Introduction
Mozambique has recently been hit by a huge corruption scandal relating to undisclosed international loans involving the political and economic elite (Orre and Rønning 2017): indeed, the country’s corruption-related ratings have dropped into freefall over the last decade (Wolf and Klein 2016). At the same time, studies show that a majority of the Mozambican population pay bribes to key public institutions, such as the police and the social registry, and to secure access to services such as health, education, water and electricity (Hanlon 2016; CIP 2016). Small-scale or ‘petty’ everyday corruption may not appear in the main headlines or attract much attention from policy-makers and researchers (Blundo and de Sardan 2006; Stahl, Kassa, and Baez-Camargo 2017), but it is the type of corruption with the most immediate implications for poor people in developing countries.
This article shows that corruption is structural and omnipresent in Mozambican society, effectively legitimising corrupt practices at all levels with detrimental effects for the country’s development and poverty reduction. On the basis of fieldwork in a number of urban and rural settings it argues that small-scale corruption is so common that it has become an integrated part of daily life for the large majority of the urban and rural poor. While most of the poor relate to corruption through tacit acceptance and acts of compliance, its practical implications are most severe for the very poorest who cannot afford to take part in corrupt exchanges and are excluded from vital social relationships and social services.
The article contributes to the literature on corruption in Africa by making a distinction between what we call ‘illicit’ and ‘habitual’ corruption. While all types of corruption in Mozambique are formally illegal as reflected in elaborate national legal provisions and the signing of international anti-corruption conventions, taking people's own (emic) perspective this is not equally clear: our studies ‘from below’ show that civil servants charging for public services and favouring ‘their own’ is so common that paying for services – or being excluded from them due to inability to pay – has become an integral part of being poor and marginalised and has lost its ‘illegal’ connotations.
The structural or systemic oppression of the poor is ‘deposited’ in individuals in the form of socialised norms or tendencies that guide thinking and behaviour towards acceptance of/compliance with corruption. The embeddedness of habitual corruption in poor shanty towns and villages is reinforced by the lack of interest on the part of public authorities and dearth of legal institutions to hold the corrupt to account. While reminiscent of de Sardan’s (1999) ideas of a ‘moral economy’ and ‘banalisation’ of corruption, the emphasis of the ‘habitual’ is more on the embodiment of systemic oppression and unequal relations of power than on anchoring corruption in ‘ordinary everyday practice’.
Corruption studies usually rest on formal and normative models of corruption, defined as the ‘abuse of public office for private gain’ (World Bank 2002) through embezzlement of public funds, favouritism and nepotism. Corruption research has been dominated by political science and economics, with a focus on political/bureaucratic structures, on the one hand, and individual case studies of corruption on the other (Klitgaard 1991; Johnston 2006; Rose-Ackerman 2007). These studies have typically combined analyses of the ‘ecosystem’ of resources and constraints that surrounds corrupt practices (Mungiu-Pippidi 2015) with ‘principal-agent’ thinking and the possible existence of ‘principled principals’ who are willing to hold corrupt officials to account (Søreide and Williams 2014) – usually with prescribed remedies in the form of various monitoring devices and punishment regimes aimed at reducing the opportunities and incentives for corrupt practices (Persson, Rothstein, and Teorell 2013).
Anthropology, for its part, has a long history of studying informal political and economic exchanges and moral economies ‘from below’, arguing that such exchanges are inherently ambiguous cultural categories (Mauss 1954 [1925]; Graycar and Jancsics 2016). The discipline’s history of dealing with corruption in the formal/normative sense is less established, but the same types of issues have usually been raised:1 what are the conditions that encourage corrupt practices to flourish, and how are such behaviours manifested and interpreted in different contexts (Haller and Shore 2005; Torsello and Venard 2016)? The anthropological focus has generally been on specific local cases of corruption and how corruption is embedded in cultural logics of ‘negotiations, gift-giving, solidarity, predatory authority and redistributive accumulation’ (de Sardan 1999, 25; see also Blundo and de Sardan 2006). Smith (2001), for example, shows how corruption among the Igbo of Nigeria must be understood in the intersection between national structures of power and inequality and everyday instances of patronage as they occur in networks of kin, community and interpersonal association. And Anders (2006) approaches the issue of corruption among civil servants in Malawi not in the official sense of ‘deviant behaviour’, but as part of a set of social obligations based on affectionate affiliations that have a strong influence on people’s social lives and position.
In order to understand corruption as practice, there is a need to combine analyses of the structural/systemic conditions encouraging corrupt practices and corruption as perceived/lived experiences of people susceptible to it. Based on data from long-term quantitative and qualitative research in rural and urban Mozambique,2 this article shows that people distinguish between what we call ‘illicit’ larger-scale corruption and ‘habitual’ everyday corruption, both in their local vernacular and in their agency as citizens. ‘Illicit’ corruption receives considerable attention in Mozambique, not only from ‘watchdog’ institutions, the media and donors (USAID 2005; CIP 2016), but also from people living in urban shanty towns and rural villages. They have come to regard larger-scale corrupt practices as an intrinsic part of the life of high-level politicians or civil servants, who inhabit a different realm than themselves (UTRESP 2010; Wolf and Klein 2016).
People’s relations with habitual everyday corruption are a reflection of structural coercion and the ‘internalisation’ or ‘embodiment’ of corruption as a common practice, reminiscent of Bourdieu’s (1990) notion of ‘habitus’ or ‘socialised norms and tendencies that guide thinking and behaviour’ (see also Ortner 2006; Navarro 2006). There is a set of structuring principles and common schemes that generate representations and practices of habitual corruption. While there is also room for agency and acts of resistance, such options vary with economic and social position and are most restricted for the poor and marginalised. Also, obtaining education and a higher-level job in the public sector is difficult and rare, and usually a result of concerted efforts by family, friends and other contacts, which constitute an intricate system of ‘patronage’ or dependency relationships. In this way, the distinction between an official’s role as civil servant and private citizen, the very basis for the normative definition of corruption, becomes blurred (Haller and Shore 2005; Anders 2006).
Below, we first briefly outline the historic and structural/systemic dimensions of corruption in Mozambique, to place the issue in context and demonstrate its wide prevalence and deep manifestations. In the next section, cases and perceptions of higher-level corruption are presented, to provide a comparative perspective and show how it legitimises corruption also at lower levels. Everyday local corruption is explored in the fourth and main section of the article, first, through exposure of various types of habitual corruption in different geographical areas and sectors to show its proliferation and diversity, and, second, by examining the case of the District Development Fund (FDD) in a region in northern Mozambique. Since 2007, the FDD has injected a total of 70 million meticais (one million euros) into local communities for development and job creation, involving a broad set of public, private and traditional institutions and actors while effectively excluding those who are too poor to take part in corrupt exchanges in the scheme.
Structural context
Mozambique is one of the poorest countries in the world and has consistently ranked among the 10 worst performers in the United Nations Development Programme (UNDP) Human Development Index, which measures income, educational attainment and longevity (UNDP 2017; World Bank 2018). The country has a history of authoritarian and divisive Portuguese colonial rule; a post-independence period under one dominant state party, Frelimo; and a period of brutal civil war between Frelimo and the Renamo rebel movement, resulting in the partial rending of the social fabric. In the most recent period, a de facto strong and centralised state apparatus has been combined with liberal economic policies, which have been pushed through by the World Bank and other donors, and increasingly juxtaposed with corrupt practices throughout Mozambican society (Newitt 1995; Castel-Branco 2014; CIP 2016).
The country’s first president, Samora Machel, had a policy of zero tolerance towards corruption both during the war of independence from 1963 until independence in 1975, and up to his death in 1986: people who committed what was considered to be a serious crime could face capital punishment (Abrahamsson and Nilsson 1995). The strong political stand against corruption, together with better control over what was a fairly transparent society, tended to limit the development of petty corruption (Joseph Hanlon, personal communication). Samora Machel also issued a warning in a speech he gave at the installation of the transitional government in 1974:
We want everyone’s attention on one basic issue: the necessity of leaders living in accordance with Frelimo’s policies, and the insistence that they in their behaviour represent the sacrifices of the masses. Power, and the opportunities that exist around the leaders can corrupt the strongest men. (Quoted in Johansson 2013, 2)
These historical trajectories currently converge in parallel yet interlinked structural systems (Orre and Rønning 2017). From the top to the bottom of the political hierarchy, the party and the state are effectively two sides of the same coin, with the former being essential for careers and the latter a unique source of income with additional benefits in the form of housing, subsidised loans, travel allowances and options for embezzlement.3 Political leaders and senior government officials, from the president down, are also extensively and openly involved in private business enterprises that are reliant on both national and international capital. At the lower end of the political hierarchy (the district, localidade and village in rural areas, and the district, bairro and quarters in urban areas), the party/state nexus is also in play, with influential party members and traditional authorities (chiefs or queens with their sub-leaders) increasingly co-opted by the state through their receipt of a monthly allowance and other benefits.4
Consequently, corrupt practices involving public officials and private citizens are part of an almost all-encompassing political economy, with checks and balances being ‘systemic’ rather than ‘individual’ in nature. Whistle-blowing represents a threat to a systemic culture, and non-compliance effectively shuts people out of the system altogether (Hodges and Tibana 2005; CIP 2016). Added to this is a culture with deep historical roots of submission to, and quiet acceptance of, decisions made by the authorities, be they traditional power-holders, political leaders, public servants or the business elite (Bertelsen 2016).
In terms of legislation, Mozambique has ratified all relevant international anti-corruption conventions and initiatives (CIP 2016). Until 2012, the country’s domestic anti-corruption law (Law No. 6/2004 – see Imprensa Nacional 2004) distinguished between active corruption by private citizens to obtain favours from the state and passive corruption by public servants for private gains that are detrimental to the interests of the state. The law did not include provisions on political corruption, favouritism and nepotism. In 2012, however, Mozambique enacted a package of anti-corruption legislation, which significantly improved on the previous laws. However, there remain significant deficiencies in the enforcement of the law due to lack of capacity, corruption in the legal branch, and insufficient judicial autonomy (Wolf and Klein 2016).
‘Illicit’ corruption
Within the structural context outlined above, national data and case studies have revealed an increase in and proliferation of illicit corruption in Mozambique. The cost of corruption between 2002 and 2014 has been estimated at US$4.9 billion, with serious implications for the state budget and social services (CIP 2016). The seriousness and increasing prevalence of corruption is confirmed in indexes such as the Worldwide Governance Indicators, the Ibrahim Index of African Governance, the Corruption Perceptions Index and the Afrobarometer. In the 2017 Corruption Perceptions Index, for example, Mozambique received a score of 25 on a scale that ranges from 0 (most corrupt) to 100 (least corrupt) ranking it number 153 out of 180 countries (down from 35/100 and 119/180 only three years earlier), while Afrobarometer reported that 48% of the population were of the opinion that corruption had increased and only 20% felt that it had decreased the previous year (quoted in Wolf and Klein 2016, 13).
People use the term corruption (corrupção) for illicit corruption, but there are also colloquial expressions such as bolada (‘jackpot’) for schemes that involve, for example, embezzlement, counterfeiting and smuggling, which are not pursued by the relevant legal authorities and where the perpetrators are left in peace with their takings. They also use expressions such as cabrito come onde está amarado (‘goats eat where they are tied up’), and mamar (to ‘nurse’ or ‘nourish’), to indicate if not acceptance, then at least tacit acknowledgement of the fact that people who have been able to acquire higher positions in the state apparatus use them to enrich themselves, their families and other people who depend on them, or on whom they depend.
Historically, the best-known corruption case in Mozambique, involving US$34 million and a number of senior government officials, relates to fraud by the now failed Mozambican Development Bank (BPD) and the Commercial Bank of Mozambique (BCM) around 2000. These cases included the assassination of Siba Siba Macuacua, an economist and prospective director of BPD, who had threatened to reveal the identity of corrupt officials, and the killing of Carlos Cardoso, a journalist, who was murdered while writing an article about BCM. In the first case, nobody has yet been held to account, and in the second, convictions have led to imprisonment, but the money has never been recovered. There was initially a strong reaction from the donor community, but in the end they failed to take action (Fauvet and Mosse 2003).
Furthermore, it is common practice in Mozambique for senior politicians, military personnel and other state employees to be directly involved in negotiations between foreign interests and private Mozambican companies. In 2011, the government, through the ministry of transport and communication, bought 250 buses from an Indian company. These were to be distributed throughout the country, but while 50 were delivered to the Maputo municipality, the rest were never put to use. The president of Mozambique was the main shareholder of the subsidiary of the Indian company, which received US$20 million for the purchase and maintenance of the buses (Indian Ocean Newsletter 2012). While people knew of the president’s involvement in private business ventures, such practices were not openly questioned, partly because of his position and authority.
There is also widespread corruption in local and intermediate public entities in Mozambique. In a case involving directors of education in the provinces of Nampula and Cabo Delgado from 2011, it was discovered that salary sheets included a large number of ghost staff (fantasmas). The salaries were being channelled directly into the accounts of the provincial directors, as well as to heads of district-level departments (CIP 2012). This case is well known among teachers in the provinces, who live and work in often remote localities without proper housing and other facilities, while earning a salary considerably lower than that of the directors. Cases such as these are actively referred to when lower-level public employees defend their own practices.
Recent developments have put the challenge of poor governance and corruption in Mozambique into an even starker perspective (Martini 2012; CIP 2016). Undisclosed foreign loans discovered in 2016 – including the so-called EMATUM case in which the government took up a loan of approximately US$850 million in the name of a non-producing publicly owned tuna-fishing company and a similar scheme of US$1.2 billion – have contributed significantly to a total current (2017) public debt of an estimated 78% of GDP. Not only has this debt hampered the growth of the Mozambican economy and led to sharp declines in government spending and investments, but it has also led to a crisis of confidence and a halt in donor payments (Orre and Rønning 2017).
Corruption in the corrupção sense is endemic at all levels of government. People in positions that enable them to engage in corruption in these institutions are well aware that this practice exists from the very top to the very bottom, and there are few examples of action being taken by police and judicial institutions, donors or whistle-blowers, that have had any positive outcomes in the fight against corruption, for the institutions or for individuals. Within such a structural context, the top echelons continue because they have little to fear and they support each other, and the lower ranks use the practices of their superiors to defend their own corrupt practices. They also often come under pressure from peers and family to provide them with something ‘in return’ for having been given the opportunity to be educated and thus to achieve public office.
Classical monitoring devices and punishment regimes used for larger-scale ‘illicit’ corruption are ineffective in this context, because there are no actors with the incentive or means to hold corrupt officials to account (Persson, Rothstein, and Teorell 2013). Even with the most recent scandal involving undisclosed foreign loans, referred to above, there are signs that donors will eventually give in to Mozambique, because it is regarded as a ‘donor darling’. Arguments have been put forward to the effect that the two international banks that provided the loans ‘failed to exercise sufficient due diligence, which would have shown that the loans were not government guaranteed, prices had been inflated, project revenues were unrealistic and the IMF was not informed’; that there are strong grounds for Mozambique to repudiate the debt; and that donors should not take further action (Wolf and Klein 2016, 4; Hanlon 2019).
‘Habitual’ corruption
Even though larger-scale corruption has detrimental implications for the political economy of Mozambique and for the population at large, people do not perceive it to be part of their everyday life. They generally do not understand the link between embezzlement of public funds and public investments, and many have not heard – or do not believe – that political leaders ‘steal’ (roubam). Rather, people in rural villages and urban shanty towns are constantly susceptible to demands for a wide range of smaller, informal ‘extra-legal’ transactions for public services, which affect them more directly and effectively contribute to make corruption ‘habitual’ – or ‘banalised’ as de Sardan (1999, see also Blundo and de Sardan 2006) would argue.
According to Afrobarometer (2016), 34% of respondents who had used a public service in the past 12 months had paid bribes to a public official. For the poorest, even small ‘token appreciations’, such as 10 meticais (0.2 euros) to have a child move up a class, gain access to life-saving medicines, or obtain the identity card necessary for formalising land ownership, may be too much for the household budget. At the same time, we know that many poor households do not use public services, either because they are not accessible or because they cannot afford to pay (Tvedten, Tuominen, and Rosário 2016). According to the Mozambican Centre for Public Integrity (CIP 2016), 68% of the population has been confronted with requirements to pay bribes.
People use the term refresco (literally, ‘soft drink’) for these transactions, and see them as an intrinsic part of everyday life which just has to be dealt with. Everybody knows that the people who request a refresco are actually asking for a contribution in cash or – more rarely –in kind. The notion of refresco is internalised or embodied through structural oppression of people at the bottom of the social hierarchy, and is seen as an inevitable part of their efforts to improve their lives. Below, we give examples of the everyday corruption encountered during our research into poverty and well-being in different parts and social formations of Mozambique in order to indicate its prevalence and scope. We then take a more comprehensive and integrative approach to the everyday corruption people encounter as citizens of local communities in the district of Lago in Niassa Province.
One of the most common and detrimental examples of everyday corruption in Mozambique is a widespread system in which people have to pay intermediaries in order to gain formal employment. In the country’s capital Maputo and its urban district of KaMaxakeni, for example, security guard positions are popular (Paulo, Rosário, and Tvedten 2011). Men seeking employment do not go directly to the head office of the relevant company (many of which are owned by senior government/Frelimo officials), but must identify particular employees who charge the equivalent of one or two months’ wages to facilitate the application process, usually in cooperation with those in management positions or the owners. This has come to be seen as part of the process of seeking lower-level public employment, but the practice effectively shuts out the poorest and most marginalised from the formal labour market who are not able to establish such contacts and/or cannot afford to forfeit income.
In the district of Buzi in Sofala Province (Picardo, Tvedten, and Paulo 2012), the Institute of Social Security (INAS) has local representatives whose job it is to identify needy people who meet the requirements for support and to organise monthly payments to them in the villages and bairros. When INAS representatives came to one of these villages with money to distribute, the local representative had made a list of 60 older people. However, what the INAS representatives did not know was that half of those who turned up were people other than those named on the lists (elderly people here rarely have identity cards): these people attended instead on condition that the local representative would receive 50% of their payment. Because the representative was also a traditional leader, no one openly questioned why many of the older people did not attend the meeting, and many of the elderly people themselves see their non-inclusion in the public social protection system as part of their being ‘poor and destitute’ (umbwa in the local vernacular).
In the district of Mossuril in Nampula Province, not everyone attends school and even fewer visit the local health unit (Tvedten, Paulo, and Tuominen 2009). A key reason for this is that although both services are in principle free of charge, in fact people have to pay to pass examinations, to be moved up a class, for maternity care and delivery of babies, and to buy such coveted items as mosquito nets against malaria, and contraceptives against transmission of HIV/AIDS. Many are unaware that these extra payments are illegal and seem to accept them as part of public services. Some who are aware accept the terms on the grounds that education and health workers earn very little and would leave a remote place like Mossuril were it not for the option of earning a little extra. The people who suffer most from this practice are poor women, who are de facto responsible for their children’s education and health, but who cannot pay the ‘entrance fees’.
Finally, in the district of Chokwe in the province of Gaza, the chief or régulo is responsible for administering land on behalf of his community under the current system of communal land holding (Tvedten, Paulo, and Tuominen 2010). People are not supposed to pay for their plots (machambas) but should be allocated land on the basis of community belonging and need. As in many other parts of Mozambique, however, there is significant pressure on land arising from commercial agricultural interests, extractive industries and tourism (many such companies, again, are owned or co-owned by higher-level public servants/Frelimo cadres). Interested enterprises and individuals tend to contact the local régulo, who in the case of Chokwe sold large tracts of the most fertile land to a sugar company without involving or informing the local population. Given the position of traditional authorities in these communities, nobody openly opposed the sale: the acts of the régulo were not questioned, and people feared that complainants would end up with no land at all.
In all these cases, the combination of long-term social oppression and marginalisation makes poor men and women embody and perceive corruption as an inevitable part of everyday life. They do not openly question such payments, because they know that such an action would result in them not getting access to the service they need and that they would be likely to remain marginalised from or excluded by the public and traditional authority in question. In some cases, people also acknowledge that in order to retain poorly paid teachers and health workers in marginal rural and urban areas, they must offer some level of refresco. The alternative, in the Mozambican context of a dearth of qualified personnel, is that the community would be left without any such services.
Perspectives from below
While the examples of habitual corruption outlined above may leave the impression of separate and individual cases, an important aspect of locally based small-scale corruption is the way it forms an integrative whole. Corruption seen from ‘below’ and the perspective of the poor themselves – rather than from ‘above’ through pre-defined categories, which are the focus of most corruption research – forms a unified system, in which it is difficult to know where the flow of ‘corrupt’ practices begins and ends.
One such local setting is a string of 12 villages in the district of Lago, Niassa Province. The province is among the poorest in Mozambique (INE 2015) and has a political organisation with districts and administrative posts heavily controlled by Frelimo. The villages have a total population of 8000, most of whom are farmers, fishermen and small-scale traders – with a few successful local entrepreneurs usually in the transport and building sectors. They have a Muslim, matrilineal and polygamous socio-cultural organisation, implying hierarchical structures albeit with a relatively independent position of women (Tvedten et al. 2012; Tvedten and Orre 2016).
Following de Sardan (1999) on the importance of understanding local language constructions of what corruption ‘means’, the local euphemism for ‘corruption’ among the Nyanja living in the area is chiphupu – which refers to lack of ‘openness’ rather than misuse of funds. The national District Development Fund was initiated in 2007 and has become the main source of money in the communities and the main source for investment in small-scale businesses. Official figures show that a total of 9,961,000 meticais – a considerable amount in this poor area – was allocated between 2007 and 2015. Formally, priority is to be given to associations and women, rather than individuals and men. Most of the funds to date have been granted for projects in artisanal fisheries, various commercial outlets and (to a much lesser extent) agricultural projects in a context where agriculture is essential for the poor and for poor women in particular. The sums per project vary between 70,000 and 177,000 meticais – or 14 to 34 times the average annual household income (Tvedten, Tuominen, and Rosário 2016).
The selection of beneficiaries for the scheme is through a process in which people in the communities are expected to develop a project proposal consistent with a set of criteria, including potential for ‘income generation’ and ‘employment creation’. Applications are to be vetted first by a traditional leader (régulo or raínha) or a neighbourhood secretary (secretário do bairro) whose primary responsibility is to assess the ‘competence and capacity’ of the applicants; the applications then go to an elected local consultative council consisting of ‘influentials’, such as small-scale businesspeople, teachers and nurses, who are to assess the potential of the application; finally, they are surveyed by the head of the administrative post for ‘technical approval’. The applications that make it this far are then sent for a concluding assessment at the district level by a district consultative council and the district administration. The system is thus open not only to ‘embezzlement of public funds’, but also to ‘favouritism’ and ‘nepotism’, to use formal corruption terminology.
In following the application processes in a number of village households,5 the first encounter with refresco at the community level comes with the need for help in formulating the written application for which fees are often charged, or a portion of possible future funding demanded, despite the fact that the service is in principle free. Once an application has been crafted, it is handed to traditional leaders (queens or raínhas in this case), who tend to favour applications from their own matrilineage or clan in accordance with tradition. In the case of bairro secretaries, who are de facto political appointees of the ruling party Frelimo, political affiliation counts and tends to exclude many people in a situation where the opposition (MDM and Renamo) has strong support. Finally, the influentes in the consultative council tend to favour peers or ‘business-associates’, sometimes explicitly, on the grounds that they will be able to cooperate at a later stage.
In all cases, the ‘exchange’ is a combination of the applicant’s material contribution and political allegiance, which feeds into future claims for social security and protection. This is similar to traditional systems of exchange associated with the extended family, the clan, traditional authorities and religious institutions, for example, except that money and political allegiance have emerged as a de facto ‘entrance fees’ at the expense of community- and family-based affiliation. The results have been that the poorest, who usually have neither the money nor the necessary social contacts, are effectively excluded from the scheme;6 that villages where political opposition is strong are completely left out; and that women represent a clear minority among the beneficiaries. Still, at this level, few openly complain about the decisions made and seem rather to ‘internalise’ them as ‘the order of things’.
As a case in point, Mwnaiba Amido7 is the first wife of a successful fisherman who is also an ‘influential’ in the village and a member of the consultative council. Mwnaiba received a contribution of 150,000 meticais from the District Development Fund, the first woman to do so, and she started a business together with two other women. They initially cooperated well, buying fish, drying it and reselling it in Metangula (the district capital) and Lichinga (the provincial capital). However, after a while the business began to decline, and the partners managed to repay no more than half the money. Contrary to a regulation that no new loan should be given before a former loan is repaid, Mwnaiba received a further contribution from the District Development Fund of 100,000 meticais, and now buys dried fish on her own and sells it in the market in Lichinga, where prices are higher than in Metangula. While people acknowledge that her influential husband is the main reason for her getting support, nobody questions her right to this type of favour and she is respected for what she has done.
Moving on to the lowest level of government, the administrative post, the heads of post tend to be recruited from outside the area in which they are to serve – ostensibly to avoid collusion with extended families and other close affiliates. Being on the ‘frontline’ of the state, so to speak, these officials have a difficult task, with few resources and limited decision-making powers. In order to develop their power at the local level, they need to assert their influence carefully by making use of clientelistic networks. In this way, ‘corruption becomes the space in which the State dissolves at local level and is replaced by a plethora of socio-cultural practises and relations’ (Torsello 2011, 8).
In the communities in question, this has partly been achieved by channelling resources to power-holders rather than to the poorest: for instance, by upgrading the house of the influential raínha rather than building water posts in marginal communities. By the same token, the assertion of power by heads of the administrative posts is most easily achieved with respect to villagers with some means. Local civil servants are known to charge up to 3000 meticais as a precondition for even receiving a proposal for the District Development Fund – which the very poorest cannot afford – and even then there is no guarantee the applicant will eventually obtain project funding.
To our knowledge, nobody has openly or publicly protested about the practice either to higher political levels or to the public employee himself. This seems to be the combined result of the power logic in relation to authorities, and fear of losing possible funding in the future (funds are released annually). Mário Mbaia, for example, along with three other people, wrote a proposal for a project with a total value of 50,000 meticais on the commercialisation of fish. The idea was to work in two groups, with one buying and drying and the other selling in Metangula. The business was to be developed by investing the surplus in fishing nets and, eventually, in a boat, to make the men ‘real’ fishermen. However, the project never made it past the Posto Administrativo because, according to Mário, ‘it is all politics.’ Mário was particularly frustrated because he had paid a ‘contribution’ (chiphuphu) of 3000 meticais just to be part of the process. He did not receive a proper explanation for the rejection of his project and has not tried his luck again. Instead, he has managed to buy nets through his own efforts and still works towards his goal of becoming a ‘real’ fisherman.
At the higher, district administration level, decisions are made on the basis of a combination of technical merit and ‘personal qualification’ – often a euphemism for the right political affiliations and strong standing in the community. In principle, decisions are taken by the district consultative council – by representatives from government, civil society and district ‘influentials’ – but the final ‘technical’ assessment is made by the district administration. In addition, a number of project applications are made by people outside the district, effectively reducing the number of local applications accepted for funding. These are often made by politicians or higher-level bureaucrats in the province or come from influential businesspeople. The final number of projects approved per year in the district of Lago is usually about 60, or what is believed to be about 10% of the total applications made. Many projects end up getting only part of the money applied for, with the district administration claiming it did not receive all the ‘seven million’ envisaged from central and provincial government – but with popular opinion being that the money is either redirected to other purposes or stolen.
Corruption and embezzlement have thus effectively strangled a programme that, at its outset, was very important, in a context where lack of access to funds for investment is a major constraint on development and poverty reduction. The people who have benefited are primarily those who already have a strong position in the district and communities, and nearly all are men. At higher levels, among public servants and economic power-holders, the programme is viewed as a welcome ‘contribution’ by a state that pays them poorly and gives them few tools for implementing policies. In this sense, the situation represents a form of ‘competition’ between central state and local government.8 Among beneficiaries in the communities, the programme represents vital additional funds for further strengthening their economic enterprises and position. Perhaps the clearest indication of the ‘common interests’ among those on the ‘inside’ is the dismal record of repayment of funds, which in principle is to be done within 24 months at an interest rate of 1%. After nine years, only 3.5% of allocated funds have been repaid at district level, and 7% in the string of villages, and no real effort seems to have been made to remedy the problem, even though it threatens the existence of the entire programme (Tvedten, Tuominen, and Rosário 2016).
In local rural communities such as those under discussion (and around 70% of the Mozambican population live in such places), there is a strong tradition of adhering to authority relations (traditional leaders vs ‘commoners’; men vs women; old vs young; people with formal positions and education vs the poor and uneducated; religious leaders vs the congregation, for example), and systems of exchange of services and prestation have deep cultural roots. However, there is a qualitative difference between an exchange for undefined outstanding claims (part of the harvest for help in times of need, or physical protection for political support, for instance) and an exchange of money. Most people understand that they are being exploited when they have to pay for services that should be free, or when they are excluded from programmes such as the District Development Fund because of poverty, gender or political affiliation. However, it is still uncommon for poor people to openly oppose local power-holders and public officials. Instead, people tend to withdraw and accept their marginal position as an inevitable aspect of being poor and/or women.
The challenge for breaking the system of ‘habitual’ corruption is the intimate linkages among the various authorities, patronage and dependency. Traditionally, people would go to the head of the extended family or clan with worries and complaints of this sort, but poverty and monetisation have weakened extended family links and many households find themselves isolated from vital social relationships. Traditional authorities should be the link between their own populations and external entities (such as the state), but they have largely been co-opted through monthly allowances and other benefits and find it increasingly difficult to oppose a state that feeds them and gives them most of the funds upon which they base their authority. Finally, there is no socio-cultural tradition and social space for the poor and marginalised to bypass institutions and people of authority.
However, there are examples of attempts to hold the corrupt to account by challenging the structural and cultural constraints on their agency. In the string of villages, for example, a small group of women openly objected to the lack of contraceptives at the local health post at the time the first HIV/AIDS cases were discovered in the community, and at the unavailability of mosquito nets to protect their children against malaria. They insisted that this was because health personnel had sold the contraceptives to men, who saw their use as an act against God, and had sold the mosquito nets to fishermen, even though both were supposed to be free and for the protection of the community. Nothing happened until the day the women voiced their concerns in speech and song during a rare visit by a senior politician’s wife, who is said to have reported the incident as a gender and mother–child issue rather than as ‘corruption’ per se. The practice of selling contraceptives and mosquito nets to those other than the intended beneficiaries has, at least for the time being, stopped.
Conclusions
Corruption in Mozambique has high costs: for the state that is tapped for vital resources that could be used for development and poverty reduction; for donors, who see part of their funds misused, and lose credibility with home constituencies; for the poor, who have to use part of their meagre resources to pay extra for goods and services; and for the ultra-poor, who are effectively excluded from services and relations with the state because they cannot pay what is necessary. Corruption is pervasive and affects practically all Mozambican households in one way or another: indirectly, by sustaining an elite-dominated economic system and diverting public resources, and directly, by constraining access to public resources and limiting choice.
At the same time, both larger-scale ‘illicit’ and smaller-scale ‘habitual’ corruption are largely internalised as part of the existing political, economic and socio-cultural order. This rests on the fact that dispositions made by higher authorities – from the president to traditional leaders – are rarely openly questioned. There is a widely accepted or embodied perception that authorities have a right to make decisions on behalf of the population. Moreover, higher-level public positions provide people with opportunities to extract the rents that are essential to maintaining the patronage system upon which politics in Mozambique is based. Classic anti-corruption measures fail for the simple reason that there is an absence of stakeholders willing or able to act as ‘principled principals’ and hence to enforce existing laws and policies.
At lower levels and among poorly paid teachers, nurses, security guards, waste-collectors and others, smaller-scale everyday corruption or refresco is in many ways what ‘makes the system tick’. It is pragmatically accepted because it facilitates life, either by maximising efficiency in achieving objectives, or by minimising risks or the chances of getting into trouble with, for instance, the police or courts (Persson, Rothstein, and Teorell 2013; see also Nystrand 2014). Without such contributions, the services might become even poorer, since employees may lose motivation and interest in their jobs and certain areas might be left without any qualified personnel. In this way, corruption may be seen as the means by which people make sense of politics and the state, or a form of discourse to achieve particular benefits that are scarcely allocated (Bubandt 2006). While smaller-scale corruption may not be considered ‘legal’, along with gift-giving, brokerage, solidary networks and redistributive accumulation, for example, it is, however, susceptible to ‘moralities’ that differ from those in play in larger-scale and higher-level corruption practices.
In fact, the most immediate and serious problem relating to small-scale corruption affecting the poor is the way it excludes the poorest sections of the population who cannot pay for basic services and social relationships. Unable to pay even the smallest ‘entrance fee’ or to provide other amenities, the poor and destitute are effectively marginalised and do not have adequate access to health, education and the other services needed to improve their lives. And their poverty and lack of voice makes them accept the endemic system of refresco from which they are firmly excluded as an inevitable aspect of being poor.
Analyses of corruption from ‘above’, and the use of formal and normative definitions of ‘embezzlement of public funds for private gains’, capture neither the complexity nor the variations in perceptions and practices of corruption in countries such as Mozambique. Rather, there is a need to combine analysis of the political, economic and socio-cultural structures making corruption possible with an anthropological approach to corruption as lived experiences particularly among the poor who are most directly and dramatically affected by it.