El v nth Edition | D c mb r 2017 Rethinking Urbanization in Rwanda: from Demographic Transition to Economic Transformation Rwanda Economic Update Rethinking Urbanization in Rwanda: from Demographic Transition to Economic Transformation December 2017 TABLE OF CONTENTS Acronyms ..................................................................................................................................................................... i Foreword ...................................................................................................................................................................... ii Executive Summary ..................................................................................................................................................... iii Part I: Recent Economic Developments .................................................................................................................... 1 1.1 The Global and Regional Context ......................................................................................................... 2 1.2 The Real Sector..................................................................................................................................... 2 1.3 The External Sector............................................................................................................................... 5 1.4 Monetary and Financial Sectors............................................................................................................ 7 1.5 Fiscal Sector Developments.................................................................................................................. 9 1.6 Outlook ................................................................................................................................................ 14 Part II: Rethinking Urbanization In Rwanda: From Demographic Transition to Economic Transformation ����� 17 2.1 Introduction.......................................................................................................................................... 18 2.2 Urbanization Trends and Forms .......................................................................................................... 19 2.3 Rwanda’s Urban System and Economic Geography........................................................................... 25 2.4 Urbanization, Job Creation, and Poverty Reduction............................................................................ 28 2.5 Policy Discussions and Recommendations for Rethinking Rwanda’s Urban Strategy........................ 34 References ................................................................................................................................................................. 37 LIST OF Figures Figure 1: Global and regional economic growth (Percent) .................................................................................. 2 Figure 2: GDP growth, 2013–17 (Percent) .......................................................................................................... 3 Figure 3: Output growth by sector, 2014–17 (Percent)........................................................................................ 4 Figure 4: Formal trade, percent change .............................................................................................................. 6 Figure 5: Inflation, percent change ...................................................................................................................... 7 Figure 6: Rwf : US$ exchange rate, 2013–17...................................................................................................... 7 Figure 7: Credit and GDP growth, percent change.............................................................................................. 8 Figure 8: Deposit and T-bill interest rates, 2015–17 (Percent) ............................................................................ 8 Figure 9: Sources of public spending, 2006–16 (Percent of GDP)......................................................................11 Figure 10: Composition of public spending, 2007–16 (Percent of GDP)...............................................................11 Figure 11: Public investment by sector, average for 2008–16 (Percent)...............................................................11 Figure 12: Fiscal deficit, 2007–16 (Percent of GDP) ............................................................................................ 12 Figure B4.1.1: Share of official aid in GDP/budget ................................................................................................ 13 Figure B4.1.3: Sectoral distribution of aid (Percent) .............................................................................................. 13 Figure B4.1.2: Aid flows for Rwanda, 2006–15...................................................................................................... 13 Figure 13: Grant classification (Percent of GDP).................................................................................................. 13 Figure 14: Public debt, 2007–16 (Percent of GDP) .............................................................................................. 13 Figure 15: Tax years needed to repay public debt, 2007–16................................................................................ 14 Figure 16: Urbanization and per capita GDP (a) and agricultural employment (b) in benchmarked countries and Rwanda ........................................................................................................................................ 18 Figure 17: Evolution of urban settlements in Rwanda, 2002 to 2015 ..................................................................... 20 Figure 18: Rural-to-urban migration in Rwanda, the share and the scale .............................................................. 21 Figure 19: Built-up areas, 2002 and 2012 .............................................................................................................. 22 Figure 20: Urban development in Musanze, Muhanga and Huye: Base year (red), latest year (yellow)................ 23 Figure 21: Population density and distance to the city core in secondary cities .................................................... 24 Figure 22: Built-up ratio and distance to the city core in secondary cities.............................................................. 24 Figure 23: City population and GDP estimated from luminosity of lights at night, 2012 ......................................... 26 Figure 24: Link between share of nonfarm employment and MPI, population density, and built-up ratio, 2002 and 2012................................................................................................................................................ 29 Figure 25: Estimated links between increasing density and nonfarm job creation and poverty reduction by baseline density ............................................................................................................................... 30 Figure 26: Estimated links between increasing density and nonfarm job creation by geography........................... 31 Figure 27: Link between increasing density and poverty reduction by cost of travel to Kigali and market accessibility ............................................................................................................................... 32 Figure 28: Number and share of nonfarm employment in greater Kigali and secondary cities .............................. 32 Figure 29: Market access index and establishment of new firms ........................................................................... 33 Figure 30: Nearly one-third of Rwanda’s poor live within 20 km of a secondary city.............................................. 36 LIST OF TABLES Table 1: Growth rates, percent .......................................................................................................................... 5 Table 2: Rwanda: Major export (exports and reexports) and import products, percent change ........................ 6 Table 3: Rwanda’s public finances, 2015/16 to 2017/18 .................................................................................... 10 Table 4: Rwanda’s medium term macroeconomic projections ........................................................................... 15 Table 5: Rwanda’s urban system ....................................................................................................................... 26 Table 6: Economic activity indicators for Rwandan cities, city and town data, 2014.......................................... 27 Acronyms BNR National Bank of Rwanda DSA Debt Sustainability Analysis EDPRS Economic Development and Poverty Reduction Strategy EICV Integrated Household Living Conditions Survey FY Fiscal Year GDP Gross Domestic Product ICT Information and Communication Technology IMF International Monetary Fund LTR Land Tenure Regularization MAI Market Accessibility Index MINECOFIN Ministry of Finance and Economic Planning MPFSS Monetary Policy and Financial Stability Statement MPI Multidimensional Poverty Index NISR National Institute of Statistics of Rwanda NPLs Nonperforming Loans NST National Strategy for Transformation NUP National Urbanization Policy OECD Organization for Economic Cooperation and Development REU Rwanda Economic Update Rwf Rwanda Franc SSA Sub-Saharan Africa TFP Total Factor Productivity US$ United States Dollar WDI World Development Indicators Rwanda Economic Update • Edition No. 11 i FOREWORD T he Rwanda Economic Update (REU) reports on and synthesizes recent economic developments; places them in a medium-term, regional, and global context; and analyzes the implications of these developments and policies for the outlook of the economy. These reports attempt to make an analytical contribution to the implementation of Rwanda’s national development strategy. Each edition has a special feature analyzing more deeply a selected topic. The report is intended for a wide audience of policy makers, business leaders, other market participants, the community of analysts engaged in Rwanda’s economy, and civil society. The eleventh edition of the REU was jointly prepared by the World Bank Rwanda Macroeconomics and Fiscal Management Global Practice team and the Social, Urban, Rural and Resilience Practice team. The teams were led by Aghassi Mkrtchyan (Senior Economist) and Narae Choi (Urban Specialist). Peace Aimee Niyibizi (consultant) contributed to the analysis of recent macroeconomic developments. Somik Lall (Lead Urban Economist) provided guidance to the team for the special topic. The report was prepared under the overall guidance of Diarietou Gaye (Country Director), Abebe Adugna (Practice Manager), Bernice Van Bronkhorst (Practice Manager for Urban and Disaster Risk Management) and Yasser El-Gammal (Country Manager). Sylvie Ingabire and Karima Ladjo (Team Assistants) supported the team. Naoko Kojo (Senior Economist), Ruslan Piontkivsky (Senior Economist), and Mark Roberts (Senior Urban Economist) were peer reviewers. Although this report does not represent the official views of the Rwandan authorities, the macroeconomic unit of the Ministry of Finance and Economic Planning (MINECOFIN) and the National Bank of Rwanda (BNR) were engaged in its formulation and provided valuable comments. The World Bank team appreciates their contributions. ii Rwanda Economic Update • Edition No. 11 EXECUTIVE SUMMARY Recent Economic Developments rate of new applications, and approval of fewer D new loans. Growth in deposits was affected by an evelopments in the Rwandan economy as increase in the government’s issuance of T-bills. of the first half of 2017 have been mixed. These factors have held back credit to the economy. On the upside, Rwanda’s external imbalances have eased on account of improvements in Despite an increase in the headline budget global commodity prices, global and regional deficit the overall fiscal stance was not strengthening of growth recovery, and the external expansionary because of unchanged fiscal adjustment the government undertook in 2016. deficit excluding grants. With spending higher Inflation pressures have also eased as Rwanda than projected, the headline fiscal deficit widened and the region recover from droughts; and after in FY2016/17 to 4.9 percent of GDP, up from the depreciations in 2015 and 2016, the currency has 3.5 percent deficit in FY2015/16. At 9.5 percent stabilized. On the other hand, economic activity of GDP, the fiscal deficit excluding grants was has continued to be subdued, with an annualized unchanged from the previous fiscal year due growth of just 3.4 percent in the first half of 2017, to a decline in spending under grant-financed on account of weak domestic demand. projects. Revenue collection is still subdued, and external grants continue to decline. Compared to Growth decelerated across the board. Despite previous year, total public spending was lower by recovering from the droughts, growth in agriculture 0.2 percentage points of GDP, and total revenues remained low, and a contraction in construction were lower by 1.7 percentage points. The decline put the brakes on industrial activity. Growth in in total revenues was largely due to the 1.3 services decelerated driven by weak retail and percentage points decline in grants. wholesale trade. Mirroring these developments, both private consumption and investments had The larger deficits and public guarantees declined in real terms by mid-year. Exports, for investment projects of recent years have however, saw healthy growth as export prices rose meant higher public debt. According to the and the exchange rate became more competitive, joint IMF and World Bank debt sustainability while a substantial contraction in imports markedly improved the trade balance. assessment Rwanda’s risk of debt distress remains low. However, public debt has grown Inflationary pressures subsided significantly steadily, and is expected to reach 46 percent of in the first nine months of 2017, while exchange GDP by year-end 2017. This has substantially rate was relatively stable. From its recent peak narrowed Rwanda’s fiscal space. of 8.1 percent in February 2017, headline inflation declined to 3.8 percent by September largely due To sustain growth and maintain the necessary to reductions in food price inflation which fell to 8.3 fiscal space, public investments must be percent from the February peak of 17.6 percent. better targeted. Total factor productivity slowed Exchange rate depreciation continued in 2017, down substantially during the investments push albeit at much slower rate than in 2015 and 2016. by the government in 2013-2015 which may suggest that economic returns on recent public The banking system remains well capitalized investments may have been lower than expected. but risks intensified. This is evidenced by This can undermine Rwanda’s growth prospects increased non-performing loans, higher rejection and raise concerns about fiscal sustainability. Rwanda Economic Update • Edition No. 11 iii Executive Summary Now needed is a balanced approach to address The special topic of this update examines investment needs in areas that have potential the role of urbanization as one of the for higher economic returns and for crowding in important factors that may facilitate growth private investments. and poverty reduction in the long term. Long term economic growth will depend on Addressing fiscal contingencies is important Rwanda’s ability to improve productivity through for debt sustainability. The reform program better use of capital and human resources. recently adopted by the authorities to address Among other important determinants such as fiscal risks in the energy sector is an important improved business environment, skills, good step forward. It aims at fiscally sustainable governance and infrastructure, well-managed expansion of electricity services by containing urbanization can greatly contribute to productivity fiscal impact of the electricity sector and through agglomeration, density and improved improving the operational efficiency, affordability, connectivity. There is evidence that urbanization and accountability of electricity services. in Rwanda is already positively associated with jobs and poverty reduction. This highlights the In the short to medium term, the growth importance of policy discourse on urbanization outlook is positive. It is estimated that by mid- as part of formulating country’s growth agenda. 2017 economic growth had already bottomed out and is now likely to recover to 5.2 percent by Rethinking Urbanization in Rwanda: From yearend and will accelerate in 2018 as agriculture Demographic Transition to Economic and investment, both public and private, pick Transformation up. Economic activity will also benefit from the In its vision to become a middle-income expected recovery of prices of such traditional country by 2020, the Government of Rwanda exports as minerals, tea, and coffee. A more has identified off-farm job creation and competitive exchange rate is already supporting urbanization as drivers of economic growth non-traditional exports, which may become and national development. To realize this an important source of growth. As the region vision, a key objective of the country’s economic recovers from drought, the agriculture outlook for development and poverty reduction strategy the medium term is positive. The government’s (EDPRS) II (2013–18) is to transform Rwanda’s renewed commitments to scale up investments in economic geography by facilitating urbanization agriculture, especially irrigation, will reinforce the and supporting secondary cities as poles of medium-term outlook. growth. The transformation would require The main risks to the growth outlook are transitioning 50 percent of the population from associated with possible weak external farm to off-farm jobs. Specific targets for 2020 environment, persisting external imbalances, also include a 35 percent urban population and continued debt accumulation, and a weak creation on 1.8 million new off-farm jobs. private sector. If global prices of minerals, coffee, and tea continue to be low, they will subdue The 2012 Census found that 16.5 percent production and exports. The pace of economic of Rwandans were living in urban areas – transformation will largely depend on the extent but Rwanda has been urbanizing rapidly. to which government expectations for its major Between the 2002 and 2012 censuses, the tourism and connectivity investments materialize. degree of urbanization did not differ much, Continued lack of private sector response to the despite observable changes that suggested improved investment climate is another risk. greater urbanization. The apparent stability was iv Rwanda Economic Update • Edition No. 11 Executive Summary a result of a change in the census definition of additional 8 percent (5 percent of total GDP). urban areas that made it difficult to compare the Kigali also accounts for 27 percent of all nonfarm figures directly. An alternative analysis using a jobs created between 2011 and 2014 and more simple definition (minimum population of 5,000 than 50 percent of all formal private sector firms and a population density greater than 1,000) and employment in them. demonstrates that between 2002 and 2015 Rwanda’s urban population increased from 1.49 Outside Kigali, the urban corridor between million to 3.46 million—by almost 2 million people Rubavu and Musanze has emerged as the or 132 percent—at an average rate of 6.7 percent largest concentration of economic activity. per year. They benefit from proximity to Goma in the Democratic Republic of Congo (DRC) as well as The increase in the urban population has densely populated fertile agricultural land and the been accompanied by the physical expansion presence and investment from relatively large of cities, notably in the periphery of Kigali and formal firms and tourism. Together, the two cities around secondary cities, mostly at low levels account for 7 percent of nonagricultural GDP, of density. Rwanda’s zoning system contributes 4 percent of jobs, and 5 percent of firms. They to low-density development in urban peripheries also have a higher than average concentration by strict building standards regulating, for of formal firms and jobs—an indication of more example, minimum plot sizes and maximum productive economic activity. The economic plot coverage ratios. This is likely to lead some performance of other secondary cities varies, but investors and household to areas where zoning all have a relatively small share in job creation. has not yet been completed or planning codes Some secondary cities have not reached a are less rigidly enforced. The trend is reflected sufficient size in population or density of firms to in move of population toward the fringes of Kigali create agglomeration economies. City. Of all recent migrants to sectors1 bordering Kigali City, 43 percent came from the city itself. Urbanization has been accompanied by poverty reduction primarily in areas with Rwanda’s urban system is dominated by high density and good connectivity, but Kigali, which is also the economic center of its potential has not been fully realized. Rwanda. Nightlights-based estimation of district- The changes in population and poverty in level GDP2 concluded that the three districts Rwanda’s 416 sectors between 2002 and 2012 of Kigali,3 which represent 11 percent of total demonstrate that a 10 percent increase in population and less than 3 percent of total land population density was associated with a drop area, accounted in 2012 for 40 percent of total in the multidimensional poverty index (MPI) of GDP and 61 percent of total nonagricultural 1.2 percent. Despite the general contribution of GDP. Kigali’s primacy is even more evident urbanization to poverty reduction, it is mainly in comparison to what other cities contribute areas with higher population density that have to the national economy. The six secondary successfully translated increasing density to cities together account for 19 percent of total poverty reduction. In other words, an increase nonagricultural GDP (13 percent of total GDP), in density does not necessarily reduce poverty and the three fast-growing towns4 contribute an in areas with low-density and little access to 1 Sector is an administrative unit in Rwanda; there are 416 of them. Unless specified they should not be confused with economic sectors. 2 GDP estimates use satellite imagery from the National Oceanic and Atmospheric Administration (NOAA) and employment and firm data from Establishment Census 2014. 3 Kigali City consists of three districts: Gasabo, Kicukiro, and Nyarugenge. 4 Nyamata, Gicumbi, and Rwamagana/Kayonza are included in the analysis because their urban population has grown fast, although from a low base, and they demonstrate economic potential. Rwanda Economic Update • Edition No. 11 v Executive Summary markets: except for areas within Greater Kigali or jobs. Rwanda is one of most land-scarce close to secondary cities, agglomeration effects economies in Africa yet has the region’s highest on poverty reduction are minimal. Given the population density; the expanding rural population low-density expansion of these cities, current will encroach on land available for agriculture and urban planning and practice should be revised thus continue to shrink farm sizes. A large share not only to increase economic density but also of employment and nonfarm income opportunities improve welfare. will depend on linking rural and urban economies, including through trade and migration. The Policy Recommendations government’s role is to facilitate this process by creating an enabling environment wherever Rwanda’s towns and cities should be market potential has emerged, rather than managed as a portfolio. Kigali will continue to directing the spatial allocation of resources by be the leading economy, with other cities having predetermining where place-specific investments different roles. Concerns over Kigali’s size will be made. and growth have led policymakers to promote secondary cities as substitutes for Kigali. Densification is critical given that the link However, as observed, businesses are more between urbanization, poverty reduction, and likely to locate in Kigali, where they benefit from job creation quickly tapers off with distance a range of financial and related services and a diversity of product and labor market, not available from a city’s core. The interaction between elsewhere in the country. Concerns for Kigali urbanization, poverty reduction, and job creation becoming a congested and ill-managed city can in secondary city centers declines steeply in be addressed through better urban management. areas 5−10 kilometers (km) away from the city For other cities, they should be part of the cores. Yet opportunities for leveraging the benefit economic strategy, not of a demographic strategy of urbanization can be found in these peri-urban (i.e. becoming distributors of the population away areas if they were better connected to the city from Kigali). This requires assessing and defining cores both physically and economically, as well their roles in the economic transformation based as to their rural hinterlands. Current planning and on their economic performance and potentials. zoning guidelines in these cities can be applied in ways that increase their economic density and Policy should focus on strengthening the connectivity, factoring in future as well as current linkages between rural and urban economies and future demands of their citizens. rather than on simply creating more nonfarm vi Rwanda Economic Update • Edition No. 11 PART ONE RECENT ECONOMIC DEVELOPMENTS Rwanda Economic Update • Edition No. 11 1 Recent Economic Developments 1.1 The Global and Regional Context Sub-Saharan Africa (SSA) is recovering from the sharp slowdown of the past two years. After slowing in 2016, global growth is SSA growth is projected to double from last year’s projected to strengthen to 2.7 percent in 2017 1.3 percent to 2.6 percent this year. Most of the (Figure 1). Since the beginning of the year, the rebound will come from major oil exporters such global economy has recovered well, particularly as Nigeria and Angola. The upturn is supported in advanced economies. In these economies, by rising commodity prices, stronger external strengthened investments and exports are demand, and better weather conditions in some reviving economic activity. Higher demand for sub-regions. However, growth is projected to be imports is contributing to a recovery in the global uneven across the region, and for the second goods trade, an uptick in commodity prices, and consecutive year, regional per capita output boosting the recovery for commodity exporters. growth is expected to be negative. The slowdown in growth appears to be bottoming out for commodity exporters and is projected 1.2 The Real Sector to reach just 1.8 percent for the year. More promisingly, growth in emerging markets and As of June 30, Rwanda’s economy had grown developing economies is projected to recover at an annualized 3.4 percent—a considerable from 3.6 percent in 2016 to 4.1 percent. drop from the 5.9 percent at the end of 2017. The slowdown, which appears to track the economic slowdown caused by the aid crisis in 2013 (Figure Figure 1: Global and regional economic growth (Percent) 2), began in 2016 and was driven by drought 8 and the completion of large investment projects. 7 The fiscal consolidation in FY 2015/2016, which 6 was designed to reduce external vulnerabilities, 5 had temporarily reduced the growth of domestic 4 demand until fiscal policies became expansionary 3 again in the second half of FY2016/2017. 2 1 In addition to these temporary factors, growth 0 4.4 3.2 2.4 2.6 2.8 2.7 2.4 2.7 2.9 has also been slowed by such long-term 2010 2011 2012 2013 2014 2015 2016 2017 2018 structural factors as anaemic productivity World AEs CEs EMDEs SSA growth and low returns on investments . Total Source: World Bank, Global Economic Prospects Notes: AEs: Advanced economies. EMDEs: Emerging markets and developing factor productivity (TFP) has been decelerating in economies. CEs: Commodity exporters; SSA: Sub-Saharan Africa. Rwanda since 2011; the main driver of growth in Key Points: Rwanda’s economy continued to slow in the second quarter of 2017 but now appears to have bottomed out and growth is projected to rebound. On the supply side: Agriculture is recovering, though slowly, but construction continues to decelerate, while the growth in services remained below the average of past three years. On the demand side: Government consumption combined with the recovery in net export were the main sources of growth, while domestic demand took a hit, with negative growth in private consumption and investments. 2 Rwanda Economic Update • Edition No. 11 Recent Economic Developments recent years has been a government investment underscores how dependent the economy is push. With external grants declining and on government-led investment; and the minimal borrowing space for expanding public investment growth in productivity highlights how important it now limited due to higher public and publicly is to allocate production factors more efficiently if guaranteed debt, that economic model might higher and sustained economic growth is to be have reached its limits. The growth slowdown achieved. Vision 2050, which the government is underscores how dependent the economy is currently drafting, focuses on stimulating growth on government-led investment; and the minimal and reducing poverty by building up private growth in productivity highlights how important it sector and market institutions to encourage the is to allocate production factors more efficiently most productive use of Rwanda’s resources if higher and sustained economic growth is to be and talent. achieved. Vision 2050, which the government is currently drafting, focuses on stimulating growth Figure 2: GDP growth, 2013–17 (Percent) and reducing poverty by building up private sector 12 and market institutions to encourage the most 8.9 9.1 productive use of Rwanda’s resources and talent. 10 8.6 8.6 7.7 7.6 7.8 8.0 7.9 8 7.0 7.0 As of June 30, Rwanda’s economy had grown 5.8 5.4 4.9 5.9 6 at an annualized 3.4 percent—a considerable 4.7 4.2 drop from the 5.9 percent at the end of 2017. The 4 3.4 slowdown, which appears to track the economic 2 slowdown caused by the aid crisis in 2013 (Figure 0 4.7 7.8 2.3 4.0 7.5 5.910.56.7 8.0 9.3 8.110.08.9 7.4 5.4 2.4 1.7 4.3 2), began in 2016 and was driven by drought Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 2013 2014 2015 2016 2017 and the completion of large investment projects. Year-on-year Annualized rate The fiscal consolidation in FY 2015/2016, which Source: Data from the National Institute of Statistics of Rwanda (NISR), was designed to reduce external vulnerabilities, had temporarily reduced the growth of domestic In the first half of 2017, growth slowed in all demand until fiscal policies became expansionary main sectors of the economy (Figure 3). As again in the second half of FY2016/2017. of June 2017, industrial growth had retreated to an annualized rate of 1.9 percent, largely due In addition to these temporary factors, growth to a contraction in construction. In 2014–16, has also been slowed by such long-term construction and the government’s investment structural factors as anaemic productivity push had fuelled the industrial growth that growth and low returns on investments5. Total averaged 8.9 percent. Growth in domestic trade factor productivity (TFP) has been decelerating in had averaged 8.2 percent for the preceding three Rwanda since 2011; the main driver of growth in years, but by June 2017 a falloff had slowed recent years has been a government investment growth in services to 5.5 percent. Although by push. With external grants declining and June agriculture had begun to recover from the borrowing space for expanding public investment droughts of 2016, having risen by an annualized now limited due to higher public and publicly 3.1 percent, it was nevertheless still far below guaranteed debt, that economic model might the 5.1 percent growth it had averaged in the have reached its limits. The growth slowdown preceding three years. The 10th Rwanda Economic Update. 1 Rwanda Economic Update • Edition No. 11 3 Recent Economic Developments Figure 3: Output growth by sector, 2014–17 categories in Rwanda, went down 6 percent (Percent) y-o-y—another drag on manufacturing growth. 12 Growth in mining and quarrying decelerated to 10.4 10 1.4 percent in the first half of 2017, led downward 8.9 8.8 8 7.5 8.2 8.1 by coltan and wolfram (Rwanda’s traditional 5.8 mineral products). Although higher export prices 6 5.2 5.5 5.2 4.2 helped the wolfram sector, export volumes have 4 3.4 3.1 2.9 contracted. Meanwhile, production of coltan was 1.9 2 somewhat higher but export prices were low. 0.2 0 H1 of 2017 H1 of 2017 H1 of 2017 H1 of 2017 H1 of 2016 H1 of 2017 H1 of 2016 H1 of 2017 H1 of 2016 H1 of 2017 H1 of 2016 H1 of 2017 2014-2016 2014-2016 2014-2016 2014-2016 Services grew 5.2 percent in the first half GDP Agriculture Industry Services GDP Agriculture Industry Services of 2017, down from a three-year average of 8.8 percent (Figure 3). The slow growth in the Annualized Year-on-year Source: NISR data. services sector was caused by limited domestic The recovery in agriculture has varied by demand and private consumption. Wholesale crop. Food crop production drove the 4.2 percent and retail trade activities declined by 9.4 percent expansion in output in the first half of 2017 in compared to the same period in 2016 (Table year-on-year (y-o-y) terms (Figure 3); and food 1). Transport, ICT, and financial services also crops, which account for more than 65 percent underperformed. Growth in financial services of agricultural output, grew by 3.8 percent (Table went down from 6.9 percent in the first half of 2016 1). The recent Seasonal Agricultural Survey6 to 2.2 percent, mainly because of slower credit found production up by 3.5 percent in Season growth. Hotels and restaurants, fortunately, grew A of 2017 after falling by 8.7 percent the year 12.5 percent in the first half of 2017, capitalizing before. Production volumes of tubers, roots, on large investments there in recent years. and fruits increased substantially, but those of cereals, pulses, and bananas dropped. Cereals The growth slowdown was mainly caused in particular suffered from a disease that attacked by slack domestic demand, though that maize. Both coffee and tea were hit by second- was partly offset by higher external demand. round effects of the 2016 droughts.7 Overall, crop Domestic demand (the sum of consumption exports fell by 3.9 percent. and investments) contracted by 3.5 percent y-o-y in the first half of 2017 (Table 1): private Industrial output growth was only 0.2 percent consumption fell by 5.5 percent and investments y-o-y (Figure 3), the lowest semi-annual by 5.1 percent. These drops were partly offset by growth since 2009, largely because of a 5.5 a 12 percent growth in government consumption. percent contraction in construction. Growth As for external demand, exports in real terms in manufacturing was relatively strong at 6 grew by 21.8 percent y-o-y as imports dropped percent, although that was less than the average by 10.2 percent. As a result, the negative trade for the past three years (Table 1). Beverages balance8 was reduced by 26.8 percent, thus and tobacco, the second biggest manufacturing contributing to the first-half economic growth. 6 After every harvest, to estimate agricultural production by district the National Institute of Statistics of Rwanda (NISR) carries out a farm survey covering areas under cultivation, plot size, cropping system, production and yield, inputs (seeds, fertilizers. and pesticides) and production activities (e.g., irrigation and anti-erosion practices). http://statistics.gov.rw/datasource/seasonal-agricultural-survey-2017. 7 According to the National Agricultural Export Development Board, the severe dry season in the second half of 2016 depressed production of coffee and reduced rainfall in the first half eroded production of tea. National Agricultural Export Development Board Reports are available at http://www.naeb.gov.rw/index.php?id=174 8 Exports minus imports. 4 Rwanda Economic Update • Edition No. 11 Recent Economic Developments Table 1: Growth Rates, Percent H1-2017 2013 2014 2015 2016 year-on- year GDP growth 4.7 7.6 8.9 5.9 2.9 Production side Agriculture 3.6 6.4 5.0 3.9 4.1 Food crops 3.7 8.6 3.6 3.1 3.8 Export crops –3.5 –2.7 14.0 1.6 –3.9 Industry 9.3 11.1 8.9 6.5 0.2 Mining 20.4 24.6 –4.8 10.0 1.4 Manufacturing 4.6 7.7 8.4 7.2 6.0 Construction 10.9 10.1 15.3 4.9 –5.5 Services 5.4 7.2 10.3 7.2 5.2 Trade & transport 6.2 6.8 11.2 6.9 –4.6 Wholesale & retail trade 5.6 8.5 12.9 6.0 –9.4 Other services 5.1 7.3 9.9 7.3 8.6 Net taxes –3.3 7.7 14.4 4.7 –10.8 Demand side Domestic demand 4.4 9.5 13.1 5.6 –3.5 Government consumption –0.3 20.8 5.1 9.1 12.0 Household consumption 4.0 8.9 13.2 4.0 –5.5 Investment 8.2 5.4 17.9 8.5 –5.1 Net exports 2.5 21.2 36.6 4.3 –26.8 Source: NISR data. Note: A negative sign of net exports symbolizes a reduction in the trade deficit, positively affecting GDP growth. 1.3 The External Sector 20.2 percent of GDP as of December 2016 to 16.5 percent for the year ending in June 2017, Rwanda’s formal trade imbalance9 has eased demonstrating how much external adjustment substantially (Figure 4). In the first half of 2017, that Rwanda has undergone in 2017. Exports export values expanded by 39.8 percent and were driven by solid growth in non-traditional imports contracted by 10.6 percent (Table 2), exports, recovery in the prices of some traditional narrowing the trade deficit from US$902 million exports, and the more competitive exchange rate in the first half of 2016 to US$671 million in the after more than 20 percent nominal depreciation same period of 2017. The trade deficit fell from in 2015-2017. Key Points: Strong exports and fewer imports narrowed Rwanda’s trade deficit in the first half of 2017. Solid growth in non-traditional exports and a more competitive exchange rate spurred growth in exports. Imports contracted as domestic demand weakened. Given that data on current account and balance of payments are published only annually basis, the REU-11 uses only formal trade 9 data for Rwanda’s traditional and nontraditional exports and re-exports, and for imports of consumer, capital, and intermediary goods, energy, and lubricants. Rwanda Economic Update • Edition No. 11 5 Recent Economic Developments Figure 4: Formal trade, percent change went up 20.1 percent (Table 2). Nevertheless, their share in total goods exports continued to decline, dropping from 34.5 percent in 2016 to 30 percent. High prices pushed up the value of tea exports by about 31 percent, and tin exports went up 55 percent because of both higher prices and higher export volumes. The value of coffee exports remained almost unchanged, while that of wolfram and coltan declined. The contraction of imports by 10.6 percent in the first half of 2017 was largely driven by Source: NISR and BNR data. lower demand for capital and intermediary goods (Table 2). Capital goods, which account The solid growth in non-traditional exports for about 29 percent of formal goods imports (CIF) was led by minerals. Exports of such minerals declined by 25 percent in the first half of 2017, as beryllium, unwrought lead, iron ore, and while intermediary goods which account for 26 gemstones almost tripled in the first half of percent of formal goods imports contracted by 2017, accounting for 19.2 percent of total 6 percent during the same period. The decline goods exports. After declining by 17.4 percent in intermediary imports was mostly due to a in 2016, the value of traditional export products 27 percent drop in imports of construction Table 2: Rwanda: Major export (exports and reexports) and import products, percent change H1 of 2011 2012 2013 2014 2015 2016 2017 Export growth 52.4 24.5 18.7 4.7 –6.8 7.1 39.8 Domestic exports 60.3 6.9 16.9 –0.8 –12.3 –1.7 47.9 Traditional products 63.5 –5.1 24.2 –7.1 –19.8 –17.4 17.9 Coffee 33.0 –18.4 –9.8 8.7 4.0 –5.7 0.2 Tea 14.7 2.9 –15.6 –6.7 40.0 –12.5 30.7 Minerals 123.4 –10.1 65.9 –9.9 –42.1 –26.6 18.5 Cassiterite 129.4 –45.4 15.5 17.8 –52.4 1.6 54.9 Coltan 108.8 47.5 136.5 –22.1 –36.8 –40.0 –1.4 Wolfram 125.7 63.9 14.4 –11.5 –34.8 –31.5 –13.0 Other products 43.0 82.0 –6.9 26.5 11.7 34.2 95.2 Other Minerals 113.3 –96.3 78.5 1,543.4 288.8 156.0 172.3 Reexports 3.8 189.8 25.0 22.5 7.6 26.1 27.6 Import growth 33.8 11.7 6.4 6.2 –2.8 –3.1 –10.6 Consumer goods 24.0 8.7 5.4 3.6 5.8 4.9 –3.2 Capital goods 27.3 21.6 5.4 7.7 1.6 9.3 –25.0 Intermediary goods 35.1 13.5 5.9 13.8 –5.2 –16.6 –6.1 o/w construction materials 24.5 1.5 4.5 4.8 4.9 –27.4 –26.6 Energy and lubricants 63.0 0.4 10.8 –4.2 –21.1 –18.3 6.1 Source: BNR data. 6 Rwanda Economic Update • Edition No. 11 Recent Economic Developments materials. Imports of consumer goods, which in 2016, the Rwandan franc (Rwf) depreciation represent the largest share of imports, declined had slowed to 2.5 percent in the first 9 months by 3.2 percent. The value of energy imports, in of 2017 (Figure 6) and was holding steady contrast, went up by 6percent, despite the 0.8 against other regional currencies. By September percent decline in volume. 2017, Rwanda’s real effective exchange rate has depreciated by 5.5 percent mirroring the 1.4 Monetary and Financial Sectors movements of the nominal effective exchange rate. According to the most recent IMF Inflationary pressures subsided significantly assessment, the real effective exchange rate is in the first nine months of 2017. From its recent largely in line with macroeconomic fundamentals. peak of 8.1 percent in February 2017, headline inflation declined to 3.8 percent by September Although the National Bank of Rwanda has (Figure 5) largely due to reductions in food cut its policy rate, credit growth remains price inflation which fell to 8.3 percent from the February peak of 17.6 percent. Even though rural subdued. In response to a steep decline in food price inflation plunged from 26percentin credit growth and the easing of inflationary and February to 11 percent in September, it still exchange rate pressures, the National Bank remained several percentage points higher than of Rwanda (BNR) cut its key policy rate by 50 headline inflation.10 basis points, to 6 percent between December 2016 and May 2017. Meeting in September, the The exchange rate has stabilized after BNR’s Monetary Policy Committee decided to depreciating for about two years. Following keep the rate unchanged for the rest of 201711. a 10 percent depreciation against the US dollar Nevertheless, credit growth has been subdued, Figure 5: Inflation, percent change Figure 6: Rwf : US$ exchange rate, 2013–17 30 900 25 850 20 17.6 800 15 10 8.1 750 10% 2.5% 3.7% depreciation depreciation 5 depreciation in 2016 as of Oct.2017 700 in 2014 7.4% 0 depreciation in 2015 -5 650 -10 600 Sep-12 Mar Sep-13 Mar Sep-14 Mar Sep-15 Mar Sep-16 Mar Sep-17 Dec. Mar Jun Sep Dec. Mar Jun Sep Dec. Mar Jun Sep Dec. Mar Jun Sep. Food-Rural Headline Food-Urban Rural Inflation 2013 2014 2015 2016 2017 Source: NISR data. Source: BNR data. Key Points: Inflation has returned to the low single digits. Exchange rate pressures have eased as external imbalances narrow. Higher risks and government crowding out are delaying the recovery of credit growth. 10 Urban inflation is used for headline inflation. 11 https://www.bnr.rw/fileadmin/AllDepartment/services/Public_notices/doc01560320170927111322.pdf Rwanda Economic Update • Edition No. 11 7 Recent Economic Developments reaching only 7.2 percent in August. To control 2017. Since 2016 interest rates on T-bills excess liquidity, the BNR has been intervening in have exceeded bank deposit rates (Figure 8), the money market through repo operations which making investments in government bonds more went up to Rwf 795 billion in the first half of 2017, attractive. There is evidence that the nonbank 16 percent higher than for all of 2016. investors, especially large institutional investors, are drawing down their bank deposits to invest Banks have remained cautious in extending more in government securities. The growth rate new loans amid increased risks. NPLs in deposits was 11 percent (as of June 2017), the increased from 6.2 percent in December 2015 to slowest since 2014.12 8.2 percent in June 2017. New loans contracted by 2.4 percent y-o-y in the first half of 2017. Rwanda’s banking sector remains well Most affected were commerce and hospitality capitalized. The capital adequacy ratio (total (restaurants and hotels), for which new loans capital to risk weighted assets) of the banking approved declined by 21.6 percent y-o-y. sector stood at 21 percent in June 2016, slightly According to the MPFSS, the bank loan rejection lower than 22 percent as of December 2016 rate was 17 percent in the first half of 2017—5 but still well above the regulatory minimum of percentage points higher than in the first half 15 percent. Most recent round of stress-testing of 2016. Historically close relationship between conducted by the BNR found that, even after credit growth and domestic demand in Rwanda applying consecutive credit default shocks, the may indicate that the banks are becoming risk large majority of banks would remain safely averse during the slump in domestic demand above the regulatory minimum capital adequacy (Figure 7). requirement. The system is resilient to liquidity and exchange rate shocks. The slowdown in credit and deposit growth coincided with the rise in government The banking sector’s liquidity, efficiency issuance of domestic debt. The issuance of and profitability indicators, however, have T-bills increased 22 percent as of September weakened over the last two years. The slowdown Figure 7: Credit and GDP growth, percent change Figure 8: Deposit and T-bill interest rates, 2015–17 (Percent) 20 40 10 15 30 8 10 20 5 10 6 0 0 -5 -10 Mar-12 Dec-12 Sep-13 Jun-14 Mar-15 Dec-15 Sep-16 Jun-17 4 Sep-15 Dec-15 Mar-16 Jun-16 Sep-16 Dec-16 Mar-17 Jun-17 Sep-17 Domestic demand GDP Credit growth (right axis) Deposit rate T-bill rate Source: BNR and NISR data. Source: BNR data Note: Credit growth is adjusted to the GDP deflator. According to the August 2017 Monetary Policy and Financial Stability Statement, the total investment in government securities by 10 the non-bank sector (individuals and non-bank institutions) has increased by more than 65 percent, leading to 20 percent decline in their deposits in the banking sector. https://www.bnr.rw/fileadmin/AllDepartment/MonetaryPolicy/upload/FinancialStatement/ AnnualFinancialStatements/MPFSS_2017_4_Print_Final.pdf 8 Rwanda Economic Update • Edition No. 11 Recent Economic Developments in economic activity, rising problems in certain percentage points. The decline in total revenues asset classes (real estate/tourism/hotels) and was largely due to the 1.3 percentage points increased banking sector competition through decline in grants, to 4.6 percent of GDP. At 9.5 entrants of new players are likely contributing percent of GDP, the fiscal deficit excluding grants factors. The liquid assets13 to total deposits was unchanged from the previous fiscal year ratio stood at 39 percent by June 2017, still (Table 3). comfortably above the regulatory minimum ratio of 20 percent, but significantly lower compared to Revenue performance in FY2016/17 remained 51 percent in December 2014. The loan to deposit weak. Overall, domestic revenue came in at 18 ratio has been on the rise as well, reaching 97 percent of GDP compared to 18.4 percent in the percent in March 2017. Indicators of banking previous year (Table 3). Tax revenues were 15.5 sector efficiency (such as cost to income ratio; percent of GDP, slightly down from 15.8 percent. overhead to income) and profitability have been Taxes on goods and services fell from 8.1 percent on a stagnating or declining trend over the last of GDP in FY 2015-16 to 7.6 percent. The Budget two years, which may indicate the limitations Execution Report of 2016/1715 attributes the of efficiency gains through scale in the context problem to lower private consumption, especially of a small economy, a slowdown in economic of beer, soft drinks, and petroleum products. Taxes growth and increased competition through new from international trade and non-tax collections market entrants. remained unchanged from the previous fiscal year, but collection of direct taxes went up by 0.2 1.5 Fiscal Sector Developments percentage points of GDP, to 6.6 percent. Recent Fiscal Developments and the Outlook for Recurrent expenditures and net lending were FY2017/18 both slightly higher in FY2016/17. While the The overall fiscal stance measured by overall spending was lower than in the previous the deficit excluding grants was almost fiscal year (Table 3), recurrent spending was unchanged compared to FY2015/16. Rwanda’s higher at 15 percent of GDP, up from 14.7 percent headline fiscal deficit widened in FY2016/17 to in FY2015/16. Capital expenditures dropped from 4.9 percent of GDP (Rwf 310 billion), up from the 11.4 percent of GDP to 10.7 percent while net 3.5 percent deficit in FY2015/1614. Total public lending, at 1.6 percent, was slightly higher than spending was lower by 0.2 percentage points the previous year’s 1.4 percent. Almost 70 percent of GDP, and total revenues were lower by 1.7 of total net lending for the year was for Rwandair. Key Points: Although the headline deficit in FY2016/17 was higher than in the previous year, the decline in grant financing kept the overall fiscal stance unchanged. Measured as a percent of GDP domestic revenues were almost unchanged. Rwanda’s fiscal space narrowed significantly as grant financing declined and public debt rose, which underscores the importance of focusing public investments on programs with the highest economic returns and addressing fiscal contingencies through reform such as the ongoing energy sector reforms. 13 Liquid assets consist of cash, balances held with the BNR, balances due from financial institutions, and trading and other securities. 14 Based on cash accounting. 15 http://www.minecofin.gov.rw/fileadmin/templates/documents/BUdget_Management_and_Reporting_Unit/Budget_Execution_ Reports/2016-2017_Budget_Execution_Report/2016-2017_Budget_Execution_Report.pdf Rwanda Economic Update • Edition No. 11 9 Recent Economic Developments The fiscal deficit is projected to decline to of the 2015-18 Revenue Mobilisation Strategy 4.2 percent of GDP in FY2017/18. Domestic target by an estimated 0.3 percentage points. revenues are expected to come in at 15.7 On the expenditure side, total public spending is percent of GDP, 0.2 percentage points higher budgeted at an estimated 26.8 percent of GDP than in FY2016/17. Sluggish revenue momentum (Rwf 2092.4 billion).16 Recurrent expenditures is mostly explained by the introduction of the are projected at 14.7 percent of GDP, capital new investment code and the Made in Rwanda expenditures at 9.9 percent, and net lending at campaign. Tax revenues are expected to fall short 2.3 percent of GDP. Table 3: Rwanda’s public finances, 2015/16 to 2017/18 2016/17 2017/18 2015/16 Revised Actuals Proj. % of % of GDP budget Rwf billion GDP Rwf billion % of GDP 2014 Revenue and grants 24.4 1,567.1 1,615.8 22.7 22.6 Total revenue 18.4 1,240.5 1,285.6 18.0 18.1 Tax revenue 15.8 1,101.4 1,104.2 15.5 15.7 Direct taxes 6.4 448.6 468.4 6.6 6.6 Taxes on goods and services 8.1 560.3 544.6 7.6 7.7 Taxes on international trade 1.3 92.5 91.2 1.3 1.4 Non–tax revenue 2.6 139.1 181.4 2.5 2.4 Total grants 5.9 326.6 330.2 4.6 4.5 Budgetary grants 3.2 180.6 184.2 2.6 2.3 Capital grants 2.7 146.0 146.0 2.0 2.2 Total expenditure and net lending 27.5 1,897.8 1,945.6 27.3 26.8 Current expenditure 14.7 1,040.0 1,069.7 15.0 14.7 Wages and salaries 3.8 258.3 295.8 4.2 3.9 Purchases of goods and services 2.9 207.4 194.1 2.7 2.8 Interest payments 0.9 78.6 72.2 1.0 1.2 Transfers 4.9 360.3 348.2 4.9 5.2 Exceptional social expenditure 2.2 135.4 159.4 2.2 1.2 Capital expenditure 11.4 741.9 759.5 10.7 9.9 Domestic 7.1 398.0 418.2 5.9 6.0 Foreign 4.3 343.9 341.3 4.8 3.9 Net lending 1.4 115.8 116.4 1.6 2.0 Overall deficit Including grants –3.5 –350.7 –349.8 –4.9 -4.5 Excluding grants –9.5 –677.3 –680.0 –9.5 –9.0 Financing 3.5 350.7 349.8 4.9 4.5 Foreign financing (net) 3.6 356.9 322.6 4.5 4.2 Domestic financing 0.7 –6.1 48.9 0.7 0.3 Errors and omissions –0.7 0.0 –21.7 –0.3 0.0 Source: MINECOFIN. The IMF report (CR17/117)noted that some tax incentives, through eliminating VAT on inputs, were offered to stimulate domestic 16 production of imported goods (textiles and leather products, cement, computers and other office accessories). 10 Rwanda Economic Update • Edition No. 11 Recent Economic Developments Long-term Fiscal Trends 15.1 percent, health which received 13.2 percent and agriculture which received 10.4 percent on Public spending has grown as a share of average during this period (Figure 11). Rwanda’s GDP in the past decade.17 It went up from 22 percent of GDP in 2006 to 30.3 percent in 2014, though it declined to 27.1 in 2016. The Figure 10: Composition of public spending, 2007–16 (Percent of GDP) increase was supported by growth in revenue, 35 which went up by 6.6 percentage points of GDP between 2006 and 2016. Because external grants 30 were declining, in recent years the government 25 has had to borrow more to fund public spending 20 (Figure 9). 15 10 Figure 9: Sources of public spending, 2006–16 (Percent of GDP) 5 35 0 30 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Recurrent Investment 25 Source: MINECOFIN and NISR data 20 15 Figure 11: Public investment by sector, average for 2008–16 (Percent) 10 Energy, 5 Transport, 15.4% 15.1% 0 Social: Education, 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 4.9% Other Tax revenue Non- tax revenue Grants infrastructure Borrowing Public spending sectors, Other, 8.8% 20.6% Source: MINECOFIN and NISR data. Agriculture, The increase in total public spending has 10.4 Social been exclusively for capital expenditures Protection, 2.4% Private -driven and net lending as recurrent expenditures sectors, Other Social: Health, 11.6% 13.2% have held steady (Figure 10). This reflects , sectors 18.1% Rwanda’s public-investment-led growth Source: MINECOFIN data. model in recent years, which gave priority to investments in order to drive growth. In 2013, The headline fiscal deficit has widened since capital spending peaked at 13.2 percent of GDP 2012 as external grants declined. For the last which is very high by international standards. Public investment projects were instrumental in four years, the headline fiscal deficit has been improving Rwanda’s infrastructure, as evidenced relatively stable at about 5 percent of GDP (Figure by the Global Competitiveness Report and the 12). The deficit exclusive of external grants, World Bank Logistics Performance Index.18 however, declined from 14.4 percent of GDP The largest proportion of public investment for in 2013 to 9.4 percent in 2016, highlighting the the period of 2008–16 went to energy with an risks that the government is facing in maintaining average of 15.4 percent. The other key areas for spending programs at historical levels amid public investment were transport which received declining external grants. Fiscal data are on a calendar-year-basis in the Macro-Framework public dataset, December 2016; (data were obtained from MINECOFIN). 17 Africa Competitiveness Report 2017 (http://www3.weforum.org/docs/WEF_ACR_2017.pdf) &International LPI (https://lpi.worldbank.org/ 18 international). Rwanda Economic Update • Edition No. 11 11 Recent Economic Developments Figure 12: Fiscal deficit, 2007–16 assessment the risk of debt distress remains (Percent of GDP) low19—public debt has grown steadily, more 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 than tripling from US$1.2 billion (20.5 percent of 2 GDP) in 2010 to US$3.6 billion (44.4 percent) in 0 0.1 2016 (Figure 14). It is expected that at year-end -2 -0.7 -1.2 -4 -1.7 -1.9 -2.3 2017 the ratio of public and publicly guaranteed -6 -5.1 -5.4 -5.3 -4.3 debt in GDP will reach 46 percent. At year-end -8 2016 external borrowing accounted for about -10 -10.0 -9.4 80 percent of public debt, 35.8 percent of GDP. -12 -10.5 -11.6 -12.6 -12.0 -11.6 Concessional loans accounted for 60 percent -14 -13.1 -16 -13.7 -14.4 of total public debt but the share of commercial Overall fiscal deficit Excl. grant fiscal deficit debt has been growing. Domestic debt rose Source: MINECOFIN and NISR data. from 4.7 percent of GDP in 2011 to 8.6 percent in 2016. Although the stock of domestic debt is The larger deficits and public guarantees much lower than external, interest payments on for various investment projects have meant domestic debt accounted for almost half of total higher public debt. Although Rwanda’s debt interest payments in 2016 because of higher remains sustainable—according to the joint IMF interest rates. and World Bank’s most recent debt sustainability Box 1: Rebasing Rwanda’s national accounts Since 2006 Rwanda has enjoyed large aid inflows but they have been trending down in recent years (Figure B4.1.1). Official aid flows went up from US$605 million (19.2 percent of GDP) in 2006 to a peak of US$1,262 million (about 19.5 percent of GDP) in 2011, but then reversed direction,sliding to US$1,081 million (13 percent of GDP) in 2015. From 2006 through 2015 foreign aid inflows averaged 16.4 percent of GDP. As a percentage of public spending, aid peaked at 46 percent in 2009 before declining steadily to 19 percent in 2016. Between 2006 and 2015 per capita aid almost doubled, from US$66 to US$120, before falling to US$93 in 2015. For 2006–15, it averaged US$92, nearly double the average for all low-income countries (Figure B5.1.2). The official aid was mostly channelled through the public sector, with the bulk of it going to social sectors (Figure B5.1.3). According to OECD data, almost 70 percent of the official aid was channelled through the public sector during 2007-15, up from 10.5 percent in 2002-06. This reflects increasing donor confidence in Rwanda’s management of its finances—the progress of which is documented in the 2016 Public Expenditure Framework Assessment (PEFA). The PEFA report concluded that “most aspects of the Public Financial Management system are functioning at a satisfactory level – one that should allow the government to attain its fiscal and budgetary objectives.” a Grants played an important role in funding public spending but they have been declining recently. Total grants averaged 9.2 percent of GDP 2006–16. In 2016, total grants were 5.1 percent of GDP, of which capital grants amounted to 2.2 percent of GDP. 19 Staff Report for the 2017 Article IV Consultation, Seventh Review Under the Policy Support Instrument, and Second Review Under the Standby Credit Facility- Press Release; Staff Report; and Statement by the Executive Director for Rwanda, IMF 2017. a https://pefa.org/sites/default/files/RW-May17-PFMPR-Public%20with%20PEFA%20Check%282016%20Framework%29.pdf 12 Rwanda Economic Update • Edition No. 11 Recent Economic Developments Figure B4.1.1: Share of official aid in GDP/budget Figure B4.1.2: Aid flows for Rwanda, 2006–15 25 50 US$ US$ million 46 45 150 1,400 44 44 42 43 40 130 1,200 20 32 110 1,000 29 30 26 90 800 15 21 19 20 70 600 50 400 10 10 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 30 200 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Share of public expendigure (right axis) %GDP Aid flows, right axis Per capita aid Figure B4.1.3: Sectoral distribution of aid Figure 13: Grant classification (Percent) (Percent of GDP) 2.8 2.1 2.3 15 12 19.1 3.6 3.6 25.5 3.7 9 3.5 30.7 69.3 3.5 5.5 4.7 6 4.3 5.0 8.2 3.6 7.6 7.7 2.2 6.4 7.0 3 4.2 4.7 1.1 16.5 3.4 2.8 2.7 2.9 0 Other channels Public, Social Sectors Public, Infrastructure sectors Public, Production sectors Public, Multisector Public, Programme assistance 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Public, Action relating to debt Public, Unspecified Budgetary grants Capital grants Source: MINECOFIN, NISR, OECD and WDI data’ Source: As Rwanda’s indebtedness has risen, its Figure 14: Public debt, 2007–16 (Percent of GDP) fiscal space has narrowed. Fiscal space is defined as the “room in a government’s budget 50 that allows it to provide resources for the desired 40 8.6 purpose without jeopardizing the sustainability of its financial position or the stability of the 30 8.3 7.1 economy.”20 It can be measured in “tax years” as 6.2 the ratio of the public debt stock to the tax base.21 20 8.8 5.7 5.9 6.5 4.7 5.4 35.8 For Rwanda, the number of tax years went up 23.2 27.2 10 21.3 17.4 from 1.7 in 2012 to 3 in 2016, clearly signalling a 15.1 13.9 14.1 14.0 16.1 narrowing of the fiscal space (Figure 15). 0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 External Public Debt Domestic Public Debt Source: MINECOFIN and DSA data. Heller, P. S., (2005). “Understanding Fiscal Space.” IMF Policy Discussion Paper PDP/05/4, International Monetary Fund, Washington, DC. 20 Aizenman, J., and Y. Jinjarak, (2010). “De Facto Fiscal Space and Fiscal Stimulus: Definition and Assessment.” NBER Working Paper 21 No.16539, National Bureau of Economic Research, Cambridge, MA. Rwanda Economic Update • Edition No. 11 13 Recent Economic Developments Figure 15: Tax years needed to repay public debt, higher indebtedness. It is important to monitor 2007–16 such risks and take necessary reform actions 4 to address them. In this context, the reform program in the energy sector recently adopted by the authorities is an important step forward 3.0 3 for Rwanda. The program, which is supported 2.5 by a Development Policy Operation of the World 2.2 2.1 2.1 Bank, aims at fiscally sustainable expansion of 2 1.8 1.7 1.7 1.7 1.7 electricity services by containing fiscal impact of the electricity sector and improving the operational 1 efficiency, affordability, and accountability of 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 electricity services.23 If fully implemented by 2019, Source: MINECOFIN and WDI data. this multi-year program will reduce the fiscal risks of energy sector expansion. The recent investment push has narrowed the fiscal space. Contracting debt for well- designed public investments can in the long term 1.6 Outlook be debt-neutral because investments heighten In 2017 growth, at about 5 percent, will be the potential for growth, expanding the capacity well below the historical average for Rwanda to service debt. The large public investments of but is projected to accelerate in 2018 to 5.9 recent years have not yet achieved the desired percent and return to a growth trajectory impact on Rwanda’s growth potential. The close to 7 percent in 2019 as investment slowdown in total factor productivity22 during the activity, both public and private, accelerates period of the government’s investment push of and agriculture becomes more productive 2012-2015 suggest that the economic returns on (Table 4). Economic activity will also benefit from recent investments may be lower than expected. the expected recovery of prices of such traditional This may undermine Rwanda’s growth prospects exports as minerals, tea, and coffee. A more and raise concerns for fiscal sustainability. The competitive exchange rate is already supporting narrowing fiscal space in turn points to the non-traditional exports, which may become importance of adopting a balanced approach an important source of growth. As the region for public policy, ensuring that public investment recovers from drought, the agriculture outlook for is focused in areas with high economic returns the medium term is positive. The government’s while also crowding-in private investments. This renewed commitments to scale up investments in is important for sustaining growth and maintaining agriculture, especially irrigation, will reinforce the the necessary fiscal space, over the near-term as medium-term outlook. Construction of the new well as over Rwanda’s ambitious Vision 2050. airport will boost construction in 2018–19. Fiscal contingencies may also undermine The main risks to the growth outlook are Rwanda’s fiscal space and debt sustainability. associated with possible weak external Fiscal contingencies may arise from weak financial environment, persisting external imbalances, performance of publicly owned enterprises continued debt accumulation, and a weak (such as utilities) as well as from various forms private sector. If global prices of minerals, of public private partnerships and may lead to coffee, and tea continue to be low, they will 10th Rwanda Economic Update. 22 Rwanda Energy Sector Development Policy Loan. http://projects.worldbank.org/P162671?lang=en 23 14 Rwanda Economic Update • Edition No. 11 Recent Economic Developments Table 4: Rwanda’s medium term macroeconomic projections 2014 2015 2016 2017 f 2018 f 2019 f Real GDP growth, at constant prices 7.6 8.9 5.9 5.2 5.9 6.8 Agriculture 6.7 5.0 3.9 5.1 5.0 5.0 Industry 11.0 8.8 6.7 4.8 5.0 6.0 Services 6.9 10.5 7.1 5.1 6.4 8.0 Inflation (Consumer Price Index) 1.8 2.5 5.7 5.7 5.0 5.0 Current Account Balance (% of GDP) -12.0 -13.4 -14.4 -11.7 -12.4 -11.7 Fiscal Balance (% of GDP) -5.3 -5.2 -4.6 -4.9 -5.0 -5.2 Debt (% of GDP) 38.7 36.4 44.5 45.7 47.1 48.7 Primary Balance (% of GDP) -4.2 -3.9 -3.4 -3.5 -3.6 -4.7 Sources: World Bank, Macroeconomics and Fiscal Management Global Practice, and Poverty Global Practice. Notes: e = estimate, f = forecast. (a) Calculations based on 2009-UNHS and 2012-UNHS. (b) Projection using point-to-point elasticity (2009-2012) with pass-through = 1 based on GDP per capita in constant LCU. “ (c) Actual data: 2012. Nowcast: 2013 - 2016. Forecast are from 2017 to 2019 subdue production and exports. Any delayed term economic growth will depend on Rwanda’s exchange rate adjustment may also affect ability to improve productivity through better incentives to invest in the nascent non- use of capital and human resources. Among traditional export sector. The pace of economic other important determinants such as improved transformation will largely depend on the extent business environment, skills, good governance to which government expectations for its major and infrastructure, well-managed urbanization tourism and connectivity investments materialize. can greatly contribute to productivity through Continued lack of private sector response to the agglomeration, density and improved connectivity. improved investment climate is another risk. There is evidence that urbanization in Rwanda is already positively associated with jobs and The special topic of this update examines poverty reduction. This highlights the importance the role of urbanization as one of the of policy discourse on urbanization as part of important factors facilitating growth and formulating country’s growth agenda. poverty reduction in the long term. Long Rwanda Economic Update • Edition No. 11 15 PART TWO RETHINKING URBANIZATION IN RWANDA: FROM DEMOGRAPHIC TRANSITION TO ECONOMIC TRANSFORMATION 24 24 Findings and analyses presented in this special topic draw from the analytical work conducted by the urban and poverty teams at the World Bank (2017), Reshaping Urbanization in Rwanda: Economic and Spatial Trends and Proposals. It consists of four notes: (i) Urbanization and the Evolution of Rwanda’s Urban Landscape; (ii) Internal Migration in Rwanda; (iii) Urbanization, Job Creation, and Poverty Reduction in Rwanda; and (iv) Profiling Secondary Cities in Rwanda – Dynamics and Opportunities. Rwanda Economic Update • Edition No. 11 17 Rethinking Urbanization In Rwanda 2.1 Introduction capita GDP. In fact, few countries have reached upper-middle-income status without substantial For Rwanda, urbanization has been an urbanization (Figure 16 (a)). Urbanization explicit developmental strategy. In its vision contributes to economic development by to become a middle-income country by 2020, facilitating the process of structural transformation, the Government of Rwanda has identified whereby workers move out of agriculture to more urbanization and creating off-farm jobs as central productive sectors, thereby helping raise national to economic growth and national development. average GDP per worker. Rwanda’s economic Accordingly, the country’s Economic Development development in is consistent with the international and Poverty Reduction Strategy II (EDPRS) for trends: its rapid GDP growth was accompanied 2013–18 has as a priority the transformation of by a shift away from agriculture, suggesting Rwanda’s economic geography, to be achieved by structural transformation is taking place in the facilitating urbanization and promoting secondary country (Figure 16 (b)). Urbanization has played cities as centers of nonagricultural economic a positive role in this. Even under conservative activities. Among goals set were to transition 50 assumptions that urban workers earn on average percent of the population from farm to off-farm twice the rural wage for similar work in similar jobs and reach an urbanization level of 35 percent. sectors of employment, urban areas are likely Those priorities are likely to continue into the next to have accounted for about a third of national generation of development strategies, consisting structural change and almost a half of national of Vision 2050 and the National Strategy for economic growth over the past 15 years.26 Transformation I (NST) I for 2017–24. In order to reap the benefits of urbanization, With the vision of embracing urbanization, the government has made significant efforts Rwanda recognizes the positive relationship in a concentrated period to lay a foundation between urbanization and economic for urban development. Beginning in the development that has been observed globally: early 2000s, the necessary laws, policies, and the higher the urbanization rate, the higher per institutions were improved or put in place; Figure 16: Urbanization and per capita GDP (a) and agricultural employment (b) in benchmarked countries and Rwanda A. Level of urbanization and income B. Urbanization and agricultural share of employment 100 100 90 90 Agricultural share of employment (%) 2002 (%) 80 80 Urban share of populationn 2012 70 70 60 60 50 50 40 40 30 30 20 20 10 10 500 1,000 10,000 25,000 50,000 10 20 30 40 50 60 70 80 90 100 GDP per capita (PPP, constant 2011 $) Urban share of population (%) All countries, average 2000−2016 China, 1980−2015 All countries Korea, Rep. China Vietnam Rwanda Korea, Rep., 1960−2016 Vietnam, 1969−2004; 2011−2015 Rwanda, 2002 and 2012 Source: Staff calculation based on World Development Indicators 2016. 25 Background paper for the Urbanization Chapter of the ongoing Rwanda Future Drivers of Growth Study (jointly conducted by the Government of Rwanda and the World Bank). 26 Huye, Muhanga, Musanze, Nyagatare, Rubavu, and Rusizi. 18 Rwanda Economic Update • Edition No. 11 Rethinking Urbanization In Rwanda and investments were made, particularly in 2.2 Urbanization Trends and Forms six designated secondary cities.27 During the period under the EDPRS II alone, policies such The share of urban population in Rwanda is as the National Urbanization Policy (2015), the increasing faster than official records suggest National Housing Policy (2015) and the National According to official statistics, Rwanda is Informal Urban Settlement Upgrading Strategy 16–17 percent urbanized. The 2012 census (2017) were adopted among others. Notably, found urbanization in Rwanda—the share of the the National Urbanization Policy (NUP) provides national population living in urban areas—to be an overarching guidance on urban development 16.5 percent in 2012 Census; and the figure as through four pillars that are central to building measured by the fourth Integrated Household productive and livable cities – Coordination, Living Conditions Survey (EICV4) in 2014 is 17.3 Densification, Conviviality and Economic Growth. percent. Between the 2002 and 2012 censuses, In addition, the Rwanda Spatial Development the level of urbanization in Rwanda does not differ Framework was prepared (currently under much despite the readily observable changes on revision) to orient the implementation of the NUP, the ground towards further urbanization. This is by offering an analysis of the real and desired spatial development situations, with focus on the mainly because the two censuses define “urban specialization, spatial coordination and linkages areas” differently,27 which makes it difficult to between urban areas. evaluate what is trending in the urban population. This special topic report examines the An analytical experiment using a definition relationship between urbanization and nonfarm of urban based on settlement population job creation, as a contributor to economic size and density demonstrates that large transformation, and poverty reduction, as scale urbanization has already taken place in a common outcome of urbanization. First, Rwanda. When a simple definition–a minimum Rwanda’s trends in terms of the changes in population of 5,000 and a population density of urban population, density, and built-up areas greater than 1,000–is applied to a population map are analysed, as is how internal migration gridded by 1km2 cells to identify urban settlements, contributes to the urbanization process. Second, the level of urbanization increased from 15.8 the urban system in Rwanda is described, with percent in 2002 to 26.5 percent in 2015 (Figure specific attention to understanding the economic 17).28 An analysis of geocoded population data performance of Kigali and six secondary cities and using the same definition found that in the same their potential for enhancing Rwanda’s economic period Rwanda’s urban population increased geography. Third, the discussion explores by almost 2 million people—132 percent— whether and to what extent urbanization—in from 1.49 million to 3.46 million for an average the sense of increased density and enhanced annual rate of 6.7 percent. The spatial analysis connectivity—has resulted in job creation and using this alternative definition also identifies poverty reduction. The report concludes with rapid urban expansion surrounding the capital recommendations for how Rwanda might rethink city of Kigali and along the emerging corridor and reshape its urban strategies. between Rubavu and Musanze, which provides 27 For the 2002 Census, 15 cities were delineated and all settlements within their catchment areas were considered urban, even those that previously would have been classified as rural. The 2012 Census defined urban-rural status based on the smallest administrative entity, the village. Villages that had significant infrastructure (e.g., schools, electricity, banks, and markets) were considered urban. 28 The 2015 WorldPop datasets for Rwanda was used as the population map. The spatial distributions of population were estimated by applying a machine-learning method to the census data, land-cover images, and various other datasets. See http://www.worldpop.org.uk/ data/summary/?doi=10.5258/SOTON/WP00223. This is the only place where the Worldpop data is used; other analyses are based on official NISR data. Rwanda Economic Update • Edition No. 11 19 Rethinking Urbanization In Rwanda Figure 17: Evolution of urban settlements in Rwanda, 2002 to 2015 A. 2002 Source: Staff calculation based on WorldPop data B. 2015 Source: Staff calculation based on WorldPop data 20 Rwanda Economic Update • Edition No. 11 Rethinking Urbanization In Rwanda an empirical basis for the trends observed on dominant form of internal population movements the ground. Clearly, this analytical exercise in 2011-2014 (34 percent). It is also important to intensifies the need to refine what is an urban note that currently, internal migration is defined area to better reflect the reality of density and as moving from one district to another; changes population growth in Rwanda. of residence within districts are not considered internal migration and captured in the data.30 Rwandans have moved toward density for economic reasons as well as in search of land The Districts of Kigali and the Eastern The combination of natural growth and Province were the main destinations for increased migration to urban areas has recent internal migration, representing both a contributed to urbanization. In 2005–15 move toward economic density and a move Rwanda experienced impressive declines in in search for land, respectively. Nearly one both infant mortality and fertility,29 but rather third (29 percent) of recent migrants headed unusually, between 2010 and 2015 urban fertility to Kigali; the districts of Gasabo and Kicukiro rates increased, indicating that relatively more were especially popular destinations (EICV4).31 children are being born in urban than in rural Another 33 percent headed to the Eastern areas. Moreover, in 2014 about 27 percent of Province, mainly to the districts of Nyagatare, the population in Kigali and 14 percent in the Gatsibo, Kayanoza, and Rwamagana. This secondary cities were recent migrants, compared represents a dual migration pattern in Rwanda: to 7 percent in rural areas. Of all internal migration a move toward density, with districts of Kigali city between 2011 and 2014, 20 percent went from attracting many migrants, and a parallel move rural to urban areas (Figure 18). However, rural- away from density, with a high share of migrants to-urban migration is not the most common type flocking to the Eastern Province – the least of migration; rural-to-rural migration remained the densely populated province in Rwanda. Figure 18: Rural-to-urban migration in Rwanda, the share and the scale 9% of people moved between 2012 and 2014 Among them, 20% moved from in Rwanda rural to urban areas 20% 9% 80% These rural-urban migrants account for 14% of the urban residents as of 2014 14% 91% 86% Rwanda population 2014 Rwanda urban population 2014 Source: Staff calculation based on EICV 2014. 29 For instance, infant mortality rates (number of deaths of children under 1 year of age per 1,000 live births) decreased from 69 to 32 in urban areas and from 108 to 44 in rural areas. 30 Both rural-to-urban migration and urban-to-rural migration within districts are not counted as part of the population movement. For details on methodology and results, see World Bank. 2017. “Rwanda Economic Geography and Urbanization. Note 2: Internal Migration in Rwanda.” 31 Among three districts in Kigali City, Kicukiro and Gasabo Districts, which still included significant peri-urban and rural areas with less expensive land, had more inflow than Nyarugenge District. Rwanda Economic Update • Edition No. 11 21 Rethinking Urbanization In Rwanda Migration to secondary city districts has been The current planning regime contributes to low- rather limited. People in Rwanda are far more density expansion in urban peripheries likely to move to the capital than to a nearby Urban population growth has been secondary city district, except Rubavu District. accompanied by rapid expansion of built-up Of secondary city districts,33 only Nyagatare and areas, particularly in the periphery and fringe Rubavu districts had positive net migration rates, zones of Kigali. Because the definitions of urban indicating that more people came into than left area used in the 2002 and 2012 censuses differed, between 2002 and 2012. Although current data satellite imagery was used as an alternate source do not capture migration from rural areas to of data to evaluate changes in the amount of built- secondary cities within the same districts, that up land in Rwanda—land covered by man-made, other secondary city districts saw negative net impervious surfaces, such as roofs, roads, and migration rates explains slower than average other infrastructure. It produces a good proxy for population growth in those secondary cities (see urban expansion. Figure 19 shows that over the Table 4). People arriving in the secondary cities are years the periphery and fringe zones of Kigali have also mainly from within their provinces and from experienced the most extensive expansion, with Kigali City.34 For instance, of all recent migrants in significant growth of built-up areas in the core city Huye, 60 percent came from other places within districts of Nyarugenge, Kicukiro, and Gasabo. In the Southern Province and 12 percent from Kigali neighboring Rulindo, Kamonyi, and Rwamagana, (EICV4). In contrast, Kigali attracts migrants from built-up areas have also increased noticeably. all over the country. Furthermore, calculations While this suggests an urban expansion trend, based on census data found that for 27 of the some of these areas will still be predominantly 30 districts, the share of migrants leaving the rural in terms of settlement pattern and economic district to go to Kigali is higher, and often far activities, with relatively low-density settlement higher, than the share going to a secondary city. and agriculture-based economies. The exception is Rubavu, which draws the bulk of migrants g from the neighboring districts of Spatial expansion can also be seen in Rutsiro and Nyabihu. This confirms the status secondary cities, and the rate is generally of Rubavu as the only urban center other than outpacing the observed population growth Kigali, which has significant appeal. rates. Analysis of the multi-temporal spatial Figure 19: Built-up areas, 2002 and 2012 Source: World Bank analysis of satellite images. 32 Since the available data captures only inter-district migration, the report refers to the districts where secondary cities are located as ‘secondary-city districts’. 33 In addition to being the main destination of internal migrants, Kigali is also the main source of migrants moving elsewhere. Of all recent internal migrants, 27 percent came from one of the three districts in Kigali (EICV4). 22 Rwanda Economic Update • Edition No. 11 Rethinking Urbanization In Rwanda expansion of secondary cities and three fast- kilometer [km2]), average density of the 416 growing towns34 revealed that all have expanded sectors went up from 725 persons per km2 in their urban footprint in the past decade.35 In the 2002 to 878 persons per km2 in 2012. Population more compact or spatially constrained cities, density is relatively high in Kigali and the six such as Rubavu and Muhanga, spatial expansion secondary cities and their surrounding areas. is more in line with population growth. The largest Around the secondary cities, however, density spatial expansions are in the cities more recently tapers off about 10 km from the center and established, notably Nyagatare and Kayonza, increased only slightly in those sectors after 2002 and those unconstrained by physical barriers to (Figure 21). Similar patterns are observed for the growth, such as Huye. Figure 20 shows the results built-up ratio—the percentage of area covered by of the Landsat data multi-temporal analysis for manmade surfaces (that is, built-up area), such Musanze, Muhanga, and Huye between two time as roofs, roads, and other infrastructure. While points when imagery is available. Musanze, for the overall built-up ratio is rather low in any instance, expanded over 120 percent between case, except in Rubavu and Musanze, it drops 1999 and 2015, at an annual average of 5.95 steeply beyond the 5 km-radius from the core of percent; between 2002 and 2012 the population secondary cities (Figure 22). Combined with the growth rate was 3.3 percent. level of sprawl, this raises the urgency of better managing development of peripheral settlements People and infrastructure are concentrated and increased density, without which it becomes in the core of secondary cities but dispersed difficult and expensive to provide a centralized outside the core. Using sector-level population network of infrastructure. density (the number of persons per square Figure 20: Urban development in Musanze, Muhanga and Huye: Base year (red), latest year (yellow) A. Central Musanze, 1999 and 2015 B. Central Muhanga, 2001 and 2015 C. Central Huye, 2002 and 2015 Total expansion, 1999–2015: 120% Average annual Total expansion 2001–2015: 81% Average annual Total expansion 2001–2015: 81% Average annual expansion: 6% expansion: 4.4% expansion: 4.4% Source: World Bank. 2017. “Rwanda Economic Geography and Urbanization. Note 4: Profiling Secondary Cities in Rwanda – Dynamics and Opportunities. 34 Nyamata, Gicumbi, and Rwamagana/Kayonza have grown quickly in recent years and thus are included in the analysis where relevant. 35 For methodological details, see World Bank. 2017.”Rwanda Economic Geography and Urbanization. Note 4: Profiling Secondary Cities in Rwanda – Dynamics and Opportunities.” 36 In the analysis, the distance between two sectors is calculated either as Euclidian distance between their centroids or as generalized travel costs Rwanda Economic Update • Edition No. 11 23 Rethinking Urbanization In Rwanda Figure 21: Population density and distance to the city core in secondary cities Huye Nyagatare 8 10 8 10 6 6 4 4 2 2 0 0 0 5 10 15 20 0 5 10 15 20 Muhanga Rusizi 8 10 8 10 6 6 4 4 2 2 0 0 0 5 10 15 20 0 5 10 15 20 Rubavu Musanze 8 10 8 10 6 6 4 4 2 2 0 0 0 5 10 15 20 0 5 10 15 20 Source: Staff calculation based on 2002 and 2012 Census. Note: In the left panel, the x-axis indicates distance from the center of the secondary city, and the y-axis indicates population density (hundred persons per km2). Blue dots indicate sectors in 2002 and red triangles sectors in 2012. Red lines and blue lines indicate smoothed trend lines in 2002 and 2012. In the right panel, a ring indicates a radius of 20 km from the center of the secondary city. Figure 22: Built-up ratio and distance to the city core in secondary cities Huye Nyagatare 40 40 30 30 20 20 10 10 0 0 0 5 10 15 20 0 5 10 15 20 Muhanga Rusizi 40 40 30 30 20 20 10 10 0 0 0 5 10 15 20 0 5 10 15 20 Rubavu Musanze 40 40 30 30 20 20 10 10 0 0 0 5 10 15 20 0 5 10 15 20 Source: Staff calculation based on 2002 and 2012 Census Note: On the left panel, the x-axis indicates distance from the center of the secondary city, and the y-axis indicates built-up ratio (%). Blue dots and red triangles indicate sectors in 2002 and 2012, respectively. Red lines and blue lines indicate smoothed trend lines in 2002 and 2012, respectively. On the right panel, a ring indicates a radius of 20 km from the center of the secondary city. 24 Rwanda Economic Update • Edition No. 11 Rethinking Urbanization In Rwanda The system of zoning in Rwanda may among urban residents. It may also be a concern contribute to low-density development in in other urbanizing areas to which master urban peripheries. Urban areas that are under planning and zoning are applied. In some of these new local plans become subject to both zoning areas, most people work in agriculture and live in requirements and new forms of property taxation. informal housing, but zoning regulations applied Both these factors function as incentives to in the context of master planning will likely insist develop property just outside formal urban areas. on formal urban developed plots. In that case As the zoning standards limit both residential existing buildings may be formalized only if they settlement and commercial activity through conform to building regulations, which is unlikely building regulations, minimum plot sizes, and or may take time and resources for the majority maximum plot coverage ratios, this is likely to of urban dwellings. lead some investors, and thus economic activity more generally, to areas where zoning has not 2.3 Rwanda’s Urban System and yet been completed or codes are less rigidly Economic Geography enforced. The trend is visible in some districts Rwanda’s urban system is dominated by surrounding Kigali where rapid urbanization and Kigali, though part of the urban population urban expansion are occurring; it is also reflected is clustered in the six official secondary in the movement of population toward the fringes cities. According to the 2012 Census, Kigali of Kigali City. Of all the recent migrants in sectors accommodates nearly half of the country’s urban bordering Kigali City, 43 percent came from the population and is seven times larger than the next city itself. Considering that education is one urban centre, Rubavu. The six secondary cities common element in all migration patterns, in account for just 22 percent of the total urban which the more educated people are, the less population (Table 5). The three fast-growing likely they are to move out of urban areas, this towns account for 5 percent of the total urban implies that people without the education and population and their growth rates are generally skills to afford living in a city are pushed to the higher than those of secondary cities, although primarily rural fringes. from a very low base. This leaves 24 percent of the urban population resident in small settlements Furthermore, zoning regulations can in some of fewer than about 15,000 people. Rwanda’s cases function as a deterrent to inclusive cities are also growing at substantially different and efficient urban development. Most land in rates. With an average annual growth rate of 3.3 Rwanda, at least in theory, is individually owned since the Land Tenure Regularization (LTR) percent, the six secondary cities are growing program began. However, much is being zoned more slowly than the national urbanization rate after the LTR process, which means that it can of 4.1 percent, which is largely influenced by the depend on luck of whether the zoning applied rate of urbanization in Kigali (4.2 percent). to a plot suits the needs and capacities of the owner. The disconnect between zoning and Kigali is also the economic center of Rwanda. current land use may also require resettlement Nightlights-based estimation suggests that the and redevelopment and may increase the risk of three districts of Kigali City, which represent 11 pushing marginalized groups further out of the percent of the total population and less than 3 urban core. There are indications that this could percent of Rwanda’s total land area, in 2012 have been a problem in Kigali, in terms of creating accounted for 40 percent of the country’s total both uncertainty among investors and insecurity GDP and 61 percent of total nonagricultural Rwanda Economic Update • Edition No. 11 25 Rethinking Urbanization In Rwanda Table 5: Rwanda’s urban system Share of Growth Rate Population Total Urban 2002-12 Population Kigali 859,332 49% 4.2 Rubavu 149,209 9% 3.7 Musanze 68,930 4% 3.3 Huye 52,768 3% 1.9 Muhanga 50,608 3% 2.1 Rwamagana-Kayonza 39,491 2% 5.6 Gicumbi (Byumba) 34,544 2% 4.1 Rusizi 28,488 2% 1.1 Nyagatare 17,274 1% 7.9 Nyamata 17,076 1% 6.3 Other centers 419,964 24% N/A Total urban population 1,737,684 100% 4.1 Source: Census 2012. Note: Secondary cities in bold. GDP (Figure 23).37 Kigali’s primacy is further growing towns contribute an additional 8 percent illustrated further in comparison to what other (5 percent of total GDP). Kigali also accounts for cities contribute to the Rwandan economy. The 27 percent of all nonfarm jobs created between six secondary cities together account for about 2011 and 2014 and more than 50 percent of all a third of Kigali’s contribution to the national formal private firms and employment in those economy, i.e. 19 percent of total nonagricultural firms (Table 6). GDP (13 percent of total GDP). The three fast- Infrastructure data shows the extent of Figure 23: City population and GDP estimated from luminosity of lights at night, 2012 investment in Kigali relative to the rest of the country, reflecting the priority of the capital city as a regional hub for Eastern and Central Africa. According to the census, in households in Kigali City, 73 percent use electricity, compared with 9-15 percent on average in other provinces; 89 percent have access to improved water sources; and 28 percent have members who have access to the Internet, which is at least seven times the national average. About 7 percent of city households have vehicles, compared to the nationwide average of 1 percent. However, 66 Source: Census 2012; GDP estimates based on night-time lights data from NOAA. percent of the population still lives in dwellings 37 In recent years, the intensity of night lights as measured from space has increasingly been used to estimate economic activity; based on the assumption that as almost all consumption and investment activities in the evening or night require lighting, the intensity of night lights can be used as a proxy for the intensity of economic activity. See Bundervoet, Maiyo, and Sanghi (2015) for details of the methodology; they show that the nightlight approach works in Sub-Saharan Africa, and that in Rwanda the trend of annual GDP and the nightlight-based GDP has been similar over the past 20 years. 38 For the level of infrastructure and service provision in secondary cities, see World Bank 2017. Note 4 above. 39 The three districts along the corridor (Musanze, Nyabihu, and Rubavu) account for one-third of non-Kigali urban population. 26 Rwanda Economic Update • Edition No. 11 Rethinking Urbanization In Rwanda Table 6: Economic activity indicators for Rwandan cities, city and town data, 2014 Share of Nonfarm Formal Nonfarm Formal Economic National Employment Nonfarm Firms Nonfarm Density City Nonfarm Employment Firms (Jobs per GDP Population) Huye 4.1% 9,141 1,713 2,854 232 0.17 Muhanga 2.8% 3,560 1,364 1,326 218 0.07 Musanze 3.3% 6,109 2,538 2,891 242 0.09 Nyagatare 1.8% 5,598 811 2,023 128 0.32 Rubavu 3.7% 12,855 2,993 5,163 341 0.09 Rusizi 3.5% 9,329 1,469 3,317 203 0.33 Gicumbi 2.7% 5,976 1,120 1,498 56 0.17 Rwamagana- 3.6% 4,526 1,116 1,739 167 0.11 Kayonza Bugesera 1.9% 2,769 488 1,155 69 0.16 Kigali 60.8% 113,093 57,260 31,831 4,654 0.13 Rwanda 100% 493,302 106,73 154,236 8,726 - Source: Nighttime lights estimates using satellite imagery from NOAA; employment and firm data from Establishment Census 2014. characterized as unplanned. Since access to When jobs in the informal sector are also infrastructure and services in secondary cities considered, the share of these districts rises to is generally adequate for their current size and 11 percent. growth rate,38 the primacy of Kigali seems to arise from the unique economic opportunities it offers. Together, Rubavu and Musanze form a corridor of urban development in the northwest of the There is some evidence that other economic country; other transnational agglomerations clusters are emerging outside Kigali, with the have potential for vibrant cross-border trade, largest concentration of economic activity particularly in Rusizi. The distance between the in the Rubavu-Musanze corridor.39 Although secondary cities of Musanze and Rubavu is only Kigali is still central to the Rwandan economy, about 50 km, and the sectors along the national between 2002 and 2012 its share of national road that connects both cities have experienced non-farm wage employment dropped from 50 to rapid urbanization since the early 2000s. The 39 percent, and the share of Kigali in total GDP corridor merges with the city of Goma just inside declined from 50 to 40 percent. Outside Kigali, the DRC border. The two Rwandan cities have the cities of Rubavu and Musanze account for 7 also benefited from densely populated fertile percent of nonagricultural GDP, 4 percent of jobs, agricultural land and the presence and investment and 5 percent of firms. They also have a higher of relatively large formal firms, as well as tourism. than average concentration of formal firms and Elsewhere in Rwanda, the southwest is densely jobs, an indicator of more productive economic populated throughout, and certain sectors activity. Of the 106,073 formal private-sector jobs (Bugarama) have urbanized fast, although from a accounted for by firms in the 2014 Establishment low base. This urban cluster could be developed Census, 57,260 (54 percent) are located in Kigali, as a trade and services hub serving the cities and the three districts of Rubavu, Nyabihu, and of Bukavu and Uvira in DRC and Bujumbura in Musanze account for another 7,388 (7 percent). Burundi. Gicumbi town is close to the Ugandan Rwanda Economic Update • Edition No. 11 27 Rethinking Urbanization In Rwanda border at Gatuna and to Kigali as well. The most the census datasets do not include information noteworthy case is Rusizi, which forms part of a about consumption, they have enough information much larger contiguous transnational urban area to calculate the multidimensional poverty status with the city of Bukavu in DRC. Rusizi is home to for each person and, by aggregating individual around 800,000 and 44 percent of its workforce scores, the MPI for each sector. As a measure is employed in wholesale and retail trade—the of the degree of urbanization, population density highest share of any city. and the rate of built-up areas at the sector level are used. Other secondary cities have not reached a sufficient population size or density of firms Urbanization – i.e. population density and to create agglomeration economies. According built-up ratio – is closely associated with to Establishment Censuses, between 2011 and nonfarm job creation and poverty reduction 2014 11 percent of all firms and jobs (16 percent at the sector level in 2002 and 2012. Sectors of formal firms and 12 percent of formal jobs) with higher nonfarm employment shares tend to were created in the six secondary cities. Together, have a lower MPI in each year (Figure 24). The Rubavu and Musanze contributed half of the relationship between nonfarm employment and formal jobs created (6 percent). The majority of population density (and built-up ratio) is not so firms in secondary cities (89 percent) are micro linear. In sectors with a population density of less enterprises with one to three employees—as is than 500 persons per km2 the share of nonfarm also the case nationally (90 percent). Analysis employment is very low; it rises in tandem with of land use also shows that these cities are population density for relatively dense geographic primarily residential and depend heavily on the sectors, those with a population density of 500 rural economy and thus agricultural productivity.40 to 1,500). The share of nonfarm employment While this dependency may cause low-density reaches 80 percent or more in a few high-density expansion of some cities, it suggests that geographic sectors, those with a population strengthening the rural-urban economic linkages density greater than 1,500). Similar relationships around these cities can be one way to release are observed between urbanization and poverty their economic potential. reduction: the higher the population density or built-up ratio, the lower the MPI. 2.4 Urbanization, Job Creation, and Poverty Reduction Increasing density is critical for creating With the varying economic performance of nonfarm jobs and reducing poverty the cities, to what extent has urbanization in Rwanda contributed to nonfarm job creation Increasing density is related to a higher share and poverty reduction since 2002? The of nonfarm employment and greater poverty country’s 416 sectors in 2002 and 2012 Censuses reduction. Even after controlling for changes in are used to analyze the relationship between the educational levels, panel regression analysis hints changes in their population density, the share of at a positive link between changes in population nonfarm employment, and the MPI.41,42 Although density and the share of nonfarm employment. 40 World Bank. 2017. “Rwanda Economic Geography and Urbanization. Note 4: Profiling Secondary Cities in Rwanda – Dynamics and Opportunities.” 41 Rwanda’s MPI is a weighted average of nine indicators: years of schooling, child school attendance, child mortality, electricity, sanitation, drinking water, flooring, cooking fuel, and asset ownership. Households are ‘poor’ if their deprivation score is greater than 33 percent and ‘in severe poverty’ if over 50 percent. 42 Rwanda’s MPI is a weighted average of nine indicators: years of schooling, child school attendance, child mortality, electricity, sanitation, drinking water, flooring, cooking fuel, and asset ownership. Households are ‘poor’ if their deprivation score is greater than 33 percent and ‘in severe poverty’ if over 50 percent. 28 Rwanda Economic Update • Edition No. 11 Rethinking Urbanization In Rwanda Figure 24: Link between share of nonfarm employment and MPI, population density, and built-up ratio, 2002 and 2012 50 MPI and share of nonfarm employment Sectors with nonfarm share < 40% 50 40 40 30 MPI MPI 30 20 20 10 10 0 0 20 40 60 80 100 0 10 20 30 40 Share of nonfarm employment (%) Share of nonfarm employment (%) 2002 2012 2002 2012 Share of nonfarm employment and density Sectors with density < 1500 100 80 Share of nonfarm employment Shere of nonfarm employment 80 60 60 40 40 20 20 0 0 0 5000 10000 15000 20000 25000 0 500 1000 1500 Population density Population density 2002 2012 2002 2012 Share of nonfarm employment and built-up Sectors with built-up ratio < 20 100 100 Share of nonfarm employment Shere of nonfarm employment 80 80 60 60 40 40 20 20 0 0 0 20 40 60 80 100 0 5 10 15 20 Built-up ratio Built-up ratio 2002 2012 2002 2012 Source: 2002 and 2012 Census data. Note: Bars indicate the percentage change in share of nonfarm jobs corresponding to a 10 percent increase in population density at the geographic sector level. Error bars indicate 90 percent confidence intervals. Rwanda Economic Update • Edition No. 11 29 Rethinking Urbanization In Rwanda When sector-level time-invariant characteristics, severe poverty rate (Figure 25 B). Indeed, few observed and unobserved, are controlled for, a 10 sectors have kept poverty rates below 20 percent percent change in population density is linked to unless they had more than 1,000 persons per km2. a 1.4 point higher share of nonfarm employment (Figure 25 (a)). Similarly, increasing density The link between increasing density and relates to poverty reduction. When all sector nonfarm job creation is particularly strong characteristics are controlled, the estimation in Greater Kigali and the cores of the six results of the base specification show that a secondary cities. Figure 26 summarizes the 10 percent increase in population density was regional variations in estimated density effects for associated with a 1.2 percent lower moderate- nonfarm employment share. A 10 percent increase poverty MPI and a 1.6 percent lower severe- in density is associated with a 2.4 percentage poverty MPI. point higher nonfarm employment share within 20 km from Kigali and a 4.6 percentage point higher Importantly, the linkage is stronger in sectors share in areas within 5 km of secondary cities. In with higher density at the baseline (i.e. density contrast, increasing density is less clearly related as of 2002). Sectors that were already dense to nonfarm job creation in areas more than 10 km at the baseline benefit more from increasing from secondary cities, indicating that those areas density in terms of nonfarm employment share: may not be fully integrated into the economy of a 10 percent increase in population density the secondary cities—and that simply increasing was associated with higher shares of nonfarm population (or density) does not automatically employment of 1.2 points for sectors with a lead to job creation. density of less than 500; 2.1 points for sectors with a density of 500 to 1,000; and 1.7 points for Similarly, the estimated density effects for sectors with a density greater than 1,000 (Figure poverty reduction are particularly strong 25 A). Similarly, high density is closely associated within the 5km radius of secondary cities. A 10 with reductions in poverty. In 2002 for sectors with percent increase in density is associated with a a density greater than 1,000, a 10 percent increase 6.0 percent lower moderate poverty rate and a in population density leads to a 3.4 percent lower 6.5 percent lower severe poverty rate within 5 km moderate poverty rate and a 5.6 percent lower from one of the six secondary cities. However, Figure 25: Estimated links between increasing density and nonfarm job creation and poverty reduction by baseline density A. Nonfarm job creation B. Poverty reduction 10% increase in population density is associated with... 10% increase in population density is associated with... Percentage point change in employment share % change in poverty rate 1.38 Rwanda Rwanda 0.54 1.15 Density <500 Density <500 0.46 2.09 Density 500-1000 Density 500-1000 0.56 1.73 Density ≥1000 Density ≥1000 0.83 0 0.5 1 1.5 2 2.5 3 3.5 -9 -7 -5 -3 -1 1 Nonfarm Wage Moderate poverty Severe poverty Poverty without urban components Source: Staff calculation based on 2002 and 2012 Census. Note: Bars indicate the percentage point change in share of nonfarm and wage jobs (panel A) and the percentage change in MPI (panel B) corresponding to a 10 percent increase in population density at the geographic sector. Error bars indicate 90 percent confidence intervals. 30 Rwanda Economic Update • Edition No. 11 Rethinking Urbanization In Rwanda Figure 26: Estimated links between increasing density and nonfarm job creation by geography 10% increase in population density is associated with... Percentage point change in employment share 1.38 Rwanda 0.54 2.38 <20km from Kigali 0.85 2.30 <30km from Kigali 0.88 4.55 <5km from secondary cities 2.00 2.49 <10km from secondary cities 0.94 1.98 <20km from secondary cities 0.74 0 1 2 3 4 5 6 7 Nonfarm Wage Source: 2002 and 2012 Census data. Note: Bars indicate the percentage change in share of nonfarm jobs corresponding to a 10 percent increase in population density at the geographic sector level. Error bars indicate 90 percent confidence intervals. this density benefit does not reach areas beyond market accessibility. As illustrated in Figure 27, the 5 km radius. Once sectors within 10 km of the estimated link between increasing density one of the six secondary cities are included, the and poverty reduction becomes smaller as the link between an increase in population density cost to travel to Kigali becomes higher. Sector- and poverty reduction become unclear (not level market accessibility is measured as the statistically significant at the 10 percent level).44 market accessibility index (MAI) which is the sum of the population accessible from the sector Cities need to be better connected to their within a certain travel time. Calculated in this hinterlands and the regional market way, the MAI indicates how well the sector is The finding of the prominent density effect in connected with areas with large population base Kigali is consistent with the recent analysis of (that is, market size). As shown in Figure 27, a the trend of nonfarm job creation in Rwanda. 10 percent increase in density was associated Between 2011 and 2014 about 30,000 nonfarm with a 1.8 percent lower MPI for the sectors with jobs were created annually in Rwanda, one-third MAI being 20, whereas the sectors with the MAI of which were located in Kigali.45 Larger new firms, being 70 would have the MPI reduced a lot more particularly in the formal sector, have tended to at about 3.2 percent. About 80 percent of the 416 locate in Kigali rather than the countryside, and sectors have a MAI of less than 20. firms already in Kigali have been growing and adding jobs. In more general terms, increasing Access to markets is an important density leads to nonfarm job creation in sectors determinant of secondary city performance, closer to Kigali in terms of lower transport cost, particularly since none of the cities are large and connected to larger populations (i.e., access enough to produce a viable internal market to better markets). for manufactured goods. At the urban core of the secondary cities, nonfarm employment The link between increasing density and accounted for 66 percent but fell with distance poverty reduction, as for job creation, is also from the center, decreasing to 43 percent at 5 km a function of the cost to travel to Kigali and from the city (Figure 28). Even in Greater Kigali, 44 Among the four provinces, the Northern and Eastern Provinces have benefited from increasing density. 45 World Bank. 2016. Rwanda: Firm Growth and Job Creation Study. Rwanda Economic Update • Edition No. 11 31 Rethinking Urbanization In Rwanda Figure 27: Link between increasing density and poverty reduction by cost of travel to Kigali and market accessibility MPI change linked to a 10% increase in density MPI change linked to a 10% increase in density by travel cost to Kigali by market accessibility 0.5 1 1.5 2 2.5 3 3.5 4 4.5 10 20 30 40 50 60 70 80 90 0 0 -0.5 -1 -1.3 -1 -1.2 -1.8 MPI change (%) MPI change (%) -1.6 -1.4 -1.7 -2.2 -1.5 -1.9 -2 -2.5 -2.1 -2.8 -2 -2.3 -3.0 -2.5 -3 -3.2 -3.4 -2.5 -2.8 -3.5 -3 -4 -3.5 -5 -4 -4.5 -6 Travel cost to Kigali Market accessibility index Source: 2002 and 2012 Census data. Note: Y-axis indicates the expected percentage change in the Multidimensional Poverty Index (MPI) when population density increases by 10 percent. Dashed lines indicate 90 percent confidence intervals. The MPI is based on severe poverty. Figure 28: Number and share of nonfarm employment in greater Kigali and secondary cities 350 300 250 Thousands 200 150 100 50 132 164 63 120 40 88 27 60 32 43 40 57 (93%) (97%) (65%) (83%) (24%) (41%) (44%) (66%) (40%) (43%) (14%) (18%) 0 2002 2012 2002 2012 2002 2012 2002 2012 2002 2012 2002 2012 Kigali <5km Kigali 5 - 10km Kigali 10 - 20km 2nd cities 2nd cities <5km 2nd cities 5 - 10km Non-farm Farm Source: 2002 and 2012 Census data. Note: Bar labels indicate the number and share of nonfarm employment. Note: Kigali includes geographic sectors within 30 km of the center of the City of Kigali; secondary cities include sectors within 20 km from the center of secondary cities (Huye, Muhanga, Musanze, Nyagatare, Rubavu, Rusizi). in 2012 the nonfarm employment share was 83 is still a challenge for smaller cities. Kigali has percent within the 5 to 10 km radius, falling to the highest market access in the country, followed 41 percent within the 10 to 20 km radius. This by the border cities of Rubavu and Rusizi. indicates that nonfarm and farm employment Accordingly, more firms have been established coexist within as well as around secondary cities, in Kigali, with Rubavu and Rusizi recording the which can play a greater role in serving the local highest number of new firms outside Kigali (Figure economy if the economic links between them and 29). In general, though, urban areas outside of their rural surroundings can be strengthened. Kigali are poorly connected to markets and will require investment in additional infrastructure, Although the government has taken initiatives particularly in roads. Although some Rwandan to improve transport connectivity, enhancing cities located close to the borders are relatively accessibility to national and regional markets well-connected to neighboring countries, such as 32 Rwanda Economic Update • Edition No. 11 Rethinking Urbanization In Rwanda Figure 29: Market access index and establishment of new firms accessible, with more opportunities to acquire nonfarm jobs and to diversify their economic activities. Recent research supports this in Tanzania (Christiaensen, de Weerdt, and Todo 2013) and India (Gibson, Datt, Murgai, and Ravallion 2017). Further research is needed to see if it applies to a geographically smaller country like Rwanda.46 Policy Discussions and 2.5 Recommendations for Rethinking Rwanda’s Urban Strategy The main findings of this study are these: • Rwanda has been urbanizing more rapidly than national statistics previously Source: World Bank. “Rwanda: Transport Connectivity and Growth Potential.” estimated. The increase in the urban population DRC, Burundi, and Uganda, it will not be easy has been accompanied by the physical to create potential agglomeration economies with expansion of cities, notably in the periphery more distant cities in those countries. Relative of Kigali and around secondary cities, though connectivity, in terms of both time and costs, is mostly at the low level of density. still stronger to Kigali. For instance, Musanze • Rwanda’s urban system is dominated, and Rubavu are better connected to Kigali than both demographically and economically, to Ugandan cities. The distance from Rubavu to by Kigali. Other economic clusters have Kabale is 14 percent further than to Kigali, and emerged, some with noticeable shares of job the transport cost 38 percent higher. creation and employment share. The largest cluster one outside Kigali is along the corridor The economic potential of cities may vary between Rubavu and Musanze. due to their location and endowment; but some recent research emphasizes the role • Although urbanization, measured in terms of secondary cities as drivers of poverty of increasing density and the built-up ratio, reduction. While big cities function as the engines has been correlated to nonfarm job creation for economic growth, the absolute number of and poverty reduction, Rwanda has not yet rural poor who can migrate to those cities tends fully realized its potential. The link is stronger to be limited. Further, the lack of manufacturing in areas with higher population density but and formal jobs in African cities makes economic quickly tapers off with distance from the city survival difficult for migrants, as reflected by cores. To translate density into job creation higher unemployment and underemployment and poverty reduction, market accessibility and and by urban-to-rural migration patterns there. transport connectivity (in terms of travel time The rural poor may find secondary cities more and cost) are equally important. 46 Christiaensen, De Weerdt, and Todo (2013) analyzed a panel of 3,300 individuals in rural Kagera, Tanzania, from 1991 to 2010, and found that a majority of those who escaped from poverty diversified their economic activities in rural areas or moved to secondary rather than big cities. Another study by Christiaensen and Todo (2014) analyzed household surveys of 51 low- and middle-income countries between 1980 and 2014 and found that rural diversification and secondary towns played an important role in poverty reduction. People also migrate to secondary cities as both transit and destination places for jobs and other opportunities: http://blogs.worldbank.org/jobs/secondary-towns- migration-and-jobs-creating-action-space Rwanda Economic Update • Edition No. 11 33 Rethinking Urbanization In Rwanda As Rwanda prepares for a longer-term strategy At the same time, challenges in places of both to reach high-income status by 2050, its origin and destination should be addressed. urbanization strategy should also be revised to As one of most land-scarce economies in Africa align with its ambition. with the region’s highest population density, for Rwanda the expanding rural population will further Rwanda needs to focus on increasing the urban raise pressure on land available for agriculture dividend rather than the urban population. and continue to lower farm sizes. The other side Having a target for urbanization (35 percent in of the policy challenge is to keep people from Vision 2020) as a policy goal is not alone the moving for the wrong reasons – mainly being best guide for urbanization to facilitate economic pushed out of their home areas by adverse development. For example, policies that improve circumstances here. To avoid migration for the basic service delivery in rural areas might reduce wrong reasons, the government should work to migration to cities. This is a positive outcome but eliminate or alleviate the push factors, such as would work against the target for urbanization. As agricultural decline due to pressures of population Rwanda urbanizes further, its policies should be growth or environmental degradation, and lack of reoriented more toward maximizing what cities adequate public services. Migration due to push contribute to national growth and wellbeing rather factors is unlikely to have agglomeration benefits than merely increasing the size or density of the but is likely to exacerbate the urban congestion urban population. To provide empirical support that policy-makers strive to avoid. for such a policy shift, several methodological challenges must first be overcome. Introducing Rwanda’s towns and cities should be a refined and consistent definition of an urban managed as a portfolio. Kigali will continue to area is critical; it is good that the government be the leading economy, with other cities playing is currently revising the definition for the different roles in the national economic geography. next census. To better understand the role of Concerns over Kigali’s size and growth have led migration in urbanization, within-district migration policymakers to promote secondary cities as should also be captured in the new census and substitutes for Kigali. However, businesses are household surveys. likely to continue to locate in Kigali, where they benefit from a range of diverse financial and The policy approach to internal migration related services and diverse product and labor needs reframing, with a view to better markets, not available elsewhere in the country. leveraging the gains from population For other cities to be part of the economic rather movement, rather than simply controlling it. than a demographic strategy (i.e., becoming Agglomeration forces and economic opportunities destinations for population moving away from will inevitably pull workers and families to cities, Kigali), their roles should be defined based on their most likely to Kigali given its primacy in the economic potential and performance. For towns Rwandan economy. Policy concerns about Kigali at the lower end of the urban portfolio in terms becoming a congested and ill-managed city may of size, density, location, and economic vocation, be legitimate but can be addressed through better improvements in the scale of agriculture and urban management. Given the city’s aspiration agro-processing may increase demand for their to drive growth for the country and in the region, goods and services and thus help to strengthen the goal for policy should not be simply to keep rural-urban economic links. people from moving to Kigali but to find best ways to accommodate the flow of people, while making the city more productive. 34 Rwanda Economic Update • Edition No. 11 Rethinking Urbanization In Rwanda Strengthening the linkages between rural and services should be directed to where and urban economies is one way to support endogenous forces are driving growth, which both areas. The government’s role should not will tend to result in better returns on investment. be to direct spatial allocation of resources but to Focusing investment initially on fewer cities that create an enabling environment for investment. already have competitive advantages will deliver There may be short-term potential for some economic growth more efficiently. enhancements to agricultural productivity, but nonfarm activities will certainly provide greater Investment in other cities should focus on future growth prospects. While some rural areas improving basic services and strengthening will transform become more urbanized in their their links with surrounding rural areas. economies and their physical densities, a large In other secondary cities, it is important is to share of employment and nonfarm income maximize the utility of existing infrastructure opportunities will depend on links between rural through adequate operations, maintenance, areas and urban economies, including trade and and where relevant cost recovery, and to ensure migration. One notable example of place- specific that resources for infrastructure and service investments is the plan for building grouped improvement are directed to maximize the return villages as highly-serviced rural settlements for both social and economic development. every 5 to 10km in each sector. The Rwanda Housing Authority’s current plans elaborate 2,235 Densification is critical given that the links grouped villages in 12 districts and proposes between urbanization and poverty reduction grouped villages for the remaining districts. While and job creation quickly taper off with distance it is questionable if the level of services to be from the city cores. The links observed in provided and the number of proposed settlements city centers decline steeply in areas 5−10 km are adequate given the costs of building such away from secondary city cores. Opportunities villages, particularly when finite public resources for leveraging the benefit of urbanization are will have to be divided (e.g., between urban found in these peri-urban areas, which in turn and rural settlements), more fundamentally it is should be better connected to the city cores critical to revisit this plan from the perspective of both physically and economically. Without the national economic transformation in terms of proper land use management and extended the role of these settlements in leveraging rural- services and infrastructure, rapid expansion at urban links. the periphery of Kigali would create negative externalities. Currently, 30 percent of Rwanda’s Investments should be targeted to emerging non-monetary poor live within 20 km of one of economic clusters with greatest potential to the six secondary cities (Figure 30). Extending achieve the scale to meet the demand from economic opportunities to them is vital for their domestic and regional markets. The emerging participation in the peri-urban economy and for economic corridor along the districts of Musanze, reducing poverty in the periphery. Nyabihu, and Rubavu benefit from both fertile agricultural land and proximity to the border with Sectoral efforts to plan and manage urban the DRC and thus are naturally favored by market areas should continue, and efficient urban forces. In comparison, some cities and towns are planning is crucial. Urban population growth too small to generate agglomeration economies, has been accompanied by rapid expansion of reflecting their uneven economic endowment built-up areas in the periphery and fringe zones of and locational advantage. For the short to Kigali and in most secondary cities. Since density medium term, public investment in infrastructure has been shown for having a positive relationship Rwanda Economic Update • Edition No. 11 35 Rethinking Urbanization In Rwanda Figure 30: Nearly one-third of Rwanda’s poor live within 20 km of a secondary city 2,500 2,000 Thousands 1,500 1,000 500 0 2002 2012 2002 2012 2002 2012 2002 2012 2002 2012 2002 2012 2002 2012 2002 2012 Kigali <5km Kigali 5 - 10km Kigali 10-20km Kigali 20-30km Secondary cities Secondary cities Secondary cities Secondary cities 10 - 20km 5km 5 - 10km 10 - 20km Poor Non-poor Source: Staff calculation based on 2002 and 2012 Censuses. with job creation and poverty reduction, and planning regulations. This would recognize opportunities indeed exist in increasing density in the living standards and housing conditions of the these areas that are currently not well connected majority of Rwandans, while preparing for future to the urban economy. It would be worthwhile for growth and responding to growing demands. In the government to revisit current planning and particular, the efficient management of Kigali’s zoning guidelines to synchronize them with actual land market and effective but also pragmatic use development and current demands, and to allow of zoning regulations will be central; here further for a more flexible and participatory use of zoning analysis is needed. 36 Rwanda Economic Update • Edition No. 11 REFERENCES · AECOM International Development Europe. (2017). “Government of Rwanda-Public Expenditure and Financial Accountability (PEFA) Assessment 2016.” https://pefa.org/sites/default/files/RW-May17-PFMPR- Public%20with%20PEFA%20Check%282016%20Framework%29.pdf · BNR. (2017). “Exchange Rate Archives. Kigali.” https://www.bnr.rw/index.php?id=89 · BNR. (2017). “External Sector Statistics. Kigali.” https://www.bnr.rw/index.php?id=123 · BNR. (2017). “Interest rates structure. Monthly Interest Rate, Kigali.” https://www.bnr.rw/index.php?id=88 · BNR. (2017). “Monetary Survey Statistics. Kigali.” https://www.bnr.rw/fileadmin/AllDepartment/Statistics/ MonetaryStatistics/Monetary_Survey_2017.pdf · BNR. (August 2017). “Monetary Policy and Financial Stability Statement. Kigali.” https://www.bnr. rw/fileadmin/AllDepartment/MonetaryPolicy/upload/FinancialStatement/AnnualFinancialStatements/ MPFSS_2017_4_Print_Final.pdf · BNR. (Sept. 2017). “Press Release: Quarterly Financial Stability Committee and Monetary Policy Committee Meetings. Kigali.” https://www.bnr.rw/fileadmin/AllDepartment/services/Public_notices/ doc01560320170927111322.pdf · Bundervoet, Tom, Laban Maiyo, and Apurva Sanghi. (2015). “Bright Lights, Big Cities: Measuring National and Subnational Economic Growth in Africa from Outer Space, with an Application to Kenya and Rwanda.” Policy Research Working Paper No. 7461. World Bank, Washington, DC. © World Bank. https:// openknowledge.worldbank.org/handle/10986/22883 License: CC BY 3.0 IGO. · Christiaensen, L., and Y. Todo. (2014). “Poverty Reduction during the Rural-Urban Transformation: The Role of Missing Middle.” World Development 63: 43–58. · Christiaensen, L., J. De Weerdt, and Y. Todo. (2013). Urbanization and Poverty Reduction: The Role of Rural Diversification and Secondary Towns. Agricultural Economics 44: 435-447. · Gibson, J., G. Datt, R. Murgai, and M. Ravallion. (2017). “For India’s Rural Poor, Growing Towns Matter more than Growing Cities.” World Bank Policy Research Working Paper 7994. · Heller, P. S. (2005). “Understanding Fiscal Space.” IMF Policy Discussion Paper PDP/05/4. Washington, DC: International Monetary Fund. · IMF. (2017). Staff Report for the 2017 Article IV Consultation, Seventh Review under the Policy Support Instrument, and Second Review under the Standby Credit Facility (Vol. Country Report No. 17/217). Washington, DC. http://www.imf.org/en/Publications/CR/Issues/2017/07/13/Rwanda-Staff-Report-for-the- 2017-Article-IV-Consultation-Seventh-Review-Under-the-Policy-45083 · MINECOFIN. (2017). Updated MacroFramework Public Dataset. MINECOFIN: http://www.minecofin.gov. rw/index.php?id=173&L=registration.register · National Agricultural Export Development Board. (2017). Monlty Reports. Kigali. http://www.naeb.gov.rw/ index.php?id=174 · NISR (2014). “Fourth Population and Housing Census Thematic report: Migration and spatial mobility. Rwanda: NISR. · NISR. (2017). Gross Domestic Product (GDP). Featured Reports, Kigali. http://www.statistics.gov.rw/ statistical-publications/subject/gross-domestic-product-%28gdp%29 · NISR. (2017). “Seasonal Agricultural Survey. Kigali.” http://statistics.gov.rw/datasource/seasonal- agricultural-survey Rwanda Economic Update • Edition No. 11 37 · NISR. (Sept. 2017). “Consumer Price Index (CPI). Statistical Reports, Kigali.” http://www.statistics.gov.rw/ publication/consumer-price-index-cpi-september-2017 · NISR and MINICOFIN. (Various years). “Integrated Household Living Conditions Survey (EICV).” Kigali: NSIR. · NISR and MINICOFIN. (Various years). “Demographic and Health Survey (DHS).” Kigali: NSIR. · NISR and MINICOFIN. (2012). “Fourth Rwanda Population and Housing Census.” Thematic Report: Measurement and Mapping of Non-Monetary Poverty. · NISR and MINICOFIN. (Various years). “Establishment Census.” Kigali, NSIR. · OECD. (2015). “Development Aid-Creditor Reporting System (CRS).” http://stats.oecd.org/ · Republic of Rwanda. (2017). “Budget Execution Report for the Fiscal Year 2016/17. Ministry of Finance and Economic Planning (MINECOFIN), Kigali.” http://www.minecofin.gov.rw/fileadmin/templates/ documents/BUdget_Management_and_Reporting_Unit/Budget_Execution_Reports/2016-2017_Budget_ Execution_Report/2016-2017_Budget_Execution_Report.pdf · World Bank Group. (2017). “Global Economic Prospects, June 2017 : A Fragile Recovery.” Washington, DC: World Bank Group. doi:10.1596/978-1-4648-1024-4 · World Bank Group. (2017). “International Logistics Performance Index Dataset (2016).” https://lpi. worldbank.org/international · World Bank Group. (2017). “Rwanda Economic Update, August 2017: Sustaining Growth by Building on Emerging Export Opportunities. Rwanda Economic Update, Edition 10, Kigali.” https://openknowledge. worldbank.org/handle/10986/28111 · World Bank Group. (2017). “World Development Indicators (WDI). World Bank Open Data.” Washington, DC. Retrieved October 2017, from http://databank.worldbank.org/data/reports.aspx?source=world- development-indicators · World Bank. (2016). “Rwanda: Transport Connectivity and Growth Potential.” Kigali: World Bank. · World Bank. (2017). “Rwanda Economic Geography and Urbanization.” Note 2: Internal Migration in Rwanda. · World Bank. (2017). “Rwanda Economic Geography and Urbanization.” Note 3: Urbanization, Job Creation, and Poverty Reduction in Rwanda. · World Bank. (2017). “Rwanda Economic Geography and Urbanization.” Note 4: Profiling Secondary Cities in Rwanda – Dynamics and Opportunities. · World Economic Forum. (2017). “The Africa Competitiveness Report 2017-Addressing Africa’s Demographic Dividend. Geneva: World Economic Forum.” http://www3.weforum.org/docs/WEF_ ACR_2017.pdf Produced by Poverty Reduction and Economic Management Unit Africa Region. Photo credits: © World Bank.