INTRODUCTION
Scholarly engagement continues to be split regarding state intervention as a strategy that should be pursued by late developing countries.Footnote 1 State intervention was stigmatised by arguments made in the 1970s and 1980s, which associated such policies with wasteful rent seeking (Krueger Reference Krueger1974; World Bank 1981).Footnote 2 Such reasoning dominated the policy prescriptions proposed by donors in the decades that followed. Policy prescriptions included governance reforms such as securing property rights, promoting transparency and building a ‘facilitating’ business environment. However, the successful examples of South Korea and Taiwan (Amsden Reference Amsden1989; Wade Reference Wade1990) did not embrace the policies that donors suggested. These countries and other East Asian late developers allocated rents to productive enterprises through state intervention. If anything, the governance reforms proposed by donors were more closely associated with stabilising growth rates after periods of growth accelerations (Khan Reference Khan, Ocampo, Sundaram and Vos2007; Sen Reference Sen2013). Recently, discussions about state intervention and industrial policy have been reinvigorated (Stiglitz et al. Reference Stiglitz, Lin, Monga and Patel2013). The debate now focuses on the kind of state interventions that should be encouraged in late developing countries. The World Bank's Justin Lin (Reference Lin2012) argues for a ‘facilitating state’ that marches neatly in line with comparative advantage, while heterodox scholars (Chang Reference Chang2002) argue that countries should take big risks through ‘picking winners’ and promoting infant industries.Footnote 3
Conducting industrial policy is always a political act. The allocation of subsidies and benefits to certain enterprises favours certain individuals or groups over others. In 2009, Douglass North and his colleagues (North et al. Reference North, Wallis and Weingast2009) embraced such ideas, reacting against the Good Governance paradigm (which North had helped establish). Their study (North et al. Reference North, Wallis and Weingast2009) began by assuming that the challenge of late development begins with the problem of endemic violence. In order to solve that problem, ruling elites sustain power through organising Limited Access Order (LAO) societies. In LAOs, ruling elites organise powerful members of society into a coalition of military, political, religious and economic elites. These elites create limits on the access to valuable political and economic functions as a way to generate rents. Powerful individuals who receive rents are motivated to cooperate with the ruling coalition and be peaceful. These individuals choose not to resort to violence because such actions would threaten their access to rents. This forms the basis of a stable elite bargain, which underpins any political settlement (Di John & Putzel Reference Di John and Putzel2009).Footnote 4
As bargains are struck between elites and threats emerge from excluded rivals, dominant coalitions must develop instruments to ensure their power is greater than that of anyone who may threaten their position. Establishing systems of centralised patronage was one way in which ruling elites were able to manage threats from rival elites (Di John & Putzel Reference Di John and Putzel2009). The existence of such structures reduced the possibility of substantial capital accumulation and political power outside formalised channels. Kelsall (Reference Kelsall2013) provides examples of countries across Asia and Africa that successfully centralised rents, orienting them to strategic industrial polices, which had long-term objectives.
Political parties and militaries are often preferred avenues through which rents are centralised. These institutions embody the ideology that legitimises the power of ruling elites. Political parties and militaries have established businesses in several countries. Agebaz (Reference Agebaz2013) showcases the use of such enterprises in Malaysia, Taiwan, Ethiopia and Rwanda. Malaysia's ruling party businesses provided evidence of cronyism but also created employment opportunities and empowered the Malay middle class (Agebaz Reference Agebaz2013). Taiwan's Kuomintang used party businesses to diversify the Taiwanese economy. Different power centres slowly emerged through empowering party conglomerates and this provided ‘a soft landing towards multi-partyism’ (Agebaz Reference Agebaz2013: 1478). Ethiopia's most prominent party-owned business – The Endowment Fund for the Rehabilitation of Tigray (EFFORT) –was also used to provide initial investments in strategic sectors. EFFORT retained monopoly control over sectors while some companies operated in liberalised environments.Footnote 5 Some scholars argue that EFFORT has been associated with corruption and that its presence ‘hinders competitors from raising capital’ (Hagmann & Abbink Reference Hagmann and Abbink2011: 588). Though many of the incidences of party capitalism that have been highlighted have contributed to long-term developmental benefits, some scholars have also argued that these enterprises may have also been used for other unproductive or corrupt purposes.
Arguments regarding the use of party and military enterprises in Rwanda are developed by two opposing sets of authors. One group (Booth & Golooba-Mutebi Reference Booth and Golooba-Mutebi2012; Kelsall Reference Kelsall2013) argues that such enterprises show ruling elites have committed to ‘long-horizon rent centralisation’. They argue that such enterprises have invested in strategic sectors and contributed to structural transformation. The other group (Gokgur Reference Gokgur2012; Reyntjens Reference Reyntjens2013) takes the neoliberal argument, assuming that the close relationship between such enterprises and the government, coupled with the existence of a ‘narrow’ group of elites, is evidence that such enterprises are used for unproductive purposes. Like other critics of Rwanda, this group contends that such control over resources may eventually result in political instability. Both these groups argue that elites work through consensus. However, the first group argues that such ‘consensus’ is motivated by developmental impulses, while the second group argues that elites work together to enrich themselves. This article argues that these enterprises are legitimised by the vulnerability experienced by ruling elites. Ruling elites are compelled to stay disciplined in line with their goals because of the insecurity of their positions. Such insecurity forces ruling elites to commit to delivering economic development. This is in agreement with many scholars, including Golooba-Mutebi (Reference Golooba-Mutebi2008), who argue that ruling Rwanda Patriotic Front (RPF) elites perceive their leadership to be legitimised by achieving economic development. Such vulnerability was central to the experience of many developmental states (Castells Reference Castells, Henderson and Appelbaum1992; Putzel Reference Putzel and Masina2002; Doner et al. Reference Doner, Ritchie and Slater2005). It has also contributed to successful industrial policies elsewhere on the African continent (Whitfield & Buur Reference Whitfield and Buur2014).
This article begins by showing that ‘vulnerability’, rather than ‘consensus’, drives elites in Rwanda to commit to economic development. Most scholars tend to characterise the political settlement in Rwanda as ‘inclusive’ or ‘exclusive’, generally supposing that the former fosters economic development while the latter acts as a prelude to civil war. This article argues that rather than being perceived as ‘inclusive’, ‘inclusive enough’ or ‘exclusionary’, the Rwandan case shows that ‘countries might oscillate, Albert Hirschman style, between more exclusionary and more inclusive arrangements for assuring stability’ (Levy Reference Levy, North, Wallis, Webb and Weingast2013).Footnote 6 Oscillating bargains between elites better describe the dynamic nature of power shifts within political settlements.
In order to achieve productivity in sectors, governments must build alliances with capitalists. If market competition is embraced, the government must find ways to steer private actors in line with national targets. In Uganda, the formation of effective alliances (with capitalists who worked with the ruling party) facilitated growth in particular sectors (Kjaer Reference Kjaer2015). The government must build partnerships with capitalists while excluding others and oscillating between exclusionary and inclusive arrangements. Though party and military enterprises may provide ruling elites with the benefit of centralising rents, the use of such enterprises antagonises individuals who miss out. The following sections detail the debate on the use of party and military enterprises in Rwanda followed by two case studies of particular sectors: pyrethrum and mining. Horizon's investments in the pyrethrum sector resulted in increased productivity and exports, as well as progress in value-addition through the domestic production of pesticides. Despite these benefits, continued productivity stalled because of fluctuations in international prices. The accumulation strategies that have been facilitated in this sector are likely to have occurred alongside negative consequences for segments of the population (Huggins Reference Huggins2014), which is common among processes of capitalist accumulation. Horizon's investments have been much more productive than those of previous private sector owners, although global price fluctuations limited the potential for growth. Meanwhile, selling state-owned assets in the minerals sector, as well as liberalising trade-and-export, has shown that the Rwandan government has chosen to avoid centralising rents in the sector and to disband existing networks with actors in the Democratic Republic of Congo (DRC). Since foreign actors now dominate the sector, the government has relied on legislation to retain control over policymaking. Liberalising too quickly has reduced the government's capacity to enforce legislation and has forced it into a ‘capability trap’ – where governments adopt reforms to ensure continued flows of external financing but do not retain the institutions to ensure the functioning of those reforms (Andrews et al. Reference Andrews, Pritchett and Woolcock2013). There have been achievements in increasing production and exports, as well as promoting value-addition. However, the government struggles to discipline foreign investors and ensure that the continued development of domestic mining occurs in line with strategic goals.
Neither strategy pursued by the government has achieved perfect outcomes. Instead, these examples show that the Rwandan government works ‘reactively’ to achieve developmental goals. The government navigates political vulnerabilities (and the demands of donors) before deciding how best to pursue alliances with firms in different sectors. In both sectors, the government's control over policies has been blocked (either by international prices or by fast-paced liberalisation, which left the government with inadequate capacity to manage the private sector). However, both sectors show evidence that vulnerability motivates ruling elites to pursue economic development, rather than pursue short-term goals.
‘CONSENSUS’ AND ‘INCLUSIVENESS’ IN POLITICAL SETTLEMENTS
The distribution of power among individuals and groups is dynamic during the process of late development. Scholarship (Khan Reference Khan2010; Acemoglu & Robinson Reference Acemoglu and Robinson2012) and policy work (DfiD 2010; OECD 2011) have favoured ‘inclusive’ elite bargains or political settlements above exclusive ones. Favouring ‘inclusive’ political bargains over ‘exclusionary’ ones has a long tradition. Ingram (Reference Ingram2014) locates Lijphart's (Reference Lijphart1977) ‘consociational theory’ as influential in this tradition. Lijphart argued that power-sharing structures could achieve stable democratic governments in plural societies where political alignments were fragmented through ethnic or religious divisions. Lindemann (Reference Lindemann2008) built on such ideas, arguing that political stability across Sub-Saharan Africa is determined by the capacity of ruling political parties to overcome historical levels of high social fragmentation and maintain ‘inclusive’ elite bargains. He suggests this was the case in Zambia where the presence of an ethnically inclusive coalition explained the absence of civil war in the post-independence period. Meanwhile, Uganda's exclusionary elite bargains ‘produced enduring antagonism between the country's major tribal groups and thereby became a key driving force behind the various insurgencies since 1962’ (Lindemann Reference Lindemann2010: 59). However, the empirical support for favouring ‘inclusive’ elite bargains above their ‘exclusive’ counterparts is questioned by others. Vu (Reference Vu2007) argues that elite polarisation was associated with developmental outcomes in East Asia. Meanwhile, ‘inclusive elite coalitions’ in Botswana were associated with structural inequality and destitution in some marginal groups (Hickey Reference Hickey2013). The consensus has shifted around favouring inclusive coalitions, with the language ‘shifting to achieving an ‘inclusive enough’ settlement or coalition’ (Ingram Reference Ingram2014: 10).
Rwanda scholars have also failed to take into account the possibility that a mixture of inclusive and exclusionary political dynamics take place during development transitions. Scholars who are positive about the RPF government have argued that the current government is ‘more inclusive than at any other time after independence’ (Golooba-Mutebi Reference Golooba-Mutebi2008: 29). The Rwandan Alliance for National Unity (RANU) was created as an ‘inclusive’ political party. RANU was renamed the RPF in 1987 and continued to retain the same inclusive rhetoric. Both Hutu and Tutsi refugees were recruited to the ‘liberation effort’. Senior Hutu RPF figures Alexis Kanyarengwe, Pasteur Bizimungu, Seth Sendashonga and Theoneste Lizinde assumed short-lived leadership positions in the first government after 1994. Hutu businessman Silas Majyembere contributed funds to the RPF's military effort. By 2000, many of the leading Hutu RPF members left Rwanda or lost their political positions (Prunier Reference Prunier2009; Reyntjens Reference Reyntjens2013). Those who left were replaced with other Hutu RPF members. For example, Bernard Makuza took the place of Pierre-Celestin Rwigema as Prime Minister in 2000. Despite these changes, Booth & Golooba-Mutebi (Reference Booth and Golooba-Mutebi2013: 13) argued that the RPF operated ‘a policy of robust inclusiveness: a willingness to adopt into the nation building project almost anyone willing to join it’.
Critics (Reyntjens Reference Reyntjens2004, Reference Reyntjens2006, Reference Reyntjens2011; Ansoms Reference Ansoms2009; Beswick Reference Beswick2010; Ingelaere Reference Ingelaere2010; Gokgur Reference Gokgur2012) argued that the RPF has consolidated control over politics and the economy among an increasingly narrow coalition of elites. The RPF has been accused of attacking several opponents abroad, including Kayumba Nyamwasa, Patrick Karegeya, Seth Sendashonga and Theoneste Lizinde (see ICG 2002; HRW 2014). Former RPF members – Kayumba Nyamwasa, Theogene Rudasingwa, Patrick Karegeya and Gerald Gahima – established the Rwanda National Congress (RNC) in 2010. The RNC has become a very vocal opponent of the RPF abroad (Nyamwasa et al. Reference Nyamwasa, Rudasingwa, Karegeya and Gahima2010).
Very few ‘historicals’ (individuals who held key positions during the liberation effort in the 1990s) retain key military and political positions.Footnote 7 However, the Rwandan government is not without Hutu ministers. Francois Kanimba has been Minister for Trade and Industry since 2010 and worked as the Governor of the Central Bank previously. Former Prime Minister Bernard Makuza now heads the Senate. The two Prime Ministers that succeeded Makuza are also Hutus (Pierre Damien Habumuremyi and Anastase Murekezi). Critics often highlight instances when individuals leave Rwanda as signs of an ‘exclusionary’ state. However, some prominent individuals who left Rwanda have also returned. Former Prime Minister Rwigema returned to Rwanda and was elected as representative of Rwanda to the East African Legislative Assembly in 2012. Other former opponents including Gerard Ntashamaje and Evode Uwizeyimana have also been reintegrated. Former Rwanda Defence Forces (FAR) officers, who fought against the RPF, have also been reintegrated into Rwandan society. By 1998, 38,500 ex-FAR officers were reintegrated into the army (Jowell Reference Jowell2014). Senior figures have been given positions of responsibility. Examples include Paul Rwarakabije (Rwanda Correctional Services), Jerome Ngendahimana (Deputy Chief of Staff – Reserve Force), Daniel Ufitikirezi (RSSB), Andre Habyarimana (Head of Reserve Force, Northern Province), Evariste Murenzi (Commanding Officer of the Rwanda Mechanised Infantry Battalion) and Albert Murasira (CSS Zigama). Additionally, the lower ranks of the military are said to contain more Hutus than Tutsis (Golooba-Mutebi Reference Golooba-Mutebi2013).
Critics are right to highlight that severe inequality persists in Rwanda. Though supporters (Booth & Golooba-Mutebi Reference Booth and Golooba-Mutebi2014) stress that surveys showed that inequality fell between 2006 and 2011 in Rwanda, the country remains the most unequal country in East Africa (by both the Gini and Palma indexes).Footnote 8 However, the government has attempted to redistribute the economic growth experienced over the last two decades. Redistribution has largely taken the form of providing health and education services. The Mutuelle de Sante (mutual health insurance) system in Rwanda is one of the most extensive community-based health insurance schemes across the continent and extends coverage to over 90% of the population. The government has also implemented the ‘Vision 2020 Umurenge Programme’ (VUP), which aims at assisting the most vulnerable portions of the population to exit poverty. Under VUP, extremely poor households where adults were not able to work were entitled to cash transfers, while households with adults who could work were given employment in public works projects (Devereux Reference Devereux2012). Providing access to wages for an educated, young population is a challenge (Somers Reference Somers2012), with more than half of the population under 19 years old in 2015.Footnote 9 Matching skills received through university education with jobs is another problem. In 2015, it was reported that 45% of masters degree holders in Rwanda were underemployed (Asaba Reference Asaba2015). Though the government has attempted to redistribute some of the gains from economic growth, inequality and poverty continue to be serious issues.
DEBATING PARTY CAPITALISM
Critics of the Rwandan government (Gokgur Reference Gokgur2012) have argued that the RPF's control of the economy is evidence of its exclusionary policies. Gokgur (Reference Gokgur2012) applies the term ‘partystatal’ collectively to such groups. However, this term is misleading since party-owned enterprises differ from military-owned enterprises or enterprises owned by private individuals who pool investments in line with national goals. Instead, ‘investment groups’ is the term assigned to these enterprises in Rwanda. Investment groups are divided into ‘formal’ investment groups and ‘informal’ investment groups. Formal investment groups include those companies which have a formal connection to the military (RDF) or the party (RPF). ‘Informal’ investment groups comprise those companies (e.g. Rwanda Investment Group) where ‘private’ investors have pooled their own money to create investment groups. President Paul Kagame has also directed the formation of different informal, regional investment groups (even investing his own money) (Kayobotsi 2012 int.; Mbundu 2012 int.). The central mandate of investment groups was to address the strategic needs of the nation and to invest in new sectors, build the sector through a subsidiary company and exit the sector once the subsidiary company had broken even (and when a suitable price had been offered) (Gatera 2012 int.; Mirenge 2012 int.). Respondents often cite two examples (tourism and telecom) where these strategies were successful.
Prime Holdings – an RPF-owned investment company – invested in the refurbishment of two hotels – the Kigali Intercontinental Hotel and the Kivu Sun Hotel in Gisenyi. The government prioritised such investments despite many donors advising against such investments (Mirenge 2012 int.). In 2007, the Serena Group signed a 30-year lease agreement to take control of the two hotels. The Aga Khan Fund for Economic Development now owns the Serena Hotels. Between 2001 and 2012, the tourism sector attracted the most investments in Rwanda (RDB 2013). Foreign brands have been attracted to the sector, with the Marriott Hotel currently under construction in Kigali and the Kempinski Hotels taking over management of the Hotel Des Milles Collines. Other foreign investors, wealthy Rwandans and smaller domestic operators own assets in the sector. In 2012, the travel and tourism sector directly employed 54,000 Rwandans and its total contribution to employment was equivalent to 135,800 jobs (6·4% of total employment) (UNCTAD 2014: 26). Between 1996 and 2013, tourism revenues increased from $12 million to $281.8 million.
In 1998, Tri-Star (now CVL) partnered with the MTN Group to establish a GSM mobile network in Rwanda. MTN Rwanda became Rwanda's dominant mobile telephony company. However, the telecom sector was gradually liberalised, with Tigo and Airtel occupying substantial shares of the market. CVL has gradually reduced its shareholding in MTN and has committed to selling its remaining 20% stake on the Rwanda Stock Exchange in 2015 (Namata Reference Namata2015). Another example is CVL's Bourbon Coffee. Bourbon coffee shops were originally established in Kigali, the UK and the US. Now, there are more than 15 coffee shops in Kigali. Nigerian investors – Kaizen Noir – have also opened shops in Lagos and Kigali. Local Rwandans and foreign nationals have also established coffee shops.
Tri-Star Investments was the first RPF investment group. It grew out of the RPF Production Unit. The Production Unit functioned as a treasury for political contributions to the RPF's ‘liberation effort’ in the early 1990s. The initial funders included sympathetic diaspora and Banyarwanda refugees from Uganda and other neighbouring countries. After the war, the Production Unit was transformed into Tri-Star (a holding company which was fully owned by the RPF). Initial investments were made in line with the national interest e.g. the provision of social services, rebuilding roads and transport links.
Some things must be contextualised within our history – especially where Rwanda has come from since 1994. We had to find ways to provide basic services to people. The previous government had stolen most of the money from the treasury. Tri-Star invested in things the people needed. People saw there was potential and then Tri-Star left. This was a way the government motivated people to invest. Actually, it was to help privatisation and to entice new investors to come here. (Gatare 2012 int.)
Tri-Star was involved in a variety of sectors including metals trading, mobile telephony, road construction, housing and food processing. One of its subsidiaries – Rwanda Metals (RM) – was involved in the trading of minerals from the Congo. RM was sold to a Botswana-based firm in 2002 amidst initial outcry over allegations that the firm traded ‘conflict minerals’ (Booth & Golooba-Mutebi Reference Booth and Golooba-Mutebi2012). There is disagreement about the extent of profits made by companies like RM.Footnote 10 Companies such as these were vital in centralising profits through the Congo Desk towards the war effort. These companies were unlikely to have become avenues through which individuals accumulated profits. Several loyal RPF officers occupied positions within RM, including Dan Munyuza (a junior military officer at the time and the current Deputy Inspector General of the Rwandan Police) and Francis Karimba (RM's former director and the current Commercial Attaché at the High Commission of Rwanda in South Africa) (Cuvelier & Raeymaekers Reference Cuvelier and Raeymaekers2002; Cuvelier & Messiant Reference Cuvelier and Messiant2004). Tri-Star did not enter other traditional export sectors i.e. coffee or tea.
Tri-Star was renamed Crystal Ventures Ltd (CVL) in 2009. The company, like its earlier incarnation, was a ‘first-mover’ in several sectors including telephony, civil works and property management. In 2014, CVL had a 50% stake or more in 12 companies operating in Rwanda. CVL's subsidiaries were the largest players in the sectors in which they operated. CVL had joint ventures with several foreign and domestic companies, including other investment groups. CVL employs between 70,000 and 100,000 Rwandans and values itself at $500 million (Mazimpaka Reference Mazimpaka2013), roughly 1/13th of the value of the entire economy.
Booth & Golooba-Mutebi (Reference Booth and Golooba-Mutebi2012) argued that CVL acted as an attractive partner for international investors, enticing investors to come to Rwanda. Gokgur (Reference Gokgur2012) disagreed with such claims. She suggested that the presence of investment groups inhibited domestic entrepreneurship and foreign investors. Gokgur claimed that domestic entrepreneurs and investors were reluctant to compete in environments where such companies held majority controls over operations. However, the presence of investment groups has not stopped businesses from being registered. Increasing numbers of companies continue to be registered across the country (Namara 2012 int.). In 2011, SMEs accounted for 95% of all registered businesses and accounted for 84% of private sector employment (AfDB 2014). Most of these companies did not survive beyond three years (MINICOM 2010). Complaints usually included problems in accessing finance, high costs of doing business, difficulties in finding a market for goods and heavy taxes (Mukubu 2015 int.). Investment groups are now encouraged to support the creation of small and medium enterprises (SMEs). The government, through its Economic Development and Poverty Reduction Strategy 2 (EDPRS 2), encourages investment groups and other large companies to develop supply chains through backward linkages with SMEs (MINECOFIN 2013). Thus, there is no evidence that foreign investment or the growth of domestic companies is hampered by benefits received by investment groups.
Sure, it isn't easy to enter the market. But these companies are not the problem. I have an advantage because my partners and I are young. We have international experience and prior expertise. These groups are not efficient enough. It is easy to blame them but there are other reasons why companies don't survive. (Domestic Investor 2013 int.)
While working as a consultant in Rwanda, Gokgur (Reference Gokgur2012) identified the presence of investment groups as an impediment to attracting foreign investment. Donors approached the government and questioned them about their use of these investment groups (EU, UK and Dutch Embassy Officials 2012 int.). The RNC and other critics highlighted the obstructive presence of these companies and accused the government of giving these groups benefits that were not accorded to other individuals in the private sector (Nyamwasa et al. Reference Nyamwasa, Rudasingwa, Karegeya and Gahima2010; Rudasingwa Reference Rudasingwa2013). In 2012, after criticism of CVL's performance, there was a change in management (showing the government's sensitivity to perceptions around these companies). Trusted Kagame confidante Nshuti Manasseh was replaced by Jack Kayonga, the former head of the Rwanda Development Bank (BRD). Elias Baingana, who had served as DG of the National Budget in the Ministry of Finance and Economic Planning (MINECOFIN), was brought in as Chief Operating Officer.
The government defended CVL's failures and attributed them to mismanagement rather than corruption. However, top officials at investment groups have also been accused of corruption. In 2010, Horizon went through a number of changes when top management officials including Lt. Col. Paul Semana and five other top-ranking military officers were accused of embezzling up to $700 million (Ndikubwayezu Reference Ndikubwayezu2009). John Zigira (Semana's successor) was also arrested for embezzling funds. Both Semana and Zigira were retired during an episode of mass retirements in 2013. Changes in the organisational structure of the company followed. In 2010, Eugene Haguma became the new CEO of Horizon and Gabriel Bizimungu was installed as the MD of SOPYRWA. This incident shows that while these companies are not averse to becoming avenues for personal corruption, the government disciplines those who are accused of such activities. Conversely, it could also be argued that charges of corruption are used by ruling elites to legitimise choices to oust certain officers (since both Zigira and Semana were rumoured to have been close to Nyamwasa, who fled the country in 2010). Nevertheless, it is clear that the Rwandan government prioritises the centralisation of rents. When ‘centralisation’ is threatened, either when there is a possibility that certain individuals may become rivals or through individual acts of corruption, the dominant coalition works to consolidate control over rents.
Formal investment groups also included RDF-owned companies.
These companies work as a ‘demonstration effect’ and the military companies are at the centre of it. The thinking is that the military is disciplined and successful. Let's use these resources and get them to do public works, logistics etc. The military will push themselves better than others because they are trained in the ideology. (Consultant A 2012 int.)
The creation of Horizon Group (in 2007) led to a distinction between the party and the military in the economic sphere. The Ministry of Defence does not own Horizon directly. Instead, Horizon has two shareholders – Military Medical Insurance (MMI) and Credit Savings Society Zigama (CSS Zigama).Footnote 11 The military was previously involved in undertaking socio-economic projects in line with national priorities. For instance, the military built the first coffee washing stations in Rwanda. Horizon has three subsidiary companies: Horizon Construction, Horizon Logistics and Horizon SOPYRWA. The board of Horizon did not include any military officers, apart from its CEO. Officially, Horizon and other military companies operate autonomously. In reality, hiring-and-firing decisions and other forms of pressure are exerted directly from the Ministry of Defence (or higher up) (Nyamvumba 2011 int.; Haguma 2013 int.). There are two other military investment groups (Agro-Processing Industries Ltd and Ngali Holdings), which have not been identified hitherto in the literature.
There are many informal investment groups that operate in Rwanda. Among these groups is Rwanda Investment Group (RIG), which was created in 2006. The creation of RIG was part of a national drive to pool investments in strategic sectors. There are also regional investment groups and sector-specific investment groups. The Petrocom Group, headed by Egide Gatera, is another example of such a group.
It was all the President's initiative. Businessmen sometimes complained about lack of funds when they were asked to invest in national projects. He called some of the most important Rwandan businessmen to his farm and asked them: ‘Can you think bigger than sole ownership?’ From that meeting, RIG started getting capital. RIG went into energy and cement. These investments are made where the country needs it most. (Mirenge 2012 int.)
Originally, RIG's shares were split between Hatari Sekoko, Tribert Rujugiro, the National Social Security Fund (NSSF), Tri-Star and various private individuals (Gokgur Reference Gokgur2012).Footnote 12 The share base transformed over time. Within the first two years of operation, RIG invested over 11 billion RwF, mostly in energy, cement and real estate (RIG 2009). Investments today are restricted to energy and cement – two strategic sectors. Investments in energy are crucial to solving infrastructural constraints. Cement is among Rwanda's top imports. The cement company, CIMERWA, had previously been managed by CBMC, a Chinese company, under a Build, Own, Operate and Transfer (BOOT) agreement that expired in July 2006. The government then sold a 90% stake in the company to RIG. ‘Since RIG's acquisition, the company has been operating with an installed capacity of 100,000 tonnes per year to serve the domestic market, although the estimated domestic demand is approximately 370,000 tonnes’ (Gathani & Stoelinga Reference Gathani and Stoelinga2013: 168). After RIG failed to raise additional funding, it was forced to sell a large percentage of its shares to the BRD and RSSB (Gathani & Stoelinga Reference Gathani and Stoelinga2013). RIG is a relatively smaller enterprise than formal investment groups.
The existing academic literature is quick to judge the outcomes associated with the use of investment groups. The investment groups should be analysed on a sector-specific basis with emphasis on understanding rent seeking as a process. The net effect of rent seeking should be equated with the net social benefits associated with rent outcomes minus the cost involved in seeking rents (Khan Reference Khan, Khan and Sundaram2000). The next section details the example of Horizon SOPYRWA, describing how investments by Horizon resulted in immediate gains. Individual owners failed to make the most of ‘first-mover advantages’ that were originally allocated to them. Judging the strategy as a ‘process’ in relation to wider social benefits shows evidence of control grabs, according to Huggins (Reference Huggins2014).Footnote 13 However, Huggins (Reference Huggins2014) (like Gokgur) simplistically assumes that these companies are being used to make elites rich (without any evidence). It is more likely that control grabs are being used to speed up the pace of class formation in rural areas. Though extremely exploitative in the short-term, the government perceives such policies as ways to modernise agriculture methods, making the most productive use of its land and providing employment in the long term through diversification (Office of the President 2012 int.). Through this strategy, local elites are empowered to organise labour to adopt certain production methods. Horizon was successful in organising labour to production strategies that increased productivity. The government was willing to facilitate investments when rents were under centralised control. Although there was a negative impact for those who worked in the sector, Horizon's investments achieved results (increasing production and exports, while also upgrading activities through value-addition). Exogenous factors, including a fall in international prices, have blocked progress. However, it is clear that Horizon's investments resulted in more productive outcomes than those of previous investors.
INVESTMENT GROUPS IN ACTION: THE PYRETHRUM SECTOR
The Rwandan government focused on three goals in revitalising the pyrethrum sector. First, it aimed at ensuring that farmers and wageworkers were organised to cultivate pyrethrum (and that they received necessary inputs) in order to ensure that consistent quantities of pyrethrum were supplied to the factory. Second, owners of the factory would invest in the necessary technological innovations to improve the quality of pyrethrum extracts produced in the factory. Third, a portion of the pyrethrum produced in Rwanda would be used to develop end-use pesticides for the domestic market.
Pyrethrum is a natural flower-based pesticide, which is used as an alternative to synthetic chemical pesticides to control a wide range of pests. It is considered to be among Rwanda's four ‘traditional’ exports, including coffee, tea and mining. Pyrethrum was introduced in Rwanda in 1936. In 1963, Gregoire Kayibanda's government expanded the pyrethrum industry by granting two hectares of land to each individual farmer in the Virunga region, located in north-west Rwanda. The rich volcanic soil in the north-west and the high altitudes suited the production of pyrethrum. Rwanda is one of the few countries in the world (including Kenya, Ecuador, Tanzania, Australia and Papua New Guinea) that produces significant quantities of pyrethrum. In the 1970s, the government merged a ‘planters association’ with an industrial facility, creating OPYRWA – a company that processed the flowers for export as a raw pyrethrum extract. The Habyarimana government increased production until the global market suffered in the 1980s and 1990s. Until 2001, domestic pyrethrum production remained at low levels because of conflict in north-western Rwanda. The government privatised OPYRWA in 2001. The company was sold to four businessmen, including Paul Muvunyi who served as Chairman and Managing Director. OPYRWA was among the 31 public companies privatised between 1997 and 2003.
Before Horizon's takeover
OPYRWA was renamed SOPYRWA by the private investors. Then Finance Minister Donald Kaberuka urged Muvunyi and other owners to invest in production and improvements in quality (Engineering News Online 2002).Footnote 14 As international prices for pyrethrum increased, domestic production of pyrethrum flowers steadily decreased between 2004 and 2008 (Figures 1–3). While SOPYRWA was in ‘private’ hands, production declined from 1350 metric tonnes (mt) of dried flowers in 2004 to 209 mt in 2008. Exports increased in 2006 and 2007 because regional production was being sent to Rwanda after a fire in Kenya's pyrethrum processing factory (Haguma 2013 int.).Footnote 15 Muvunyi claimed that farmers received increased prices after the factory was rebuilt in 2007, with Horizon's additional investments (Namara Reference Namara2007). Farmers complained about not receiving payments and switched to growing potatoes (Bizimungu 2011 int.). SOPYRWA was accused of financial management leading to cash shortages in the refinery, buyers not being supplied and a failure to invest in new seed stocks, provide extension services and maintain the factory (Consultant B 2012 int.). Many foreign buyers including Valent BioSciences, Whitmeyer and SC Johnson did not receive the pyrethrum that they expected to receive, as per contracts signed with SOPYRWA. The company filed for bankruptcy and Muvunyi and his partners were accused of moving resources out of SOPYRWA into a hotel business (Bizimungu 2011 int.; Haguma 2012 int.; Uwitonze 2012 int.).
Figure 1. Comparison of Quantities of Processed Pyrethrum exported and Pyrethrum Flowers harvested. Source: MINECOFIN and Huggins (Reference Huggins2013).
Figure 2. Comparison of Annual Unit Prices of Pyrethrum Exports. Source: MINECOFIN.
Figure 3. Comparison of Annual Pyrethrum Export Revenues. Source: MINECOFIN.
Post-Horizon takeover
SOPYRWA's private owners were unable to incentivise farmers to produce pyrethrum in the region. Horizon took over the company and invested in the revitalisation of the processing plant. The government facilitated Horizon's accumulation strategies by encouraging farmers to join cooperatives. Horizon used these cooperatives to organise labour for pyrethrum production (Huggins Reference Huggins2014). Consequently, domestic production of pyrethrum has increased. SOPYRWA claims to have the best quality pyrethrum extract in the world and produces five per cent of global production. Those involved in the sector claimed that global supply currently only meets half (or even a quarter) of global demand (Bizimungu 2011 int.; Namata Reference Namata2009; USAID 2010).
At first, Horizon bought 70% of the plant. Karisimbi Business Partners – a Kigali-based consultancy that was involved in SOPYRWA's turnaround – cited the change in ownership as vital to revitalising the sector (Kalan Reference Kalan2012). Horizon invested substantially in upgrading the processing plant and worked with the government to ensure farmers in the region dedicated 40% of their land to growing pyrethrum. The rise in unit price and total revenues from pyrethrum was dramatic (Figures 2 and 3). Exports did not reach 2006 levels but have shown steady improvement (Figure 1). Horizon and the government collectively set ambitious targets. Managers travelled to other countries (e.g. Australia and Kenya) to learn from best practices around the world. In 2012, the government asked SOPYRWA to treble its revenues over the following three years. Additional land was also allocated to growing pyrethrum. In 2011, Horizon established a joint venture with UK-based Agropharm Ltd, named Agropharm Africa Ltd. This joint venture was established to refurbish the pyrethrum refining plant and collaborate with Horizon in the production of end-use pesticides. Such value-addition initiatives have proved to be a success, with domestically produced end-use pesticides used within Rwanda for the past two years.
The government and donors facilitated Horizon's investments. Horizon appealed publicly to farmers to grow pyrethrum in public meetings and in newspaper articles (Gasore Reference Gasore2013). To address the concerns of farmers, Horizon worked with USAID to organise farmers into 24 cooperatives and savings and credit cooperatives. USAID (2010) reported that farmers were being paid at least 20–40% more than in 2008. The Ministry of Agriculture and Animal Husbandry (MINAGRI) works closely with Horizon to organise labour in service of Horizon's accumulation strategies (e.g. through providing access to inputs and extension services) (MINAGRI official 2012 int.). This ‘organisation of labour’ was legitimised by donor choices to support the empowerment of cooperatives. The choice to empower cooperatives was a way for the state to speed up processes of rural differentiation and facilitate accumulation strategies.
The government uses assumptions of ‘mutual benefits’ associated with cooperatives to mask the reality that increases in pyrethrum production depend on the exploitation of those who grow the crop. Pro-cooperative ideals parallel the mainstream poverty literature that ‘projects an analytically ungrounded ideal (the stable, efficient and homogeneous peasantry) onto a misrepresented reality’ (Cramer & Pontara Reference Cramer and Pontara1998: 126). Success is cast in terms of production and increases in quality and number of cooperatives formed, feeding donor perceptions of progress. Critical scholars (Huggins Reference Huggins2014) argue that predatory elites dominate most cooperatives in the agriculture sector since they are administered from the top-down rather than from bottom-up. Huggins (Reference Huggins2014) cites one cooperative in Musanze where farmers preferred to sell their land to the cooperative at below-market rates rather than join and risk confrontation with the powerful cooperative president. He claims similar trends have developed in the pyrethrum sector and argued that farmers are coerced into growing pyrethrum (Huggins Reference Huggins2014: 376).
Though Horizon's investments were effective initially, progress has hit roadblocks recently. Between 2012 and 2014, international pyrethrum prices fell by more than 20%. In the same period, unit prices decreased by more than 32% and the volume of pyrethrum exports fell by 73·6% from 35·9 tonnes to 9·46 tonnes (BNR 2014). The farmgate price for producers (1,080 RwF) was not reduced (Nkurunziza Reference Nkurunziza2014). Poor weather conditions and difficulty in accessing the market had a detrimental effect. Pyrethrum exports more than halved between 2012 and 2013, reaching $3.98 million in 2013 and $1.83 million in 2014. Farmers began to publicly criticise Horizon for forcing them to grow pyrethrum despite low returns (although the government officially encouraged farmers to intercrop with potatoes). One farmer complained: ‘We should not be stopped from cultivating food crops on the land … if we grow potatoes, we get enough food and money to send our children to school’ (Habimana Reference Habimana2013). In the same article (Habimana Reference Habimana2013), the Managing Director of SOPYRWA and the Head of Horticulture Production at the National Agriculture Exporting Board argued that ‘farmers needed to change their mindset’ and that farmers were offered services, which would bring the value of one kilo of pyrethrum flowers to nearly 3,000 RwF. Government officials complained that farmers’ yields remained low despite the provision of free organic manure and improved seeds (Nkurunziza Reference Nkurunziza2014). Decreases in pyrethrum production showed that the government and Horizon did not force large numbers of farmers to continue growing pyrethrum (Kayonga 2015 int.).
Horizon's investments in the pyrethrum sector paint a mixed picture. Initial investments resulted in effectively organising labour to cultivating pyrethrum. Horizon also invested in improving the quality of pyrethrum extract produced domestically. However, when international prices fell, all actors in the sector (including farmers and Horizon) suffered. Though Horizon has contributed to the growth of the pyrethrum sector domestically (including through producing pesticides), it continues to face difficulties in accessing the international market. This leaves local actors susceptible to global price shifts. Though there have been recent problems, no other investor is likely to have been able to withstand the problems faced in the sector over the last two years. Positive outcomes include Horizon's investments in refurbishing the local factory and in developing a strategy of value-addition through the production of end-use pesticides.
BUT WE HAVE MINERALS TOO: RWANDA'S DOMESTIC MINERALS SECTOR
The transformation of the domestic minerals sector has followed a very different path. The domestic minerals sector has been tarred by its links with ‘conflict minerals’ emanating from the DRC. Such a narrative (e.g. Polgreen Reference Polgreen2008; Allen Reference Allen2009) has left the domestic minerals sector ignored by scholars. This section shows that the government has found it difficult to centralise rents from the export of domestic minerals. In order to capitalise on the full potential of domestic mineral assets, a choice was made to privatise state-owned concessions. Embracing market competition has forced the government to rely on legislation to discipline private actors. However, rapid privatisation led to government officials leaving for private sector employment, leaving the government with little capacity to monitor and enforce legislation or react to the demands of new private sector operators. Despite these weaknesses, the minerals sector has established itself as the country's primary export in recent years (in close competition with tourism as the top foreign exchange earner). The government has made progress in ‘cleaning’ the image of the mineral sector, embracing tagging systems and distancing the domestic minerals sector from the ‘conflict minerals’ narrative. However, the government struggles to direct actors to work in line with its strategic goals.
History
Industrial mining in Rwanda began in 1928, with the export of 68 tonnes of Cassiterite (Uwizeyimana Reference Uwizeyimana1988). The first mining code was established in 1938, which detailed the provision of concessions, as well as licences for research, exploration and extraction. Other discoveries followed: wolfram (in 1937), gold (in 1933) and coltan (in 1947). Government officials blame the colonial administration for neglecting Rwanda and instead preferring to invest in the Congo (Uwizeye 2012 int.). Unlike in Katanga, concessions were allocated competitively in Rwanda and no chartered company gained a monopoly or was incentivised to develop a long-term plan to capitalise on the full potential of mineral deposits within Rwanda (Hillman Reference Hillman1997). Very few large-scale industrial mines were established in Rwanda during the colonial era. Most investors prioritised extraction rather than exploration to ensure short-term profits. Both the colonial administration and the first two independent Rwandan governments struggled to induce investors to make full use of their concessions.
However, Habyarimana did make more progress than Kayibanda. In 1973, Habyarimana successfully consolidated the concessions of various mineral operators under a single company – Société Minière de Rwanda (SOMIRWA) (Uwizeyimana Reference Uwizeyimana1988). Belgian-owned Geomines owned 51% of the company's share capital and the government retained 49% of shares. Heavy losses hit SOMIRWA in the early 1980s. In 1982, losses amounted to 2·2 billion RwF before SOMIRWA finally went bankrupt in 1985 (Bezy Reference Bezy1990). In January 1989, Régie d'Exploitation et de Développement des Mines (REDEMI) was formed to operate the former SOMIRWA concessions. REDEMI operated in a total area of 104,000 ha with a capital of 97 million RwF (MINIRENA 2009). In 1988, the government also established a union of cooperatives, the Coopérative de Promotion de l’Industrie Minière Artisanale au Rwanda (COPIMAR). COPIMAR was able to generate profits almost immediately through buying and selling the production of its members. The cooperative even repaid its original loan (Ruhigira 2012 int.). Exports gradually decreased during the civil war in the early 1990s, which preceded the genocide.
Rwanda's Post-1994 Minerals Sector
The RPF government did not prioritise the minerals sector during the 1990s. Today, capitalising on domestic mineral resources has become central to providing revenues to reduce the trade deficit and providing wage employment. Emphasising the existence of a domestic minerals sector also diverts attention from the negative press that accompanies the ‘conflict minerals’ narrative.
State-owned REDEMI retained ownership of domestic concessions through the 1990s and early 2000s. During this time, very few new mining sites were developed. In the 1990s, COPIMAR was relatively active in organising small-scale miners although many small-scale miners operated independently of COPIMAR. Government departments suffered from a lack of expertise. Very few geologists and officials from the previous regime stayed on (Uwizeye 2012 int.). Smuggling within Rwanda was likely to have occurred with help from government officials (Imena 2012 int.). There was little emphasis on the minerals sector during this period.
Privatisation of state-owned assets began in the early 2000s. The ‘trade and export’ side of the mineral sector was initially liberalised in the late 1990s. But the Rwandan state, through REDEMI, dominated control of concessions and trade-and-export. Two international comptoirs operated alongside REDEMI and some locally owned companies. ‘Informal’ comptoirs also operated. In 2004, REDEMI exported about 60% of cassiterite from Rwanda. Some of these exports came from the REDEMI-operated cassiterite processing facility in Rutongo. COPIMAR and the Metal Processing Association (MPA), among other companies, also produced and exported cassiterite (Pourtier Reference Pourtier2004). REDEMI exported most of the wolfram in Rwanda (approximately 65% in 2004). A large share of the wolfram was produced in the REDEMI-owned concession in Nyakabingo. COPIMAR and other companies exported the remaining share of wolfram (Yager Reference Yager2004). REDEMI operated processing facilities in Gatumba for the production of Coltan.
In 2004, REDEMI operated eight concessions and received a small amount of production from independent traders who purchased minerals from small-scale miners (Garrett Reference Garrett2008). In total, REDEMI controlled 20 concessions although 12 were not in operation. By the end of 2005, only two of the concessions that had previously been under operation remained under REDEMI control. The Rwanda National Innovation and Competitiveness Report from October 2006 listed 55 private sector companies in the sector in 2005. Another report found that companies in the minerals sector comprised 23% of all companies operating in Rwanda (Temesgen et al. Reference Temesgen, Ezemenari and Munyakazi2006). Most of these companies were small comptoirs who exported small quantities of minerals after buying them from artisanal miners.
Between 2007 and 2010, a process of rapid privatisation of state assets took place. During this period, many of the 38 large-scale mining licences were issued to (almost entirely) foreign investors. These investors were from countries including South Africa, USA, Germany, Botswana, China and Russia. The majority of these investors obtained vast concessions at low prices. Most of these entrants took advantage of fast-paced privatisation and were granted four-year contracts (Uwizeye 2011 int.). The government retained shares in the two largest concessions, Gatumba and Rutongo. Joint ventures were established and two companies were created to operate these concessions – Gatumba Mining Concessions and Rutongo Mines Ltd. Between 2007 and 2011, registered investment stood at nearly USD 81·5 million. The domestic minerals sector went from one major concession holder (REDEMI) in the early 1990s to 170 registered operators and 434 active mine permits by 2013 (Perks Reference Perks2013). In 2015, there were more than 700 active permits (Imena 2015 int.).
Foreign actors also export most minerals produced in Rwanda. In 2007, the Metal Processing Association (a foreign-owned company) already exported half of the country's mineral export value. Now, two large foreign firms – Minerals Supply Africa (MSA) and Phoenix Metals – export around 75% of Rwanda's minerals (Uwizeye 2015 int.). Domestic operators complained about added costs associated with the imposition of the mineral tagging system, as a result of the Dodd-Frank Act (Rubagenga 2012 int.).Footnote 16 One such company was ROKA Global Resources, which won the RDB's Investor of the Year award in 2011.
The phase of privatisation of state-owned assets and liberalising trade-and-export has been accompanied by growth in mineral production and increases in mineral exports. Mineral exports only recovered to 1970 levels in 2004, with annual cassiterite exports only reaching the 2000-tonne mark (consistently achieved in the 1970s) in 2004. Gradually, the RPF government has begun to exceed the performances of its predecessors. Between 2004 and 2012, annual cassiterite exports were between 3000 and 5000 tonnes, excluding 2011 when cassiterite exports were nearly 7000 tonnes. Since 2006, Wolfram exports have exceeded 1970s levels (with the exception of 2008). Coltan exports were much more than at any time in the 1970s. In the late 2000s, government officials admit that a large share of these exports comprised minerals that originated from the DRC.
The government played an important role in attracting private investment to the sector. It recognised the difficulties that would be associated with maintaining centralised control of rents in a sector where high-ranking political and military contacts had already developed personal relationships with foreign buyers and regional suppliers. Additionally, the sector required large-scale investment, research and access to skills and technology, which could only be achieved through bringing in foreign investors. In the last few years, several new mining sites have opened. In January 2015, President Kagame rejoiced at the rejuvenation of the mining sector by offering a free plane ticket to anyone who doubted that minerals were being produced in Rwanda (Ngabonziza Reference Ngabonziza2015).
Figure 4. Total Annual Mineral Production (1996–2013). Source: MINECOFIN.
Figure 5. Total Annual Mineral Export Revenues (1996–2013). Source: MINECOFIN.
The DRC and Rent Centralisation
The domestic minerals sector in Rwanda has been transformed against the backdrop of violence in the DRC. The Rwandan government fought the Congo Wars initially to protect national security interests. Later, Rwandan interests were slowly transformed to what Jackson (Reference Jackson2002) called an ‘economisation of conflict’ where actors reoriented their goals in order to create profitable opportunities. However, Kigali dictated that profits obtained from commercial networks be re-invested in line with national priorities, rather than to make elites rich. In comparison to the networks of other countries, Rwandan commercial networks were disciplined with strict centralised control from Kigali (UNSC 2001; Vlassenroot & Romkema Reference Vlassenroot and Romkema2002). Others (Reyntjens Reference Reyntjens2001; Stearns Reference Stearns2013) argued that the RPF maintained a ‘Congo Desk’ or managed an informal network of Rwandan-supported companies named Le Système. Rwanda Metals and Grand Lacs Metals were examples of such companies. During this time, many military officers forged relationships, on the basis of kinship, personal and historical loyalties, with their Tutsi brethren across the border (Longman Reference Longman and Clark2002). Even the most disciplined ‘ideological’ project, as Stearns (Reference Stearns2011: 301) calls Rwanda's incursion into the Congo, had contradictions.
The ‘economisation’ of conflict’ acted as a trade-off against redeveloping the Rwandan domestic minerals sector. Given the international pressure against Rwanda, the government took the decision to clean its image. The government took a number of measures to achieve this. Rwanda was among the first countries in the region to implement the International Tin Research Institute's (ITRI) tagging system. Since then, the Rwandan government has made significant changes to how the domestic mining sector is monitored (Global Witness 2013). The government returned 82 tonnes of minerals to the DRC government in 2011 (BBC 2011). A decision was made to sever ties with rebel groups in the DRC. This came at a cost for RPF-allied businessmen such as Tribert Rujugiro, who had individual interests in the Kivus that were protected by Bosco Ntaganda's forces. Stearns (Reference Stearns2012: 59) notes that Rujugiro invested in land in Masisi, which was worth ‘millions of dollars’. Since the RPF could no longer centralise control over the export of minerals from the DRC, it made the choice to embrace market-led reforms in the hope that new investments could tap into domestic mineral resources and provide new sources of foreign exchange for Rwanda. It was hoped that these investments (though in the hands of foreign owners) could be disciplined in line with national goals.
Disciplining Foreign Investment
REDEMI had no capacity to mine and couldn't exploit all key concessions. Privatising was the right thing in terms of getting finance but we didn't get the best of investors. It was difficult because the mining sector was not known in Rwanda. (Biryabarema 2012 int.)
The Rwandan government struggled to force most private companies to work in line with the 2009 mining law, especially since many investors were granted concessions before it was established. The 2009 mining law required investors to engage in exploration before extraction. Various permits were included: a prospecting permit (two years); followed by an exploration permit (maximum of four years); exploitation of a large concession (30 years) or a small mine (five years). Most of these companies based their business plans on trading first and then investing in exploration. Others were simply there to raise the value of their company on the basis of survey reports and some were simply waiting for ‘a big discovery’ (Foreign investor 2012 int.). The government complained that many large companies, including Natural Resources Development (NRD), did not respect the mining law. NRD operated five concessions and had 110 regular employees, with 1,100 artisanal miners selling their production to the company (Garrett Reference Garrett2008). NRD was accused by the government of being more interested in trading than extracting minerals from their own concessions. Since then, two different groups have contested ownership of NRD and one group has claimed rights over the company's assets.
The government has found it difficult to discipline investors. ‘We didn't know the value of our minerals. We gave them on a first-come, first-serve basis. Then we realised we didn't want to just give away concessions without knowing the value of our mines’ (Biryabarema 2012 int.). Government officials complained they had little power to punish companies, which did not meet their demands. For example, Bayview Group was reported to have sold its concession to NRD without informing the government. The government was unable to react to this illegal sale for over a year (Uwizeye 2013 int.).
Funding gaps have plagued the sector. The minerals sector did not even receive half the money that was initially promised between 2009 and 2012. Government capacity was inhibited by reduced numbers and ageing manpower. Only four out of 40 mining technicians were under 40 years old. Forty per cent of mining technicians were over 57 years old in 2009 (MINIRENA 2009). In 2012, three senior geologists reached retirement age, making the situation even worse. Many private companies also hired the few geologists that were working for the government. In 2013, there were five geologists left in the government, three of whom were set to retire later in the year (Uwizeye 2013 int.). The government also struggled to fund prospecting and exploration in 21 Potential Target Areas (PTAs). In 2012, MINIRENA's budget only allowed it to employ consultants in four of these sites. Foreign investors were not incentivised to spend large amounts of money on research without the promise of confirmed returns (De Zahony 2013 int.).
Despite initial difficulties, the government has gradually reacted by making its law more flexible and engaging more with the private sector. ‘We are developing our legal framework to regulate private companies, which control 100% of the sector. The sector is still young and we are learning how to work together. But we will evolve side by side’ (Imena 2015 int.). The Rwanda Mining Association (RMA), which includes all private actors in the mining sector, boasted over 80 members in 2014. In order to develop Rwandan expertise within the sector, 30 students were sent to foreign countries to obtain degrees in Geology. The first batch of graduates is expected to return in 2015.
A new mining law was established in 2014 which restricted access to land at 400 ha each, instead of the large areas of land that were earlier assigned to companies. Minister of State Imena is quoted: ‘each block will be independent – we will have the right to take any block the investor is not using’ (Namata Reference Namata2014). The new mining law also departs from the previous rigid time spans that accompanied the granting of licences. The previous law only allowed the government to grant five-year licences or 30-year licences. Since July 2014, the government has provided more flexible licences depending on investors. The RPF narrative tags these reforms as ‘investor-friendly’ (Imena 2015 int.). However, these reforms can be interpreted as ways in which the government has tried to increase its own bargaining power vis-à-vis investors. In July 2013, Rwanda began operating the East African's Community's highest royalty tax on minerals – four per cent on ordinary minerals and six per cent on precious metals. Such reforms show that the government continues to find ways to ensure returns from the mining sector are retained domestically.
The example of the Gatumba Mining Concession (GMC) indicates how the Rwandan government has improvised to discipline investors who were slow or acted as roadblocks to the fulfilment of national targets. GMC was a joint venture between a private company and the government. Production was low until 2012 in the Gatumba concession, which was among the largest in Rwanda (about 22,000 hectares). Investors claimed to have invested $12 million in mining equipment and training and installing processing plants. Government officials claimed the processing plants that GMC procured did not match the types of deposits in the concession. The company complained about ‘vested interests’, even complaining that local officials incited residents to raise environmental issues. Some said that GMC was just ‘unlucky geologically’ (Schutte 2013 int.). In 2014, mining sites under GMC's control were closed down. The government carved up the concession into smaller units and offered these smaller units for sale to other investors. When the government found that GMC neither had the funds nor the willingness to extract minerals and use the concession, GMC's board of directors (led by government representatives) closed operations (Mugisha Reference Mugisha2014a, Reference Mugisha2014b).
Beneficiation
The RPF prioritised beneficiation in the sector. In 2006, consultancy reports circulated with the finding: If Rwanda could smelt all exported cassiterite today, it would increase export revenue by USD 6 million (33% increase) and for every additional 10% of cassiterite transformed, revenue would rise by USD 1·3 million.
In the late 2000s, plans were put in place to establish a Kigali Mineral Campus (KMC). KMC would be established in order to provide opportunities of diversification into higher value-addition activities. This would make Rwanda a hub for the export of minerals. The Habyarimana government had built a tin smelter in the 1980s. Beneficiation reforms begun with a focus on making the existing tin smelter operational again.
In 2001, the government sold the smelter to European-based NMC Metallurgie (who later changed their name to Phoenix Metals). The company invested in the maintenance and repair of the factory. However, despite Phoenix's efforts and investments, the smelter was not operational in 2013. Raphael Ritter de Zahony, the Managing Director of Phoenix, attributed the lack of activity to low, unreliable quantities of raw materials, the high cost and unreliable supply of electricity and the difficulty in finding buyers for Rwandan minerals. Zahony complained that institutional changes within the sector and government indecision regarding subsidising electricity have slowed down the pace of reforms. Former Minister of Infrastructure Vincent Karega was close to providing incentives to make the smelter operational until he was transferred to South Africa as High Commissioner (De Zahony 2013 int.). In 2013, government officials were in discussions with Phoenix Metals to guarantee a cheaper price of electricity to Phoenix metals (which had been discussed since 2006). At the time, Phoenix's representatives said that even if subsidies were granted, beneficiation would still be difficult. Phoenix would have to pass a difficult Conflict-Free Smelter audit. Malaysia Smelting Company (MSC) was passing through this audit for the third time in 2013.
Progress was much faster than expected. By 2015, Phoenix was undergoing the audit (after MSC had eventually passed it). Phoenix expected to begin exporting tin in January 2015. Phoenix will still have to be audited for six months, develop a guaranteed supply of cassiterite for the smelter and retain access to electricity at a subsidised rate. However, impressive strides have been made. If Phoenix successfully passes the audit, Rwanda will have the only conflict-free tin smelter in Africa. Such developments would affect the global tin supply chain, with the Rwandan government hoping to incentivise other African tin producers to smelt their tin in Rwanda.
The strategies used by the Rwandan government in the minerals sector are very different to those used in the pyrethrum sector. In the early 2000s, the domestic minerals sector was largely inactive. While rents emanating from the DRC may have been centralised, the domestic minerals sector was characterised by smuggling and theft. Politically, refocusing investments on the domestic minerals sector has several benefits. It has reduced the possibility of rival elites accessing rents from the DRC or through informal trade networks within Rwanda (although withdrawing such rents was a source of friction). The minerals sector is now among the top sources of foreign exchange for Rwanda. Employment also increased. 58,000 miners were employed in the sector in 2011, compared with 22,000 in 2006. However, there were costs. The government chose to privatise assets and place control of the sector in foreign hands. The government continues to face challenges in developing capabilities to enforce legislation and monitor and discipline foreign investors. However, government officials are reacting to the needs of actors across the value-chain, investing in improving their own capability and attempting to regulate the sector effectively.
CONCLUSION
This paper has argued against the current debate regarding the use of investment groups in Rwanda. Economic development is being undertaken in an environment of vulnerability where ruling elites oscillate between ‘inclusionary’ and ‘exclusionary’ arrangements with other actors. This paper shows that choices to use investment groups or embrace market forces depend on politics pertaining to particular sectors. In the pyrethrum sector, Horizon Group was able to effectively organise labour to cultivate pyrethrum (with more success than private investors). Horizon also invested in producing better-quality pyrethrum and engaged in value-addition. In the mining sector, investment groups were not used because (i) the Rwandan government aimed to ‘clean’ its image of involvement in ‘conflict minerals’; (ii) there was a danger that individual elites would use their own networks and work against the aim of centralising rents; (iii) large-scale foreign investment (and expertise) was required in order to capitalise on potential mineral revenues.
However, neither strategy pursued in these sectors has been associated with entirely positive outcomes. Horizon's investments have been associated with the displacement of segments of the population (Huggins Reference Huggins2014) and a reduction in international prices has limited the continued expansion of pyrethrum cultivation. The fast-paced privatisation that accompanied the growth of the minerals sectors occurred while the government struggled to retain expertise and capacity to monitor and discipline foreign investors in the sector.
Importantly, this paper also shows how the dominant coalition in Rwanda has understood the importance of centralising rents in particular sectors while preferring to embrace wholesale privatisation in others. The Rwandan government adapts its strategy to the needs of specific sectors and political requirements, rather than staying disciplined in line with a long-term plan.
It is never one or the other for us. We privatised early on but then it became difficult. Sometimes, foreign investors don't listen. Other times they work with us. It is not easy to do any diversification. There are many barriers. We are always learning and finding new ways to achieve our goals. (Rwamucyo 2014 int.)
Ruling elites are reluctant to empower individuals or facilitate the investments of domestic private sector actors (who could later become rivals). Decisions are taken in an environment of uncertainty and threat among elites in Rwanda, rather than through a generalised consensus. Market-led reforms are undertaken with an understanding that it may empower foreign actors and reduce domestic control in these sectors. However, the government remains confident of finding ways to discipline foreign investors in line with national goals. None of these strategies are perfect. Progress is blocked by fluctuations in international prices, donor demands, domestic constraints and political pressures. However, the RPF has navigated this environment and retained a commitment to upgrading economic activities while adapting to the vulnerability it has experienced.
