Improving Basic Education—The Governance Challenge
Improving Basic Education—The Governance Challenge
Abstract and Keywords
Building on political settlements analysis, this chapter lays out the multi-level framework used throughout the book to explore how political and institutional context influence the governance of basic education in South Africa at national, provincial, and school levels. It reviews the literature on the influence of politics on bureaucratic performance, and on the potential and limits of school-level participatory, horizontal governance. It details why South Africa’s institutional arrangements for education provide an ideal natural experiment for exploring the influence of divergent contexts on the operation and performance of subnational bureaucracies—and also the potential of horizontal governance as an institutional complement in settings where bureaucracies work well, and as an institutional substitute in settings where educational bureaucracies are dysfunctional.
The world over, economic inclusion has risen to the top of the development discourse. A well-performing education system is central to achieving inclusive development—but the challenge of improving educational outcomes has proven to be unexpectedly difficult. Access to education has increased, but quality remains low, with weaknesses in governance comprising an important part of the explanation.
‘Achieve universal primary education’ was adopted unanimously at the United Nations in 2000 as the second of the Millennium Development Goals. Major gains have been achieved. As of 2015, 91 per cent of children in the relevant age cohort in developing countries were enrolled in primary schools, up from 83 per cent in 2000. But 2008 results from the Trends in International Math and Science Study (TIMSS) found that more than half of developing-country students who took the test scored below the ‘low’ threshold benchmark set for the test. Of the 250 million children worldwide who cannot read, write, or add, more than half are in school (Unesco, 2014).
The reasons for the difficulties in improving learning outcomes are many (Sniltsveit et al., 2016). The 2018 World Development Report, Learning to Realize Education’s Promise, distinguishes usefully between the proximate and underlying causes of learning shortfalls. The proximate causes include the difficult socio-economic context in which many children live, with the result that many children arrive at school without being ready to learn; the lack of resources to provide teachers, facilities, or schoolbooks; and shortfalls of teaching and managerial skills, and of teacher training. Underlying these proximate (p.4) causes are shortfalls in the effectiveness with which the human, financial, and physical resources available for educating children are used effectively. This leads to a consideration of governance, and its political determinants.
A burgeoning literature, synthesized by Kingdon et al. (2014) and Hickey and Hossain (2018), identifies a multiplicity of distinct political explanations for lagging quality. These include:
• Incentives of teachers (and teachers’ unions) which are not necessarily consistent with the goal of enhancing quality.1
• The mediating role of political and bureaucratic institutions, which can undercut the incentives for politicians and public officials to give priority to improving educational quality.2 One especially relevant political mechanism comprises the use of patronage to sustain political support; this generates strong incentives to expand access, hire more teachers, and build more buildings—but weak incentives to confront inefficiencies within the system.3
• Relatively limited demand for skills emanating from employers and the labour market, especially in low-income countries.4
• Limitations in the ability of parents to effectively demand quality education.
This proliferation of political explanations is a mixed blessing. As Kingdon et al. (2014: 46–7) put it:
The literature on the political economy of education is under-developed in geographic scope, robustness of methods utilized, and theoretical richness…The array of theoretical frameworks developed since the emergence of the ‘new institutionalism’ of the 1980s has not been well utilized…Future work needs to develop more nuanced political theories about change and particularly about how alternative structural, historical and institutional conditions determine varied possibilities and constraints.
The need is thus not for more studies which show in general terms that ‘politics matters’ for educational outcomes, but rather for in-depth work that delineates some specific causal mechanisms through which specific political drivers exert (p.5) their influence—and how better insight into these drivers can support better informed and more effective policymaking and implementation.
This book aims to advance the state of the art of political analyses of education policymaking and implementation by exploring in depth a specific policy-related question: What should be the balance between hierarchical and horizontal institutional arrangements for the public provision of basic education? The analysis uses the ‘new institutionalism’ (defined later in this chapter) as a key theoretical basis, is anchored empirically in a single country and, within that on the public provision of basic education. Choosing South Africa as the case study country anchors the more abstract and academic reflections on governance in a setting, and a sector, where achieving real gains is of vital concern.
Though the focus of the book is narrow, its goals are ambitious. A first goal is to contribute, academically and practically, to the crucial, ongoing challenge of improving educational outcomes in South Africa. A second is to use the South African case to contribute from a fresh perspective to the global literature on some aspects of education sector governance. A third goal is to contribute more broadly from a political and institutional perspective to the analysis of public bureaucracies, and of participatory approaches to service provision. A fourth goal is to illustrate how research approaches which put politics and institutions at centre stage can be applied at a micro level in a way which sheds light on specific practical challenges of policymaking and implementation.
1.2 Horizontal versus Hierarchical Governance
Schools can come in many forms: public schools, private schools, religious schools, secular schools; and hybrids that combine the above: charter schools, not-for-profit schools run by private foundations, voucher-financed schools.5 The research presented in this book focuses on public schools—that is, schools financed by the public sector, and operated under public auspices. In most countries, public schools remain the predominant mode for the provision of education (including in South Africa where, as of 2015, close to 95 per cent of schoolchildren were in public schools).
In discussion of public school systems, a major controversy worldwide concerns the appropriate balance between hierarchical (top-down, bureaucratic) (p.6) governance, and more ‘horizontal’ approaches, which delegate significant resources and responsibility to internal and community stakeholders at the school-level. There are (at least) two distinct variants to this controversy. One question, which is not a focus of this book, concerns where bureaucratic control should be located—nationally, subnationally, or at local government level. Here the focus is on the related but different question of the extent to which governance responsibility should be assigned to the school-level itself.
Building on the experience of some countries with strong and effective centralized systems of education (e.g. France, Russia, Japan), some practitioners argue that education should be tightly managed hierarchically—with strong, top-down control of recruitment, promotion, curriculum and the content (almost to the level of individual lessons) of classroom-level instruction. But others argue for greater flexibility at the school level, allowing for quick identification of localized problems, and development of appropriate context-specific solutions (Pritchett, 2013; Sayed, 2002; Lauglo, McLean, and Bray, 1985; Prawda, 1993; Alexander, 2000).
Certainly, a better performing public hierarchy is more desirable than a weakly performing one: The allocation of scarce public funds across the system, the assignment of personnel to the places where they are most needed, building the capabilities of the teachers and other employees who work within the system, monitoring and managing the results achieved by staff, the construction and management of infrastructure, and the provision of furniture, textbooks, and other teaching materials are quintessentially bureaucratic tasks. A school system will surely work better when they are done well than when they are done badly.
What determines whether or not a bureaucracy works well or works badly? Insofar as a bureaucracy works well, might more governance flexibility at the school-level nonetheless be a useful complement, adding to the overall effectiveness of the system? And insofar as the bureaucracy works badly, can school-level governance be at least a partial substitute, sustaining at least some performance-orientation? As will become evident, the answers to each of these questions depends in important part on context.
1.2.1 Bureaucracies are Embedded in Politics
Since at least the 1980s, discourse on the performance of public bureaucracies has been predominantly managerialist in orientation. The result, in education and more broadly, was a worldwide wave of reforms (including in South Africa) under the rubric of the ‘New Public Management’. But, as Pollitt and Bouckaert (2000), World Bank (2004), Grindle (2013), Levy (2014), and Yanguas (2017) explore in depth, this managerial framing ignores the ways in which bureaucracies are embedded in politics.
(p.7) The World Bank’s (2004) World Development Report framed the relationship between politics and bureaucracy as a ‘long route’ of accountability—a hierarchical chain linking citizens (as principals) to politicians, politicians (as principals) to policymakers, and policymakers (as principals) to the bureaucracy. The long route of accountability is dauntingly complex. For the system to work, each of the links in the chain needs to be robust; such robustness can be achieved only under very distinctive political conditions.
The interface between politics and bureaucracy can undercut the efficacy of the long route of accountability in at least three ways. First, political principals set the goals which bureaucracies should pursue—and politicians are more successful in some settings than in others in reconciling their multiple competing interests and objectives in ways that provide clarity of purpose to public officials (Wilson, 1989). Second, political principals (plus other arm’s-length check and balance institutions) are charged with the task of overseeing bureaucracies, and holding them accountable for performance—and, again, there is substantial variation across settings in the extent to which they take on this responsibility. Third, and of particular relevance to the education sector, the ways in which political leaders cultivate and sustain political support may run counter to the kinds of actions needed to improve educational outcomes.
Consider patronage and clientelism—the use by political leaders of public resources and positions as mechanisms for rewarding friends, punishing opponents and, more broadly, sustaining political support. One key consequence of a patronage orientation is to give strong support to expansion of access to education (with all of its associated promises to voters, new jobs in the sector, and new procurement contracts)—but in practice to give low priority to improving quality (for which gains become evident only over the longer term, with many of the required measures likely to meet resistance, insofar as they involve holding public employees and others to account).
Numerous analyses of the politics of education have highlighted how patronage and clientelism shape sectoral decision-making. Grindle (2004) details the incestuous links between teachers’ unions and the education bureaucracy in a half-dozen Latin American countries. Kingdon et al. (2014) summarize the voluminous literature on the ways in which patronage-related political considerations have shaped decision-making in education, including Sharma’s (2009) analysis of discretionary and patronage-based appointment and transfer processes in India; Kingdon and Muzammil (2003, 2009) and Kingdon and Teal’s (2010) exploration of the role of teachers and teachers’ unions in shaping education policymaking and implementation in that country; and also (in broader, comparative empirical studies) Hoxby (1996) and Beteille (2009).
(p.8) Chapters 3–7 of this study analyse comparatively how the political-bureaucratic interface works, and what its effects are on performance in South Africa’s education sector, at the national level (Chapter 3) and in South Africa’s Western Cape and Eastern Cape provinces (Chapters 4–7). The two provinces operate under similar formal rules, and have similar levels of access to fiscal resources—but differ widely from one another in their modes of operation, their political contexts, and their performance. Building on our analysis of how provincial political dynamics influence the bureaucracy, we explore also what kinds of policy approaches might improve educational outcomes, even in difficult contexts.
1.2.2 Horizontal Governance—its Potential and Limitations
Turning to horizontal governance, Bruns, Filmer, and Patrinos (2011) and Westhorp et al. (2014) review the literature on the potential and limitations of a variety of horizontal governance mechanisms. Bruns et al. focus on the results of methodologically rigorous randomized control trials and ex-post impact evaluations of bottom-up, transparency and participation initiatives to strengthen bottom-up accountability. They report a mixed picture. Studies show gains from participatory initiatives in El Salvador, some parts of India, Kenya and Pakistan, but multiple studies (including some conducted in India) show no positive effect. Pritchett (2013) takes a more positive view, with a detailed depiction of how a networked education ecosystem might provide an improved platform for performance.
What might be some reasons for this uneven pattern? In exploring this question, it is useful to begin by laying out some theoretical arguments as to how horizontal governance—delegation of significant authority and thus flexibility to internal and community stakeholders at the school level—might improve educational outcomes. One set of arguments is intrinsic to the educational task—and thus potentially applies both in settings where the bureaucracy works well, and in settings where it works less well. The second is more directly relevant to settings where the bureaucracy works less well.
To begin with the intrinsic arguments, Wilson (1989), Israel (1987), and World Bank (2004) distinguish among sectors according to the heterogeneity and monitorability of their production activities. Top-down hierarchical governance, they argue, is most effective where production can be standardized, and where the monitorability of outputs and/or outcomes is straightforward. By contrast, where production is more heterogeneous, and outputs/outcomes are less readily monitorable, more flexibility needs to be accorded to front-line production units, with a correspondingly greater role for horizontal (‘principal-principal’/peer-to-peer) governance arrangements.
• By creating scope to customize provision in ways which are responsive to the local context. (In the context of education, this concerns questions of approaches to pedagogy, and adaptation of curriculum—questions which, it is important to note, fall wholly outside the scope of the present study.6)
• By improving motivation—with the ‘zone of autonomy’ at the service provision front-line hypothesized to provide the opportunity for internal leaders to motivate their teams effectively, including by fostering an environment of continuing learning on the part of staff as well as students. This is a classic argument for improving the effectiveness of ‘street-level’ bureaucracies (including schools) which operate at a distance from organizational hierarchies (Wilson, 1989; Lipsky, 2010; Carpenter, 2001).
• By creating scope for the utilization of local-level information of a kind to which higher-level hierarchical authorities lack access—and thereby enhancing processes for the selection of good quality staff and leaders, and the efficacy of efforts to hold staff and leaders accountable for their performance. Sah and Stiglitz (1986), North (1990), and Aghion and Tirole (1997) explore the informational dimensions of organizational governance and decentralization in general terms.
Note that, although all three of the above propositions potentially are relevant regardless of whether or not hierarchies perform well, a necessary condition for their efficacy is that the rules of the game set by higher levels provide some local-level autonomy.
The more context-specific argument highlights the role that horizontal governance potentially might play as a partial institutional substitute for relatively weak bureaucracies—enabling some schools to operate as relative ‘islands of effectiveness’ within a broader sea of dysfunction. The key channel here is the empowerment of developmentally oriented local stakeholders (including professionally committed teaching staff) to hold school staff accountable for making their best effort. As Levy (2014) explores, a necessary condition for ‘empowerment’ to be effective as a means of strengthening accountability is that developmentally oriented stakeholders indeed have sufficient influence to be able to ‘trump’ predatory actors seeking to capture school-level resources (teaching and administrative positions, contracts, other discretionary resources) for private purposes.
(p.10) The question as to whether parents and other school-level stakeholders have the potential to hold schools accountable for performance has been the subject of much controversy. Based on their review of the literature, Kingdon et al. (2014) are sceptical: they report multiple studies which show that local power relations disempower poor parents in ways that inhibit them from having an impact—from Indonesia (Chen, 2011), to Ghana (Essuman and Akyeampong, 2011), to Honduras and Guatemala (Altschuler, 2013) and India (Rawal and Kingdon, 2010). A quote from Kingdon et al. (2014: 2) provides a flavour of the criticism:
while decentralization [to school-level] is a widely advocated reform, many of its supposed benefits do not accrue in practice because in poor rural areas the local elite closes up the spaces for wider community representation and participation in school affairs.
But other studies—undertaken in El Salvador (Jimenez and Sawada, 1999, 2003), India (Pandey, Goyal, and Sundararaman, 2011), Ghana (ESID, 2016), and Indonesia (Pradhan et al., 2014)—report a more positive impact. The mixed evidence lends itself to the interpretation that the relative power of ‘developmental’ and ‘predatory’ actors at the local level is indeed key to the effectiveness of horizontal governance. Chapters 8 and 9 of this book explore in detail how these school-level interactions play out in practice in the two South African provinces.
1.3 South Africa as a Case Study
Over the past three decades, one response in many countries to weaknesses in the performance of education bureaucracies has been to try and shift responsibility for delivering (public) education from national to subnational and school levels.7 As Grindle (2004) details for Latin America, the reform of historically centralized systems often has been difficult to achieve politically; reforms have been scattershot and uneven.
In contrast to Latin America, the 1996 South Africa Schools Act (promulgated two years after the advent of democracy, at a moment of unusual political cohesion) put in place a carefully designed set of institutional arrangements which located very substantial responsibility for delivering basic education at the provincial level, and gave quite broad authority, (p.11) including for recruitment, to school governing bodies in which parents comprised a majority. (Chapter 2 reviews in detail the trajectory of education sector reform in South Africa.) This combination of strong delegation to provincial and school levels, and large differences in provincial-level political dynamics creates an ideal opportunity for exploring comparatively the influence of politics on bureaucratic performance—and on the relative efficacy of hierarchical and horizontal governance in public schools.
Taking a long-run view, transforming the system of education is one of the most crucial tasks confronting South Africa’s democracy. Notwithstanding the democratic ‘miracle’ of 1994, the legacy of apartheid left the country saddled with among the highest levels of inequality in the world. The longer-term sustainability of democracy depends on moving the economy in a more inclusive direction—and this, in turn, depends on a transformation of basic education from a system which (as Chapter 2 details) was geared to the reproduction of subservience on the part of the majority of the country’s population, to one that builds skills, citizenship, and enhanced economic prospects for all.
As a middle-income country with a clear commitment to redirect policy in a pro-poor direction, South Africa had available relatively abundant fiscal resources for improving education. The country’s legacy of economic, social, and political dualism meant that (alongside a dysfunctional educational system for the majority) it also had in place the knowledge, institutions, and track record of a relatively high-performing public education system—albeit one that historically had been targeted only towards elites). And the country’s sophisticated bureaucracy seemingly had the capability to effectively use the available money and capacity to address the challenge of providing a quality education for all.
By the early 1990s, South Africa already had achieved near universal enrolment of the relevant age cohort in primary education. The secondary school enrolment rate rose from 51 per cent in 1985 to 91 per cent by 2007. However, a consequence of South Africa’s apartheid legacy was that only a very small minority of the country’s students have had access to a quality education. Overall, South African students scored lowest among twenty-six low-middle and middle-income countries for which Pritchett (2013) reports comparable data from international standardized tests.
As Chapter 2 details, in the years immediately following its first democratic election in 1994, South Africa embarked on an extraordinarily far-reaching transformation of the governance, fiscal, and curricular arrangements of its education system. Yet for all that the new arrangements enabled the country to distance itself from the unequal arrangements inherited from apartheid, it has proved extraordinarily difficult to reverse the inequality in educational outcomes. Some recent efforts to identify the causes of these continuing (p.12) difficulties have identified governance as a key contributory factor.8 A major research project to identify binding constraints, synthesized in Van den Berg, Spaull, Wills, Gustafsson, and Kotze (2016), concludes that the root causes of South Africa’s low educational outcomes generally fall into one of two categories: ‘a lack of accountability (including weak institutional functionality and undue union influence), and a lack of capacity (including weak teacher knowledge and pedagogical skill, wasted learning time and insufficient opportunity to learn’). Gustafsson and Taylor (2016), using an innovative empirical strategy, quantify the impact on performance of differences in implementation capacity at the provincial level. Zengele (2013) hones in on how teachers’ unions influence recruitment in ways which are inconsistent with a commitment to quality. This study complements these findings by drilling down into some specific mechanisms of policymaking, implementation by the bureaucracy, and school-level governance, and their political influences.
Three sets of features of the institutional arrangements put in place by the 1996 South Africa Schools Act are especially relevant for the present research. First, responsibility for policymaking, for resourcing the system, and for setting the overall regulatory framework was retained at the national level. The latter included the negotiation and promulgation of an elaborate system for performance management. How politics shaped the eventual form of this system is the subject of Chapter 3.
Second, responsibility for implementation was delegated to the country’s nine provinces, which differed substantially from one another both politically and institutionally. This makes for a natural experiment, where (with policy, regulation, and financing automatically controlled for) the focus can be on the impact of politics and institutions on the operation of the provincial education bureaucracies responsible for policy implementation. Differences in politics and institutions are especially stark between the Western Cape and Eastern Cape provinces, providing a strong platform for comparative analysis of the interactions between politics, institutions, and how bureaucracies operate—and the effect on educational outcomes. Chapters 4 and 5 explore in detail why each of the two provinces’ bureaucracies perform the way they do; Chapter 6 examines econometrically the effect of differential bureaucratic performance on educational outcomes; Chapter 7 builds on the case studies to explore comparatively and conceptually how background political and institutional contexts shape the performance of public bureaucracies.
Third, the new institutional arrangements assigned substantial school-level responsibilities (including the recruitment of the school principal and senior teachers) to school governing bodies (SGBs) in which parents are required to (p.13) be in the majority. This provides an excellent opportunity for exploring the interactions between education hierarchies, and more ‘horizontal’ school-level governance arrangements in the divergent Western Cape and Eastern Cape contexts—with the research in both provinces focusing on lower-income communities with limited experience of school-level governance during the apartheid era. Chapters 8 and 9 use in-depth case studies to explore school-level governance dynamics, with a particular focus on how the interplay between developmental and predatory stakeholders affects performance. Integrating the school-level findings and the analysis of educational bureaucracies, Chapter 10 lays out an overarching ‘all for education’ approach to working at scale, so that the benefits of quality education can be made available to all of South Africa’s citizens.
1.4 Analytical Framework
This section lays out the framework used in this book to analyse the relationship between context and the governance of public education—including (but not only) the relative merits of hierarchical and horizontal approaches to governance. The framework builds on an ongoing movement in the development discourse over the past decade—away from technocratic visions of ‘best practice’, and towards a focus on more ‘good fit’ approaches to policymaking and implementation.
One task for ‘good fit’ analysis is to delineate a practical framework for distinguishing among different contexts. There is a burgeoning literature which explores how political context affects educational outcomes. One set of contributions (Ansell, 2008; Bourguignon and Verdier, 2005; Hicken and Simmons, 2008; Stasavage, 2005) focuses on regime type, specifically whether democratic or authoritarian setting lead to better educational outcomes.
Recent work has moved beyond ‘regime type’ to explore more broadly how a variety of different types of country-level ‘political settlements’ influence development policymaking, implementation, and outcomes. Contributions include Khan’s (2010) pioneering analysis, and North, Wallis, and Weingast’s (NWW) (2009) conceptualization of ‘limited access orders’. Building on these efforts, Levy (2014) lays out a typology of country-level political settlements, distinguishing between them according to whether their configuration of political power is dominant or competitive, and whether the institutional rules of the game are personalized or impersonal. This two-fold distinction generates four ‘ideal types’ of political settlement:
• Dominant-personalized—where elite cohesion is high, and power is exercised top-down by the leadership, with limited constraint.
• Personalized competitive (or competitive clientelist)—where elite cohesion is low, the settlement revolves around agreement that political power should change hands on an electorally competitive basis, but the rules of the game are personalized.
• Competitive with rule-of-law—where politics is competitive, and impersonal rules prevail.
In a companion volume to this one (also sponsored by the Effective States and Inclusive Development research programme), Hickey and Hossain (2018) and co-authors use this typology for a comparative cross-country analysis of the politics of education quality in Bangladesh, Cambodia, Ghana, Rwanda, and Uganda, plus South Africa.9 This book extends the typology ‘vertically’—it explores how institutional and political contexts shape development policymaking and implementation at each of a variety of distinctive levels.10
Figure 1.1 lays out the multi-level framework. As the figure suggests, each level is nested within a set of incentives and constraints shaped in part by the higher level; in turn, each level shapes, in part, the incentives and constraints which prevail at lower levels. At the peak level is the country’s national political context—the national political settlement, the national-level (p.15) institutional arrangements for the education sector derived from that settlement, and national-level policies for the education sector. At the next level down is the subnational context, shaped partly by the national level, and partly by distinctive, provincial-level drivers. The provincial-level political context in turn affects the operation of provincial education bureaucracies. All of this cascades down to the de jure and de facto school-level governance arrangements, and thence to educational outcomes.
A multi-level framework along these lines provides a platform for giving practical content to the idea of ‘good fit’. ‘Good fit’, Levy and Walton (2013) hypothesize, can be framed in terms of the alignment between the political and institutional patterns which prevail at a higher level, and the patterns which prevail at levels beneath that:
• Where the higher- and lower-level configurations are aligned, we can say we have a ‘good fit’—and thus potentially the best feasible outcome.
• Where they are misaligned, we can say we have a ‘poor fit’—there exists the possibility of improving the development outcome by reshaping lower-level institutional arrangements and policy choices to align better with the political and institutional arrangements which prevail at higher levels.
Based on this formulation, the relative merits of hierarchical and horizontal governance of schools are hypothesized to depend in important part on the specific patterns of incentive and constraint which cascade down from higher levels to the levels beneath them.
The ways in which the two core analytical concepts—institutions and politics—are used to distinguish among contexts varies as one moves from the over-arching country-level political settlement to lower levels. It is thus helpful to spell out carefully each of the core concepts, and the relationships between them.
Institutions can be defined colloquially as ‘the rules of the game’. North (1990) provides a more formal definition of institutions as ‘humanly-devised constraints which govern human interaction’. He distinguishes between the institutional rules of the game on the one hand, and the players of the game (‘organizations’ or ‘stakeholders’) on the other. Building on this distinction, ‘politics’ is defined here as comprising the stakeholders involved in decision-making, their interactions with one another, and their relative power.
As North (1990), North, Wallis, and Weingast (2009), Ostrom (2005), and Acemoglu and Robinson (2006) explore theoretically, governance arrangements and stakeholder dynamics are interdependent: the rules of the game set the parameters for the interactions among stakeholders; stakeholders, in turn, work to try and (re-)shape the rules of the game to their advantage. (p.16) As Levy (2015) explores, in analysing the institutional arrangements for public governance, two sets of distinctions are key:
• The distinction between hierarchical and more horizontal, peer-to-peer modes of structuring authority—with the former organized around vertical relationships between ‘principals’ and ‘agents’, and the latter organized around negotiated understandings among multiple principals, who are more or less equal in power.11
• The distinction between impersonal and personalized rules, monitoring and enforcement mechanisms—with the former applied equally, by an impartial body, to all who have standing, and the latter structured around ‘deals’ among influential actors.
Table 1.1 combines the latter two distinctions, giving us four distinct sets of ‘ideal type’ institutional arrangements for governing public policymaking and implementation:
• The top-right cell delineates the classic impersonal and hierarchical mode of decision-making and implementation.
• The top-left cell delineates a mode of decision-making and implementation which also is hierarchical, and thus governed via nested principal–agent relationships, but is one where compliance on the part of agents follows from the personalized authority of the leadership, rather than a system of rules.
• The bottom-right cell comprises a pattern where multiple stakeholders, each with significant independent authority, agree on how they will work together, and codify these agreements in formal, enforceable rules.
• In the bottom-left cell neither formal rules nor a well-defined hierarchy of authority is in place. Stakeholders may or may not reach agreement as to whether and how to co-operate. Insofar as they do, such agreements are dependent on the shared understandings of the specific parties involved.
Table 1.1. A Public Governance Typology
In practice, any specific governance arrangement is likely to be a hybrid combination of the four ideal types defined by the cells, with the relative (p.17) weight varying from setting to setting. (A useful heuristic way of describing these weights, applied throughout this volume, is to allocate 100 points across the four cells.)
The middle column of Table 1.2 summarizes the explanatory variables used to analyse how the institutional context influences South Africa?s educational outcomes at each of the multiple levels considered in this book. As the table signals:
• At the national level, higher-level institutions enter in two distinct ways:
◦ The overarching 1994 political settlement provided an enabling framework for new sectoral governance and budgetary arrangements, arrangements which were responsive to central concerns of both black and white South Africans vis-à-vis public education (see Chapter 2 for details).
◦ The Education Labour Relations Council (ELRC) provides a formal arena for government and organized labour to deliberate and reach agreement on national-level policies for managing the performance of education bureaucracies (see Chapter 3).
• At the provincial level (explored in depth in Chapters 4–7), higher-level (and inherited) institutional arrangements—including those which govern electoral competition and the internal operation of political parties—account for the quality of education bureaucracies.
• At the school level (the focus of analysis in Chapters 8–10), both higher-level and concurrent institutional arrangements account for the observed educational outcomes.
Table 1.2. How Context Influences Outcomes—A Multi-Level Analysis
Outcome to be explained
(‘rules of the game’)
(stakeholders and power)
Higher-level contextual variables:
New education sector budgetary & governance arrangements
Derived from inclusive political settlement
Higher- and concurrent-level variables:
Quality of performance-based management
Formal bargaining arrangements of ELRC
Interactions among key stakeholders under ANC umbrella
Higher-level contextual variables
Quality of education bureaucracies(chapters 4–7)
(i) Extent of electoral competition
(ii) Configuration of power within governing political party
(iii) Quality of inherited bureaucratic institutions
(iv) Social composition of non-elite groups
Higher-level contextual variables:
School-level educational outcomes
(i) Quality of provincial education bureaucracy
(iii) de facto interactions with officials from department of education with union officials, and with other political stakeholders
(ii) De jure arrangements for school-level governance
Concurrent level variables
(iv) De facto school-level governance arrangements
(v) Interactions between principal, teachers, school governing body, and other external stakeholders
The right-hand column of Table 1.2 directs attention to some key explanatory variables used to analyze how, at each level, ‘politics’—stakeholders, their interactions with one another, and their relative power—influences South Africa’s educational outcomes. Following North, Wallis, and Weingast (2009) and Khan (2010), it is useful to distinguish between two broad types of stakeholder interactions: interactions among elites, and the ways in which non-elites are incorporated politically.
Consider, first, interactions among elites. Two distinct types of inter-elite interactions are explored in this book’s case studies:
• interactions within a political party (governing or otherwise)—illustrated by the strategic interplay among key stakeholders under the African National Congress’s governing umbrella at both the national level (Chapter 3) and within the Eastern Cape (Chapter 5).
(p.18) Turning to the incorporation of non-elites, a first question concerns the mechanisms of incorporation. Especially relevant here is the distinction between broad-based, ‘programmatic’ approaches, centred around the provision of services to citizens, and more clientelistic modes of incorporation. (In South Africa a third approach, the repression of non-elite stakeholders, has an especially long and sordid history.) Chapters 4–7 explore in depth the divergent mechanisms through which non-elites are incorporated in the two case study provinces.
A second question concerns which non-elite actors are (analytically) relevant. Building on a definition suggested by Kelsall and vom Hau (2017), the relevant non-elite groups can be delineated in terms of the ‘social composition’, that is:
all those [elite and non-elite] groups that [can threaten the authority of the political leadership relevant to the problem at hand].12 The social [composition] (p.19) influences…whether or not incumbent leaders will pursue policies intended to benefit broad or narrow segments of the population. Other things being equal, the broader the social composition, the more inclusive the development outcomes.
Chapter 7 explores how the divergent social composition of non-elite groups in the Western Cape and Eastern Cape influenced the operation of their public bureaucracies, and thus educational outcomes.
One final pattern of stakeholder interaction (identified in the bottom part of Table 1.2) warrants attention—stakeholder interactions at the school level. Figure 1.2 details schematically five distinct causal mechanisms through which a variety of stakeholders potentially can influence school-level governance arrangements, and thereby performance:
• Causal mechanism (i)—The influence of hierarchical governance by the education bureaucracy.
As per the earlier discussion, this could have a variety of divergent consequences: it potentially could support efforts to improve educational outcomes; it could be a driver of patronage; or (insofar as the bureaucracy is rendered ineffective by broader political influences) it could be largely without influence.
• Causal mechanisms (ii) and (iii)—between the school principal and the school’s teaching staff.
(p.20) Causal mechanism (ii) is consistent with a central finding of empirical research on the determinants of school performance—namely that the quality of school leadership is an important proximate explanatory variable (Hallinger and Heck, 1996; Leithwood, Patten, and Jantzi, 2010). As per Figure 1.2, we broaden this hypothesis somewhat by considering also the possibility of a two-way relationship, with the organizational culture among the school’s teacher cadre having a (relatively weaker) causal effect on the approach to management of the school principal (Wimpelberg in Stringfield, 1993; Taylor et al., 2013; Leithwood, Harris, and Hopkins, 2008).
Turning to causal mechanisms (iv) and (v), we have:
• Causal mechanism (iv)—The role of bottom-up governance—specifically the extent to which SGBs, with the support of parents and the local community more broadly, are able to provide offsetting governance support and oversight.
• Causal mechanism (v)—Although not included explicitly in Figure 1.1, the school-level research also probed for the influence on school-level governance of the teachers’ unions (SADTU and NAPTOSA) and political parties.
As Chapters 8–10 detail, how the above five causal mechanisms play out varies both across provinces, among schools within each province, and within schools over time. School-level interactions are shaped in the short-to-medium-term by the direction provided by school principals—and, over the longer-term, by the local power dynamics between developmental and predatory stakeholders, and their influence on the de facto governance arrangements within schools. More broadly, the case studies suggest that the school-level outcomes are not fore-ordained by local context, but are contingent and cumulative—with individual ‘agency’ by stakeholders playing a significant role.
1.5 A Preview of the Research Findings
The chapters which follow explore in depth the research questions introduced in section 1.2, using the analytical approach laid out in section 1.4. Chapters 7 and 10 synthesize the findings vis-à-vis, respectively, hierarchical and horizontal governance. Major findings include:
• First, the policies promulgated and the observed patterns of bureaucratic implementation are shaped by distinctive political and institutional drivers—with the causal drivers carefully documented in the research. As Chapter 3 details, (p.21) national-level political and institutional drivers account for the limitations of the performance management arrangements negotiated for the education sector. At provincial level, Chapters 4–7 underscore the stark differences between the Western and Eastern Cape educational bureaucracies in the quality of management—and the roots of these differences in the starkly divergent background political contexts of the two provinces.
• Second (and unsurprisingly), a well-functioning bureaucracy emerges as a valuable asset. As Chapters 4 and 5 detail, the Western Cape does well (but the Eastern Cape poorly) the core bureaucratic tasks of managing resources; assigning personnel to where they are most needed; monitoring and managing on the basis of performance. Econometric analysis in Chapter 6 reinforces the conclusion that differences in bureaucratic performance, rather than demographic or resource differences, account for the Western Cape’s superior educational outcomes relative to other South African provinces. But,
• Third, a well-functioning bureaucracy does not provide a sufficient governance platform for achieving good educational outcomes. The Western Cape school-level case studies in Chapter 8 show how a combination of strong, top-down bureaucracy and weak horizontal governance can result in ‘isomorphic mimicry’—in which a school seemingly is compliant with formal processes, but in practice is locked into a low-level equilibrium of mediocrity. The comparative econometric analysis in Chapter 6 confirms that, notwithstanding its strong top-down management (and significant resource advantages), the Western Cape school system underperforms that of Kenya—even after socio-economic differences and teacher skills and experience are accounted for. Finally,
• Fourth, horizontal governance emerges as a partial institutional substitute for hierarchical weakness. The econometric analysis in Chapter 7 shows a strong, significant positive effect on educational outcomes of ‘parental contribution to building construction and maintenance’ (high in the Eastern Cape relative to the Western Cape). The Eastern Cape school-level case studies in Chapter 9 detail how pro-active engagement on the part of school governing bodies and parents helped sustain and turn around school-level. But participation is no panacea; the case studies also uncover instances of capture by predatory interests.
As these findings underscore, politics, institutional arrangements, management systems, and stakeholder dynamics are interdependent. There are thus multiple governance constraints to education sector reform. There also are multiple potential governance entry points (both hierarchical and horizontal) for improving educational outcomes—with the efficacy of any of them (p.22) depending as much on how politics and power interact within specific contexts as on the technocratic expertise of reformers.
Given the urgency of the task of improving educational outcomes, a pre-occupation with institutional and political complexity might, to some, seem unnecessary. Indeed, the temptation can be strong to argue that getting better results is simply a matter of ‘political will’. However, as the global experience of efforts at education sector reform underscore, a ‘just do it’ approach is misguided. All too often, the result can be confrontation, with a high likelihood of negative, unintended consequences. Effective reform requires skilful engagement with stubborn governance realities.
Building on the empirical findings laid out in Chapters 2–9, the book’s final chapter lays out an incremental and cumulative strategy for engaging these realities—an ‘all for education’ approach to reform. The approach combines two complementary aspects of change. One aspect comprises practical initiatives which aim to strengthen governance by working around, and engaging constructively with, reform constraints. The second aspect comprises a broader reframing of the ideas surrounding how the provision of education and other public services might be improved. Especially relevant here is the call for ‘active citizenship’ in South Africa’s 2012 National Development Plan:
Active citizenship requires inspirational leadership at all levels of society…Leadership does not refer to one person, or even a tight collective of people. It applies in every aspect of life…To build an inclusive nation the country needs to find ways to promote a positive cycle, where an effective state, inspirational leadership across all levels of society, and active citizens reinforce and strengthen each other.
(The Presidency, 2012)
If South Africa were actually to embrace active citizenship as the path to improving outcomes, the consequences for the education system could be far-reaching. In many schools an activated citizenry could shift the balance decisively between developmental and predatory actors in favour of the former. Within the bureaucracy, new momentum could emerge for learning-oriented engagement, surfacing the limits of pre-occupations with ‘process compliance’ for its own sake. Teacher unions might increasingly embrace a vision of teaching as a profession, as a calling, and move decisively away from a narrow pre-occupation with the rights of teachers as employees. New possibilities would arise for adapting national policies in ways that enhance a focus on educational outcomes. Civil society activism might more systematically target those aspects of education sector governance which have strong impacts on learning. A top-down vision of ‘education for all’ is insufficient to meet the frontier challenge of improving outcomes. What now is called for is ‘all for education’.
Acemoglu, D. and Robinson, J. A. (2006) Economic Origins of Dictatorship and Democracy. New York: Cambridge University Press.
Aghion, P. and Tirole, J. (1997) ‘Formal and real authority in organizations’, Journal of Political Economy 105, 1–29.
Alexander, R. (2000) Culture and Pedagogy: International Comparisons in Primary Education. Oxford: Blackwell.
Altschuler, D. (2013) ‘How patronage politics undermines parental participation and accountability: community-managed schools in Honduras and Guatemala’, Comparative Education Review 57(1), 117–44.
Ansell, B. W. (2008) ‘Traders, teachers, and tyrants: democracy, globalization, and public investment in education’, International Organization 62(2), 289–322.
Béteille, T. (2009) ‘Absenteeism, transfers and patronage: the political economy of teacher labor markets in India’. Dissertation, Stanford University, California.
Bourguignon, F. and Verdier, T. (2005) ‘The political economy of education and development in an open economy’, Review of International Economics 13(3), 529–48.
Bruns, B., Filmer, D., and Patrinos, H. (2011) Making Schools Work: New Evidence on Accountability Reforms. Washington DC: World Bank.
Busemeyer, M. R. (2014) Skills and Inequality: Partisan Politics and the Political Economy of Education Reforms in Western Welfare States. Cambridge: Cambridge University Press.
Carpenter, D. (2001) The Forging of Bureaucratic Autonomy. Princeton: Princeton University Press.
Chen, D. (2011) ‘School-based management, school decision-making and educational outcomes in Indonesian primary schools’, Policy Research Working Paper Series 5809. Washington DC: World Bank.
ESID (2016) ‘How politics shapes the quality of education in Ghana’, ESID briefing number 17, University of Manchester.
Essuman, A. and Akyeampong, K. (2011) ‘Decentralization policy and practice in Ghana’, Journal of Education Policy 26(4), 513–27.
Gift, T. and Wibbels, E. (2014) ‘Reading, Writing, and the Regrettable Status of Education Research in Comparative Politics’, Annual Review of Political Science 17(1), 291–312.
Grindle, M. (2004) Despite the Odds: The Contentious Politics of Education Reform. Cambridge: Harvard University Press.
Grindle, M. (2013) Jobs for the Boys. Cambridge: Harvard University Press.
Gustafsson, G. and Taylor, S. (2016) ‘Treating schools to a new administration: Evidence from South Africa of the impact of better practices in the system-level administration of schools’. Stellenbosch Economic Working Paper Series WP No. 05/16, Stellenbosch University, Department of Economics.
Hallinger, P. and Heck, R. H. (1996) ‘Reassessing the principal’s contribution to school effectiveness: A review of empirical research, 1980–1995’, Educational Administration Quarterly 32(1).
Hicken, A. and Simmons, J. W. (2008) ‘The personal vote and the efficacy of education spending’, American Journal of Political Science 52(1), 109–24.
Hickey, S. and Sen, K. (2017) ‘Politics, power and development: Identifying alternative pathways to social justice’, mimeo.
Hoxby, C. (1996) ‘How teachers’ unions affect education production’, Quarterly Journal of Economics 111(3), 671–718.
Israel, A. (1987) Institutional Development: Incentives to Performance. Washington DC: World Bank/Oxford University Press.
Jimenez and Sawada (1999) ‘Do community-managed schools work? An evaluation of El Salvador’s EDUCO program’, World Bank Economic Review 13(3), 415–41.
Jimenez and Sawada (2003) ‘Does community-management help keep kids in school?’ Centre for International Reseach on the Japanese Economy, Discussion Paper F-236. University of Tokyo.
Khan, M. (2010) ‘Political Settlements and the Governance of Growth-Enhancing Institutions.’ London: mimeo.
Kelsall, T. and vom Hau, M. (2017) ‘Beyond Institutions: A Political Settlements Approach to Development’, paper presented at ESID research workshop on political settlements, University of Manchester.
Kingdon, G. G., Little, A., Aslam, M., Rawal, S., Moe, T., Patrinos, H., Beteille, T., Banerji, R., Parton, B., and Sharma, S. K. (2014) ‘A Rigorous Review of the Political Economy of Education Systems in Developing Countries. Final’. Education Rigorous Literature Review.
Kingdon, G. and Muzammil, M. (2003) The political economy of education in India. Oxford: Oxford University Press.
Kingdon, G. and Muzammil, M. (2009) ‘A political economy of education in India: the Case of Uttar Pradesh’, Oxford Development Studies 37(2), 123–44.
Kingdon, G. and Teal, F. (2010) ‘Teacher unions, teacher pay and student performance in India: A pupil fixed effects approach’, Journal of Development Economics 91(2), 278–88.
Kosack, S. (2012) The Education of Nations. New York: Oxford University Press.
Leithwood, K., Harris, A., and Hopkins, D. (2008) ‘Seven strong claims about successful school leadership’, School Leadership and Management 28(1), 27–42.
Leithwood, K., Patten, S., and Jantzi, D. (2010) ‘Testing a conception of how school leadership influences student learning’, Educational Administration Quarterly 44(4), 455–7.
Levy, B. (2014) Working with the Grain: Integrating Governance and Growth in Development Strategies. New York: Oxford University Press.
Levy, B. (2015) ‘Governance Reform: Getting the Fit Right’, Public Administration and Development 35(4).
Levy, B. and Walton, M. (2013) ‘Institutions, Incentives and Service Provision: Bringing Politics Back in’, Working Paper 18, Effective States and Inclusive Development Research Centre, University of Manchester.
Lipsky, M. (2010) Street-level Bureaucracy: Dilemmas of the Individual in Public Service. New York: Russell Sage Foundation.
North, D. W. (1990) Institutions, Institutional Change and Economic Performance. New York: Cambridge University Press.
Ostrom, E. (1990) Governing the Commons. New York: Cambridge University Press.
Ostrom, E. (2005) Understanding Institutional Diversity. Princeton, NJ: Princeton University Press.
Pandey, P., Goyal, S., and Sundararaman, V. (2011) ‘Does Information Improve School Accountability? Results of a Large Randomized Trial’. South Asia human development sector; discussion paper series no. 49. Washington DC: World Bank.
Pradhan, M., Suryadarma, D., Beatty, A., Wong, M., Gaduh, A., Alisjahbana, A., and Artha, R. P. (2014) ‘Improving Educational Quality through Enhancing Community Participation: Results from a Randomized Field Experiment in Indonesia’, American Economic Journal: Applied Economics 6(2), 105–26.
Pollitt, C. and Bouckaert, G. (2000) Public Management Reform: A Comparative Analysis. Oxford: Oxford University Press.
Prawda, J. (1993) ‘Educational decentralization in Latin America: Lessons learned’, International Journal of Educational Development 13(3), 253–64.
Pritchett, L. (2013) The Rebirth of Education. Washington DC: Center for Global Development.
Ravitch, D. (2010) The Death and Life of the Great American School System. New York: Basic Books.
Ravitch, D. (2013) Reign of Error: the hoax of the privatization movement and the danger to America’s Public Schools. New York: Alfred A. Knopf.
Rawal, S. and Kingdon, G. (2010) ‘Akin to my teacher: does caste, religious or gender distance between student and teacher matter? Some evidence from India’, DoQSS Working Paper 10–18, Department of Quantitative Social Science Institute of Education, University of London.
Russakoff, D. (2015) The Prize: Who’s in Charge of America’s Schools. New York: Houghton Mifflin Harcourt.
Sah, R. K. and Stiglitz, J. E. (1986) ‘The architecture of economic systems: Hierarchies and polyarchies’, American Economic Review 76(4), 716–27.
Sayed, Y. (2002) ‘Democratising education in a decentralised system: South African policy and practice’, Compare 32(1), 35–46.
Sharma, R. (2009) ‘The internal dynamic’. In R. Sharma and V. Ramachandran (eds), The Elementary Education System in India. New Delhi: Routledge.
Sniltsveit, B., Stevenson, J., Menon, R., Phillips, D., Gallagher, E., Geleen, M., Jobse, H., Schmidt, T., and Jimenez, E. (2016) The impact of education programmes on learning and school participation in low- and middle-income countries: a systematic review summary report, 3ie Systematic Review Summary 7. London: International Initiative for Impact Evaluation (3ie).
Stasavage, D. (2005) ‘Democracy and education spending in Africa’, American Journal of Political Science 49(2), 343–58.
Taylor, N., Gamble, J., Spies, M., and Garisch, C. (2013) ‘School leadership and management’. In N. Taylor, S. van der Berg, and T. Mabogoane (eds), Creating effective schools. Cape Town: Pearson.
UNESCO (2014) Teaching and Learning: Achieving Quality for All. EFA Global Monitoring Report, 11.2013/14. Paris: UNESCO.
Van der Berg, S., Spaull, N., Wills, G., Gustafsson, M., and Kotze, J. (2016) Identifying Binding Constraints to Educational Improvement. Synthesis Report for the Programme to Support Pro-poor Policy Development (PSPPD). Stellenbosch University: Research on Socio-Economic Policy.
Wales, J., Magee, A., and Nicolai, S. (2016) ‘How Does Political Context Shape Education Reforms and Their Success?’, https://www.odi.org/sites/odi.org.uk/files/resource-documents/10808.pdf.
Westhorp, G., Walker, B., Rogers, P., Overbeeke, N., Ball, D., and Brice, G. (2014) ‘Enhancing Community Accountability, Empowerment and Education Outcomes in Low and Middle-Income Countries: A Realist Review.’ London: Department for International Development.
Williamson, O. (2000) ‘The New Institutional Economics: Taking Stock, Looking Ahead’, Journal of Economic Literature 38(3), 595–613.
Wilson, J. Q. (1989) Bureaucracy. New York: Basic Books.
Wimpelberg, R. (1993) ‘Principals’ roles in stable and changing schools’. In C. Teddlie, and S. Springfield (eds), Schools make a difference: Lessons learned from a 10-year study of school effects (pp. 165–86). New York: Teachers College Press.
World Bank (2004) Making Services Work for Poor People, World Development Report 2004. Washington DC: World Bank and Oxford University Press.
World Bank (2018) Learning: to realize education’s promise. World Development Report 2018. Washington DC: World Bank.
Yanguas, P. (2017) ‘Varieties of state-building in Africa: Elites, ideas and the politics of public sector reform’. ESID Working Paper No. 89. Manchester.
Zengele, T. (2013) ‘A critical Analysis on Processes of Filling Promotional Posts in South African Schools: A Transactional Leadership Perspective’, Mediterranean Journal of Social Sciences (4)3.
(2) This is a central theme of the present study, with multiple references throughout this volume. Important contributions from different perspectives than that adopted here include Kosack (2012).
(5) Pritchett (2013) is a strong (implicit) advocate of these more market-like options. Ravitch (2010, 2013) offers a more chastened picture—as a former advocate sobered by the many unintended consequences of the American school reform movement. Russakoff (2015) provides an extraordinarily rich depiction of the uphill, contentious struggle to implement a hybrid agenda of school reform from 2010–13 in Newark, NJ, USA.
(6) For a sampling of the voluminous literature on the impact of pedagogical approaches on educational outcomes, see Snilstveit et al. (2016).
(7) It is important to underscore that this volume does not take any normative position as to what patterns of inter-governmental (de-)centralization are preferred. The analytical and normative pre-occupation of this chapter concern the balance between hierarchical and horizontal patterns of governance vis-à-vis authority at the level of schools.
(8) See other chapters in this volume for many references in addition to those highlighted here.
(9) See also Wales, Magee, and Nicolai (2016) for a comparative review of the results of eight country studies conducted by the Overseas Development Institute which explore the relationship between different types of ‘political settlement’ and educational outcomes.
(12) This formulation is broader than Kelsall and vom Hau’s framing which focused narrowly on the social ‘foundation’ (rather than ‘composition’) of a political settlement (rather than of ‘contexts’ more broadly). In Kelsall and vom Hau’s variant of political settlement analysis, the relevant social foundation comprises those groups which ‘have the power to overturn the settlement, and which are not repressed or resisted by government.’