83775 Rwanda Economic Update December 2013 | Edition No. 5 Seizing Opportunities for Growth With a Special Focus on Harnessing the Demographic Dividend THE WORLD BANK GROUP Working for a World Free of Poverty Rwanda Economic Update Seizing Opportunities for Growth With a Special Focus on Harnessing the Demographic Dividend TABLE OF CONTENTS Abbreviations and Acronyms ........................................................................................ i Foreword ................................................................................................................................... ii Overview ..................................................................................................................................... iv I. Recent Economic Developments and Prospects .......................................... 1 I.1 Weakening Growth in the Real Sector ................................................................................... 2 I.2 The External Sector: Narrower Trade Balance, Weaker Terms of Trade ............................... 8 I.3 Monetary Policy, Exchange Rate Policy, and Inflation: Growing Concerns .......................... 13 I.4 Fiscal Policy: Effect of Aid Distress on Development Agenda .............................................. 18 I.5 Economic Outlook and Risks ................................................................................................. 22 I.5.1 Outlook for the Short Term: Tighter Focus on Policies .................................................... 22 I.5.2 Prospects for the Medium Term: Tied to the EDPRS 2 ..................................................... 24 II. Special Focus: Harnessing the Demographic Dividend .......................... 29 II.1 The Demographic Dividend: Why Fertility and Age Structure Matter for Economic Growth ..... 31 II.2 Explaining Rwanda’s Fertility Drop ...................................................................................... 34 II.2.1 Key Evolutions Between 2005 and 2010 ........................................................................ 34 II.2.2 What Drove the Decline in Fertility? .............................................................................. 40 II.3 Rwanda’s Potential Demographic Dividend .......................................................................... 43 II.3.1 Forecasting Income Levels for Rwanda Based on Demographic Projections ................ 44 II.3.2 Getting Policies Right to Seize the Moment ................................................................... 46 II.4 Beyond the Long Term: Preparing for an Aging Population ................................................ 51 References ................................................................................................................................ 57 Appendixes .................................................................................................................................. 59 List of Figures Figure 0.1: Economic growth decelerated in H1 2013 ..................................................................... v Figure 0.2: … pulled down by sharp slowdown in services sector growth ...................................... v Figure 0.3: Export-import coverage improved, as export growth outpaced import growth ............. v Figure 0.4: GDP growth is expected to remain relatively modest in 2013 and accelerate in 2014 ..... v Figure 0.5: Bursting at the bottom in 2010… .................................................................................. vi Figure 0.6: … means thicker in the middle by 2050 ........................................................................ vi Figure 0.7: Rising education and wealth reduced fertility between 2005 and 2010 ......................... vi Figure I.1: Rwanda’s growth trended to the EAC average in H1 2013 ........................................... 2 Figure I.2: … weighed down by weak domestic demand ................................................................ 2 Figure I.3: Slower growth of services pulled down real GDP growth … ........................................ 4 Figure I.4: … leaving other sectors to sustain growth ..................................................................... 4 Figure I.5: Growth slowed in all services subsectors ...................................................................... 5 Figure I.6: Construction continued to drive industrial growth ...................................................... 5 Figure I.7: Agricultural growth reversed from the downward trend in H1 2013 .......................... 7 Figure I.8: Export-import coverage improved as import growth decelerated … .......................... 8 Figure I.9: … leading to a modest narrowing in Rwanda’s trade deficit ...................................... 8 Figure I.10: Rwanda’s coffee is vulnerable to prices and production ............................................. 10 Figure I.11: Rwanda’s terms of trade deteriorated .......................................................................... 10 Figure I.12: … driven by volatile prices of Rwanda’s export products .......................................... 10 Figure I.13: Gross reserves increased in H1 2013 ........................................................................... 11 Figure I.14: Inflation remains moderately low… ............................................................................ 14 Figure I.15: … but there are concerns about rising food prices ...................................................... 14 Figure I.16: Tightened monetary conditions slowed private credit growth ..................................... 15 Figure I.17: Loans to financial services and mortgages contracted sharply .................................... 15 Figure I.18: T-bills rates were higher than deposit rates for 15 consecutive months ...................... 16 Figure II.1: The youth dependency ratio in Sub-Saharan Africa is projected to drop substantially over the next 40 years ............................................................................ 32 Figure II.2: Rwanda’s labor force will grow more rapidly than that of other countries… ............. 34 Figure II.3: … and youth dependency ratios will drop much faster ............................................... 34 Figure II.4: Where has all the child wish gone? ............................................................................. 35 Figure II.5: The odds of surviving childhood in Rwanda increased dramatically since 2000….... 36 Figure II.6: … reducing fertility ..................................................................................................... 36 Figure II.7: Average years of education increased most for the youngest women… ..................... 37 Figure II.8: … as did primary school completion .......................................................................... 37 Figure II.9: Female labor force participation is increasing, especially among younger women ....... 38 Figure II.10: Young women are turning to nonagricultural activities .............................................. 38 Figure II.11: Household wealth increased between 2005 and 2010 ................................................. 39 Figure II.12: Rising education and wealth are reducing fertility ..................................................... 41 Figure II.13: Improvements in education have reduced fertility, especially among women 15–24 ....... 42 Figure II.14: Increasing wealth contributed to the fertility decline in all age groups, especially the young ..................................................................................................... 43 Figure II.15: The move to nonfarm employment reduced fertility, especially among women 15–24 ..... 43 Figure II.16: Bursting at the bottom in 2010…................................................................................. 44 Figure II.17: … means thicker in the middle by 2050 ...................................................................... 44 Figure II.18: A larger share of working-age adults has been associated with higher income levels ...... 45 Figure II.19: The 2012 revision projects larger changes in the age structure of the Rwandan population .................................................................................................... 45 Figure II.20: The drop in fertility has potentially significant implications for economic growth .... 45 Figure II.21: Further rapid reduction in fertility could pay off ......................................................... 45 Figure II.22: A variety of policies is needed if Rwanda is to reap the demographic dividend ........ 46 Figure II.23: In the low-fertility scenario, successive cohorts will be smaller by 2050 ................... 49 Figure II.24: Rwanda’s window of opportunity will remain open until 2070 .................................. 51 List of Tables Table I.1: Fiscal outcomes in FY 2012/13 were better than projected .............................................. 19 Table I.2: Rwanda’s GDP growth is expected to slow in 2013 before accelerating in 2014 ............ 22 Table II.1: The decline in fertility in Rwanda between 2005 and 2010 was greatest among younger women ................................................................................................... 35 Table II.2: Significant changes in the determinants of fertility between 2005 and 2010 ................... 38 Table II.3: Proximate determinants of fertility are moving in the right direction .............................. 40 list of boxes Box I.1: What accounted for Rwanda’s growth in the past decade? ................................................ 3 Box I.2: Setting the basis for a private sector–led growth .............................................................. 6 Box I.3: Rwanda’s capital and intermediate imports slowed ........................................................... 9 Box I.4: Toward self-reliance ........................................................................................................... 11 Box I.5: Beyond gorilla trekking ..................................................................................................... 12 Box I.6: Rwanda’s real effective exchange rate depreciated .......................................................... 15 Box I.7: BNR - Beyond our control? ................................................................................................. 17 Box I.8: Increased capital spending in both absolute and relative terms .......................................... 21 Box I.9: Stairway to the future ......................................................................................................... 21 Box I.10: Rwanda’s experience in managing economic crises and their aftermath .......................... 24 Box I.11: Looking ahead ...................................................................................................................... 25 Box II.1: The demographic transition ................................................................................................ 33 Box II.2: Once again, change is made by youth ................................................................................ 36 Box II.3: Under pressure .................................................................................................................... 41 Box II.4: Boom or bust ....................................................................................................................... 47 ABBREVIATIONS AND ACRONYMS BNR Banque Nationale du Rwanda (National Bank of Rwanda) DHS Demographic and Health Survey EAC East African Community EDPRS Economic Development and Poverty Reduction Strategy EICV Enquête Intégrale des Conditions de Vie des ménages (Integrated Household Living Conditions Survey) FY Fiscal Year GDP Gross Domestic Product IMF International Monetary Fund MINECOFIN Ministry of Finance and Economic Planning NEER Nominal Effective Exchange Rate NPLs Non Performing Loans NISR National Institute of Statistics of Rwanda ONAPO Office Nationale de la Population (National Population Office) PPP Purchasing Power Parity RDB Rwanda Development Board REER Real Effective Exchange Rate Rwf Rwandan franc UN United Nations   i Rwanda Economic Update | Edition No. 5 Foreword T he Rwanda Economic Update reports and synthesizes recent economic developments and places them in a medium-term and global context. It analyzes the implications of these developments and policies for the outlook of Rwanda’s economy. This report attempts to make an analytical contribution to the implementation of Rwanda’s national development strategy. Each edition of the report includes a special feature on a selected topic. The report is intended for a wide audience, including policy makers, business leaders and other market participants, and the community of analysts engaged in Rwanda’s economy. The fifth edition of the Rwanda Economic Update was prepared by the Rwanda Poverty Reduction and Economic Management (PREM) team at the World Bank. The team was led by Yoichiro Ishihara (Senior Economist). Peace Aimee Niyibizi (Economist) led Part I; Tom Bundervoet (Senior Poverty Economist) led Part II. Other team members who contributed to the fifth edition are Jane Bogoev (Economist), Toru Nishiuchi (Economist), and Valence Kimenyi (Economist). Apurva Sanghi (Lead Economist) supervised the team. Diarietou Gaye (Country Director), Carolyn Turk (Country Manager), and Pablo Fajnzylber (Sector Manager) provided overall guidance. The team was supported by Sylvie Ingabire (Team Assistant) and Barbara Karni (Editor). Although this report does not represent the official views of the authorities, the macroeconomic unit of the Ministry of Finance and Economic Planning (MINECOFIN) and the National Bank of Rwanda (BNR) were engaged in the formulation of this report and provided valuable comments. The Bank team appreciates their contributions. The findings, interpretations, and conclusions expressed herein are those of the authors and do not necessarily reflect the views of the World Bank’s Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this report. For more information about the World Bank and its activities in Rwanda, please visit www.worldbank.org/rw. To be included in the email distribution list of this semiannual series and related publications, please contact singabire@worldbank.org. For questions and comments about this publication, please contact Yoichiro Ishihara (yishihara@worldbank.org). Rwanda Economic Update | Edition No. 5 ii Overview R wanda’s economic growth slowed in the first half of 2013. Weighed by a slowdown in domestic demand, the economy grew by 5.9 which accounted for 40 percent of goods exports in the first half of 2013, recorded solid growth as a result of new investments in the sector, which percent, a modest pace relative to the average of helped to widen Rwanda’s export base. 8 percent in the past three years. Decelerating GDP growth mirrored the low growth of services Is the economic slowdown temporary or the and was the lowest half-year growth rate since beginning of further deceleration and what is 2010, when the domestic economy was hard hit the growth forecast for 2014? T by the combination of the global financial crisis he World Bank foresees that growth will and a domestic credit crunch. This edition of the pick up in the second half of 2013 and Rwanda Economic Update examines three key further accelerate in 2014. The May 2013 edition questions: What caused the recent economic of the Rwanda Economic Update projected that slowdown? Is the economic slowdown temporary Rwanda’s economy would grow at 7.0 percent or the beginning of further deceleration, and in 2013 and 7.5 percent in 2014. This projection what is the forecasted growth for 2014? What are reflected the government’s successful efforts to policy options for the authorities? contain the first-round spillover effects of the aid shortfall in 2012. However, the slowdown in What caused the recent economic slowdown? the first half of 2013 was deeper than predicted. T he recent economic slowdown mainly reflected lagged effects of the aid shortfall, which affected domestic demand in the first In light of recent economic developments, the World Bank revised down its growth projections to 6.6 percent for 2013 and 7.2 percent for 2014. half of 2013. Increased domestic borrowing by the government, on one hand, partly offset What are policy options for the authorities? R the aid decline, but on the other hand tightened ecent developments suggest that monetary conditions, reducing the availability of macroeconomic management has become financial resources for the banking sector (mainly more challenging, as policy buffers were for households and private business). Together significantly exhausted. On the monetary with the lower level of public capital spending, front, the recent reduction in the policy interest the credit slowdown led to the contraction of rate provided an important policy signal to the domestic demand. Growth in the services sector economy. However, the impact on economic was particularly affected and sharply decelerated activity is likely to be limited, as the transmission to 4.2 percent in the first half of 2013, down from of monetary policy actions remains weak. The 11.4 percent in the second half of 2012. The decline in aid inflows, for the time being, is likely contraction of domestic demand was partially to constrain the BNR’s ability to meet market offset, however, by higher external demand for demand for foreign exchange. On the fiscal front, Rwandan exports. The trade deficit narrowed Rwanda is adjusting to a lower level of foreign aid slightly, as growth of exports accelerated and and on its way to medium-term fiscal consolidation, outpaced growth of imports. Despite declining by expanding domestic resource mobilization international prices for Rwanda’s major export with a view to increase the fiscal space needed items, traditional exports of goods expanded to implement its medium term development significantly in real terms. The mining sector, agenda. Until domestic resource mobilization Rwanda Economic Update | Edition No. 5 iv Overview Rwanda: recent economic developments in figures Figure 0.2: … pulled down by sharp slowdown Figure 0.1: Economic growth decelerated in H1 2013… in services sector growth Real GDP growth, percent Changes from the previous year, percent 12 16 10.1 10 13.1 8.6 12 11.4 8.2 10.9 7.7 8 8.9 8.9 6.3 5.8 5.9 Percent 6 8 6.9 4 4.2 4 2 0 0 H1 2010 H2 2010 H1 2011 H2 2011 H1 2012 H2 2012 H1 2013 H1 2010 H2 2010 H1 2011 H2 2011 H1 2012 H2 2012 H1 2013 Year-on-year 2010-2013 average Year-on-year 2010-2013 average Figure 0.3: Export-import coverage improved, as export Figure 0.4: GDP growth is expected to remain relatively growth outpaced import growth modest in 2013 and accelerate in 2014 60 Actual and projected yearly GDP growth, percent 9 8.3 8.2 50 8.0 8 7.2 7.2 7 6.6 40 6.2 6 Percent 30 5 4 20 3 10 2 1 0 0 Sep-10 Mar-11 Sep-11 Mar-12 Sep-12 Mar-13 Sep-13 2002-08 2009 2010 2011 2012 2013p 2014p Coverage rate (percent) Exports Growth (Y-o-y) Imports Growth (Y-o-y) (av) Source: NISR, BNR and World Bank staff estimates and the expansion of the export base advance rapid drop comes after a decade and a half of further, the authorities will have to maintain a high and stable fertility, during which Rwanda’s prudent fiscal policy stance by emphasizing population grew rapidly and youth dependency increased efficiency and effectiveness in public ratios (the average number of dependent children spending, in line with available resources, per working-age adult) steadily increased. ensuring timely disbursement of foreign aid, and Rwanda’s fertility rate is now well below the prioritizing expenditures aimed at promoting average for Sub-Saharan Africa of 5.1. the development of the country’s productive sectors. The sharp drop in fertility will have substantial implications for the age structure Turning to the special focus, Rwanda has the of the population. Rwanda’s population is potential to reap a demographic dividend characterized by a youth bulge and a relatively F ertility rates in Rwanda dropped rapidly in the second half of 2000s. The average number of children per woman fell from 6.1 in small labor force (Figure 0.5). About 53 percent of the population is in the labor force (age 15–64); children under the age of 15 make up 42 percent 2005 to 4.6 in 2010, a 25 percent decline. This of the population. Driven by the recent drop in v Rwanda Economic Update | Edition No. 5 Overview fertility, the share of the labor force is projected facilitating higher investment. Smaller families to increase to 64 percent by 2050, and the share also tend to invest more in the health and education of youth (younger than 15) is projected to decline of their children, laying the foundations for a to 28 percent (Figure 0.6). The size of Rwanda’s virtuous cycle of fertility decline, human capital labor force will almost triple, rising from 5.7 accumulation, and wealth creation, labeled the million in 2013 to 15–16 million by 2050. “demographic dividend.” Figure 0.5: Bursting at the bottom in 2010… (Rwanda’s population by age group and gender) Improved education and household wealth are driving a fertility transition I 80+ 70-74 ncreasing levels of female education and 60-64 household living standards are the main 50-54 drivers of the fertility drop. Average educational attainment of women in Rwanda increased by Age group 40-44 30-34 almost a year between 2005 and 2010, explaining 23 percent of the decline in fertility (Figure 0.7). 20-24 Improved living standards at the household level, 10-14 related to rapid growth and poverty reduction 0-4 1,500 1,000 500 0 500 1,000 1,500 over the past decade, accounted for 15 percent of Population (thousands) the decline in fertility. Higher participation in the Male2010 Female2010 labor market, in particular increased engagement Source: UN 2013 in the nonfarm labor market, has also been Changes in the age structure open a window of associated with the decline in fertility. opportunity for rapid and sustained economic growth. As today’s youth grow up, the size and Young women are at the core of the sharp share of the labor force will grow dramatically drop in fertility. Women younger than 25 years in the coming decades, with the potential to experienced the strongest decline in fertility substantially increase economic growth. With between 2005 and 2010. The decline was driven fewer children to support, families will be able mainly by the increase in educational attainment. to save more of their disposable income, thereby The fraction of young women who completed Figure 0.7: Rising education and wealth reduced fertility Figure 0.6: …means thicker in the middle by 2050 between 2005 and 2010 (Rwanda’s projected population by age group and gender) (Contribution, percentage points) 80+ 50 70-74 Higher female labor 60-64 40 5.4 market participation 50-54 Age group 30 15.4 40-44 Increasing household wealth 30-34 20 20-24 10-14 22.7 10 Improving female education 0-4 1,500 1,000 500 0 500 1,000 1,500 Population (thousands) 0 Male 2050 Female 2050 Source: UN 2013 Source: DHS 2005, 2010 Note: Figures show medium-fertility scenario Rwanda Economic Update | Edition No. 5 vi Overview primary school almost tripled between 2005 and and creating a business-friendly environment and 2010, and significantly more young women were a flexible labor market able to accommodate the still in school in 2010 (37 percent) than in 2005 burgeoning working-age population. Although (20 percent). Overall, the improvement in female these policies are not a recipe for success, they education accounts for two-thirds of the drop go a long way toward turning demographic in fertility in young women. Young women are opportunity into economic success. also increasingly taking up nonfarm economic activities, which has also been associated with a Many elements of Rwanda’s current policy drop in fertility. framework are conducive to realizing a demographic dividend. These features include The population shifts set in motion by the strong governance, a stable macroeconomic strong fertility drop bode well for economic framework, relative price stability, strong growth. Before the sharp decline in fertility, investments in education, and a business-friendly real GDP per capita (adjusted for differences environment. The new focus on urbanization is in purchasing power) was projected to double also important, given its benefits in terms of lower between 2010 and 2050. Taking account of the fertility and nonfarm job creation. Important fertility drop, real GDP per capita has the potential challenges remain, however, in particular the to triple by 2050. This means that the recent drop low domestic savings rate, which holds back in fertility has increased the long-run estimate of investment, and the high prevalence of early per capita GDP by 50 percent. Even larger gains childhood malnutrition. Enrollment in secondary are possible if further decreases in fertility can school remains low, though it is expected to rise be achieved. Based on the projected population sharply in coming years, creating a skills shortage shifts, a substantial acceleration of growth is and barring many youth from accessing jobs in expected after 2025, when the labor force will the formal sector. start expanding rapidly and youth dependency ratios will drop. Dependency rates are projected to fall through 2070, which means that the demographic The demographic dividend needs to be earned conditions for rapid growth will persist for A lthough it provides a golden opportunity for rapid economic growth, falling fertility—and the resulting increase in the another six decades. After 2070, the growth in the old-age population will outpace overall population growth, leading to an elderly bulge share of the working-age population—will not and a heavy fiscal burden related to old-age automatically yield a demographic dividend. pensions. From a demographic point of view, Cashing in on the population shifts requires a the coming decades present Rwanda with an conducive policy framework aimed at fostering opportunity to accelerate growth. Realizing the a productive workforce through investments opportunity will require sustaining the sound in child health and post primary education; policy framework and making additional efforts boosting savings, by sustaining price stability and to foster a healthy, skilled, and productive encouraging competition by financial institutions; workforce and spur job creation. vii Rwanda Economic Update | Edition No. 5 PART ONE Recent Economic Developments and Prospects R wanda’s economic growth rate decelerated to 5.9 percent in the first half of 2013, a level similar to that in the East African Community (EAC). The deceleration mainly reflected the lagged effects of the decline in aid Rwanda experienced in the second half of 2012. Following substantial domestic government borrowing, credit growth to the private sector slowed. As a result, consumption contracted and investment demand was feeble, leaving net exports to support growth. On a positive note, the trade deficit narrowed slightly, as growth of exports outpaced growth of imports. However, declines in intermediary and capital imports raise concerns about growth in the near term. I. Recent Economic Developments and Prospects 1.1. Weakening Growth in the Real Sector R wanda’s economic growth decelerated in the first half of 2013, slowing to about the average level in the EAC (Figure I.1). The exchange rate eased, reflecting the resumption of aid inflows, strong exports, and stable imports. The Rwandan franc (Rwf) depreciated by 1.8 previous Rwanda Economic Update noted the percent against the U.S. dollar—a slower rate of prompt and adequate fiscal and monetary policies depreciation than the 3.1 percent in H2 2012. undertaken in the wake of the aid shortfall of Treasury-bill rates declined from 12.4 percent in mid-2012.1 It projected robust annual growth of December 2012 to 10.8 percent in June 2013. The 7.0 percent in 2013 and acceleration to 7.5 percent combined effect of the resumption of aid (more in 2014. In the first half of 2013 (H1 2013), the than half of the delayed financing—US$140 economy started experiencing the lagged effects million—was disbursed) and the issuance of a of the aid shortfall. Year-on-year GDP growth Eurobond increased the stock of international slowed to less than 6.0 percent for the first time reserves with the central bank by US$220 since 2010,after sustaining high GDP growth in million to US$1,069 million, equivalent to 4.4 the second half of 2012 (Figure I.2). months of imports. In H2 2012, the impact of aid shortfall was Despite steady financial market indicators, contained. The shortfall created a fiscal gap of Rwanda’s domestic economy started feeling US$230 million (equivalent to 3 percent of GDP). the lagged effect of the aid shortfall. Low levels The direct effect to the economy was subdued. of aid inflows and heavy government domestic Inflation remained modest, credit to the private borrowing in H2 2012 squeezed the banking sector remained strong, and real annual GDP sector’s room for expanding credit to the private growth was sustained at 7.7 percent, propelled by sector. As a result, credit growth decelerated robust domestic demand. sharply in H1 2013, reducing domestic demand for goods and services. Domestic demand Some financial market indicators improved contracted 1.4 percent in H1 2013 (year-on-year), in H1 2013. The depreciating trend of the reducing GDP growth by 1.6 percentage points. Figure I.1: Rwanda’s growth trended to the EAC average in H1 2013… Figure I.2: …weighed down by weak domestic demand (Year-on-year percentage growth) (Year-on-year growth composition, percentage points) 12 20 10 15 10.1 8 10 8.5 8.6 8.2 7.7 5.8 6.3 5.9 4.1 6 5 4 0 2 -5 0 -10 H1 09 H2 H1 10 H2 H1 11 H2 H1 12 H2 H1 13 H1 2009 H2 H1 2010 H2 H1 2011 H2 H1 2012 H2 H1 2013 Rw UG TZ KE Regional Average Domestic demand External demand GDP growth Source: EAC countries statistics offices and World Bank staff estimates. Source: NISR and World Bank staff estimates Note: For Tanzania, growth rate of the first quarter 1 Following the United Nations’ publication, in July 2012, of a report alleging Rwanda’s involvement in the conflict in the North Kivu Province of the Democratic Republic of Congo, various donors delayed or cancelled their planned aid to Rwanda for FY2012/13. 2 Rwanda Economic Update | Edition No. 5 I. Recent Economic Developments and Prospects Net exports offset the contraction in domestic Contraction of consumption has been a key demand, as growth of exports continued to driver behind weaker domestic demand. All increase, outpacing growth of imports. For the domestic demand components of GDP, including first time in recent history, real GDP growth was investment and particularly consumption, supported entirely by external demand in the exhibited restrained performance compared expenditure account (Box I.1). with the previous half year. Consumption, the Box I.1 What accounted for Rwanda’s growth in the past decade? GDP measures value added produced, final expenditures, or all factor incomes in an economy. The first approach suggests that GDP sums the values added by all producers. The expenditure approach involves counting all final spending on goods and services in the economy: consumption expenditures by households, gross investment, government purchases of goods and services, and net exports (exports minus imports). In the third approach, GDP describes the sum of incomes directly generated by productive activity. Data for the two first approaches are published quarterly by the National Institute of Statistics of Rwanda (NISR). In the past, Rwanda’s growth was driven largely by domestic demand. Between 2001 and 2012, real GDP expanded by 8 percent, driven by large increases in domestic demand, reflecting substantial increases in public spending as a result of larger resource flows from increased aid and higher domestic revenues. Domestic demand grew at an average annual rate of about 8.2 percent; its aggregate contribution to GDP growth was 9.7 percentage points. Household consumption grew by 7.0 a year on average. It was responsible for the bulk of GDP growth, with an aggregate contribution of 5.8 percentage points (equivalent to 72 percent of growth). Investment grew by 13.0 percent a year, accounting for 30 percent of GDP growth. Government consumption rose 8.9 a year and contributed 19 percent to GDP. Counterweighing these factors were net exports, which reduced GDP growth by 1.7 percentage points. Rwanda’s high growth created a domestic economy that is increasingly dependent on nontradable goods. Rwanda has seen growth in manufacturing lag overall growth. Manufacturing growth of 7.2 percent was lower than the overall growth rate of 8.0 percent. Its share in GDP was just 6.7 percent in 2012. The bulk of GDP growth was contributed by nontradable goods and services, which accounted for more than 70 percent of growth. The services sector expanded by 10.2 percent a year in 2000-12, boosted by high growth rates in trade, transport, communication, and financial services, which were liberalized in the early 2000s. Public expenditure–led services (public administration, education, and health) showed robust annual growth of 9.3 percent and contributed 13.9 percent to GDP growth (1.1 percentage points), supported mainly by the increase in public spending. The construction sector expanded by 13 percent a year and contributed 11.7 percent (0.9 percentage points) to economic growth. This increase partly reflected higher public capital spending (Box I.8). Agriculture expanded by 4.9 percent a year and contributed 28.1 percent (2.0 percentage points) to economic growth in 2001-12. Table B1.1.1: In the last decade, Rwanda’s high growth was supported by domestic demand … Annual Growth Rates (percent) Contribution to Real GDP Growth (Percentage points) 2000-12 H1 2013 2000-12 H1 2013 Domestic demand 8.2 -1.4 9.7 -1.6 Consumption 7.3 -3.8 7.3 -3.6 Households 7.0 -5.4 5.8 -4.2 Public 8.9 3.0 1.5 0.6 Investment 13.1 8.6 2.4 2.0 Construction 14.6 21.0 1.9 3.5 Durable goods 9.7 -24.8 0.5 -1.5 External trade 8.9 -39.0 -1.7 7.5 Export 16.5 46.7 1.9 7.4 Imports 11.6 -0.2 3.6 -0.1 Source: NISR and World Bank staff estimates Rwanda Economic Update | Edition No. 5 3 I. Recent Economic Developments and Prospects Table BI.1.2: …that is increasingly dependent on non-tradable goods Annual Growth Rates Contribution to Real GDP Growth (percent) (Percentage points) 2000-12 H1 2013 2000-12 H1 2013 Agriculture 5.2 5.4 2.0 1.7 Food crops 5.3 5.4 1.7 1.5 Export crops 2.7 25.2 0.0 0.2 Industry 9.8 13.6 1.3 2.0 Mining 10.4 13.6 0.0 0.2 Manufacturing 7.2 0.9 0.4 0.0 Construction 12.7 20.4 0.9 1.7 Services 10.2 4.2 4.4 2.1 Trade 11.1 6.8 1.3 0.9 Transport & communication 14.6 6.6 1.0 0.6 Financial services 9.0 9.9 0.3 0.3 Public expenditure led services 9.3 5.5 1.1 0.7 Adjustment 6.3 2.3 0.4 0.1 Memorandum Tradable activities 5.5 4.7 2.4 1.6 Non-tradable activities 10.5 6.5 5.3 4.3 Source: NISR and World Bank staff estimates main growth driver in the past, contracted by 3.8 slowed sharply to 8.6 percent in H1 2013, down percent (year-on-year), reducing GDP growth from 21.5 percent in H2 2012. The slowdown of by 3.6 percentage points (Box I.1). Household investment growth is attributed to contraction of consumption contracted by 5.4 percent (year- investment in durable capital goods. In contrast, on-year) in H1 2013, following expansion of 6.9 construction investments held up. Weak domestic percent in H2 2012. Government consumption demand is reflected in the contraction of imports expanded at a much slower pace (3.0 percent) (0.2 percent in year-on-year terms), down from than in the H2 2012 (11.3 percent), as the effect of 31 percent the preceding half year. the aid shortfall emerged. Growth of investment Figure I.3: Slower growth of services pulled down real GDP growth … Figure I.4: … leaving other sectors to sustain growth (Annual growth, percent) (Contribution to real GDP growth, percentage points) 20 9 8 7 15 6 4.0 5 4.0 5.6 5.1 2.1 10 2.8 4 3 0.2 2.4 2.0 1.2 5 2 2.7 1.1 1.4 1 1.8 1.6 1.7 1.0 0.9 0 0 2009 2010 2011 2012 H2 2012 H1 2013 2009 2010 2011 2012 H2 2012 H1 2013 Agriculture Industry Services GDP growth Agriculture Industry Services Adjustment GDP growth Source: NISR and World Bank staff estimates Source: NISR and World Bank staff estimates 4 Rwanda Economic Update | Edition No. 5 I. Recent Economic Developments and Prospects Weaker growth mirrored the faltering growth 2012, reflecting a decline in mortgage loans. of the services sector (Figure I.4). Services, Financial services grew 9.9 percent, down from which represent 45 percent of GDP, grew at 22.7 percent in H2 2012. Growth in public 4.2 percent in H1 2013 (year-on-year), down expenditure–led services continued its downward from 11.4 percent in H2 2012 (Figure I.3). trend, expanding 5.5 percent, down from 10.9 Strong growth in services has been the main percent in H1 2012 and 7.4 percent in H2 2012. driver behind strong economic growth since 2010, compensating for smaller contributions Growth in industry accelerated, as by agriculture and industry. Moderate growth construction activities remained strong. in services in H1 2013 is attributed to weak Overall industrial growth soared to 13.6 percent public consumption and the slowdown in credit in H1 2013, up from 9.2 percent in H2 2012 expansion to services (Figure I.17). The weaker and 7.2 percent in H1 2012. This performance performance of the services sector in H1 2013 was a result of robust construction activities, reduced its contribution to growth, although especially in nonresidential components, as part it remained the single biggest contributor to of the implementation of the master plan for growth (Box I.1). Strong growth was observed Kigali City. Construction grew 21.3 percent in in industry and agriculture, making them key H1 2013 (year-on-year), contributing more than drivers of growth for the first time since 2010. 85.0 percent to the growth of industry and 29.4 percent to overall real GDP growth (Figure I.6). Growth moderated across all services. Trade H1 2013 was the fourth half year since 2011 in services (retail and wholesale, which together which the construction sector recorded growth account for 13 percent of GDP), grew by 6.8 higher than 20 percent. Construction was one of percent in H1 2013 (year-on-year), down from the sectors that recorded significant improvement 11.6 percent in H2 2012, reflecting a slowdown in regulations, as observed in the World Bank’s in the volume of Rwanda’s international trade 2014 Doing Business report (Box I.2). For (Figure I.5). As a result, the aggregate contribution instance, the time required to obtain a location of trade services fell below 1 percentage point and building permit was reduced from 164 days for the first time since 2010. Growth in transport in 2013 to 104 days in 2014. The improvement and communication services slid to 6.6 percent reflects implementation of an electronic platform in H1 2013, down from 18.8 percent in H2 2012. for obtaining building permit applications and the Real estate and business services contracted by streamlining of procedures. 6.8 percent, after 7.0 percent real growth in H2 Figure I.5: Growth slowed in all services subsectors Figure I.6: Construction continued to drive industrial growth (Annual growth, percent) (Contribution to growth, percentage points) 20 Financial services 17.6 16 Trade services 13.6 12 Transport, communication 9.2 Education, Health & 8.4 Pub. Admin. 8 7.2 Other services 4 Hotels & restaurants 0 Real estate, business -4 -10 0 10 20 30 2010 2011 2012 H2 2012 H1 2013 H1 2013 H2 2012 H1 2012 Construction Mining Manufacturing Utilities Industry Growth Source: NISR and World Bank staff estimates Source: NISR and World Bank staff estimates Rwanda Economic Update | Edition No. 5 5 I. Recent Economic Developments and Prospects Box I.2 Setting the basis for a private sector–led growth: Rwanda’s Doing Business indicators soared in 2014 Rwanda moved up from 54th place in the world in 2013 to 32nd in 2014 in ease of doing business. The government placed the highest priority on business environment reforms and continued efforts to promote Rwanda as an attractive business and investment destination to promote private sector–led growth. Among 10 reform indicators, Rwanda made improvements in 6 indicators (Table BI.2.1). This remarkable performance makes Rwanda the second-best reformer in the world and the second-most business friendly country in Africa, behind Mauritius but ahead of South Africa (Table BI.2.2). Rwanda performed better than the average for Sub-Saharan Africa on all indicators except trading across borders and revolving insolvency (Table BI.2.3). Although Rwanda has made several reforms to improve trade logistics environment, the costs to exports and to imports are higher than the average for Sub-Saharan Africa. Resolving insolvency is more difficult in Rwanda, with a high recovery cost and low rate of recovery. Table BI.2.1: Rwanda increased its Doing Business ranking on six indicators Indicators 2014 2013 Change in Table BI.2.2: Within Africa, only Mauritius has rank rank rank a higher Doing Business ranking than Rwanda Starting a Business 9 8 -1 2014 2013 Change in Country rank rank rank Dealing with 85 122 37 Construction Permits Mauritius 20 20 No change Getting Electricity 53 52 -1 Rwanda 32 54 22 Registering Property 8 62 54 South Africa 41 41 No change Getting Credit 13 24 11 Kenya 129 122 -7 Protecting Investors 22 32 10 Uganda 132 126 -6 Paying Taxes 22 25 3 Burundi 140 157 17 Trading Across Borders 162 160 -2 Tanzania 145 136 -9 Enforcing Contracts 40 40 No change Source: Doing Business database 2014 Resolving Insolvency 137 166 29 Source: Doing Business database 2014 Table BI.2.3: Rwanda improved almost all of its Doing Business indicators in 2014 Sub Sub Saharan Rwanda Rwanda Saharan Africa Africa average average Activity Time required to complete activity Cost required to complete activity Starting a Business Days 2.0 29.7 Percent of per income 4.4 67.4 Dealing with Construction Permit Days 104 171.1 Percent of per income 375.7 736.8 Getting Electricity Days 30 141 Percent of per income 4,018.7 4,819.9 Registering Property Days 12 58.9 Percent of property value 0.2 9.0 Paying Taxes Hours 113 314 per year Export Days 26 31 US$ per container 3,245.0 2,108.0 Import Days 30 38 US$ per container 4,990.0 2,793.0 Enforcing Contracts Days 230 652 Percent of claim 78.7 51.1 Resolving Insolvency Years 2.5 3.1 Percent of estate 29.0 23.0 Source: Doing Business database 2014 Doing Business 2014 report names Rwanda the number two improver globally and top improver in Sub- Saharan Africa. Rwanda has moved from 150th in 2008 to 32nd in 2014 with consistent reforms every year. Rwanda reduced the number of days it takes to obtain a construction permit from 166 in 2013 to 104 days in 2014 as a result of the implementation of the electronic platform for permit application. Registering property in 2013 took almost a month (25 days), and transfer fees amounted to 5.6 percent of the property value. In 2014, the process will take 12 days and cost 0.2 percent of the value. These improvements were results of the elimination of the requirement to obtain a tax clearance certificate and the introduction of the web-based land administration information system for processing land transactions. 6 Rwanda Economic Update | Edition No. 5 I. Recent Economic Developments and Prospects Stellar improvement in doing business indicators has not, however, been translated into increases in foreign investment inflows in Rwanda. Doing Business focuses on laws and regulations that matter for small and medium-size domestic enterprises; these indicators may not be relevant to larger, foreign-owned firms. In addition, key constraints such as infrastructure and landlockedness are beyond the scope of doing business. Although the quality of laws and regulations, and the extent to which this quality is reflected in their implementation, may be a useful sign to foreign investors of the quality of the business environment, there are binding constraints that need to be explored and addressed to unleash the potential growth of Rwanda’s private sector, and therefore strong FDI inflows. These constraints include inadequate infrastructure, including limited electricity and high transportation costs; lack of adequate skills; a restrictive regulatory environment; and market and coordination failure. Other industrial subsectors recovered from contracted by 15.8 percent, as a result of adverse the contraction of 2012. Strong mining growth weather conditions.2 Agricultural performance is (36.6 percent) was led by increased production likely to be lower in H22013, because of a lean of tin and coltan on the back of new investments, food harvest in the second season as the long including foreign investment in the sector. season rains (normally February–May) ended According to the Rwanda Development Board several weeks early this year. In the second (RDB), Rwanda registered US$70 million worth season, agricultural production increased by only of investments in the mining sector in 2012, up 0.62 percent.3 from just US$24 million in 2011. Growth in the Figure I.7: Agricultural growth reversed from manufacturing sector returned, as contraction in the downward trend in H1 2013 furniture production ceased and growth in food (Annual growth, percent) production continued to accelerate in H1 2013. 10 8.6 Agriculture expanded by 5.4 percent in H1 8 2013 (year-on-year), up from 2.5 percent in H2 6 5.4 2012 (Figure I.7). The increase was attributed 3.6 4 mainly to a 5.4 percent increase in food crops. 2.5 Solid outcomes in food crops were propelled by 2 an 11.4 percent rise in volumes, led by production 0 of roots and tubers, which accounted for 53.8 percent of Rwanda’s harvest. After contracting in -2 2012, export crops grew 25.2 percent in H1 2013. H2 2011 H1 2012 H2 2012 H1 2013 Coffee production expanded by 39.4 percent Food Export Others Agricultural growth (year-on-year). In contrast, tea production Source: NISR and World Bank staff estimates 2 Coffee and tea respond differently to weather conditions. Coffee needs the use of insecticides/pesticides to protect the green fruit (known as cherries) from pests and diseases. With heavy rains, there is a risk of proliferation of pests and diseases, as heavy rains do not allow the effective application of fertilizers and pesticides. In addition, heavy rains lead to a falling of flowers, which reduced harvests. Tea leaves are vulnerable to dry and erratic weather conditions. 3 Rwanda’s agriculture is mainly rain fed, and agricultural seasons are determined by the rainy seasons. The short rainy season lasts from mid-September to mid- December (followed by a short dry season lasting until February). The long rainy season stretches from the middle of March to the beginning of June. There are thus three seasons: Season A, which involves planting from September to October and harvesting from November to January/February); Season B, which involves planting from February to March and harvesting from May to July; and Season C, which lasts from July–September and corresponds to the long dry season, during which cultivation takes place largely in marshlands. Harvest results for Season A affect growth in the first and second quarters of each calendar year; harvest results for Season B are reflected in third and fourth quarter growth rates. Rwanda Economic Update | Edition No. 5 7 I. Recent Economic Developments and Prospects 1.2 The External Sector: Narrower Trade Balance, Weaker Terms of Trade R wanda’s formal trade balance improved slightly in H1 2013, thanks to strong performance in traditional export products (primarily coffee and coltan) and weak import demand. Strong export performance reflected mainly increases in volumes; prices remained low, weakening Rwanda’s terms of trade. The tourist sector remained the largest single earner of foreign currency. The level of gross international reserves improved in May-June 2013, on the back of aid resumption, the Eurobond issuance, and weak import demand. T he trade deficit narrowed marginally in H1 2013, as export growth outpaced import growth (Figure I.8 and Figure I.9). Export Rwanda’s traditional product exports rebounded in H1 2013. After a 9.4 percent decline in 2012, export earnings from traditional values expanded by 46.3 percent (year-on-year) products (coffee, tea, and minerals) increased by to US$289.9 million, driven by solid performance 51.4 percent to US$168 million, equivalent to 58 in traditional export products, primarily coltan percent of total goods exports. The value of coffee and coffee. Coltan exports almost tripled over exports increased by 68.3 percent (year-on-year) H1 2012, becoming the single-largest export to reach US$20.2 million (7.5 percent of total item, accounting for 23.8 percent of total goods exports). With expectations of further decline in exports and 60 percent of mineral exports. Driven coffee prices, Rwandan coffee exporters decided by low levels of domestic activities, import to sell their stocks, which they had hoarded growth decelerated to 1.1 percent in H1 2013, during H2 2012. Tea exports—historically, down from increases of 30.0 percent in H1 2012 Rwanda’s second-largest export—contracted by and 19.1 percent in H2 2012. The slowdown 5.5 percent (year-on-year), as a result of both was driven mostly by contractions in imports of lower volumes and lower prices. The value of capital goods, intermediate goods, and food (Box tin exports recovered following bad performance I.3). As a result, the trade deficit narrowed by 6.5 in 2012, growing 15.5 percent (year-on-year), percent to US$1,193 million in June 2013. mainly as a result of higher international prices in Figure I.8: Export-import coverage improved as import Figure I.9: … leading to a modest narrowing in Rwanda’s growth decelerated … trade deficit (Annual percentage change) (Formal trade, million US$) 60 2,000 -1,400 50 -1,200 1,500 -1,000 40 -800 30 1,000 -600 20 -400 500 10 -200 0 0 0 Sep-10 Mar-11 Sep-11 Mar-12 Sep-12 Mar-13 Sep-13 Dec-11 Jun-12 Dec-12 Jun-13 Coverage rate (percent) Exports Growth (Y-o-y) Imports Growth (Y-o-y) Exports Imports Trade balance (right Axis) Source: NISR and World Bank staff estimates Source: NISR and World Bank staff estimates 8 Rwanda Economic Update | Edition No. 5 I. Recent Economic Developments and Prospects the first quarter of 2013. However, tin continued This decline largely reflected volatility in coffee to lose its position, with its share of overall good production; prices were at historic highs. The export values sliding from 25.0 percent in 2011 to volume of coffee exported in 2012 was at 64 10.8 percent. Export values of wolfram exhibited percent of recent high levels (26,500 tons in 2006), good results, growing 5.1 percent (year-on-year). whereas export earnings were almost 13 percent higher than the 2006 level. Therefore, increases Rwanda’s export crops remain vulnerable to in coffee export values were driven mainly by external and domestic shocks. Over the last high prices on international markets, which make five years, the share of traditional export crops Rwanda’s coffee more vulnerable. However, declined from more than 50 percent of total all Rwanda’s export items are vulnerable to goods exports in 2006 to 20 percent in 2012. international price and production volatility. The Box I.3 Rwanda’s capital and intermediate imports slowed Rwanda’s imports expanded by 1.1 percent (year-on-year) in H12013. Following a slowdown in investment demand, imports of capital goods declined by 5.5 percent, and intermediate goods declined by 2.1 percent in H1 2013 (in H1 2012, capital goods imports rose by 51.7 percent and intermediate goods imports rose by 32.0 percent). Imports of consumer goods increased by 8.3, down from 13.3 percent in H1 2012, and imports of energy products rose by 5.5 percent, down from 26.4 percent in H1 2012. Consumer goods remain the largest import expenditure, accounting for 28.6 percent of total goods imports. Intermediary goods accounted for 27.6 percent, capital goods for 26.3 percent, and energy products for 17.5 percent. Trends suggest that imports continue to expand as investment gears up for more growth. This pattern is evidenced by impressive growth rates in imports of capital goods (which reached US$471 million) and intermediate goods (which reached US$498 million) in 2012 (figures are free-on-board). In 2004, Rwanda imported just US$66 million of capital goods and US$60 million of intermediary goods. The annual nominal rate of increase was thus 28 percent for intermediate goods imports and 30 percent for capital goods imports. These imports have played a major role in compensating for the shortfall in the domestic supply of capital and intermediate goods. A cursory look at Rwanda’s total imports since 2004 indicates a shift in structure of import volumes toward raw materials. The share of intermediary goods increased to 47.4 percent of imports in 2012, up from 30.0 percent in 2004. The composition of intermediary goods also indicates strong growth in the share of construction materials, from 35 percent of intermediary goods in 2004 to 56 percent in 2012. This evidence is consistent with the sizable role of construction activities in recent economic growth rates where they generated more than 20 percent of real GDP growth in 2011-13. Growth in the volume of imported construction materials decelerated to 18 percent by June 2013,down from 26 percent by June 2012. However, imports of construction materials are expected to bounce back as large infrastructure projects come on stream. Figure BI.3.1: Import growth decelerated, driven by Figure BI.3.2: Both real investment and imports rose contraction in capital and raw materials between 2006 and 2013 (Percentage annual change in import values) (Billions of Rwf) 600 Other consumer goods 500 y = 21.137x + 175.83 Energy products 400 Food 300 y = 14.097x + 125.84 Intermediary goods 200 Capital goods 100 H1 H2 H1 H2 H1 H2 H1 H2 H1 H2 H1 H2 H1 H2 H1 -10 0 10 20 30 40 50 60 2006 2007 2008 2009 2010 2011 2012 2013 H1 2013 H2 2012 H1 2012 Gross Capital Formation Imports Source: BNR, NISR, and World Bank staff estimates Rwanda Economic Update | Edition No. 5 9 I. Recent Economic Developments and Prospects government recognizes this vulnerability and is 16.1 percent. Rwanda’s high concentration of making efforts to diversify exports, as reflected in exports leaves it highly vulnerable to volatility the Second Economic Development and Poverty in global prices. The high concentration of the Reduction Strategy (EDPRS 2 [Box I.4]). export basket in basic commodities is associated with large swings in terms of trade and a more Figure I.10: Rwanda’s coffee is vulnerable to prices and production volatile growth performance (Arezki, Lederman, (Index, 2006=1) and Zhao 2011). Rwanda’s export basket is 3.0 concentrated in a few exports items—coffee, tea, 2.35 2.30 and minerals (tin, coltan, and wolfram)—which 2.5 account for more than 65 percent of total exports.4 1.76 Therefore, the export price index is subject to the 2.0 volatility of international prices of these items. Between December 2011 and June 2013, export 1.5 1.00 prices of coffee have declined by 33.6 percent; 0.64 international tea prices increased by 2.8 percent. 1.0 0.63 0.69 The export price of tin declined by 7.4 percent, 0 the export price of wolfram fell by 10.7 percent, Dec-04 Jan-06 Jan-07 Jan-08 Jan-09 Jan-10 Jan-11 Jan-12 and the export price of coltan increased by 42.3 Volume Price percent. Import prices rose significantly in 2012 Source: BNR and World Bank staff estimates and 2013 (Figure I.11). Rwanda’s terms of trade continued to Tourism continues to be the leading export weaken. The terms of trade calculated based earner. In H1 2013, tourism receipts have on the national account declined by 18 percent increased by 21.9 percent to US$142.5 million. in 2012 and by another 8 percent in H1 2013 This robust growth was driven mainly by a 14 (year-on-year). This deterioration was caused percent increase in the number of tourist arrivals, mainly by low export prices and the steady to 665,000. In addition, in the 12 months ending increase in import prices. In the 18 months June 2013, tourist receipts from gorilla trekking ending June 2013, export prices declined by activities almost tripled, to US$11.9 million in 10.8 percent while import prices increased by H1 2013, mainly as a result of the increase in the Figure I.12: … driven by volatile prices of Rwanda’s Figure I.11: Rwanda’s terms of trade deteriorated export products (Index, 2006=100) (Index, December 2011=1) 2.0 160 140 1.6 120 1.2 100 0.8 80 60 0.4 2008 2009 2010 2011 2012 H1 2013 Dec-11 Mar-12 Jun-12 Sep-12 Dec-12 Mar-13 Jun-13 Sep-13 Terms of trade Export Price Index Import Price Index Coffee Tea Tin Coltan Wolfram Source: BNR and World Bank staff calculation Source: BNR and World Bank staff calculation 4 In 2008–12, coffee accounted for 16.6 percent, tea for 19.1 percent, and minerals for 32.5 percent of total exports. 10 Rwanda Economic Update | Edition No. 5 I. Recent Economic Developments and Prospects Box I.4 Toward self-reliance Rwanda aspires to diversifying its export base The EDPRS 2 identifies three main bottlenecks to Rwanda’s external competitiveness: • low external connectivity • high trade-related cost, as a result of landlockedness • heavy concentration of a few traditional export items and their vulnerability to terms of trade shocks. To enhance external competitiveness, the EDPRS 2 proposes prioritizing interventions that will improve external connectivity, transform Rwanda’s logistics system, and build soft and hard sector-specific infrastructure to accelerate growth in the commodity and tourism sectors. To increase Rwanda’s external connectivity, the EDPRS 2 prioritizes two interventions. The first is completion of phase 1 of the development of a major new international airport in Bugesera by 2017. The second is investment in an ambitious expansion of RwandAir during the EDPRS 2 period (2013–18). The government will also play a leading role in finalizing the planning of and procurement for a railway connection to Mombasa and Dar es Salaam, construction of which is expected to start before the end of EDPRS 2. To transform Rwanda’s logistics system and strategically position the system to strengthen Rwanda’s growing exports and re-exports to Burundi and the eastern Democratic Republic of Congo, the EDPRS 2 proposes the following interventions: • establishment of an integrated logistics system based on a Kigali logistics platform linked to regional logistics centers • improvement of the sea-land logistics system by building off-dock container depots • completion of one-stop posts at all Rwandan border posts to facilitate both large- and small-scale cross-border trade • removal of nontariff barriers on the northern and central corridors. To strengthen export promotion and enhance sector-specific hard and soft infrastructure in key export sectors, the EDPRS 2 focuses on five areas: • review of the institutional set-up for export promotion and strengthening of the export promotion department at the RDB • overhaul of Rwanda’s mining sector, on the basis of a new mining law, a new mining model contract for investors, and targeted investments in exploration • investment in a major tea expansion program, increasing the area under production by 18,000 hectares by 2018 • engagement in a large and systematic expansion of extension services and capacity building in the coffee sector, modeled on successful approaches tested domestically • completion of the Kivu belt (road) and investment in developing a tourism circuit. Rwanda already has two main strategies to improve external trade. The first is the National Export Strategy of March 2011, which aims to accelerate growth and diversify exports. This strategy is currently under review. The second is the National Cross-Border Trade Strategy, adopted in October 2012, which focuses on exports to neighboring countries. Source: EDPRS 2 cost of gorilla permits, which rose by 50 percent Figure I.13: Gross reserves increased in H 2013 (US$ million and months of prospect imports) to US$750 for foreign nonresidents, US$375 for foreign residents, and Rwf 30,000 for Rwandans 1,200 6 as of June 1, 2012. Since 2005, Rwanda has 5 1,000 intensified campaigns promoting itself as the gorilla destination. The country also has other 800 4 excellent tourism assets (Box I.5). 3 600 Foreign reserves surpassed the level they had 400 2 reached before the aid shortfall. After sliding 1 200 to their lowest levels since December 2011, gross international reserves increased by US$220 0 0 Dec-10 Apr-11 Aug-11 Dec-11 Apr-12 Aug-12 Dec-12 Apr-13 Aug-13 million to reach US$1,069 million in June Million US$ Import Months (Right axis) 2013 (Figure I.13). Although this is the highest Source: BNR and World Bank staff estimates Rwanda Economic Update | Edition No. 5 11 I. Recent Economic Developments and Prospects Box I.5 Beyond gorilla trekking There are other excellent tourism charms in Rwanda Rwanda’s tourism sector—the country’s largest earner of foreign exchange—is growing. After declining by 8.6 percent between 2008 and 2009, as a result of global turmoil, the number of tourist arrivals gradually increased (Figure BI.5.1). Rwanda hosted 1,075,790 tourists in 2012, 53.9 percent more than four years earlier. Receipts grew even faster, rising 61.5 percent over the same period. Tourism accounted for 30 percent of Rwanda’s foreign exchange (equivalent to 4 percent GDP). The main characteristics of international tourists to Rwanda are the large share of visitors for business and conferences (46 percent in 2008-12); the small share of tourists for leisure (8.9 percent [Figure BI.5.2]); and the small share of tourists from developed countries (12.0 percent). Rwanda is well known for its mountain gorillas. Since 2000, it has marketed itself as the gorilla destination. The Volcano National Park attracted more than 48 percent of total park visitors between 2008 and 2012; it accounted for more than 95 percent of park revenues and about 5 percent of total tourist receipts. The gorilla-naming ceremony called “Kwita Izina” has played an important role in marketing Rwanda’s tourism, attracting a number of international celebrities and conservationists. Each year since its launch in 2005, mountain gorillas born in the previous 12 months are named. Ceremonies provide a good platform to promote Rwanda as a destination as well as to protect the animals and conserve their habitat. Partly thanks to the sensitization of the gorilla-naming ceremony and gorilla tourism in general, poaching has been significantly reduced and the number of gorillas increased steadily Figure BI.5.1: The number of tourists has increased Figure BI.5.2: … dominated by visitors for business and steadily since 2008 … conferences (Sum of consecutive 12 months) (Sum of consecutive 12 months, thousands) 1,200 50 1,400 40 1,200 1,000 30 Transit & others 1,000 800 20 800 10 600 Business & conference 600 0 400 08 09 10 11 12 9 0 1 2 3 400 -0 -1 -1 -1 -1 - - - - - -10 ec ec ec ec ec n n n n n Ju Ju Ju Ju Ju D D D D D 200 200 Visiting friends and relatives -20 -30 0 Leisure 0 Year-on-year growth Thousands Dec-08 Jun-09 Dec-09 Jun-10 Dec-10 Jun-11 Dec-11 Jun-12 Dec-12 Jun-13 Source: RDB and World Bank staff estimates Source: RDB and World Bank staff estimates Rwanda has other excellent tourism assets, which create a wider foundation for tourism. The Akagera National Park hosts a range of wildlife, including elephants, hippopotamuses, giraffes, and zebras. The Nyungwe National Park has a large tract of mountain forest and is rich in biodiversity. It recently opened a canopy walk, the only one of its kind in East Africa. The canopy walkway is 40 meters above ground level and 90 meters long. It stands above a steep, deep valley that affords clients a dizzying platform from which to view the incredible biodiversity of this rare forest. Lake Kivu has recreational facilities and potential for significant tourism development. Looking forward, Rwanda is targeting high-end and conference tourism. To this end, the Kigali Convention Center, with world-class conference facilities, will to be operational in 2014. It will host regional and international conferences that will further stimulate demand and exports from the service industry. Finalizing the Kivu Belt road will enable Rwanda to achieve a number of key objectives under the EDPRS 2. They include (a) connecting a number of important secondary cities to domestic markets and Eastern DRC; (b) spurring investments in the tourism sector along the Kivu Belt; and (c) creating a national tourism circuit, enabling travelers to combine a visit to the Volcano National Park, Lake Kivu, Nyungwe Forest and other notable touristic sites in one circuit. Source: RDB nominal level ever reached, it remains lower than Aid inflows resumed. Almost a year after the the 5.1 months achieved at the end of 2011. The suspension and delay of aid, donor sentiment increase in reserves resulted from the resumption toward Rwanda improved, in particular following of aid inflows, proceeds from the Eurobond, and the signing of a regional peace agreement in the deceleration of central bank intervention in February 2013. Multilateral donors, as well as foreign exchange markets. a number of bilateral donors, resumed budget 12 Rwanda Economic Update | Edition No. 5 I. Recent Economic Developments and Prospects support operations. Despite these positive Despite the Eurobond issuance, Rwanda’s developments, budget support for FY2012/13 fell debt profile improved and the debt remains short of the budgeted amount by 1.2 percent of sustainable. According to the new Debt GDP (about US$90 million). Sustainability Analysis conducted by the World Bank and the International Monetary Fund (IMF), Taking advantage of positive market Rwanda’s debt sustainability indicators are below sentiments, Rwanda successfully issued its desirable thresholds under all scenarios examined. first Eurobond—a US$400 million, 10-year Rwanda’s risk of debt distress is judged to have bond—in April 2013. The issue was priced improved from “moderate” to “low” risk. The to yield 6.875 percent and attracted a US$3.5 improvement was driven by two factors. First, billion order book, more than 8.5 times the the Country Policy and Institutional Assessment, issue size and more than half the country’s which assesses the quality of a country’s present GDP. Half of the proceeds will be used to retire policy and institutional framework, classified unfavorable government-guaranteed loans to the Rwanda as a strong performer, leading to higher Kigali Convention Center (US$120 million) and thresholds for assessing debt sustainability. RwandAir (US$80 million). The remaining funds Second, Rwanda’s export performance in 2013 will finance completion of the convention center improved significantly on the back of new (US$150 million) by the end of 2013 and the investments, including by foreign investors, in building of a 28-megawatt hydropower project the mining sector. The improvement in Rwanda’s (US$50 million). As of June 2013, the loans had debt distress risk rating is expected to make been paid and construction of the convention additional sources of concessional financing center was underway. available to Rwanda. The main risk to Rwanda’s debt sustainability remains the narrow export base and high dependent on aid. I.3 Monetary Policy, Exchange Rate Policy, and Inflation: Growing Concerns I nflation rates and the Rwandan franc remained stable in H1 2013, but inflation rose in June and the Rwandan franc depreciated sharply in July. In H1 2013, the rate of credit growth declined markedly, posing significant challenges to financing. With their balance sheets impaired by the increase in the volume of non performing loans, commercial banks substituted lending activities to the private sector with Treasury bills. To stimulate more financing to the economy, the central bank eased its monetary stance. The results are not yet evident. I nflation stabilized in H1 2013, but upward pressure is gathering. Inflation rates remained in single digits and were stable, following lower declined from 7.9 percent in December 2012 to 2.6 percent in May 2013, largely as a result of an adequate supply of food crops throughout H1 food and fuel prices, and a steady import price 2013. Core inflation, which excludes fresh food index (Figure I.14). Headline inflation rate and energy prices, increased to 5.2 percent in remained below 6 percent during H1 and was May 2013, falling to 3.5 percent in the following 3.7 percent in June 2013. The food price index months. Import prices also remained low in Rwanda Economic Update | Edition No. 5 13 I. Recent Economic Developments and Prospects Figure I.15: … but there are concerns about Figure I.14: Inflation remains moderately low… rising food prices (Year-on-year change in inflation, percent) (Year-on-year change in inflation, percent) 7 16 6 14 12 5 10 4 8 3 6 4 2 2 1 0 0 -2 Jun-12 Sep-12 Dec-12 Mar-13 Jun-13 Sep-13 Jun-12 Sep-12 Dec-12 Mar-13 Jun-13 Sep-13 Core Headline Food Energy Local prices Import prices Source: BNR and World Bank staff estimates Source: BNR and World Bank staff estimates H1 2013, mainly reflecting inflation trends in its level of end 2012. This rapid depreciation Rwanda’s main trading partners (Box I.6), low could be attributed to the reduction of the BNR’s import growth, and slower depreciation of the intervention. BNR, the main player in Rwanda’s Rwandan franc. After June 2013, inflation rose, foreign exchange market, reduced by 43.5 as a result of the poor food harvest of the second percent its interventions, to US$206 million. The agricultural season (Season B). Driven by higher Rwandan franc also depreciated against all major prices of vegetables, the food price index was up trading partner currencies, indicating real and 7.8 percent (year-on-year) by September 2013. It effective depreciation (Box I.6). increased by 3.4 percentage points between the second and third quarter. In addition, the import Monetary conditions tightened in H1 2013, price index is gradually rising, as a result of the thus leading to weak private lending activities. rapid depreciation of the Rwandan franc that Growth of broad money and deposits decelerated began in July 2013. to 3.1 percent in May 2013, falling from more than 20 percent 12 months earlier. The growth The depreciation of the Rwandan franc of loans to the private sector decelerated for 12 began accelerating in July 2013. After a rapid consecutive months. The credit growth declined depreciation of 3.1 percent against the U.S. dollar from the peak of 36.7 percent in September 2012 in H2 2012, the depreciation eased in H1 2013, to to 18.3 percent in June 2013, and further dropped 1.8 percent. However, the depreciation accelerated to 9.9 percent in September 2013 (Figure I.16). in July 2013, with the value of the Rwandan franc The volume of new loans increased by only 2.2 falling by 3.0 percent between July and October. percent in June 2013 compared with 53.8 percent Overall, the Rwandan franc depreciated by 4.8 for the same period in 2012. The slowdown was percent between January and October 2013, observed across all sectors except manufacturing thus 1.0 percentage point more than in the same (Figure I.17). Mortgage loans, which represent a period in 2012. The spread between the official fifth of total new loans, contracted by 16.5 percent and market exchange rates narrowed in H1 2013. in the 12 months ending June 2013, and new loans After a wider spread of about 5 percent by end- to financial services declined by 31.6 percent. 2012, the spread had narrowed to less than 1.0 Remarkably, the flow of credit to trade and percent by end May 2013. However, the spread transport services, the two drivers of the economic started widening, averaging 2.4 percent between resilience in H2 2012, significantly declined. The June and October, although it remained far below volume of new loans to trade, restaurant, and hotel 14 Rwanda Economic Update | Edition No. 5 I. Recent Economic Developments and Prospects Box I.6 Rwanda’s real effective exchange rate depreciated Since the beginning of the aid shortfall in mid-2012, Rwanda’s real effective exchange rates (REER) depreciated by 3.5 percent (Figure BI.6.2). The narrow gap with the nominal effective exchange rate (NEER) suggests that the depreciation of the REER was caused mainly by the nominal depreciation of the Rwandan franc against currencies in major trading partnersa (Figure BI.6.1). In the 12 months ending June 2013, the Rwandan franc lost 5.2 percent of its value against the U.S. dollar; the decline was partially offset by weaker depreciations against neighbor country currencies. Inflation rates have been low in both Rwanda and its trading partners. Figure BI.6.1: Although inflation rates remained low in Figure BI.6.2: … Rwanda’s real exchange rate continued both Rwanda and its trading partners … to depreciate in H1 2013 (Year-on-year percentage change) (2005=100) 10 81 8 Real Effective Exchange Rate (REER) Rwanda 80 > Depreciation 6 79 4 78 Trading Partners (weighted average) 2 ------------- 77 0 76 Appreciation < -2 75 Gap between Rwanda's Inflation -4 and that in trading partners Nominal Effective Exchange Rate (NEER 74 73 Jan MarMay Jul Sep Nov Jan MarMay Jul Sep Nov Jan MarMay Jul Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul 2011 2012 2013 2011 2011 2011 a Rwanda’s major trading partners are Uganda, Kenya, Tanzania, Burundi, the United States, the Euro area, South Africa, Sweden, Switzerland, and the United Kingdom. Source: BNR and World Bank staff estimates activities rose by just 9.5 percent (down from Low credit growth can be attributed to two 60.7 percent for the same period in 2012), and main factors. First, lending to the government new loans to transport services rose 6.6 percent was a preferred option in H1 2013, thus crowding (down from 26.5 percent). out private sector credit. Between June 2012 and Figure I.16: Tightened monetary conditions Figure I.17: Loans to financial services and slowed private credit growth mortgages contracted sharply (year-on-year change, percent) (Annual percentage change) Year-on-year Bn Rwf growth Manufacturing 40 160.0 Community services 30 120.0 Trade, restaurant and hotel Transport and warehousing 20 80.0 Agriculture Others 10 40.0 Mortgage Financial services 0 0 Jan-12 May-12 Sep-12 Jan-13 May-13 Sep-13 -40 0 40 80 120 T-bills Stock (Right axis) New loans (Right axis) Deposit Broad Money Credit Jun-13 Jun-12 Source: BNR and World Bank staff estimates Source: BNR and World Bank staff estimates Rwanda Economic Update | Edition No. 5 15 I. Recent Economic Developments and Prospects June 2013, the stock of short-term public debt percent by the end June of 2013. The share of (Treasury bills) increased by 60 percent, more 6- to 12- month T-bills was stable, at about 40 than three times the increase in private sector percent of the T-bill stock. Second, with balance credit (18.3 percent). Preference for T-bills was sheets impaired by rising non performing loans a result of attractive interest rates, which were (NPLs), commercial banks have become prudent higher than deposit rates for 15 consecutive in extending further credit to a decelerating months ending in August 2013 (Figure I.18). The domestic economy. The NPLs ratio increased attractiveness of T-bill rates diverted depositors from 6.0 percent in December 2012 to 6.9 percent from keeping money into their accounts into in June 2013. the banking system, reducing the growth of bank deposits to 3.1 percent (year-on-year) in To stimulate more financing of economic May 2013. This behavior was reflected by the activity in H2 2013, the central bank reduced increasing share of nonbank holders of T-bills and its policy rate by 50 basis points, to 7.0 percent the constant portion of longer T-bill maturities. in mid-June 2013. It had kept its policy rate at The share of nonbank holders of T-bills expanded 7.5 percent for 13 months, expecting inflationary from less than 1.0 percent in June 2012 to 29.5 pressures from substantial domestic government borrowing. The results were mixed. Although the Figure I.18: T-bills rates were higher than deposit rates declining trend in broad money growth credit for 15 consecutive months (Percent) stopped, credit growth decelerated further, to 9.9 percent in September 2013 (real growth of 4.8 19 percent), leaving lending rates at the level before 17 revision of the policy rate. This delayed impact on credit growth corroborates the weakness in 15 the effectiveness of Rwanda’s monetary policy, 13 which is limited by a number of factors, such as the heavy reliance on cash in the economy 11 and an underdeveloped financial market that is 9 dominated by a small and concentrated banking system (Box I.7). The pick-up in the growth of 7 Apr-12 Jul-12 Oct-12 Jan-13 Apr-13 Jul-13 broad money and deposits could be attributed to Deposit T-bills Lending the increased level of foreign assets held by the Source: BNR and World Bank staff estimates central bank. 16 Rwanda Economic Update | Edition No. 5 I. Recent Economic Developments and Prospects Box I.7 BNR - Beyond our control? Factors affecting effectiveness of Rwanda’s monetary policy Rwanda’s central bank is responsible for monetary policy, including the maintenance of price stability. It follows a monetary targeting framework, using reserve money as its operational target and broad money as its intermediate target. To implement monetary policy, the central bank relies on a variety of instruments, including reserve requirements, open market operations (the buying and selling of government securities by the central bank in order to expand or contract the amount of money in the banking system), the central bank policy rate, foreign exchange interventions and the rediscount rate. Since late 2012, the BNR has transitioned from strict reserve money targeting to flexible reserve money targeting for effective monetary policy. For more than 15 years since 1995, the BNR maintained a strict reserve money targeting framework—a framework based on a target on reserve money with an assumption that the demand for money and the money multiplier are stable.a However, the BNR has come to realize that this assumption does not always hold. Therefore, in November 2012, it introduced a flexible reserve money targeting framework with a reserve money band of ±2 percent around the central reserve money target. The new framework aims to provide (a) flexibility to accommodate changes in money multipliers and velocity caused by money demand shocks, financial deepening,and other factors and (b) a platform to react to unanticipated economic shocks to output and inflation, which may cause monetary aggregates to deviate substantially from ex ante monetary target. A relationship between reserve money and broad money seem stable in short-run and long-run (Figure BI.7.1). This suggests that the BNR can control money creation to the extent that it controls reserve money. However, the relationship between broad money and inflation is not always stable. Until the fourth quarter of 2009, inflation followed the development of broad money with a few months’ lag. This relationship started to break in early 2010, partly reflecting supply-side factors. For example, domestic food production dramatically increased (food accounts for 35 percent of the consumer price index basket), and inflationary pressures in neighboring countries subdued. The central bank policy rate was introduced in 2005 to signal the bank’s monetary policy stance (or equivalently to influence market interest rates). The impact of monetary policy on prices depends on the responses of intermediate variables, such as market interest rates, exchange rates, and credit. Market interest rates have barely responded to the central bank policy rate, and the response of lending rates remains weak. The response of the deposit rate has been improving since 2009, when two regional banks (Kenya Commercial Bank and Equity) entered the financial sector, suggesting that increased competition and deepening of the financial sector are key for effective monetary policy. The impact of monetary policy on price stability is limited as a result of heavy reliance on cash in the economy, a small banking system, persistent excess liquidity in the banking system, and external/domestic shocks. Despite recent developments in the payment system (such as the introduction of automatic teller machines and mobile money), most Rwandans still rely on cash for daily transactions. Consequently, about 70 percent of reserve money—an operating target for the BNR—is currency in circulation, leaving the BNR with only bank deposits (amounting to 30 percent of reserve money) to implement monetary policy. The impact of monetary policy on market interest rates depends on the behaviors of banks. Moreover, experience shows that price trends are mainly driven supply factors such as domestic food production and international fuel prices, which are beyond control of monetary policy. Persistent excess liquidity and an inactive interbank market are at the helm of the limited monetary policy effectiveness. Banks in Rwanda continue to hold excess reserves, because they have limited financial instruments in which to invest. In H1 2013, for example, the average ratio of excess reserves to total deposits reached 40 percent, up from 17 percent in 2012. This excess liquidity enables banks to exercise their lending behavior irrespective of monetary policy decisions. Monetary policy instruments for mopping up this excess liquidity have been mostly short term and unable to bring liquidity level to the levels required for effective monetary policy. The size and inactivity of interbank market also limits the price discovery mechanism (market-based interest rate determination), limiting the signaling impact of BNR policy rate. According to the Seventh IMF Staff Review Report, the BNR has established strategies to strengthen monetary policy transmission. They include issuing monetary policy Treasury bills for absorbing longer- term and more permanent liquidity (in consultation with the Treasury), developing a bond issuance program, providing instruments with longer maturities by extending the maturity period of repos, and developing a secondary market. Rwanda Economic Update | Edition No. 5 17 I. Recent Economic Developments and Prospects Figure BI.7.1: Reserve money tends to affect broad Figure BI.7.2: Structural excess reserves reduce money in the longrun rather than the short-run the impact of monetary policy effectiveness (Year-on-year percentage changes) (Excess reserves as percent of total deposits) 50 80 60 40 40 30 20 20 0 10 -20 -40 0 -60 -10 Feb-05 Apr-06 Jun-07 Aug-08 Oct-09 Dec-10 Feb-12 Apr-13 -80 Jan-07 Dec-07 Nov-08 Oct-09 Sep-10 Aug-11 Jul-12 Jun-13 Broad Money Inflation Reserve money Source: BNR and World Bank staff estimates Source: BNR and World Bank staff estimates a The money multiplier is the rate at which money is created in response to changes in reserve money. Money demand refers to the demand for money by economic agents to make transactions. I.4 Fiscal Policy: Effect of Aid Distress on Development Agenda D espite Rwanda’s sound aggregate fiscal position in the face of aid shortfall in the annual budget ended in June 2013, the fiscal space for development spending was lessened. The overall fiscal deficit was lower than the revised budget, despite the 1.5 percent of GDP shortfall in budgetary grants relative to projections in the revised budget of February 2013. The authorities covered some of the shortfall through greater than expected domestic revenue collection. Capital spending and net lending operations were weaker than projected, as the government ensured the fully execution of nondiscretionary spending for wages, interest payments, and social expenditures. R wanda’s fiscal policy came under pressure in FY2012/13, leading to spending adjustment. The budgetary grant was projected government introduced a hiring freeze, cut non- priority current spending allocations across line ministries and public institutions, and classified as at 5.3 percent of GDP (equivalent to 18.8 percent “contingent expenditures” some expenditure that of the budget) in July 2012, more than 90 percent could be executed when aid disbursements from of which was supposed to be realized by the end multilateral agencies accrue to the budget. The of December 2012. In the event, less than 40 revised budget took into account increased net percent (2 percent of GDP) was realized, pushing lending operations intended for the completion the government to resort to increased domestic of key strategic investments. Nevertheless, borrowing, a build-up in arrears, and the the overall deficit target of the original budget, postponement of some capital spending in order excluding operations related to the Eurobond, to sustain priority spending. In February 2013, was maintained, to cushion the economy from the budgetary grant was revised downward to 4.2 downside risks. percent of GDP. In response to the aid shortfall, the 18 Rwanda Economic Update | Edition No. 5 I. Recent Economic Developments and Prospects The actual budget deficit in FY 2012/13 was equivalent to 7.4 percent of GDP, the deficit much lower than expected (Table I.1). By was fully financed and the government built June 2013, the overall fiscal deficit was 5.3 up its reserves at the central bank, to about 3.1 percent, less than the 6.1 percent in the revised percent of GDP. More important, by end June budget, despite a 1.5 percent of GDP shortfall in 2013, the government had fully paid accumulated the budgetary grant in the revised budget. The arrears and implemented almost all contingent low deficit was caused primarily by a surge in expenditures. It also ensured full execution of domestic revenue collection but also by lower nondiscretionary spending for wages, interest capital spending. Domestic revenue collection payment, and social expenditures. exceeded projections by 0.9 percent of GDP, almost equally divided between tax and non-tax Tax collection is gradually improving. Over the revenues. The fiscal deficit was financed mainly last two fiscal years, tax collection increased by by external borrowing. With net foreign loans 0.8 percent of GDP (0.3 percent in FY2011/12 Table I.1: Fiscal outcomes in FY 2012/13 were better than projected (Percent of GDP) 2012/13 Original Revised Actual Revenue and grants 25.5 24.6 23.9 Domestic revenue 15.2 15.1 16.0 Tax revenue 13.4 13.7 14.2 Non-tax revenue 1.7 1.4 1.8 Grants 10.4 9.4 7.9 Budgetary grants 5.3 4.2 4.1 Capital grants 5.1 5.2 3.8 Total expenditure and net lending 28.0 30.5 29.0 Current expenditure 14.2 13.6 13.8 Wages and salaries 3.8 3.8 3.7 Purchases of goods and services 2.7 2.5 2.7 Interest payments 0.4 0.6 0.7 Transfers 5.6 5.1 5.0 Exceptional social expenditure 1.8 1.5 1.7 Capital expenditure 13.5 13.4 12.3 Domestic 5.8 5.4 5.2 Foreign 7.8 7.9 7.1 Net lending 0.2 3.5 3.0 Eurobond related operations - 3.0 2.5 Change in arrears (- reduction) -0.2 -0.2 -0.2 Overall deficit Excluding grants -13.0 -15.5 -13.2 Including grants -2.6 -6.1 -5.3 Financing 2.6 6.1 5.3 Foreign financing (net) 2.7 8.3 7.4 Domestic financing -0.1 -2.3 -2.1 Memo: Overall deficit (excl. Eurobond related operations, including grants) -2.6 -3.0 -2.8 Source: MINECOFIN and World Bank Staff calculation Rwanda Economic Update | Edition No. 5 19 I. Recent Economic Developments and Prospects and 0.5 percent in FY2012/13) to 14.2 percent Capital spending appears to have gained in FY2012/13. As a result of ongoing tax some traction. Capital spending increased by 17 administration measures, tax revenues were up percent in FY2012/13, reaching 12.3 percent of nearly 18 percent, reaching 101.7 percent of GDP (42.3 percent of total spending), although the revised target in FY2012/13. A series of tax lower than budgeted. The increase shows administration measures focus on increasing the government’s commitment to alleviating compliance at low costs for both taxpayers and infrastructure bottlenecks by increasing its the Rwanda Revenue Authority and limiting tax contribution to capital spending. Over the last evasion. The government aims at accelerating three fiscal years, the share of capital spending domestic resource mobilization to 17.7 percent gradually increased, averaging 12.2 percent of of GDP in FY2015/16 as a part of the IMF’s GDP (or 43.6 percent of total spending [Box new policy instrument support. The increase will I.8]). To achieve this expenditure reorientation, allow additional spending on capital projects. the government contained the growth of recurrent spending, especially spending on goods and Achieving better fiscal performance has services. As a result, the share of recurrent not been without cost. The aid shortfall has spending declined to 13.8 percent of GDP in diminished the fiscal space for capital spending, FY2012/13, down from 14.9 percent the previous through lower capital spending and net lending fiscal year. Despite the increase in capital operations. By the end of June 2013, about spending, spending on service delivery was not Rwf 1,335.6 billion (93.7 percent of the revised compromised. The government maintained the target) had been spent—a lower execution rate momentum of budget allocations to education, than in previous years. A low execution rate health, energy, social protection, agriculture, of 82.9 percent was recorded in net lending and transfers to districts. Moreover, most capital activities, as some Kigali Convention Center expenditures are in these sectors (investment in construction activities were postponed to the energy, water sanitation and hygiene, irrigation, current fiscal year. The aid shortfall also affected skills developments, and so forth). capital spending, with a shortfall of almost 10 percent of its revised target. The government The 2013/14 budget offers a clear signal about had to postpone or decelerate execution of some the importance of capital spending (Box I.9). ongoing agriculture and infrastructure projects. It clearly points to fiscal consolidation, by taking In addition, the domestic debt of the government into account the decline in external budget surged. By the end of June, the government’s stock support while creating space for critical spending of T-bills had increased by almost 60 percent for EDPRS 2 priorities. Total expenditure and (year-on-year), to Rwf 146.1 billion (equivalent net lending are projected at 30.1 percent of to 3.2 percent of GDP). Increased government GDP in FY2013/2014. Capital expenditure is domestic borrowing has been a better alternate projected to increase to 13.7 percent of GDP, for the banking sector, thus restraining the credit and the share of recurrent spending is projected expansion to the private sector. at 14.1 percent of GDP. 20 Rwanda Economic Update | Edition No. 5 I. Recent Economic Developments and Prospects Box I.8 Increased capital spending in both absolute and relative terms Development spending as a share of GDP has more than doubled since 2003. Starting at just 5.1 percent in 2003, development spending reached 12.3 percent of GDP in 2012, almost a half (44.8 percent) of total spending. Between 2003 and 2012, the compound annual growth rate for capital spending was almost 30 percent—much higher than the 16.5 percent for wages and salaries or the 14.0 percent for goods and services (Figure BI.8.1). In nominal values, development spending increased by a factor of almost 11, from just Rwf 51 billion in 2003 to Rwf 536.3 billion in 2012. The government contribution also increased, with domestic capital spending increasing from 4.8 percent (equivalent to 1.0 percent of GDP) to 18.6 percent of total spending (equivalent to 5.1 percent) in 2012. The foreign-financed component averaged 21.8 percent of spending (5.3 percent of GDP) in 2003–12. To achieve this expenditure reorientation, the government contained the growth of recurrent spending. It sustained recurrent expenditure at an average of 14.8 percent of GDP in the 10 years ending in 2012. The figure is projected to stabilize at about 14 percent of GDP in the medium term (Figure BI.8.2). Since the adoption of the decentralization process, transfers to districts form the largest share of recurrent spending, currently accounting for a fifth of total spending. Figure BI.8.1: Public capital spending has been increasing … Figure BI.8.2: … narrowing the gap with recurrent spending (Compound annual growth rate 2003–12, percent) (Percent of GDP) 20 Domestic Cap. Spending 40.6 Transfers 30.2 Recurrent 15 Foreign Cap. spending 25.5 Wages and salaries 16.5 10 Capital Goods and services 14.0 5 Excep. Soc. spending 7.8 Interest payments 7.2 0 0 5 10 15 20 25 30 35 40 45 2000 2002 2004 2006 2008 2009/10 2011/12 2013/14 Proj. Source: BNR and World Bank staff calculations Source: MINECOFIN and World Bank staff calculations Box I.9 Stairway to the future The 2013/14 budget makes headway on the medium-term agenda In the face of declining aid, the government has renewed its commitment to domestic revenue mobilization. The share of the budget financed by budget support is projected to fall to less than 30 percent, with the rest financed by domestic collections. To increase collections, the government prepared a comprehensive revenue mobilization plan. For the 2013/14 budget, tax revenue collections are projected to rise to 15.1 percent of GDP, as a result of a series of revenue administration measures aimed at strengthening tax collection. In addition to ongoing measures aimed at increasing compliance at low costs and limiting tax evasion, new measures include: • review of the Investment Code (removal of exemptions from duties on construction materials and employment tax discounts); • introduction of a royalty tax on mining; • revision of the Double Taxation Avoidance Agreement (DTAA) with Mauritius; and • enhancement of value-added tax (VAT) collection, with implementation of electronic billing machines. The budget creates space for further capital spending. The capital budget was increased by 28 percent to 46 percent of the budget (13.7 percent of GDP). Capital allocations remain focused on various ongoing projects in energy (the energy roll-out program,which involves the construction of hydropower stations and distribution); transportation (road construction);and irrigation, as well the construction of schools and health facilities. Export promotion projects will also benefit from these allocations. About half of the budget is allocated to the four thematic areas of the EDPRS 2 Source: 2013/14 budget speech Rwanda Economic Update | Edition No. 5 21 I. Recent Economic Developments and Prospects I.5 Economic Outlook and Risks T he World Bank now projects annual economic growth in 2013 of 6.6 percent, down from 7.0 percent in the May 2013 edition of the Rwanda Economic Update. The downgrade is based on macroeconomic trends in the first three quarters of the year. Although government actions limited the impact of the decline in aid, the economy is now feeling the lagged effects. Moderation is evident in some economic indicators, including credit expansion to the private sector and imports of capital and intermediate goods; there has been a sharp depreciation of the Rwandan franc; and the harvest for the second season was disappointing, slowing sector growth in the second half of 2013. Rwanda’s economy is projected to grow by 7.2 percent in 2014. I.5.1 Outlook for the Short Term: percent. Growth in 2014 is projected to be just Tighter Focus on Policies 0.3 percentage point lower than earlier (7.2 T percent; [Table I.2]). The revised growth rate he May 2013 edition of the Rwanda projections reflect a variety of factors, including Economic Update projected growth of 7.0 the lean agricultural harvest in the second season; percent in 2013 and 7.5 percent in 2014. Those recent declines in overall imports, including projections were based on an assessment that construction materials; and the slowing of credit domestic activity was robust and not showing growth, which may contribute to a slowdown signs of significant fragility. Over the last several in construction activities. Services are expected months, the economy started feeling the lagged to regain some dynamism, as the government is effects of the aid shortfall, through deceleration expected to accelerate its spending in the second of credit to the private sector and public spending. half of 2013. Private consumption growth dampened and public consumption growth considerably decelerated The inflation outlook is contingent on the in the first half of 2013. These effects were not movements in food and fuel prices and observed in the second half of 2012. exchange rate pass-through. Recent rises in food prices, which reflect food shortages, are In view of these developments, the growth expected to continue through the end of 2013. projection for 2013 was lowered to 6.6 Further depreciation of the exchange rate and Table I.2: Rwanda’s GDP growth is expected to slow in 2013 before accelerating in 2014 (Annual percent change) 2013 projection 2014 projection H1 2013 H2 2013 Item 2011 2012 November Actual Projection November May 2013 May 2013 2013 2013 GDP 8.2 8.0 7.0 6.6 5.9 7.3 7.5 7.2 Agriculture 4.7 3.0 3.3 3.5 5.4 1.7 3.8 5.0 Food crops 5.0 3.2 3.1 3.1 5.4 1.0 3.1 5.1 Export crops 2.9 -9.3 12.2 20.8 25.2 18.2 28.9 9.9 Industry 17.6 7.2 12.2 12.4 13.5 11.4 10.1 9.4 Manufacturing 8.1 -2.9 5.2 2.8 0.8 4.5 7.2 8.0 Construction 23.6 15.4 16.7 16.9 21.3 15.1 11.8 10.7 Services 8.9 12.2 8.0 7.1 4.2 9.8 9.5 8.5 Public expenditure led services 14.3 9.1 7.2 7.8 5.5 10.0 6.5 12.5 Other services 7.0 13.3 8.4 6.9 3.8 9.8 10.6 7.0 Source: NISR and World Bank Staff estimates 22 Rwanda Economic Update | Edition No. 5 I. Recent Economic Developments and Prospects an increase in domestic fuel prices are upside The government’s policy buffers seem to be risks to the inflation outlook. In November diminishing in the face of slowing growth in the 2013, domestic fuel prices at the pump were short-term. The increase in domestic borrowing increased by 3 percent, and experience shows has already crowded out the private sector. that the movements in fuel prices tend to affect Although the signaling effect of the policy rate is overall prices in Rwanda. The weakening of the limited, it is important that the BNR continue to Rwandan franc is expected to increase import show its monetary stance to the financial sector. prices, which account for 20.5 percent of the In the face of uncertainty about aid inflows, household consumer basket. It is important to it is likely that the BNR will be constrained to monitor inflation trends in the case of recovery in meet the market demand for foreign exchange. domestic demand. On fiscal policy, accelerating domestic resource mobilization and ensuring timely disbursement Rwanda’s external accounts remain vulnerable of foreign aid are critical to maintain a certain to uncertainty in the global economy. Although level of public expenditures. The composition conditions improved in recent months, the global of public expenditures can be revisited to economy continues to suffer from the effects of prioritize expenditures for productive sectors. the recession in the United States and Europe. Until domestic resource mobilization advances Advanced economies recently gained some and uncertainty about aid inflows to Rwanda, speed, led by the United States, where activity the authorities will have to implement a prudent will move into higher gear as fiscal consolidation fiscal policy stance, emphasizing efficiency and eases and monetary conditions stay supportive. effectiveness in public spending. Moreover, in April 2013, projections of world economic growth were revised downward by the In implementing monetary and fiscal policies, IMF’s World Economic Outlook of October 2013 the government needs to be more careful than to 2.9 percent for 2013 and 3.6 percent 2014, it was in the past. Slower growth may reinforce down from 3.3 percent for 2013 and 4.4 percent demand for further loosening of monetary and for 2014. This downward revision was anchored fiscal policies, with the risk of stoking inflationary by slowing growth in China and a growing pressures and reversing fiscal consolidation. number of emerging markets, which are coming Potential increases in public consumption would off cyclical peaks. However, old problems—a require the government to use its gross reserves fragmented financial system in the euro area or borrow from the domestic market, depleting and worrisomely high public debt in all major gross reserves or further crowding out the private advanced economies—remain unresolved and sector. Offsetting these risks will depend on could trigger new crises. Therefore, Rwanda’s how Rwanda combines policy instruments to external position remains vulnerable. Lower bolster growth in the nearterm. A cursory look at commodity prices and reduced aid and foreign experience suggests that the aftermath of crises direct investment flows will erode the gains of appears to be harder to manage than the crises Rwanda’s main export products recorded in themselves and that both monetary policy and the first half of 2013 Rwanda’s current account fiscal policy are limited in their ability to react to deficit is expected to persist at about 10 percent downside risks (Box I.10). of GDP, roughly in line with the 2012 level.   Rwanda Economic Update | Edition No. 5 23 I. Recent Economic Developments and Prospects Box I.10 Rwanda’s experience in managing economic crises and their aftermath The government has gained significant experience in crisis management since 2008/09, when Rwanda had to manage both external and internal crises, including the global financial crisis and the domestic credit crunch, which led to a sharp decline in real GDP growth, from 11.2 percent in 2008 to 6.2 percent in 2009. In 2012/13, the government had to manage the aid shortfall. During both episodes, it managed to preserve the domestic economy from first-round effects, thanks to its sage fiscal and monetary policies. However, the aftermath of crises appears to be harder to manage than crises themselves. The government’s response to crises reflects the prevailing economic environment and policies. In 2009, the government accelerated its fiscal expansionary stance, backed by strong aid inflows. Since 2006, it has embraced an expansionary fiscal policy to reduce poverty by improving education, health, infrastructure, and agricultural productivity. Between 2008/09 and 2009/10, the overall deficit before grants increased from 10.7 percent to 13.6 percent of GDP. In addition, the government undertook a series of seemingly contradictory monetary policies. On the one hand, it adopted measures to stimulate liquidity, injecting liquidity into the market in January 2009 (for the first time since 2004), through a reverse repo operation; opting not to roll over maturing T-bills; reducing the reserve requirement ratio from 8 percent to 5 percent in February 2009; and introducing a 3- to 12-month refinancing facility against collateral (T-bills, bonds, and foreign deposits) at a fixed rate of 11.5 percent a year in July 2009. On the other hand, it increased the policy rate from 8 percent to 9 percent in January 2009. Second-round effects affected the economy in 2009. Credit growth, which was decelerating in early 2009, became negative throughout the three months ending November 2009. Although GDP growth remained high during the intervention period, it slowed as a result of weaker domestic demand. Although consumption lost momentum, it was the only component that largely sustained growth; investment and external demand contracted severely. Expansionary fiscal policy boosted consumption expenditures. At the same time, the central bank maintained refinancing facilities, stimulating credit to the private sector. In the wake of the aid shortfall in 2012, the government’s responses were in line with its fiscal consolidation path. On the fiscal side, the government sought to preserve the overall deficit target of the original budget while sustaining priority spending. In line with this fiscal consolidation, it maintained the overall fiscal deficit close to the original target. On the monetary side, responses aimed to cushion the economy from downside risks of fiscal responses. The central bank kept its policy rate at a relatively high level, anchoring inflation expectations from sizable government borrowing on the domestic market. With positive spillovers from the maintenance of public spending, growth in domestic demand expanded beyond domestic production, requiring high imports. Increased demand for imports put pressures on the foreign exchange market, prompting the central bank to intervene by selling U.S. dollars to foreign exchange bureaus. Inflation remained modest, and credit growth to the private sector was high. In 2013, lagged effects of the decline in aid hit the main growth driver in the previous crisis: domestic demand. Investment and public consumption weakened, and private consumption contracted. The challenge is that both monetary and fiscal policies have only limited ability to react to downside risks from lagged effects of the decline in aid. There is no room for an expansionary policy stance. I.5.2 Prospects for the Medium-Term: • rural development, to bring the national poverty Tied to the EDPRS 2 rate below 30 percent; T • productivity and youth employment, to ensure he EDPRS 2 aims to accelerate growth that growth and transformation are supported and reduce poverty, including extreme by appropriate skills; and poverty. It seeks to do so while reducing aid • accountable governance, to improve service dependency, thereby increasing self-reliance. delivery and increase citizen participation in and The policy builds on the strengths and addresses satisfaction with the delivery of development. the shortcomings of the EDPRS 1. These thematic areas are supported by eight The EPDRS 2 includes four thematic areas: foundational issues (long-term priorities, on • economic transformation, to achieve high and which substantial progress was achieved during sustained growth and restructure the economy the EDPRS 1) and seven cross-cutting issues. toward services and industry; 24 Rwanda Economic Update | Edition No. 5 I. Recent Economic Developments and Prospects The foundational issue areas are macroeconomic policy formulation, more prudent macroeconomic stability, demographics, food security, basic assumptions than those embodied in the EDPRS education, primary health care, rule of law, public 2 should be used. The government has already financial management, and decentralization. The acknowledged the need to pursue this approach, cross-cutting issue areas are capacity building, including in the budget framework paper for the environment, gender, regional integration, 2013/14–2015/16, which assumes annual real HIV/AIDS and noncommunicable diseases, GDP growth of 7.0–7.5 percent in the medium- disaster management, and social inclusion. term. The macroeconomic framework underlying Achievement of EDPRS 2 objectives will not the EDPRS 2 reflects the government’s be possible without reaping opportunities ambitious growth and poverty reduction goals such as the demographic dividend. Rwanda is for the next decade. In order to achieve middle- entering a demographic window of opportunity, income status by 2020, Rwanda needs to grow characterized by falling youth dependency at an average annual real rate of 11.5 percent ratios and a substantial increase in the share between 2012 and 2020. Attaining this target of working-age adults in the population. The requires significant acceleration from the already increased share of working-age adults will high growth levels of the last decade. EDPRS 2 provide an opportunity to accelerate growth. growth targets seem to be more ambitious than However, Rwanda has to create a conducive potential growth (Box I.11). A key challenge environment if it is to reap the “demographic to attaining the target is the need to fund and dividend.” Part II of this report focuses on the implement public and private investment in view demographic dividend and the implications for of declining official external financing (at least in economic growth in the coming decades. relative terms). For the purposes of medium-term Box I.11 Looking ahead How fast could Rwanda grow? Rwanda is one of the fastest-growing economies in Sub-Saharan Africa, with average annual growth of about 8 percent since 2001. An estimate of Rwanda’s potential growth rate can help determine whether such high growth is sustainable in the medium or long term. Potential growth refers to the growth rate of the economy when all production factors (capital, labor, and technology) are fully utilized and aggregate economic activity is not producing any inflationary pressures. The estimation of the potential growth rate is based on a Cobb-Douglas production (Annex IA). Estimated results indicate that Rwanda’s potential GDP will grow steadily through 2020. Potential GDP growth in 2012 was expected to reach 9.0 percent before decelerating to 8.0 percent in 2013 and 7.4 percent in 2020 (Figure BI.11.1 and Figure BI.11.2). According to these results, real GDP will double by 2020. These findings suggest that the growth rates of the past decade may be sustainable over the medium and long term if the economy efficiently uses the available production factors. The estimated average potential growth rate for 2012–20 of 7.8 percent almost equals the average growth rate in the past decade of 7.7 percent, suggesting that Rwanda’s economy expanded at a rate close to its potential. These findings are confirmed by applying the statistical Hodrick-Prescott (HP) filter method to Rwanda’s growth over the past decade. The HP filter method decomposes actual GDP into its long-term trend (“potential GDP”) and cyclical components. The average growth rate of the estimated long-term trend component of Rwanda’s GDP for 2001-11 equals 7.8 percent (see Figure BI.11.2), suggesting that the economy grew at its full potential. The same growth rate may be expected in the future if the economy is functioning close to its potential. Rwanda Economic Update | Edition No. 5 25 I. Recent Economic Developments and Prospects Figure BI.11.1: Rwanda expects the same growth in 2012–20 as in the previous decade … Figure BI.11.2: … mainly driven by physical capital (Estimated potential GDP growth in 2001–11 and projected (Projected potential growth rate and decomposition of potential GDP growth in 2012–20, percent) Rwanda’s GDP, 2012–20, percentages) 10 Average potential 10 growth rate 9.0 9 Historically estimated Predicted potential 8.0 7.9 9 7.8 7.7 7.6 potential GDP growth rate GDP growth rate 8 7.5 7.4 7.4 7 8 6 7 5 4 6 3 2 5 1 0 4 2012 2013 2014 2015 2016 2017 2018 2019 2020 2001 2003 2005 2007 2009 2011 2013 2015 2017 2019 Physical Capital Human Capital Labor TFP GDP Growth Source: World Bank staff estimates based on data from WBI Source: World Bank staff estimates based on data from WBI The input factor decomposition of the estimated increase in the average years of schooling in Rwanda potential growth rates of Rwanda for 2012-20 (see could be the explanation of the small contribution of the Figure BI.11.1) indicates that physical capital makes the human capital to GDP growth. largest contribution (3.2 percentage points), followed by total factor productivity (2.7 percentage points) and These findings are similar to the findings in the human capital–adjusted labor (1.9 percentage points). IMF’s 2013 Fall Regional Economic Outlook Report Decomposing the human capital—adjusted labor for Sub-Saharan Africa. This report estimates that input on human capital and labor reveals that human the average contribution to GDP growth over 2000-10 capital by itself contributes only 0.2 percentage points was 2.8 percent for labor, 2.4 percent for capital, and to overall GDP growth. This small contribution can be 2.3 percent for total factor productivity. It estimates explained by the low average years of schooling (3.2 the contribution of human capital to growth at just 0.2 years in 2000 and 4.0 years in 2011). This slow pace of percentage points over the 2000-2010 period. 26 Rwanda Economic Update | Edition No. 5 I. Recent Economic Developments and Prospects Annex IA: The Cobb-Douglas Production Function Estimation of the potential growth rate is based on a Cobb-Douglas production function, which estimates potential output (Y) as a function of the stock of physical capital (K), human capital–adjusted labor (H), and technology (A),which determine total factor productivity (TFP): Superscript α indicates the share of physical capital, superscript 1–α indicates the share of human capital–adjusted labor in output, and superscript γ measures the returns to scale of the inputs. The value of γ can be equal to, greater than, or less than 1, suggesting constant, increasing, or decreasing returns to scale. H is calculated as a function of the labor force adjusted for employment and participation rates and the level of human capital, which is estimated as a function of returns to education and average years of schooling. In estimating the potential growth rate of Rwanda’s GDP, the researcher needs to make assumptions about the values of the parameters of the model (α, γ, returns to education, and the capital depreciation rate). In addition, the initial and final values of the input variables (the working-age population ratio, participation rate, unemployment rate, TFP growth rate, gross capital formation, and average years of schooling) for the forecasting period need to be exogenously determined. The initial values refer to the last available data for the variables used (2011 in this case). The final value refers to their expected value at the end of the forecasting period (2020). The baseline scenario assumes no structural shift in economic relations. Hence, the parameters of the model are assigned values that are predefined in the economic literature. The initial and final values of the input variables are assumed to be equal for the working-age population ratio, labor participation rate, unemployment rate, TFP growth rate, and gross capital formation. The input and final values of the average years of schooling variable are assumed to be different: the initial value equals the most recent available data from the dataset, whereas the final value is projected by the linear trend fitted from the past 11 years of this variable (see Annex Table IA.). Annex Table IA.1: Assumed values of the parameters and initial and final values of the input variables in the model used (percent, except where otherwise indicated) Parameter Value Input variable Initial value Final value (2011) (2020) Share of physical capital (α) 35 Working-age population ratio 53.5 53.5 Share of human-adjusted capital (1-α) 65 Participation rate 89.6 89.6 Returns to scale of the inputs (γ) 1 Unemployment rate 4.7 4.7 Returns to education 5 Total factor productivity (TFP) growth rate 2.7 2.7 Capital depreciation ratio (years) 5 Gross capital formation 25.6 25.6 Average years of schooling 4.0 4.5 Source: Values of parameters are from Ghosh and Kraay 2000 and Kuepie, Nordman, and Roubaud 2009 Rwanda Economic Update | Edition No. 5 27 PART TWO Special Focus: Harnessing the Demographic Dividend “Increasing levels of female education and of women’s participation in the nonfarm labor market are behind the rapid drop in fertility transition, thus kick-starting Rwanda’s demographic transition” II. Special Focus: Harnessing the Demographic Dividend A lthough the economy is cooling off in the short term, favorable demographic dynamics provide an opportunity for young women are increasingly turning to nonfarm employment, resulting in a decline in the share of young women engaged in agriculture. These accelerating growth in the long term. Fertility changes were associated with a salient shift in rates dropped sharply between 2005 (6.1 children fertility preferences: young women wanted to per woman) and 2010 (4.6), kick-starting have 4.4 children in 2005 but just 3.0 in 2010 Rwanda’s demographic transition. During this (Rwanda DHS 2005, 2010). transition, the size of the labor force is projected to increase sharply while the share of dependent The drop in fertility will have significant children (younger than 15) in the population will consequences for the age structure of the decrease. As noted in the EDPRS 2, Rwanda population and substantial implications for will have the opportunity to take advantage of a economic development. The share of working- bulging labor force over the coming decades to age adults in the population is projected to increase dramatically increase the productive capacity of from 53 percent in 2010 to 64–67 percent by 2050 the economy. This unique demographic window (depending on the pace of the fertility decline), of opportunity for accelerated growth is known with the youth dependency ratio (the average as the “demographic dividend.” number of dependent children per working-age adult) falling from 0.84 to 0.45 or 0.37 by 2050. The drop in fertility was unusually rapid. Over These population shifts bode well for economic a period of just five years, the average number development. Estimates from a demographically of children a woman can expect to have at the based forecasting model suggest that income end of her life decreased by 1.5. The decline per capita adjusted for purchasing power could was driven mainly by a marked improvement almost triple between 2010 and 2050 if fertility in female education. Between 2005 and 2010, declined at a medium pace and more than triple average educational attainment of women 15– if faster reductions in fertility could be achieved. 49 increased by almost one year, explaining Combined with the economic transformation more than a fifth of the drop in fertility. agenda of the EDPRS 2, the population shifts Increasing household wealth and rising women’s have the potential to spur rapid growth over the participation in the labor force were important decades to come. secondary drivers behind the fertility decline. In particular, the increased participation of women This demographic dividend is not guaranteed. in the nonfarm labor market was associated with Missing this golden opportunity for rapid a drop in fertility. economic growth would have the opposite effect, resulting in economic drag and widespread Young women are at the core of Rwanda’s underemployment of human resources. Cashing fertility transition. The fertility drop has been in on the demographic dividend requires a greatest for women 15–24. The decline was driven conducive policy framework aimed at fostering by a marked evolution in their socioeconomic a productive workforce through investments fortunes: average educational attainment for in child health and post-primary education; young women 15–24 increased by 36 percent boosting savings by sustaining price stability between 2005 and 2010, the proportion of young and encouraging competition in the financial women that never went to school fell by half, and institutions; and creating a flexible labor market almost twice as many young women were still in able to accommodate the burgeoning working- school in 2010 than in 2005. On the labor market, age population. Although these policies are not 30 Rwanda Economic Update | Edition No. 5 II. Special Focus: Harnessing the Demographic Dividend a recipe for success, they go a long way toward to sketch the drop in fertility and illuminate its enabling countries to make the most of beneficial main drivers. The potential implications for population shifts. future income levels are examined in section three, which also sheds light on the policies that This special focus is organized as follows. are required, though not necessarily sufficient, The first section explains why rapid changes to take advantage of the demographic shifts. in fertility matter for economic growth and The last section briefly elaborates on the end of describes the projected age-structure effects of the demographic transition, characterized by a the recent decline in fertility in Rwanda. The renewed rise in dependency rates and an aging second section exploits data from the 2005 and population. 2010 Demographic and Health Surveys (DHS) II.1 The Demographic Dividend: Why Fertility and Age Structure Matter for Economic Growth F ertility is linked to economic growth through its influence on the age structure of a population. Sharp drops in fertility reduce the youth dependency ratio, the average number of youths per working-age adult, and raise the proportion of adults of working age in the population. The net result is a larger labor force and fewer dependents, which creates a demographic opportunity for accelerated growth. Rwanda will experience this opportunity during the coming decades: the proportion of working-age adults is projected to increase from 53 percent in 2010 to 64 percent in 2050, while the youth dependency ratio will drop from 0.84 to 0.45, according to the medium-fertility scenario of World Population Prospects. T he relationship between population size and population growth and economic performance has been a subject of debate number of dependent children per working-age adult). Countries with high dependency ratios are likely to devote substantial resources to children, for centuries. The pessimistic view can be creating a heavy fiscal burden and limiting the traced back to Malthus, who argued that rapid scope for productive investments in other areas.6 population growth cancels out improvements In contrast, countries with a large proportion in income, leading to lower per capita incomes. of working-age adults in the population—and The more optimistic view argues that increased hence a low youth dependency ratio—have an population pressure will spur productivity gains opportunity to dramatically raise output levels and technological innovations, offsetting any and living standards. negative impact of population growth on per capita income.5 Every country experiences dramatic changes in age structure as it moves through the stages Recently, scholars have argued that is not of the demographic transition—the move from the size of the population that matters but a high-fertility/high-mortality society to a low- rather its structure. Across the developing fertility/low-mortality one (Box II.1). During world, children under the age of 15 make up the early stages of the demographic transition, almost half of the population, translating into sharp drops in child mortality are not matched by high youth dependency ratios (the average corresponding drops in fertility, leading to rapid 5 This is a highly simplified summary of the debate, but it captures the main strands of thought. 6 Of course, investments in children’s health and education are highly productive in the longer run. Rwanda Economic Update | Edition No. 5 31 II. Special Focus: Harnessing the Demographic Dividend population growth and high youth dependency extent explain Ireland’s economic success in the ratios. As fertility eventually begins to fall in 1990s (Bloom and Williamson 1998; Bloom and response to lower child mortality, each successive Canning 2003). In this context, it is encouraging cohort of children tends to be smaller. The change that many countries in Sub-Saharan Africa started produces a bulge cohort of children that gradually their fertility transition in the 1990s (although makes its way through the age structure of the the transition seems to have stalled in the 2000s population. As this cohort grows up and fertility [Bongaarts 2008]). continues to decline, it creates a bulge of working- age adults. Because fertility is falling, there are The age structure effects of the transition are relatively few children. And because mortality plotted in Figure II.1. In Sub-Saharan Africa rates for the older generations are still high, there as a whole, the period between 1950 and the are relatively few elders to take care of. The result early 1990s was characterized by a small and is more people in the labor force, fewer people to declining labor force (in relative terms) and a support, and a unique demographic opportunity high youth dependency ratio. In the mid-1990s, for economic growth called the “demographic the proportion of working-age adults began to dividend” (Bloom, Canning, and Sevilla 2003). increase and dependency ratios began to fall, as fertility started to decline from a high base. This Falling fertility does not promote economic trend is projected to continue over the coming growth only through changes in the age decades: the proportion of working-age adults is structure of the population. As families become projected to increase from 53 percent in 2010 to smaller, economic behaviors tend to change, too. 61 percent in 2050, and the number of dependent Women with fewer children are more likely to children per active adult is estimated to drop from enter the labor market, adding to the increased 0.8 to 0.5 (UN 2013). labor supply resulting from the shifts in age structure. Because the opportunity cost of Figure II.1: The youth dependency ratio in Sub-Saharan Africa is projected to drop substantially over the next 40 years having extra children is higher when women (Proportion of working-age adults in total population and youth participate in the labor market, employed dependency ratio, percent) women typically choose to have fewer children, 1.0 leading to a virtuous circle of falling fertility 0.9 and wealth creation. When families have fewer children, they also tend to invest more in their 0.8 children’s education, laying the foundations for 0.7 an educated and productive future workforce. 0.6 Lower youth dependency ratios are also linked to higher domestic savings rates, facilitating 0.5 productive investment (Bloom, Canning, and 0.4 Sevilla 2003; Bloom and others 2009; Becker, 0.3 Murphy, and Tamura 1990; Galor 2006). 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Proportion working age Youth Dependency Ratio The economic impacts of the demographic Source: UN 2013 Note: Figures are based on the medium-fertility scenario dividend can be substantial. Estimates indicate that population dynamics related to the demographic dividend accounted for up to one- Between 2005 and 2010, total fertility rates in third of the East Asian Miracle (the extraordinary Rwanda dropped from 6.1 children per woman growth experienced by many East Asian to 4.6. Over a period of just five years, the average countries between 1965 and 1990) and to a large number of children a woman can expect to have 32 Rwanda Economic Update | Edition No. 5 II. Special Focus: Harnessing the Demographic Dividend Box II.1 The demographic transition Societies move from a high-mortality/high-fertility state to a low-mortality/low-fertility one in four distinct phases The demographic transition refers to the transition from high mortality and fertility rates to low mortality and fertility rates as a country develops and moves from a preindustrial to a modern society. The transition involves four or possibly five stages (see Figure BII.1.1). In stage 1, birth rates and death rates are high and roughly in balance. Populations are young and population growth slow. In stage 2, death rates drop rapidly, thanks to improvements in public health, basic education, and food supply. Birth rates, however, remain high, resulting in rapid population growth. In stage 3, birth rates begin to fall, thanks to access to contraception, urbanization, improvements in female education and women’s status, and a gradual shift in the nature of economic Figure BII.1.1: The demographic transition involves activity away from subsistence agriculture. Population at least four stages growth levels off somewhat but remains high because (Evolution of birth rates, death rates, and total population of the large cohorts of people born in stage 2. Finally, throughout the demographic transition in stage 4, a new equilibrium is reached where both birth and death rates are low and population is more 1 2 3 4 5 or less stable. A number of developed countries (for example, Germany, Italy, and Japan) have now entered a fifth phase of the transition, defined by a shrinking Birth Rate population as birth rates fall below replacement level. Births/Deaths per 1,000 Most countries in Sub-Saharan Africa are now in the third stage of the demographic transition. This stage is characterized by falling youth dependency Death Rate ratios and a gradual increase in the proportion of working-age adults. The steeper the decline in fertility in this stage, the steeper the decline in dependency ratios and the greater the potential for accelerated Total Population economic growth through changes in the age structure. With its recent steep decline in fertility, Rwanda is Time relatively well positioned to reap the rewards of a sharply expanding labor force in the decades to come. Source: https://www.e-education.psu.edu/geog030/node/205 at the end of her reproductive period dropped will account for 67 percent of Rwanda’s total by 1.5. This decline is one of the fastest in the population by 2050, up from 53 percent in 2010. history of the DHS.7 Rwanda’s fertility level is This figure is substantially higher than that of now the same as Kenya’s and much lower level other African countries.10 At the same time, the than the average for Sub-Saharan Africa of 5.1. youth dependency ratio in Rwanda is projected According to the latest DHS data, eight countries to drop faster than in other African countries, to (excluding South Africa) have lower fertility 0.45 (medium-fertility scenario) or 0.37 (low- rates than Rwanda.8 fertility scenario) by 2050, down from 0.84 in 2010 (see Figure II.3). The rapid fertility decline has implications for the age composition of the population in Together with overall population growth, the coming decades. In the medium-fertility changes in the age structure will have far- scenario, by 2050, the working-age population is reaching consequences for Rwanda. The projected to reach 64 percent of the population population is projected to increase from 10.5 in Rwanda, 61 percent in Sub-Saharan Africa, million in 2012 to 22.3 million (low fertility) and 60 percent in the EAC (see Figure II.2).9 In or 25.3 million (medium fertility) in 2050, the low-fertility scenario, working-age adults with the size of the labor force almost tripling, 7 Other countries that experienced rapid drops in total fertility rates include Kenya (drop of 1.3 in five years), Ghana (drop of 1.2 in five years), and Nigeria (drop of 1.3 in nine years). 8 The countries are Cape Verde, Ghana, Mauritania, Gabon, Sudan, Zimbabwe, Lesotho, and Namibia. 9 The UN Population Division projects fertility using two probabilistic models, according to which fertility tends toward 2.1 children per woman in the long run. The future fertility paths for a country are generated by taking into account past fertility trends in the country, in countries in the same region, and in all countries. The probabilistic approach permits the simulation of 100,000 future fertility paths for each country. The median of all the projected fertility paths is used as the medium-fertility scenario. Rwanda Economic Update | Edition No. 5 33 II. Special Focus: Harnessing the Demographic Dividend from 5.7 million to 15.0 million (low fertility) substantially increase national output and income or 16.3 million (medium fertility) by 2050. levels. Failing to do so will result in large-scale Coping with this situation will be the defining unemployment and underemployment of human challenge for Rwanda over the decades to resources and the squandering of this unique come. If put to productive use, the dramatically demographic opportunity. increased labor force offers the opportunity to Figure II.2: Rwanda’s labor force will grow more rapidly Figure II.3: … and youth dependency ratios will drop than that of other countries… much faster (Proportion of working-age adults in population, percent) (Youth dependency ratio, percent) 0.7 1.1 0.9 0.6 0.7 0.5 0.5 0.4 0.3 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Rwanda Medium Fertility Rwanda Low Fertility Rwanda Medium Fertility Rwanda Low Fertility SSA Medium Fertility EAC (excl. Rwanda) Medium Fertility SSA Medium Fertility EAC (excl. Rwanda) Medium Fertility Source: UN 2013 Source: UN 2013 II.2 Explaining Rwanda’s Fertility Drop T he total fertility rate in Rwanda dropped from 6.1 children per woman in 2005 to 4.6 in 2010. This rapid drop was driven mainly by young women. The drop in fertility comes during a time of rapid social and economic development, marked by improvements in female education, increasing household wealth, increasing participation of women in the labor force, and a sharp uptake in the use of contraception. Higher female education levels emerge as the main driver of the drop in fertility, explaining 23 percent of the decline in fertility between 2005 and 2010. Increased household living standards accounted for another 15 percent, and higher labor force participation explained 5 percent. The effect of education is particularly salient for young women, explaining two-thirds of their decline in fertility between 2005 and 2010. II.2.1 Key Evolutions Between 2005 and 2010 number of children born per woman—dropped more for younger women than for older ones. Fertility declined across the board but Women 25–29 in 2005 had on average 2.3 especially among younger women children; women the same age in 2010 had only A t the time of the 2005 DHS, women 15–49 had on average 2.7 children. By 2010, this figure had dropped to 2.4, a decline of 10 percent. 1.9 children, a decrease of more than 20 percent. A similar pattern is evidenced for women 20–24 group. Declines are weaker for older women: Table II.1 shows that cumulative fertility—the cumulative fertility dropped by 6 percent for 10 The low-fertility scenario projects fertility at 0.5 children below the medium-fertility case. 34 Rwanda Economic Update | Edition No. 5 II. Special Focus: Harnessing the Demographic Dividend women 30–34, 5 percent for women 35–39, As often observed in other countries, the drop and about 10 percent for women 40–49. These in fertility coincided with a sharp drop in child changes indicate that between 2005 and 2010, mortality. A main hypothesis in the demographic women, especially younger women, decided to literature is that declines in child mortality reduce have fewer children (or have them later). Fertility the need for parents to have many children. As actually increased for the youngest women (15– parents realize that fewer children are likely to 19 at the time of the surveys). The size of the die in childhood, they can give birth to fewer increase was negligible, however (0.01 children). children in order to attain their desired number of offspring. As shown in Figure II.5, child mortality Even more salient than the drop in fertility in Rwanda halved between 2000 and 2010, with is the drop in the ideal number of children. fertility responding five years later. In 2005, Rwandan women wanted to have 4.5 children by the end of their reproductive period. Female education and labor force participation By 2010, the number had dropped to 3.4. The are on the rise ideal family now contains one child fewer than it did in 2005. As shown by Figure II.4 this change was driven by the younger generations. In 2005, women 15–19 wanted to have almost F emale education and women’s participation in the labor force are among the main determinants of fertility. Both variables are five children. In 2010, they wanted fewer than believed to influence fertility by raising the three children. Although the desired number cost of women’s time (the opportunity cost) of children in 2010 is lower than in 2005 for (Mincer 1963; Becker 1981). Women with more all women, the effect is much less pronounced education command better opportunities in the for the older generations. The ideal number of labor market, which makes having extra children children for women 45–49 in 2005 was 4.8, only costly because of the time “lost” bearing and marginally more than in 2010 (4.4), for example. raising children. As a result, more educated The larger changes for younger women hint at a women decide to have fewer children to limit change in mentality and preferences among the the time away from the labor market. In the same young generation, associated with the sense of way, expanding labor market opportunities for development and modernization that has taken women raises the cost of having extra children, hold in Rwanda over the past decade (Box II.2). driving down fertility levels. Figure II.4: Where has all the child wish gone? Table II.1: The decline in fertility in Rwanda between (Number of children women wish to have at the end of the 2005 and 2010 was greatest among younger women reproductive period) (Number of children ever) Age Percentage 6 2005 2010 difference interval 15–19 0.04 0.05 25.0 5 20–24 0.77 0.62 –19.5 25–29 2.33 1.86 –20.2 4 30–34 3.66 3.44 –6.0 35–39 4.96 4.7 –5.2 40–44 6.2 5.5 –11.3 3 45–49 7.02 6.37 –9.3 Total 2.68 2.42 –9.7 2 Source: DHS 2005, 2010 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Age Group 2005 2010 Source: DHS 2005, 2010 Rwanda Economic Update | Edition No. 5 35 II. Special Focus: Harnessing the Demographic Dividend Figure II.5: The odds of surviving childhood in Rwanda increased dramatically since 2000… Figure II.6: … reducing fertility (Under five mortality per 1,000 live births) (Total fertility rate) 250 7 6.2 207 6.1 200 6 5.8 182 5.5 163 5 150 135 4.6 102 100 4 50 3 1992 2000 2005 2008 2010 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 Source: DHS 1992, 2000, 2005, 2010 Source: DHS 1992, 2000, 2005, 2010 Rwanda experienced notable improvements in women 15–49 increased by almost one year (from female education between 2005 and 2010. The 3.9 in 2005 to 4.7 in 2010), and the proportion magnitude of the increase is remarkable given of women who never went to school dropped the short time period between the surveys. The from more than 23 percent to less than 16 percent average number of years of education among (Table II.2). In line with the rapid expansion in Box II.2 Once again, change is made by youth Cohort effects dominate the drop in fertility When studying the evolution of an age-related variable such as fertility, researchers often want to know whether the evolution reflects general shifts in social, economic, or cultural environments that affect all age groups simultaneously or rather changing behaviors or preferences across groups of people who were born and grew up during a specific time period. The first set of influences is called Period effects. They capture variation over time that affects a whole population. The second set of influences is called Cohort effects. They refer to the effects of having been born during a particular time period and reflect the impact of early-life exposure to socioeconomic, behavioral, or environmental circumstances that persist over time to produce differences in life-course outcomes for specific cohorts. Cohort effects are particularly important, because they are the essence of social change. Figure BII.2.1: Young cohorts are driving down fertility Applying a statistical model to the data from the (Cohort contribution to fertility, controlling for age and 2005 and 2010 DHS shows that Rwanda’s fertility time period, percentage) drop reflects both period and cohort effects. The 80 decline in fertility can be explained by both the rapid social and economic development that Rwanda 60 experienced between 2005 and 2010 and by changing behaviors and preferences across successive cohorts. 40 The cohort effects emerge as dominant, explaining 20 up to four times as much of the drop in fertility as the period effects. Figure BII.2.1 shows that the 0 younger cohorts are driving the drop: controlling for the effect of age (older women will on average -20 have more children, regardless of what cohort they belong to), being part of an older cohort increases -40 fertility, whereas being part of a younger cohort -60 reduces it. The cohort effects are significantly negative for the cohorts born in 1981 or later, -80 cohorts that entered their reproductive period 1956 1961 1966 1971 1976 1981 1986 1991 after the 1994 events and grew up in a period of unprecedented change in Rwanda. Source: DHS 2005, 2010 36 Rwanda Economic Update | Edition No. 5 II. Special Focus: Harnessing the Demographic Dividend access to education, the proportion of women the youngest group were still in school in 2010, still in school at the time of the survey increased compared with only 20 percent in 2005 (Annex from 9 percent in 2005 to more than 15 percent Table IIA.1). in 2010. The improvement in education between 2005 and 2010 is helped by the normal passing of The decline in education for women in the time: moving from 2005 to 2010 sheds the oldest middle age group may seem puzzling. In both cohort in 2005 (women who were 45–49 in 2005) developed and developing countries, each cohort and replaces them with the youngest cohort in usually has more education than the previous 2010 (women who were 15–19 in 2010). The one. The fact that this pattern did not occur for older cohort had on average of only 2.2 years women 25–34 is most likely the result of the of education; the youngest women in the 2010 genocide: women who were 25–34 during the survey already had 5.4 years of education 2010 survey were born between 1976 and 1985. (and counting). Many of them were in school when the events erupted, bringing a premature and permanent end The improvements in education between 2005 to their educational aspirations. and 2010 were driven mainly by the younger cohorts. Figure II.7 and Figure II.8 show the Women’s participation in the labor market evolution of education indicators by three age also increased significantly between 2005 categories: 15–24, 25–34, and 35–49. The and 2010. In the 2010 DHS, almost 84 percent average number of years of education increased of women 15–49 were economically active, up by 1.4 for the youngest group, decreased by 0.2 from 73 percent five years earlier (Table II.2). for the middle group, and increased by 1.0 for The increase is particularly salient for nonfarm the oldest group (Figure II.7). Primary school occupations: the percentage of women with completion almost tripled between 2005 and 2010 a main occupation outside agriculture almost for the youngest group, declined marginally for doubled between 2005 (10 percent) and 2010 (19 the middle group, and increased by 8 percentage percent). Participation rates in agriculture also points for the oldest group (Figure II.8). In increased, albeit modestly (from 62 percent in addition, more than 37 percent of women in 2005 to 65 percent in 2010). Figure II.7: Average years of education increased most for the youngest women… Figure II.8: … as did primary school completion (Years of education) (Percent completing primary school) 6 25 23.3 5.3 20.9 5 4.7 4.5 20 3.9 4.1 4 15 3.1 12.7 3 12.0 11.9 10 8.9 2 1 5 0 24-34 35-49 0 15-24 Age Group 15-24 24-34 35-49 2005 2010 2005 2010 Source: DHS 2005, 2010 Source: DHS 2005, 2010 Rwanda Economic Update | Edition No. 5 37 II. Special Focus: Harnessing the Demographic Dividend Table II.2: Significant changes in the determinants of fertility between 2005 and 2010 (Means of key variables for women 15–49) Variable 2005 2010 Mean difference Cumulative fertility (number of children at time of survey) 2.68 2.42 –0.26*** Age (years) 28.30 28.40 0.1 Years of education 3.90 4.73 0.83*** Percentage that completed primary 13.10 19.00 5.9*** Percentage still In school 9.00 15.40 6.4*** Percentage that never went to school 23.40 15.50 –7.9*** Percentage literate 59.10 68.40 9.3*** Years of education of household head 3.57 3.95 0.38*** Percentage who work 73.10 83.70 10.6*** Percentage with non-agricultural occupation 10.60 19.00 8.4*** Percentage with agricultural occupation 62.40 64.70 2.3*** Household wealth (asset index) 0.14 0.22 0.362*** Number of observations 11,321 13,671 Source: DHS 2005, 2010 Note: *** Statistically significant at the 1 percent level Young women are driving the higher young women engaged in agriculture remained participation rates. The increase in labor force almost unchanged between 2005 (50 percent) and participation was higher for younger women (21 2010 (49 percent). percent) than for women in the middle (13 percent) and oldest (8 percent) age groups (Figure II.9). To summarize, female education and The increase in labor force participation rates participation in the labor market improved among young women is solely the result of their substantially between 2005 and 2010. Both increased engagement in nonfarm occupations. changes likely increased the value of women’s In 2010, more than 24 percent of young women time and hence the cost of having children. had a main occupation outside agriculture, more This evidence is supported by data from the than twice the percentage in 2005 (11 percent) 2006 and 2011 Integrated Household Living (Figure II.10). In contrast, the proportion of Conditions Surveys (EICV), which show that Figure II.9: Female labor force participation is increasing, Figure II.10: Young women are turning to nonagricultural especially among younger women activities (Percentage of women working) (Percentage of women with main occupation outside agriculture) 100 100 90.4 91.6 84.8 80.3 80 80 73.4 60.8 60 60 40 40 24.3 20 20 17.3 11.0 12.3 12.9 8.3 0 0 15-24 24-34 35-49 15-24 24-34 35-49 2005 2010 2005 2010 Source: DHS 2005, 2010 Source: DHS 2005, 2010 38 Rwanda Economic Update | Edition No. 5 II. Special Focus: Harnessing the Demographic Dividend the proportion of women running a nonfarm in 2005 and above average (positive) in 2010. business, a relatively lucrative activity, increased These results are in line with the high economic from 8 percent in 2006 to 13 percent in 2011 growth rates during the period (8.2 percent a year and women’s real wages in the nonfarm sector between 2005 and 2010) and the sharp reduction increased by 9 percent. Young women have been in poverty (from 57 percent in 2005/06 to 45 at the forefront of the rise in education and labor percent in 2010/11). force participation, two trends that are reflected in the larger fertility drop for women 15–24. Figure II.11: Household wealth increased between 2005 and 2010 (Score on standardized wealth index) Household wealth is increasing W 1 omen’s economic conditions are expected to affect their demand for children, though the direction of the effect Score on Wealth Index .5 is theoretically ambiguous. On the one hand, an increase in income or wealth could raise the 0 demand for children, as the household has more resources to take on additional children. On the -.5 other hand, children are usually more expensive for better-off households and women because of the higher opportunity value of their time. -1 Empirically, the relationship between income or 2005 2010 wealth and fertility is negative for most countries Source: DHS 2005, 2010 Note: The horizontal line within each box represents the median wealth at most times (Jones, Schoonbroodt, and Tertilt score for the survey round. The lower line of the box is the first quartile; the upper line is the third quartile 2008; Weerasinghe and Parr 2002). Although the DHS does not collect information Proximate determinants of fertility are on income, it is possible to construct a composite changing quickly household wealth index based on households’ ownership of a range of assets (Filmer and Pritchett 2001). We construct a household-level S tudies of fertility often make distinguish between the underlying socioeconomic determinants of fertility and so-called wealth index based on ownership of 22 assets.11 proximate determinants. Proximate The resulting wealth index is a standardized determinants clarify the mechanisms through variable (has mean of zero), with higher values which the underlying determinants of fertility representing higher levels of wealth. exert their effect. For instance, female education is an underlying socioeconomic determinant of Household wealth increased between 2005 and fertility. It influences fertility through its impact 2010. In Figure II.11, the box for 2010 is higher on age at marriage (educated women tend to get than the box for 2005, which indicates higher married later) and use of contraception (educated wealth levels in 2010 (the horizontal line within women are more likely to use contraceptives). the box represents the median wealth score for Age at marriage and use of contraception are the respective survey round, which is higher in proximate determinants of fertility, the channels 2010 than in 2005). The average score on the through which the “real” determinant (female wealth index was –0.14 in 2005 and 0.22 in 2010. education) exercises its effect. The wealth index was below average (negative) 11 The index includes information on households’ quality of housing, sources of water and energy, asset holdings, and other measures. See Annex IIA for the construction of the wealth index. Rwanda Economic Update | Edition No. 5 39 II. Special Focus: Harnessing the Demographic Dividend Between 2005 and 2010, the proximate II.2.2 What drove the decline in fertility? T determinants of fertility moved in a way he drop in fertility between 2005 and that reduced fertility. The average age at first 2010 could reflect a wide array of factors. union (first cohabitation with a partner) was Women’s education and participation in the labor pushed back by half a year between 2005 and market increased, and household wealth levels 2010, and first birth was delayed by about four are on the rise. Child mortality decreased, family months. These changes are important, as women planning campaigns attempted to sensitize the who delay their first birth typically have fewer population about the need to limit family size, children over the course of the reproductive and use of contraception increased sharply. In period. The percentage of women without any exposure to mass media decreased from 44 addition, the period between 2005 and 2010 percent to 31 percent, and the percentage of witnessed rapid economic growth and strong women without any exposure to family planning poverty reduction. messages fell from almost 60 percent in 2005 to 33 percent in 2010. To examine the extent to which the factors mentioned above were associated with the drop The use of modern contraception recorded in fertility, we use a statistical decomposition an exponential increase between 2005 and method (the method is detailed in Annex IIC). 2010. The proportion of women using modern In simple terms, the method examines how much contraception more than quadrupled, rising of the decline in fertility can be explained by the from less than 6 percent in 2005 to more than observed changes in underlying determinants 25 percent in 2010. More than half of married (such as improved female education and women were using modern contraceptives in increased labor force participation). In this way, 2010, up from 10 percent in 2005. Rural women the contribution of each factor to the drop in are not lagging behind: more than 25 percent of fertility can be quantified. The decomposition rural women, married or not, were using modern method uncovers the main correlates of the contraception in 2010, compared with 24 percent decline in fertility but does not necessarily of urban women. The overwhelming majority of imply causation. As the drop in fertility and the women (91 percent) obtained their contraceptives evolution in underlying determinants have been from government facilities, suggesting that most salient among young women, we perform government efforts to tackle population growth the decomposition by age group. are bearing fruit (Box II.3). Table II.3: Proximate determinants of fertility are moving in the right direction 2005 2010 Mean difference Cumulative fertility (number of children) 2.68 2.42 –0.26*** Age (years) 28.34 28.41 0.07 Percent ever in union 0.623 0.613 –0.01 Age at first union 19.99 20.50 0.51*** Age at first birth 21.13 21.45 0.32*** Percent with no exposure to mass media 0.436 0.305 –0.131*** Percent with no exposure to family planning messages 0.589 0.334 –0.255*** Percent currently using modern contraception 0.056 0.252 0.196*** Number of observations 11,321 13,671 Source: DHS 2005, 2010 Note: *** Statistically significant at the 1 percent level 40 Rwanda Economic Update | Edition No. 5 II. Special Focus: Harnessing the Demographic Dividend Box II.3 Under pressure Efforts to curb population growth seem to be paying off Rapid population growth and increasing land pressure were already recognized as important challenges during the colonial period. It was not until 1981 that Rwanda introduced family planning activities, however, with the establishment of the National Population Office (ONAPO). In 1990, the first National Population Policy was formulated, aiming to reduce total fertility rates from 8.3 in the 1983 National Fertility Survey to 4.0 by 2000. After the 1994 genocide, family planning activities were suspended, because of the unpopularity of birth control in light of the massive loss of life. Concerns about rapid population expansion resurfaced when the 2000 DHS and 2002 population census highlighted the persistence of high fertility and rapid population growth. In response to the increase in fertility between 2000 and 2005, the government embarked on a massive family planning campaign. The campaign aimed to raise awareness of the necessity of reducing population growth. It trained health service providers, local leaders, secondary school teachers, and journalists; conducted mass media campaigns; and made contraceptives widely available and affordable throughout the country. The campaign also aimed at changing Rwandans’ perceptions of the ideal family size. The 2010 DHS data suggest that the efforts have not missed their intended targets. Total fertility rates dropped by 25 percent between 2005 and 2010, contraception use increased by factor of more than four, and the ideal number of children decreased from 4.5 to 3.4. Source: Muhoza, Rutayisire, and Umubyeyi 2013; Westoff 2012 In explaining the decline in fertility, we participation, and household wealth explain focus only on the underlying socioeconomic almost half of the observed drop in fertility determinants. Proximate determinants are not between 2005 and 2010. The other half of the included in the analysis, as they are themselves the decline remains unexplained by the logic of result of changes in the underlying determinants. the model, because although household—and The decomposition focuses on the contribution individual-level characteristics affect fertility, a of the socioeconomic evolutions to the drop substantial part of the fertility change observed in fertility and not on the role of behavioral during fertility transitions is a result of changing changes (delayed childbirth, delayed marriage, social norms and expectations, which are not use of contraception) that are induced by the included in the model (Canning 2011). socioeconomic evolutions. Figure II.12: Rising education and wealth are reducing fertility (Percentage contribution to drop in cumulative fertility The main results of the statistical between 2005 and 2010) decompositions are summarized in Figure 50 II.12. The improvement in female education emerges as the main driver of the drop in fertility, Higher female labor 40 5.4 market participation explaining about one-fifth of the decline. The increasing levels of household wealth are an 30 15.4 important secondary driver, accounting for Increasing household wealth an additional 15 percent of the fertility drop. Women’s increased participation in the labor 20 market has been associated with the decline in 22.7 fertility, although the association is weak. As 10 Improving female education argued later, the increased engagement in non- agricultural occupations has had a significantly 0 larger effect on fertility. Together, the Source: DHS 2005, 2010 improvements in female education, labor force Rwanda Economic Update | Edition No. 5 41 II. Special Focus: Harnessing the Demographic Dividend Female education is the main driver behind on the wealth index) explains about 20 percent of the decline in fertility the decline in fertility of women 15–24, 11 percent I mprovements in female education between for women 25–34, and 4 percent for women 35– 2005 and 2010 emerge as the main driver 49. The association between wealth and fertility of the decline in fertility. In the full sample is not statistically significant for women 35–49. (all women 15–49), the improvement in female education explains almost 23 percent of the Women’s work, especially outside the farm, drop in cumulative fertility. Education was limited fertility especially important in bringing down fertility levels of young women: 65 percent of the drop in fertility among women 15–24 is associated T he increase in the proportion of women who work is associated with the decline in fertility. The magnitude of the association with improvements in educational attainment, is weak, however: women’s participation in the compared with about 23 percent in the oldest labor market increased by 15 percent between age group (Figure II.13). For women 25–34, 2005 and 2010, explaining 5 percent of the the worsening of educational outcomes between drop in fertility. The type of work appears more 2005 and 2010 (because of the genocide) held important for fertility than whether or not a back the fertility decline, adding an estimated 4 woman is working. The increased uptake of percent to their fertility. nonfarm activities explains 8 percent of the overall fertility drop.12 In contrast, traditional Figure II.13: Improvements in education have reduced fertility, especially among women 15–24 farm activity is not associated with fertility. (Percentage contribution of improvements in educational attainment to drop in fertility between 2005 and 2010) The move to non-agricultural employment 70 65.1 predominantly affected the fertility of young 60 women. The proportion of women 15–24 with 50 a main occupation outside agriculture more than 40 doubled between 2005 and 2010 (Figure II.10). The change contributed almost 13 percent to their 30 22.8 drop in fertility (Figure II.15). Women in the 20 other age groups also increasingly turned to non- 10 agricultural occupations, though it had had only 0 marginal effects on their observed fertility drops. -3.9 -10 15-24 24-34 Age Group 35-49 To summarize, changes in the underlying Source: DHS 2005, 2010 socioeconomic determinants explain almost half of the drop in fertility between 2005 and 2010. For women 15–24, the fertility drop Higher levels of household wealth added to the is fully explained by the large changes in their decline in fertility socioeconomic fortunes. Together, improved T he increasing wealth levels of households between 2005 and 2010 explain 15 percent of the drop in fertility. Though wealth levels education, higher household wealth, and higher participation in the labor market explain more than 98 percent of the decline in fertility increased more for households of women 35–49 among this cohort of women. For women in and 25–34, the increase in household wealth older age groups, the part of the fertility drop mattered more for the fertility of women 15–24. explained by the socioeconomic determinants The observed increase in household wealth (score is substantially lower. 12 There is a likely two-way causation here, with women who have fewer children being more likely to have a nonfarm activity. 42 Rwanda Economic Update | Edition No. 5 II. Special Focus: Harnessing the Demographic Dividend Figure II.14: Increasing wealth contributed to the fertility Figure II.15: The move to nonfarm employment reduced decline in all age groups, especially the young fertility, especially among women 15–24 (Percentage contribution of increasing household wealth to drop in (Percentage contribution of increased nonfarm employment fertility between 2005 and 2010) to drop in fertility between 2005 and 2010) 30 30 20.6 20 20 12.7 10.8 10 10 3.8 4.4 1.5 0 0 15-24 24-34 35-49 15-24 24-34 35-49 Age group Age Group Source: DHS 2005, 2010 Source: DHS 2005, 2010 II.3 Rwanda’s Potential Demographic Dividend T he size of the labor force will increase substantially over the coming decades, in both absolute and relative terms, potentially boosting income levels. According to a demographically based forecasting model, real GDP per capita could triple between 2010 and 2050 in the medium-fertility scenario and more than triple if a more rapid reduction in fertility is achieved. Estimation of the same model before the sharp drop in fertility highlights the economic importance of fertility: Before the sharp fertility drop, the model estimated real GDP per capita (PPP) to almost double between 2010 and 2050, while taking account of the drop real GDP per capita is projected to almost triple. Reaping economic rewards from the demographic changes is not automatic, however. To make the most of the demographic opportunity, countries need to invest in health and education to foster a productive future workforce, create flexible labor markets capable of accommodating a bourgeoning labor force, and encourage price stability and competition in financial institutions to increase the domestic savings rate. T he sharp drop in fertility accelerated Rwanda’s demographic transition and will have dramatic consequences for the age a youth bulge. Working-age adults will account for more than 64 percent of the population, and the share of youth will drop to 29 percent in the structure of the population. Rwanda is currently medium-fertility scenario. These age structure characterized by a youth bulge (43 percent of the changes hold significant economic promise. population is below 15) and a relatively small labor force (53 percent of adults are of working Over the past 60 years, the share of working- age). The 2010 population pyramid reflects the age adults in the total population hardly typical pattern of a developing country—wide changed. In 1950, working-age adults accounted at the bottom and narrowing fast as one moves for 53 percent of Rwanda’s population, exactly up the age distribution (Figure II.16). By 2050, the same as the share in 2010. In between, the the population pyramid will look completely share of the labor force fluctuated between a different (Figure II.17). It will be bell shaped and low of 46 percent in the late 1980s and a high characterized by a bulging labor force rather than of 53 percent in the late 2000s. Fluctuations in Rwanda Economic Update | Edition No. 5 43 II. Special Focus: Harnessing the Demographic Dividend Figure II.16: Bursting at the bottom in 2010… Figure II.17: … means thicker in the middle by 2050 (Rwanda’s population by age group and gender) (Rwanda’s projected population by age group and gender) 80+ 80+ 70-74 70-74 60-64 60-64 50-54 50-54 Age group Age group 40-44 40-44 30-34 30-34 20-24 20-24 10-14 10-14 0-4 0-4 1,500 1,000 500 0 500 1,000 1,500 1,500 1,000 500 0 500 1,000 1,500 Population (thousands) Population (thousands) Male2010 Female2010 Male 2050 Female 2050 Source: UN 2013 Source: UN 2013 Note: Figures are based on the medium-fertility scenario the share of the labor force have been associated and the quality of the institutional and policy with changes in income levels: between 1960 framework. The estimated income levels should and 2010, a 1 percent increase in the share of the therefore be interpreted as the income levels that labor force was on average associated with a 1.1 are attainable in light of a growing labor force percent increase in real GDP per capita (Figure and decreasing dependency rates. II.18).13 This relationship suggests that the large increase in the share of the labor force projected The 2011 World Bank policy note “Demography for the coming decades could have significant and Economic Growth in Rwanda” already economic effects. applied this demographic model to Rwanda. It was based on the population projections in the II.3.1 Forecasting income levels for Rwanda 2010 revision of World Population Prospects. The based on demographic projections 2010 revision did not take the recent sharp drop in fertility into account and hence underestimated G iven the association between age structure and income levels, what can the demographic projections say about future the magnitude of the forthcoming changes in the age structure of the Rwandan population, as illustrated by Figure II.19. It projected the share income levels in Rwanda? To answer this of working-age adults to increase to almost 62 question, we use a demographic forecasting percent by 2050 (the dashed blue line) and the model that estimates real GDP per capita in share of people under 15 to drop to 33 percent purchasing power parity (PPP) based on a (the dashed red line). The 2012 revision, which range of demographic variables, including the takes the sharp fertility drop into account, share of the working-age population, the share projects a more rapid transition: the share of of dependent youth, and life expectancy.14 The working-age adults is projected to rise to more model is detailed in Annex IID. The model than 64 percent and the share of people under focuses exclusively on demographic variables to 15 to fall to less than 29 percent by 2050. As a estimate income levels; it does not take account more rapid transition has implications for income of other important determinants of growth, such growth (at least within the logic of the model), as the quality of human resources (as opposed this section re-estimates the model based on the to the quantity, which is included in the model) updated population projections.15 13 The association is stronger when controlling for the effects of the genocide. In this case, a 1 percent increase in the share of working-age adults is associated with a 1.5 percent increase in GDP per capita. 14 Because the model is estimated based on data from 108 countries (see annex II.C), GDP levels need to be adjusted for differences in prices across countries to be comparable. For this reason, GDP per capita is expressed in PPP. For low-income countries, PPP GDP levels will be higher than nominal GDP levels, though growth in GDP expressed in PPP is typically lower than growth in nominal GDP. 15 Ideally, we would have used the updated projections from the 2012 Population and Housing Census. As the updated projections were not yet available at the time of writing, we used projections from the 2012 revision of the World Population Prospects. The national projections would most likely result in even sharper changes in the age structure. 44 Rwanda Economic Update | Edition No. 5 II. Special Focus: Harnessing the Demographic Dividend Figure II.18: A larger share of working-age adults has been associated with higher income levels Figure II.19: The 2012 revision projects larger changes (Relationship between share of working-age adults and in the age structure of the Rwandan population real GDP per capita in Rwanda, 1960–2010) (Percentage of youth and working-age adults in total population) 350 70 64.4 GDP per capita (constant 2000 US$) 300 60 61.8 53.0 50 250 44.7 40 200 32.8 30 150 28.8 20 100 2010 2015 2020 2025 2030 2035 2040 2045 2050 45 47 49 51 53 55 Under 15 (2010) Under 15 (2012) Share of working age adults (percent) Working age (2010) Working age (2012) Sources: World Bank staff calculations based on data from Source: UN 2013 World Development Indicators and UN 2013 Note: Figures are based on the medium-fertility scenario The drop in fertility has potentially significant to US$2,817. The real growth acceleration is economic importance. The 2011 policy note expected to kick in after 2025, which coincides with estimated that in the medium-fertility scenario, the fastest projected increase in the working-age real GDP per capita in PPP would increase by 90 population (see Figure II.2) and the fastest decline percent between 2010 and 2050, reaching almost in the youth dependency ratio (see Figure II.3). US$2,000 per capita by 2050 (the blue line in Figure II.20).16 Using the new projections, and The 2011 World Bank policy note “Demography still assuming a medium-fertility scenario, real and Economic Growth in Rwanda” already GDP per capita (PPP) is expected to almost triple applied this demographic model to Rwanda. between 2010 and 2050 (the red line in Figure It was based on the population projections in the II.20). The recent drop in fertility adds almost 2010 revision of World Population Prospects. The US$900 (in PPP) to the long-run estimate of 2010 revision did not take the recent sharp drop GDP per capita, increasing it from US$1,939 in fertility into account and hence underestimated Figure II.20: The drop in fertility has potentially significant implications for economic growth Figure II.21: Further rapid reduction in fertility could pay off (Estimated real GDP per capita in US$, medium-fertility scenario) (Estimated real GDP per capita in US$, 2012 revision) 3,000 2,817 4,000 3,485 Real GDP per capita (estimated) Real GDP per capita (estimated) 3,000 2,817 1,939 2,000 2,000 1,025 1,000 1,025 1,000 0 0 2010 2015 2020 2025 2030 2035 2040 2045 2050 2010 2015 2020 2025 2030 2035 2040 2045 2050 2012 revision 2010 revision Low fertility Medium fertility Source: UN 2013 Source: UN 2013 16 Real GDP per capita is expressed in 2005 prices and adjusted for differences in purchasing power. Rwanda Economic Update | Edition No. 5 45 II. Special Focus: Harnessing the Demographic Dividend the magnitude of the forthcoming changes in The next section elaborates on the policies that the age structure of the Rwandan population, as are believed to be required for a country to make illustrated by Figure II.19. It projected the share the most out of the demographic transition. of working-age adults to increase to almost 62 percent by 2050 (the dashed blue line) and the II.3.2 Getting policies right to seize the moment A share of people under 15 to drop to 33 percent lthough it presents a golden opportunity (the dashed red line). The 2012 revision, which for sustained growth, the demographic takes the sharp fertility drop into account, dividend is neither automatic nor guaranteed. projects a more rapid transition: the share of Seizing the moment may entail substantial working-age adults is projected to rise to more economic gains. Failure to cash in on the than 64 percent and the share of people under demographic shifts may create large-scale 15 to fall to less than 29 percent by 2050. As a underemployment of human resources and mean more rapid transition has implications for income permanently forgoing the opportunities afforded growth (at least within the logic of the model), by the demographic window. A number of policies this section re-estimates the model based on the and institutions have been shown to be conducive updated population projections.17 to realizing the full potential of the demographic dividend. Although these policies are not a recipe These estimates suggest that population for success, they are a required component of any changes hold significant economic potential policy and institutional framework seeking to for Rwanda. Given that purchasing power exploit the demographic shifts. parity–adjusted GDP growth rates (as used in the model) are usually significantly lower than The enabling policy environment for the growth rates that are not adjusted for purchasing demographic dividend is depicted in Figure power (the metric used to classify countries as II.22. The demographic transition is triggered low, middle, or high income), we would expect by sharp drops in fertility, leading to notable even bigger impacts on growth. For instance, changes in the population age structure. For growth in GDP per capita adjusted for purchasing these changes to yield a dividend, investments in power averaged 4 percent a year between 2006 human capital are needed to improve the health and 2011 and 5 percent a year between 2001 and and education of today’s children, in order to 2011. Growth in Gross National Income (GNI) create a skilled and productive labor force several per capita (Atlas method) was much higher: 13 percent a year between 2006 and 2011 and 10 Figure II.22: A variety of policies is needed if Rwanda is to reap the demographic dividend percent a year between 2001 and 2011. These differences suggest that high growth rates over the coming decades are a distinct possibility. EDUCATION Figure II.20 and Figure II.21 shed light on how the Rwandan economy could expand in light of the steadily increasing labor force and GOVERNANCE DEMOGRAPHIC DIVIDEND decreasing dependency ratio. By themselves, POPULATION STRUCTURE HEALTH however, demographic shifts will not lead to higher economic growth and income levels. To take advantage of a much larger labor force, ECONOMICS investments are required to increase the quality and productivity of the labor force and to sustain or even accelerate the pace of the fertility transition. Source: Population Reference Bureau (www.prb.org) 17 Ideally, we would have used the updated projections from the 2012 Population and Housing Census. As the updated projections were not yet available at the time of writing, we used projections from the 2012 revision of the World Population Prospects. The national projections would most likely result in even sharper changes in the age structure. 46 Rwanda Economic Update | Edition No. 5 II. Special Focus: Harnessing the Demographic Dividend decades down the road. A bigger, healthier, and creation, and good governance is needed to better-educated labor force will, however, not establish legal systems and rules of law to boost in itself lead to accelerated per capita growth. domestic savings and attract foreign investments. Economic interventions to create an investment- Combined, these policies greatly increase the friendly climate and a flexible labor market are probability of turning a demographic opportunity required to lay the foundations for rapid job into a vehicle for growth (Box II.4). Box II.4 Boom or bust Lessons from the East on the importance of the right policy environment Between 1965 and 1990, East Asian economies grew at an average rate of more than 5 percent a year, resulting in what has become known as the East Asian Miracle. Real GDP almost quadrupled in only 25 years, and poverty levels plummeted (the poverty headcount dropped from 58 percent to 17 percent in Indonesia and from 37 percent to less than 5 percent in Malaysia [World Bank 1993]). The East Asian Miracle unfolded during an exceptionally rapid demographic transition. Fertility rates fell from 5.5 in the early 1960s to 2.0 in the early 1990s, with the share of working-age adults increasing from 56 percent in 1965 to 66 percent in 1990 (UN 2013). Research suggests that up to a third of the East Asian Miracle can be explained by rapid population shifts (Bloom and Williamson 1998). A conducive policy framework helped East Asian countries cash the demographic dividend. Access to and the quality of education, already high relative to other developing countries at the start of the 1960s, expanded rapidly during the demographic transition, leading to a skilled and productive labor force. Strong macroeconomic management aimed at maintaining price stability and stable real interest rates boosted domestic financial savings, which in turn enabled high levels of private domestic investment, a main driver of growth. In addition, efforts to capture savings from small and rural savers proved successful, adding to the accumulation of domestic capital. Strong technocratic governments created business-friendly environments by implementing legal and regulatory reforms and avoiding high tariffs on capital goods. In contrast to many other developing economies, East Asian governments refrained from passing minimum wage legislation. As a result, wages have been downwardly flexible in response to changes in labor demand. Until recently, India was widely believed to be the 21st century Asian miracle. Fertility rates dropped rapidly during the 1980s and 1990s, giving rise to a rapidly growing labor force and declining dependency rates. Currently, the Indian labor force is growing by about 12 million people a year. India will soon have a fifth of the world’s total working-age population and will overtake China as the largest labor force by about 2020. The IMF projects that these population shifts could add up to 2 percentage points to per capita growth in GDP over the next two decades (Aiyar and Mody 2011). There is growing concern that India is at risk of squandering this huge opportunity, however. Despite producing some of the world’s most brilliant minds, India provides substandard basic education. Many people leave school ill-prepared for even basic jobs. More than 90 percent of girls complete primary school, but only 56 percent are enrolled in secondary school. Indian women have much lower literacy rates than men, limiting their productivity in the labor force (Gribble and Bremner 2012). The health situation is worrisome, too. The 2013 Global Hunger Index ranks India 63rd among 120 countries, calling the country’s hunger levels “alarming.” Apart from India, only three countries outside Africa (Haiti, Timor-Leste, and the Republic of Yemen) have similar hunger levels. More than 40 percent of India’s children are malnourished, jeopardizing their physical and intellectual development. High inflation in recent years has reduced the once impressive savings rate, as households seek refuge in gold and real estate. The labor market is hardly flexible: factories cannot fire without the state’s permission (although businesses can get around this rule by hiring labor from third parties). Rwanda’s policy framework bears more resemblance to that of the East Asian tigers than to India’s. Just like the East Asian economies during the 1970s and 1980s, Rwanda stresses prudent macroeconomic management and price stability and aims to create a business-friendly environment though regulatory reforms. Education is being dramatically expanded, resulting in higher education levels among youth. Efforts in reproductive health have reduced child mortality and fertility. Health in early childhood remains worrisome, however: more than 40 percent of Rwandan children are chronically malnourished. Despite relatively low inflation, the domestic savings rate in Rwanda is very low (0.4 percent of GDP in 2010); dramatic boosting is needed to enable more private domestic investment. A key difference between Rwanda and the East Asian countries at the outset of their demographic transition is that East Asian countries could build on the supply of cheap and skilled labor whereas Rwanda’s labor is relatively expensive and low skilled. Rapidly building transferrable skills is critical if Rwanda is to make the most out of its demographic transition. Rwanda Economic Update | Edition No. 5 47 II. Special Focus: Harnessing the Demographic Dividend Strengthening human capital for a productive average educational attainment is expected to and flexible workforce increase substantially over the coming decade. H uman capital is critical to realizing a demographic dividend. Investments in health and education foster the development of Expanding educational opportunities needs to be matched by improvements in health, in particular for infants and children. Ill health a skilled and healthy labor force equipped to and malnutrition in early childhood severely take full advantage of the beneficial population impede physical and intellectual development shifts. Expanding access to education, especially and are causally linked to worse socioeconomic at the secondary and higher levels, is key to outcomes later in life, ranging from lower providing young people with the analytical educational attainment in adolescence to lower and entrepreneurial skills they need to take on productivity and wages as an adult. In order for higher-value jobs in the formal sector or access today’s children to realize their full potential as loans and financial services that allow them working adults, targeted interventions need to be to grow their formal or informal businesses. designed to drastically reduce the incidence of Promoting secondary and higher education chronic malnutrition from the current level of 44 for girls is particularly important, because it percent. delays marriage and first pregnancy, leading to lower lifetime fertility and a sharper more rapid A healthier and better educated workforce will demographic transition. In addition, secondary only increase growth if productive employment school–educated women are more likely to work opportunities are available. Although there is outside the home, adding to the skilled and formal no magic bullet for job creation, flexible labor workforce. markets are believed to be vital to accommodate a burgeoning labor force. Flexibility means that Rwanda’s labor force is largely unskilled. Close businesses are able to rapidly expand and contract to two-thirds of the working-age population did their operations in light of changing environments not complete primary education and less than 5 and increase or decrease salaries in response percent completed secondary education.18 Good to market conditions. To create sufficient jobs, progress was however made over the past decade. priority should be given to labor-intensive sectors Primary school completion rates increased from such as manufacturing, retail trade, hospitality, 23 percent in 2000 to 70 percent in 2010, and and agriculture, which contribute substantially to net enrollment is almost universal (99 percent). both employment and GDP growth. Despite notable achievements, access to post- primary education remains low. Gross enrollment An open economy and a business-friendly in secondary school tripled between 2000 (11 environment are important for accommodating percent) and 2010 (36 percent), but remains below a bulging labor force, as the experiences of the Sub-Saharan average of 40 percent (World East Asia and Latin America in the second Development Indicators 2013). Post-primary half of the 20th century reveal. Both regions education will require substantial boosting to experienced rapid demographic transitions, but equip the future workforce with the transferrable Latin America largely failed to reap the rewards: skills it will need to move into different types per capita growth between 1975 and 1995 of jobs as the labor supply expands and the averaged 6.8 percent a year in East Asia and 0.7 economy diversifies. Given the government’s percent for Latin America. Research shows that commitment to educational expansion through half of the growth gap between the two regions the 9-year basic education (compulsory) and can be explained by the closed economies and 12-year basic education (entitlement) program, business-unfriendly policies of Latin American 18 Based on EICV3. 48 Rwanda Economic Update | Edition No. 5 II. Special Focus: Harnessing the Demographic Dividend countries during this period, which led to sluggish fertility since 2005. The speed of the transition and job creation (Bloom and others 1999). Rwanda’s the magnitude of the demographic dividend will business-friendly climate, as highlighted by the depend on further fertility declines. Although the 2014 Doing Business report, is an important asset youth dependency ratio will steadily fall over the for Rwanda on the road to earning a demographic coming decades, and the proportion of working- dividend. age adults steadily rise, individual cohorts will still continue to be bigger than the previous ones A competitive financial sector to attract savings up until 2050. In Figure II.17, this is reflected by T the lower bars being consistently larger than the he demographic transition can lead not next-to-lower bars (medium fertility scenario). only to a bulge in labor but also to a bulge in capital. With fewer children to support, Faster declines in fertility will accelerate the working-age adults can save more of their demographic transition and bring forward income, resulting in increasing domestic savings the demographic dividend. Figure II.23 shows rates. This mechanical effect is compounded by Rwanda’s projected age distribution in 2050 under the life-cycle effect of savings, whereby people the low-fertility scenario. In this case, successive save least when they are young and most in late child cohorts will become smaller and smaller and middle age. As the demographic transition raises the size of the labor force will grow more rapidly the proportion of older adults, savings rates are in relative terms, bringing forward the potential further boosted. dividend. This scenario implies driving down the total fertility rate from 4.6 in 2010 to 3.7 by 2020 The increase in savings can provide the capital and 2.9 by 2030. Replacement fertility levels of needed to expand economic enterprise during 2.1 would then be achieved by 2050 (UN 2013). the labor force bulge. The demographic transition Achieving this scenario will require continued will lead to higher savings rates if households efforts to further decrease child mortality, and businesses have a safe and relatively especially among poorer households, and boost profitable means of saving. To promote savings, the use of modern contraception, which despite the government must provide price stability the recent increase is still low at 25 percent of and encourage competition and efficiency in women 15–49. the financial institutions to make savings more attractive. Small and rural savers should be Figure II.23: In the low-fertility scenario, successive cohorts will be smaller by 2050 encouraged to participate in the formal financial (Projected population by age group and gender) system by designing savings products that cater to their needs. Higher savings rates have the 80+ potential to substantially boost private domestic 70-74 investment, a key factor of growth during East 60-64 Asia’s demographic transition (Box II.4). 50-54 Age group 40-44 Reproductive health and family planning for 30-34 smaller families 20-24 R apid drops in fertility kick-start the demographic transition by lowering the proportion of dependent children and youths 10-14 0-4 -1,500 -1,000 -500 0 500 1,000 1,500 Population (thousands) in the population. Rwanda’s demographic Male 2050 Female 2050 transition has been ignited by the sharp drop in Source: UN 2013 Rwanda Economic Update | Edition No. 5 49 II. Special Focus: Harnessing the Demographic Dividend The data suggest that there is ample scope for percent of households (5.4) than for women in the further rapid declines in fertility in Rwanda. richest households (3.4). Expanding reproductive In 2005, the gap between the total fertility rate health services to draw in the poorest households (the number of children an average woman has would most likely reduce child mortality among at the end of the reproductive period) and the the poor, resulting in lower fertility rates and a total wanted fertility rate (the number of children sharper more rapid demographic transition. an average woman would have at the end of the reproductive period if all unwanted births were Urbanization can be a catalyst for the avoided) was 1.5—that is, the average woman demographic dividend W in Rwanda in 2005 could expect to have 1.5 ell-managed urbanization could be children more than she actually wanted. The total an important catalyst for creating fertility rate dropped between 2005 and 2010, but the conditions for reaping the demographic so did the total wanted fertility rate, leaving the dividend. Urbanization has many advantages, gap between the two at 1.5 in 2010.19 Addressing several of which are conducive for realizing the this gap by further expanding the reach of family dividend. For similar levels of education and planning campaigns and access to contraception wealth, urban women tend to have fewer children has the potential to significantly reduce fertility. than rural women, which means urbanization Nineteen percent of married or cohabitating would speed up the fertility drop and lead to women expressed an unmet need for family a sharper demographic transition. A sharper planning in 2010, highlighting the need to expand transition would lead to larger income gains family planning education campaigns.20 (Figure II.21). Urban children are also likely to achieve higher levels of schooling, equipping Improvements in reproductive health are them with the skills they need to participate important to achieve further reductions in the formal workforce. Urbanization is also in infant and child mortality. Rwanda has associated with higher economic growth rates performed well in recent years (see Figure II.15), and the creation of nonfarm jobs, which Rwanda but child mortality rates are still 60 percent will sorely need in light of increasing pressure on higher for the poorest 20 percent of households arable land. In addition, urbanization to secondary (119 deaths per 1,000 live births) than for the towns, an element the EDPRS 2, is linked to richest 20 percent (75). As a result, fertility rates accelerated poverty reduction (Christiaensen, are much higher for women in the poorest 20 De Weerdt, and Todo 2012). 19 The total wanted fertility rate differs from the desired number of children pictured in Figure II.4. The total wanted fertility rate is calculated by applying the current age-specific wanted fertility rates to women in different stages of the reproductive period, while the desired number of children is a simple average regardless of age. 20 Women with un-met need are women who are fecund and sexually active but are not using any method of contraception and report not wanting more children or wanting to delay the birth of their next child. 50 Rwanda Economic Update | Edition No. 5 II. Special Focus: Harnessing the Demographic Dividend II.4 Beyond the Long Term: Preparing for an Aging Population R wanda’s demographic window of opportunity is expected to remain open until 2070. After that, dependency rates will start increasing again, as the bulge of working-age adults retires and old- age dependency rates shoot up. This change will be accompanied by a sharp increase in the fiscal burden related to old-age pensions. During the demographic transition, countries therefore need to put in place policies to care for a rising number of elderly people, without which many will face destitution in their final years. T he population shifts set in motion by the sharp drop in fertility have created the opportunity for accelerated growth over the Figure II.24: Rwanda’s window of opportunity will remain open until 2070 (Overall dependency ratio) decades to come. This demographic window of 1.1 opportunity will remain open until the bulge in working-age adults retires, at which time old-age 0.9 dependency ratios will rise steeply. The end of the demographic dividend will be associated with a sharp increase in the fiscal burden, as a larger old- 0.7 age pension burden will have to be shouldered by a smaller labor force. 0.5 The beneficial age structure shifts are not 0.3 expected to end any time soon. The overall 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 dependency ratio—defined as the sum of Source: UN 2013 (World Population Prospects 2012 revision) dependent young (under 15) and elderly (65 Note: Figures are based on the medium-fertility scenario and older) people divided by the working-age of-opportunity phase of the transition, countries population—is projected to decline from 0.9 to need to put in place policies to provide for the 0.5 in 2070, after which the elderly population rising number of old people, without which many will start growing more rapidly than the labor of them will face destitution in their final years. force (Figure II.24). Taking into account the pace of the population shifts (Figure II.2 and Figure Many developed countries now face rapidly II.3) and the estimates from the demographic aging populations (and in some countries, such forecasting model (Figure II.20 and Figure II.21), as Germany, Italy, and Japan, even shrinking Rwanda’s potential for a demographic dividend is populations). From a demographic point of view likely to be at its highest in the 35 years between at least, the coming five decades will present the 2025 and 2060. developed world with significant challenges and high fiscal burdens while offering the developing Eventually, the demographic transition is world, and Rwanda in particular, an appealing characterized by a decline in the ratio of opportunity. workers to dependents. During the window- Rwanda Economic Update | Edition No. 5 51 Annexes Annex Table IIA.1: Construction of the wealth index Factor Standard Poorest 40 Middle 40 Richest 20 Asset indicators included in Wealth Index Mean loading deviation percent percent percent Has piped water in dwelling or yard 0.149 0.036 0.186 0.000 0.000 0.193 Uses water from public tap 0.025 0.254 0.435 0.111 0.319 0.411 Uses water from spring –0.066 0.485 0.500 0.474 0.429 0.243 Uses surface water –0.017 0.080 0.271 0.272 0.101 0.033 Has a flush toilet 0.105 0.012 0.109 0.000 0.000 0.065 Has a ventilated improved pit –0.064 0.311 0.463 0.643 0.088 0.111 Has a pit latrine 0.042 0.641 0.480 0.287 0.896 0.817 Has no latrine –0.020 0.031 0.173 0.065 0.011 0.004 Lives in a house with concrete floor 0.165 0.151 0.358 0.000 0.000 0.806 Lives in a house with an earth or mud floor –0.164 0.841 0.365 1.000 0.994 0.167 Uses charcoal for cooking –0.016 0.467 0.490 0.724 0.245 0.423 Uses coal or lignite for cooking 0.082 0.029 0.169 0.000 0.009 0.137 Uses wood for cooking –0.016 0.422 0.494 0.213 0.641 0.371 Uses crop residue for cooking –0.015 0.067 0.249 0.055 0.097 0.024 Has electricity 0.168 0.075 0.263 0.000 0.004 0.391 Owns a radio 0.076 0.550 0.497 0.213 0.741 0.837 Owns a television 0.159 0.039 0.194 0.000 0.000 0.210 Owns a refrigerator 0.125 0.012 0.110 0.000 0.000 0.065 Owns a bicycle 0.024 0.133 0.340 0.041 0.157 0.273 Owns a motorcycle 0.047 0.008 0.090 0.000 0.000 0.044 Owns a car 0.105 0.008 0.088 0.000 0.000 0.042 Owns a mobile phone 0.121 0.242 0.428 0.000 0.283 0.660 Asset index 0.000 1.000 –0.58 –0.22 1.420 Sources: DHS 2005, 2010 and World Bank staff calculations. 52 Rwanda Economic Update | Edition No. 5 Annex IIB: Evolution of Fertility Determinants by Age Group Annex Table IIB.1: Determinants of fertility by age group, 2005 and 2010 Women 15–24 Women 25–34 Women 35–49 Determinant Mean Mean Mean 2005 2010 2005 2010 2005 2010 difference difference difference Cumulative fertility 0.387 0.324 –0.063 2.938 2.531 –0.407 5.993 5.445 –0.548 Age 19.32 19.3 –0.02 29.2 28.98 –0.22 41.5 41.49 –0.01 Ever given birth (percent) 0.235 0.229 –0.006 0.884 0.854 –0.03 0.972 0.964 –0.008 Number of years of education 3.91 5.33 1.42 4.72 4.51 –0.21 3.06 4.06 1 Completed primary (percent) 0.089 0.233 0.144 0.12 0.119 –0.001 0.127 0.209 0.082 Still in school (percent) 0.206 0.374 0.168 0 0 0 0 0 0 Never went to school (percent) 0.128 0.061 –0.067 0.209 0.15 –0.059 0.422 0.303 –0.119 Literate (percent) 0.65 0.775 0.125 0.652 0.68 0.028 0.438 0.552 0.114 Number of years of education household head 3.43 3.82 0.39 4.23 4.24 0.01 3.14 3.79 0.65 Woman works (percent) 0.608 0.734 0.126 0.803 0.904 0.101 0.848 0.916 0.068 Woman has nonagricultural occupation (percent) 0.11 0.243 0.133 0.123 0.173 0.05 0.083 0.129 0.046 Household wealth (asset index) –0.055 0.286 0.341 -0.16 0.217 0.377 –0.261 0.122 0.383 Ever in union (percent) 0.238 0.215 –0.023 0.876 0.836 –0.04 0.968 0.957 –0.011 Age at first union 18.63 19.19 0.56 20.1 20.69 0.59 20.42 20.76 0.34 Age at first birth 19.3 19.56 0.26 21.13 21.55 0.42 21.81 22.02 0.21 No exposure to mass media (percent) 0.4 0.274 –0.126 0.433 0.318 –0.115 0.493 0.335 –0.158 No exposure to family planning messages (percent) 0.626 0.364 –0.262 0.54 0.31 –0.23 0.582 0.317 –0.265 Currently using modern contraception (percent) 0.021 0.098 0.077 0.095 0.405 0.31 0.073 0.307 0.234 Number of respondents 4,951 5,655 3,205 4,317 3,165 3,699 Sources: DHS 2005, 2010 and World Bank staff calculations. Rwanda Economic Update | Edition No. 5 53 Annexes Annexes Annex IIC: Statistical Decomposition Method and Results 21 Annex Table IIC.1: Decomposition of the drop in fertility, 2005–10 (1) (2) Cumulative fertility in 2005 survey 2.68 2.68 Cumulative fertility in 2010 survey 2.42 2.42 –0.26*** –0.26*** Difference between 2010 and 2005 [0.036] [0.036] –0.084*** –0.084*** Part explained by covariates [0.029] [0.029] –0.183*** –0.182*** Unexplained (changes in coefficients) [0.022] [0.022] Percentage of fertility drop explained by Increase in education (number of years) 22.7*** 21.9*** Increase in labor force participation 5.4*** In agricultural employment 0.8** In nonagricultural employment 8.1*** Increase in household wealth 15.4*** 12.7*** Total number of observations 24,992 24,992 Number of observations 2005 11,321 11,321 Number of observations 2010 13,671 13,671 Source: DHS 2005, 2010 and World Bank staff calculations Note: *** statistically significant at the 1 percent level; ** statistically significant at the 5 percent level; * statistically significant at the 10 percent level. Robust standard errors in brackets 21 Performing the decomposition using the estimated coefficients from the pooled sample ( ) risks inappropriately transferring some of the unexplained part of the difference to the explained part. To avoid doing so, we include a group indicator (dummy for survey period) in the pooled model (Jan 2008). 54 Rwanda Economic Update | Edition No. 5 Annexes The results of the decomposition by age group are presented in Annex Table IIC.2. The drop in cumulative fertility between 2005 and 2010 is greater for women 15–24 (16 percent) than for women 25–34 (14 percent) or 25–49 (9 percent). For the youngest group, the increase in education explains almost two-thirds of the decline in fertility. Improved standards of living in their households explain 21 percent; the uptake of nonfarm economic activities explains another 13 percent. Taken together, these three factors explain almost 98 percent of the drop in fertility for women 15–24. Overall, education emerges as the main driver the fertility decline for all but women in the middle age group. Women in this age group surveyed in 2010 were exposed to the genocide during their school-going years, which resulted in lower educational attainment than women of the same age surveyed in 2005. The decline in educational outcomes for this group negatively affected the drop in fertility, reducing the drop in fertility by 4 percent. Annex Table IIC.2: Decomposition of the fertility decline by age group, 2005–10 15–24 25–34 35–49 Cumulative fertility 2005 survey 0.387 2.938 5.993 Cumulative fertility 2010 survey 0.324 2.531 5.445 –0.063*** –0.407*** –0.548*** Difference in fertility between 2010 and 2005 [0.015] [0.044] [0.065] Percentage difference –16.3 –13.8 –9.1 –0.077*** –0.065** –0.11*** Explained by covariates [0.009] [0.026] [0.028] 0.014 –0.342*** –0.438*** Unexplained (changes in coefficients) [0.013] [0.039] [0.063] Percentage of fertility drop explained by Increased education (number of years) 65.1*** –3.9** 22.8*** Increased agricultural employment 0 –1.2 –0.4 Increased nonagricultural employment 12.7*** 1.5* 4.4*** Increased household wealth 20.6*** 10.8*** 3.8 Total number of observations 10,606 7,522 6,864 Number of observations in 2005 4,951 3,205 3,165 Number of observations in 2010 5,655 4,317 3,699 Source: DHS 2005, 2010 and World Bank staff calculations. Note: *** statistically significant at the 1 percent level; ** statistically significant at the 5 percent level; * statistically significant at the 10 percent level. Robust standard errors in brackets. Rwanda Economic Update | Edition No. 5 55 Annexes Annex IID: The Demographically Based Forecasting Model 56 Rwanda Economic Update | Edition No. 5 References ▪ Aiyar, S., and A. Mody. 2011. “The Demographic Dividend: Evidence from the Indian States.” IMF Working Paper 11/38, International Monetary Fund, Washington DC. ▪ Arezki R, L. d. 2011. “The Relative Volatility of Commodity Prices: A Reappraisal.” IMF Working Paper, December, International Monetary Fund, Washington DC. ▪ Becker, G. 1981. “A Treatise on the Family.” Cambridge, MA: Harvard University Press. ▪ Becker, G., Murphy, K, and R. Tamura. 1990. “Human Capital, Fertility, and Economic Growth.” Journal of Political Economy 98 (5): 12–37. ▪ Bloom, D., and D. Canning. 2003. “Contraception and the Celtic Tiger.” Economic and Social Review 34 (3): 229–47. ▪ —. 2011. “Demographics and Development Policy.” Program on the Global Demography of Aging Working Paper 66, Harvard School of Public Health, Cambridge, MA. ▪ Bloom, D., D. Canning, D. Evans, B. Graham, B., P. Lynch, and E. Murphy. 1999. “Population Change and Human Development in Latin America.” Background paper for IPES 2000, Harvard Institute for International Development, Cambridge, MA. ▪ Bloom, D., D. Canning, and J. Sevilla. 2003. “The Demographic Dividend. A New Perspective on the Economic consequences of Population Change.” RAND Corporation, Santa Monica, CA. ▪ Bloom, D., and J. Williamson. 1998. “Demographic Transitions and Economic Miracles in Emerging Asia.” World Bank Economic Review 12 (3): 419–55. ▪ Bongaarts, J. 2008. “Fertility Transitions in Developing Countries: Progress or Stagnation?” Studies in Family Planning 39 (2): 105–10. ▪ BNR (Banque Nationale de Rwanda). 2013. “Monetary Development: Current Trends and Short-Term Policy Stance.” MPC Working Document. http://bnr.rw/index.php?id=343. ▪ Canning, D. 2011. “The Causes and Consequences of the Demographic Transition.” Program on the Global Demography of Aging Working Paper 79, Harvard School of Public Health, Cambridge, MA. ▪ Christiaensen, L., De Weerdt, J. and Y. Todo 2013. “Urbanization and Poverty Reduction: The Role of Rural Diversification and Secondary Towns.” Agricultural Economics 44(4–5): 435–47. ▪ Cohen, B. 1998. “The Emerging Fertility Transition in Sub-Saharan Africa.” World Development 26 (8): 1431–61. ▪ Filmer, D., and L. Pritchett. 2001. “Estimating Wealth Effects without Expenditure Data—or Tears: An Application to Educational Enrolments in States of India.” Demography 38 (1): 115–32. ▪ Doing Business Indicators (database). World Bank, Washington, DC. http://www.doingbusiness.org/data. ▪ Galor, O. 2006. “Human Capital, Fertility and Growth.” Working Paper 2006-22, Brown University, Department of Economics, Providence, RI. ▪ Ghosh, S. R., and A. Kraay A. 2000. “Measuring Growth in Total Factor Productivity.” PREM Note 42, World Bank, Poverty Reduction and Economic Management, Washington, DC. ▪ Gribble, J., and J. Bremner. 2012. “Achieving a Demographic Dividend.” Population Bulletin 67 (2). ▪ IMF (International Monetary Fund), 2013. “Sub-Saharan Africa -Building Momentum in a Multi-Speed World.” Regional Economic Outlook. Washington, D.C. ▪ —. (forthcoming). Rwanda “Seventh Review under the Policy Support Instrument and Request for a New Policy Support Instrument.” IMF Country Report. ▪ Jann, B. 2008. “The Blinder-Oaxaca Decomposition for Linear Regression Models.” Stata Journal 8 (4): 453–79. ▪ Jones, L., A. Schoonbroodt, and M. Tertilt. 2008. “Fertility Theories: Can They Explain the Negative Fertility-Income Relationship?” NBER Working Paper 14266, National Bureau of Economic Research, Cambridge, MA. ▪ Kuepie, M., C. J. Nordman, and F. Roubaud. 2009. “Education and Earnings in Urban West Africa.” Journal of Comparative Economics 37: 491–515. ▪ Lindh, T., and B. Malmberg. 2007. “Demographically Based Global Income Forecasts up to the Year 2050.” International Journal of Forecasting 23: 553–67. Rwanda Economic Update | Edition No. 5 57 References ▪ Mincer, J. 1963. “Opportunity Costs and Income Effects.” In Measurement in Economics, ed. C. Christ. Stanford, CA: Stanford University Press. ▪ Muhoza, D., Rutayisire, P., and A. Umubyeyi. 2013. “Measuring the Success of Family Planning Initiatives in Rwanda: A Multivariate Decomposition Analysis.” DHS Working Paper 94. Calverton, Maryland: ICF Macro ▪ UN (United Nations). 2011. “World Population Prospects: The 2010 Revision.” Department of Economic and Social Affairs, New York. ▪ —.2013. “World Population Prospects: The 2012 Revision.” Department of Economic and Social Affairs, New York. ▪ Weerasinghe, D., and N. Parr. 2002. “Effect of Wealth on Marital Fertility in Sri Lanka.” Journal of Health, Population and Nutrition 20 (2): 112–19. ▪ Westoff, C. 2012. “The Recent Fertility Transition in Rwanda.” Population and Development Review 38 (supplement): 169–78. ▪ World Bank. 1993. “The East Asian Miracle: Economic Growth and Public Policy.” Washington DC: World Bank. ▪ —. 2011. “Demography and Economic Growth in Rwanda.” World Bank Policy Note,Washington DC. ▪ World Development Indicators (database). World Bank, Washington, DC.http://data.worldbank.org/data-catalog/world- development-indicators. 58 Rwanda Economic Update | Edition No. 5 Appendixes Appendixes Appendix 1: Rwanda: Selected Economic Indicators 2009 2010 2011 2012 H1 2013 GDP Growth Rate (percent) 6.2 7.2 8.2 8.0 5.9 Agriculture 7.7 5.0 4.7 3.0 5.4 Industry 1.3 8.4 17.6 7.2 13.6 Services 6.2 9.0 8.9 12.2 4.2 Fiscal Framework (percent of GDP) 1 2 Total revenues 24.4 25.6 24.8 25.7 23.9 Domestic revenues 12.7 12.5 13.9 14.5 16.0 Tax revenues 12.2 12.0 13.3 13.6 14.2 Non-tax revenues 0.6 0.5 0.6 0.9 1.8 Grants 11.6 13.1 10.9 11.2 7.9 Budgetary Grants 7.8 9.0 6.2 6.5 4.1 Capital Grants 3.8 4.0 4.7 4.7 3.8 Total Expenditures & net lending 25.3 25.7 28.3 26.9 29.0 Recurrent Expenditure 14.2 14.7 15.1 15.0 13.8 Capital expenditure 9.9 10.1 12.6 11.8 12.3 Domestic 4.7 5.1 6.3 5.7 5.2 Foreign 5.2 5.0 6.3 6.2 7.1 Net lending 1.2 0.9 0.5 0.0 3.0 Budget deficit (cash basis) Excluding grants -12.9 -13.6 -14.7 -12.7 -13.2 Including grants -1.3 -0.5 -3.8 -1.5 -5.3 External Sector Exports (year-on-year growth) -12.2 26.5 56.2 27.3 34.2 Imports (year-on-year growth) 13.2 8.7 44.5 25.6 1.1 Gross Reserves (Million US$) 744.3 813.3 1,050.0 850.3 1,068.8 Gross Reserves (months of imports of GS) 5.4 4.5 5.1 3.6 4.6 Consumer Price index (percentage change) End of period 5.7 0.2 8.3 3.9 3.7 Period average 10.7 2.3 5.7 6.3 4.0 Exchange rate (Rwf/US$) End of period 571.2 594.45 604.1 631.4 642.7 Period average 568.3 583.3 600.3 614.3 636.5 Source: NISR, BNR, MINECOFIN 1 On a fiscal-year basis (July–June). For example, the column ending in 2011 refers to FY2010/11 2 On a fiscal year basis after 2009 60 Rwanda Economic Update | Edition No. 5 Appendixes Appendix 2: Rwanda - Gross Domestic Product by Kind of Activity 2009 2010 2011 2012 2013 H1 H2 H1 H2 H1 H2 H1 H2 H1 (Rwf billion at constant 2006 prices) Gross Domestic Product 1,072 1,110 1,134 1,205 1,205 1,327 1,304 1,430 1,380 Agriculture 381 395 397 418 399 454 413 465 436 Food crops 324 335 337 355 338 388 352 398 371 Export crops 9 10 10 12 10 13 8 13 10 Livestock 17 17 17 18 18 18 19 19 19 Forestry 28 29 29 30 30 31 31 32 32 Fisheries 3 3 3 3 3 3 4 4 4 Industry 149 152 157 168 180 203 189 221 215 Mining and quarrying 5 5 4 5 6 8 6 7 9 Manufacturing 60 68 66 74 68 83 69 77 70 of which: Food 26 28 27 31 28 33 28 34 29 Electricity, gas and water 2 3 3 3 3 3 4 4 4 Construction 81 76 84 86 103 108 110 134 132 Services 481 496 515 550 560 599 633 667 660 Wholesale and retail trade 138 137 142 155 153 174 173 195 185 Hotels and restaurants 20 21 22 23 22 24 25 25 25 Transport, storage, communication 88 93 96 101 97 110 117 131 124 Finance, insurance 28 26 31 36 40 40 45 49 49 Real estate, business services 74 83 77 81 80 78 89 84 83 Public administration 52 52 57 61 65 71 76 77 80 Education 54 54 59 59 69 69 73 73 79 Health 16 17 19 20 20 20 22 21 21 Other personal services 12 13 13 14 14 13 14 14 14 Adjustments 61 67 65 69 66 71 68 76 70 Less: Imputed bank service charge -16 -13 -16 -18 -21 -25 -26 -27 -30 Plus: VAT and other taxes on products 77 80 82 87 87 96 94 103 100 (Rwf billion at current prices) Gross Domestic Product 1,437 1,548 1,579 1,701 1,783 2,031 2,050 2,314 2,292 Agriculture 478 534 520 538 554 669 653 784 743 Food crops 402 456 440 452 468 572 555 676 634 Export crops 11 12 13 15 13 16 11 18 15 Livestock 24 24 25 26 27 29 30 31 32 Forestry 36 37 37 40 39 45 49 52 53 Fisheries 5 5 5 6 6 7 7 7 8 Industry 209 222 230 260 288 337 316 379 374 Mining and quarrying 8 8 8 14 22 26 19 20 29 Manufacturing 87 104 100 118 110 142 120 139 126 of which: Food 38 43 43 50 47 61 51 63 54 Electricity, gas and water 3 3 3 4 4 4 5 6 5 Construction 112 107 120 125 151 165 172 215 214 Services 658 700 738 794 827 905 953 1,011 1,040 Wholesale and retail trade 191 194 203 229 224 266 264 300 284 Hotels and restaurants 29 32 33 35 34 38 41 40 41 Transport, storage, communication 104 120 124 132 133 155 161 183 172 Finance, insurance 34 30 39 42 53 47 63 67 74 Real estate, business services 133 149 142 151 151 155 172 164 165 Public administration 64 65 72 78 86 95 106 108 116 Education 68 72 84 84 103 105 100 102 141 Health 21 23 25 27 26 26 29 29 29 Other personal services 15 16 16 17 17 17 18 18 19 Adjustments 92 93 91 108 115 120 127 139 135 Less: Imputed bank service charge -22 -19 -23 -26 -32 -38 -42 -44 -50 Plus: VAT and other taxes on products 114 111 115 135 147 158 169 183 185 Source: NISR Rwanda Economic Update | Edition No. 5 61 Appendixes Appendix 3: Gross Domestic Product by Expenditure 2009 2010 2011 2012 2013 H1 H2 H1 H2 H1 H2 H1 H2 H1 (Rwf billion at constant 2006 prices) Gross Domestic Product 1,073 1,110 1,134 1,205 1,205 1,327 1,304 1,430 1,380 Total final consumption expenditure 1,043 1,097 1,110 1,211 1,172 1,272 1,255 1,369 1,208 Government 184 176 194 203 202 213 239 237 246 Private (includes changes in stock) 859 920 916 1,008 970 1,059 1,016 1,132 962 Gross capital formation 247 220 238 262 272 289 300 351 326 Gross fixed capital formation 247 220 238 262 272 289 300 351 326 Construction 160 149 167 171 204 216 219 268 265 Durable capital goods 87 70 71 91 67 73 81 83 61 Resource balance -216 -206 -214 -268 -239 -234 -251 -291 -153 Exports of Goods and NFS 109 113 110 124 162 184 208 258 305 Goods (fob) 39 44 47 60 75 100 126 175 214 Non-factor services 71 69 63 64 88 84 82 83 91 Imports of goods and NFS 326 319 324 392 401 418 459 548 458 Goods (fob) 215 212 216 269 269 291 322 397 324 Non-factor Services 111 107 108 123 132 127 137 151 134 Domestic demand 1,290 1,316 1,348 1,473 1,444 1,561 1,555 1,720 1,534 External Demand -216 -206 -214 -268 -239 -234 -251 -291 -153 (Rwf billion at current prices) Gross Domestic Product 1,437 1,548 1,579 1,701 1,783 2,031 2,050 2,314 2,292 Total final consumption expenditure 1,387 1,533 1,581 1,685 1,730 1,929 2,029 2,229 2,122 Government 229 225 256 268 274 294 330 331 370 Private (includes changes in stock) 1,158 1,307 1,325 1,418 1,456 1,635 1,699 1,898 1,752 Gross capital formation 341 303 330 357 381 437 456 541 538 Gross fixed capital formation 341 303 330 357 381 437 456 541 538 Construction 221 210 237 247 301 330 343 430 429 Durable capital goods 121 93 94 111 80 107 112 111 108 Resource balance -292 -287 -332 -342 -328 -334 -435 -457 -368 Exports of Goods and NFS 144 158 150 179 237 276 261 313 399 Goods (fob) 47 61 61 88 108 147 131 180 248 Non-factor services 97 97 89 92 129 129 130 134 150 Imports of goods and NFS 436 445 482 521 566 611 696 770 767 Goods (fob) 288 296 322 357 380 425 488 557 543 Non-factor Services 148 149 160 164 185 186 208 213 224 Domestic demand 1,729 1,835 1,911 2,043 2,111 2,366 2,484 2,770 2,660 External Demand -292 -287 -332 -342 -328 -334 -435 -457 -368 Source: NISR 62 Rwanda Economic Update | Edition No. 5 Appendixes Appendix 4: Rwanda - Tourism Sector Data Tourism Arrivals Park Visits Visiting Holiday/ Business / Transit & friends & TOTAL VOLCANOES AKAGERA NYUNGWE TOTAL Vacation conference others relatives January 8,934 29,762 39,935 23,532 102,163 1,901 2,061 672 4,634 February 8,975 21,977 40,240 20,721 91,913 2,002 2,032 686 4,720 March 7,402 23,797 43,085 24,375 98,659 1,927 2,124 641 4,692 April 6,747 30,593 35,003 17,367 89,710 862 1,234 338 2,434 May 7,923 25,648 36,875 17,148 87,594 1,151 2,017 390 3,558 2013 June 9,342 26,397 32,436 16,515 84,690 2,379 2,740 581 5,700 July 9,584 30,170 43,370 12,500 95,624 3,208 3,673 820 7,701 August 12,033 32,274 31,295 18,269 93,871 3,254 3,345 705 7,304 September 7,862 25,998 29,906 16,248 80,014 3,034 2,845 604 6,483 Total 78,802 246,616 332,145 166,675 824,238 19,718 22,071 5,437 47,226 January 8,126 28,711 27,977 15,794 80,608 2,738 2,186 655 5,579 February 8,775 19,956 33,441 18,481 80,653 2,516 1,856 681 5,053 March 7,848 21,236 33,684 18,413 81,181 1,945 1,315 444 3,704 April 5,890 22,691 33,828 16,554 78,963 1,443 1,269 416 3,128 May 5,167 23,405 40,168 17,569 86,309 1,627 1,492 420 3,539 June 7,364 23,697 29,491 19,104 79,656 2,690 2,384 634 5,708 2011 July 9,663 25,186 36,097 22,034 92,980 3,149 3,457 855 7,461 August 10,693 33,299 34,014 22,242 100,248 3,219 2,984 994 7,197 September 10,102 25,112 32,531 19,878 87,623 2,843 1,786 477 5,106 October 8,961 25,105 35,042 22,869 91,977 2,906 1,443 521 4,870 November 5,810 27,292 45,012 30,102 108,216 1,583 2,605 689 4,877 December 9,403 33,096 39,663 25,214 107,376 1,824 2,423 634 4,881 Total 97,802 308,786 420,948 248,254 1,075,790 28,483 25,200 7,420 61,103 January 6,775 21,843 22,282 4,443 55,343 2,100 1,724 576 4,400 February 5,622 17,039 36,471 9,861 68,993 2,208 1,340 556 4,104 March 6,386 19,788 39,795 10,783 76,752 2,114 1,732 387 4,233 April 5,849 18,508 35,888 11,045 71,290 1,336 1,416 528 3,280 May 5,488 24,671 29,334 7,852 67,345 1,311 1,431 488 3,230 June 5,889 26,912 19,902 13,375 66,078 2,378 1,663 690 4,731 2011 July 8,559 31,862 24,551 14,626 79,598 3,079 2,826 1,253 7,158 August 9,298 31,205 21,172 15,393 77,068 3,055 2,836 1,386 7,277 September 5,169 23,514 27,529 5,164 61,376 2,623 1,323 831 4,777 October 5,379 26,234 45,220 15,658 92,491 2,671 2,465 601 5,737 November 6,250 25,531 46,243 14,611 92,635 1,787 1,277 381 3,445 December 10,955 27,807 47,176 13,102 99,040 2,159 2,424 597 5,180 Total 81,619 294,914 395,563 135,913 908,009 26,821 22,457 8,274 57,552 Source: BNR and NISR Rwanda Economic Update | Edition No. 5 63 Appendixes Appendix 5: Rwanda - Inflation indicators (year-on-year percent change) Month Overall Food prices Core Energy prices Import Prices January 1.1 -1.9 0.7 2.4 1.7 February 2.6 0.9 1.6 3.7 2.9 March 4.1 4.2 2.8 4.9 5.1 April 5.0 6.2 3.6 4.5 5.7 May 4.5 4.7 4.3 4.8 6.9 June 5.8 7.2 5.8 5.0 8.6 2011 July 7.1 10.6 7.0 6.0 9.2 August 7.5 10.3 8.2 7.3 10.4 September 6.6 6.3 9.0 7.3 10.7 October 7.8 9.6 8.9 7.3 10.1 November 7.4 9.1 8.1 6.7 9.0 December 8.3 11.2 8.3 9.3 8.6 January 7.8 12.7 7.1 8.4 7.9 February 7.9 15.5 6.0 5.8 6.0 March 8.2 15.5 5.3 8.3 4.9 April 6.9 12.8 4.8 6.9 3.8 May 8.3 15.1 5.4 10.8 3.1 June 5.9 11.3 3.7 6.6 2.6 2012 July 5.6 10.4 3.0 8.8 2.6 August 5.8 12.6 2.5 5.4 1.2 September 5.6 13.7 2.1 2.8 1.2 October 5.4 12.1 2.5 5.5 2.7 November 4.5 9.8 2.8 5.9 2.9 December 3.9 7.9 2.5 5.7 3.2 January 5.7 8.3 4.7 5.6 3.0 February 4.8 4.7 5.1 8.5 4.0 March 3.2 1.9 4.8 4.6 3.4 April 4.4 4.1 5.2 6.4 4.0 May 3.0 2.6 3.6 2.5 3.5 2013 June 3.7 4.4 3.4 0.9 1.9 July 3.5 4.0 3.6 -0.9 1.5 August 4.0 4.9 3.6 2.0 2.7 September 5.1 7.8 3.3 2.8 2.5 October 5.1 8.2 3.2 0.3 1.2 November 4.6 6.4 3.4 0.2 2.3 Source: BNR and NISR 64 Rwanda Economic Update | Edition No. 5 Appendixes Appendix 6: Rwanda - Key Interest Rates (percent) Treasury Bill rates Key Average Average Interbank Month Repo Deposit Lending rate Weighted Rate rate rate 28 days 91 days 182 days 364 days Average Rate January 6.7 6.1 6.4 7.2 - 7.2 February 6.0 7.5 16.9 6.7 6.2 6.4 7.2 - 7.0 March 6.0 7.5 16.6 6.7 6.4 6.9 7.4 - 7.2 April 6.0 8.7 16.6 6.9 6.4 6.8 7.2 - 7.1 May 6.0 7.9 16.9 6.9 6.2 6.7 7.2 - 7.0 June 6.0 8.0 17.0 7.0 6.1 6.5 6.9 - 6.8 2011 July 6.0 6.8 16.6 6.9 6.1 6.4 7.2 - 6.8 August 6.0 7.7 17.0 6.9 6.1 6.2 7.2 - 6.7 September 6.0 7.7 17.0 6.9 6.3 6.5 6.9 - 6.7 October 6.5 7.4 17.0 7.4 6.8 7.0 7.2 - 7.2 November 7.0 8.0 16.5 7.5 6.8 7.2 7.7 - 7.8 December 7.0 8.0 16.7 8.1 7.0 7.3 7.6 - 7.6 January 7.0 7.4 17.0 7.3 7.1 7.3 7.7 8.4 7.6 February 7.0 8.3 16.3 6.9 7.1 7.6 7.4 8.0 7.6 March 7.0 8.2 16.3 7.7 7.4 7.6 7.9 7.8 7.7 April 7.0 8.1 16.9 8.0 7.8 7.6 7.9 8.5 7.9 May 7.5 9.9 16.7 8.6 7.9 8.1 8.3 8.9 8.3 June 7.5 7.9 16.8 9.0 8.8 9.6 9.4 9.1 9.3 2012 July 7.5 8.9 16.5 9.1 9.4 10.2 - - 9.8 August 7.5 8.6 17.1 9.5 10.6 10.2 10.5 11.7 11.1 September 7.5 8.5 17.1 10.8 11.5 12.1 12.0 12.7 12.3 October 7.5 9.2 16.6 10.9 11.9 12.4 12.5 - 12.1 November 7.5 11.2 16.7 11.9 11.8 12.5 12.7 - 12.4 December 7.5 10.7 16.5 11.1 11.8 12.6 12.8 - 12.4 January 7.5 11.3 17.1 11.1 12.1 12.6 12.8 - 12.4 February 7.5 10.3 17.0 10.4 11.6 12.3 12.7 - 12.2 March 7.5 10.4 17.2 10.0 11.0 12.1 12.6 12.8 12.1 April 7.5 10.7 17.3 10.9 11.2 12.3 12.8 13.0 12.0 May 7.5 10.6 17.6 11.1 11.0 12.0 12.4 12.7 12.0 2013 June 7.0 10.6 17.7 9.6 10.0 10.7 11.3 11.7 10.8 July 7.0 8.5 17.2 9.6 8.9 9.6 10.0 10.7 9.7 August 7.0 10.5 17.5 7.6 7.8 8.3 8.9 9.3 8.6 September 7.0 9.0 17.8 7.0 6.8 6.9 7.3 7.8 7.1 October 7.0 9.5 17.4 6.7 6.2 6.5 6.7 7.6 6.8 November 7.0 - - 6.1 5.5 5.9 6.2 7.0 6.1 Source: BNR Rwanda Economic Update | Edition No. 5 65 Appendixes Appendix 7: Rwanda – Exchange Rate (Monthly Average) Ugandan Kenya Tanzanian Burundian Month US dollar Euro UK Pound Shilling Shilling shilling Franc January 596.75 796.76 939.46 0.26 7.37 0.41 0.49 February 600.24 818.70 967.47 0.26 7.37 0.40 0.49 March 599.53 838.94 969.66 0.25 7.12 0.40 0.49 April 601.27 867.09 982.15 0.25 7.17 0.40 0.49 May 599.28 860.98 979.81 0.25 7.02 0.40 0.49 June 600.00 863.18 974.24 0.25 6.76 0.38 0.49 2011 July 600.51 856.74 967.83 0.24 6.66 0.38 0.49 August 599.75 860.21 981.83 0.22 6.56 0.38 0.49 September 599.89 826.21 947.67 0.22 6.39 0.37 0.49 October 601.29 822.51 945.55 0.22 5.98 0.36 0.50 November 601.77 817.69 952.40 0.24 6.51 0.35 0.49 December 603.45 796.17 942.33 0.25 6.99 0.38 0.48 January 604.37 779.26 936.44 0.25 7.11 0.39 0.47 February 605.26 800.37 955.50 0.26 7.40 0.39 0.47 March 606.75 801.24 959.52 0.25 7.44 0.39 0.47 April 607.01 799.45 971.24 0.25 7.39 0.39 0.46 May 608.58 780.82 970.12 0.25 7.32 0.39 0.45 June 609.94 764.00 947.89 0.25 7.32 0.39 0.44 2012 July 612.95 752.14 955.23 0.25 7.40 0.39 0.44 August 613.63 760.29 963.86 0.25 7.43 0.40 0.43 September 618.08 793.51 994.28 0.25 7.43 0.40 0.43 October 625.24 810.86 1,006.08 0.25 7.47 0.40 0.43 November 628.77 806.64 1,003.95 0.24 7.46 0.40 0.43 December 630.99 827.21 1,018.50 0.24 7.46 0.40 0.42 January 631.29 838.05 1,008.81 0.24 7.38 0.40 0.42 February 633.25 846.82 981.39 0.24 7.36 0.40 0.41 March 634.98 824.27 957.00 0.24 7.52 0.40 0.41 April 637.38 829.03 974.68 0.25 7.69 0.40 0.41 May 640.13 831.41 979.34 0.25 7.73 0.40 0.41 2013 June 641.66 846.19 993.12 0.25 7.61 0.40 0.42 July 645.22 843.25 980.34 0.25 7.55 0.41 0.42 August 649.01 864.16 1,005.03 0.25 7.53 0.41 0.43 September 653.26 871.37 1,033.65 0.26 7.60 0.41 0.43 October 661.29 901.19 1,064.45 0.26 7.88 0.42 0.43 November 664.30 897.29 1,068.75 0.27 7.84 0.42 0.43 Source: BNR 66 Rwanda Economic Update | Edition No. 5 Appendixes Appendix 8: Rwanda – Gross International Reserves Month Rwf billion US$ million 2011 December 634 1,050 January 597 987 February 582 960 March 546 899 April 514 845 May 464 762 June 526 859 2012 July 473 771 August 451 733 September 449 721 October 471 750 November 477 757 December 535 850 January 465 736 February 436 688 March 444 698 April 452 707 May 656 1,024 2013 June 687 1,069 July 659 1,018 August 657 1,012 September 681 1,035 October 691 1,045 Source: BNR Rwanda Economic Update | Edition No. 5 67 The World Bank, Rwanda Blvd. de la Revolution SORAS Building P. O. Box 609 Kigali, Rwanda Telephone: + 250 252 591300 Fax: +250 252 576385 www.worldbank.org/rw Produced by Poverty Reduction and Economic Management Unit Africa Region Photo credits: © Peace Aimee Niyibizi and Rogers Kayihura/World Bank Design by: Robert Waiharo