Thirty one years after the end of the Apartheid system, the World Bank still rates South Africa to be the most unequal country on Earth, with a Gini Index of 0.63, four points ahead of the second-placed nation, Namibia (the Gini Index measures inequality from 0.00 to 1.00, where 0.00 represents all wealth being distributed equally among the population, and 1.00 represents a single person owning everything within a nation). It has been calculated that the richest 1% of the South African population own 55% of the country's wealth, while the top 10% own more than 85%. Thus wealth inequality within South Africa is equal to that of the entire globe.
Most modern research into inequality in South Africa concentrates on the post-Apartheid era, although it is generally accepted that the country's long history of racial discrimination and political repression is linked to the current state of inequality. However, the question of when this inequality first arose has not really been addressed by historians.
In a paper published in the South African Journal of Science on 26 November 2025, Johan Fourie of the Laboratory for the Economics of Africa’s Past at Stellenbosch University presents the results of a study in which he examined historical records from the Dutch Cape Colony of the seventeenth and eighteenth centuries and the British Cape Colony of the early nineteenth century, in order to attempt to understand the origins of the severe inequality seen in South Africa today.
Fourie notes that the Cape Colony was a deeply unequal place, even compared to modern South Africa. There were three distinct populations present, European settlers, enslaved people, and the indigenous Khoesan population (note, this is not always considered to be an acceptable name for this group, as it is linked to a history of persecution and discrimination, however, no term is universally deemed acceptable, and it is the term used by Fourie within his study). The economic inequality between these three groups, was predictably stark, but even within these groups inequality was very high.
Most societies in the developed world became steadily more equal in the decades after the Second World War, however, whilst this continued for the remainder of the twentieth century in continental Europe and Japan, the countries of the English-speaking word have become increasingly unequal since the 1970s, a pattern which has been repeated in India and China. Estimating inequality in pre-twentieth century societies is much harder, due to a paucity of records. Calculations have been made for the mid nineteenth century for the US and some western European countries, suggesting that they were more unequal than today. Studies looking at societies from before the nineteenth century are harder still, as census data simply did not exist in most places (the US undertook its first census in 1790, while the first census in England and Wales was taken in 1801). In Africa, such data is generally absent well into the twentieth century.
The Cape Colony provides a rare opportunity to examine wealth distribution in a pre-industrial society, as it was administered by colonial officials from first the Dutch East India Company and then the British Empire. who kept meticulous records of financial transactions, albeit largely with the aim of extracting wealth from the colony. Furthermore, Cape Town has never been affected by any major military conflict, something which as led to the destruction of records in many other parts of the world.
Map of the Dutch Cape Colony in 1795, from George McCall Theal (1816). History of Africa south of the Zambesi - from the settlement of the Portuguese at Sofala in September 1505 to the conquest of the Cape Colony by the British in September 1795. G. Allen & Unwin Ltd, London. Internet Archive. Previous studies of inequality in South Africa have largely concentrated on the twentieth century, and the entrenching of disparities between racial groups by the Apartheid system, and associated political exclusion, social stratification, and control of the labour market. However, this system only entrenched divisions which were already present.
This concentration on twentieth century inequality has led to two visions of inequality in South Africa emerging, a 'Liberal' view which sees inequality as having arisen as a result of European conquest, with people becoming impoverished as a result of being dispossessed from their traditional lands, and a 'Radical' view, which views poverty as a result of a prolonged struggle over land, labour and markets. However, the debates between these two camps have generally been based around the way in which the problems should be viewed, with neither side concentrating particularly on the measurement of inequality.
The Cape Colony was founded by the Dutch East India Company (Vereenigde Oostindische Compagnie) in 1652, as a resupply station for ships on their way to Asia. It subsequently grew into a settler colony with a large agricultural economy, where at least a portion of the population enjoyed a standard of living comparable to that of Europe. This colony subsequently expanded into the African interior, with an expanding agricultural economy built upon the labour of enslaved and Khoesan workers, and producing grain, wine, and livestock products.
Landing of Van Riebeeck at the Cape of Good Hope by Charles Davidson Bell (painted in about 1850). Wikimedia Commons. By 1825 the Cape Colony comprised 11 districts, with more than 10 420 settler households. At the abolition of slavery in the colony in 1834, a census found over 37 000 enslaved individuals. Records of Khoesan labourers are less clear, though this appears to have increased steadily over time.
Fourie was able to access a variety of records to facilitate his study. The most important of these were annual tax censuses, which were kept by both the Dutch East India Company and British colonial officials, and many of which have been transcribed in the past decade by a team of historians at the Cape Archives. Another source of information was probate inventories (records of people's assets at the time of their death, and how these were divided amongst their heirs), which were collected by the Master of the Orphan Chamber, a public official esponsible for administering estates when individuals died intestate, left heirs under 25 or unmarried or had heirs who were untraceable. A third source of information was the Slave Emancipation Dataset, which provided detailed information about the ownership of slaves at the time of emancipation. This survey was carried out by officials appointed from London, who toured the colony documenting who owned slaves, how many slaves they owned, and estimating the value of those slaves.
Fourie began by calculating annual Gini coefficients for the district of Stellenbosch-Drakenstein based upon annual censuses taken between 1685 and 1844. This dataset comprised 142 219 individual household records, with an average of 995 households each year. To calculate relative wealth, six assets were examined, Horses, Cattle, Sheep, wheat reaped, vines and wine.
Three trends became immediately obvious from this dataset. The first was that wealth distribution in the Cape Colony was remarkably uneven, even when only free households were included. Second is that this applies to all assets, with no asset producing a Gini coefficient less than 0.5 at any point in the record. The third was that whole there were some periods of volatility in the records, these largely reflect points at which the data collecting method was changed, rather than political events.
Fourier then combined all six asset datasets into a single bucket, using values for each item derived from the Orphan Chamber dataset. This again produced a record of remarkably uneven wealth distribution, with the Gini coefficient remaining between 0.6 and 0.75 throughout the eighteenth century, then rising further in the nineteenth century (although this was lower than for the most extreme asset types).
Looking at the share of wealth held by the wealthiest portion of the population, Fourier found that the top 1% of the population held an average of 12% of the total wealth over the eighteenth century at, but that this increased over time, with their holding on average 18% of the wealth over the period 1800-1830. This was true for smaller wealth divisions too, with the wealthiest 0.1% of the population holding an average of 2% of the total wealth in the eighteenth century and 3.7% of the total in the period 1800-1830, while the top 0.01% owned an average of 0.5% of the total wealth through the eighteenth century, and an average of 0.8% of the total wealth in the period 1800-1830. Fourier notes that because these figures reflect only agricultural assets, that they cannot be directly compared to modern statistics, but that they do demonstrate evidence for early wealth inequality.
Fourier also calculated a Gini coefficient for slave ownership, which again was very high, consistently remaining above 0.7. Adding slaves into the general basket, however, reduced the average slightly, which Fourier suggests may reflect many slaves being involved in non-taxable activities, such as domestic servitude.
However, slaves are not just assets, they are people, and members of the community. Since most slaves could be assumed to have no assets (this is not always strictly the case), Fourier first tried representing each male slave as the head of a household with zero assets. This skewed wealth distribution even further, reaching 0.92 by the end of the eighteenth century. Since slaves do have to be provided for in some way, Fourier next repeated the experiment with the assumption that slave households had resources equivalent to six sheep per year. This produced a Gini coefficient with varied between 0.89 and 0.91 over the course of the eighteenth century, comparable for the figures calculated for the slave-owning communities of the Caribbean. Thus the Cape Colony would have been one of the most unequal societies of the early modern world.
Because these records are only from a single district, Stellenbosch-Drakenstein, there remains the possibility that this district was uniquely uneven in its wealth distribution. Records were not available for all districts throughout the period, but Fourier was able to assemble a roughly coeval record set by using 1825 for the Cape District, Stellenbosch, Graaff-Reinet, Swellendam, Albany, Beaufort, George and Uitenhage; 1824 for Clanwilliam and Worcester; and 1823 for Cradock. Using this data, he created Gini coefficients for Horses, Oxen, Cattle, Sheep, wheat and wine. This record included 10 420 individual households.
The Gini coefficient remained above 0.5 for all agricultural measures in all districts, with the single exception that Albany District does not produce any wine. In all other districts, wine produced a Gini coefficient of 0.8 or higher. Distributions were most even for Horses and Oxen. Fourier suggests that Stellenbosch was at the upper end of wealth distribution inequality for two reasons; high wine production and a large number of slaves.
'Mossel Bay on the Indian Ocean' by Robert Cocking, 1818. British Library/Wikimedia Commons. Next Fourier looked at information derived from probate records. Potentially, this could be used to examine household possessions, something absent from the taxation records, though Fourier instead chose to remain with agricultural assets, as this allowed easier comparison with records derived from other sources. Probate records were available for all districts for the entire period, with little variation in how records were made.
Using these records, Fourier examined five agricultural assets, Cattle, Horses, Sheep, ploughs, and wagons, for the eighteenth and early nineteenth centuries. This produced similar results to the taxation records, consistently showing a Gini coefficient of greater than 0.6 for all assets, and from 1770 onwards, a Gini coefficient greater than 0.7, indicating a high rate of wealth inequality among settlers from the earliest days of the colony.
The slave emancipation report records the value of each slave owned at the time of abolition, plus the compensation paid to each slave-owner, with a total of 36 417 individual records. All records come from the year 1834. Gigi coefficients constructed for slave ownership by district showed a high level of inequality, but not as great as for other assets, ranging from 0.29 for Somerset and Uitenhage to a 0.47 for Worcester. Gini coefficients for compensation for the loss of slaves ranged from 0.45 in Somerset to 0.55 in the Cape District.
However, these figures only show inequality among slave-owning colonists, with less than 30% of the settler population at the time actually owning slaves. Once the 13 033 non-slave-owning settler households were added to the equation, the Gini coefficients shifted considerably, giving coefficients for slave-ownership between 0.49 in Stellenbosch (where slave-ownership was comon) and 0.75 in Beaufort (where slave-ownership was unusual). Coefficients for compensation were even higher, ranging from 0.83 in Somerset to 0.92 in the Cape District.
However, European settlers and their slaves were never the only people present in the Cape Colony. By the early nineteenth century a number of Khoe settlements had been incorporated into the colonial administrative system, with some, notably mission stations, having produced documentation which is now being transcribed and analysed by modern scholars. To try to represent this population, Fourier looked at three such settlements, Bethelsdorp, Swartrivier, and Groenekloof, concentrating on records from 1825.
Bethelsdorp, near Algao Bay, was founded in 1804 by the London Missionary Society, and became an important settlement where the Khoe people were able to gain some protection and autonomy within the colonial system, with a population of 425 households by 1814. However, the settlement suffered from poil soil quality, and a variety of other economic problems.
The Swartrivier settlement in Swellendam District grew more organically, attracting pastoralist Khoe who worked as herders of labourers for European farmers, often under exploitative conditions. By 1825 Swartrivier was home to 364 households.
Groenekloof, located on the coast 55 km to the north of Cape Town, was founded by the Moravian Missionary Society in 1808, and became a haven for freed slaves and Khoesan, By 1825 the settlement comprised 108 households, practising subsistence farming, or working as labourers on settler-owned farms.
Tax records were collected for these settlements in the same way as they were for settler communities, although the number of variables per household was generally lower, probably as a reflection of these households owning fewer assets. In analysing these households, Fourier looked at five assets, Cattle, Horses, Oxen, Sheep and wheat reaped.
Creating Gini coefficients for these assets revealed severe inequality within all three settlements. Only a single asset type (Cattle at Bethelsdorp) produced a coefficient lower than 0.75, and that was 0.72. The distribution of many assets approached near perfect inequality (i.e. the situation where one person owns all of the asset). For example, only ten families owned sheep in Bethelsdorp and only one family owned Sheep in Groenekloof. Thus, inequality among the Khoe was typically more severe than among European settlers. Including this group with the general population pushed inequality even further, due to the greater wealth typically held by settler households. This led to a Gini coefficient for Cattle ownership of 0.92, and measures for Horses, Oxen and wheat all in excess of 0.77.
Fourier's findings reflect severe inequality (generally interpreted as a Gini coefficient of greater than 0.6) for all assets, and both within and between communities, throughout the history of the Cape Colony. Such inequality had clearly arisen long before the beginning of the twentieth century. While comparing historic income and wealth to their modern equivalents is notoriously problematic, within-period comparisons can determine the distribution of assets, and therefore how equal or unequal societies were.
See also...
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)