INTRODUCTION: THE ACCOUNTABILITY POLITICS OF REDUCING HEALTH INEQUALITIES: Learning from Brazil and Mozambique

INTRODUCTION: BRAZIL, MOZAMBIQUE AND THE POLITICS OF ACCOUNTABILITY FOR HEALTH EQUITY

Inequality is a key political issue of our times. It has political consequences, fuelling conflict and raising legitimacy challenges for regimes around the world, in democratic and non-democratic settings alike. At the centre of these challenges is the question of accountability: who can be held accountable, on what basis, how and by whom for tackling or failing to tackle which inequalities. In the field of global health, inequality has long been a key issue. A significant body of work - in global health and health systems research - has approached the issue of health inequalities and inequities by highlighting the role of social determinants (Marmot, 2015; Wilkinson; Pickett, 2006; Barreto, 2017). A more recent and less developed stream of work frames them in terms of political determinants: in other words, seeing them as issues that may be addressed through politics as well as policy.

As a contribution to this literature, and building on the decade-old consensus around the importance of political will in the expansion of access to health services, this Novos Estudos dossier focuses on the politics of ensuring accountability for health equity (Nelson; Bloom; Shankland, 2018) - or for reducing health inequalities. It does this by unpacking the ways in which accountability politics has shaped the key enabling and disabling factors influencing the extent to which countries in the Global South in general - and Brazil and Mozambique in particular - have been able to translate their historical political commitments to equitable access to health services into durable institutional changes. These changes have included innovative health system organisation arrangements, sustained human and financial resource mobilisation and improved management capacities to deal with increasingly complex pluralistic health systems.

Thus, our analytical gaze has centred on the political struggles over time to create better systems, and on the ways in which political factors (including electoral politics, health system governance and institutional design and state-society relations) have interacted to produce more pro-poor or pro-equity patterns of health outcomes. While the series of articles we present here discusses the roles that power, accountability and politics have played in shaping patterns of access to health services in just two countries, we believe that our findings have implications that go well beyond these cases. In many ways, the trajectories of Brazil and Mozambique over the last four decades exemplify broader patterns that have emerged during a period of intense reflection and change in global health. We hope that the findings presented in this dossier can serve as a resource not only to inform conjecture about future health equity dynamics in both countries, but also to support mutual learning among other unequal societies about the most effective strategies for moving towards global health equity.

CONTEXT: THE CHANGING GLOBAL AND DOMESTIC POLITICS OF ACCESS TO HEALTH SERVICES

When Tedros Adhanom Ghebreyesus became Director General of the World Health Organisation (WHO) in 2017, he announced that his organization would give top priority to the goal of making rapid progress towards universal health coverage (UHC). The following year, 2018, saw three highly significant UHC-related anniversaries: forty years since the Alma Ata Declaration which made access to primary health care for all a global goal; seventy years since the foundation of one of the world’s best-known universal health systems, the UK National Health Service (NHS); and thirty years since the universal right to health was enshrined in the Brazilian Constitution and underpinned the foundation of the Sistema Único de Saúde (SUS). In October 2018, a major international meeting organized by the WHO and Unicef issued the Declaration of Astana, which reaffirmed the values and principles of Alma Ata and the fundamental right of every human being to the enjoyment of the highest attainable standard of health.

But amidst the celebrations and exhortations, there is increasing unease. Seismic political change in many countries - including Brazil, the United Kingdom and the United States - has cast doubt on the durability of what had previously seemed to be a broad consensus around the importance of universal rights and international cooperation. With the end of a period of rapid growth in the Global South, fiscal crises and austerity measures have called into question governments’ willingness and ability to sustain increased levels of health spending, whether in “rising powers” like Brazil or low-income countries like Mozambique. Health service provision has become an issue raised not only in election campaigns but also in street protests, as Brazil witnessed in 2013. These trends have served to remind us of both how intensely political the issue of access to health services has always been, and how important the historical context is for shaping the social contract between states and citizens that mediates this access (Bloom et al., 2008).

The historical context of the original Alma Ata Declaration was the end of the period of decolonisation and the expansion of Cold War superpower rivalry across the Global South. The Declaration expressed the global consensus on post-colonial and post-revolutionary health system development strategies and provided a vision of an ideal health system, but the experience of translating this global vision into reality has been mixed. In 1978 recently decolonized countries like Mozambique - initially held up as a global exemplar for its commitment to delivering on the promise of Alma Ata - rushed to promise their citizens that the victorious struggle for freedom from colonial rule would be followed by an equally victorious struggle to ensure health care for all. Then, as the Cold War fed into proxy wars in Africa as well as Asia, these countries became battlegrounds in which the fight for supremacy between different ideological and geopolitical forces inflicted severe collateral damage on their health systems and the health of their citizens. With the end of the Cold War, superpower rivalry was replaced by debt crisis and neoliberal hegemony, bringing a new set of challenges for state-led visions of universal provision.

Since the 1970s most countries have experienced a substantial expansion of their network of health facilities and in the numbers of health workers. And there are now very few localities where people cannot obtain a wide variety of drugs, if they can afford to pay for them. Overall, there have been reductions in excess mortality. However, major differences in health and in access to health services persist between countries, between localities within countries (along the urban-rural divide as much as within rapidly growing urban areas) and between different social groups. The issue of inequality, while not particularly new to the field of epidemiology or public health, has gained renewed international attention in recent years under the umbrella of the Sustainable Development Goals (SDGs). The WHO itself has recently called for a “movement” to address these persisting inequalities and achieve universal health coverage by 2030 (Nelson et al., 2018). Thus, the challenge of ensuring greater equity and access to services has clearly been placed on the table.

Realistic strategies for achieving this goal will need to build on the lessons of the past forty years and note the reasons why some countries have made a lot of progress in creating more inclusive health systems while others have failed to do so. They also need to take into account the big changes that have taken place during the past forty years. Many countries have experienced a growing importance of markets and greater integration into the global economy. This has been mirrored by a move away from the central planning that characterized early attempts to implement universal access in the health sector. This move has been associated with the emergence of pluralistic systems (Bloom et al., 2008), in which people seek medical care from a variety of public and private providers.

Many countries have experienced rapid urbanization and ageing of the population with associated changes in health problems. There have also been big changes in the way people organise in political parties, social movements and other civil society organisations to influence, amongst other things, the performance of the health sector. At the same time, the proliferation of channels of information through the mass media and, more recently, social media is changing the ways that people seek health information and hold providers of health services to account. Countries need health system development strategies that take into account this rapidly changing reality.

Another important change has been the weakening of the Post World War II global order, which was marked by the dominance of a small number of countries, often associated with the external imposition of policy agendas such as the post-1980s neoliberal model that came to be characterized as the Washington Consensus. One illustration of the change is the difference between the Millennium Development Goals (MDGs), issued in 2000, and the SDGs of 2015. The former were designed top-down and included specific targets, which strongly influenced the design of aid-funded programmes and programmes of debt relief. The latter are much more general and express a global consensus reached after a complex process of consultation. Having (re) emerged as important providers of “South-South” development cooperation, a number of large middle-income countries, including Brazil and China, had an important influence on this process. There is no consensus on modes of international finance or on the role of the SDGs in financial transfers to low income countries. On the contrary, one characteristic of this new context would appear to be an increased emphasis on national sovereignty and, therefore, on the influence of national governments and local politics on the development and performance of health systems.

THE POLITICS OF REDUCING HEALTH INEQUALITIES: LEARNING FROM BRAZIL AND MOZAMBIQUE

This means that it is very timely to re-examine the specifically national and local politics shaping access to health care and the governance of health systems, whilst also looking across countries to identify common patterns. Brazil and Mozambique, more than sharing a common language and Portuguese colonial past, have both negotiated social contracts that included a strong commitment to creating universal health systems and enacting the right to health. In both countries, a socio-political movement led to the construction of a national compact with a vision similar to that in the Alma Ata Declaration: pro-equity national health compacts. While the symbolic moment of Mozambique’s universal health pact is identified with independence in 1975 and claimed by the ruling Mozambique Liberation Front (Frelimo), in Brazil it is identified with the enacting of the universal right to health in the 1988 Federal Constitution and claimed by the social movement for health reform, the movimento sanitarista. Following these foundational moments, the governments of both countries made health central to their offer to their populations during the consolidation of the post-colonial regime in Mozambique and the democratic regime in Brazil.

The subsequent history of both countries provides important lessons on the factors, and particularly the kind of politics, that influenced the outcomes in terms of equality of access to services and in the distribution of health outcomes. What has happened to the initial promises? To what extent have the two countries been able to overcome a series of geographical and socioeconomic structural inequalities and promote equal access to health services nationally? And what can be learned from the two countries’ experiences of designing, planning, managing and operationalising health systems in highly unequal settings over several turbulent decades that can help to guide the growing universal health coverage movement?

For the past three years, a group of researchers and practitioners from Brazil, Mozambique and the United Kingdom have been working together to generate evidence that can help to answer these and other questions. Our project, The Accountability Politics of Reducing Health Inequities: Learning from Brazil and Mozambique, which was funded by a grant from the Joint Fund for Poverty Alleviation Research of the Economic and Social Research Council (ESRC) and Department for International Development (DFID),¹ set out to support efforts to translate national and international commitments to achieve universal health coverage into better access to high priority services by the poor and marginalised by identifying relevant lessons from the experiences of Brazil and Mozambique.

The project was based on the assumption that although the two countries are very different, the fact that they shared the experience of a foundational compact between state and citizens based on a principle of universal access to health services meant that there would be value in comparing their trajectories over the decades since that compact was first established, as well as the contemporary dynamics of their health systems. A comparison could yield insights into the barriers faced by contemporary efforts to achieve UHC as well as into the drivers of successful innovation to deliver it. Such a comparison could also potentially provide insights into whether and how rights-based social compacts can sustain pro-universality policy strategies even in adverse contexts, in which government health systems are facing political as well as fiscal pressures.

The group agreed that this comparison would need to be multi-dimensional as well as longitudinal, and should include primary qualitative data collection as well as secondary data analysis. In order to deliver on this ambitious agenda, the project brought together a coalition of three research organisations (the Institute of Development Studies - IDS from the United Kingdom -, the Brazilian Centre from Analysis and Planning - Cebrap from Brazil - and Kula Applied Research & Studies from Mozambique) and two health rights NGOs (N’weti from Mozambique and Saúde Sem Limites from Brazil) with a multidisciplinary team that included political scientists, communication researchers, anthropologists, health economists, epidemiologists and historians as well as civil society activists.

From the outset, our aim was to intervene in policy and practice rather than simply to act as detached observers. To this end, the project has emphasised engagement with key actors in the health system at every level - from the Minister of Health in Mozambique to the São Paulo Municipal Health Secretariat to a local indigenous health council in Amazonas - and has helped to convene global conversations involving policymakers from the WHO, the UK Department for International Development, the Swiss Agency for Development and Cooperation and the Open Society Foundations, among others, as well as promoting South-South exchange in partnership with the international development cooperation and global health diplomacy arm of Fiocruz, Brazil’s national health research and training agency. The research was supported by dialogue with experts from academia, government and civil society, who agreed to join Reference Groups for the work in Brazil, in Mozambique and across the two countries. All of these engagements served as opportunities for data gathering as well as policy influence, since they allowed us to co-construct, refine and validate our conceptual framework and emerging findings.²

INTELLECTUAL AGENDA: POLITICAL DETERMINANTS AND ACCOUNTABILITY POLITICS FOR HEALTH EQUITY

Alongside its commitment to engage with policy and practice, the project has engaged with theory in a key field of work on social policy governance in general and health system governance in particular: conceptualising and analysing accountability and its relationship with health equity. This is because we believe that tackling health inequalities is fundamentally a political issue, and that accountability is a central aspect of the politics of health. When promises of universal access are made, who is held to account for delivering them, and how? When health rights remain unevenly unfulfilled, what accountability failures have taken place, and what power relations shape these failures? In sum, what accountability dynamics operate in relation to efforts to reduce health inequalities and ensure universal access?

By asking these questions, our research endeavours contribute to the growing body of work that seeks to understand the politics of social policies, and health policies in particular, but also to the body of work on accountability and health accountability, offering a rigorous examination of the politics of accountability for health equity drawing on policy-relevant empirical evidence from countries in the Global South. In the remaining of this section, we will briefly outline our accountability politics approach to understanding shifting trajectories in health inequalities and discuss what it means for the intellectual field in which we are engaged.

We understand the politics of accountability for health equity to be a key component of what is described in the health systems literature as the political and institutional determinants of health system performance and of health inequalities - what Ellen Immergut has framed as “health politics” (1992). As subsequently suggested by Ilona Kickbusch, “looking at health through the lens of political determinants means analysing how different power constellations, institutions, processes, interests, and ideological positions affect health within different political systems and cultures and at different levels of governance” (Ottersen et al., 2014; Kickbusch, 2015).

We consider these political lenses to be important in our study because while the evidence on the social determinants of health inequalities has been thoroughly analysed for several countries (including Brazil and Mozambique), political and institutional determinants of health system performance and of the ways in which health systems are organised have been less widely studied. Power and political relations have a historical materiality in health systems, given the historically constructed nature of these systems and their ideological underpinnings, or national health compacts, which shape rights and responsibilities, as well as contemporary and always shifting manifestations in the ways that the health systems and policies operate on a daily basis.

When it comes to the study of accountability in health, is important to note that looking at how health systems work and deliver on their promises to expand access to health to citizens in a given community through the lenses of accountability is not a new endeavour. On the contrary, accountability - often referred to as a “buzzword” in the international development discourse (Cornwall, 2007) - has been mainstreamed in several international debates on how to make health systems work better and/or work for all. Several theoretical models as well as a proliferation of policy tools - a characteristic feature of the accountability field (Joshi; Houtzager, 2012; Eyben, 2008) - have appeared. While some models emphasised the role of political and managerial decentralisation, others highlighted the link between increased citizens’ voice and government responsiveness to poor and vulnerable citizens’ health needs in enabling the so-called “short route to accountability” to improve the performance of health systems (World Bank, 2003; Flores et al., 2009; Coelho; Waisbich, 2016). Over the past decade or more, these models have become increasingly influential in national and global health debates and policy responses. However, and at the same time, their limitations have become increasingly apparent, as it has become harder to ignore the complexity of actually existing accountability relations in pluralistic, often highly fragmented health systems, and also the intensely political nature of accountability issues that had initially been framed as straightforwardly “technical”. It is now clear that if the UHC movement is to succeed it will require more systematic attention to health politics and more comprehensive approaches to conceptualising health accountability politics.

In an attempt to respond to the challenges of conceptualising health accountability beyond what Erica Nelson and colleagues refer to as the prevailing “empty rhetoric of accountability for “improved health service delivery” or “more resilient health systems’” (Nelson et al., 2018, p. 3), we have jointly constructed an analytical framework designed to make sense of (a) the national compacts, (b) the historical trajectories and (c) the contemporary dynamics of the health systems in Brazil and Mozambique, focusing on the socio-political-institutional conditions and arrangements that have enabled reductions in health inequalities in some settings and/or localities. In other words, we have sought to explain what has worked for reducing particular forms of health inequalities, as well as what has led to the persistence of inequalities under different accountability arrangements, and the implications for different groups.

Our framework is based on three main assumptions. The first of these is that power plays a central role in shaping the accountability politics of health systems, which must be understood as historically constructed through longer-term processes of political contestation and institutional change, shaping actors’ understandings of health rights and responsibilities (Cornwall; Shankland, 2013).

Second, that it is essential to examine the role played by a “tripod” of accountability dimensions (political, social and managerial) operating and interacting in each setting. The interactions among these dimensions produce what can be conceptualised as “accountability ecosystems” (Halloran, 2016) or even, we suggest, as “accountability regimes”, insofar as they work as assemblages of norms, logical underpinnings, tools and incentives that shape different and coexisting forms of public accountability relations between the state and its citizens, but also between non-state actors (domestic and international) who are also engaged in policy and service delivery.

Our use of the concept of “regimes” to think about accountability arrangements has been influenced by three bodies of literature. The first is the “welfare regimes” literature (exemplified by the work of Esping-Anderson [1990], and Gough and Wood [2004]), with its emphasis on historically-contextual differences in the ways in which states have crafted their welfare models. This aligns with our emphasis on the role of foundational compacts in shaping the expectations that citizens have of their national health systems. The second comes from the International Relations literature (for instance the work of Krasner [1983], and Grant and Keohane [2005]), where (international) regimes are understood as a set of norms, rules and procedures (both formal and informal) that shape and constrain states’ behaviour under a multi-level and multi-layered governance systems. In the global health field, it is clear that the arenas shaped by international regimes, characterised by multiple authorities and complex responsibility chains, are increasingly visible in domestic public affairs and health governance in virtually all states, including Brazil and Mozambique. Finally, our use of the concept is informed by the literature on urban citizenship territories (for instance the work of Oosterlynck et al., 2013), which highlights the different regimes that may operate in different directions in the same territory. This approach has informed our case study focus on disentangling the concurrent effects of policies at different territorial levels and the existing intra-national differences in citizen-state relations.

Lastly, our framework is informed by the assumption that analysis should take into account the relational nature of the three accountability dimensions that we mobilise (political, social and managerial). A “relational gaze”³ applied to the concept of accountability pays attention to the patterns of social relations - or social mechanisms (Bovens, 2010) - that both shape and are shaped by those providing and receiving services. Relationality can also be seen in the way in which the three different accountability dimensions interact to constitute a multi-layered, multi-level, multi-actor accountability regime.

As such, this framework operationalised in our study is an attempt to go beyond the existing health sector accountability models, building on existing, but often fragmented, knowledge towards a more robust but also more nuanced and multi-layered approach to understanding the conditions under which health systems can shift towards greater inclusiveness. In deploying our “tripod” of accountability dimensions, we try to go a step beyond reiterating that in health systems accountability politics matters and that it is shaped by multi-dimensional relational power. We seek to characterise what kinds of power relations operate on and shape health systems, by looking at different spaces of interaction situated at different levels, from the local to the global (Janes; Corbett, 2009; Gaventa, 2006.). We also look not only at each of the three types of accountability dimensions examined in the studies - political, social and managerial - but also at the synergies and tensions between them, and the ways in which these synergies and tensions can either enable or disable health systems’ efforts to meet the challenge of inclusivity and move towards greater equity.

In examining political accountability, we look at how the vertical relationships between the state and its citizens operate to shape health outcomes. What is at stake here are the ways in which accountability relationships between elected politicians, health managers and bureaucrats, popular and professional leaders and health service users are expressed in political terms. In our research, different cases have operationalised this dimension through looking at a set of independent variables, including electoral dynamics, the degree of political competition, the degree of decentralisation, the nature of the health-related social contract (that is, the place of health in the imaginary of citizens and in state-society relations), and the extent and nature of health-related citizen mobilisation and protest.

For managerial accountability, we initially focused on health system management contracts and plans, before broadening our investigation to examine the nature of pluralistic service delivery arrangements and relationships between health managers and public or private service providers - whether local, national or international. In this field, our research mobilises variables such as resource allocation models, types of providers, the degree of fragmentation in health provision, types of contracts and levels of management capacity.

Lastly, in seeking to analyse social accountability, we look at both at social accountability tools and at institutionalised spaces for citizen participation in health-related policies and service provision. By analysing both tools and spaces, or what Joshi and Houtzager call the “widget” and the “watchdog” approaches (Joshi; Houtzager, 2012), we align ourselves with their emphasis on the importance of considering the interplay between the two. We also acknowledge the importance of non-institutionalised social claim-making and social relations in shaping accountability, including episodes of “rude accountability” and informal pressures on frontline bureaucrats (Hossain, 2010).

In analysing how the three accountability dimensions interact in each setting, we note that these interactions may involve positive synergies - for instance in creating the incentives for pro-equity policies to emerge and be sustained by elected officials and policymakers, or in creating the conditions for civil society (and citizens at large) to be able to engage in “strategic approaches” (Fox, 2015) that connect social accountability interventions with wider political and institutional change - which can contribute to longer-term change in the structural inequalities that influence health equity outcomes. However, as some of our cases demonstrate, the interactions may be characterised more by tensions than by synergies, as for example when a lack of managerial incentives to improve services for marginalised groups leads to frustration and failure to deliver on political promises, or when social accountability interventions fail to gain a purchase on the wider political economy of health systems that are dominated by service providers whose primary accountabilities lie outside the localities or even the countries where they operate. One of our key conclusions is that pro-equity health outcomes are strongly influenced by the degree of alignment of these various accountability dimensions with each other and with the overall thrust of the original social contract or the national compact around universal health coverage.

THE CASE STUDIES

Our empirical studies, which were carried out in four different research sites (São Paulo and the Rio Negro region in Brazil, and Maputo and Zambezia in Mozambique), have continuously challenged us to adapt and refine our model. Throughout the research process, insights coming from the field pointed to a need to pursue more robust ways to analyse accountability politics on the ground, breaking down the multiple and intertwining accountability dimensions that shape trajectories of health inequalities in countries like Brazil and Mozambique in a manner that is sensitive to historical and geographical differences.

The papers in this Novos Estudos dossier focus on the political and institutional influences on health service provision and health outcomes. Each of them deals with relationships of accountability shaping health policies, services and outcomes through a lens that distinguishes between three different accountability dimensions: political, social and managerial. The papers have engaged with the overall challenge of understanding the ways in which these various accountability dimensions aligned with each other to deliver on the promises of the national compact or failed to do so. The papers reflect different methodological strategies and do not devote the same attention to each of the three dimensions. However, taken together the stories that are represented in the set of papers in this special section highlight trajectories that include progressive responsibilisation of the state by citizens and external actors, but which have been unevenly translated into full accountability for health equity, despite the promises of universal coverage by both the Brazilian and Mozambican states. Moreover, it is clear that the urban poor and the rural poor (a category that often includes a high degree of overlap with groups marginalised on political, social and/or ethnic grounds) are subject to very different accountability politics, with a much stronger prevalence of market accountability relations in urban areas and a greater influence of traditional social norms and practices on health-seeking behaviour in rural areas.

Besides this introductory text, this section features four papers. The first is a comparative study - by Rômulo Paes-Sousa, Leonardo Chavane and Vera Schattan P. Coelho - on the historical trajectories of health inequalities in Brazil and Mozambique since the 1990s. This is followed by two articles on Brazil, one by Vera Schattan P. Coelho, Luís Marcelo Marcondes and Marina Barbosa on the trajectories of health inequalities in the city of São Paulo since 2001, and the other by Luciana Benevides Ferreira, Danilo Paiva Ramos and Alex Shankland on the Rio Negro Special Indigenous Health District (in the Brazilian state of Amazonas). Finally, there is one article - by Cristiano Matsinhe and Denise Namburete - that assesses health equity challenges in the cities of Maputo and Quelimane, in Mozambique, through the prism of the political accountability relations that result from the country’s incomplete decentralisation processes.

When taken together, what can the experiences of these four sites - the megalopolis of São Paulo, of the remote Rio Negro Indigenous Health District, of Mozambique’s capital and largest city Maputo and of Quelimane in historically marginalised Zambezia province - possibly say about what combinations of accountability dimensions work best under which political and institutional conditions for reducing health inequalities? Our logic of case selection, featuring both major urban centres and rural marginalised areas in each of the countries, enabled us to examine the micro-, meso- and macro-level trajectories of change in health systems performance and in accountability and associated power relations among managers, providers and citizens, in order to arrive at a better understanding of what works for different poor and marginalised groups in different contexts. Needless to say, these cases cannot be taken as a systematic representation of either urban or rural areas in the two countries. However, the parallels and contrasts between them has provided scope for comparison of different combinations of elements and different causal pathways - including where one or more of the elements of a pro-poor accountability regime is absent or limited - thereby permitting a combination of within-case contribution analysis and cross-case comparison.

So what stories do those two countries and the four cases tell us? Both Brazil and Mozambique are highly unequal countries, and although they show different trajectories when it comes to reducing health inequalities both can point to some important historical gains. Observing these trajectories over a period of three decades or more, it is clear that there has been a clear improvement in a series of basic health indicators and reduction in regional/geographical health inequalities. These inequalities historically related to uneven patterns of socioeconomic development and to processes of political marginalisation, resulting in a South-North divide in both countries. Reductions in maternal and child mortality serve as important indicators of the extent of this decline in regional inequality, and they have proved to be very sensitive to the poverty reduction gains both countries achieved in the past decades. Nevertheless, there are significant exceptions to these general trends, including a notably worse performance in reducing the health inequalities affecting some rural populations in Brazil (particularly indigenous peoples) and some historically marginalised provinces in Mozambique (particularly Zambezia). Overall, although both countries have recorded significant progress, Brazil’s processes of decentralisation and correlated expansion of primary health programmes, specifically targeting areas with lower per capita income, help to explain why it has secured more comprehensive results than Mozambique, whose politics remain very much shaped by the vivid legacies of the civil war.

An important part of this story in Brazil has been its clear trajectory of progressive national sovereignty - and health sovereignty - within which both government actors and civil society have taken responsibility for expanding access to health coverage. This made possible the development of strong institutional arrangements to influence health system performance. Despite the strength of the broad policy and political framework governing these arrangements, as noted above, their performance has not been uniform. We have chosen two cases in Brazil that illustrate different experiences.

The first is that of São Paulo, a mega-city with the largest economy of any municipality in Brazil and a strong and highly competent local government. São Paulo has experienced substantial economic growth and has dense networks of public and private organisations, with strong managerial capacity. Although its health department receives fiscal transfers from the central government, it also relies heavily on local tax revenue. The population of São Paulo can claim access to health services as a right that is established both nationally and locally, and this is reflected in the politics of accountability. The study focuses on the degree to which the alignment of the three dimensions of accountability has led to reductions in inequalities between relatively rich and poor sub-regions within the city.

The São Paulo case tells a story of sustained political competition (and alternance between administrations from the Workers’ Party - PT - and two centre-right parties, the Social-Democrats - PSDB - and the Democrats - DEM) and the virtuous combination of pro-poor social mobilisation with adoption of innovative management tools leading to significant reduction in health inequalities between regions with higher and comparatively lower human development indices. In São Paulo, health service provision has evolved into a complex ecosystem of service providers. These are not necessary public, but government has taken the responsibility for expanding access to health services - with a great emphasis on primary health care - to the most historically marginalised groups, geographically located in the outskirts of the megalopolis. The authors tell a promising story of decentralisation working for the most vulnerable, at least in terms of access to basic services. In their analysis of the trajectories over the past twenty years, political, managerial and social accountability have aligned to deliver on a pro-poor health compact and have indeed contributed to setting the municipal agenda around prioritising the poorer and more vulnerable areas, achieving some kind of equity in health outcomes in a highly unequal and complex urban setting.

The second Brazilian case focuses on the Rio Negro Special Indigenous Health District in the Amazon region. There, health services are almost entirely funded by earmarked fiscal transfers from the national level and managed by a federal government department, rather than by municipal authorities who are accountable to the local electorate. The relationship that has emerged between local indigenous communities and the department in distant national capital of Brasília that determines how their services will be organised resembles that of the residents of heavily aid-dependent low-income countries - a category that includes Mozambique - on the international donor agencies that fund (and often organise) their services. This “gift” relationship has meant that people have not been able to make claims for health rights on local governments, since there is a mismatch between levels of and spaces for accountability and governance, given that those ultimately responsible for their health policies and services - at the regional or national level - are not the same politicians who are elected at the local level. The chains of communication are long, and the system has not been very responsive either to longstanding service delivery deficits or to needs arising from emergencies (such as the malaria outbreak that is discussed in the case study). Indigenous groups in the Rio Negro, and elsewhere in Brazil, have invested in what Jonathan Fox calls “vertical integration”, channelling their demands for accountability up this long chain to Brasília, but despite initial success in securing and sustaining a substantial budget for health services, health indicators among indigenous populations - including those living in the Rio Negro - remain significantly worse than those for the Brazilian population as a whole. In the absence of supportive managerial and political accountability regimes at the local level, repeated episodes of social mobilisation have not resulted in long-lasting improvements in health conditions.

While Brazil experienced intense political competition but relative civil peace, Mozambique experienced a long civil war, during which its health infrastructure was targeted. For many years, it was extremely difficult for the government to deliver on its social compact and the promises made at independence, as the health system lost many health workers and competent managers. Mozambique’s health services became highly dependent on a proliferating array of international donors. This led to a profound fragmentation of service provision and very complex ecosystem of accountability relationships - a true “Tower of Babel” (Namburete, 2018).

Decades of intense state re-building have improved overall health status, but health inequalities have remained high. Government and external actors (mainly bilateral aid agencies from industrialised countries and transnational non-governmental organisations) have increased health-related spending and created a necessary, but insufficient bandage for Mozambique’s wounded health system through gift-inspired models of service delivery, resulting in a high level of fragmentation and blurred responsibility chains. Innovative social accountability experiments have sprung up, but as the government reconstruction effort is still not complete, and the power of international donors and NGOs is so entrenched in the health system, these initiatives end up channelling rights-claims towards the actors who do not actually decide on priorities and resource allocation within the system. As the comparative paper on the historical trajectories of health inequalities shows, health indicators and significant equity gains happened in Mozambique since the end of the civil war, particularly between the richer and the poorer provinces, yet further progress in reducing health inequalities in Mozambique seems likely to require more attention to stubborn inequalities affecting specific historically marginalised localities - and potentially to require a new social compact that can give greater purchase to a combined use of accountability tools across the three dimensions: political, managerial and social.

Following our assumption that understanding persistent health inequalities requires greater attention to political determinants, the authors of the Mozambique paper argue that a further major challenge has resulted from the country’s unfinished and dysfunctional political decentralisation processes. These have taken different forms in the cities of Maputo and in Quelimane, but in both sites they respond to similar logics - above all a lack of political will on the part of central government to share power - which results in blurred lines of accountability for health equity. The authors argue that this gradual and incomplete process of “decentralisation in small doses” has left the reorganisation of the health system incomplete in several areas of Central and Northern Mozambique, contributing to reproducing regional health inequalities. Unlike in the case of São Paulo, in the Mozambican case, emerging political competition - with a high degree of party-based polarisation - has not led to pro-poor outcomes because the fear of the opposition capturing power in different localities and provinces in the country has created incentives for the central government and political elites in Maputo to resist the decentralisation of responsibilities and resources to the levels at which citizens and civil society are working to construct more effective social accountability mechanisms.

Finally, when assessed in tandem, the trajectories of health inequalities in Brazil and Mozambique tell us two important stories. The first is that foundational national universal health compacts, followed by sustained alignment of political, social and managerial accountability to deliver on pro-equity promises have, in the case of Brazil and Mozambique, achieved important inequality reduction gains in the past twenty years. Those gains are particularly visible for basic health indicators, such as maternal and infant mortality, and have been strongly influenced by other key political and policy developments, including post-conflict reconstruction (in Mozambique), effective decentralization (in Brazil) and comprehensive poverty reduction strategies (in both countries).

The second story is one of persistent inequalities (overlapping regional marginalisation with other politically significant constitutive differences such as gender, race and ethnicity), as in the case of the province of Zambezia in Central Mozambique and the Rio Negro Indigenous Health District in the Brazilian Amazon. Those stubborn inequalities point to the limits in the formulation and/or implementation of national compacts, and raise questions about current dynamics of national and health sovereignty that need to be addressed by national government and civil society alike. The studies include cases where state and health system organisation, inter-institutional arrangements, management capacity issues and/or participation strategies have resulted in insufficient or unclear responsibility chains and blocked or nonexistent accountability channels. These are the places where efforts to meet global and national health promises are lagging behind.

We conclude this introductory piece with a final reflection on what the stories presented here mean not only for Brazil and Mozambique, but also for other countries in Latin America and Southern Africa - and beyond, for the academics and practitioners who make up the wider Universal Health Coverage movement. Given its combination of fine-grained historical, longitudinal and epidemiological analysis, we consider our analytical framework and our findings to be relevant not only to unpacking the past twenty years of health trajectories in Brazil and Mozambique, but also as resources that can be drawn upon to think prospectively about the way inequalities might evolve from now on - whether in Brazil, in Mozambique or globally. The studies we have carried out using this framework have illustrated the need to build institutions that have the capacity to align the specific accountability dimensions we have analysed with each other and with the overall thrust of the original social contract for equitable access to health services.

When we first started work on this project, we decided to name it Unequal Voices, as we wished to highlight our assumption that all accountability dimensions and channels were shaped by different configurations of power relations, and therefore that not all voices counted equally when it came to shaping accountability for health equity. Brazil and Mozambique tell us complex stories of multiple accountability dimensions interacting with each other to produce highly varied outcomes: in some cases working synergistically to support delivering on promises of more inclusive health service provision, and in others generating contradictory pressures that undermine efforts to reduce health inequalities. Through the case studies we have heard many voices speaking of how existing accountability channels are broken and calling for both re-designed responsive management mechanisms and new social accountability tools to be deployed at the local level to ensure that politicians and policy-makers become more responsive. We have also heard about innovative political, social and managerial accountability strategies that have proven their potential to transform the access to services of poorer and more marginalised people. Building on this potential in the increasingly complex social, political and economic contexts that are emerging in Brazil and Mozambique will require a fresh effort to renew the promises of “health for all” that both countries made so powerfully to their citizens at the foundational moments of their health systems.

REFERENCES

• Barreto, Mauricio Lima. “Desigualdades em saúde: uma perspectiva global”. Ciência & Saúde Coletiva, v. 22, n. 7, 2017, pp. 2.097-108.
• Bloom, Gerald; Standing, Hilary; Lloyd, Robert. “Markets, Information Asymmetry and Health Care: Towards New Social Contracts”. Social Science & Medicine, v. 66, issue 10, 2008, pp. 2.076-87.
• Bovens, Mark. “Two Concepts of Accountability: Accountability as a Virtue and as a Mechanism”, West European Politics, v. 33, no. 5, 2010, pp. 946-67.
• Coelho, Vera Schattan P.; Waisbich, Laura. “Participatory Mechanisms and Inequality Reduction: Searching for Plausible Relations”. Journal of Public Deliberation, v. 12, no. 2, 2016.
• Cornwall, Andrea. “Buzzwords and Fuzzwords: Deconstructing Development Discourse”. Development in Practice, v. 17, no. 4-5, 2007, pp. 471-84.
• Cornwall, Andrea; Shankland, Alex. “Cultures of Politics, Spaces of Power: Contextualizing Brazilian Experiences of Participation”. Journal of Political Power, v. 6, no. 2, 2013, pp. 309-33.
• Esping-Andersen, Gøsta. The Three Worlds of Welfare Capitalism. Cambridge: Polity Press, 1990.
• Eyben, Rosalind. “Power, Mutual Accountability and Responsibility in the Practice of International Aid: A Relational Approach”. IDS Working Paper. Brighton: Institute of Development Studies, 2008.
• Flores, Walter; Ruano, Ana Lorena; Funchal, Denise Phe. “Social Participation within a Context of Political Violence: Implications for the Promotion and Exercise of the Right to Health in Guatemala”. Health and Human Rights, v. 11, no. 1, 2009, pp. 37-48.
• Fox, Jonathan. “Social Accountability: What Does the Evidence Really Say?”. World Development, v. 72, 2015, pp. 346-61.
• Gaventa, John. “Finding the Spaces for Change: A Power Analysis”, IDS Bulletin, v. 37, no. 6, 2006, pp. 23-33.
• Gogh, Ian; Wood, Geof. Insecurity and Welfare Regimes in Asia, Africa and Latin America. Cambridge: Cambridge University Press, 2004.
• Grant, Ruth. W.; Keohane, Robert O. “Accountability and Abuses of Power in World Politics”, American Political Science Review, v. 99, no. 01, 2005, pp. 29-43.
• Halloran, Brendan. “Accountability Ecosystems: Directions of Accountability and Points of Engagement”. Research Briefing. Brighton: IDS, 2016.
• Hossain, Naomi. “Rude Accountability: Informal Pressures on Frontline Bureaucrats in Bangladesh”. Development and Change, v. 41, issue 5, 2010, pp. 907-28.
• Immergut, Ellen. Health Politics: Interests and Institutions in Western Europe. Cambridge: Cambridge University Press, 1992.
• Janes, Craig R.; Corbett, Kitty K. “Anthropology and Global Health”. Annual Review of Anthropology, v. 38, no. 1, 2009, pp. 167-83.
• Joshi, Anuradha; Houtzager, Peter P. “Widgets or Watchdogs? Conceptual Explorations in Social Accountability”. Public Management Review, v. 14, issue 2, 2012, pp. 145-62.
• Kickbusch, Ilona. “The Political Determinants of Health: 10 Years on”. BMJ, 350, h81, 2015.
• Krasner, Stephan D. (ed.). International Regimes. Ithaca: Cornell University Press, 1983.
• Marmot, Michael. The Health Gap: The Challenges of an Unequal World. London: Bloomsbury, 2015.
• Namburete, Denise. “When a Health System Turns into a Tower of Babel: Mozambique’s Experience”. The Impact Initiative, 14/11/2018. Available at: <http://www.theimpactinitiative.net/blog/tower-babel-mozambique-experience>.
  » http://www.theimpactinitiative.net/blog/tower-babel-mozambique-experience
• Nelson, Erica; Bloom, Gerald; Shankland, Alex. “Introduction: Accountability for Health Equity: Galvanising a Movement for Universal Health Coverage”. IDS Bulletin, v. 49, no. 2, 2018, pp.1-16.
• Oosterlynck, Stijn et al. “Exploring the Multi-Level Governance of Welfare Provision and Social Innovation: Welfare Mix, Welfare Models and Rescaling”. ImPRovE Working Papers 13/12, Herman Deleeck Centre for Social Policy, University of Antwerp, 2013.
• Ottersen, Olle Peterson et al. “The Political Origins of Health Inequity: Prospects for Change”. Lancet, v. 383, issue 9.917, 2014, pp. 630-67.
• Wilkinson, Richard Guy; Pickett, Kate E. “Income Inequality and Population Health: A Review and Explanation of the Evidence”. Social Science & Medicine, v. 62, issue 7, 2006, pp. 1.768-84.
• World Bank. “World Development Report 2004: Making Services Work for Poor People”. World Bank-Oxford University Press, 2003.

• 1
  The different sub-components of this research also benefited from other sources of funding and support, namely the Open Society Foundations, the Centro de Estudos da Metrópole (CEM-Cepid), and the Citizen Engagement Programme.
• 2
  For a summary of the research and policy activities conducted under the project, as well as a full-list of the members of the Reference Groups, see: https://www.ids.ac.uk/projects/unequal-voices-the-politics-of-accountability-for-equity-in-health-systems/.
• 3
  A comprehensive discussion on the relational dimension of accountability can be found in Rosalind Eyben’s (2008) defence of a relational approach to mutual accountability in foreign aid relations.

Publication Dates

• Publication in this collection
  05 Sept 2019
• Date of issue
  May-Aug 2019