MEDIATING BUREAUCRATS

The concept of ‘mediating bureaucrats’ has not been widely used or defined, despite occasional references to it here and there. For example, Chambers (Reference Chambers and Farmer1977: 365) makes casual reference to the existence of a ‘mediating bureaucracy’, in contrast to situations in which there is no organised bureaucracy in rice-growing communities, but he does not provide any clarity regarding what the notion refers to more broadly, except that it relates to ‘the state’ in one way or another. Torry (Reference Torry2016: 21) comes closer to a useful definition when, in a recent work, he discusses ‘institutions that mediate’, thus facilitating specific relationships. He is concerned with ‘mediation’ between secular and religious institutions or organisations that are sometimes also referred to as structures. Even through Torry seems to conflate institutions and organisations, as is common in the social sciences (see Douglas Reference Douglas1986 for a critique), more usefully he points to the importance of mediation as a particular kind of practice that resolves conflicts in society. Who or what that entity is or is related to is a question Torry leaves open, which is useful for our own understanding of the term ‘mediating bureaucrats’, which is also premised on a Weberian understanding of bureaucracy.

Drawing on Weber, Evans (Reference Evans1995: 12), in his important work on ‘embedded autonomy’, pointed out that the state and its bureaucrats cannot be kept autonomous from the economic sphere if successful innovation and economic development is to be achieved. States and their bureaucrats must respond to ‘demands [for] more intimate connections to the private economic agents, a state that is more “embedded” in society than insulated from it’ (Evans Reference Evans1995: 32). As we shall argue later, this is an important observation, but the significance of mediation in the literature on the state and business is primarily focused on the ‘embeddedness’ of state bureaucrats in the sense of their institutionalised engagement and relations with industry actors. For obvious reasons we are also inspired by Evans, but his understanding of bureaucrats is that in essence they are ‘external’ to the economy, whereas we emphasise movements between party, state and industry in different constellations where what is inside and outside is less clearly demarcated. This leads us to the social embeddedness of actors.

The embeddedness concept has been interpreted and used in various ways, but we draw primarily on Mark Granovetter's (Reference Granovetter1985) understanding, which focused on actors’ embeddedness rather than societies’ structural embeddedness. By highlighting the social embeddedness of all societies and economies, he also provided a way out of the dichotomies between traditional and modern societies common in embeddedness analysis in a Polanyian tradition (Block Reference Block2001; Krippner et al. Reference Krippner, Granovetter, Block, Biggart, Beamish, Hsing and O'Riain2004) meaning that ‘economic actions and actors always have to be understood in relation to the context in which they are embedded’ (Nystrand Reference Nystrand2015: 8). Granovetter's actor-focused approach to embeddedness is based on a critique of both over- and under-socialised views of economic behaviour and suggests that actors are embedded in ‘concrete, ongoing, systems of social relations’ (Granovetter Reference Granovetter1985: 487).Footnote ⁶ The mediating bureaucrats are here seen as such socially embedded actors, who embody the experiences of taking up formal positions in the state, the government, business and party and thereby become mediators between these social realms.

One important context in which meditating bureaucrats act in our understanding is related to what Lipset (Reference Lipset1960), drawing on Weber (Reference Weber, Roth and Wittich1978 [1968]), focuses on: the bureaucracies themselves. That is, Lipset concentrates less on the institutional aspect and more on the bureaucracy and the individual bureaucrat within it, arguing that bureaucracies are major integrating mechanisms in society. This is because their members are in the nature of ‘an impartial expert rather than an interested party’ (Lipset Reference Lipset1960: 52) due to their allegiance to Weberian bureaucratic norms. As such they are important in mediating conflicts in society, as they take on a mediating role in the way they act. Here Lipset follows Weber in arguing that this provides some stability during social change or other transitional periods, as bureaucratic procedures ‘tend to reduce conflicts to administrative decisions by experts, [thereby effacing] issues from the political arena’ (Lipset Reference Lipset1960: 52). A bureaucracy's mediating role is in part due to what we might call the ability to depoliticise contentious decisions, in our case to implement contentious policies by mediating and translating between ideological positions so that resistance to reforms is smoothed.

Here it is important to remember that Weber not only discussed state bureaucracies, more generally he described and discussed the key characteristics of different ideal-typical forms of public administration, government and corporate business. In this modern way of understanding bureaucracy more generally, an administrative system governs any large society in both its public and private aspects. A bureaucracy can therefore refer to the centralised hierarchical structure of a public administration as much as to a corporate business or a multinational company. More broadly, the ideal-typical ‘bureaucracy’ Weber referred to, in both the public and private sectors, as well as in different combinations, was characterised by being ‘fundamentally [related to] domination through knowledge’ (Weber (Reference Weber, Roth and Wittich1978 [1968]: 225). This knowledge involved understanding the hierarchy of the organisation, with its formal as well as informal lines or chains of command related to fixed areas of activity. However, Weber and most of those who have studied bureaucracies have emphasised the formal aspects of the rigid division of labour, with bureaucrats involved in the regular and continuous execution of assigned tasks.Footnote ⁷ Keith Hart (Reference Hart2008), by contrast, has repeatedly pointed out that, in the shadow of the formal aspect of bureaucracy, with all its benefits, we usually find bureaucrats with an intimate understanding of the informal working of a given bureaucracy who simply could not operate based on the formal procedures alone. In Hart's (Reference Hart2008: 11) own words: ‘The formal and informal aspects of society are already linked, of course, since the idea of [the informal] is entailed by the institutional effort to organise society along formal lines. “Form” is the rule, an idea of what ought to be universal in social life’.Footnote ⁸

However, the formal sphere of society in all its public and private aspects does not just exist in the abstract, it ‘consists also of the people who staff bureaucracies [in public and private entities] and their informal practices’ (Hart Reference Hart2008: 12). It is this double aspect of bureaucracy we want to capture, using the concept of ‘mediating bureaucrats’ who can translate between the different political and economic ideologies and logics in which they are socially embedded. This implies navigating the uneven terrain of not only formal rules and regulations, but also what Blundo & Olivier de Sardan (Reference Blundo and Olivier de Sardan2006: 5) refer to as the ‘tacit codes and practical norms’ at play in public services. They pointed out that following a formal rule guiding social activities is one thing, understanding the ‘tacit codes and practical norms’ allowing bureaucrats to translate between different political and economic ideologies and logics quite another. Although Blundo & Olivier de Sardan explored everyday corruption in public services in West Africa, what we consider important is mediating bureaucrats’ ability to translate and navigate between the formal and informal, between different rules and ‘tacit’ codes.

The tacit aspect of bureaucrats’ ways of operating was also pointed out as important by Buur et al. (Reference Buur, Baloi and Mondlane Tembe2012b), who identified the emergence of an ‘embedded and mediating bureaucracy’ as crucial in connecting the state to business interests in Mozambique. The authors suggested particularly that the cadres had embodied knowledge – a certain ‘knowing-how’, while also ‘knowing-that’ (Ryle Reference Ryle1949) – for example, that ideology and political projects mattered to party and government officials. They were therefore able to draw on both intimate tactical forms of knowledge and the rationales and concerns guiding the government, the state and industry.

In line with these arguments, by ‘mediating’ we mean that bureaucrats can translate and liaise between business and political interests, technical and technocratic concerns, and administrative and legal procedures. This clearly requires translating multiple political, ideological and pragmatic concerns without necessarily letting official policies be ‘reinterpreted beyond recognition’ by the bureaucracy (Kaviraj Reference Kaviraj1990: 13). In Mozambique, bureaucrats’ historical experiences of being socially embedded in the party-state, government and private industry on the one hand, and their understanding of the enmeshing of these domains on the other, gave them a unique position for mediating and translating.

Besides translating, we wish to point out one further way of understanding mediating bureaucrats that is intimately related to ‘knowing how’ and tacit forms of knowledge, but which also focuses more on pragmatism. Often the experiences of the cadres or bureaucrats we encountered when studying the rehabilitation of the sugar industry and the liberal reforms in which it was embedded can best be described as ‘muddling through’ (see Lindblom Reference Lindblom1959, Reference Lindblom1979; see also Buur Reference Buur, Kjær, Pedersen and Buur2015 for a more recent use of this concept). The Cambridge Dictionary defines ‘muddling through’ as ‘to manage to do something although you are not organised and do not know how to do it’, while the Merriam-Webster Dictionary suggests that it means ‘to achieve a degree of success without much planning or effort’ (see Buur Reference Buur, Kjær, Pedersen and Buur2015). For Lindblom (Reference Lindblom1959, Reference Lindblom1979), ‘muddling through’ or ‘incrementalism’ is a way of trying to find a middle way between a rational actor model and bounded rationality – neither long-term, goal-driven policies and analytical rationality on the one hand, nor laissez-faire approaches on the other.

In the context of this article, muddling through should be understood as a way of working in and through a state-based and state-controlled planning apparatus – a top-down exercise – as well as a de facto need to take a gradual approach to addressing particular problems based on the embodied knowledge that success in economic planning is rarely reliant on blueprints for action. As the cadres we focus on here became astutely aware, planning is rather a result of contingent processes of implementation and unintended consequences (see Buur et al. Reference Buur, Mondlane and Baloi2011, Reference Buur, Baloi and Mondlane Tembe2012b). This is a modus operandi that comes closer to the Lévi-Straussian figure of the bricoleur (Lévi-Strauss Reference Lévi-Strauss1966), with its focus on addressing concrete problems by combining different elements in new constellations to produce new solutions through an iterative process based on the resources, knowledge and experiences at hand.

We argue that this group of cadres – the ‘mediating bureaucrats’ – cannot be defined in simplistic and one-dimensional terms as belonging to the state, the government, business or party, as is commonly assumed. Instead the ‘mediating bureaucrats’ in charge of the rehabilitation of the sugar industry and the liberal reforms in which it was embedded were people who, over the last decade before the privatisation reforms, had taken up formal positions in all these social realms in different configurations. But just because they are socially embedded, as Nystrand argues, ‘they are not following norms slavishly, but continuously negotiate with and reflect on social norms’ (Nystrand Reference Nystrand2015: 9).Footnote ⁹ The mediating bureaucrats’ embeddedness in different social realms means that they negotiate and reflect on norms embedded in these varying realms, and thereby embody the capacity to mediate between state, government, business and party.

As we shall discuss in the following sections, the existence of mediating bureaucrats during the post-1990 liberal reforms in Mozambique was important because of the considerable mutual mistrust between the actors involved. On the one hand, parts of the state bureaucracy and the political elite did not trust the intentions of foreign investors and were apprehensive of the effects of allowing them to run former state industries. On the other hand, in most cases the foreign investors who began investing in Mozambique after 1992 had limited experience of working with a de facto one-party state apparatus and political system.

MUDDLING THROUGH MULTIPLE EXPERIENCES

An important characteristic of the sugar sector is that all the bureaucrats and leaders who were in charge of the rehabilitation of the sugar industry, including the union leaders we interviewed at the national and provincial levels, had been working in it in both ‘private’ and ‘public’ capacities since the 1970s. In different constellations, they had been educated by and/or worked for colonial-era private sugar estates, the post-colonial centralised state and the Frelimo party as cadres before taking up positions in the party, the state and privatised companies after rehabilitation was initiated at the end of the civil war in 1992. This group of cadres had acquired their specific skills from those experiences, what the sector had achieved in its heyday during colonialism, when it was the biggest formal employer in Mozambique, the leading sector with the biggest company (the colonial-era Sena company, whose core industry was sugar production controlling two sugar estates) and the third biggest exporter and foreign-currency earner. They had also experienced the struggles of the late 1970s under the centralised state system, when the six sugar companies were either nationalised or subject to other state intervention, as well as the total breakdown of the industry during the 1980s, when the civil war intensified. They had also been part of the unsuccessful attempts to rehabilitate the state-owned industry using foreign loans from, for example, the African Development Bank, at the end of the 1980s to the early 1990s (see Buur et al. Reference Buur, Mondlane and Baloi2011, Reference Buur, Baloi and Mondlane Tembe2012b). These multiple and overlapping experiences of success and failure became a resource during the rehabilitation of the industry that was embedded in the liberal reforms that took shape after 1990.

The story told to one of the authors in 2011 by a former director of cane production on one of the sugar estates in Sofala province in Mozambique captures well the multiple experiences that informed his work. His experiences spanned both the colonial and post-independence eras, when the sugar industry was state-owned, after which he took up positions at the sugar institute in Maputo and at the provincial level in Sofala, where he was director when we met him:

Being a provincial line-ministry director today was in large part made possible by the colonial sugar company and the schooling it provided, as well as the experiences this cadre acquired within the framework of the post-colonial party-state. Like other cadres who had come out of the twin experiences of the colonial and state-owned sugar industry, he saw the demise of the sugar sector as tragic: these cadres knew it could have been avoided with a mixed economy, and they suggested that, had the post-colonial state pursued its centralising policies less rigidly and allowed a greater degree of flexibility, it might have been enough to meet changing market conditions (world market conditions seriously deteriorated after 1974, when sugar prices plummeted). Flexibility did not apply to the international market alone but also to responses to internal conditions such as the civil war, labour incentives and productivity, and the production, organisation and mobilisation of inputs.

The former director of the Maragra estate in southern Mozambique after independence, when the state took over after roughly 300,000 people of predominantly Portuguese origin were in part forced out and in part fled out of fear of the incoming post-colonial regime (see Weimer et al. Reference Weimer, Macuane, Buur and Weimer2012; Sumich Reference Sumich2018), suggested that these problems were nested in one another:

This particular cadre, who had been trained as an agrarian economist, only to be given the daunting task of managing the newest of the six sugar estates and mills when productive units declined as skilled people and owners abandoned the industry, suggested that, despite such problems:

The cadre later moved out of Mozambique and relocated to Portugal, where we met, but he maintained close contact with other former cadres. He was well aware of developments in the industry and had been very supportive of state control of it after independence and its aim of changing the rural countryside to be more accommodating of Frelimo party ideas (see below for a discussion of the New Society). But this cadre was also very clear about the challenges faced by this state-directed industry.

As the above account of their experiences clearly suggests, these cadres, who had experience spanning several decades as industry, state and party actors, had a keen eye for the problems a state-based system could encounter from a market-driven procurement system. On one level they knew what the aim was and what was required to increase production, or even just to produce, and also how difficult this could be in a highly centralised state system. They were aware of the policies and aims involved and were highly motivated, but they also knew that none of this guaranteed success. At another level, even if these cadres did not fully understand the operation of the international market, then at least they had some experience of its modus operandi and the constraints involved in trying to market sugar in an international system. Although ostensibly free and equal for all, this market was actually characterised by high levels of protection by some countries, while ‘dumping price’ practices undercut the viability of others’ production (INA 1996; LMC & Global Sugar Consulting 2000).

We suggest that the ability of people like these cadres to ‘muddle through’ with regard to policy changes from the colonial via the post-colonial to the liberal reforms of the 1990s is absolutely key for understanding why policies in the sugar sector were actually implemented, in contrast to other sectors that were promoted just as much, but fared far less well (see Pitcher Reference Pitcher2002 for cotton; Buur et al. Reference Buur, Mondlane and Baloi2011 for cashew). This group of cadres had intimate knowledge of success and failure they could draw on, as they encountered new obstacles and needed to find pragmatic ways to deal with real-life policy decisions after liberalisation in 1994. What makes it important is that they captured both the how and the what of expanding industry interests and needs on the one hand, and the diverse and often contradictory mixture of state-party and government political imperatives on the other. These were concerns and problems that would emerge with the reform of the sugar sector and the wider political and economic liberalisation reforms in which the sector was embedded. Here the key to muddling through in unfamiliar territory was to use the experiences these cadres had acquired while straddling the colonial and post-colonial eras in an ad hoc fashion in the new context of the liberal reforms.

These cadres had seen the sector fall victim to dumping prices in trying to rely on sugar exports for revenue purposes during the post-independence period, and they had also come to realise that all sugar-producing countries base their production on two pillars: internal markets and external trade agreements (see LMC 2004; Buur et al. Reference Buur, Mondlane and Baloi2011). Entering into such trade agreements with the USA and the EU in particular, as well as with east European countries, did not just happen by itself but required coordinated international lobbying by both the industry and the Foreign Ministry, which formally signed such agreements.Footnote ¹⁰ The changing positions and shifting constellations of party, state and industry personnel among the mediating bureaucrats meant that cadres in charge of the reforms had tacit knowledge of how to work within changing policy environments.

For example, even though ideologically the trade union president for the sugar industry was aligned with the old guard of socialists around the now late Marcelino dos Santos and therefore favoured a state-owned sugar industry, he had himself experienced the problems of the state-owned and centrally governed industry after 1975. He had also taken part in the rehabilitation efforts during the 1980s and realised that, without strong foreign partnerships, finance, technology and market access, rehabilitation would be an uphill battle (President SINTIA, Maputo 2009 Int.). He therefore met with leaders within Frelimo's nomenklatura, who were against the reforms on socialist ideological grounds, to get them to understand that the reform could industrialise and urbanise peasants through other means that could nonetheless still be seen as a continuation of Frelimo policies. Just when he was acting as a go-between when attempts to block the illegal trade in sugar needed the support of the socialist ideologues in the party, the Chissano administration confronted the business empire of the future President Guebuza. We argue that it was this particular blend of skills and experiential knowledge that made it possible for mediating bureaucrats such as this union leader to ‘muddle through’ using their experiences and their factual and tacit knowledge to straddle various ideological frontiers. Despite their experiences of decline and the more or less total breakdown of the sector, they were adamant that it would be possible to get the flagship of modern rural industrialisation in Mozambique up and running again, but from their experience they had also realised that political backing in itself would not be enough. As a consequence, assistance from outside Mozambique would be required as well, and they acknowledged that private investments could be important. As one cadre who had been active in the party after independence and later became the national director of INA and, in the new millennium, the first ‘black’ Mozambican director of one of the privatised companies argued in 2010:

It was knowledge and experience of the different spheres and the inner workings of the party, the state and business in their fragmented forms that were important for the emergence of those we call ‘mediating bureaucrats’ in the Mozambican sugar industry. The cadres were ‘bureaucrats’ in the Weberian sense (Weber Reference Weber, Gerth and Mills1946: 196–204), as they had been working in and had been in charge of the industry after independence and were embedded in the liberalising reform initiatives after 1994. It was cadres who had muddled through during the colonial, post-colonial and post-civil war liberalising reforms who realised their own limitations as the drivers of a state-centralised industry after independence, as well as the need to involve the state in the liberal reforms. They used this tacit knowledge to deal with the constraints of the global market economy in which they had to operate when marketing sugar, procuring technology and inputs, and trying to upgrade and make the industry more effective after 1994. This particular set of experiences consisted in paradoxically running down an industry and then trying to rehabilitate it within a liberalising regime without unnecessarily upsetting important ideological hardliners from the Marxist–Leninist period, as we indicated above. We will discuss this further in the next section, where we consider the ability of mediating bureaucrats to translate multiple concerns.

TRANSLATING MULTIPLE CONCERNS

The mediating bureaucrats who were embedded in the liberal reforms that were introduced after 1986 but were speeded up after 1990 could intercede and translate between foreign direct investment (FDI) and the state's, government's and Frelimo party's needs as they changed over time. The mediating bureaucrats embodied and, in different configurations, were able to draw on experiences that gave them an intimate understanding of the different aspirations of investors, the state, the government and party officials, while still being able to play on and promote each of these aspirations when necessary during the reform process, as they translated between multiple concerns. In the process, they changed the socio-economic matrix within which the sugar industry operated. As a result, it went from being a sector revolving around a profoundly party-state-industry-based nexus in the 1980s to something closer to a mixed liberal economy in the 2000s, when the sugar industry, while still largely reliant on the protection of state tariffs and state co-ownership, operated more and more independently.

Translating liberal reforms and privatisation into a form that could appropriate the ideological terrain and win over the ideological hardliners who felt that the post-1990 liberal reforms were abandoning the post-independence socialist project was not easy. A key concern for the Chissano regime, particularly after 1994 and the first democratic elections that followed the first wave of liberalising reforms, was job creation after the civil war and the closing of redundant state-owned industries in urban areas. This happened when people began returning from internally displaced camps (IDPs) after the 1992 General Peace Accord. A policy that could, with a degree of translation, cater for both issues was to create an ‘urban labour class’ in rural areas by rehabilitating the sugar industry.

A former state-industry cadre, who among other roles had been Director of the Buzi sugar estate before taking up various positions in the privatised sugar industry after 2000, elaborated in an interview in 2010 on the sugar industry's track record as ‘the biggest (private) employer in Mozambique before independence’ and as a ‘key export commodity’ that ‘made colonial investors rich’, as well as highlighting the sugar industry's ‘proven capacity to monetise and industrialise the rural areas’ (see also the National Sugar Policy, INA 1996). Here in particular the industry's capacity to change labour relations and turn ‘petty peasants into labour that could work in the industry’ was valued (Former Director at Buzi sugar estate, Maputo, 2010, Int.). Job creation in the sugar industry allowed these cadres to connect the particular ideological history of creating a New Society after independence with liberalising reforms in Mozambique – two histories that were in opposition. However, being able to tap into this history of changing rural areas was important in mobilising support for key aspects of the strategy behind the rehabilitation (see INA 1996), as it legitimised liberalising state ownership with some state participation. The liberalisation of the industry and the involvement of FDI were contested politically within the ruling Frelimo coalition, where members of the old hardline ideological faction did not want to back down over retaining the Marxist-Leninist tenets of the post-independence state project, in which state control was imperative. However, if FDI and liberal reforms could somehow be presented not only as supporting Frelimo policies but also as delivering this imperative in a new way, this could soften the implacability of their resistance.

One key area for understanding how post-independence concerns relating to transforming society came to be translated into the domain of liberal reforms is the policy of ‘urbanising the rural’, which suggests a continuation of post-independence policies to create a Homem Novo (New Man) and a Sociedade Nova (New Society) (see Buur Reference Buur, Shiranagama and Kelly2010; Sumich Reference Sumich2018; Buur & Sumich Reference Buur and Sumich2019). After 1996, party, state and industry cadres and officials implementing the sugar rehabilitation policy repeatedly emphasised that, wherever the sugar industry had asserted itself, one saw a greater transformation of rural areas compared with other agricultural, industry and infrastructural investments. In particular, when the sugar industry became operational after 2000 (see Buur et al. Reference Buur, Baloi and Mondlane Tembe2012b), there were a number of knock-on effects, including expansions of local trade, public services, banking and credit, and infrastructure (roads and communications). In general, the rural economy around the industrial plants changed, as far more resources and services (markets, banking and telecommunications, as well as the transport of cane, workers and agricultural labour) emerged, and with it came improvements in the provision of infrastructure related to transport, education, health, water and other state services. The argument was that the sugar-producing areas stood out as ‘islands’ of modernity in a landscape of superstition and authority based on culture and ethnicity. The fact that the sugar industry was known for providing superior, quasi-urban social services in housing, education, health and water made it special for the Frelimo leadership.

The ability to link the transformative capacity of the sugar sector in rural areas during the liberal reforms to Frelimo's long-term post-independence socialist drive towards the idealised, modern and rational Homem Novo and Sociedade Nova that would transform colonial society and its social relations (Buur Reference Buur, Shiranagama and Kelly2010; see also Sumich Reference Sumich2013) was important in creating political support for the rehabilitation of the industry. This deep link between post-independence concerns related to social transformation and post-socialist concerns with market relations nonetheless needed some translation.

Samora Machel's vision as the first president had been for a society cleansed of its traditional divisions: ‘No tribes, no regions, no races, no religious belief … to divide us’ (Machel Reference Machel1981: 43). After independence, rural transformation was attempted through ‘collectivisation’: peasants, or family-based subsistence farmers, as they were largely but wrongly regarded (given that they were integrated into the world market in multiple ways), were forcibly concentrated into collective villages or state farms with little consideration for their market-based livelihoods, which mainly took the form of cash-crop production and seasonal migrant labour. This forcible and often quite violent attempt at ‘uprooting’ the populations of the rural hinterlands and gathering them into villages was expected to bring with it access to education and health, to political training, to the party-state authorities and so on, thus creating the conditions for the internal liberation of Man, not only from traditional society, but also from all elements of capitalist exploitation and colonial divisions.Footnote ¹¹

The transformative capacity of the sugar industry is what the president of the sugar producers’ organisation (APAMO) referred to as ‘urbanising the rural’ or ‘making hostile populations more positive to change’ (President APAMO, Maputo 2008 Int.). This was also what Prime Minister Diogo later suggested in a series of interviews during 2008–2010 with one of the authors, as well as in her book A sopa da madrugada: das reformas à transformação económica e social em Moçambique, 1994–2009 (Diogo Reference Diogo2013). She forcefully pointed out that a major concern of the Chissano government, as people returned from internal and external refugee camps after the peace accord in 1992, was access to services they had been accustomed to, but also that returnees needed to be accommodated so that they did not feel abandoned.

The APAMO president had been part of the rehabilitation process from the outset in 1995 and had taken up positions in the post-independent state-party before becoming president of the sugarcane producers. As a core example of a mediating bureaucrat, he had served as the state industry's representative at the Buzi sugar estate during the civil war and later as advisor to the governor of Sofala during liberalisation. The APAMO president was well placed to appraise the contradictions that had to be reconciled between on the one hand the project to create the New Man and Society after independence in 1975 through socialist relations, and on the other hand, after the end of the civil war in 1992, the transformation that was to take place through and with capitalist relations, as well as the measures required to facilitate this displacement of goals. The old ideological vanguard was not naïve, but it trusted mediating bureaucrats such as the APAMO president, who had worked closely with them as a loyal party cadre in state-owned enterprises, line ministries and at the provincial level when they had all taken up high positions in the Frelimo state as governors of key provinces, ministers and directors of state industries. The story of how the old ideological project of the New Society was linked to the rehabilitation of the sugar industry as part of the liberal reforms was echoed by all interviewees regardless of faction. That this was a paradox or contradiction in terms was exactly the point – being able to bridge and translate the divide was what allowed the mediating bureaucrats to find the political support to push forward the reforms. One way in which this group of hesitant leaders was, if not convinced, then eased into support related to ownership. Continued ownership of the sugar industry by the state was contested by both the old ideologues, who disliked being involved with FDI, and the foreign investors, who severely distrusted the intentions of the Frelimo state with respect to its ownership, as they feared interventions.

MEDIATING BETWEEN COMPETING INTERESTS

A surprising feature of the rehabilitation process was that, until around 2012, the Mozambican state (through the Banco de Moçambique) was still a main player in the sugar sector. Today it still has shares in more than one company, but until after 2012 it was actually a majority shareholder in three of the four companies (Buur et al. Reference Buur, Mondlane and Baloi2011). This was after the first two waves of privatisation involving heavyweight South African and Mauritian capital, management and knowledge infusion. The Mauritian stake was later taken over by large French-Brazilian capital interests, and one of the South African companies, Illovo, was bought up by one of the biggest players in the sugar industry globally, Associated British Foods plc. Besides big corporate international sugar interests, a few private investors, some of whom were former sugar-company owners from the colonial era such as the Petiz family from Portugal (involved with Maragra and later on cane production at Buzi in Sofala), became involved after the restructuring and privatisation (Buur et al. Reference Buur, Baloi and Mondlane Tembe2012a, Reference Buur, Baloi and Mondlane Tembe2012b). In a country apparently overrun by neoliberal reforms (Hanlon Reference Hanlon1991; Obarrio Reference Obarrio2014), it came as a big surprise that the state still had major interests in the industry. This first became clear when individual sugar estates wanted to promote individualised marketing strategies when international sugar prices peaked after 2012, after they had been forced by the state to pursue collective marketing through the National Sugar Distributor (DNA) after 1994 (Buur et al. Reference Buur, Mondlane and Baloi2011, Reference Buur, Baloi and Mondlane Tembe2012b).

The reasons why the privatisation of the sugar industry did not mean the Mozambican state relinquishing its ownership of or control over the sugar sector have both political and economic underpinnings, and they were intimately related to the particular way in which access to finance for the rehabilitation of the industry was initially provided. Here the mediating bureaucrats as ‘socially embedded actors’ played a key role in mediating between different competing interests emanating from industry actors and government.

In order to access cheap loans for the rehabilitation process, besides what the investing companies could bring to the table themselves, the Mozambican state ‘borrowed’ money from special funds like OPEC and the Kuwait Foundation and handed it over to the sugar companies (Diogo 2010 Int.; Whitfield & Buur Reference Whitfield and Buur2014). In exchange, the state retained 51% of the shares after the foreign companies had bought themselves into the industry after 1994. The state percentage in fact first saw a large cut in 2008, when private investors began a new round of upgrading and expansion of cane production in order to increase productivity further. This expansion was in large part stipulated by the EU's accompanying measures (AMSP 2009). As new investments were made, the state share of ownership slowly diminished further. For the state this initial arrangement was beneficial: as the private investors serviced the ‘state loan’, the state's credit ranking increased so that it could take out new state loans for other projects at cheaper rates (see Buur et al. Reference Buur, Mondlane and Baloi2011). This arrangement not only created links between the FDI-driven sugar companies, the state bureaucracy and the government, it also enmeshed ownership structures in three out of the four sugar estates that were rehabilitated.

This was a risky arrangement for both the Frelimo government and the state, as well as the international companies. While the state and the party relinquished day-to-day management of the sugar estates to foreign companies and their staff, they maintained close control over their boards and over how sugar rents from creating a protected internal market were distributed and to whom. First, as a guarantee and a way of monitoring the companies, the government appointed one or two representatives to the boards at the outset of the privatisation process, and three companies still had such representatives as late as 2016. The appointees were highly ranked cadres whom we regard as members of the mediating bureaucracy, people from either the party-state or the state-business sector. For example, the former director of the National Sugar Institute (INA), who was previously in charge of drawing up the rehabilitation strategy, was appointed to all the administrative boards. He paid special attention to Tongaat-Hulett's investments in Xinavane in the south and Mafambisse in the centre of the country, where he later became executive director until 2017. The new INA director who took up the mantle after 1996 later served on the board of Marromeu until 2009. Other members of the boards were high-level Frelimo or state officials with similar experiences from the era of centralised industrialisation, civil-war management and the transition to a mixed economy during liberalisation. They were cadres who had alternated between party, state and industry.

According to Luísa Diogo, the former Prime Minister, members of the mediating bureaucracy who served as state representatives on the boards and in the leadership of companies were important because they ‘served as our eyes and ears, as many members of the party did not trust FDI’ (Former Prime Minister Luísa Diogo, 2010 Int.). What Diogo is referring to here is the mistrust of the liberalising reforms on the part of the hardcore socialist ideologues, as they feared the country would be taken over by foreign capital, thus undermining Frelimo's legitimacy and its control over society and the economy. However, by placing trusted cadres in key positions on boards and in the management of the industry, this allowed the state to prevent breaches of contract and to ensure that the companies honoured the financing deals, which ultimately had implications for the government's credit-worthiness. Particularly during the early years after the end of the civil war, the formal links with company boards allowed the state to follow the rehabilitation process closely, monitoring it so as to ensure that the politically important aspects of the rehabilitation plan were being followed, such as job creation and social service provision. A known aspect of this was that the industry suspended automated cane-cutting, thus making sure that thousands of temporary cutters’ jobs would be available every year during the harvest up to and after Christmas.

This special arrangement also allowed the mediating bureaucrats to mediate industry concerns, particularly when companies were confronted with labour unrest (see Buur & Sumich Reference Buur and Sumich2019) or when they struggled to have important inputs released from the ports when they were kept there because of red tape. In particular, as a director of one of the sugar companies, a former INA director and longstanding state representative on several estates, argued:

Second, while the companies certainly felt that they had become enmeshed in the webs of influence within the Frelimo state as it emerged after the GPA of 1992 and during the liberalising reforms, and therefore were put under pressure to support election campaigns (providing cars, lorries and other contributions), creating jobs and minimising import costs, it also provided ample ground for trade-off agreements between the companies and the party-state. Such trade-off negotiations and decisions, which benefited both parties, required that the right people, those who understood the needs of international capital and how to translate them into national political concerns, could be mobilised. Here the mediating bureaucrats were greatly valued, as they could bridge different concerns so that policy-wise they could support each other, even if they were somehow mutually contradictory. International investors, for example, needed some protection for the internal market in order to be able to recover the costs of their investing in a situation of risk. By legislating and enforcing the implementation of a flexible tariff for sugar imports, the government pursued a means of protecting infant industries (Buur et al. Reference Buur, Baloi and Mondlane Tembe2012b). Aligning related but ideologically different policy trajectories became a major input that mediating bureaucrats could provide, thus allowing reform of the industry to continue, despite being contested by World Bank liberalists, anti-liberalist NGOs and post-independence centralists. During this process the mediating bureaucracy were important, in that they facilitated the transition of the sugar industry from being state-owned to being the predominantly foreign-owned industry it is today.

In short, the mediating bureaucrats in the sugar companies were able to influence state and government actors to assist the companies when, for example, the tariff system that protected the industry from outside competition needed to be adjusted or renewed (Buur et al. Reference Buur, Baloi and Mondlane Tembe2012b; Buur Reference Buur2020). With what we have referred to as mediating bureaucrats placed strategically in boardrooms as ‘socially embedded actors’ in different capacities in each of the companies, conflicting expectations and interests could be aligned. Because this group of cadres had knowledge and sufficient political savvy due to their deep-rooted historical understanding of what other actors, whether in the state, the party, government or industry, needed in order to continue the association, they could get things moving. They could even overcome the devastating flooding that caused damage reaching hundreds of millions of US dollars, as in 2000/2001, 2007/2008 and again in 2013 (UNDP 2014).Footnote ¹²

This group could draw on empathy from having seen the industry erode after independence, as they stayed on when, after independence, the political project imploded, trying to rescue what could be rescued, as well as issuing informed communications about the challenges and possibilities that could be mutually beneficial for the state, the government and investors based on some degree of ‘mutual interest’ (Whitfield & Buur Reference Whitfield and Buur2014). In a dominant one-party state there are always people who, due to party-state connections, a history of common struggle or a shared organisational history, have influence beyond their title or official duties. They may have close to what Ferris et al. (Reference Ferris, Treadway, Kolodinsky, Hochwarter, Kacmar, Douglas and Frink2005: 127) refer to as ‘political skill’, defined as: ‘The ability to effectively understand others at work, and to use such knowledge to influence others to act in ways that enhance one's personal and/or organisational objectives’. Cadres with such understandings can allow trust and ability to be listened to, creating the basis for being able to exert influence that could amount to political savvy. Political savvy for mediating bureaucrats meant that they could translate and communicate difficult political and economic realities to all parties: industry, state and government. Such mediation allowed the rehabilitation to move forward until 2019, when it stalled due to the Idai cyclone, the effects of the disappearance of the lucrative European internal market, industry actors’ corporate strategies and the government turn to gas and minerals as the most important sectors to use for support for purposes of political survival (see Buur Reference Buur2020). However, that is a story for the future.