provide public services, resulting in worse outcomes for gold mining districts than similar non-mining districts over the long run. However, the opposite results can be found in the Peruvian silver mining economy: the concentrated wealth of mining districts also meant a more concentrated political elite, which was therefore more capable of extracting concessions for public services from the Spanish crown. And further research questions whether the effect is driven by slavery and coerced labor at all, instead of being driven by the role of slavery on wealth inequality and by the impact of wealth inequality on the political process.

So this is a very time-and-place sensitive process, which means that some care is needed to evaluate how slavery shaped the growth of American capitalism. One such (recent and influential) paper finds that, at the border between slave and free states, the price of land was cheaper on the slaver side. This would imply that farmers might have preferred to work and own on the slave side, since land would be cheap; however, even with higher wages on the slave side, these farmers refused to move to the slave side, perhaps showing some kind of opposition to it. However, this paper suffers from some issues. Basically, land prices were correlated with population density more than slavery (meaning that the farmers refused to move to bad land, not to slave land), and were significantly more correlated with population density than slavery. In fact, it is very likely that slavery helped population density relative to no-slavery, not that it hurt it - though slavery did have a negative effect on urbanization in the South (and therefore in economic development), mainly through its negative effect on industrialization.

In this sense, slavery did not impede the growth of American capitalism, but rather, boosted it: after Emancipation (i.e. the abolition of slavery), the value of land in former slave territory plummetted. This was due to the fact that slavery provided cheap labor to farmers, who could offset the negative factors on their land value with coercion before, but not after, slavery, at which point land prices fell in slave states but not comparable land on free states. This advantage from slave labor quickly spread into the rest of the cotton economy: if the American South had been forced to pay market wages to the slaves, Southern cotton would not have been competitive on the global market. Cotton, therefore, became the main export commodity of the United States, and the largest source of revenue for the Federal Government - which allowed for the expansion of the border to the West, and which pumped liquidity into the capital markets. Industrialization up North was also aided by the textile industry and cheaper cotton than in the counterfactual, no-slavery scenario.

… old times there are not forgotten

In a nutshell, therefore, slavery both did and didn’t finance industrialization in the US and Britain, and did and didn’t (negatively) affect institutional quality. So what happened after slavery was abolished?

By far the most ridiculous economics-infused argument about slavery is that Emancipation represented a violation of the slaveowners’ property rights, which resulted in lower economic confidence. Besides the fact that this is, self-evidently, insane, there’s also the fact that dissolving an extractive institution can result in long-term benefits: when Henry VIII dissolved the British monasteries and took their land (they found out he played rough), he was also de facto abolishing the exploitative feudal land tenure relationships. This is because lords were forced to free their serfs after the Black Death, but the Church did not - until Henry VIII. So he, inadvertently, also freed England’s last remaining serfs, who over time became the British gentry who later invested in industry (and also in the Glorious Revolution).

Well, that’s fine and interesting, but what were the effects of slavery on economic performance? Firstly, the slaves became free citizens, which was good. According to one study, Emancipation did decrease the economic output of the cotton economy, but also drastically increased the economic productivity of the broader economy over a longer span of time. This was caused by the fact that allowing for slaves to leave their relatively unproductive employment in the cotton sector reduced economic misallocation enormously - while as capital the slaves were valuable and productive, as labor they were not, at least not to the same margin. The estimates on the impact on productivity from this reallocation range from 4% to 35%, or around 7 to 60 years of economic advancements.

Not everything was fine and dandy for the freed slaves, though. After Emancipation, it took their descendants two generations to catch up in literacy, education, and employment to free Black people outside the South, and over the next 160 years, Black wealth grew significantly more slowly than White wealth, given the vastly different initial conditions under slavery - and vastly different accumulation trajectories, marked by discrimination and poverty.

In particular, after Emancipation, slaves were not in fact entirely free - many were stuck in the same occupation under their former owners, just as “free labor” (often underpaid and overworked). Lacking assets, and education, and skills to relocate, plus the political violence of the Confederate South, the freed Black population was ill-suited to move forward. Many went on to work on other sectors of the economy, but a large number remained in argiculture - where landowners refused to sell their land to former slaves, particularly in cotton-growing regions: only 20% of farm owners in areas specialized in cotton were Black, 35% in areas that could grow it but didn’t exclusively, and 56% in areas suited for general agriculture.

Given the large dependence on the former Confederate economy on cotton (roughly 30% of farms were locked-in on cotton), a large number of Black workers simply moved on to work on farms, compared to White workers, who had a good shot of owning their own farm. For White and Black workers, two thirds of the disparity in posterior farm ownership is explained by race. Additionally, in 1892, a plague known as the boll weevil attacked cotton crops, forcing divestment into other, general purpose food crops - which was associated with large relative gains in the socioeconomic outcomes of Black farmers born after the weevil’s arrival, since their parents were broken out of the exploitative cotton economy by the pest. Contrarily, the former slaveowners largely saw their wealth bounce back: the former Southern planter elite saw a large decline in its wealth with Emancipation, due to the loss of slave assets and lower land prices, and this loss was positively associated with the share of slavery-linked assets, compared to similarly wealthy pre-war families. Nonetheless, the children of former slaveowners recovered relative to comparable sons, and their grandchildren surpassed them - thanks to social networks primarily, with intermarriage in the ex-slaveholding elite playing a crucial role.

While remaining in the South as farmers was extremely economically harmful, as mentioned, many did leave - and saw significantly improved standing. After the Great Mississippi Flood of 1927, a large number of Black agricultural workers moved out to urban areas, which led to landowners to both resist their out-migration (which would have forced them to bid up wages for the remaining labor), and to invest in mechanization of agricultural production. This helped spur economic convergence between the South and the North, following decades of post-bellum stagnation. However, this resulted in a large increase in redidential segregation due to the fact that Black homeowners largely moved next to each other, and White homeowners moved out - resulting in higher wages for newcomers but lower property values, and reduced work opportunities over time.

Additionally, slavery continued impacting political institutions after its abolition. The Federal government (briefly) attempted to shore up racial equality with political and economic redistribution (known broadly as Reconstruction), which largely failed due to activism on behalf of White Southerners a lack of political will to see it through. Part of this involved the building of an institution known as the Freedmen’s Bureau, which aimed to help slaves and launch institutional reforms. The Bureau, in the short term, managed to improve political relations and reduce vote share for the Democrats (in the South back then, the racist party), as well as reduce wealth inequality. Similarly, the boll weevil infestation also managed to reduce political repression - in the 1892 to 1922 period, Southern society became less violent and less repressive, which endured over the backlash - violence was lower, KKK activity was lower, non-white electoral participation was higher, and Civil Rights protests were less common. This was mediated, again, through Black out-migration, which was able to pressure White society into providing certain political concessions - another Acemoglu, Johnson, Robinson type bargain. However, after a few years (particularly after Reconstruction finished falling through) a violent backlash started which continued through the 20th century: Democrat vote share went up substantially, as did racial violence in various forms, and higher Ku Klux Klan activity - as well as higher wealth inequality and lower socioeconomic mobility. Violent racial incidents as a whole would obviously have negative economic effects, and especially violent and destructive ones like the Tulsa Race Massacre of 1921 have been found to have substantial impacts.

Having the economic effects (bad) and the political effects (good for 30 years, then bad for basically the remainder of the 20th century), we can move onto culture. Was American culture impacted by the mores of the former Confederacy? Well, yes. Looking at the diaspora of former Confederates out of the South, a study find that the newcomes promoted the construction of Confederate memorials, exacerbated racial violence, and boosted forms of exclusion previously uncommon in those areas, drastically reshaping racial outcomes in those areas. Similarly, movies and entertainment originated from the Confederacy promoted racist values and racial violence, and swelled the ranks of the Ku Klux Klan across the United States. Similarly to Black migration into cities, White urban patterns also shifted: Whites responded to Black in-migration by leaving cities, and desegregation lowered housing prices due to similar factors. This second migration of Whites resulted, again, in racially and religiously conservative values spreading to their new homes.

One final effect is on Black culture: Black first and last names are generally distinctive because of the role of slavery, Emancipation, and the slave trade, and Black people in the United States are still impacted in their mindset by it - by exhibiting a higher level of zero-sum values, and thus supporting government redistribution, and opposing immigration; these beliefs, directly and not just through their impact on policy, affect economic growth. Similarly, slavery has been found to have an impact on regions of Africa with a higher share of slave raids, reducing trust in government and others; it would not be unreasonable to assume that the direct victims of these raids also showed similar belief systems. One final point is female labor force participation: Black women have had, since the 1880s, lower labor force participation than White women, even after controlling for possible confounding factors; the difference seems largely racial, and particularly attributed to socialization and social norms linked to slavery. More recent and credible research back this up, and point to slaver practices still shaping customs and beliefs to these days.

Conclusion

Trying to reach a definitive conclusion on a thorny, fraught area of research like slavery seems like folly; however, based on the arguments cited above, it seems like slavery was harmful to economic, political, and cultural institutions, and while it did contribute to Western development, it does not seem to be a determining factor. Either way, the past is the past, and the lessons to be learned from slavery are clear: on the virtues of civil government and of action against non-civil institutions, as well as a rejection of hollow, exclusionary projects - as applied to the present.

1. I don’t want to get into the Time On The Cross of it all, but overall, the book was very influential and was largely interested in centering slaves on the narrative about slavery, but it was also criticized by economists and posterior cliometricians (“economic history statisticians”) even if its influence is undeniable.

2. As a weird fact on the side, Williams was also Prime Minister of Trinidad and Tobago from 1962 to 1981 (time of his death) and is considered the nation’s founding father.