Slavery, racial inequality, and education

Evidence on human capital accumulation by race

The economic literature on racial inequality in education in the US is extensive [2], [3]. The evidence documents the evolution of racial differences, both in the quality and the quantity of education. Following the American Civil War, African Americans had essentially no exposure to formal schooling. This was a legacy of the extremely high rates of illiteracy that prevailed in the southern plantations, where the vast majority of African Americans were employed. The next generations of the descendants from African slaves, initially still located mostly in the south, were able to complete far fewer years of schooling than white people. They also had access to racially segregated public schools, which provided a qualitatively inferior education in comparison to what was available to white people [4]. By 1940, as a result of the combination of low educational attainment and inferior educational quality, the racial gap was still very wide. Subsequently, however, as successive generations of black children received more and better schooling, the racial gap narrowed, with an eventual impact on earnings equalization.

Persistence of racial education inequality

Overall, the evidence on the evolution of racial educational differences points to long-term convergence, but also to the persistence of disparities along racial lines [4]. This takes into account a wide number of dimensions, such as literacy rates, years of educational attainment, spending per pupil and returns to literacy.

Figure 1 shows the evolution of the racial gap, at the high school and the bachelor level, from 1940 to 2000. For each year, the histograms represent the ratio between the share of white and black people, in the corresponding populations, holding at least a high school or a bachelor degree. Full racial equality would correspond to a value of one on the vertical axis. However, starting for instance from the first histogram, it reveals that in 1940 the share of white people with a high school education is between three and four times the corresponding share of black people.

Overall, the figure shows that over the 1940–2000 period, white people are on average more educated than black people at both levels. The gap is even larger at the bachelor level. Over time, the gap decreases for both levels, but at a lower rate for the bachelor level. The decline of the gap is much more marked up to 1980, while progress has been much slower since then. Despite its gradual reduction, the gap is still present until the end of the period: in 2000 the ratio is still above one for high school degree and almost two for the bachelor degree. These trends confirm the persistence of deeply-rooted racial disparities, despite the presence of a convergence process.

Slavery and racial education inequality

The institution of slavery is highly likely to be a factor conditioning race relations and, in particular, racial inequality along multiple dimensions, including human capital and education. The historiography of slavery in the US is also extensive. In the aftermath of the American Civil War the legacies of slavery determined extremely high rates of illiteracy among black people. The obstacles subsequently encountered by black children in acquiring education are widely believed to represent the channel through which inequalities were perpetuated.

The correlation between the racial educational gap in 1940 and the share of slaves over population in 1860 is extremely high: 0.90 and 0.81 at the high school and bachelor level, respectively. For the state-level data (Illustration), empirical estimates of the determinants of racial educational inequality during the period 1940–2000 confirm a causal effect of historical slavery. This effect continues through from its influence on the initial gap, both at the high school and bachelor level [4]. In other words, those states that relied more intensively on the use of slave labor in 1860 still exhibited larger contemporary racial disparities in education in 2000, through a channel that is represented by the initial gap in attainment.

The transmission of human capital is therefore the link between slavery and current racial gaps in schooling attainment, which confirms that the origins of educational inequality are indeed very deeply rooted in the history of this country [1], [4].

Channels of transmission

What is driving the correlation between slavery and education is the local nature of schooling provision and funding, in conjunction with the political exclusion of the descendants of African slaves, at least up to the 1960s. Indeed in the former slave states in the US black people have been persistently disenfranchised, both through de jure and de facto methods.

The property tax is the principal source of local funding and therefore for public educational support. Under the “Separate but Equal” educational policies applied in southern states until the 1960s, local officers could divert state funding for black people in order to subsidize education for white people. As a result, they could impose a lower property tax and spend less on the overall education budget. Even after legally mandated school segregation was abolished, the counties that had previously chosen low education expenditures adhered to a lower tax rate, with negative effects on public school funding and, in particular, on education for black people.

Empirical evidence supports the relevance of political mechanisms for the determination of education outcomes. For example, the disenfranchisement provisions enacted following the abolition of slavery in the late 19th century in the southern states are associated with a fall in black educational inputs and with low-quality southern schooling [5]. Within the state of Mississippi, which had one of the highest proportions of slaves, there is also a significant and positive effect of black political participation in the late 19th century on contemporary black educational outcomes [6].

Implications for income distribution

The influence of slavery on education disparities has even broader implications, since through its impact on the racial gap in human capital accumulation it, in turn, affects income inequality between races and ultimately overall income inequality. In other words, the distribution of per capita income is more unequal today in counties associated in the past with a larger proportion of slaves in the population [1]. The positive correlation between racial education inequality and income inequality is illustrated in Figure 2, which is based on cross-county census data from 2000.

Evaluation of policies

The key Supreme Court decision to overturn the “Separate but Equal” doctrine that allowed racial segregation in US schools (Brown v. Board of Education, in 1954), initiated a long process involving subsequent rulings as well as federal legislation. Since then, education policy has always been at the heart of government intervention aimed at reducing racial disparities.

In the desegregation years between the 1960s and the 1970s, substantial gains were made, with a tapering in subsequent decades. The declared goal of recent reform programs for the US schooling system has been the removal of the persistent and pervasive racial and ethnic educational gaps. This is the case both for “No Child Left Behind,” which was the federal education program enacted in 2002 under President Bush, and for “Race to the Top,” the current program created in 2010 by President Obama. Both programs have relied on the introduction of standards and assessments of students and teachers based on test scores, with a focus on special support for the worst-performing schools and on special initiatives directed at developing pre-schooling and vocational education. The emphasis put by both administrations on these programs is consistent with the evidence presented earlier. This evidence shows how deeply educational inequality is rooted in the history of the country. It still remains to be seen whether these policies can fulfill their goals and expedite the process of convergence to full equality.

Brazil

Brazil was the destination of almost half of the African slaves who were shipped across the Atlantic. African slaves sent to Brazil outnumbered those sent to the US by a ratio of about ten to one. Brazil was also the last country in the Americas to abolish slavery, which it did in 1888.

However, in part because of the custom of inter-racial marriage, slavery in Brazil never resulted in the forms of segregation that have been observed in the US. Empirical evidence on this country has uncovered a negative correlation between past slavery (proxied by measures of land inequality, which are strongly correlated with the presence of crops suitable for the use of slave labor) and quantitative and qualitative measures of contemporary education (namely, the share of the population between 15 and 17 years of age attending secondary education in 2000, and a 2005 index of school quality provided by the government) [7].

A parallel investigation, which focused on the period 1889–1930, has shown that Brazilian states with fewer slaves were able to exploit positive trade shocks and invest the resulting export tax revenues in elementary education expenditures, thus achieving higher literacy rates. An opposite, negative effect on education is observed instead, under the same circumstances, in states with more slaves. The resulting distribution of human capital across states in Brazil persists to the present day [8].

However, contrasting evidence emerges from a county-level analysis of the state of São Paulo, which is the largest state in the country. In São Paulo, the intensity of slavery shows no discernable long-term impact on a variety of current education indicators [9].

Colombia

In order to investigate the impact of slavery on long-term development in Colombia, it is possible to exploit the variation in the presence of gold mines in different municipalities, since gold mining was strongly associated with demand for slave labor. The empirical findings show that the historical presence of slavery is associated with reduced measures of various contemporary education indicators and, more generally, increased poverty and land inequality. Even though the question of racial education inequality is again not explicitly addressed, it can be observed that districts more heavily associated with slavery have higher shares of population of African descent. This suggests negative implications for the influence of slavery on the racial dimension of education inequality [10].

Puerto Rico

Even though this Caribbean country did experience African slavery, the available evidence concerns the impact of the libreta system, which was an alternative form of labor coercion introduced in 1849 following the 1845 agreement between Spain and Britain to enforce the abolition of the slave trade. This de facto substitute of slavery remained in place until 1874. Using the response to regional variation in coffee prices and demand for unskilled labor, these findings indicate that in coffee-growing regions, during the coercive period, higher coffee prices had no effect on literacy rates, whereas after the abolition of coercion higher coffee prices reduced literacy rates.

These results suggest that the abolition of forced labor reduced the incentive to accumulate human capital. However, despite appearing puzzling at first glance, they are not inconsistent with the intuition that coercive labor institutions should exert a negative influence on long-term human capital accumulation, since a side-effect of abolition was an increase in the relative wages of unskilled laborers [11].

Peru

Focusing on a segment of modern-day southern Peru that transects the Andean range, it is possible to gauge the impact of yet another form of labor coercion. The mining mita system, which was an extensive forced mining labor system, was instituted by the Spanish government in 1573 and abolished in 1812. It is worth noting however that, rather than involving black slaves, the mita system was applied to the indigenous population. Analysis has shown that the mita’s impact on education is fading through time. While a negative impact can be observed in earlier periods, by the 2000s there is no remaining evidence of a negative influence, across a range of education indicators [12].