In developed and developing countries alike, officials, academics, and educators are concerned about the widespread problem of failing and poorly funded public schools. The problem is not unique to developing countries such as India and Brazil, but extends to developed countries such as the US, suggesting income differences alone cannot be responsible for the poor quality of public schools worldwide. Recent academic research suggests weak teachers and ineffective instruction are the main culprits. While teachers’ unions are blamed for the poor performance of US schools and students (Coulson 2010), high rates of teacher absenteeism and crumbling infrastructure are the hallmarks of the poor quality and quantity of public schooling in developing countries (Chaudhury et al. 2006, Alcazar et al. 2006).
A popular solution in policy circles is to decentralise the provision of public schools to lower levels of government. In this view, decentralisation improves outcomes because local constituents can hold elected officials responsible for public service delivery and enables a tighter fit between local preferences and public schooling. This idea dates back to Oates’ (1972) classic work on decentralisation and is backed by case studies of the US and other developed countries. A recent article in The Economist (2012) cites Florida’s unique success at school reform in contrast to the federal “No Child Left Behind” law. Several articles attribute the early success of US public education to successful decentralisation of education to states and local school boards (Goldin and Katz 2008). Policymakers and academics working for multilateral organisations have translated these US success stories into recipes for developing countries, perhaps optimistically hoping that decentralisation can solve the problems of bureaucratic failure, low public funding, and the poor quality of public schools (e.g. USAID 2005, Florestal and Cooper 1997, and Behrman et al. 2002).
We, however, are cautious in extrapolating from the positive historical experience of the US to developing countries today. Scholars of the political economy of decentralisation have long been wary of decentralised policymaking because it can lead to elite capture (e.g. Bardhan and Mookherjee 2006). Indeed, our study of the origins of primary education in Brazil, Russia, India, and China (the BRICs) in the late 19th and early 20th centuries suggests decentralisation is unlikely to improve outcomes unless local populations can hold elected leaders or influential elites accountable (see Chaudhary et al. forthcoming). In the absence of local measures to reduce elite capture, central governments may need to intervene to improve the accountability of local governments. While the BRICs enjoy recognition today for their common status as fast-growing emerging economies, their relevance for studying the historical origins of public primary education lies in their level of comparative economic and political underdevelopment, the large scale and variability of their politics and geography, and the availability of previously unexplored data on basic schooling. We also view the BRIC experience as representative of the majority of the world in the early 20th century that had limited publically provided schooling.
In the early 20th century, the BRICs accounted for more than 50% of population (1910), but only 12%, 20%, 8%, and 4% of school-age children in Brazil, Russia, India, and China, respectively, were enrolled in primary school in 1910, compared to more than 80% in the United States. The relative poverty of the BRICs may have negatively influenced the demand for schools, but income alone cannot account for the differences in enrolment. Figure 1 plots enrolment rates against GDP per capita for several countries in 1910. Primary enrolment rates in the BRICs are below what we would expect, given their income levels. All four countries lie below the fitted line in the graph, suggesting that factors other than income were also at play.
Figure 1. BRIC primary school enrollments in comparative perspective, c.1910
Source: Chaudhary et al. (forthcoming).
Although Brazil, Russia, India, and China had different political structures by the early 20th century, each country had some form of “elite democracy” that did not allow for the broad representation of the population in policymaking bodies. The absolutism of Qing China gave way to elected bodies after the 1911 revolution, but these highly circumscribed bodies gave little voice to the non-elites. British India had no substantive elections before 1919. Colonial officials chaired provincial and local councils, although nominally elected and appointed local members could provide feedback to the councils. In Russia, most provinces had quasi-representative local governments (the zemstvo) after 1864, and there was an elected national assembly after 1906, but the nobility continued to hold de jure and de facto sway in both institutions. After 1889, Brazil held national and state elections with a restricted and easily co-opted electorate dominated by different elite groups.
Across the four countries, central governments largely absolved themselves from the direct provision of public education and substantial autonomy was devolved, formally or by default, to local governments. In post-1889 Brazil, a strong form of federalism entailed the devolution of fiscal authority – especially export tax revenues – to provinces, which contributed to extreme inter-regional disparities in spending and outcomes. The creation of the zemstvo in much of European Russia transferred property taxes and related spending authority to local governments. Although central colonial authority was strong in British India, even here the provision of public primary schools was decentralised to local councils, albeit without any substantial tax authority. In China, decentralisation was more accidental than policy driven. Political instability in the late Qing and the Republic allowed local actors to take military or political power relative to the central authorities.
In such decentralised polities with limited democracy, elites were able to capture local governments and influence education policy. In the face of low incomes, most elites were unwilling to support strong redistributive programmes to increase the quantity and quality of public schools. But variation in local socio-economic conditions, along with significant differences among local elites, influenced the willingness to fund public schools across space and time.
The devolution of political and fiscal authority in China in the late-Qing and Republican eras was aided by the collapse of central authority, and increasing social and military conflict. Although this prompted a shift to modern forms of public primary education, elite control of school policies combined with limited fiscal resources were key stumbling blocks. More public schools were opened in commercially developed areas where traditional elites were replaced by newer elites with ties to modern industry. Such new elites in China were less constrained by central authorities and more pro-schooling because of their gains from a more educated labour force, unlike Indian local elites, who were perhaps more anti-public schooling because they perceived fewer direct political or economic benefits. Indian elites, defined by both caste and economic status, successfully lobbied the Colonial Government to divert public funds away from primary schools. Decentralisation was ineffective because of both low public spending and the high degree of caste and religious diversity within Indian society. Such divisions combined with the social hierarchy of Indian society enabled local elites to co-opt the policymaking process.
Brazil and Russia – marginally richer and possessing broader forms of elite democracy – saw greater investments in public primary schooling than India and China. In Brazil, fiscal federalism and literacy restrictions on voting after 1891 increased elite support for education in states and municipalities with high export tax revenues or where the elite required greater voter turnout to maintain political power. Until the 1900s, variation in the support for mass public education in Russia was driven by differences in who controlled local political institutions. The zemstvo offered some channels for the expression of broad popular interest in schooling, but elite control of these local institutions meant that it was generally those districts with more liberal nobility that invested more in public education. The founding of a national assembly in 1906, with some popular representation, coincided with more funding for basic education from the central government, but it took the Soviet Union’s centralised policies (and increased funding) to push the country towards universal primary education.
Decentralisation in the face of weak or absent democratic mechanisms led to local elite capture of political institutions, compounding the constraints of low income, high opportunity costs, and limited public funds in the BRICs. Little money was spent on mass schooling, as elites often attended private schools and saw little need to engage in redistributive policies that may have challenged their political and economic control (as also described for Latin America by Engerman and Sokoloff 2002). Democratically driven central government policies and investments may have helped overcome low incomes and local constraints (Gallego 2010), as was perhaps evident following the (slightly) representative national elections in Tsarist Russia after 1906. But as the Soviet example, or even the cases of Meiji Japan and 19th century Prussia, would suggest, strong but relatively undemocratic central initiatives can force backward countries towards universal primary schooling. Our focus is on historical public spending and the extension of mass schooling, unlike the concerns today which revolve around quality. That said, similar issues of local elite capture could also influence the quality of public schools. Policymakers in developing countries should heed the historical experience of BRICs in designing effective decentralisation policies today.
Alcázar, Lorena F, Halsey Rogers, Nazmul Chaudhury, Jeffrey Hammer, Michael Kremer and Karthik Muralidharan (2006), “Why are teachers absent? Probing service delivery in Peruvian primary schools”, International Journal of Educational Research 45(3):117-136.
Bardhan, Pranab and Dilip Mookherjee (2006), Decentralization and Local Governance in Developing Countries, MIT Press.
Behrman, Jere R, Anil Deolalikar, and Lee-Ying Soon (2002), “Promoting Effective Schooling through Education Decentralization in Bangladesh, Indonesia and Philippines”, Asian Development Bank, ERD Working Paper No. 23.
Chaudhury, Nazmul, Jeffrey Hammer, Michael Kremer, Karthik Muralidharan and Halsey Rogers (2006), “Missing in Action: Teacher and Health Worker Absence in Developing Countries”, Journal of Economic Perspectives, 20(1):91-116.
Chaudhary, Latika, Aldo Musacchio, Steven Nafziger, and Se Yan (forthcoming), “Big BRICs, Weak Foundations: The Beginning of Public Elementary Education in Brazil, Russia, India, and China”, Explorations in Economic History.
Coulson, Andrew J (2010), “The Effects of Teacher Unions on American Education”, Cato Journal 30(1):155-170.
The Economist (2012), “The Floridian school of thought”, 25 February.
Engerman, Stanley L, and Kenneth L Sokoloff (2002), "Factor Endowments, Inequality, and Paths of Development among New World Economies", Economia 3:41-102.
Florestal, Ketleen and Robb Cooper (1997), “Directions in Development: Decentralization of Education”, Legal Issues, The World Bank.
Gallego, Francisco A (2010), “Historical Origins of Schooling: The Role of Democracy and Political Decentralization”, Review of Economics and Statistics 92(2):228-243.
Goldin, Claudia and Lawrence Katz (2008), The Race between Education and Technology, Harvard University Press.
Oates, Wallace E (1972), Fiscal Federalism. Harcourt Brace Jovanovich.
USAID (2005), ‘’Decentralization in Education”, EQ Review 3, No. 4.