Republic of Congo Poverty Assessment Report Education, Jobs and Social Protection for a Sustainable Reduction of Poverty Executive Summary Republic of Congo Poverty Assessment Report Education, Jobs and Social Protection for a Sustainable Reduction of Poverty May 2017 Executive Summary Standard Disclaimer: This volume is a product of the staff of the International Bank for Reconstruction and Development/ The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Copyright Statement: The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@ worldbank.org. Photography: World Bank, Franck Sidney Chrysantheme Bitemo, Clementine Maoungou and Clarence Tsimpo. Design and layout: The Word Express, Inc. Acknowledgements T he World Bank greatly appreciates the close collaboration with the Government of the Republic of Congo (the Ministry of Planning and Integration1 and the National Statistics Office,2 in particular) in the preparation of this report. The team preparing this report was led by Clarence Tsimpo Nkengne (Senior Economist, GPV01) and consisted of Fulbert Tchana Tchana (Senior Economist, GMF07), Etaki Wa Dzon (Economist, GMF07), and Quentin Wodon (Lead Economist, GEDGE). This report was prepared under the guidance of Ahmadou Moustapha Ndiaye (Country Director, AFCC2), Djibrilla Adamou Issa (Country Manager, AFMCG), Pablo Fajnzylber (Practice Manager, GPV01), Pierella Paci (Acting Practice Manager, GPV01), Yisgedullish Amde (Country Program Coordinator, AFCCD), Emmanuel Pinto Moreira (Program Leader, AFCC2), and Johannes G. Hoogeveen (Lead Economist, GPV01). Kathleen Beegle (Program Leaders, AFCW1), Dean Mitchell Jolliffe (Senior Economist, DECPI), and Vasco Molini (Senior Economist, GPV01) were peer reviewers for the report. In addition, the team is grateful for comments and input from Andrew Dabalen, Prospere Backiny, Paolo Verme, Franck Adoho, Gyorgy Bela Fritsche, Rose Mungai, Mahine Diop, Meskerem Mulatu, Amadou Oumar Ba, Ephraim Kebede, Michel Niama (Director-General of the Economy), Samuel Ambapour Kosso (Executive Director, CNSEE), Théophile Bassissila (Economist, CNSEE), Jean-Christophe Boungou Bazika (CERAPE), Euloge Bikindou-Boueya (UERPOD), and the developments partners’ community. Josiane Maloueki Louzolo (Program Assistant, AFMCG), Aimee Niane (Program Assistant, GPV01), Senait Yifru (Operations Analyst, GPV01), and Arlette Sourou (Senior Program Assistant, GPV01) provided assistance and various supports in the preparation of the report. Thanks are especially due to Martin Buchara (Program Assistant, GPV01) for excellent support during the preparation of this report. In addition, the report benefited from proofreading by Nathan Weatherdon and Rachel Nalubega. 1 Ministère du Plan et de l’Intégration. 2 Centre Nationale de la Statistique et des Etudes Economiques (CNSEE). 3 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) Abbreviations and Acronyms ARV Anti-retroviral CBN Cost of Basic Needs CNSEE Centre Nationale de la Statistique et des Etudes Economiques CERAPE Centre d’Etudes et de Recherche sur les Analyses et Politiques Economiques CONFEMEN Conférence des ministres de l’Éducation des Etats et gouvernements de la Francophonie CPI Consumer Price Index DHS Demographic and Health Survey DPT Diphtheria, Pertussis, and Tetanus DRC Democratic Republic of Congo ECOM Enquête Congolaise auprès des Ménages FAO-STAT The Statistics Division of Food and Agriculture Organization of the United Nations FBO Faith-based Organization FGT Foster-Greer-Thorbecke GAVI Global Alliance for Vaccines and Immunization GDP Gross Domestic Product GIE Gini Income Elasticity GNI Gross National Income GPS Global Positioning System HiA Health in Africa IBRD International Bank for Reconstruction and Development ICT Information and Communication Technology IFC International Finance Corporation MDGs Millennium Development Goals MFM Macroeconomics and Fiscal Management MOHP Ministry of Health and Population (Ministère de la santé et la population) NDP National Development Plan NHA National Health Accounts NIS National Institute of Statistics NSO National Statistical Office OLS Ordinary Least Squares 5 PEEDU Projet Eau, Electricité et Développement Urbain PPP Purchasing Power Parity PRSP Poverty Reduction Strategy Paper PSU Primary Sampling Unit RIF Recentered Influence Function ROC Republic of Congo SCD Systematic Country Diagnostic SCI Statistical Capacity Index SDI Service Delivery Indicators SMEs Small and Medium Enterprises SNDE Société Nationale de Distribution d’Eau SNE Société Nationale d’Electricité SSA Sub-Saharan Africa STATCAP Statistical Capacity Building UERPOD Union pour l’Etude et la Recherche sur la Population et le Développement WDI World Development Indicators Vice President: Makhtar Diop Senior Director: Carolina Sanchez-Paramo Country Director: Ahmadou Moustapha Ndiaye Country Manager: Djibrilla Adamou Issa Practice Manager: Pablo Fajnzylber Acting Practice Manager Pierella Paci Task Team Leader: Clarence Tsimpo Nkengne 6 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) Executive Summary T his poverty assessment analyzes recent trends in monetary and nonmonetary aspects of poverty and economic vulnerability in the Republic of Congo (ROC), based on two nation- ally representative and broadly comparable household expenditure surveys conducted by the National Institute of Statistics (NIS) in 2005 and 2011. The study determines the drivers of poverty reduction by systematically looking at demographic, labor, and human capital dimen- sions. The report also discusses cross-cutting issues relevant for poverty reduction, such as service delivery, marginalization of autochthons, and others. This study aims to provide policy makers with the knowledge needed to improve the effectiveness of their programs to reduce and finally eradicate extreme poverty in the Republic of Congo.3 Between 2005 and 2011, the Republic of Congo experienced a strong macroeconomic performance that was driven by oil revenues The country’s history has been characterized by ups and downs related to the performance of the oil sector, changes from communism to market economy, civil war, and social unrest. Since the 1960s, the country has experienced four economic phases: 1960–1972; 1973–1984; 1985–1999; and 2000–2014. These periods correspond, respectively, to the pre-oil economy; the first oil boom; the oil price crisis; and the second oil boom (World Bank 2016a). The country also experienced social unrest with a military coup and civil war. The various governments, also in line with the cold war, have moved back and forth from communism towards a market economy. All these affected the economy and people’s perceptions, especially toward the importance of the state as the main job provider. Dutch disease followed the discovery of oil. In the 1960s, the economy was diversified, but with the discovery of oil followed the reduction of the relative importance of other sectors. Most importantly, the agriculture and the manufacturing sectors shrank due to reduced competi- tiveness. As a consequence, the country imports most of its goods, even food. 3 Economic growth, economic diversification job creation and poverty reduction are at the core of the National De- velopment plan (Republic of Congo 2012a, 2012b). 7 The Republic of Congo experienced strong eco- day had declined in the Republic of Congo from 50.2 nomic growth between 2002 and 2015. This strong eco- percent in 2005 to 37.0 percent in 2011. nomic growth was driven by higher oil prices and political Despite the improvement in living standards, the stability. On average, during this period, The Republic of country is still underperforming given its potential Congo’s growth rate stood at 4.5 percent. Between 2005 and status as a middle-income country. Cross-country and 2011, the country grew strongly at an average annual comparisons suggest that the level of international rate of 5.4 percent mainly driven by high oil revenues from poverty in the Republic of Congo is still much higher oil production and the decision of the Government to step than in other comparable middle-income countries. up its investment in infrastructure in 2006. Countries with a similar level of economic development generally have much lower poverty rates. Moreover, the Between 2005 and 2011, the strong international poverty rate (US$1.90 –2011 PPP) in the Republic of Congo is quite close to the Sub-Saharan macroeconomic performance Africa (SSA) average of 42.6 percent. coincided with substantial poverty reduction Poverty reduction was sustained The strong macroeconomic performance translated beyond 2011, but it will be difficult into a substantial reduction of the proportion of for the country to reach the goals of the population living in poverty between 2005 and eradicating extreme poverty by 2030 2011. The proportion of the population living below the national poverty line declined from 50.7 percent in 2005 Microsimulation results suggest that the poverty to 40.9 percent in 2011, a decrease of 9.8 percentage decline has continued beyond 2011, although at a points which is in line with the gross domestic product slower pace. It is estimated that the slowdown of eco- (GDP) growth rate observed during that period. Overall, nomic growth linked to the oil sector led to a slower around 143,000 people moved out of poverty. Changes reduction in poverty after 2011. Between 2005 and in the poverty gap and squared poverty gap follow similar 2011, poverty declined by 1.63 percentage points patterns to those observed for the poverty headcount. annually. In 2016, due to slower economic growth, it is Nationally, despite population growth between the two estimated that the share of the population living below years, the number of poor decreased to 1,658,000 in the national poverty line was around 34 or 35 percent. 2011, from 1,801,000 in 2005. This corresponds to a poverty decline by 1.52 percentage Similar to the national poverty rate, the inter- points annually between 2011 and 2016. national extreme poverty rates have declined sig- Our projections up to 2030 show that, unless nificantly. National poverty measures are used for performance and inequality improve substantially, within-country poverty analysis and for deriving the it will be difficult but not impossible for the country economic policies for poverty eradication. However, the to reach the goals of eradicating extreme poverty by national poverty figures are generally not comparable 2030. Under a very optimistic scenario where the coun- across countries. To compare the Republic of Congo with try is assumed to achieve a 10 percent annual growth other countries in Africa and worldwide, the so-called rate between 2021 and 2030, the US$1.90 PPP per international extreme poverty measures are generally day poverty rate will be 3.6 percent in 2030. Under a used. The most common international poverty line is more realistic scenario based on projected growth by US$1.90 expressed in 2011 purchasing power parity World Bank staff, we project that the US$1.90 PPP (PPP) U.S. dollars. The share of the extremely poor by per day poverty rate will still be high, around 15 per- international standards living below US$1.90 PPP a cent in 2030. 8 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) The prosperity that the Republic in 2005 to 69.4 in 2011. In rural areas, the depth and of Congo enjoyed with oil windfall severity of poverty has also increased, that is, the rural poor have become poorer. Between 2005 and 2011, did not trickle down to the entire the number of poor increased in rural areas to 951,000 population up from 795,000. While rural areas accounted for 44.3 percent of the poor in 2005, this had increased to 57.4 Inequality levels remain high. There was a slight increase percent by 2011. of the Gini coefficient, although not statistically signi- Poverty is predominantly a rural phenomenon. ficant (0.460 in 2005 and 0.465 in 2011). This slight In addition, urban poverty remains very important, increase is coherent with the fact that poverty decreased especially Brazzaville. In rural areas, seven out of ten more in the largest cities than in other urban areas and (69.4 percent) people are poor; 57.4 percent of poor peo- rural areas. Per equivalent adult consumption among ple live in rural areas. Brazzaville, despite a relatively low the richest 10 percent of households in the Republic of poverty incidence (21.6 percent), has a big share of poor. Congo was 17.2 times that of the poorest in 2005; it Close to 20 percent of poor people live in Brazzaville. increased to 20.0 times by 2011. Thus, rural areas and Brazzaville account for 70 percent By international standards, inequalities are of the overall population and 77 percent of total poor. higher in the Republic of Congo. Cross-country As a consequence of inequality increase and the comparisons suggest that inequality is high in the urban/rural gap, prosperity was not shared with the Republic of Congo. The Republic of Congo is ranked poorest segments of the population. Growth incidence among the most unequal societies based on the World curves suggest that growth was not pro-poor nation- Development Indicators (WDI) data. Data on inequal- ally. The poorest actually experienced a deterioration in ity in 105 countries are available in the WDI beyond their living standards according to consumption-based 2010. The Republic of Congo ranked 90 out of 105 measures of poverty. Those in the middle of the distri- countries on the Gini. bution and a small share of the wealthiest households The autochthons, who represent about 1 percent experienced the large positive growth. of the population, stand out as the most marginal- Success in reducing poverty has resulted in many ized group in the Republic of Congo. Monetary pov- households that are living just above the poverty line erty headcount for autochthons is more than twice the who remain vulnerable to falling below the poverty poverty rate of the remaining population. Close to nine line in the face of a negative shock. As poverty receded, out of ten autochthons are poor. The marginalization of both the vulnerable (insecure nonpoor) and middle class autochthons is characterized by very limited access to groups expanded, with the middle class group expand- social services, including health and education, as well ing faster than the vulnerable/insecure nonpoor group. as the labor market. Thus, they contribute and benefit While only 20.6 percent of the population had con- very little from economic activities. sumption more than twice the poverty line in 2005, this The story that emerges from poverty estimates is increased to 26.3 percent by 2011. In the meantime, 32.8 that of a dual economy. Poverty is becoming an increas- percent of the population, while technically ‘nonpoor’, ingly rural phenomenon, and this should be a much is consuming at a level below an average XAF 550,000 bigger reason for concern. Most of the poverty reduc- yearly (in nominal 2011 CFA francs) per equivalent tion was observed in the two largest cities of Brazzaville adult. Given that a large share of the vulnerable rely and Pointe Noire. What was most worrisome was an on either agriculture or informal activities, which are increase in poverty headcount as well as in the number susceptible to significant output volatility, many remain of poor people in rural areas. In rural areas, the poverty at serious risk of falling back into poverty, at least on a headcount went up by 4.6 percentage points from 64.8 temporary basis. Executive Summary 9 On nonmonetary dimensions of Despite improvement over the last decade, well-being, the Republic of Congo access to electricity is very low compared to expecta- tions. Connection rates in the country remain below is performing less than the main expectations compared to peers. Coverage rates have comparators increased substantially, from 26.7 percent in 2005 to 42.5 percent in 2011. The improvement in coverage Performance is way below expectations on most of the is the result of improvement in both access (network other critical education sector indicators, including availability) and take-up. In 2011, close to seven out primary school completion.4 The Republic of Congo of ten households (68 percent) lived in neighborhoods is performing below expectations with regard to primary with electricity network; this is up from 57 percent in school completion. The primary school completion rate 2005. Reflecting the improvement in monetary poverty, was 74 percent in 2012. The primary school completion take-up rate also increased from 47 percent to 62 percent rate has been fluctuating over the last decade. between 2005 and 2011. There has been substantial improvement in child Although the share of the population with access and maternal mortality. However, the country still to safe water slightly increased during the last decade, performs below expectations with regard to maternal it is still far below what is expected. The Republic of mortality and has not reached any of the health-related Congo is performing below expectations with regard to Millennium Development Goals (MDGs). Child and access to improved water. Between 2005 and 2015, access maternal mortality are often used as a measure of the to improved water increased slightly from 72 percent efficiency of the health sector in a given country. Between to 77 percent. However, the country is still performing 2005 and 2012, under-five mortality, which measures below expectations. the probability of children dying between birth and the Access to improved sanitation remains very fifth birthday, dropped from 95.3 to 52.6 per 1,000 low and as a consequence, the country is performing births. Thanks to this improvement, the country is now below expectations on this dimension as well. The performing as expected given its gross national income Republic of Congo is performing below expectations (GNI) level. The story is a bit different regarding mater- regarding access to a safe toilet. In 2014, only 43 percent nal mortality. Despite improvement, the country is still of the population had access to improved sanitation. The performing lower than its peers on maternal mortality. situation is even worse in rural areas where only 13 per- Maternal mortality rate declined from 781 to 426 deaths cent of the population have access to an improved toilet. per 100,000 live births between 2005 and 2012. In information and communication technol- The country is performing above expectations ogy (ICT), the Republic of Congo is performing with regard to stunting, but malnutrition is still high. very well in terms of the mobile cellular network, Stunting, defined as low height for age and an indica- but performance is below expectations in terms of tor of chronic malnutrition, decreased from 31 percent Internet access. The country seems to have harvested in 2005 to 25 percent in 2011. As a result, the country the low hanging fruit as materialized by the strong per- is now performing as expected in comparison to peers. formance of access to mobile phone network. However, Nevertheless, the level of stunting remains quite high. the country is still struggling to reach the next level. High malnutrition reduces agricultural productivity, Access to Internet is very low. It is estimated that only contributes to poverty, and affects education and intel- 7 percent of the population were using the Internet in lectual potential of school children (for example, stunting causes children to start school late because they look too 4 The findings are not specific to the Republic of Congo; as demons- small for their age, and will also be a cause of absentee- trated by de la Brière et al. (2016), all resource-rich countries in SSA ism and repetition of school years). fare poorly on nonmonetary dimensions of well-being. 10 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) 2014. Beyond the low quality, prices remain far too high the unemployment rate is 32.7 percent for those ages 15 for the general public and could be the main reason for to 29. The corresponding figures for those ages 30–49 and the low usage of Internet. 50–64 are 15.6 and 8.3 percent, respectively. Therefore, The country is also performing below expecta- a program focusing on youth employment should be tions with regard to connectivity and road density. boosted and scaled up as much as possible, given the In 2014, the country barely had 5 km of roads per 100 importance of the youth in the overall population. km2 of land, which is way below expectations. A pos- Access to formal wage jobs is very limited for the sible explanation for this very low density of road could youth. They are more likely be employed on their own be the fact that the vast majority of the population is account or by a household as family helper. As we saw concentrated in the two main cities. Still, the low den- in the previous section, the youth are more likely to be sity of road will definitively translate into connectivity unemployed. Moreover, even when they manage to find a problems and result in economic inefficiencies. job, they are likely to be employed on their own account as is the case for the whole population. On the other hand, The current generation does not youth, ages 15–29, are more likely to be employed by a household (15 percent) or by small and medium enter- seem to be benefitting from the oil prises (SMEs) (12 percent). Thus, a policy toward support- windfall ing the emergence of strong SMEs could be an indirect way to insure the creation of quality jobs for the youth. During the period of strong economic growth, the The formal sector, either public or private, has economy created jobs, but this was offset by popula- failed to create quality jobs for the population. As tion growth. Between 2005 and 2011, the number of a consequence, the vast majority of the labor force is employed increased by 180,000. In the meantime, the employed on their own account. A bit more than three potentially active population (ages 15–64) increased by out of five workers (63 percent) work on their own 264,000. Population growth appears to be a challenge, account, running a business with no employees or being as not only does the country have to create jobs, but involved in subsistence agriculture, and another quarter there must be enough new jobs to at least match the of workers work for a household without pay. The public fast-growing working-age population. administration is the main provider of formal wage jobs. Unemployment seems to have increased between One out of ten workers (14 percent) is employed in the 2005 and 2011, which could represent a puzzle to public administration or by a parastatal firm. The role understand trends in well-being. It seems that the of the private sector in providing jobs to the popula- impressive economic growth was jobless, or at least did not tion is not negligible. Up to 13 percent of workers are have significant impact on unemployment, especially the employed either in large private firms (5 percent) or in youth. This is not surprising given that growth was driven SMEs (8 percent). by the oil sector which is not labor intensive and has limited linkages with the rest of the economy. Moreover, these findings also suggest that the massive infrastructure Education improves income and program under the ‘municipalization’ slogan also failed, welfare only at the secondary level at least between 2005 and 2011, to create substantial and beyond and enough jobs to have a consequential impact on the labor market. This is not surprising either as most of the The level of education matters a lot on the job market. construction work is just creating temporary jobs. Those with a high level of education are more likely to The youth are severely affected by unemploy- work as managers or skilled workers. On the other hand, ment. If we consider the second definition, for example, those with no education or primary education are more Executive Summary 11 likely to work as unskilled workers or laborers. Up to 80 Investment in human capital percent of those with tertiary education are employed as managers. There is also an important proportion Skills and good health are prerequisites for success on the (40 percent) of those with upper secondary education job market. Good health and education increase chances that work as either managers or skilled workers. Upper of finding a job and of being more productive, thus earning secondary education appears to be the threshold beyond higher returns. Our analysis suggests that in the Republic of which education actually matters for accessing quality jobs. Congo, education really did not make a big difference below Close to half of those with no education or primary educa- upper secondary. Thus, the fight against poverty should tion only work as unskilled workers or laborers. Still, there focus on dropouts and transition to secondary education. is a non-negligible proportion of those with no education Educating and equipping the population with the neces- who are either employed as skilled workers or managers. sary skills will be critical as the country moves ahead with Earnings increase with age (an implicit measure its goal of achieving economic diversification.5 Vocational of experience) and with education. Education affects training should also play a key role, especially in the short earnings only when one reaches the secondary level. The run, for those who already dropped out. gains appear only as of secondary education since there are few differences in earnings between those with less Boost agricultural productivity and than primary education and those who completed pri- commercialization mary school but did not pursue their education further. The better educated an individual is, the higher Increasing agricultural productivity will be critical the level of earnings of the individual. The effect of edu- for rural poverty alleviation. The rural population cation on earnings after controlling for type of employ- relies heavily on agriculture as a main income source. ment is substantially significant for those reaching upper The evidence suggests that the country depends on secondary and tertiary education. This has implications imports to satisfy food needs. Yet, availability of arable for the development of a national strategy for education. land and opportunities in the fishing and livestock areas The objective should be to provide education, including provide huge opportunities for the country to achieve vocational training, up to at least upper secondary level. the food sovereignty. In doing so, the rural population Reducing dropouts and improving transition at each could generate enough income to move out of poverty. sublevel will be critical to achieve such a goal. For increased agricultural productivity to be effective, it should be accompanied with other actions such as impro- For poverty reduction to happen ved connectivity to market, research and development for better inputs, and so on. in a sustainable way, the following actions should be considered Expand coverage of formal social to overcome some of the many safety nets programs challenges Through a program such as LISUNGI,6 the Government Going forward, one of the main development challen- could aim at providing cash transfers to the poor and ges for the country is to translate oil wealth into better services delivery, better human capital, and quality jobs 5 Commercial farming, agro processing, and services (ICT and for its population, and to support the poor and vulne- tourism, among others). rable so as to insure pro-poorness and inclusiveness of 6 LISUNGI means helps or assistance in Lingala. The LISUNGI program is a strategic pillar of the social protection system that is growth. The following five areas of actions can be derived design to support the country in improving delivery of services from from the analysis: multiple sectors. 12 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) vulnerable, including the autochthons. The financial also be important to take measures to facilitate private cost for a hypothetical cash transfer could be quite high, investments aimed at creating jobs outside of the oil about XAF 171.2 billion in 2011 prices, representing sector, including through efforts to improve electricity about 10 percent of the government budget. In 2011 generation and distribution and increase access to credit prices, the national poverty line was estimated at XAF for the private sector. 274,113 per year and per equivalent adult. On average, the distance of the poor from the national poverty line Provide better services to the is 15.4 percent. This indicates that if it was possible to population perfectly target the poor, it will take average annual payments of XAF 103,260 to eliminate poverty in the As illustrated, the Republic of Congo is performing Republic of Congo.7 Overall, the budget for such a cash below expectations on most of the nonmonetary transfer program is estimated at XAF 171.2 billion in dimensions of well-being. The country is underperfor- 2011 prices. This represents slightly less than 10 percent ming when compared to peers on access to education, of the government budget and about 14 percent of annual health, electricity, safe water, sanitation, ICT, and roads. oil revenues. Despite the high cost, such a program if Improving the availability and quality of service to the linked to some conditionality could represent investment population will help the country establish itself as a true in human capital that will result in important medium- middle-income country. and long-term benefits for the country. In the case of a conditional cash transfer, for example, the likelihood of a household using the windfall in building human capital will be higher. This will, if all other conditions are also in place,8 result in a better educated, better nourished, and healthier population. Support the expansion of the private 7 Of course this is just a hypothetical assumption, as there are many sector for job creation other factors at play, including education that takes time and the likelihood of some poor spending the allocation on useless items such as consumption of products associated with ill health (alcohol, The private sector is playing an important role in the tobacco, etc.). Republic of Congo and will continue to do so. It will 8 Such as quality of service delivery. Executive Summary 13 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) Executive Overview T his poverty assessment analyzes recent trends in monetary and nonmonetary aspects of poverty and economic vulnerability in Republic of Congo (ROC), based on two nation- ally representative and broadly comparable household expenditure surveys conducted by the NIS in 2005 and 2011. The study determines the drivers of poverty reduction by systemati- cally looking at demographic, labor, and human capital dimensions. The report also discusses cross-cutting issues relevant for poverty reduction, such as service delivery, marginalization of autochthons, and others. This study aims to provide policy makers with the knowledge needed to improve the effectiveness of their programs to reduce and finally eradicate extreme poverty in the Republic of Congo. History has been marked by political instability and dependency on the oil sector The country’s history has been characterized by ups and downs related to the performance of the oil sector, changes from communism to market economy, civil war, and social unrest. Since the 1960s, the country has experienced four economic phases: 1960–1972; 1973–1984; 1985–1999; and 2000–2014. These periods correspond, respectively, to the pre-oil economy; the first oil boom; the oil price crisis; and the second oil boom (World Bank 2016). The country also experienced social unrest with a military coup and civil war. The various governments, also in line with the cold war, have moved back and forth from communism toward a market economy. All these affected the economy and people’s perceptions, especially toward the importance of the state as the main job provider. Dutch disease followed the discovery of oil. In the 1960s, the economy was diversified, but with the discovery of oil followed the reduction of the relative importance of other sectors (Figure 1). Most importantly, the agriculture and the manufacturing sectors shrank due to reduced competitiveness. As a consequence, the country imports most of its goods, even food. The Republic of Congo experienced strong economic growth between 2002 and 2015. This strong economic growth was driven by higher oil prices and political stability. On average, during this period, the Republic of Congo’s growth rate stood at 4.5 percent (Figure 2). Between 2005 and 2011, the country grew strongly at an average annual rate of 5.4 percent mainly driven by high oil revenues from oil production and the decision of the Government to step up its invest- ment in infrastructure in 2006. 15 FIGURE 1: Sector Shares in GDP, 1960–2010 (%) 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0 1960 1962 1964 1966 1968 1970 1972 1974 1976 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 Services Extractive Industries Manufacturing Agriculture Forestry Source: World Bank Economic Data, 2014. FIGURE 2: Trend in Per Capita GDP FIGURE 3: Poverty and Per Capita GDP Trends 2500 60 $5,600 GDP per capita (constant 2005 US$) 50.7 50 $5,400 50.2 Poverty Rate (%) 2000 40.9 GDP per capita 40 $5,200 37.0 1500 30 $5,000 20 $4,800 1000 10 $4,600 0 $4,400 500 2005 2006 2007 2008 2009 2010 2011 Year 0 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 $1.9/day PPP National Poverty line GDP per capita Year Source: Authors’ calculation using WDI, the 2005 and 2011 Enquête Congolaise auprès des Ménages (ECOM) surveys. Source: WDI. a decrease of 9.8 percentage points which is in line Between 2005 and 2011, the strong with the GDP growth rate observed during that macroeconomic performance period. Overall, around 143,000 people moved out coincided with substantial poverty of poverty. Changes in the poverty gap and squared reduction poverty gap follow similar patterns to those observed for the poverty headcount. Nationally, despite popu- The strong macroeconomic performance translated lation growth between the two years, the number into a substantial reduction of the proportion of of poor decreased to 1,658,000 in 2011, down from the population living in poverty between 2005 1,801,000 in 2005. and 2011. The proportion of the population living Most of the reduction in poverty was observed below the national poverty line declined from 50.7 in the two largest cities of Brazzaville and Pointe percent in 2005 to 40.9 percent in 2011 (Figure 3), Noire. In Brazzaville, the poverty headcount went 16 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) FIGURE 4: National Poverty Rate by Region Poverty reduction was sustained 80 beyond 2011, but it will be difficult 70 for the country to reach the goals of 60 eradicating extreme poverty by 2030 50 40 Microsimulation results suggest that the poverty 30 20 decline has continued beyond 2011, although at a 10 slower pace. It is estimated that the slowdown of eco- 0 nomic growth linked to the oil sector led to a slower Brazzaville Pointe Other Semi- Rural ROC reduction of poverty after 2011. Between 2005 and Noire municipalities urban 2011, poverty declined by 1.63 percentage points 2005 2011 annually. In 2016, due to slower economic growth, it is Source: Authors’ calculation using WDI, the 2005 and 2011 Enquête estimated that the share of the population living below Congolaise auprès des Ménages (ECOM) surveys. the national poverty line was around 34 or 35 percent. This corresponds to a poverty decline by 1.52 percentage points annually between 2011 and 2016. Our projections up to 2030 show that, unless down by 20 percentage points from 42.3 in 2005 to performance and inequality improve substantially, it 21.6 in 2011. Pointe Noire also experienced a sharp will be difficult but not impossible for the country decrease in poverty (13 percentage points), down to to reach the goals of eradicating extreme poverty by 20.3 in 2011 from 33.5 in 2011 (Figure 4). By con- 2030. Under a very optimistic scenario where the coun- trast, in rural areas, there was an increase in poverty. try is assumed to achieve a 10 percent annual growth In rural areas, the poverty headcount went up by 4.6 rate between 2021 and 2030, the US$1.90 PPP per percentage points from 64.8 in 2005 to 69.4 in 2011. day poverty rate will be 3.6 percent in 2030. Under a The story that emerges from these estimates is that of a more realistic scenario based on projected growth by dual economy, with much lower and faster decreasing World Bank staff, we project that the US$1.90 PPP per poverty in urban areas, especially the two large cities, day poverty rate will still be high, around 15 percent than in rural areas. in 2030. Similar to the national poverty rate, the inter- national extreme poverty rates have declined sig- nificantly. National poverty measures are used for There were also important within-country poverty analysis and for deriving the improvements in nonmonetary economic policies for poverty eradication. However, the dimensions of well-being national poverty figures are generally not comparable across countries. To compare the Republic of Congo with Subjective poverty as well as poverty in terms of assets other countries in Africa and worldwide, the so-called ownership decreased substantially. There was a 10 international extreme poverty measures are generally percentage point decline in the share of those living in used. The most common international poverty line is households struggling to satisfy their food needs (a proxy US$1.90 expressed in 2011 PPP U.S. dollars. The share definition of subjective poverty). Ownership of modern of the extremely poor by international standards living assets increased while ownership of traditional assets below US$1.90 PPP a day had declined in the Republic deteriorated. There was a notable increase in the propor- of Congo from 50.2 percent in 2005 to 37.0 percent in tion of households who own a mobile phone, a TV set, a 2011 (Figure 3). modern chair, or an iron (Figure 5). As a consequence, the Executive Overview 17 FIGURE 5: Asset Ownership, 2005–2011 (%) Share of the Population in Poverty, TABLE 1:  Nonmonetary Dimensions, 2005–2011 (%) Mattress or bed Phone Subjective poverty Assets-based poverty Radio 2005 2011 2005 2011 Television Brazzaville 42.6 31.2 50.5 21.3 Modern chair Pointe Noire 29.6 23.0 43.1 20.2 Iron Other municipalities 41.0 35.0 71.6 52.7 Bicycle Semi-urban 54.3 34.7 82.4 59.4 Canoe Rural 50.2 37.6 80.3 69.1 Computer Total 42.9 32.2 62.5 40.7 Motorcycle Source: Authors’ calculation using the 2005 and 2011 ECOM surveys. Car or Truck 0 10 20 30 40 50 60 70 80 90 100 2005 2011 and status as a middle-income country. Cross-country Source: Authors’ calculation using the 2005 and 2011 ECOM surveys. regression estimates suggest that the level of international poverty in the Republic of Congo is still much higher than in other comparable middle-income countries magnitude of the reduction in the share of the population (Figure 6). Countries with a similar level of economic living in households considered assets-poor is even larger.9 development generally have much lower poverty rates. Importantly though, the reduction in assets-based poverty Moreover, the international poverty rate (US$1.90 – is again observed for all areas, including rural areas. 2011 PPP) in the Republic of Congo is quite close to the SSA average of 42.6 percent. The country is performing below expectations on monetary poverty 9 A factorial analysis is conducted for possible assets, and the poverty Despite the improvement in living standards, the line is set to be as close as possible to the 2011 monetary estimate. country is still underperforming given its potential Projecting this to 2005 allowed to explore the trend. FIGURE 6: Poverty Headcount (2011 PPP) Versus GNI Per Capita in a Cross-Country Setting 1. Poverty headcount at US$1.90 a day (2011 PPP) 2. Poverty headcount at US$3.10 a day (2011 PPP) 80 100 60 at $1.90 a day (2011 PPP) at $3.10 a day (2011 PPP) 80 Poverty headcount ratio Poverty headcount ratio ROC 2005 ROC 2005 40 ROC 2011 60 ROC 2011 20 40 0 20 –20 0 0 2000 4000 6000 8000 10000 0 2000 4000 6000 8000 10000 GNI per capita, PPP (constant 2011 international $) GNI per capita, PPP (constant 2011 international $) Source: Authors’ calculation using the WDI data and based on the Gable, Lofgren, and Osorio-Rodarte (2015) approach. 18 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) Inequality and poverty reduction (SCD) (World Bank 2016a), the focus for the coming were mainly driven by education, years is to consolidate LISUNGI, expand for national coverage, and use it as a vehicle to deliver a bundle of sector of occupation, and a interventions and specific productive, incentive-based, reduction in average household size and income-generating activities. Growth, not redistribution, drove poverty Household size, education level, and reduction, but economic growth only labor market dynamics contributed the benefited the urban population most for poverty reduction Growth in mean expenditures contributed most of the Migration accounted for 5.6 percent of the decline poverty reduction observed between 2005 and 2011. in poverty.10 As illustrated in Figure 8, the population The stories vary depending on the areas of residence. In shift across location accounted for 5.6 percent of the Brazzaville, growth and reduction of inequality contri- poverty reduction. Population shift is happening in buted equally to poverty reduction. In Pointe Noire, favor of Brazzaville. Between 2005 and 2011, the share most of the poverty reduction was driven by growth. In of those living in Brazzaville increased by 8 percentage semi-urban areas, growth contributed substantially to points while the share of those living elsewhere decreased. poverty reduction, but part of this was offset by increased Migration appears to be a channel through which poverty inequality. By contrast, in rural areas, growth in mean is shifting from rural to urban areas. Migration itself is consumption was negative, leading to an increase in not enough to improve living conditions. Other intrin- poverty (Figure 7). Inequalities also increased in rural sic factors, including skills and education, are comple- areas, leading to an increase in poverty as well. mentarily important for migration to be successful. Yet, The overall poor performance on inequality those moving from rural areas to urban areas often lack reflects difficulties for the poor and vulnerable to such endowments. As a consequence, migration did not access quality jobs and a limited use of fiscal policy account for a bigger share of poverty reduction. to directly support them. The only serious social safety There was a positive shift toward more produc- net program is the LISUNGI, which is still in the pilot tive sectors that contributed 11.2 percent of poverty phase. As argued in the Systematic Country Diagnostic reduction. Although the share of people living in a household headed by someone who is unemployed or mpact of Growth and Changes in FIGURE 7: I inactive remained stable (about 18 percent) between Inequality on Poverty 2005 and 2011, there was a positive shift toward more 10 productive sectors, especially in the two main cities 5 (Figure 8). It is important to note that structural trans- 0 formation seems to be happening in the wrong direc- –5 tion. The manufacturing sector is shrinking, probably –10 a consequence of Dutch disease. As a consequence, a –15 growing share of households is relying on agriculture –20 and informal services for livelihood. There was a positive –25 Brazzaville Pointe Other Semi- Rural National 10 The 2011 population structure is questionable. For instance, given Noire municipalities urban the solid performance, it is difficult to understand how the share of Growth component Redistribution component those living in Pointe Noire has reduced. Unfortunately, with no access to the sample frame or detailed census, it was not possible for Source: Authors’ estimations using the 2005 and 2011 ECOM surveys. the team to confirm or invalidate the current figures. Executive Overview 19 FIGURE 8: Sectoral Decompositions of Change in Poverty (%) 1. By region 2. By head’s education 2 0.0 1 0 –0.5 –1 –1.0 –2 –3 –1.5 –4 –5 –2.0 –6 –7 –2.5 Brazzaville Pointe Noire Other municipalities Semi-urban Rural Population- shift effect Interaction effect None Primary Secondary 1 Secondary 2 Tertiary Population- shift effect Interaction effect 3. By head sector of occupation 4. Household size 2 0.5 1 0.0 –0.5 0 –1.0 –1 –1.5 –2 –2.0 –2.5 –3 –3.0 –4 –3.5 –5 –4.0 Agriculture Mining/ Manufacturing Services Not Working Population- shift effect Interaction effect 1 Individual 2 to 3 Individuals 4 to 5 Individuals 6 to 7 Individuals 8 Individuals and more Population- shift effect Interaction effect Source: Authors’ estimations using the 2005 and 2011 ECOM surveys. shift toward more productive sectors in the main cities someone with no or primary education increased in rural and adverse shift in rural areas. In the main cities, labor areas from 46 to 53 percent. Thus, intervention toward shifted out of agriculture and manufacturing toward ser- curbing rural poverty should focus, among other things, vices. In rural areas, labor shifted out of manufacturing/ on skills/education beyond primary education. processing back into agriculture, which is less productive. Further analysis highlights that education Richer data sets and further investigations are needed to makes a significant difference only when one reaches understand this paradigm in rural areas. upper secondary or tertiary levels. In a world where Gains in education contributed 14.0 percent digital competencies are playing a growing role, educa- of the poverty reduction. This is a larger share than tion and skills allow households to improve their living for location and sector of occupation. Most of the standards by accessing more productive jobs and increas- improvement in skills happened in the two main cities. ing their productivity in self-employment activities. The In Brazzaville and Pointe Noire, the share of people liv- slight increase of the share of households headed by ing in households headed by someone with primary or someone with upper or tertiary education emerges as a no education declined, while the corresponding figures significant driver of consumption growth in the decom- for secondary and tertiary education increased. On the position analysis (Figure 9). The strong positive correla- contrary, the share of people in households headed by tion of tertiary education and consumption growth is 20 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) particularly important for poor households. However, The country has curbed monetary primary and lower education have been related to dete- poverty substantially, but many rioration of welfare. challenges remain Reduction of the size of households accounted for 38.2 percent of poverty reduction. Although fertil- Inequality remains high because of the ity remains high, the Republic of Congo has started its pattern of growth demographic transition. As a consequence, the house- hold size declined by close to one person from 5.12 in At the national level, inequality levels remain high. 2005 to 4.28 in 2011.11 However, the growing propor- There was a slight increase of the Gini coefficient, tion of children is maintaining the dependency ratio at although not statistically significant (0.460 in 2005 and higher levels, especially in rural areas. Still, results sug- 0.465 in 2011), suggesting a slight increase in inequa- gest that the shift in household size contributed a large lity. This increase is coherent with the fact that poverty percentage to poverty reduction. decreased more in the largest cities than in other urban Gender and disability have an indirect impact on and rural areas. Using an alternative measure, per equi- poverty as women and disabled persons exhibit lower valent adult consumption among the richest 10 percent levels of education and more limited employment of households in the Republic of Congo was 17.2 times opportunities. Gender and disability affect the ability of that of the poorest 10 percent in 2005; it had increased individuals to acquire the required skills, and thus exclude to 20.0 times by 2011. them from productive jobs. It was not possible to conduct By international standards, inequalities are a trend analysis by ethnicity as the relevant information higher in the Republic of Congo. Cross-country com- was not collected in 2005. Still, women, the disabled, and parisons suggest that inequality is high in the Republic autochthons should be considered as marginalized groups of Congo (Figure 10). The Republic of Congo is ranked as they have limited opportunities to access higher levels among the most unequal societies based on WDI data. of education and better jobs. Data on inequality in 105 countries are available in the WDI beyond 2010. The Republic of Congo ranked 90 out of 105 countries on the Gini. As reflected by higher inequalities, prosperity FIGURE 9: Educational Attainment Associated with was not fairly shared between 2005 and 2011. Another Consumption Growth Only for Upper way to look at growth and changes in the distribution Secondary and Tertiary Levels of welfare consists of using growth incidence curves which plot growth rates in consumption at various lev- 2.5% els of welfare (Ravallion and Chen 2003). As shown in 2.0% Figure 11, growth incidence curves suggest that growth 1.5% was not pro-poor nationally. The poorest actually expe- 1.0% rienced a deterioration in their living standards accord- 0.5% ing to consumption-based measures of poverty. Those 0.0% in the middle of the distribution and a small share of –0.5% the wealthiest households experienced the large posi- –1.0% tive growth. 10th quantile 25th quantile Median 75th quantile 90th quantile Primary Secondary 1 Secondary 2 Tertiary 11 These estimates are very close to the Demographic and Health Source: Authors’ estimations using the 2005 and 2011 ECOM surveys. Survey (DHS) findings. Executive Overview 21 A growing disparity between the two FIGURE 11: Growth Incidence Curves main cities and the rest of the country 5 Annual growth rate, % Poverty is becoming an increasingly rural phenome- 3 non, and this should be a much bigger reason for 1 concern. Most of the poverty reduction was observed –1 in the two largest cities of Brazzaville and Pointe Noire. –3 What was most worrisome was an increase in poverty –5 headcount as well as in the number of poor people in 1 11 21 31 41 51 61 71 81 91 rural areas. In rural areas, the poverty headcount went Expenditure percentile up by 4.6 percentage points from 64.8 in 2005 to 69.4 Source: Authors’ estimations using the 2005 and 2011 ECOM surveys. in 2011. In rural areas, the depth and severity of poverty has also increased, that is, the rural poor have become poorer. Between 2005 and 2011, the number of poor increased in rural areas to 951,000 up from 795,000. Beyond the urban-rural dichotomy, significant Consequently, the contribution of the rural areas to spatial differences in welfare exist in the Republic of poverty is on the rise (Figure 12). Congo. Figure 13 visualizes geographical disparities in Poverty pockets still exist in suburban slums. poverty headcount by department with maps. Among The Republic of Congo is one of the most urbanized the 12 ‘departments’ in the Republic of Congo, Pointe countries in the world with more than half of the popu- Noire and Brazzaville have by far the lowest poverty rate lation concentrated in two main cities. In the cities, the with a poverty headcount of 20.3 and 21.6, respectively. poor and vulnerable often are left with no choice but to Cuvette-Ouest is the poorest department, with 79.1 per- settle in slums, with insecurity and difficult living con- cent of population living below the poverty line, followed ditions, including lack of access to basic infrastructure by Lékoumou and Cuvette with 76.1 and 70.2 percent (electricity, piped water, transport, and so on). of poor individuals, respectively. The poverty rate in the FIGURE 10: Gini Inequality for the Republic of Congo and Selected Countries 70 60 50 40 30 20 10 0 Ukraine Norway Slovak Republic Kazakhstan Sweden Romania Kyrgyz Republic Netherlands Montenegro Albania Iraq Serbia Austria Cambodia Luxembourg Cyprus Bangladesh Ireland United Kingdom Nepal France Taiwan, China Canada Sierra Leone Italy Australia Indonesia Tunisia Portugal Iran, Islamic Rep. Tanzania Sri Lanka Vietnam Georgia United States Russian Federation El Salvador Congo, Dem. Rep. Uganda Philippines Benin Djibouti Togo Ecuador Congo, Rep. Costa Rica Chile Rwanda Guatemala Honduras Lesotho Haiti Source: Authors’ estimations using the 2005 and 2011 ECOM surveys. 22 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) FIGURE 12: Contributions to the Poverty Headcount by Location (%) 2005 2011 Brazzaville, Rural, 44% Brazzaville, Rural, 57% 20% 24% Pointe Noire, 10% Pointe Noire, 15% Other municipalities, 7% Other municipalities, Semi-urban, 9% 7% Semi-urban, 6% Source: Authors’ calculation using the 2005 and 2011 ECOM surveys. province of Kouilou is 56.9. The poverty rate is quite Increased vulnerability with many at risk high for the remaining departments as well—the pov- of falling back into poverty erty rate ranges between 62 and 69 percent for Plateaux, Likouala, Bouenza, Sangha, Pool, and Niari. Given the Despite moving out of poverty, many households are high rate in most of the departments, we can rank the living just above the poverty line and remain vulne- departments into two groups: the two main cities and rable to falling below the poverty line. A way to look the rest of the country. at vulnerability is to classify the population into three FIGURE 13: Poverty Maps by Department, 2011 1. Consumption-based headcount 2. Map of the Republic of Congo (75.0,80.0] (70.0,75.0] Likouala (65.0,70.0] (60.0,65.0] (55.0,60.0] (50.0,55.0] (45.0,50.0] (40.0,45.0] Sangha (35.0,40.0] (30.0,35.0] Cuvette-Ouest (25.0,30.0] (20.0,25.0] (15.0,20.0] (10.0,15.0] Cuvette (5.0,10.0] [0.0,5.0] Plateaux Niari Lékoumou Pool Kouilou Brazzaville Pointe Noire Bouenza Source: Authors’ calculation using the 2011 ECOM survey. Executive Overview 23 groups: the poor (consumption below the poverty line), group in the Republic of Congo. Monetary poverty the insecure nonpoor or vulnerable (consumption above headcount for autochthons is more than twice the the poverty line but below twice the poverty line), and poverty rate of the remaining population. Close to nine the middle class (consumption above twice the poverty out of ten autochthons are poor. This group appears line; this would include some wealthy households as well, to be extremely marginalized. Their marginalization is but most households could be considered middle class characterized by very limited access to social services, for a country like the Republic of Congo). Figure 14 including health and education, as well as the labor shows that as poverty regressed, both the vulnerable market. Thus, they contribute and benefit very little (insecure nonpoor) and middle class groups expanded, from economic activities. The autochthons who are with the middle class group expanding faster than the not poor tend to be close to the poverty line, and thus vulnerable/insecure nonpoor group. While only 20.6 ranked as insecure nonpoor. There is a clear difference percent of the population had consumption more than between autochthons living in rural areas and autoch- twice the poverty line in 2005, this increased to 26.3 thons living in urban areas. All autochthons who percent by 2011. In the meantime, 32.8 percent of the manage to move from rural areas to urban areas are population, while technically ‘nonpoor’, is consuming ranked as middle class. at a level below twice the poverty line, meaning less that XAF 550,000 yearly (in nominal 2011 CFA francs) per The economy is not creating equivalent adult. Given that a large share of the vulne- enough jobs, especially for the youth rable rely on either agriculture or informal activities which are susceptible to significant output volatility, Issues related to job creation are at the center of the many remain at serious risk of falling back into poverty, development debate in the Republic of Congo. For at least on a temporary basis. most households in poverty, inability to meet their basic needs reflects difficulties faced by its members in Marginalization of autochthons accessing quality jobs. The close relationship between poverty and work is very clear to the population. Lack The autochthons, who represent about 1 percent of of employment is the most cited factor leading to the population, stand out as the main marginalized poverty in household responses to a question in the FIGURE 14: Poverty, Insecure Nonpoor, and Middle Class Status by Year and Location 1. By year 2. By residence area, 2011 only 100% 100% 20.6 26.3 80% 80% 28.7 60% 32.8 60% 40% 40% 50.7 20% 40.9 20% 0% 0% 2005 2011 Brazzaville Pointe Other Semi- Rural National Noire municipalities urban 2011 Poor Insecure nonpoor Middle class Source: Authors’ calculation using the 2005 and 2011 ECOM surveys. 24 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) 2011 ECOM survey about the causes of poverty. Lack Distribution of Workers by Type of FIGURE 15:  of employment was cited as a leading cause of poverty Employer and Education Level by 91.7 percent of respondents and insufficient income, 100% a closely related effect, was cited by 59.0 percent. 90% 80% During the strong economic growth, the 70% economy created jobs, but this was offset by popula- 60% tion growth. Between 2005 and 2011, the number of 50% employed increases by 180,000. In the meantime, the 40% 30% potentially active population (ages 15–64) increased by 20% 264,000. Population growth appears to be a challenge, 10% as not only does the country have to create jobs, but 0% None Primary Secondary 1 Secondary 2 Tertiary there must be enough to at least match the fast-growing Education working-age population. Public administration Public firm/parastatal The youth are severely affected by unemploy- Small and medium enterprise (SME ) Large private company ment. In 2011, the unemployment rate was 32.7 percent Association, Cooperative, Church, NGOs Household International Organization, Embassy Own account for those ages 15 to 29. The corresponding figures for those ages 30–49 and 50–64 were 15.6 and 8.3 percent, Source: Authors’ calculation using the 2011 ECOM survey. respectively. Therefore, a program focusing on youth employment should be boosted and scaled up as much as possible, given the importance of the youth in the same activity, women are paid at the same rate as their overall population. male counterparts. The formal sector, either public or private, has Education is positively correlated with earnings. failed to create quality jobs for the population. As The better educated an individual is, the higher the level a consequence, the vast majority of the labor force is of earnings of the individual. The benefit of education employed on their own account in agriculture or infor- for earnings are very significant for those reaching upper mal services. A bit more than three out of five workers secondary and tertiary education (Figure 16). This has (63 percent) work on their own account, running a implications for the development of a national strat- business with no employees or being involved in subsis- egy for education. The objective should be to provide tence agriculture, and another quarter of workers work education, including vocational training, up to at least for a household without pay (Figure 15). The public upper secondary level. Reducing dropouts and improv- administration is the main provider of formal wage ing transition at each sublevel will be critical to achieve jobs. Fourteen percent of workers are employed in the such a goal. public administration or by a parastatal firm. The role of the private sector in providing jobs to the popula- tion is not negligible. Up to 13 percent of workers are The Republic of Congo is employed either in large private firms (5 percent) or in performing below expectations on SMEs (8 percent). service delivery There is no gender-based pay gap. Analysis sug- gests that possible differences observed between women Education and men are due to other factors, including differences in education/skills, but more importantly, difference in the Performance is way below expectations on most of type of activity. Women are more likely to be employed the critical education sector indicators. For example, on their own account. With equal competencies in the the Republic of Congo is performing below expectation Executive Overview 25 FIGURE 16: Employment Rate and Earnings Profile Despite improvement in enrollment, the quality of by Education Level education is of concern and seems to be decreasing. Test 80 200,000 results suggest that a high proportion of students leaving 70 180,000 primary school do not have sufficient foundational skills. 160,000 Mothnly wage (FCFA) Share employed (%) 60 End-of-cycle tests in primary education have found that 140,000 50 120,000 two-thirds of students leaving primary school do not have 40 100,000 30 80,000 sufficient foundational skills in literacy and numeracy. Thus 20 60,000 more and more children are going to school, but it does not 40,000 10 20,000 guarantee that they are acquiring the expected knowledge. 0 0 This assessment confirms the importance of the ongoing None Primary Secondary 1 Secondary 2 Tertiary debate on the quality of service delivery, particularly educa- tion services in the country (World Bank 2015).12 Education The private sector is playing a very important Employed Mean and growing role in the provision of education ser- Median vices. Figure 18 provides data on the type of school Source: Authors’ calculation using the 2011 ECOM survey. attended. The share of students in private schools seems to have increased between the last two surveys for 2005 and 2011. In 2011, 35 percent of students were enrolled with regard to primary school completion (Figure 17). The in a private school. This is up by 15 percentage points primary school completion rate was 74 percent in 2012. compared to 2005. Such high shares of private provi- The primary school completion rate has been fluctuating sion in a country where 40 percent of the population is over the last decade. Overall, between 2005 and 2012, the poor is an indication of inadequate success in ensuring primary school completion rate increased for girls, but effective access by the Government. Is the deterioration went down for boys. In 2012, primary school completion for girls was 79 percent, up from 70 percent in 2005. 12 The World Bank Education Global Practice is currently conducting During the same period, the primary school completion a Service Delivery Indicators (SDI) survey that will provide more rate for boys went down from 75 percent to 70 percent. evidence of this. FIGURE 17: Performance is Below Expectation in Primary and Secondary School Completion 1. Primary completion rate 2. Lower secondary completion rate 120 150 Lower secondary completion rate, total (% of relevamt age group) Primary completion rate, total (% of relevamt age group) 100 100 80 ROC 2005 ROC 2012 50 ROC 2012 60 ROC 2004 ROC 1996 ROC 1996 40 0 0 2000 4000 6000 8000 10000 0 2000 4000 6000 8000 10000 GNI per capita, PPP (constant 2011 international $) GNI per capita, PPP (constant 2011 international $) Source: Authors’ calculation using the WDI data and based on the Gable, Lofgren, and Osorio-Rodarte (2015) approach. 26 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) FIGURE 18: Market Share by Type of School FIGURE 19: Reasons for Lack of Satisfaction 100% Lack of 16.4 21.6 20.6 books/supplies 20.3 25.7 90% 31.3 32.9 34.9 36.5 0.1 80% 49.4 3.1 Overcrowding 1.4 2.4 70% 0.4 1.3 3.1 2.9 60% Lack of 3.4 teachers 50% 0.2 80.9 76.2 74.4 Teachers often 73.1 71.4 40% 66.6 63.6 61.1 absent 58.9 30% 48.9 20% Facilities in bad condition 10% 0% Poor teaching Primary Secondary 1 Secondary 2 Tertiary All Primary Secondary 1 Secondary 2 Tertiary All Other problem 0 5 10 15 20 25 30 35 40 45 50 2005 2011 2005 2011 Government Religious Organization Private Source: Authors’ calculation using the 2005 and 2011 ECOM surveys. Community Other Source: Authors’ calculation using the 2005 and 2011 ECOM surveys. and maternal mortality are often used as a measure of the efficiency of the health sector in a given country. Between of the quality of education a cause or a consequence 2005 and 2012, under-five mortality, which measures of the growing role of the private sector? Further data the probability of children dying between birth and the and analysis are needed to answer such an important fifth birthday, dropped from 95.3 to 52.6 per 1,000 question. births. Thanks to this improvement, the country is now Satisfaction is low for public schools. Lack of performing as expected given its GNI level (Figure 20). books/supplies and overcrowding are the main reasons The story is a bit different regarding maternal mortality. for dissatisfaction. Satisfaction with schools, as measured Despite improvement, the country is still performing by the share of parents who do not have any complaints with schools, is low, especially in public schools. Less than 20 percent are satisfied with public schools, against FIGURE 20: Mortality Rate, under-Five close to 60 percent for private schools. The main reasons (per 1,000 Live Births) for dissatisfaction are lack of books/supplies, overcrowd- ing, and lack of teachers (Figure 19). The lack of books/ 150 supplies and the lack of teachers are predominant for Mortality rate, under-5 (per 1,000 live births) ROC 1996 children from lower socioeconomic backgrounds, while 100 ROC 2005 overcrowding is cited more often by parents of children from higher levels of well-being. 50 ROC 2014 Health 0 There has been substantial improvement in child and 0 2000 4000 6000 8000 10000 maternal mortality. However, the country still performs GNI per capita, PPP (constant 2011 international $) below expectations with regard to maternal mortality and Source: Authors’ calculation using the WDI data and based on the Gable, has not reached any of the health-related MDGs. Child Lofgren, and Osorio-Rodarte (2015) approach. Executive Overview 27 FIGURE 21: Maternal Mortality Ratio (national FIGURE 22: Type of Facility Visited Estimate, Per 100,000 Live Births) 100% 13.8 11.3 9.1 6.0 4.8 8.8 90% national estimate (per 100,000 live births) 1,500 80% 30.3 35.7 40.1 45.1 44.9 39.5 70% Maternal mortality ratio, 60% 4.2 3.1 1,000 50% 2.0 1.8 1.3 2.4 ROC 2005 40% 30% 500 ROC 2012 20% 51.6 49.9 48.7 47.1 49.0 49.2 10% 0% 0 Q1 Q2 Q3 Q4 Q5 Total 0 2000 4000 6000 8000 10000 Public FBO Private Other GNI per capita, PPP (constant 2011 international $) Source: Authors’ calculation using the 2011 ECOM survey Source: Authors’ calculation using the WDI data and based on the Gable, Lofgren, and Osorio-Rodarte (2015) approach. The country is performing above expectations with regard to stunting, but malnutrition is still high. lower than its peers on maternal mortality. Maternal Stunting, defined as low height for age and an indica- mortality rate declined from 781 to 426 deaths per tor of chronic malnutrition, decreased from 31 percent 100,000 live births between 2005 and 2012 (Figure 21). in 2005 to 25 percent in 2011. As a result, the country The distribution of private expenditure on is now performing as expected in comparison to peers. health shows that the poor will be more affected by a Nevertheless, the level of stunting remains quite high. fiscal reform concerning drugs, hospitalization, and High malnutrition reduces agricultural productivity, consultation fees. In August 2011, the Government contributes to poverty, and affects education and intel- decided that malaria drugs, anti-retroviral (ARV), and lectual potential of school children (for example, stunting C-section will be free of charge in public facilities. The causes children to start school late because they look too analysis here shows that these measures were actually pro- small for their age, and will also be a cause of absenteeism poor, but new data and additional analytical work are and repetition of school years). As shown in Figure 23, needed to access the effectiveness and the impact of such in the case of the Republic of Congo, malnutrition a measure. Since 2014, financing of inputs for these free increases with poverty. Affordability of food might be health care programs has stagnated and many services the main cause of malnutrition. In principle, this makes are either not provided (free C-section or free health it easier to design a program to fight malnutrition, as it care for pregnant women), not free, or provided against will mainly imply making sure the poor access food, for out-of-pocket payments (in the case of malaria drugs). example, through school canteens. The private sector is playing a very important and growing role in the provision of health services. Electricity Slightly more than half of the population seeks care in private facilities. Only six out of ten individuals who were Despite improvements over the last decade, access sick did seek care. A bit more than half of sick people to electricity is very low compared to expectations. seek care in nongovernment facilities (Figure 22). Such Given that the Republic of Congo is a country with high shares of private provision in a country with high substantial hydro potential, electricity could in principle poverty rate is an indication of inadequate success in be generated and distributed at a relatively low cost to ensuring effective access by the Government. a large share of the population. Coverage rates have 28 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) FIGURE 23: Stunting, Height for Age, the Republic FIGURE 25: Reason for Not Subscribing to Electricity of Congo vs. Peers 100% 60 90% Prevalence of stunting, height for age 80% 70% (% of children under 5) 40 60% 50% ROC 2005 40% ROC 2011 30% 20 20% 10% 0% 0 Q1 Q2 Q3 Q4 Q5 0 2000 4000 6000 8000 10000 Welfare quintile Total GNI per capita, PPP (constant 2011 international $) Too expensive to subscribe Consumption too expensive Source: Authors’ calculation using the 2011 ECOM and the 2012 DHS; and Complicated process Remote network (not accessile) using the WDI data and based on the Gable, Lofgren, and Osorio-Rodarte Useless Network not avaliable in the area (2015) approach.. Source: Authors’ estimates based on the 2011 ECOM survey. increased substantially, from 26.7 percent in 2005 to 42.5 percent in 2011. Unfortunately, connection rates Frequent shortages are by far the main reason in the country remain below expectations compared to for dissatisfaction. Poor quality of electricity and peers (Figure 24). high cost are also important reasons for dissatisfac- Those who are not connected to the network tion. Satisfaction with electricity service is very low. quote issues related to affordability and supply Slightly less than one out of three households is satis- as main reasons for not subscribing. Poor people fied with the electricity network service. Seven out of often don’t have electricity in their neighborhood, ten households point to shortage as reason for dissatis- and when the network is present, they cannot afford faction (Figure 26). The second reason for dissatisfac- it (Figure 25). tion is issues related to quality (quote by 13 percent of FIGURE 24: Access to Electricity (% of Population) 1. Urban areas 2. Rural areas 120 100 Access to electricity, urban Access to electricity, rural 100 80 (% of urban population) (% of rural population) 80 60 60 ROC 2012 40 40 ROC 2000 20 ROC 2012 20 ROC 2000 0 0 2000 4000 6000 8000 10000 0 2000 4000 6000 8000 10000 GNI per capita, PPP (constant 2011 international $) GNI per capita, PPP (constant 2011 international $) Source: Authors’ calculation using the WDI data and based on the Gable, Lofgren, and Osorio-Rodarte (2015) approach. Executive Overview 29 FIGURE 26: Satisfaction with the Electricity Service FIGURE 27: Improved Water Source (% of Population with Access) 100% 90% 100 80% (% of population with access) 70% Improved water source 60% 50% 80 ROC 2014 40% ROC 2005 ROC 1997 30% 20% 60 10% 0% Q1 Q2 Q3 Q4 Q5 40 Welfare quintile Total 0 2000 4000 6000 8000 10000 GNI per capita, PPP (constant 2011 international $) Too expensive Frequent shortages Poor quality Service interruption/landslide Not concerned Source: Authors’ calculation using the WDI data and based on the Gable, Lofgren, and Osorio-Rodarte (2015) approach. Source: Authors’ estimates based on the 2011 ECOM survey. households). The cost of electricity is the third reason differences between those using electricity and those pay- (11.5 percent). Cost issues are more predominant in ing for it are big. In 2011, 42.5 percent of households poor households. For the bottom quintile, cost of elec- were connected to the grid, but only 30 percent were pay- tricity is quoted by three out of ten households. ing for electricity, generating a 12.5 percentage point gap Electricity tariffs are poorly designed in the between coverage and payment. The Société Nationale de Republic of Congo. The implicit consumption subsi- Distribution d’Eau (SNDE) is facing a similar challenge. dies are poorly targeted. Connection subsidies clearly have the potential to be better targeted. A robust mea- Water and sanitation sure of the targeting performance of subsidies is the share of the subsidy benefits received by the poor divided by Although the share of the population with access the proportion of the population in poverty (Ω).13 The to improved water slightly increased during the last tariffs design is pro-poor if Ω>1. For the Republic of decade, it is still far below what is expected. The Congo, the value was 0.62 in 2011, that is, implicit Republic of Congo is performing below expectations electricity subsidies are not pro-poor in the Republic with regard to access to improved water. Between 2005 of Congo. To our knowledge, electricity tariffs have not and 2015, access to improved water increased slightly been revised for a very long time. Pro-poorness of tariffs from 72 percent to 77 percent. However, the country design should be considered in the future. is still performing below expectations (Figure 27). Evidence suggests that Société Nationale Poverty is correlated with the lack of access to safe water d’Electricité (SNE) is struggling to effectively col- (Figure 29). lect payments from residential customers, including Access to improved sanitation remains very through the installation of new meters in some areas. low and as a consequence, the country is perform- The share of households paying for their electricity is ing below expectations on this dimension as well. systematically lower than the share of households who As illustrated in Figure 28, the Republic of Congo is declare using electricity. This may be an indication of illicit connections, but it may also reflect late payment, lack of recovery of late payment, or other issues. The 13 See Angel-Urdinola and Wodon (2007) for more details. 30 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) FIGURE 28: Improved Sanitation Facilities (% of Correlation between Access to FIGURE 29:  Population with Access) Improved Water and Poverty 50 100 45 Cuvette-Ouest (% of population with access) Improved sanitation facilities 40 Monetary poverty (FGT1) 80 35 Lékoumou Cuvette 30 Likouala Plateaux Sangha 60 25 Pool Niari Bouenza 40 ROC 2012 20 Kouilou 15 20 10 y = 123.84e–0.03x 5 Brazzaville R² = 0.7462 Pointe-Noire 0 0 30 50 70 90 110 0 2000 4000 6000 8000 10000 Access to Improved Water Source (%) GNI per capita, PPP (constant 2011 international $) Source: Authors’ estimates based on the 2011 ECOM survey. Source: Authors’ calculation using the WDI data and based on the Gable, Lofgren, and Osorio-Rodarte (2015) approach. still struggling to reach the next level. Access to Internet is very low and way below expectations (Figure 32). It performing below expectations with regard to access to is estimated that only 7 percent of the population were a safe toilet. In 2014, only 43 percent of the population using the Internet in 2014. Beyond the low quality, prices had access to improved sanitation. The situation is even remain far too high for the general public and could be the worse in rural areas where only 13 percent of the popula- main reason for the low usage of Internet. At US$1,200, tion have access to an improved toilet. More of concern the Republic of Congo was ranked as one of the most is the share of the population with no toilet at all. The expensive countries in the world with regard to the price situation is particularly of concern in four departments: of international bandwidth (Mbps/month) (Jacquelot, Plateaux, Lékoumou, Cuvette, Cuvette-Ouest. The share quote by World Bank 2016a). of households with no toilet ranges from 42.8 to 30 per- cent in these departments. Here again, the gap between the two main cities and the rest of the country is very pronounced (Figure 30). Correlation between Access to FIGURE 30:  Information and communication Improved Sanitation and Poverty technology 50 45 y = 10.522e0.0335x R² = 0.51722 Cuvette-Ouest The Republic of Congo is performing very well with 40 Monetary poverty (FGT1) regard to the mobile cellular network, but performance 35 Cuvette Likouala 30 Niari Lékoumou is below expectations with regard to Internet access. Sangha Plateaux 25 Bouenza Pool ICT provides new opportunities to boost productivity and 20 Kouilou save costs. Moreover, applications in various areas such as 15 10 education, agriculture, health, and so on have proven to Brazzaville 5 Pointe-Noire be critical for economic development and well-being. For 0 now, the country seems to have harvested the low hanging 0 10 20 30 40 50 fruit as materialized by the strong performance of access Share of households with no toilet (%) to mobile phone network (Figure 31). The country is Source: Authors’ estimates based on the 2011 ECOM survey. Executive Overview 31 FIGURE 31: Population Covered by Mobile Cellular Investment in human capital Network (%) Skills and good health are prerequisites for success 100 on the job market. Good health and education increase ROC 2011 mobile cellular network (%) chances of finding a job and of being more productive, Population covered by 80 thus earning higher returns. Our analysis suggests that in the Republic of Congo, education really did not make 60 a big difference below upper secondary. Thus, the fight 40 ROC 2005 against poverty should focus on dropouts and transi- tion to secondary education. Educating and equipping 20 the population with the necessary skills will be critical 0 2000 4000 6000 8000 10000 as the country moves ahead with its goal of achieving GNI per capita, PPP (constant 2011 international $) economic diversification.14 Vocational training should Source: Authors’ calculation using the WDI data and based on the Gable, also play a key role, especially in the short run, for those Lofgren, and Osorio-Rodarte (2015) approach. who already drop out. FIGURE 32: Internet Users (per 100 People) Boost agricultural productivity and commercialization 100 Internet users (per 100 people) Increasing agricultural productivity will be critical 80 for rural poverty alleviation. The rural population relies heavily on agriculture as a main income source. 60 The evidence suggests that the country depends on imports to satisfy food needs. Yet, availability of arable 40 land and opportunities in the fishing and livestock areas ROC 2014 ROC 2005 provide huge opportunities for the country to achieve 20 the food sovereignty. In doing so, the rural population 0 2000 4000 6000 8000 10000 GNI per capita, PPP (constant 2011 international $) could generate enough income to move out of poverty. For increased agricultural productivity to be effective, it Source: Authors’ calculation using the WDI data and based on the Gable, Lofgren, and Osorio-Rodarte (2015) approach. should be accompanied with other actions such as impro- ved connectivity to market, research and development for better inputs, and so on. Looking forward: how to overcome Expand coverage of formal social some of the many challenges safety nets programs Going forward, one of the main development challen- Through a program such as LISUNGI, the Government ges for the country is to translate oil wealth into better could aim at providing cash transfers to the poor and services delivery, better human capital, and quality jobs vulnerable, including the autochthons. The financial for its population, and to support the poor and vulne- cost for a hypothetical cash transfer could be quite high, rable so as to insure pro-poorness and inclusiveness of growth. The following five areas of actions can be derived 14 Commercial farming, agro processing, and services (ICT and from the analysis: tourism, among others). 32 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) about XAF 171.2 billion in 2011 prices, representing also be important to take measures to facilitate private about 10 percent of the government budget. In 2011 investments aimed at creating jobs outside of the oil prices, the national poverty line was estimated at XAF sector, including through efforts to improve electricity 274,113 per year and per equivalent adult. On average, generation and distribution and increase access to credit the distance of the poor from the national poverty line for the private sector. is 15.4 percent. This indicates that if it was possible to perfectly target the poor, it will take average annual Provide better services to the payments of XAF 103,260 to eliminate poverty in the population Republic of Congo.15 Overall, the budget for such a cash transfer program is estimated at XAF 171.2 bil- As illustrated, the Republic of Congo is performing lion in 2011 prices. This represents slightly less than below expectations on most of the nonmonetary 10 percent of the government budget and about 14 dimensions of well-being. The country is underperfor- percent of annual oil revenues. Despite the high cost, ming when compared to peers on access to education, such a program if linked to some conditionality could health, electricity, safe water, sanitation, ICT, and roads. represent investment in human capital that will result Improving the availability and quality of service to the in important medium- and long-term benefits for the population will help the country establish itself as a true country. In the case of a conditional cash transfer, for middle-income country. example, the likelihood of a household using the windfall in building human capital will be higher. This will, if all other conditions are also in place,16 result in a better educated, better nourished, and healthier population. Support the expansion of the private sector for job creation 15 Of course this is just a hypothetical assumption, as there are many other factors at play, including education that takes time and the likelihood of some poor spending the allocation on useless items such The private sector is playing an important role in the as consumption of products associated with ill health. Republic of Congo and will continue to do so. It will 16 Such as quality of service delivery. Executive Overview 33 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) References Angel-Urdinola, D., and Q. Wodon. 2007. “Do Utility Subsidies Reach the Poor? Framework and Evidence for Cape Verde, Sao Tome, and Rwanda.” Economics Bulletin 9 (4): 1–7. de la Brière, Bénédicte, Filmer Deon, Ringold Dena, Rohner Dominic, Samuda Karelle, and Denisova Anastasiya. 2016. “From Mines and Wells to Well Built Minds: Turning Sub-Saharan Africa’s Natural Resource Wealth into Human Capital.” Washington, DC: The World Bank. Gable, Susanna, Hans Lofgren, and Israel Osorio-Rodarte. 2015. Trajectories for Sustainable Development Goals: Framework and Country Applications. Washington, DC: The World Bank. Ravallion, M., and S. Chen. 2003. “Measuring Pro-Poor Growth.” Economic Letters 78 (1): 93–99. République du Congo. 2012a. “Plan National de Développement PND 2012-2016, livre 1 (Document de Stratégie pour la croissance, l’Emploie et la réduction de la pauvreté).” ——— . 2012b. “Plan National de Développement PND 2012-2016, livre 2 (Condensé des PAS/ CDMT ministériels.” World Bank. 2014. “Enhancing Efficiency in Education and Health Public Spending for Improved Quality Service Delivery for All.” A Public Expenditure Review of the Education and Health Sectors. Washington, DC: The World Bank ——— . 2016. “Systematic Country Diagnostic: Ending Extreme Poverty and Boosting Share Prosperity on Road to the Emerging Status”, Washington, DC: The World Bank. 35 Republic of Congo – Education, Jobs and Social Protection for a Sustainable Reduction of Poverty (Overview) 1818 H Street, NW Washington, DC 20433 USA Telephone: +1 202 473 1000 Internet: www.worldbank.org