Strategic privatisation: rehabilitating the Mozambican sugar industry

transcript

This article was downloaded by: [ ]On: 27 June 2011, At: 03:30Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Review of African Political EconomyPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/crea20

Strategic privatisation: rehabilitatingthe Mozambican sugar industryLars Buur a b c , Carlota Mondlane c d & Obede Baloi c da Danish Institute for International Studies, Copenhagen, Denmarkb Wits Institute for Social and Economic Research (WISER),University of the Witwatersrand, Johannesburg, South Africac Centro de Estudos de Democracia e Desenvolvimento (CEDE),Maputo, Mozambiqued Universidade Eduardo Mondlane, Maputo, Mozambique

Available online: 21 Jun 2011

To cite this article: Lars Buur, Carlota Mondlane & Obede Baloi (2011): Strategic privatisation:rehabilitating the Mozambican sugar industry, Review of African Political Economy, 38:128, 235-256

To link to this article: http://dx.doi.org/10.1080/03056244.2011.582762

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching and private study purposes. Anysubstantial or systematic reproduction, re-distribution, re-selling, loan, sub-licensing,systematic supply or distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representationthat the contents will be complete or accurate or up to date. The accuracy of anyinstructions, formulae and drug doses should be independently verified with primarysources. The publisher shall not be liable for any loss, actions, claims, proceedings,demand or costs or damages whatsoever or howsoever caused arising directly orindirectly in connection with or arising out of the use of this material.

Strategic privatisation: rehabilitating the Mozambican sugarindustry

Lars Buura,b,c,∗, with Carlota Mondlanec,d and Obede Baloic,d

aDanish Institute for International Studies, Copenhagen, Denmark; bWits Institute for Social andEconomic Research (WISER), University of the Witwatersrand, Johannesburg, South Africa;cCentro de Estudos de Democracia e Desenvolvimento (CEDE), Maputo, Mozambique;dUniversidade Eduardo Mondlane, Maputo, Mozambique

This article argues that the rehabilitation of the sugar industry in Mozambique cannot beunderstood without including the active role played by the state and government. Itfocuses on key aspects of why and how the Mozambican sugar industry wasrehabilitated after 1996 with and through foreign direct investments. It challenges theexternalist literature on Mozambique that has commonly argued that all policydecisions are enforced by the pressure of well-meaning donors and/or ignorantinternational financial institutions preparing the ground for large internationalcorporations through neoliberal policies, privatisation and structural adjustmentprogrammes. There can be no doubt that donors in general, international financialinstitutions, and international capital have had and continue to have considerableinfluence over economic and industrial policy in Mozambique, but externalistaccounts of various persuasions have limitations and tend to present accounts of theMozambican state and government solely as victims instead of active players.

Keywords: elites; productive sectors; sugar industry; bureaucracy; politics

Introduction

This article’s primary focus is on key aspects of how the Mozambican sugar industry wasrehabilitated after 1996 with and through foreign direct investments (FDIs). We argue thatthe rehabilitation of the sugar industry in Mozambique cannot be understood withoutincluding the active role played by the state and government or what Ouma and Whitfield(forthcoming, p. 1) call ‘the active processes of constructing industries’. The literature onMozambique in the last decade has commonly argued that in Mozambique all policydecisions are enforced by the pressure of well-meaning donors and/or ignorant inter-national financial institutions (IFIs) preparing the ground for large international corpor-ations (see particularly Hanlon 1991, 1996, 2000; Castel-Branco 2002; and Hanlon andSmart 2008). Along the same lines, it is common to argue that in Mozambique internationalcapital has been dealing with a weak African state enforcing neoliberal policies (e.g. Riddell1992; Plank 1993; Saul 1993; Villalon and Huxtable 1998), because after the 1980s and1990s, with Structural Adjustment Programmes (SAPs), privatisation and liberalisation,international capital had the upper hand with close to unlimited access to the natural

ISSN 0305-6244 print/ISSN 1740-1720 online

# 2011 ROAPE Publications LtdDOI: 10.1080/03056244.2011.582762

http://www.informaworld.com

∗Corresponding author. Email: lbu@diis.dk

Review of African Political EconomyVol. 38, No. 128, June 2011, 235–256

riches of the country: land, labour and minerals/energy.1 There can be no doubt that donorsin general, IFIs and international capital have had and continue to have considerable directand/or indirect influence over economic and industrial policy in Mozambique in particular,and in Africa in general (Whitfield 2009). But externalist accounts of various persuasionshave limitations, as we will illustrate here, and tend to present accounts of the Mozambicanstate and government solely as victims rather than active players.

The general literature on business and state relations has highlighted relationshipsbetween state and private sector (Maxfield and Schneider 1997; Taylor 2007) and thesame perspective has forcefully suggested that for Mozambique privatisation was charac-terised by a ‘transformative preservation’ that not only ‘altered the roles of state institutions’but also ‘produced new alliances’ as well as conflicts in society (Pitcher 2002, p. 6), out-comes that have consolidated the power of the ruling party Frelimo over society, theeconomy and the state. This article will draw on the approach and findings of the state–business literature as it challenges externalist positions.

Mozambique has, for good or bad, been ‘a donor darling’ and has had a sustained meanannual gross domestic product growth rate of around 8.6% (World Bank 2005; de Renzioand Hanlon 2009), leading to a doubling of gross domestic product per capita from 1994 to2004 (Virtanen and Ehrenpreis 2007). The overall growth figures for Mozambique havebeen partly triggered by cashing in on the war-to-peace dividend (expected once-off econ-omic rebound), and partly by very high and continued levels of foreign donor support,which today account for over 50% of the state budget (Clement 2008; Buur and Baloi2009). While impressive, this growth level comes off a very low base. In reality, recentgrowth has returned to only pre-independence growth levels (Sousa and Sulemane 2007)and the most important growth trigger has been relatively high levels of FDI, attracted pri-marily to mega-projects in the energy, mineral and gas sectors (Bartholomew 2008, p. 12;Clement and Peiris 2008). Mega-projects account for most of the overall growth andspecific gains in industrial manufacturing and export, especially those secured throughthe US$2 billion aluminium smelter, Mozal, on the outskirts of Maputo, which bringstogether British, Japanese and South African investors (BHP Billiton, Mitsubishi Corpor-ation and the Industrial Development Corporation) as well as some Mozambican capitalinterests.

The Mozal smelter today accounts for roughly one-third of all industrial production inMozambique, but it has been growth with little sustained formal employment creation.For example, the Mozal smelter employs just around 1200 workers and many of those,it is said, are floor sweepers kept in order to boost the employment figures. Furthermore,as argued by several studies, mega-project investments generally do not bring in tax rev-enues due to a 10-year tax exemption rule, although this has recently been cut to five yearsas ‘strong incentives are not necessary any longer’2 and there are strong indications thatagreements will be renegotiated over the coming years. Nor do they create broad-basedpoverty reduction (see the critiques of investment patterns by Beaumont 2004;Bucuane and Mulder 2007; and Tvedten et al. 2006, 2009). But, in general, the FDI-dri-ven mega-investments story in Mozambique suggests that it has formally ‘transformed’the Mozambican economy, whose structure during the 1990s was dominated by agricul-ture and service provision, so that the agricultural share of gross national product declinedfrom 40% in 1995 to 25% in 2008, whereas the industrial share – mainly catered for bythe Mozal smelter and similar investments in energy and gas – rose from 15% in 1995 toaround 33% of gross national product in 2008. But, as agriculture accounts for thelivelihood of up to 70% of the population, the relatively meagre overall annual agricul-tural growth rate at the average of 6% has not been caused by an increase in productivity.

236 L. Buur et al.

Worryingly, there has been no increase in agricultural productivity since 1963 (Uaiene2010).3

We will use the case of the rehabilitation of the sugar industry in Mozambique to illus-trate how the state was not just a hopeless victim to FDI capital, but active for well over adecade in the formation of a qualified investment success story. A superficial glance overhow the industry has evolved over the last decades would suggest that Mozambique hasbeen a hapless victim of mega-project investments by international capital: major inter-national sugar conglomerates invested in only four estates and industrial plants available;the industry is enforcing an enclave economy with few benefits for the Mozambicanstate and population; and, not least, industry has forced the Mozambican state and govern-ment to create and enforce internal market protection measures that ensure that already poorconsumers pay for the rehabilitation of the industry.

The present form of support to the sugar sector took off in the mid 1990s when a policyand strategy for rehabilitation – Politica e estrategias para o desenvolvimento do sectoracucareiro – was approved. However, in contrast to other sectors, state and governmentsupport has been pretty steady and has continued well into the new millennium, and thesocio-economic effects and consequences of the sugar rehabilitation have been very differ-ent from the trajectory of other mega-investments: jobs have been created on a large scale,expansion into the small- and medium-based landholding base is taking place with irriga-tion and inputs raising productivity, social service delivery has been expanding, and trainingand skill development has been taking place with the industry being a net provider of mech-anics, welders, fitters, electricians etc. But many problems exist, and as with productiveinvestment cases they are messy and difficult to generalise from (Khan 2003). The nextsection considers the decline and recent results achieved by the sugar industry in Mozam-bique and functions as a background for the article.4 At the core of analysing key aspects ofhow the Mozambican sugar industry was rehabilitated after 1996 with and through FDIs isthe underlying question of why state and government officials have supported the rehabili-tation efforts. The third section summarises the main state, government and industry attri-butes of support related to why the Mozambican sugar industry was rehabilitated.5 Thefourth section highlights three key aspects of how the sugar industry was rehabilitated.Combined, this aims to demonstrate the active role of the state and government in the rela-tively successful sugar rehabilitation process so far.

The decline and new momentum of the sugar industry

Sugar cane has been produced in Mozambique since the end of the nineteenth century. TheSena Sugar Estates Ltd, comprising two facilities, Luabo and Marromeo, became particu-larly important, and in many ways the company was the crown jewel of the sugar industryuntil about 1975, operating among other projects a rail spur connecting the sugar refineriesto Beira’s port. Until the 1950s, expansion of the sugar sector was based on British invest-ments, after which Portuguese capital interests became prominent. Cane and sugarproduction expanded massively at the end of the 1950s and during the 1960s. In 1954,the Xinavane plant on the Incomati River was further enlarged as land concessions wereincreased (Gode 1997, pp. 16–17). Two new factories followed, first in 1969 with theMaragra plant in southern Mozambique, downriver from the older Xinavane estate onthe Incomati River (Cardoso 1993), and in 1970 the Mafambisse plant in the Beira corridorin central Mozambique along the Pungue River.

With these industrial plants, the total sugarcane production capacity was brought up to360,000 tonnes, with a production record reached in 1972 of 325,051 tonnes, which was

Review of African Political Economy 237

close to maximum production, with 60% for export. The expansion of the sugar sector –culminating in 1972 – saw it become the third biggest export sector and the biggestformal labour employer, with the Sena Sugar Estates Ltd in the mid 1960s – the largestprivate firm in Mozambique – employing more than 10,000 people. The independencewar in northern and central Mozambique, and from 1974 the emergence of the Frelimo-dominated transitional government, left investors feeling insecure, and were a directcause of the ‘disinvestment strategies’ adopted in the transition to independence (Castel-Branco 2002, p. 83). Furthermore, up to and after independence in 1975, the flight ofskilled labour of all types (government employees, state administrators, technicians, engin-eers, managers etc.) and capital added to the decline, just as global market prices made thesector less profitable. After independence the industry was considered a ‘strategic industry’and became partly nationalised and partly ‘state intervened’ as the effects of disinvestment,lack of finance, human capital constraints and so forth took their toll on the industry. From1972, the peak year with 325,051 tonnes of sugar, production declined to just over 200,000tonnes in 1975 and thereafter fell slowly to about 175,000 tonnes in 1980. A steep declineafter this saw production fall to well below 50,000 tonnes from 1982/83 onwards, afterwhich it was almost wiped out (Instituto Nacional do Acucar (INA) 2001, p. 4; Cardoso1993). In 1992, at its lowest, the production was only 13,224 tonnes. The result was asteep increase in imports of sugar, making different regions of Mozambique dependenton official and contraband imports (from Swaziland, South Africa, Zimbabwe, Malawi,Zambia and Tanzania). For the labour force the decline in the sugar sector was a tragedy.At the beginning of the 1970s around 45,000 workers were formally employed (one-third of the standing labour force employed in South African mines at the same time;First 1983). As the civil war intensified after 1980, it affected the six sugar-producingentities so badly that by 1982 four were closed and two were operating at very low capacity.By 1999, only 17,000 workers were still formally employed by the six estates, but most ofthe workers were redundant and employed only on paper as production was so low.

Since the mid 1990s, the Mozambican sugar sector has been undergoing a steadyprocess of rehabilitation, involving both partial privatisation and large-scale FDI, in bothcane production and sugar processing and refining capacities. This has seen four out ofsix sugar estates with sugar producing factories becoming productive between 1998 and2002. The areas under cultivation rose from 7266 ha in 1998 to just under 31,000 ha by2008 and 35,000 ha in 2009, with a scheduled increase to 50,000 ha after 2011(CEPAGRI 2009, p. 12; LMC 2006). From coming close to a total standstill with onlyaround 16,000 tonnes produced in 1986 and around 13,224 tonnes during the 1991/92campaign (INA 2001, p. 4), the sector had increased production to just over 200,000tonnes of sugar, with close to half of this being exported by 2004 (INA 2004, pp. 4, 18).While by 2008 the productive gains were modest, peaking in 2005 at just 265,000tonnes, this was caused by drought, poor rain distribution, delays at the start of theseason and work stoppages caused by installing two new machines in a new expansionprocess with a concerted drive toward doubling production by 2010 to just around500,000 tonnes (CEPAGRI 2009).6 By 2003 export had exceeded local consumption,with total export figures reaching 134,796 tonnes in 2008, even though the industry-owned Distribuidora Nacional de Acucar (DNA) had managed to double nationalconsumption.

It is estimated that around 32,000 direct jobs (combining permanent and temporaryjobs) have been created at the four rehabilitated sugar estates (CEPAGRI 2009, p. 15;CEPAGRI 2010) besides jobs created in outsourced service functions in land preparation,planting, maintenance and transport, as well as jobs created by independent producers and

238 L. Buur et al.

down- and upstream jobs created along the value chain, making the industry once again thebiggest non-state employer in Mozambique. Importantly, this is jobs created in the ruralareas, where those taking up employment are a mix of local populations and ruralmigrant workers (mainly for cane cutting). Salary scales in the industry are generallywell above what can be earned in rurally based agriculture industries. Social servicessuch as health, education and housing are to a large degree catered for or strongly subsidisedby all four factories, where districts and municipal administrations benefit from generouscorporate social responsibility (CSR) budgets. It is generally acknowledged that social ser-vices and access to banks around the four rehabilitated estates and plants are well abovewhat one normally encounters in rural areas (Locke 2009). The industry is ‘creating anincome multiplier effect, with sugar workers having money to spend on other goods andservices [and] the sector also creates the opportunities for other industries to develop, sup-plying goods and services to the sugar mills’ (LMC 2006, p. 7; also INA 2005b). But the re-emergence of the sugar industry in the rural areas has exposed the shallowness of thespecialised service and supply industries as the backward linkages between the sugar indus-try and local Mozambican suppliers of services and goods have been problematic, just as itis for all other FDI mega-investments. For example, only about 23% of purchases byMaragra sugar estate in 2004 were locally produced, with the rest provided by foreign com-panies or their subsidiaries (LMC 2006, p. 8).

Furthermore, the European Union’s E6 million Accompanying Measures Fund to miti-gate the effects of changing the European sugar regime (CEPAGRI 2006) has stipulated thepreviously limited outsourcing of cane production. Where only 0.4% of cane was producedby independent Mozambican producers in 2005 (INA 2005a, p. 21), this figure hadincreased to 4% by 2008 (CEPAGRI 2009, p. 24) with around 15 associations, involving1365 producers. The European Union fund also signals the first large-scale direct involve-ment of donor money as part of the rehabilitation process, which in itself is remarkable, con-sidering the levels of aid dependence experienced by Mozambique over the past twodecades.

Why was the sugar industry rehabilitated?

Why was the sugar sector singled out by the Mozambican state and government for rehabi-litation? The literature on the issue has argued that the rehabilitation strategy approved bythe European Union Council of Ministers in 1996 came about ‘under the coordinatedpressure of investors, three large international sugar corporations’ (Castel-Branco 2002,p. 179). In this optic, the sugar rehabilitation strategy from 1996 and the actual processof rehabilitating the industry was therefore more of a forced imposition than a jointeffort between the state/government and the foreign corporations.7 Mozambique’s emer-gence from a history of nationalisation and state intervention/administration of mostparts of the industrial base after independence in 1975 seems to support such an argument,though there were forces within Frelimo and among its academic backers that in general –for ideological and nationalistic reasons – were against privatisation (Hall and Young 1997;Pitcher 2002; Macuane 2010). Let us briefly engage with the question of resistance and/oraccommodation to reforms like privatisation of the industrial base.

The politics of privatisation

It is argued that as the reforms of then President Samora Machel took off, particularly afterhis death in 1986, one would for the first time see clearly demarcated ideological and

systemic divisions materialise within the Frelimo post-independence set-up that bothreflected and further ‘intensified fragmentation and dissension within the state’ and theparty (Pitcher 2002, p. 116). It is common to depict this as the emergence of three blocs:the ‘hardliners’, the ‘softliners’ and a ‘neoliberal’ bloc (Pitcher 2002). The three campsemerged just when Frelimo attempted to reform the centralised state order and disengagethe state from the party at the same time as the civil war was peaking and the economiccrisis seemed irreversible.

‘Hardliners’ saw no need for substantial reform of the state project and they refer to atleast two quite different, but intertwined positions. On the one hand, ‘ideological hardliners’recognised the need for reform as this was considered part of the conflictual basis ofMarxism, but they did not search for abandonment of the socialist principles nor of stateintervention per se. If anything was needed it was adjustment and adaptation. On theother hand, ‘systemic hardliners’ saw no need for reform or even change. Here the admin-istrative unity of party, state, political system and society is seen as absolutely natural and isimplicitly taken for granted by many in the state and party apparatus at both national andprovincial level.

‘Softliners’, in contrast, promoted reforms of the state sector and actively advanced theemergence of a private sector and free markets, but they did not want state involvement tobe totally eradicated, only reformed (streamlined and restructured), so that the state couldengage more efficiently with the private sector. Controlled private sector and market pro-motion was therefore the preferred modality for the privatisation of state assets after1987. The ‘softliners’ bloc was quite diverse and differentiated internally, with stands-offs between shifting formations, ministries and personalities regarding the pace ofreform, who the intended beneficiaries were and the specific content of reforms. Buteven though reforms were contested, ‘softliners’ saw reform as part of preservingFrelimo dominance (Pitcher 2002), such as the role of the party vis-a-vis the state andthe role of the state in the economy and social sphere. It is therefore perhaps morecorrect to call them ‘soft hardliners’ or ‘pragmatic hardliners’, as they advocated adjustmentthat reproduced Frelimo dominance and did not challenge the administrative unity of partyand state, or Frelimo control over the political system, society or economic/businesssystem.

Studies of Mozambique dealing with the state and privatisation after 1987 and econ-omic policy in relation to donor dependence have suggested that a third bloc was increas-ingly in the ascendance promoting unguarded and rampant neoliberal reforms (see, forexample, the various writings of Hanlon mentioned in this article). The neoliberal bloc con-sisted of a mixture of Western donors and state and government officials from key minis-tries – such as finance, and trade and industry – that through the 1990s and after the newmillennium were in charge of market and investment policy making and implementation.The small and scattered group of national private sector members supported this drive,which was further underpinned by the alleged merger or ‘marriage between the politicianand businessman’ (Pitcher 2002, p. 118). Descriptions of the ‘neoliberal’ bloc clearlyplay on derogatory images of the exploitive capitalist ‘selling out’ sovereign state assetsand appropriating such assets for own gain. It is doubtful that such a group can actuallybe empirically identified, at least within the Frelimo setup. In contrast it is possible toidentify within the party protagonists for the different instantiations of the ‘hardliner’and ‘softliner’ blocs. Rather the loosely organised blocs created a certain degree offluidity, and capacity to shift location made the direction of the state project less clear,but it also allowed for considerable flexibility regarding policy, as suggested by Pitcher(2002).

240 L. Buur et al.

We mention the three blocs here because the crux of the matter is that reforms have beenappropriated by the Mozambican government, party and state to secure the survival ofFrelimo as a dominant force. As the popular saying in Mozambique suggests, ‘in orderto make sure everything remains the same some change has to take place’. In quite substan-tial cases, reforms were considered necessary rather than just an imposition, and they wereappropriated and aligned to ongoing Mozambican concerns because their content and formwere in line with long-term national policy preoccupations (like poverty alleviation anddevelopment) and the survival of the system, thus both allowing for ‘hardliners’ to seethem as ‘national’ and for ‘softliners’ to find them acceptable as reforms over time founda controlled form. Here it was sometimes convenient to use radical critique and suggestionsof imposition and government weakness as it allowed the government to carry out some ofthe more controversial policies as the ‘victim’ of the international aid system, the IFIs andso forth.

We argue in this article that the state, and the political leadership in charge of the state,supported the rehabilitation of the sugar industry in the context of five related issues thatconverged and created sufficient support at different times:

. Political and pragmatic concerns related to the General Peace Accord from 1992.

. Ideological attributes.

. The need to re-establish state bureaucratic control over territory and population.

. Government needs.

. The particular experience and drive of national industry actors, creating a bureauc-racy that could mediate between FDI and state/government.

These five drivers shaped the various responses outlined above to reform and, we suggest,merged with FDI interests in Mozambique. As such, outlining the five aspects of supportshould demonstrate that the rehabilitation of the Mozambican sugar industry was not justan imposition by foreign interests. Furthermore, this also implies – as we have argued else-where – that the support was not uniform, nor can it be defined in simplistic and one-dimen-sional terms, but needs to be understood in terms of what we call attributes of support thatemerged out of a post-independence fusion of a range of state and government officials’historical experiences of success and failure in the industry and pragmatic concerns atdifferent levels, as well as longer-term ideological stances (Buur 2011a). We brieflysummarise each of the five attributes of support.

Post-conflict and pragmatic attributes of support

The government of President Joaquin Chissano, which took over after President SamoraMachel’s death in 1986, singled out the sugar sector for special treatment. This was donelong before the General Peace Accord in 1992 and based on post-independence strategiesthat referred to the potential of the sugar industry. But it is clear that as part of preparing forthe end of the civil war, a specialised agricultural ‘task force’ was constituted by high-levelmembers of government ministries and state institutions and tasked with identifying poten-tial industries (sugar, cotton, tobacco, tea, cashew etc.) that could provide income opportu-nities besides crucial export revenues. Equally important was the sector’s capacity toprovide a certain degree of service provision, so the heavy influx of people to the citiescould be stopped and demobilised soldiers would see a future in the rural areas. It is there-fore no surprise that in the important March 1995 document Establishing the basisfor economic and social development: key policies (Government of Mozambique 1995) –

which was prepared for the Consultative Group Meeting for Mozambique in Paris, wheredonors were asked for money for the reconstruction of Mozambique – the sugar sectorwas identified as a core strategic industry due to its superior track record as an export com-modity, as well as its capacity to monetise and industrialise the rural areas and its labourrelations. This concern had a particular historical and ideological background.

Ideological attributes of support

The sugar industry was known for providing superior, quasi-urban social services inhousing, education, health and water (INA 2000, p. 2). The transformative capacity ofthe sugar industry is what has been called ‘urbanising the rural’ and is an importantmarker of a softer version of the post-independence attempt at creating the idealisedmodern and rational ‘New Man’ (Homen novo) (Buur 2010). In order to fashion oneunited people in the areas that Frelimo controlled during the liberation war against the Por-tuguese, a myth was created that idealised the liberated zones: ‘Everything is directedtowards liberating man, serving the people . . . there is nothing to divide us’ (Machel1981, p. 43). After independence it was expected that, with access to education andhealth, political training, local governance, and so forth, the new society would producethe internal liberation of Man. After the end of what became a civil war in 1992, the trans-formation would also take place through and with capitalist relations.

Regaining ‘lost’ territory

All of the sugar plants and estates were situated in areas which by the end of the 1980s werecontrolled or strongly influenced by the opposition forces of Renamo. In some areas doubleadministration was de facto the order of the day long after it was abolished de jure after theGPA and the Mozambican post-1994 democratic state had taken over. In some of theseareas Frelimo, the governing party since independence, lost the 1994, 1999 and 2004 elec-tions. The re-emergence of functioning production units in rural areas became a sign of thegovernment’s will and effort to secure greater legitimacy by providing jobs and securinginvestment in transport, communication and port facilities – and, not least, providingsocial services at a higher level than in many urban areas, as it was subsidised by CSR-induced social spending from foreign investors. The re-emergence of the sugar sectorallowed for the gradual re-establishment of territorial integrity combined with politicalcontrol in contested areas. Over a 15-year period the ruling party managed to secureenough support to win both national and local elections in all the areas where the sugarindustry re-emerged.

Government needs

The government was confronted by a serious financial dilemma over state enterprises,including sugar estates and industrial plants. The post-independence state-intervened indus-try, running at a loss, was being funded by the banks and in the end by the national bank.The debt increased year after year. The state had to take over the debt as well as settleworkers’ claims and pensions, and even get rehabilitation started in order to attract inves-tors. As a strategic industry the government wanted to prepare the ground well. This wasbased on an acknowledgement of failures to protect Mozambican government prioritieswhen confronted with pressure from the Bretton Wood institutions. For example, top Chis-sano-era officials made no secret of the sugar sector’s special status during the early phases

242 L. Buur et al.

of rehabilitation, where the sector was used to prove that a more considered form ofprivatisation, with state support, could be achieved without repeating past command-economy shortcomings.8 In particular, the International Monetary Fund (IMF) and WorldBank treatment of other productive sectors like the cashew industry (Cramer 2006,pp. 266–268; Hanlon 2000; Hanlon and Smart 2008), where state protection had been with-drawn too sharply under heavy pressure by the IFIs and just after it was privatised, had direconsequences for the industry and those investing. Put simply, the government wanted toshow that Mozambique could provide support to a productive sector without turning itinto an ineffective sector. It would be ‘like in the past, with central planning and acommand economy, but done in a sensitive, effective and competent manner’.9

A mediating bureaucracy

Finally, after independence state officials and industry people began to realise both theirown limitations and some of the constraints of the capitalist world market economy. Wehave argued (Buur 2011a) that this particular set of experiences drove national industryactors and created a mediating bureaucracy that could intercede between FDI and state/government expectations and needs as they emerged or changed. An important character-istic of the sugar sector or all other sectors that have some recent success in Mozambique(chicken farming or the new momentum of the cashew industry) is that many of the stateand sector leaders, including union leaders, have worked within the industry in bothprivate and state capacities since the 1970s in different constellations. This group ofactors knew what the sector earlier had achieved, they had experienced the total breakdownduring the 1980s and felt it as tragic, as it could have been avoided with a mixed economy.This group of actors had a keen eye for the many aspects of private sector business interestsand needs on the one hand and the diverse and contradictory mix of state/political impera-tives outlined above on the other, as well as for new issues that would emerge. They knew itwould be possible to get the flagship of modern rural industrialisation in Mozambique upand running again, but they also knew that besides political backing it required assistancefrom outside Mozambique. They had experienced that relying on export of sugar forrevenue was problematic, as the free market generally was based on ‘dumping prices’and that was the reason that all sugar-producing countries based their production on internalmarkets and trade agreements.

The general support from a very diverse group of direct actors – the state and govern-ment, the sugar union and industry players – does not imply that there was no critique orthat the process of privatisation and rehabilitation did not create many uncertainties or ten-sions. Union leaders and ordinary members, for example, while generally positive – as theyhad seen the industry vanish and with it their jobs and power, also engaged in struggles tomaintain jobs deemed redundant, and keep privileges and influence that had become insti-tutionalised after independence. Questions related to land, which one could expect wouldfeature prominently and potentially could create tough resistance, were nonetheless lessprominent. This was partly because the land involved in the privatisation process and theinitial rehabilitation phase was industry land developed during the colonial era. After inde-pendence, all land was formally taken over by the state, so privatisation of the estates meanttransfer of buildings, production units, irrigation systems and so on, and creation of long-term leases or user-rights. Most sugar estates were dilapidated and in some instances (forexample at Maragra; Cardoso 1983) the land had partly been used for food productionand some resettlement of displaced people during the civil war. Privatisation and transferto foreign investors of user-rights did initially create some tensions but never on a large

scale as most peasant families could access other land, but more often found work at thesugar estate.

This contrasts sharply with conflicts emerging from the third wave of investments incane-based ethanol production that was initiated in 2008 (Schut et al. 2010), where largetracts of new land are appropriated in processes that resemble large-scale land grabs asthey often involve movement of family-based peasants. This drive towards cane-basedethanol production has involved the potential transfer of land on a far bigger scale thansugar production, where the four estates approach 50,000 ha in total even after a recentexpansion drive based on independent small- and medium-scale cane production. In con-trast, the now failed ProCana cane-based ethanol facility in Massingir in the south hadlaid claim to about 30,000 ha alone, while the Principle Energy project in Dombe in thecentral region envisaged producing cane on 18,000 ha during the first phase. In fact allof the projects have either failed or are struggling, mainly because they are based on volatileventure capital funding, as opposed to the state-guaranteed loans that initially financed therehabilitation of the sugar sector.

Key features of the rehabilitation

In this section we will engage in some detail with three aspects of the state and govern-ment’s role in coordinating the diverse political and commercial purposes and interests atplay as the foreign corporate sugar sector bought into the national rehabilitation policy.The three aspects are: financing rehabilitation; limiting opportunities; and creating aninternal market. The role of the state and government has been broader than this as itwas also involved in organising and institutionalising the sector, monitoring implemen-tation of the sugar strategy, and promoting small-scale cane producers, particularly after2006 when the European Union accompanying measures came into force and so forth.(On the European Union accompanying measures, see Locke 2009.) But the aim here isnot to provide a full narrative of the rehabilitation efforts. It is only to provide sufficientdata for arguing that the Mozambican government and state was indeed very active inthe relatively successful rehabilitation of the sugar industry, instead of solely seeing it asthe victim of external forces of diverse kinds.

The three aspects dealt with here fall in many ways between what Peter Evans tried tocapture with the terms ‘midwifery’ and ‘husbandry’ (Evans 1995, p. 12; 1997, p. 75). Mid-wifery emphasised the nurturing role of the state when prioritising between productive unitsand providing support for them. In Mozambique, nurturing was important so that the sugarindustry could in the short to medium term operate in a manner where export earnings andinternal revenues could be generated, allowing for servicing of loans taken in order to payfor the rehabilitation. These loans were facilitated by the state, as we describe in the follow-ing subsection. As the midwifery role succeeded, the role of the state changed and becamemore of a husbanding role. While the state and government continued to actively promotethe industry, largely with the same people that initiated the process (at least until 2006, whenthe effects of the new Guebuza administration made several changes to the institutionalstate set-up), it also institutionalised and organised the industry as well as continuouslyevaluated the progress of efficiency and competiveness in order to ward off critiques ofits protected internal market. This, in turn, presented possibilities to spur on expansionof cane production, as initially suggested by the policy/strategy of 1996, although thisdid not become economically important until the European Union market became availableand the European Union Accompanying Measures Fund provided some funds for expan-sion through small and medium producers organised in associations.

244 L. Buur et al.

Importantly, the three aspects dealt with here all relate in one way or another to rent cre-ation within the sugar industry (on rents and rent-seeking as a process to foster develop-ment, see Khan and Jomo 2000). In economics rents are conventionally defined as an‘excess return’ that is above ‘normal levels’ of profit made in a free market or enterprisecompetition, and usually becomes associated with a lack of competition in markets domi-nated by monopoly creation and maintenance. But there are, as Khan (2000a, 2000b) pointsout, many types of rents (transfers, regulatory pricing structures, targeted support etc.) andthey can be absolutely necessary for economic growth and industrial development to takeoff, and for the productive sector to become consolidated and profitable. This is especiallytrue in developing countries, as ‘institutional change almost always involves the creation ordestruction of rents’ (Khan and Jomo 2000, p. 3). Rents can therefore be both necessary andproblematic – something that cannot be asserted a priori. Various logic constellations canbe presented but positive/negative implications are context-specific and related to a varietyof internal/external factors. That there is a relationship between the ‘midwifery’ and ‘hus-bandry’ roles of the state and rents is no surprise as rents are essentially incomes created bypolitical interventions (Khan 2000a, 2000b). Transfers, for instance, are often the basis forasset accumulation, for example primitive accumulation that caters for the emergence ofnew capitalist and middle class groups, producers etc. or creates possibilities like thoseof small and medium cane producers far later on in the process. Transfer mechanismscan include taxes, subsidies, conversion of public property into private property and soon and can also, as illustrated here, be used to involve FDI. So where Khan is primarilyinterested in rents related to concerns such as covert class and economic entrepreneur for-mation that can form part of a development strategy, the rent dynamics in the sugar reha-bilitation process were somehow different, at least initially. In the sugar case rents aimedto assist the rehabilitation of a once-productive sector with and through FDI capitalwhere the state played not just an active but a critical role in promoting investments,upgrading technology, acquiring skills, providing loans, creating protected internalmarkets, signing up for international high-yielding trade quotas and so on. There was nodirect covert class and economic entrepreneur formation initially but rent creation definitelyformed part of a post-conflict development strategy for highly contested territories wherethe Frelimo state had only a scant or contested presence. Rents as such compensated forinitially uncompetitive cost structures while the industry got off the ground, paid backstate guaranteed loans and generally learnt to operate in the risky post-conflict Mozambicanbusiness and political landscape.

Financing rehabilitation (or creation of rent opportunities)

Initially, one of the enigmas was why the state continued to have a large shareholding in theindustry after it became privatised. If Mozambique had been solely a neoliberal victim, stateinvolvement would have vanished or been derailed. If the old hardliner faction of Frelimohad been the driving force the state would have been basically running the actual oper-ations. Instead, its role was to prepare for and finance the initial phase of rehabilitation.State shareholding was initially related to the ‘debt swap initiatives’ that formed part ofthe initial restructuring of ownership that took place before actual privatisation was effectedand partly to accessing favourable loans for the actual process of rehabilitation. Debt swapdealt with old debt accumulated before 1975 and escalating afterwards, or new loans/credits necessary for paying out workers, pensions, use of land etc. which was swappedfor shares by the Banco de Mocambique, which became a main owner of the sugar industryby swapping shares before foreign investments were made. But as Castel-Branco (2002,

pp. 201–202) has correctly noted, ‘multilateral and commercial credit was made available’to the four companies that were rehabilitated. How exactly this was done is a rather com-plicated issue but it holds the key to understanding the substantial participation of the statein the industry until recently. In Mozambique, the quality of the financial system was andstill is poor, and interest rates, which today hover around 15–20%, were even higher before– if any capital at all was available immediately after the civil war for such long-term andintensive investments as the sugar industry demanded.

An illustrative example of how financing changed the formal composition of the sectoris the shifting ownership structure of the Xinavane estate and industrial plant, the originalname of which was Sociedade Agricola do Incomati (SARL). Xinavane was never nationa-lised or intervened in like the other companies were after independence, but stayed in thehands of the South African sugar company Tongaat-Hulett Ltd, the owner of which was theeven larger multinational, Anglo American. The company was formally operational afterthe civil war but at a very low level – close to 8% of its capacity at the time of rehabilitation.Interestingly, Xinavane was not rehabilitated before the state took up substantial shares inthe estate. In 1990 SARL was dismantled and a new company was set up, Acucareira deXinavana, with the state as majority shareholder (51%) and Tongaat-Hulett at 49% withthe right to acquire more shares at a later stage. The Mozambican state’s involvementwas financed by loans from the Banco Arabe de Desenvolvimento (Arabic DevelopmentBank, BADEA), the Kuwait Fund and the OPEC Fund to the total value of US$45million (Gode 1997, p. 6). These were the loans used to rehabilitate the estate and industrialplant.

Therefore the reason for the shareholding change is that the rehabilitation was to a con-siderable extent financed through the Mozambican state, which took ‘cheap loans’ atvarious multilateral and commercial credit facilities like development banks, specialfunds and facilities providing such loans primarily to states at that time. (This, accordingto present actors, has partly changed.) In such a situation, loans negotiated for rehabilitationhad a double function. They aimed at rehabilitating a particular entity, but they also sent astrong message to the international sugar industry: ‘We want to rebuild the industry; we areserious and are ready to take chances and assist.’10 As the sugar companies got export rev-enues in – initially from the few limited preferential trade agreements with the UnitedStates and European Union and less so from free market sales characterised by ‘dumpingprice setting’ (LMC 2004, 2006) – they paid off the ‘state debt’, so to speak.11

Here export revenues, due to foreign currency constraints, were crucial for external debtpayment and paid for the actual loan-servicing besides also catering for the payment of asmaller fee to the state for the provision of access to finance.12 As debt was paid off,state shares diminished but the state’s overall credit rating improved as it had been ‘punc-tual’ in servicing debt. The state’s intervention in providing cheap finance was crucial forattracting foreign investments particularly to less attractive and politically difficult invest-ment areas like the central regions of Mozambique.

The point is that the Mozambican state (through Banco de Mocambique) was still amain player in the sugar sector after the first two waves of privatisation, with heavyweightSouth African and Mauritian capital interests (capital, management and knowledge, whichfor the latter was later taken over by large French–Brazilian capital interests) increasinglybecoming directly involved.13 Besides big corporate international sugar interests, a fewprivate investors, some of which were former sugar company owners like the Petizfamily from Portugal (who managed to get control over the Maragra estate after the civilwar and later bought Buzi with the aim of rebuilding the company), became involvedafter the restructuring and privatisation. The particular way in which access to finance

246 L. Buur et al.

was initially provided created further links between the FDI-driven sugar companies, thestate bureaucracy and government. For each company the state/government appointedone or two representatives consisting of highly ranked personnel from either the party–state or state–business sector. For example the former director of the National Sugar Insti-tute (INA), previously in charge of producing the rehabilitation strategy, was deployed to allthe administrative boards but with special attention given to Tongaat-Hulett where he laterbecame director, while the new INA director taking up the mantel after 1996 later served onthe board of Marromeo until 2009. Others were high-level Frelimo or state officials.

For the state/government, the formal administrative linkages allowed it to follow therehabilitation process closely, monitor that the politically important aspects of the rehabili-tation plan were followed and assist when needed. In particular, it could make sure that pol-itical and pragmatic concerns related to the General Peace Accord from 1992, which hadideological underpinnings such as the social service dimensions (building and maintenanceof school, health and housing facilities and water provision that also referred to an ideologi-cal project of transforming rural populations), were honoured as much as the economy atthe different estates allowed. As a result of continuous pressure from state representativesthe economic profile of the industry did become more inclusive, particularly when it cameto small- and medium-scale cane production. It is no coincidence that Xinavane today hasthe most expansive small- and medium-scale cane producing scheme, as the state represen-tative – the former INA director – got the company to test different models for such pro-duction early on in the process. State representatives on the boards also allowed the state toprevent defection and ensure that the companies honoured the finance deal, which ulti-mately had credit implications for the government.

Table 1. Ownership composition by 2000.a

Company Ownership Shares (%)

Maragra Acucar, SARL Maragra SARLb 50Maragra Comercial, SARL Illovo Sugar Ltd 50

Maragra SARL 75Others 25

Xinavane The state 51Tongaat-Hulett Ltd 49

Mafambisse Tongaat-Hulett Ltd 75The state 25

Buzi Banco de Mocambique 67The state 33

Marromeo Sena Holding Co.c 75The state 25

Luabo Sena Holding Co. 75The state 25

Source: Drawn from Instituto Nacional do Acucar (INA) (2000), p. 11.Notes: aThis ownership structure has since changed in various ways, but primarily for Sena Company (Marromeo/Luabo) as new French–Brazilian capital has taken over. State shares have generally declined, which is relatedprimarily to debt payment swaps as we have found no information related to any payments for state shares.bMaragra was the only company divided into sugar-producing and agricultural (cane) divisions. Maragra SARLhad by 1999 an ownership structure that included the Petiz family (60%), Banco de Mocambique (17%), thestate (17%) and others (6%).cThe company was privatised in 1998 and Sena Sugar Estate was owned primarily by Mauritian capital. This hassince changed and Tereos, a French company, the world’s fourth largest sugar producer, took a 50% share in SenaHoldings Ltd. In 2007, Tereos’s sugar interests were concentrated in the Brazilian subsidiary Acucar Guarani SA,the third biggest Brazilian sugar cane producer (Sena 2009, pp. 1–2).

This leads us to the second aspect of how rehabilitation was facilitated by the state andgovernment. In the next section we discuss how four of the six available estates/industrialplants were selected for rehabilitation, and the political implications of the exclusions andinclusions.

Prioritising among estates (or selecting rent seekers)

That only four out of six estates were initially rehabilitated can be explained partly by thelimits to the liabilities the state could expose itself to as the institution that de jure took outthe loans. But beyond the economistic aspects other issues also need to be consideredrelated to the political economy of the rehabilitation of the sugar industry. There can beno doubt that the government and state personnel initially wanted to rehabilitate all sixfacilities and investors had indeed been identified for five of them. There were fiveestates and industrial plants that were not operating and/or were severely dilapidatedbesides Mafambisse, which by then was already under rehabilitation but encounteringserious management problems (Buur 2011b). Was it economically feasible to maintainall the companies? Was it politically desirable to rehabilitate all the estates and mills? If pol-itically desirable but economically unfeasible, then based on what criteria should a milland/or estate continue to be closed or reopened? The issue was both economic and at itscore deeply political – and related in interesting ways to a set of issues connected to transferof rights and sharing of rents generated by political intervention and, later, marketingcoordination strategies (elimination of competition) besides state and government needsfor re-establishing bureaucratic control over territory and population.

The four that were ‘prioritised for the implementation of the rehabilitation’ wereMafambisse, Xinavane, Maragra and Marromeo (INA 1996, p. 8). Maragra and Xinavaneare in government strongholds in the south, Mafambisse is in politically contested centralMozambique and Marromeo to the centre/north. For state–industry people creating whatwe above called a mediating bureaucracy there were important strategic issues thatneeded to be addressed, considering how the regional and world markets for sugar wereorganised, besides the political aspects of prioritising. One of them was how to definethe priorities for privatisation and rehabilitation so that excess productive capacity couldbe avoided in the short to medium term while the industry was given a chance tobecome productive and feasible. Another was how to align economic concerns withdiverse political concerns related to the priorities of the state and government, particularlythe issue of territorial and population control in former Renamo areas.

In simple terms, by selecting only four estates/industrial plants and transferring to thegroup of FDIs the right to the property and exploitation of the land, the state and govern-ment made it a limited contest among a closed group of companies, which could in thefuture make the most of the rent opportunities the Mozambican state offered through a pro-tected internal market.14 Importantly, there would be no ‘excluded’ competitors as such, asnone of the estates/industrial plants deselected was operational. There could be provincialauthorities and union representatives that felt marginalised as investment was transferredelsewhere but no direct sugar competitors that needed to be negotiated with. Thoseestates and industrial plants that were not selected could potentially in the future be reha-bilitated, but the time horizon would be rather long and require substantial investments.

The total cost of rehabilitating the six sugar facilities was estimated in 1996 at aroundUS$395 million for the first phase, with the two facilities along the remote Zambezi River –Marromeo in Sofala province and Luabo in Zambezia province – the most expensive, atover US$200 million.

248 L. Buur et al.

This meant that two companies were in the south (Xinavane and Maragra), with easyaccess to the industrial and technical expertise in South Africa and two (Mafambisse andMarromeo) were situated in the politically contested areas of Sofala province (generallyconsidered opposition territory), with one of these on the border with the Zambezia pro-vince. The Luabo industrial plant had been totally destroyed, was surrounded by a rela-tively limited population and could be accessed only by boat along the Zambezi River. Itwas in the end not considered feasible to rehabilitate due to the massive financial inputneeded.15 Buzi, on the other hand, would have been the cheapest to rehabilitate andhad easy access to the port of Beira. Although the Buzi River bed was changing – athreat to the industrial plant16 – the plains around the plant and the irrigation systemwere more or less intact and it would have made a favourable place to invest.17 Butexactly because of these conditions Buzi also had the best other options available withrice, prawns, livestock and cotton as alternatives and natural gas for exploitation, as itturned out.

Prioritising was also intimately linked to making the future pricing policy effective, ascapacity had first to be incrementally increased for the protected internal market beforeaccess to more lucrative export markets could be negotiated on a larger scale. Luabo, ifreconstructed, would have been the largest estate and industrial sugar plant in Mozambique,as it was during the colonial era, and Buzi the smallest. But besides the more economisticconsiderations the selection of Marromeo instead of Buzi was at its core deeply political asparty and state players believed this was the only way in which the state could establish acredible presence in the hardliner Renamo area of Marromeo and the Sofala side of themassive Zambezi river valley (based on information from provincial state and governmentofficials and industry people). To wrest control from Renamo was difficult, but over timethe Frelimo government did manage to win both national and local elections in Marromeo.This ultimate political success was directly linked to the rehabilitation of the sugar estateand industrial plant as jobs were created and crucial services delivered, showing – as itwas put to us by provincial state and government officials – that ‘the Frelimo governmentcould provide to those supporting it’.18

Therefore considering the combined weight of political and economic gains, the reha-bilitation of the four estates/plants seems to balance fairly well the diverse political, prag-matic and economic needs of the involved stakeholders. What is clear is that Castel-Branco’s argument that prioritising was done solely in order ‘to avoid excess capacity’under pressure from ‘the three large international sugar corporations’ that ended uprunning the four sugar estates and mills (Castel-Branco 2002, p. 179) does not capturethe intricate considerations that made the diverse attributes of support converge. For theMozambican state and government there were good reasons that merged both political,state administrative and economic considerations. This was not a question of creatingunfair opportunities for a group of external actors. It is easy to forget that while the 1996rehabilitation strategy anticipated the creation of a protected internal market with sufficient

Table 2. Estimated rehabilitation and upgrading costs, 1996 (US$, millions).

Xinavane Maragra Buzi Mafambisse Marromeoa Luabo Total

49.5 50 35 50 100 110 394.5

Source: Extracted from Instituto Nacional do Acucar (INA) (2000), p. 8.Note: aFor the Marromeo/Luabo sugar plant, the costs did not include the rehabilitation of the Sena railway line, onwhich the plant depended before, nor the installation of electricity from the Cahorra Bassa Dam.

profit margins to pay for the rehabilitation, this was all policy and not a concrete reality orsomething easily done. Around 90% of all sugar consumed (private and industry) in 1996was unaccounted for, in other words provided through smuggling.19 Considerable stateintervention would be needed in areas where its capacity was notoriously weak, andwhere it would be confronted with long-established trade groups connected to fundingFrelimo and the military and security establishment, as well as the state and fiscal systemand the political domain more broadly. The creation of an internal market, then, whileeasily stated on paper, was circumscribed by considerable risks.

Creating productive rent opportunities

The pricing policy that created a protected internal market was provided for in the SugarRehabilitation Strategy of 1996 and further work would be done over the followingyears by administrative staff, consultants, the industry and researchers fine-tuning, defend-ing and legitimising its existence (e.g. Gode 1997; Food and Agriculture Organization ofthe United Nations (FAO) 2000; INA 2000, 2001; LMC and Global Sugar Consulting2000; CEPAGRI 2007; LMC and Intellica 2010). From the outset the strategy gavenominal policy guarantees for investors even before the state had taken up all the loanson behalf of the industry. It came into effect in November 1997, long before the industrywas up and running. The actual implementation of the price policy, with its flexible levyon sugar imports that catered for the creation and protection of an internal market, had towait until the new millennium. When the policy therefore was approved it had littleeffect, because there was not enough sugar produced in Mozambique to protect – therewas hardly enough to honour the few and limited preferential trade opportunities onoffer to Mozambique from the United States and European Union. But, as the industrypicked up production after the 2000 floods, and as the estates/industrial plants becameoperational (with Marromeo the last one in 2002), and cane production rose and pro-ductivity increased, it became an important political issue and created substantial public dis-cussion, with the industry pressing for its implementation while different Frelimo andgovernment groups either supported or resisted implementation (see below).

The objective for creating a protected internal market was formally to ‘Create a favour-able milieu in order to stimulate new investments in the sugar industry, promote efficientproduction and minimise the obstacles for developing the sugar sector caused by thenature of the world market (extremely volatile and based on dumping prices), the fiscalpolicy (lack of stability) and high risk confronting investors’ (INA 1996, p. 10). Thiswas necessary first of all because of the residual nature of the world market, where atthat time 70% of sugar was traded in closed and protected markets and only 30% marketedopenly – and this usually only surplus production, which most of the time was sold at pricelevels below actual production costs (INA 1996, pp. 20–26; FAO 2000, p. 8). Secondly,more implicit but still traceable in the policy is what is known as ‘the infant industry argu-ment’, which suggests that in order to reach high levels of productivity based on aneconomy of scale, market protection was a sine qua non. A third reason was the Chissanogovernment’s commitment to the industry that had created jobs, provided social servicesand expanded infrastructure to rural populations in parts of the country where the statethen had limited access or legitimacy. It was acknowledged that the economic costs wereconsiderable and export markets could not yet provide sufficient revenue even to makethe industry ‘break even’.

The 1996 strategy presented the principles for calculating the flexible sobre taxa or sur-charge that would follow international price developments (INA 1996, pp. 76–77). When

250 L. Buur et al.

the price falls below a certain historically determined reference price calculated through afixed formula over a three-year period and based on various sources, a levy or import taxcomes into effect (INA 2000, p. 14; LMC and Global Sugar Consulting 2000, pp. 1–3).20

The surcharge would be calculated month by month and made public in the press.In important ways the surcharge would become the single most controversial aspect of

the whole rehabilitation strategy, and in fact it became the litmus test for state and govern-ment commitment to the sector as it involved a number of ministries and regional coordi-nation of state and government entities for enforcement, besides the key authorities inharbour and border control. Furthermore, it came to embrace reforms of complex state insti-tutions like border controls, customs and tax/value added tax (VAT) and immigration auth-orities, as well as impacting on investment policies already in effect and curbing of familytrade houses like Delta Trading (Aga Khan-owned) and Sasseka/AFRICOM, which earnedwell on importing sugar mostly informally (informality was the rule as the formal systemwas weak). Since its creation in 1995 AFRICOM, for example, had become the premierecommodities supplier in Mozambique, importing products like sugar, rice, wheat flour,maize flour, cooking oil, spaghetti, biscuits, soap and batteries. Challenging such familytrade monopolies, which often spanned several continents and had strong economicholding power and internal cohesion besides paying generously for protection to theFrelimo party and individuals, in the name of an FDI-based national industry was noteasy. Challenging trade houses importing sugar required considerable political investmentand continual liaison internally in the governing party Frelimo, as well as in the varioustrade houses financing considerable parts of the political settlement at the end of the1990s and early part of the new millennium (something these family-based trade housesstill do, which to some extent explains why productive sector development in Mozambiqueseldom takes off).

Furthermore, the surcharge initially included only raw cane sugar and not differentialcharges for raw and refined (white) sugar as all other surcharges worldwide did. Whenthe National Sugar Institute, on behalf of the industry and the government, added aspecial charge for refined sugar, it met spirited resistance. The development of a differen-tiated surcharge tax that protected the upcoming industry also thereby hindered free price-setting and made the market imperfect – responding to already imperfect market conditionsas sugar from Swaziland and South Africa, while cheaper, continued to benefit from protec-tion measures. Resistance to the surcharge came mainly from two camps: first, upstreamindustries like the beverage industry (particularly Coca-Cola and the South African Brew-eries-owned local beer companies), which had come in with substantial investments after1990, as they were the biggest formal users of refined sugar; and secondly the IMF andWorld Bank, which became nervous as the surcharge imposed at the end of 1999 (butnot effective until after the floods in 2000) raised the ghost of the state-imposed price-setting system of the 1970s and 1980s. As such it was related to the creation of a mon-opoly-like situation which was considered unfair for market competition in general.However, the pricing system was implemented after a compromise was brokered withthe sugar user industries, and after a long and fierce battle with the IMF/World Bank ledby then deputy minister for finance and planning Luisa Diogo and including all the topMozambican journalists, President Chissano, then Frelimo general secretary ManuelTome and other top government figures, as well as many other bank and emerging businessdignitaries (for example AIM No. 170, 1 December 1999).21

The controversy surrounding the establishment of the pricing policy and the internalmarket epitomised one of the most important aspects of the rehabilitation process: thecapacity to mobilise diverse societal, economic and political groups at different times.

The Mozambican government made sure that the sugar sector got able and coordinatedsupport from the intellectual critical left, the political and administrative nomenclature,the newly constructed private sector and the union movement.22 While the struggle overthe pricing policy has been used to argue that corporate ‘investors’ pressure forced theIMF to withdraw its demand for the liberalisation of the industry’ (Castel-Branco 2002,p. 201) it seems clear that the state and government indeed were key players and coordi-nated actions were taken. To see investment priorities and pricing policy as driven unilat-erally by the interests of large international sugar corporations neglects the active role ofgovernment and state as well as the diverse motivations underpinning their interests.

Conclusions

In this article we have argued that the Mozambican state and government is not just a ‘haplessvictim’ of IFIs and other donors, but has at least a degree of autonomy and capacity to shapeeconomic development according to its own priorities. We have used the case of the rehabi-litation of the sugar industry in Mozambique after the economic collapse following indepen-dence (and the subsequent destabilisation war that became a civil war) to argue that state andgovernment involvement in the partial divestment of Mozambique’s large sugar estates hasbeen greater than commonly acknowledged by ‘externalist’ perspectives. We have illustratedthis by outlining the main reasons that the industry was rehabilitated as presented by state andgovernment officials, besides presenting three cases of how the state and government wasactively involved in the rehabilitation process. While economistic reasons were plenty andusually well argued, there were in each case also clear post-conflict pragmatic, ideological,and political reasons for the state and government’s active involvement.

The support the sugar industry has received epitomises an important comparative entrypoint for understanding the mixed results of productive agricultural sectors in Mozambiquelike cashew, chicken, cotton and tobacco. Comparatively the support for the sugar sectormay also be important when relating it to the more recent venture capital investments incane-based biofuel, which seems to have lost state and government support. The fact isthat the sugar sector got able and coordinated support from the intellectual critical left,the political and administrative nomenclature and the union movement at crucialmoments. Other sectors also attempted to bring investment to rural areas where the stateand government was weak and lacked legitimacy, but few were able to combine large-scale formal job creation with social service provision on a continual basis. While asector like cotton has provided economic income opportunities for rural populations, thecost has been multiple local conflicts over inputs, sales organisation, and pricing policiesthat often required complicated and contested state and government intervention. Weknow little about these comparative aspects, but they seem crucial to explore as Mozambi-que presently stands at a crucial juncture with mega-project investments in oil, gas, energyand minerals that create economic growth with little broad-based economic development.There is in this context an increased need to understand why productive sector investmentin agriculture has had so little success in the past. One starting point could be to look atrelative successes like the sugar rehabilitation sector and systematically explore the differ-ences and similarities with other sectors – something we hope to be able to do in the future.

AcknowledgementsThanks to the participants in the Elite, Production and Poverty research programme (EPP) and thoseat the African Studies Association, UK, meeting held at St Antony’s College, University of Oxford,

252 L. Buur et al.

who shared their generous comments. We are particularly grateful to Lindsay Whitfield, Mette Kjær,Ole Therkildsen, Jan Kees van Donge, Alison Stent and Gary Littlejohn, as well as the two anon-ymous reviewers, who commented so helpfully on the article in draft form. The usual disclaimersapply.

Notes on contributorsLars Buur is a Senior Researcher at the Danish Institute for International Studies and Research Associ-ate at the Wits Institute for Social and Economic Research, Johannesburg, and at the Center for theStudies of Democracy and Development, Maputo. He has co-edited State Recognition and Democra-tization in Sub-Saharan Africa (2007) and The security–Development Nexus (2007).

Carlota Mondlane is a Master’s student at the University of Eduardo Mondlane and a research associ-ate at the Center for the Studies of Democracy and Development in Maputo.

Obede Baloi is a lecturer in sociology at the University of Eduardo Mondlane and a Senior Researcherat the Center for the Studies of Democracy and Development, Maputo.

Notes1. Furthermore, there is, despite the relevant critique of SAPs and liberal reforms, a ‘certain intrin-

sic value in such features as reining in runaway budget deficits, an end to state monopsony, anexport-oriented foreign exchange regime, a measure of privatization, and reduced tariffs onindustrial and agricultural inputs‘ (Taylor 2007, p. 9).

2. Interview with the Investment Promotion Centre (CPI), April 2010.3. As the Ministry of Planning and Development has argued over the last year, productivity has

been constantly low. The food and cash crop sub-sectors of the agriculture sector grew primarilythrough area expansion and an increase in the labour force, with a large increase in cultivatedareas in the central region after the General Peace Accord in 1992 and the first national multi-party elections in 1994.

4. This article is based on fieldwork carried out in Mozambique between February 2008 and Feb-ruary 2011 as part of the Elites, Production and Poverty (EPP) research programme (http://www.diis.dk). The methodology for the Mozambican part of the EPP involved archive studies, open-ended, semi-structured interviews with relevant actors combined with participatory observation.The main aim was to understand why the rehabilitation policy of the industry was considereddesirable and how it was made feasible.

5. In general, we follow Haggard et al., (1997, p. 38) and differentiate between government as ashifting and relatively temporary collection of political leaders, while state officials/personnelare the more or less permanent bureaucracy of the public sector.

6. Interview with management, Marromeo, 2009.7. One can argue that this position forms part of the ‘common suspicion‘ that state and business

relations attract across quite diverse theoretical positions, be they liberal or Marxist/socialistin orientation, as illustrated by Schneider and Maxfield (1997, pp. 3-5).

8. Interviews, 2009, 2010.9. Interview with National Sugar Distributor director, 2008.

10. Interview with former Minister of Finance and Prime Minister Luisa Diogo, June 2010.11. A preferential trade arrangement with the United States – amounting to 1.3% of total US

imports – allowed Mozambique to export between 14,000 and 26,000 tonnes during the1990s, stimulating a small increase in production (INA 2001, p. 4).

12. Interviews with Luisa Diogo, 2009, 2010.13. Interestingly, the continued co-ownership by the Mozambican state has not been included in any

of the discussions of the privatisation of the sugar industry (the few studies include Gode 1997;Castel-Branco et al. 2001; Castel-Branco 2002, 2008) or privatisation in general (Pitcher 2002).

14. We here follow Khan (2000a) in so far that rents based on transfer are ‘rent-like incomes . . .created by transfers organized through the political mechanism’ (p. 35).

15. Interview with DNA, 2008.16. Interview with the former Director, October 2010.

17. This now seems to be happening, 15 years later. On 3 March 2011 it was announced that Buziwould be rehabilitated at a cost of US$120 million before 2014 by way of a Portuguese invest-ment group led by Jorge Petiz.

18. Interview with Sofala, 2011.19. For this reason, starting in October 1999, the sugar industry benefited from a five-year regime of

full exemption from customs duties and taxes on a range of equipment and material associatedwith the project, including on foreign investors’ and staff members’ personal belongings.

20. It follows the formula: sobretaxa ¼ preco de referencia – preco CIF (calculated price) (INA2000, p. 14). It is INA that ‘calculates and publishes indicative c.i.f. prices for both raw andwhite sugar, based on prevailing . . . world market prices (including transport costs locally). . . and a margin of USD80/tonne for freight and insurance . . .’ (LMC and Global Sugar Con-sulting 2000, p. 2).

21. Interview with Luisa Diogo, 2009, 2010.22. Interview with General Secretary OTM, 2008.

ReferencesBartholomew, A., 2008. Private sector development. Evaluation of development cooperation between

Mozambique and Denmark, 1992–2006. Working Paper No. 09, September. Copenhagen:Danida.

Beaumont, J.W., ed., 2004. Mozambique: a closer look at the economic reform ‘poster child’. Lifeover Debt. Discussion Paper Series. December, 2, 1–8.

Bucuane, A.J. and Mulder, P., 2007. Fiscal policy and tax incidence: exploring an electricity tax onmega projects in Mozambique. National Directorate of Studies and Policy Analysis Ministry ofPlanning and Development. Republic of Mozambique. April. Discussion Paper No. 39E (17).

Buur, L., 2010. Xiconhoca: Mozambique’s ubiquitous post-independence traitor. In: S. Thiranagamaand T. Kelly, eds. Traitors: suspicion, intimacy, and the ethic of state building. Philadelphia. PA:University of Pennsylvania Press, 24–47.

Buur, L. and Baloi, O., 2009. The Mozambican PRSP initiative: moorings, usage and future. WorkingPaper No. 2009:35. Copenhagen: Danish Institute for International Studies.

Buur, L., with Mondlane, C., and Baloi, O., Forthcoming 2011a. The white gold: the role of govern-ment and state in rehabilitating the sugar industry in Mozambique. Journal of DevelopmentStudies.

Buur, L., with Mondlane, C., and Baloi, O., Forthcoming 2011b. The Mozambican sugar case: howand why the Mozambican sugar industry was rehabilitated and supported. Working Paper.Copenhagen: Danish Institute for International Studies.

Cardoso, F., 1993. Gestao e desenvolvimento rural: Mocambique no contexto da Africa sub-sahari-ana. Lisbon: Fim de Seculo.

Castel-Branco, C., ed., 2002. An investigation into the political economy of industrial policy: the caseof Mozambique, PhD thesis, Department of Economics, School of Oriental and African Studies,University of London, London.

Castel-Branco, C., 2008. Aid dependency and development: a question of ownership? A critical view.Working Paper No. 1/2008. Maputo: Instituto de Estudos Sociais e Economicos (IESE).

Castel-Branco, C., Cramer, C., and Hailu, D., 2001. Privatization and economic strategy inMozambique. WIDER Working Paper No. 2001/64. Helsinki: United Nations UniversityWorld Institute for Development Economics Research.

CEPAGRI, 2006. Uma estrategia de adaptacao para o sector acucareiro de Mocambique. SumarioExecutivo. March. Maputo: Republica de Mocambique, Ministerio da Agricultura.

CEPAGRI, 2007. Balanco do PES 2005: acucar. January. Maputo: Republica de Mocambique,Ministerio da Agricultura.

CEPAGRI, 2009. Balanco do PES 2008: acucar. January. Maputo: Republica de Mocambique,Ministerio da Agricultura.

CEPAGRI, 2010. Balanco do PES 2010 3th Trimestre: acucar. October. Maputo: Republica deMocambique, Ministerio da Agricultura.

254 L. Buur et al.

Clement, J.A.P., 2008. Introduction and overview. In: J. A. P. Clement and S.J. Peiris, eds. Post-stabil-ization economics in Sub-Saharan Africa: lessons from Mozambique. Washington. DC:International Monetary Fund, 1–10.

Clement, J.A.P. and Peiris, S. J., 2008. Sustaining growth takeoffs: lessons from Mozambique. In: J.A. P. Clement and S. J. Peiris, eds. Post-stabilization economics in Sub-Saharan Africa: lessonsfrom Mozambique. Washington. DC: International Monetary Fund, 12–81.

Cramer, C., 2006. Civil war is not a stupid thing: accounting for violence in developing countries.London: Hurst.

de Renzio, P. and Hanlon, J., 2009. Mozambique: contested sovereignty? Dilemmas of aid depen-dence. In: L. Whitfield, ed. The politics of aid: African strategies for dealing with donors.Oxford: Oxford University Press. 246–270.

Evans, P., 1995. Embedded autonomy: states and industrial transformation. Princeton. NJ: PrincetonUniversity Press.

Evans, P., 1997. State structures, government–business relations, and economic transformation. In: S.Maxfield and B.R. Schneider, eds. Business and the state in developing countries. Ithaca. NY:Cornell University Press, 63–87.

First, R., 1983. Black gold: the Mozambican miner, proletarian and peasant. Brighton: Harvest.Food and Agriculture Organization of the United Nations (FAO), 2000. Final report: Review of the

Mozambique sugar sub-sector policy. Rome: Sugar and Beverages Group. Commodity andTrade Division, FAO.

Gode, C., ed., 1997. Industria acucareira em Mocambique: producao versus importacao do acucar,Tese de Licenciatura em Economia, Faculdade de Economıa. Universidade Eduardo Mondlane.

Government of Mozambique, 1995. Establishing the basis for economic and social development: keypolicies. Report prepared for the meeting of the Consultative Group Meeting for Mozambique,Paris, March. Maputo: Government of Mozambique.

Haggard, S., Maxfield, S., and Schneider, B., 1997. Theories of business and business–state relations.In: S. Maxfield and B.R. Schneider, eds. Business and the state in developing countries. Ithaca.NY: Cornell University Press, 36–60.

Hall, M. and Young, T., 1997. Confronting Leviathan: Mozambique since Independence. London: Hurst.Hanlon, J., 1991. Mozambique: who calls the shots? London: James Currey.Hanlon, J., 1996. Peace without profit: how the IMF blocks rebuilding in Mozambique. Oxford: James

Currey.Hanlon, J., 2000. Power without responsibility: the World Bank and Mozambican cashew nuts.

Review of African Political Economy, 27 (83), 29–45.Hanlon, J. and Smart, T., 2008. Ha mais bicicletas – mas ha desenvolvimento? Maputo: Kapicua

Livros e Multimedia.Instituto Nacional do Acucar (INA), 1996. Politica e estrategias para o desenvolvimento do sector

acucareiro. Resumo Executivo. Maputo: Ministerio da Agricultura e Pescas, INA.Instituto Nacional do Acucar (INA), 2000. O sector do acucar em Mocambique: situacao actual e

perspectivas futuras. Maputo: INA. Ministerio da Agricultura e Desenvolvimento Rural,Republica de Mocambique.

Instituto Nacional do Acucar (INA), 2001. O sector do acucar em Mocambique: situacao actual eperspectivas futuras. Maputo: INA. Ministerio da Agricultura e Desenvolvimento Rural,Republica de Mocambique.

Instituto Nacional do Acucar (INA), 2004. Balanco do PES 2003: acucar. Instituto Nacional doAcucar, Janeiro. Maputo: Republica de Mocambique, Ministerio da Agricultura.

Instituto Nacional do Acucar (INA), ed., 2005a. Capıtulo 1: Contexto e antecedentes. InstitutoNacional do Acucar. Maputo: Republica de Mocambique, Ministerio da Agricultura.

Instituto Nacional do Acucar (INA), ed., 2005b. Capıtulo 7: Avaliacao do contributo socioeconomicoda industria acucareira de Mocambique. Instituto Nacional do Acucar. Maputo: Republica deMocambique, Ministerio da Agricultura.

Khan, M., 2000a. Rents, efficiency and growth. In: M. Khan and S. Jomo, eds. Rents, rent-seeking andeconomic development: theory and evidence in Asia. Cambridge: Cambridge University Press. 21–69.

Khan, M., 2000b. Rent-seeking as process. In: M. Khan and S. Jomo, eds. Rents, rent-seeking and econ-omic development: theory and evidence in Asia. Cambridge: Cambridge University Press. 70–142.

Khan, M., ed., 2003. Corruption and the capitalist transformation: analysis, policy and the real world,Paper presented at the International Conference on Re/Constructing Corruption, University ofEast Anglia, Norwich, UK, 30 April–2 May.

Khan, M. and Jomo, K.S., 2000. Introduction. In: M. Khan and S. Jomo, eds. Rents, rent-seeking andeconomic development: theory and evidence in Asia. Cambridge: Cambridge University Press.1–20.

LMC, 2004. Appraisal of the impact of sugar pricing policy and investment in the sugar industry ofMozambique. Maputo: National Sugar Institute. Mozambique.

LMC, 2006. An adaptation strategy for the Mozambican sugar industry. Main Report. Maputo: Unitfor the Promotion of Commercial Agriculture, GSPCA, LMC International Ltd.

LMC and Global Sugar Consulting, 2000. Mozambique sugar sector review. June. Oxford: LMCInternational and Cheltenham: Global Sugar Consulting.

LMC and Intellica, 2010. A review of the sugar pricing policy in Mozambique. Draft ExecutiveSummary, September. Maputo: CEPAGRI.

Locke, A., 2009. Review of European Commission accompanying measures programmeMozambique. Draft. Maputo: European Commission.

Machel, S., 1981. Sowing the seeds of revolution. Harare: Zimbabwe Publ. House.Macuane, J., ed., 2010. Elites, pro-poor policies and development in Mozambique, Paper presented at

the International Workshop ‘Elites, Production and Poverty’, Accra, Ghana, 13–15 January 2010.Maxfield, S. and Schneider, B., 1997. Business and the state in developing countries. Ithaca. NY:

Cornell University Press.Ouma, S. and Whitfield, L., Forthcoming 2011. The making and remaking of agro-industries in

Africa. Journal of Development Studies.Pitcher, A., 2002. Transforming Mozambique: the politics of privatization, 1975–2000. Cambridge:

Cambridge University Press.Plank, D., 1993. Aid, debt and the end of sovereignty: Mozambique and its donors. Journal of Modern

African Studies, 31 (3), 407–430.Riddell, J., 1992. Things fall apart again: structural adjustment programmes in Sub-Saharan Africa.

Journal of Modern African Studies, 30 (1), 53–68.Saul, J., 1993. Recolonization and resistance in Southern Africa in the 1990s. Trenton. NJ: Africa

World Press.Schneider, B.R. and Maxfield, S., 1997. Business, the state, and economic performance in developing

countries. In: S. Maxfield and B.R. Schneider, eds. Business and the state in developing countries.Ithaca. NY: Cornell University Press, 3–35.

Schut, M., Slingerland, M., and Locke, A., 2010. Biofuel developments in Mozambique: update andanalysis of policy, potential and reality. Energy Policy, 38, 5151–5165.

Sena, 2009. Introduction to Companhia de Sena SA (CDSS), Marromeo, Mozambique. SENA:Tereos/Guarani.

Sousa, C. de and Sulemane, J., 2007. Mozambique’s growth performance, 1960–96. In: B. J. Ndulu,S. A. O’Connell, J.-P. Azam, R. H. Bates, A. K. Augustin, J. W. Gunning and D. Nijinkeu, eds.The political economy of economic growth in Africa, 1960–2000, Vol. 2: Country Case Studies.Cambridge: Cambridge University Press [CD-ROM]

Taylor, S., 2007. Business and the state in southern Africa: the politics of economic reform. Boulder.CO: Lynne Rienner.

Tvedten, I., Margarida, P., and Carmeliza, R., 2006. ‘Opitanha’ social relations of rural poverty innorthern Mozambique. Report No. 16. Bergen: Chr. Michelsen Institute (CMI).

Tvedten, I., Margarida, P., and Carmeliza, R., 2009. Monitoring and evaluating Mozambique’spoverty reduction strategy PARPA 2006–2008. A synopsis of three qualitative studies on ruraland urban poverty. Report No. 5. Bergen: Chr. Michelsen Institute (CMI).

Uaiene, R., ed., 2010. As ligacoes entre o agricultura e a industria em Mocambique, Paper presented atthe Conferencia sobre a Competitividade Industrial, Associacao Industrial de Mocambique(AIMO), Maputo, 22 September 2010.

Villalon, L. and Huxtable, P., 1998. The African state at a critical juncture: between disintegrationand reconfiguration. Boulder. CO: Lynne Rienner.

Virtanen, P. and Ehrenpreis, D., 2007. Growth, poverty and inequality in Mozambique. Country StudyNo. 10, September. International Poverty Centre. Brasilia: International Poverty Centre.

Whitfield, L., 2009. The politics of aid: African strategies for dealing with donors. Oxford: OxfordUniversity Press.

World Bank, 2005. Country Economic Memorandum: Mozambique. Report No. 32615-MZ, 27September. Washington, DC: The World Bank.

256 L. Buur et al.

Strategic privatisation: rehabilitating the Mozambican sugar industry

Documents