Jump to ContentJump to Main Navigation
Deals and DevelopmentThe Political Dynamics of Growth Episodes$

Lant Pritchett, Kunal Sen, and Eric Werker

Print publication date: 2017

Print ISBN-13: 9780198801641

Published to Oxford Scholarship Online: December 2017

DOI: 10.1093/oso/9780198801641.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (oxford.universitypressscholarship.com). (c) Copyright Oxford University Press, 2022. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.date: 16 May 2022

The Disorder of ‘Miracle Growth’ in Rwanda

The Disorder of ‘Miracle Growth’ in Rwanda

Understanding the Limitations of Transitions to Open Ordered Development

(p.217) 8 The Disorder of ‘Miracle Growth’ in Rwanda
Deals and Development

Pritish Behuria

Tom Goodfellow

Oxford University Press

Abstract and Keywords

This chapter analyses how Rwanda has achieved near miracle growth rates of above 6 per cent (excluding 2003 and 2013) since 1994. This is due to the country being led by a strong dominant party which has resulted in a stable deals environment within the country. The pursuit of growth has led the Rwanda Patriotic Front (RPF) to drive for a more open deals space, yet retaining some closed deals space for strategic interests. However, growth maintenance in the country remains dependent on commodity price fluctuations, access to foreign aid, and the maintenance of a stable political settlement. This leaves Rwanda’s growth episodes vulnerable to external shocks and negative feedback loops.

Keywords:   economic growth, political settlements, deals environment, Rwanda, rent space

8.1 Introduction

In the two decades since 1994, the Rwanda Patriotic Front (RPF) government has achieved growth rates of over 6 per cent every year (with the exception of 2003 and 2013). This has led to praise from diverse groups, ranging from international financial institutions (Tumwebaze, 2014; Lagarde, 2015) to mainstream (Collier, 2015) and heterodox scholars (Kelsall, 2013; Booth et al., 2014).1 Conventional perspectives on the drivers of economic growth in Rwanda vary, but there are three identifiable narratives, all of which tend to oversimplify the drivers of growth by placing disproportionate emphasis on one particular feature of Rwanda’s development trajectory.

The first account focuses on Rwanda’s embrace of market reforms, supported by foreign aid, as key to its success. A typical statement in this mould would be that: ‘The economic growth in Rwanda has been primarily driven by liberalization in the agricultural sector—mainly coffee and tea, the country’s main exports’ (Oro and Arias, 2012). This chapter demonstrates that liberalization in these sectors has indeed been significant. However, it has not always had a positive effect on economic growth and has been only one aspect of the growth story in Rwanda since 1994.

(p.218) A second narrative focuses on the centrality of ‘second-generation reforms’—i.e. issues such as improved education indicators, social protection, and women’s empowerment—as key to the growth success (Lagarde, 2015). Whatever their significance for human development, such explanations underplay important political economic factors underpinning economic development in key sectors.

A third narrative takes a heterodox perspective, attributing the country’s success largely to the use of party- and military-owned enterprises to intervene in strategic and long-term ways in the economy (Booth and Golooba-Mutebi, 2012). Our analysis in this chapter concurs that these strategic state interventions have been important, but challenges the linear simplicity of the focus on ‘long-term centralized rent creation’, and seeks to deepen the analysis of Rwanda’s economic strategy and state–business relations. Unlike existing analyses, we show how important variations in state–business relationships in specific sectors have influenced Rwanda’s developmental trajectory. Even within the same economic sectors, the RPF government has attempted to retain some control over rents by using party- and military-owned enterprises in some sub-sectors, while relying heavily on open deals in others. We argue that the government has employed such strategies to spur overall economic growth while retaining some state control (to ensure a continued emphasis on economic transformation) and also attempting to maintain a stable political settlement.

To explore this in depth, this chapter examines four specific sectors of the economy: coffee, mining, construction, and financial services. Tensions between facilitating rent creation and pursuing an agenda of liberalization play out in different ways in different sectors, with important implications. The strain of attempting to pursue ‘open ordered’ deals while also maintaining a political settlement has started to show in Rwanda, posing potential threats to future prospects. After describing the growth and structural transformation experience in Rwanda, we explore in detail the rents space and deals environment in the four sectors, before reflecting on feedback loops and how the Rwandan case speaks to the theoretical framework set out in this volume.

8.2 Overview of Growth and Structural Transformation Experience

Rwanda became independent in 1961. In the first few years after independence, growth fluctuated, but between 1965 and 1969 annual growth was high (above 7 per cent). Between 1969 and 1975, growth rates remained relatively low, before picking up again until 1981. Rwanda experienced a large deceleration in growth between 1981 and 1994. However, since 1994, Rwanda has experienced its longest ever acceleration episode (Kar et al., 2013). Unlike the previous growth episodes, Rwanda’s post-1994 episode has been strong enough and long enough to constitute a clear case of ‘miracle growth’. Indeed, (p.219) Rwanda was among the top ten fastest growing economies globally in the 2000s (ACET, 2014). It has also experienced some structural transformation. Figure 8.1 shows that agriculture as a proportion of GDP has gradually decreased, falling from 49.7 per cent in 1994 to 33 per cent in 2014. Meanwhile, the services sector contributed 29 per cent of Rwanda’s GDP in 1994, compared with 53 per cent in 2014. Figure 8.2 provides a breakdown of GDP composition by activity between 1999 and 2013. In terms of employment, between 2002 and 2011, the percentage of the population employed in agriculture decreased from 87 per cent to 73 per cent, with corresponding increases for the proportion of the population employed in services (from 10 per cent in 2002 to 20 per cent in 2011), and in manufacturing and extractive industries (from 3 per cent in 2002 to 6 per cent in 2011) (NISR, 2012).

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.1. Components of GDP in Rwanda by sector, 1965–2014 (% of GDP)

Source: World Bank.

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.2. GDP by kind of activity

Source: MINECOFIN (2013).

Structural transformation remains limited, however. Rwanda is one of the least transformed countries in Africa, ranking eighteen out of twenty-one countries on the African Centre for Economic Transformation (ACET) Index—though this is an improvement from the situation in 2000, when it was ranked last (ACET, 2014: 197). In 2014, Rwanda was ranked eighty-seventh out of 144 countries on the Observatory for Economic Complexity’s rankings. It remains a largely rural society, with 70–80 per cent of its population working in the agriculture sector. Coffee, tea, and minerals have accounted for over 90 per cent of Rwanda’s exports for most of its history. This has gradually changed during the RPF’s rule, and Rwanda has shown marked improvement relative to many other countries in the ACET’s ratings on diversification and technological upgrading, despite a lack of progress on measures of export competitiveness and human well-being (ACET, 2014: 33). Figure 8.3 illustrates Rwanda’s export space. Some re-exports (petroleum) have recently become increasingly prominent, as have new exports (within the agriculture sector), including wheat and rice. The government has managed to diversify its export base to some degree. However, exports remain concentrated (p.220) in the agriculture and minerals sectors and there has been no success in developing manufacturing exports.

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.3. Composition of Rwanda’s exports, 2014

Source: Atlas of Economic Complexity <http://atlas.cid.harvard.edu>.

Lack of attention to manufacturing growth has been a striking feature of development under the RPF so far, barely featuring in the government’s Vision 2020 strategy (GoR, 2000). In the Economic Development and Poverty Reduction Strategy 2, the government slightly increased its emphasis on manufacturing, stating an aim of moving ‘from an agriculture-based economy to an industry and services-based economy’ (MINECOFIN, 2013: 55). Though agriculture grew at a relatively healthy rate between 2000 and 2013, the sector’s annual growth rate has remained below the annual GDP growth of Rwanda for most years, while growth in the services sector has been the steadiest (Figure 8.4).

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.4. Comparison between agriculture, industry, and services sector with annual GDP growth rates for Rwanda, 2000–13

Source: MINECOFIN (2013).

Figure 8.5 demonstrates that construction has been a growing component and the most consistent source of growth in the industry sector. The growth that is implied in industrial sector statistics can also be misleading, given that mining is included within them. As is evident from Figure 8.5, manufacturing (p.221) (p.222) has shrunk significantly, while construction and mining have grown. The manufacturing sector is still quite young, given that most companies were destroyed during the genocide. Despite some investment and technology acquisition, several factors inhibit further growth in the sector, including small market size, difficulties in creating supply chains and distribution networks, very high transport costs, inconsistent access to electricity, and problems establishing effective management and production systems.2

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.5. Industry as a share of GDP for Rwanda, 1999–2013

Source: MINECOFIN (2013).

Growth in services was prioritized after 2000, with the aim of building a knowledge-based economy. Within this sector, finance, real estate, hotels and restaurants, and trading and transport have shown promising growth (Figure 8.6). The government has also launched a Meetings, Incentives, Conferences, and Events (MICE) strategy to augment revenues from the services sector. (p.223) The growth of the construction sector can also be understood in line with such goals. Rwanda Development Board officials claim that revenues from MICE could reach US$150 million by 2015.

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.6. Services as share of GDP for Rwanda, 1999–2013

Source: MINECOFIN (2013).

8.3 Historical Evolution of Political Settlements in Rwanda

In Rwanda, the dynamic and violent nature of political settlements has been on display since the country became independent in 1962. Before independence, Gregoire Kayibanda—who would later become Rwanda’s first president—led his party (MDR-Parmehutu) in organizing the 1959–61 revolution. Kayibanda was a strong advocate of the Hutu Power political movement, and the revolution was mobilized on the basis of ethnic divisions, with the Tutsi monarchy and party (Union Nationale Rwandaise) as the targets.

Kayibanda’s anti-Tutsi pogroms continued, with the inyenzi threat (of returning Tutsis), salient for the early 1960s. However, elite politics during the First Republic were largely contested on a regional basis. Kayibanda relied on Hutus from the southern and central regions for support, while his main threat came from the northern areas of Ruhengeri and Gisenyi. Thus, it could be said that Kayibanda’s regime was a ‘vulnerable authoritarian coalition’ from the start, with significant threats to his power from excluded factions within the majority Hutu ethnic group. Kayibanda continued to control TRAFIPRO (Travail, Fidélité, Progrès)—the coffee marketing board—and (p.224) headquartered it in the south. TRAFIPRO acted as ‘the economic arm of the regime’ (Verwimp, 2003: 163). Its control over rents led to determined opposition from the north, as southern elites attempted to extend their control across the country. By the early 1960s, Kayibanda’s regime moved towards being a ‘competitive clientelist’ regime. Initially, during the 1960s, there was an opening up of deals space and some new private actors invested in the economy. However, towards the latter half of the 1960s and early 1970s, Kayibanda used state institutions and cooperatives such as TRAFIPRO to monopolize areas of the economy. The deals space gradually shifted from opened to closed, with less private investment during this period.

The army chief of staff, Juvénal Habyarimana, led a group of northern elites within the ruling party to mount a coup in 1973 (Prunier, 1995), the main motivation being to take power away from southern Hutus (Lemarchand, 1995). Habyarimana’s Second Republic immediately distanced itself from the racially charged rhetoric of the preceding government. He publicly declared the day of the coup as ‘a day of peace and reconciliation’ (Verwimp, 2013). However, throughout his reign there were almost no Tutsi bourgmestres or préfets (Prunier, 1995). The Second Republic enjoyed significant periods of economic growth during the 1970s, as described earlier. During this time, the regime could be classified as a ‘strong dominant party’. Until 1981, Habyarimana was largely reliant on elites from his northern support base and was able to manage rivalries between different northern elite groups (Prunier, 1995). Gradually, growth began declining, beginning in 1981. This reflected the beginnings of a commodity crisis, with sharp decreases in global prices of coffee, tea, and tin weakening the Rwandan economy.

Following an attempted coup in 1980, the regime shifted to becoming a ‘vulnerable authoritarian coalition’ and later, a ‘competitive clientelist’ regime (particularly once democratic space was opened). Habyarimana became increasingly reliant on a small group commonly referred to as the akazu (‘little house’). The akazu also controlled most positions in the coffee-exporting agency, OCIR-Café, and the national tea regulatory authority, OCIR-Thé, thus dominating control over rents (Verwimp, 2003). As with Kayibanda’s reign, Habyarimana did not open the deals environment. Most of the economy remained ‘closed’ during this period, with the government controlling monopolies in most export sectors (although a limited number of deals were open).

The current RPF government, which took power in 1994 after the devastation of the genocide, has managed severe political pressures, given the fractured society it inherited and the fact that it was the first Rwandan government to be dominated by the minority Tutsi ethnic group. The political settlement that has emerged can be categorized as a ‘strong dominant party’, in the sense that factions excluded from the ruling coalition are relatively weak, as are lower-level factions within the ruling coalition itself. Some have (p.225) explicitly classified the RPF government in this way (Lavers and Hickey, 2015), while others have not used the same language but view the government similarly (Booth and Golooba-Mutebi, 2014; Kelsall, 2013; Goodfellow, 2014).

8.4 Studying the Rents Space

The preceding narrative describes how the rents space was dominated by rentiers in the coffee and tea sectors during the Kayibanda and Habyarimana regimes (and powerbrokers in much of the rest of the economy, with a gradual closing up of deals space towards the end of both regimes). This chapter demonstrates that the same is not true today. Figure 8.7 illustrates an attempt at describing the composition of GDP in Rwanda in 2013 in terms of the rents space categories. What this ‘radar’ chart shows is that overall, despite some success in diversifying exports, GDP is composed largely of activities targeted at the domestic market in the form of workhorses and powerbrokers. Why this is the case and how dynamics operate within our chosen sectors in these categories will be described in Section 8.6.

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.7. Composition of GDP in the rents space in Rwanda

To undertake the research on the contemporary deals environment, four sectors were selected to cover each of the four key categories in the rents space (Figure 8.8). Given the dynamic nature of reforms in Rwanda, however, the sectors are not a perfect fit in the rents space categories, and choosing sectors to fit these boxes was challenging. There are very few competitive, export-oriented sectors in Rwanda. Since the coffee sector has been liberalized and (p.226) private actors own assets, the coffee sector was chosen as a magician. The mining sector was chosen as a rentier. The construction sector was chosen as a powerbroker, as it is mostly geared towards the domestic market and party- and military-owned investment groups play a significant role. The financial services sector was chosen as a workhorse. The sector has been liberalized, even though the government still has a majority shareholding in the largest bank (Bank of Kigali).

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.8. Chosen sectors aligned to the rents space in Rwanda

Fresh primary research was conducted for this chapter by both authors in January 2015, building on their existing work in Rwanda. The project focused on interviewing representatives from firms working in the four sectors, as well as government, party, and military officials, where relevant. The two researchers conducted seventy-nine interviews. Some interviews conducted as part of previous research are also used in this chapter. Attempts have been made to triangulate and cross-check data that were presented by respondents. Most private-sector respondents were fairly open in discussions; some respondents had been interviewed previously and the researchers had already established a rapport with these respondents. However, it is possible that many respondents downplayed the degree to which personal connections and corruption might have operated in Rwanda. Judgements regarding deals spaces in different sectors have been made on the basis of interviews with respondents working in the sector. We recognize that such interpretations may not represent the full picture, given the limited database of interviews on which to make generalizations.

8.5 A Characterization of the Deals Space

This section begins with an overview, before turning to four sectors in detail. In all these sectors, the RPF government has faced pressure to embrace market-led reforms, while retaining a preference that domestic firms are able to engage in technology acquisition. In the late 1990s, the IMF pressured the Rwandan government to privatize state-owned enterprises. By 2011, (p.227) fifty-seven state-owned enterprises had been fully privatized, with a further twenty going through different phases of privatization, though a 2011 consultancy report found only around half of these to be operational. Acknowledging these mixed results, there is recognition among government officials that state regulation and interventions of different forms are required. However, in most sectors, they have been reluctant to pick domestic winners (other than by using investment groups, which we discuss later in this section). Instead, government figures highlight that the state should play a key role in sectors early on and then get out as soon as possible: ‘Telecoms represents a good case of how we made investments first and then liberalized. This prepared economic growth, and openness led to innovation in the sector.’3 They also emphasize that initial government investments are meant to spur private investment that would not otherwise have been forthcoming: ‘We invested where no one would invest, like in the tourism sector. It is this model of going in where others will not that has spurred growth in the Rwandan economy.’4

The choice of which strategy to embrace in specific sectors depended on a number of factors, including the degree of donor pressure, the nature of domestic competition, and the degree of technology acquisition required. Where no private investor came in, the government invested. Often, investment groups (or party- and military-owned holding companies) were used. These groups have been detailed in the existing literature (Booth and Golooba-Mutebi, 2012; Gokgur, 2012; Behuria, 2015).

Representatives of investment groups likewise claim to follow a strategy of investing in strategic sectors, increasing productivity in those sectors, breaking even, and then eventually leaving. Representatives of Crystal Ventures Ltd (CVL, formerly Tri-Star) attribute the success of many of their firms not just to government backing, but to the philosophy they have ‘inherited’ from the RPF and their commitment to rebuilding the country. There is also a sense that they have to perform well because everyone is watching.5 CVL representatives emphasize their commitment to ‘crowding in’ rather than ‘crowding out’ domestic private firms and cite some state-owned companies under their umbrella that were allowed to fail.6

Investment groups are the largest domestic players in the economy. However, the private sector has also developed an organized platform (with government support and even ‘direction’), the Private Sector Federation, which groups together ten professional and promotional chambers. It is now compulsory that for all government policies consultation with relevant private-sector (p.228) representatives has taken place. In January 2015, one MINECOFIN official said that parliamentarians sent him back to his office when he presented work on the mining sector that had not been discussed with private-sector representatives.7 However, despite these efforts to support the private sector, most SMEs struggle and few stay afloat for very long.

This overview is important, because liberalization has, as we show, often been associated with a move towards more open (in the sense of impersonal) deals. However, in these sectors, closed deals are still used for strategic investments and to ensure rents remain under centralized control. The following four subsections briefly describe the trajectory of state–business relationships in four sectors, with explicit reference to the ‘deals space’ framework. In each case, a brief overview of the sector’s history, its current firm structure, and the general evolution of the deals environment will be outlined.8

8.5.1 Coffee (Magician)

Before 1994, the coffee sector remained in a relatively stagnant ‘closed ordered’ deals space, with rents centralized under both the Kayibanda and Habyarimana regimes. New exporting companies were sometimes licensed (although they never gained a substantial foothold).9 However, there is no indication that there were any open deals operating in the sector.

Coffee production in Rwanda has never amounted to even 1 per cent of global production. However, it was Rwanda’s leading export before 1994 and it is still counted among Rwanda’s top exports today. The volume produced in Rwanda has remained relatively stagnant. In 2015, 20,000 tonnes were produced. In 1995, production was actually slightly higher. In the early 1980s, production was almost double its current amount. Though production has not increased, the value of coffee exports has increased significantly. This can be attributed to an increase in global coffee prices. However, it is also because of the government’s prioritization of the production of higher-value-added coffee. Over time, increasing numbers of new investors have entered as exporters and owners of washing stations. Such actors are competing for a relatively unchanged amount of coffee produced in the country. Though an increase in openness has not increased the overall growth of the sector (in terms of production), it has contributed to increased technological change in the sector, which has fuelled growth in terms of the value of coffee (p.229) produced. Figure 8.9 illustrates how annual GDP growth rates compare with annual export crops growth rates (coffee and tea) between 2000 and 2013.

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.9. Comparison of annual GDP growth and annual export crops growth for Rwanda, 2000–13

Source: MINECOFIN (2013).

Immediately after the RPF assumed power, trade and export operations were liberalized. Rwandex—a majority government-owned coffee export agency, which exported most of Rwanda’s coffee during Habyarimana’s reign—continued to be a prominent exporter, but RPF supporters and prominent businesspeople who had supported the RPF during the war also invested. At this time, government officials claimed that any investors were welcome.10 Though many individuals who set up companies were closely linked to the RPF, there is no evidence for the claim that the benefits they received would not have been available to other investors. In the 1990s, six new companies entered the sector, two of which went bankrupt after two years (MINAGRI, 2008). Later, other smaller companies that entered in 1994 also went bankrupt (IMF, 2000).

Swiss-based RwaCof, owned by Switzerland-based Sucafina, entered Rwanda in 1995, quickly developed their value chains, and exported large shares of Rwandan coffee (partly because of their foreign contacts).11 Sucafina had already established a presence in the region through its Uganda-based company, UgaCof. Entering Rwanda was ‘a no-brainer’ for Sucafina. There was an opportunity, since ‘many of the other players here did not have much knowledge of the sector and it was not managed’. RwaCof representatives said that competitors ‘learned, but many who started or misbehaved, did not survive’.12

RwaCof then gradually competed with Rwandex to become the largest exporting company in Rwanda. Between 2000 and 2002, RwaCof and Rwandex (p.230) collectively held about 65–75 per cent of the domestic market. Agrocoffee (owned by a loyal businessman) and SICAF (owned by the husband of the sister of the president’s wife) collectively captured more than 20 per cent of the market. Since 2004, competition in the sector intensified. Several new exporters entered the sector. During this time, the government built an institutional environment to fund the coffee sector and donors had also begun supporting the strategy. In 2009, Rwandex’s assets were bought by Scott Ford’s Rwanda Trading Company (RTC). The RTC used new innovations in the fully washed coffee chain to capture more of the market. In 2015, sixty-three coffee exporters operated in the country (nearly double the number that existed in 2012).13

All respondents in the coffee sector agreed that the government’s role in the sector was restricted to regulation and ensuring that it remained productive.14 Firm representatives also admitted to offering high prices to farmers to capture market share. However, respondents (government officials and private-sector representatives) argued that the liberalized environment in the sector was a challenge. This was primarily because fluctuating international prices forced exporters and washing station owners to speculate, which led to coffee being bought at higher prices and sold at lower prices. Significant challenges faced by most companies related to the lack of skills and the availability of working capital, facilities (such as warehouse space), and logistics.15

Today, RTC, RwaCof, and Coffee Business Company dominate trade and export operations. The Nigerian company, Kaizen, entered the sector in 2012 and in five months acquired eight washing stations. The firm’s owners initially beat out competition by paying farmers for coffee cherries as soon as they were delivered to washing stations. Kaizen’s entrance was marked by increased competition in the sector: ‘The days of big traders sitting and waiting for people to bring coffee are going. To survive in the coffee sector, you have to go closer to the source and make your contacts outside.’16

The nature of the deals space in the above description may be characterized as open ordered, given the entry of many new firms, and perceptions that government decision-making is relatively predictable. This openness, however, has tended to favour larger operators with access to expertise, international networks, and capital. To facilitate the demands of coffee companies, government officials said they established a Memorandum of Understanding with only the ten largest coffee-exporting companies.17 Local coffee companies and cooperatives continue to occupy a prominent role in the sector, however. One, Rwashoscco, received significant support (p.231) from USAID and has continued to flourish as a result of the contacts it developed abroad and the expertise gained by local Rwandans working at the cooperative. Rwashoscco’s exports multiplied nearly three to four times between 2011 and 2015, with the success attributed to high-quality and efficient management of the supply chain.18

(p.232) The growth experienced in the coffee sector in the mid to late 2000s led to a resurgence of interest in the sector domestically. The coffee sector turned around in 2004, which was the first year that coffee production volumes matched yearly government targets (Figure 8.10). As a result, ‘People started thinking there was money in coffee. However, when they tried, everyone realized it is very difficult because of the competition and small market.’19 Though foreign investors and donor assistance have contributed to ‘transfers of technology’ and expertise, some local companies warned that these ‘transfers’ were not being sustained, as larger players had begun to dominate the market. They argued that heightened competition ‘is not good for everyone’,20 and that there was very little protection for local companies: ‘It is like a family where you have five children. Two may get PhDs, the other three will not make it to high school.’21

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.10. Value of Rwanda’s coffee exports (annual variation, %)

Source: MINECOFIN (2013).

Liberalization has been accompanied by the government’s choice to prioritize adding value and shifting from the production of semi-washed coffee to fully washed coffee. The number of coffee washing stations in the country increased from two in 2000 to 229 in 2014. Though a large number of washing stations established rapidly, it was on a first-come, first-served basis. For most of the 2000s, washing station owners were making losses, except for those who received support from donors. Though government officials may have facilitated investments, banks and other actors could not support investors. This is borne out by the fact that, of the 229 washing stations constructed in Rwanda, thirty were not in operation and most washing stations operated at 50 per cent capacity (Macchiavello and Morjaria, 2015). Government officials stress that opportunities are open to new investors in this area, but that all operate under strict guidelines with regard to where washing stations are constructed, showing that, despite the problems with washing stations, the government has made substantial efforts to create an ‘open ordered’ deals environment in this sub-sector.

In contrast, the exports of packaged single-origin Rwandan coffee have been developed through relatively closed but ordered deals. Exporters of packaged coffee included USAID-sponsored Rwashoscco, Kaizen, and coffee brands developed by domestic coffee companies, owned by domestic elites. The government (National Agriculture Export Board and Development Bank of Rwanda (BRD)) has worked with partners, the Clinton Hunter Development Initiative and the Hunter Foundation, to create a coffee company—the Rwandan Farmers’ Coffee Company (RFCC)—and invest in a US$3 million coffee processing factory in Kigali. In 2015, the RFCC began operations and will (p.233) produce under the brand ‘Gorilla’s Coffee’, and sell to local, African, Asian, and European markets. The government approached international roasters—Starbucks, Costco, and Rogers Family—to partner specific, local exporting companies and cooperatives.22 In this deals space for strategic value-addition investments, the government has relied on relatively closed ordered deals, since only certain local firms (with close relationships with the government) are entrusted with risky value-addition attempts. Figure 8.11 details the evolution of the deals space in the coffee sector.

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.11. Deals space in Rwanda’s coffee sector

8.5.2 Mining (Rentier)

Figure 8.12 illustrates the turnaround in mineral exports in Rwanda in recent years. Historically, the domestic minerals sector has remained underdeveloped and closely linked with the ‘conflict minerals’ narrative in the Democratic Republic of the Congo (DRC). During the colonial era, concessions were allocated on a first-come, first-served basis. Pre-1994 governments were unable to incentivize companies to utilize their concessions fully.23 Figure 8.13 compares annual growth rates in the mineral sector with annual GDP growth rates in Rwanda.

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.12. Total mineral production (million tonnes) for Rwanda, 1996–2013

Source: MINECOFIN (2013).

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.13. Comparison of annual minerals sector growth rates with national growth rates for Rwanda, 2000–13

Source: MINECOFIN (2013).

Immediately after 1994, the RPF government showed little interest in rejuvenating the mining sector. Government-owned Régie d’Exploitation et de Développement des Mines (REDEMI) controlled all concessions during the 1990s. However, government officials claim that during most of the 1990s, issues of smuggling and theft limited growth in the domestic minerals sector. Some government officials argued that RPF officials and lower-ranked (p.234) government officials were the cause of the problem.24 This showed that there was a ‘closed disordered’ deals environment in the mining sector at the time:

Until the genocide, the sector was actually working well. After the genocide, most assets were stolen. Some powerful people also went into concessions and organized stealing minerals. For the government, it was difficult. They traced maybe one or two mining engineers in the country … But it was not only because of a lack of skills, it was because of all the procedures that were there in the ministry.25

As with the coffee sector, it is important to break down the mining sector into specific sub-sectors or nodes, each of which has a distinctive ‘deals environment’ under the RPF. These nodes are as follows: links with mineral networks in the DRC; ownership of concessions in Rwanda; the trade and export of minerals from Rwanda; and the organization of artisanal and small-scale mining (ASM). The Rwandan minerals sector has been directly linked with the minerals sector of the Kivus in Eastern DRC for most of its history. Trade networks intensified during Rwandan involvement in the Congo Wars and the RPF’s later alleged support of rebel groups. However, even critics (UNSC, 2011) agree that Kigali retained centralized control over mineral networks in the DRC. Though this may have led to individuals gaining access to rents, discipline was strictly administered from Kigali. Thus, rents from minerals networks in the DRC were managed in a ‘closed ordered deals’ environment, though this centralized control was sometimes threatened.

(p.235) Liberalizing trade and export operations picked up pace towards the end of the 2000s. In 2004, one official claimed that REDEMI exported 60 per cent of Rwanda’s minerals. The sale of REDEMI’s concessions took place rapidly, however: initially REDEMI controlled twenty concessions, but by the end of 2005 only two remained under its control. Many mining companies were registered in the mid-2000s. The Rwanda National Innovation and Competitiveness Report listed fifty-five private companies in the sector in 2005. Most were small comptoirs, who exported small quantities of minerals after buying them from artisanal miners. These companies benefited from the rapid privatization that was prioritized ahead of the establishment of a mining law in 2009. Yet government officials admitted that initially they had very little control over the activities of individuals who controlled REDEMI or who managed concessions. In two nodes (ownership of assets, and trade and export operations), the deals environment thus shifted from a ‘closed disordered’ environment to an ‘open disordered’ environment, in which the sale of concessions was subject to competitive bidding, though a degree of disorder remained. Trade and export operations operated in a similar way. Disorder was a characteristic of the rapid liberalization in this sector, with many companies operating without licences for several years.26

By 2010, thirty-eight large-scale mining licences were granted to (almost entirely) foreign investors: ‘Most mining companies said they would invest, but nothing ever came. Because of the way we did privatization, it was difficult to get these companies out after they didn’t do what was promised.’27 Most investors obtained vast concessions at low prices. 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.

The rapid shift in ‘closed’ to ‘open’ deals was not matched by investments in government expertise, which helps to explain the continued disorder. Gradually, the government adapted the mining law in line with achieving government targets. Instances where foreign companies were disciplined are evidence of this, and the government has recently become stricter.28 Thus, it could be said that there is now a gradual shift to move the deals environment from ‘open disordered’ to ‘open ordered’. However, the openness of deals exposed the ‘capability traps’ to which the government was vulnerable,29 and some private-sector operators claimed that the government had not (p.236) delivered on its promises. The government’s intent to move towards greater order in the deals environment here has thus not yet been fully realized.

The nature of deals in the sector was also impacted by the Rwandan government’s decision to embrace tagging initiatives. The work of advocacy groups, who propagated the ‘conflict minerals’ narrative, eventually contributed to the inclusion of Section 1502 in the Dodd–Frank Wall Street Reform and Consumer Protection Act of July 2010. Section 1502 directed the US Securities and Exchange Commission (SEC) to promulgate new disclosure rules for SEC-reporting companies that use ‘conflict minerals’ originating in the DRC or adjoining countries. The Rwandan government was one of the first national governments to adopt tagging initiatives. Some foreign companies then chose to invest in the Rwandan minerals sector, which was previously ignored:

Earlier, everyone was doing mining from the Congo. People who had concessions here were not using them. Then there was also the minerals ban in the DRC from September 2010 to March 2011, with no export from the DRC. It pushed companies to produce minerals in Rwanda.30

Changes in the liberalizing of trade and export operations and the sale of REDEMI’s assets are also echoed in changes in the organization of ASM, which was the predominant form of mining in Rwanda for most of the 1990s and 2000s. However, since 2009, there has been a push to formalize ASM operations. By 2013, there were 434 active permits, and by January 2015 more than 700.31 In the ASM node, government officials affirm that there has been an attempt to bring greater order to the deals environment,32 though there is some way to go before the government’s enforcement capabilities are able to match up to needs.

Fast-paced liberalization has also reduced opportunities for national champions to emerge. The growth of the Fédération des Coopératives Minières au Rwanda (FECOMIRWA) has been limited because of a lack of investment, limited availability of skilled personnel and geologists, and difficulties in dealing with the competitive environment in the domestic minerals sector. Government officials affirm that, despite efforts to help mining cooperatives and small firms, the scale of competition has impeded production increases in this area.33 It could therefore be argued that the move from ‘closed’ to ‘open’ deals has meant increased control of the sector for better-resourced actors. Local operators complained about difficulties in accessing finance, poor recovery of minerals because of inadequate mining equipment, lack of skills, and the high cost of exploration and exploitation. This is true for local operators in (p.237) trade and export operations and for those who own concessions.34 Though 70 per cent of companies are still owned by Rwandans, foreigners own most large concessions.35 The government has realized this, recognizing the importance of agreeing contracts in line with strategic priorities and bolstering enforcement capabilities within government departments.36

As with coffee, there is an increasing recognition that strategic initiatives (e.g. value addition, in the form of beneficiation) must take place in a ‘closed ordered deals’ environment. Rwanda’s existing tin smelter was sold to NMC Metallurgie (later renamed Phoenix Metals) in 2003. This occurred in a competitive bidding process, which was open to all investors. Though Phoenix later made significant investments in the smelter, the government has been unable to mobilize the necessary supply of minerals or provide enough electricity, despite guarantees made by the ministry.37 This trajectory indicates that beneficiation, although a strategic priority, operated in an ‘open disordered’ deals environment, but this has met with limited success. Attempts are therefore now being made to shift to a ‘closed ordered’ deals environment, with Phoenix being promised guaranteed supplies of electricity and other benefits (which most other firms in Rwanda would not enjoy). Figure 8.14 (p.238) illustrates the evolution of the deals space at various nodes in the minerals sector in Rwanda.

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.14. Rwanda’s deals space in the mining sector

Source: Authors’ own data.

8.5.3 Construction (Powerbroker)

Prior to 1994, most major construction projects were undertaken by foreign firms. Some domestic construction firms were also in operation, including a few that have survived to the present day. The deals environment is likely to have been fairly closed and disordered, with a small number of firms with close relations to government, a small amount of activity, and little strategic attention to the sector. At this time, however, construction had nothing like the significance to the economy that it has today.

Figure 8.15 compares Rwanda’s annual national GDP growth rates with annual growth rates in the construction sector. Unlike the two sectors discussed in Sections 8.5.1 and 8.5.2, as the graph indicates, growth in construction has regularly equalled or outstripped overall GDP growth since the early 2000s. The growth of the sector in Rwanda since 1994, and especially since 2000, should be understood in relation to the specific demand factors in this period. The devastation caused by the civil war and genocide, followed by the influx of enormous numbers of returnees, international aid, and donor representatives, all contributed to demand for construction services and materials. Physical reconstruction itself necessitated a large amount of building and infrastructure development, and urban real estate projects proliferated to house the new elite. Seizing on construction as a potentially critical sector underpinning the growth of other priority sectors, by the late 2000s the Rwanda Development Board had also introduced very attractive incentives (p.239) in the sector; for projects over US$1.8 million, there was a flat fee of 10 per cent in place of the usual import duties and other taxes.38

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.15. Comparison of annual GDP growth rates with annual growth rates in the construction sector for Rwanda

Source: MINECOFIN (2013).

This demand spurred the decision to concentrate a significant amount of investment group activity in it. However, different kinds of firms are required for different scales of activity, and the government has been keen to bring in international firms for large-scale projects. Construction differs significantly from the other sectors discussed here, in that much of the work undertaken by firms takes the form of government contracts, often funded by donor money, and generally awarded through processes of competitive tendering. Other forms of construction (i.e. private construction) do operate in a distinctive deals environment of their own, which, while not unproblematic and with some exceptions, can generally be characterized as open and ordered relative to many other developing countries (in terms of the processes of getting building permits and the necessity to adhere to planning and construction regulations).39 In what follows, however, we concentrate on the large part of construction that occurs through procurement of government contracts, which is a key ‘powerbroker’ sector in the economy. We distinguish between five key categories of firms: Chinese firms, other foreign firms, state-owned firms, other large domestic firms, and domestic SMEs.

After the genocide, a few new private firms were established to deal with immediate demand. As the government’s party- and military-owned investment groups consolidated, several construction companies were also set up under the aegis of Tri-Star/CVL and Horizon. Among these was NPD (Nyarutarama Property Developers), established in 1996. NPD bought the Belgian firm Cotraco in 2008, forming NPD-Cotraco.40 Meanwhile the military investment group Horizon had its own construction firm, Horizon Construction.

The relationship between the party-owned CVL firms and military-owned Horizon Construction merits some discussion. They are both involved in bidding for infrastructure projects and primarily view each other as competition: ‘They complain if we get favoured, they complain if they feel cheated … It’s normal competition.’41 They also maintain that they operate in an arena that is open and fair, with a significant degree of consistency and order. In the words of NPD-Cotraco:

We get harshly treated like everybody. We pay our fair share of penalties, we buy our own capital equipment, we pay 18 per cent margins on our loans … we are (p.240) doing business, just normal business … I think the perception that we are favoured was there at the start, but these guys we compete with can’t use the excuse that we are favoured any more. Maybe the competition isn’t enough, but that’s not our fault.42

One Horizon representative argued that: ‘Some might think we get jobs without competing, but let me assure you that we don’t. We always compete and often lose.’43 This might mean losing out to foreign (usually Chinese) firms, or to small local firms who have a comparative advantage due to lower overhead costs. A number of other large domestic firms also compete with NPD and Horizon, and sometimes win. Domestic firms also work together in some cases, and in joint ventures with foreign ones, and large firms are increasingly encouraged to subcontract to small ones.

The question of how Rwandan firms are treated relative to foreign ones is a source of ongoing tension. The investment groups claim to have sometimes been treated unfavourably. For example, in one case, NPD-Cotraco bid to build a particular road and the tender was cancelled because no foreign company tendered. The tender was subsequently readvertised and awarded to a foreign firm. This caused bewilderment within NPD-Cotraco, which claimed to be unaware of a stipulation that a foreign company was needed for the tendering process to be valid, suggesting a degree of disorder to the deals environment. One investment group representative claimed that 90 per cent of construction in 2014 was undertaken by foreign firms, despite growing local capacity.44 Some domestic firms even claimed that contracts were given to Chinese firms even when local firms could provide evidence of doing the job more cheaply. They also lamented their lack of knowledge, both of Chinese ‘market tricks’ and of ‘tricks of the so-called consultants’, citing the problem that foreign consultants often set the terms and budgets for construction projects before tendering begins, which intrinsically disadvantages locals.45

That the government is widely seen as providing preferential access to foreign firms is surprising, given its rhetoric of national rebuilding and the need to cultivate domestic capacity in the sector. The head of one investment group-owned firm complained that:

Rwanda is the only country I know where you can sit on your computer in China and bid for something and win it. Even in Kenya, you would have to have local content, and prove that what you were planning to do can’t be done by locals.46

(p.241) The foreign firms offered varying perspectives. A key Chinese firm representative suggested that the market was relatively open, if rather small: ‘There is no friend or favour … by African standards it really is transparent … it’s not about personal relations.’47 Yet some other foreign firms believed they lost out due to inconsistent policies and corruption, frequently expressing irritation at the number of contracts awarded to the Chinese and at collusive behaviour among Chinese firms that enabled them to undercut competitors.48 Some Chinese firms, meanwhile, believed that the government sometimes granted contracts to its own state-owned firms under preferential terms. One claimed that the government gives local firms ‘a sort of discount … so if we bid for $10 million and a Rwandese firm does too, they consider it more like $9.5 million’. He added that ‘there are no rules, it’s just certain projects that get that treatment’.49

Such claims suggest that the deals space was relatively open, with contracts being allocated across a wide range of providers relative to market size—but also that it was relatively disordered and unpredictable. Consequently, most firms felt discriminated against in some way, not because opportunities were closed to them, but because of the lack of clear rules. Ironically, the state-owned firms feel unfairly treated because they believe that the government’s concern to guard against the perception they are always favoured actually loses them some contracts—plus they cannot build as cheaply as the Chinese and do not know the market ‘tricks’ of the internationals. The non-state-owned domestic firms were actually the least inclined to complain about unfair treatment, though most are too small to compete for the big government tenders.

The openness in the sector is by no means complete, and there are cases in which the allocation of contracts is completely closed. The process of allocating contracts for four new football stadia across the country for the African Nations Championship is a case in point. Originally, the government tendered for these projects and awarded the contract to a large foreign firm. However, amid rising concern about cost, the government performed a U-turn and cancelled this contract, allocating the jobs to three domestic state-owned companies (NPD-Cotraco, Real Contractors, and Horizon) and one foreign firm (Roko).50 Contracts can also be revoked if firms are not seen to deliver on time; there are several cases where local or international firms were seen to have failed to deliver and the government has brought in the Rwanda Defence Force Engineer Regiment to take over and (p.242) finish the job.51 Horizon was also sometimes brought in at the last minute for ‘crash projects’.52 Figure 8.16 illustrates the deals space in the construction sector.

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.16. Rwanda’s deals space in the construction sector

8.5.4 Financial Services

Preceding governments retained some control over the financial sector. The Banque Commerciale du Rwanda (BCR) was incorporated as the first commercial bank in Rwanda in 1963, partly owned by the Banque Bruxelles Lambert. The government retained 42 per cent of BCR’s share capital. The Bank of Kigali (BK) was established in 1966, as a joint venture with Belgolaise SA. Later, the Banque Nationale de Paris and Dresdner Bank also invested in the bank. The government retained 50 per cent of BK’s share capital. In 1983, the Banque Continentale du Luxembourg established the Banque Continentale Africaine du Rwanda (BACAR). BCR accounted for 48.4 per cent of commercial bank assets, while BK and BACAR held 36 and 15.6 per cent of total assets, respectively (World Bank, 1991). Figure 8.17 compares Rwanda’s annual national GDP growth rates with annual growth rates in the financial services sector.

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.17. Comparison of annual GDP growth rates with annual growth rates in financial services for Rwanda

Source: MINECOFIN (2013).

(p.243) After the genocide, the local banking sector ‘was overburdened with non-performing loans and was not in a position to support the reconstruction of a capital intensive sector in the immediate aftermath of the crisis’ (Gathani and Stoelinga, 2013: 22). The government decided to license two new commercial banks—the Bank of Commerce, Development, and Industry (BCDI), and the Banque à la Confiance d’Or (BANCOR). Fick (2002) cites BCDI as an example of entrepreneurial success. BANCOR increased its share capital from RwF 300 million to RwF 1.5 billion between 1995 and 2001 (Emile, 2008). In 1999, more than forty Rwandan investors and state-owned institutions (which owned a minority share) established Cogebanque. Given the way in which these banks were set up, it is clear that the sector was relatively closed until the early 2000s. However, it is unlikely that there was a considerable degree of order in the deals environment. In the mid-2000s, the Chairman of BCDI and a senior BACAR official were both accused of embezzling funds from their banks (Behuria and Goodfellow, 2016). This showed that, for this period, there was some degree of disorder in the deals environment.

The liberalization of the financial sector gradually gathered pace in the early 2000s, with a large number of foreign acquisitions rapidly taking place from 2004 onwards and the entry of a number of international banks into the market, including Ecobank, RABO bank, the Kenya Commercial Bank, the Nigerian-based Access Bank, and the Ugandan Crane Bank.53 As of 2015, there were ten commercial banks, with the government retaining shares in three. There are also four microfinance banks, one development bank (BRD), and a military savings bank, the Credit Savings Society Zigama.

(p.244) BK has remained Rwanda’s largest bank for the last two decades. Despite liberalization, BK’s share of the market (by assets) increased from 23 per cent in 2008 to 36 per cent in 2013. In 2013, BK also had a market share of more than 30 per cent in terms of total assets, net loans, customer deposits, and equity. Though the government officially has a minority shareholding in the bank, its total shareholding (including shares owned by the Rwanda Social Security Board) stands at 54 per cent. BK representatives welcomed opening up the sector, arguing that ‘there is a lot of room in the market’ and that it helped them learn and compete at the regional level.54 There are also only three banks in which Rwandans retain a majority shareholding (BK, BPR, and Cogebanque).55 These three banks have retained 50 per cent of the market by assets between 2008 and 2014.

Most representatives of commercial banks agreed that open deals operated in the sector. One banking official said that Rwanda’s banking sector was even more open than most countries in Asia, including India, Singapore, and Pakistan.56 They generally believed that, on balance, an open environment was beneficial, pointing out that each bank has its own niche, so competition is not too damaging, as well as expressing a common belief that: ‘without liberalization, complacency sets in, and you don’t get innovation’.57 The managing director of a leading bank claimed that the banking sector:

has evolved faster than any other industry. There has been a process of leapfrogging on the technological side. We were where Zambia was in 1993. In 2009, we are probably doing more than what Zambia and Kenya have done.58

There was a sense that each bank had its own advantages: investment groups were expected to bank with the government-owned BK, while foreign banks benefited from support from their international parent companies, giving them favoured access to corporate international firms. Though representatives from other commercial banks claimed that BK received some deals ‘that were allocated without being tendered’ and that it was ‘patronized’, there were few complaints and the same respondent even said there was a ‘level playing field’.59 BK did not receive all government contracts.

The above narrative suggests that, while different banks clearly have preferential access to certain parts of the market, the sector is one in which deals are relatively open and ordered. Most stakeholders stressed predictability and revealed relatively open opportunities to engage in deals. However, the choice to embrace market-led reforms in the sector has been at odds with the need to (p.245) retain control over financing strategic investments. Some representatives of commercial banks were surprised at the government’s readiness to open up the financial sector before ‘Rwanda can stand on its own two feet’.60 Figure 8.18 illustrates the evolution of the deals space in the financial sector.

The Disorder of ‘Miracle Growth’ in RwandaUnderstanding the Limitations of Transitions to Open Ordered Development

Figure 8.18. Rwanda’s deals space in the financial sector

8.6 Conclusion

This account of the dynamics underpinning growth in specific sectors of the economy presents a much more nuanced and finely balanced situation than is offered by the conventional narratives of Rwanda’s economic development. Contrary to the idea that Rwanda’s growth has been ‘primarily driven by liberalization in the agricultural sector’, we have shown that in fact liberalization in export crops has not led to much growth, while the importance placed on value addition and strategic investments has spurred transformation in the sector. Such strategic investments have depended on closed deals. Meanwhile, the ‘developmental patrimonialism’ argument that focuses on the long-term horizons of party- and military-owned firms tells only part of the story, concealing how these operate alongside significant openness in some sub-sector deals environments.

(p.246) This openness has at times spurred innovation and facilitated large-scale investments, but also serves to threaten the order that the government has been trying to achieve in the deals environment across the economy as a whole. The increased concern with corruption, and impressive achievements in relation to containing it, can be understood as indicating a shift towards a more ordered deals environment in most sectors. However, donor pressures to embrace market-led reforms alongside the impetus to attract foreign investment have been associated with a move towards more open deals, sometimes prematurely, creating difficulties for the government in retaining control over sectors and empowering national champions. This threatens the order that the government has been trying to create.

The framework presented in this volume posits that the relationship between the political settlement and the rents space is crucial to determining everything else about deals and development. In Rwanda, under the RPF, the ‘strong dominant’ political settlement—combined with important ideological principles based on self-reliance and domestic ownership—has helped to create a rents space with significant scope for firms serving a domestic market that is substantially funded by foreign aid. As Figure 8.7 shows, GDP is dominated by powerbrokers and even more so (given liberalization) by workhorses. Aid, growth, and investment have helped workhorses (including banking and real estate) to flourish. This nexus between the political settlements and rents space has, however, done little to drive export competitiveness. Rentiers have shifted from being in the coffee sector under previous regimes to activities such as mining, but the liberalization of coffee that precipitated this has not created many magicians. More generally, the effort to create magicians has been difficult, because the government is unable to empower national champions to an extent that gives them the capacity to export in a liberalized domestic environment.

Turning to the deals environment itself, the attempt to pursue growth while maintaining a stable political settlement has led to an effort to push increasingly open deals, but with some space reserved for strategic pockets of closed deals. Government respondents suggested that a key target was to ensure that all sectors eventually operated in an environment of ‘open ordered’ deals. Yet the pressures caused by moving so rapidly towards this openness have meant that the government sometimes fails to sustain an ordered deals environment, despite the significance it places on policies promoting order. A powerbroker sector, such as construction, which depends on substantial openness due to domestic capacity deficits, exhibits a sort of state-driven open disorder, which fosters growth through openness while allowing the state to retain control through sometimes breaking, revoking, or selectively awarding contracts. Rwanda’s growth has thus depended on its ability to manage pragmatically a highly variegated deals environment—one that can appeal to donors and (p.247) investors, while maintaining a stable (yet always vulnerable) domestic political settlement. The sporadic disorder that results from this is, however, off-putting to investors in some sectors.

As the model predicts, growth and transformation in the economy (primarily towards services) have fed back into the rents space by generating opportunities for workhorses, as well as powerbrokers and rentiers in the mining sector. There have been few successful attempts at creating magicians, with the exports still reliant on primary commodities. The government has also done very little to cultivate the growth of national champions to an extent that they can compete in liberalized competitive export markets. However, it is still relatively early in the country’s development trajectory.

Also in line with the model, the Rwandan case shows how transnational factors have affected both the political settlement (for example, through foreign aid effectively supporting the strong dominant position of the current regime and influencing the government to embrace market-led reforms) and the rents space (for example, by generating new mining opportunities in Rwanda, due to the tagging of conflict minerals). However, in contrast with the model, Rwanda also shows how transnational factors can affect the deals environment directly. In construction, for example, there has been a feedback loop between donor funds for contracts and the nature of deals in the sector: the large variety of international actors funding construction has led to a proliferation of different rules and practices in tendering for contracts, causing considerable confusion. In this way, aid-driven construction growth is promoting diverse, poorly integrated institutional processes, which undermines the order of the deals environment. Meanwhile, in struggling traditional export sectors, the recourse to ‘closed ordered deals’ with loyal capitalist partners (investment groups or local elites) for strategic investments demonstrates a feedback loop between volatile demand in such sectors and the deals environment itself.

In conclusion, though some structural transformation has been achieved since 1994, growth maintenance in the country remains dependent on commodity price fluctuations, access to foreign aid, and the maintenance of a stable political settlement. Rwanda’s growth episode continues to be vulnerable to external threats—whether these take the form of international price fluctuations, diminishing donor goodwill, or rival elites who threaten the regime’s legitimacy abroad. Yet the nature of its growth in the context of widespread liberalization across sectors also leads to threats to growth maintenance that can be seen as more internal in nature, two of which stand out in particular. The first is that a high degree of openness can result in poorly coordinated economic development without sufficient enforcement capability of the state to discipline firms, so as to promote robust linkages and spillovers in the economy. The second is that growth that is highly dependent (p.248) on open (and especially foreign) competition results in periodically unpredictable and inconsistent behaviour by the government to keep a balance between foreign and domestic firms and a stable political settlement. This creates negative feedback to the deals environment by creating disorder, which may undermine investor confidence, but also contributes to a situation where the government retains very little capacity to discipline existing foreign investors in line with strategic goals.


Bibliography references:

ACET. 2014. African Transformation Report 2014: Growth with Depth. Accra and Washington, DC: African Centre for Economic Transformation.

Andrews, M., Pritchett, L., and Woolcock, M. 2013. ‘Escaping Capability Traps through Problem Driven Iterative Adaptation (PDIA)’. World Development, 51: 234–44.

Ansoms, A., and Rostagno, D. 2012. ‘Rwanda’s Vision 2020 Halfway Through: What the Eye Does Not See’. Review of African Political Economy, 39(133): 427–50.

Behuria, P. 2015. ‘Between Party Capitalism and Market Reforms—Understanding Sector Differences in Rwanda’. Journal of Modern African Studies, 53(3): 415–50.

Behuria, P., and Goodfellow, T. 2016. ‘The Political Settlement and “Deals Environment” in Rwanda: Unpacking Two Decades of Economic Growth’. ESID Working Paper No. 57. Manchester: University of Manchester.

Booth, D., Cooksey, B., Golooba-Mutebi, F., and Kanyinga, K. 2014. East African Prospects: An Update on the Political Economy of Kenya, Rwanda, Tanzania, and Uganda. London: Overseas Development Institute.

Booth, D., and Golooba-Mutebi, F. 2012. ‘Developmental Patrimonialism? The Case of Rwanda’. African Affairs, 111(444): 379–403.

Booth, D., and Golooba-Mutebi, F. 2014. ‘How the International System Hinders the Consolidation of Developmental Regimes in Africa’. Developmental Regimes in Africa Working Paper 4. London: Overseas Development Institute.

Collier, P. 2015. ‘Africa: New Opportunities, Old Impediments’. Economic Affairs, 35(2): 169–77.

Emile, B. 2008. ‘West African Banks Enter Rwandan Market’. Business Rwanda, 27 February.

Fick, D. 2002. Entrepreneurship in Africa: A Study of Successes. Westport, CT: Quorum Books.

Gathani, S., and Stoelinga, D. 2013. Understanding Rwanda’s Agribusiness and Manufacturing Sectors. Kigali, Rwanda: International Growth Centre.

Gokgur, N. 2012. ‘Rwanda’s Ruling Party-Owned Business Enterprises: Do They Enhance or Impede Development?’ IOB Discussion Paper. University of Antwerp, Belgium.

Goodfellow, T. 2013. ‘Planning and Development Regulation amid Rapid Urban Growth: Explaining Divergent Trajectories in Africa’. Geoforum, 48: 83–93.

(p.249) Goodfellow, T. 2014. ‘Rwanda’s Political Settlement and the Urban Transition: Expropriation, Construction and Taxation in Kigali’. Journal of Eastern African Studies, 8(2): 311–29.

GoR. 2000. Rwanda Vision 2020. Kigali, Rwanda: Government of Rwanda.

IMF. 2000. Rwanda: Recent Economic Developments. International Monetary Fund Staff Country Report No. 4. Washington, DC: International Monetary Fund.*.

Kar, S., Pritchett, L., Raihan, S., and Sen, K. 2013. The Dynamics of Economic Growth: A Visual Handbook of Growth Rates, Regimes, Transitions and Volatility. Manchester: University of Manchester.

Kelsall, T. 2013. Business, Politics, and the State in Africa: Challenging the Orthodoxies on Growth and Transformation. London: Zed Books.

Lagarde, C. 2015. ‘Rwanda: Taking on the Future, Staying Ahead of the Curve’. Speech made in the Parliament of Rwanda. Kigali, Rwanda.

Lavers, T., and Hickey, S. 2015. ‘Investigating the Political Economy of Social Protection Expansion in Africa: At the Intersection of Transnational Ideas and Domestic Politics’. ESID Working Paper No. 47. Manchester: University of Manchester.

Lemarchand, R. 1995. ‘Rwanda: The Rationality of Genocide’. Issue: A Journal of Opinion, 23(2): 8–11.

Macchiavello, R., and Morjaria, A. 2015. Fully Washed Coffee Exports in Rwanda: Market Structure and Policy Implications. London: International Growth Centre.

MINAGRI (Ministry of Agriculture and Animal Husbandry). 2008. Rwanda Coffee Strategy Update, 2009–2012. Kigali, Rwanda: Ministry of Agriculture and Animal Husbandry, with Ministry of Trade and Industry.

MINECOFIN (Ministry of Finance and Economic Planning). 2013. Economic Development and Poverty Reduction Strategy: 2013–2018. Kigali, Rwanda: Government of Rwanda.

NISR (National Institute of Statistics of Rwanda). 2012. Fourth Population and Housing Census 2012: Thematic Report. Kigali, Rwanda: Government of Rwanda.

Oro, Á. M., and Arias, M. B. 2012. ‘Rwanda’s Economic Success: How Free Markets Are Good for Poor Africans’. Foundation for Economic Education, 27 June. <https://fee.org/articles/rwandas-economic-success-how-free-markets-are-good-for-poor-africans/>.

Prunier, G. 1995. The Rwanda Crisis: History of a Genocide. New York: Columbia University Press.

Reyntjens, F. 2011. ‘Constructing the Truth, Dealing with Dissent, Domesticating the World: Governance in Post-Genocide Rwanda’. African Affairs, 110(438): 1–34.

Tumwebaze, P. 2014. ‘World Bank Hails Rwanda’s Growth’. New Times. 10 April.

UNSC (United Nations Security Council). 2011. ‘Report of the Panel of Experts on the Illegal Exploitation of Natural Resources and Other Forms of Wealth of the Democratic Republic of the Congo’. 15 November (S/2011/738).

Verwimp, P. 2003. ‘The Political Economy of Coffee, Dictatorship and Genocide’. European Journal of Political Economy, 19(2): 161–81.

Verwimp, P. 2013. Peasants in Power: The Political Economy of Development and Genocide in Rwanda. London: Springer.

World Bank. 1991. Rwanda: Financial Sector Review. World Bank Report No. 8934-RW. Washington, DC: World Bank.


(1) Rwanda has also been subject to extensive and important criticisms regarding limitations on freedoms and human rights, as well as increased inequality (see for example Reyntjens, 2011; Ansoms and Rostagno, 2012), though there is not space to go into these debates here.

(2) Various interviews.

(3) Interview, Emmanuel Hategeka, MINICOM (Ministry of Trade and Industry) permanent secretary.

(4) Interview, MINICOM official.

(5) Interview with foreign advisor, January 2015.

(6) Interview with CVL representative, January 2015.

(7) Interview, MINECOFIN official.

(8) Greater detail on the dynamics in these sectors can be found in Behuria and Goodfellow (2016).

(9) Interview, National Agriculture Export Board (NAEB) representative, January 2015.

(10) Interview, NAEB representative, January 2015.

(11) RwaCof bought the Gikondo coffee factory in 1997.

(12) Interview, RwaCof representative, January 2015.

(13) Interview, NAEB representative, January 2015.

(14) Interview, NAEB representative, January 2015.

(15) Interview, coffee company representatives, January 2015.

(16) Interview, foreign company representative, January 2015.

(17) Interview, NAEB representative, January 2015.

(18) Interview, Rwashoscco representative, January 2015.

(19) Interview, local coffee company, January 2015.

(20) Interview, local coffee company, January 2015.

(21) Interview, local coffee company, January 2015.

(22) Such companies include RPF cadre Alfred Nkubiri’s ENAS.

(23) See Behuria (2015) for a history of Rwanda’s minerals sector.

(24) Two interviews, Ministry of Natural Resources (MINIRENA).

(25) Interview, local mining company, January 2015.

(26) Interviews, local mining companies, January 2015.

(27) Interview, MINIRENA, January 2015.

(28) Interview, MINIRENA official, January 2015.

(29) Capability traps refer to situations 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., 2013).

(30) Interview, foreign mining investor, January 2015.

(31) Interview, Rwanda Mining Association, January 2015.

(32) Interview, MINIRENA, January 2015.

(33) Interview, FECOMIRWA, January 2015.

(34) Interview, local mining exporter, January 2015.

(35) Interview, MINIRENA, January 2015.

(36) Interview, MINIRENA, January 2015.

(37) Interview, Phoenix, January 2015.

(38) Interview with investment official, June 2014.

(39) See Goodfellow (2013, 2014) for more detail on this.

(40) In 2015, the board of directors decided to rebrand from NPD-COTRACO Ltd to NPD Ltd.

(41) Interview with NPD-Cotraco representative, January 2015.

(42) Interview with NPD-Cotraco representative, January 2015.

(43) Interview with Horizon representative, January 2015.

(44) Interview with head of investment group, January 2015.

(45) Interview with small construction firm, January 2015.

(46) Interview with head of investment group, January 2015.

(47) Interview with foreign construction firm, January 2015.

(48) Interview with foreign construction firm, January 2015.

(49) Interview with Chinese construction firm, January 2015.

(50) Interview with foreign construction firm, January 2015.

(51) Interview with architect, January 2015; interview with foreign advisor, January 2015.

(52) Interview with Horizon Construction representative, January 2015.

(53) See Behuria and Goodfellow (2016) for details.

(54) Interview with BK official.

(55) After Atlas Mara’s investments, BPR will no longer have a majority local shareholding.

(56) Interview, BCR official.

(57) Interview, BPR official.

(58) Interview, manager of foreign bank, January 2015.

(59) Interview, foreign bank official, January 2015.

(60) Interview, foreign bank official, January 2015.