CONFERENCE EDITION. Poverty Reduction in Ghana PROGRESS AND CHALLENGES. Vasco Molini and Pierella Paci

Transcription

1 CONFERENCE EDITION Poverty Reduction in Ghana PROGRESS AND CHALLENGES Vasco Molini and Pierella Paci

2

3 Conference Edition Poverty Reduction in Ghana Progress and Challenges Vasco Molini Pierella Paci

4 215 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington, DC 2433 Telephone: ; Internet: Some rights reserved This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, 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. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3. IGO license (CC BY 3. IGO) creativecommons.org/licenses/by/3./igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution Please cite the work as follows: Molini, Vasco and Pierella Paci Poverty Reduction in Ghana: Progress and Challenges. Systematic Country Diagnostic. Conference Edition. World Bank, Washington, DC. License: Creative Commons Attribution CC BY 3. IGO Translations If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to the Publishing and Knowledge Division, The World Bank, 1818 H Street NW, Washington, DC 2433, USA; fax: ; pubrights@worldbank.org. Cover photo: Jonathan Ernst/World Bank. Permission required for reuse. Cover design: Bill Pragluski/Critical Stages

5 Contents Acknowledgments... vii Executive Summary.... ix Introduction....1 Framing Ghana s Success Data Definitions and Technical Notes... 3 Notes A Tale of Success: Shared Prosperity, Poverty Reduction, and a Boost in Opportunity... 5 Two Decades of Expanding Prosperity for All Shared Prosperity Led to a Reduction in Poverty and Vulnerability The Poor over Time: Better Attributes and Living Standards... 9 Better and More Equal Opportunities for Sustainable Progress Notes Drivers of Poverty Reduction Key Economic Developments The Main Drivers of Poverty Reduction and Consumption Growth Notes Remaining Challenges Growing Inequality and Polarization in Household Consumption Persistent Spatial Disparities The Deteriorating Macroeconomic Environment Notes A Roadmap for Policy Action A Tale of Success A Less Positive Outlook A Roadmap for Policy Action Appendix A Computing Poverty Appendix B Regression Tables References Boxes 1.1 How has the typical poor household changed between 1991 and 212? Poverty Maps in Ghana iii

6 Figures 1 Real GDP Growth, ix 2 Trends in Poverty and Extreme Poverty, x 3 Distribution of Employment, by Sector and Consumption Quintile xi 4 Workforce Educational Attainment, by Sector of Employment, xi 5 Decomposition of Poverty Changes.... xii 6 Poor Individuals by Regions and Urban Rural Areas: xiii 1.1 Real GDP Growth, Growth Incidence Curves, and Trends in Poverty and Extreme Poverty, Growth, Inequality, and Poverty Decomposition, Growth Elasticity of Poverty, Vulnerability as Distance from the Poverty Line, Nationwide, Vulnerability and Poverty Depth as Distance from the Poverty Line, Rural and Urban Areas, Trends in Regional Poverty Headcounts, Main Sectors of Employment, Household Heads, by Quintile, Access to Basic Services Has Improved, Asset Index, Rural and Urban Areas and Nationwide, Sectoral Composition of GDP, (%) Poverty and Access to Employment, Adults Aged 15 64, Production Indexes, Major Crops Output per Worker, by Sector, Distribution of Employment, by Sector, Region, and Poverty Status Labor Force with at Least Basic Education, by Region, 1991 and Education and Employment Are Strong Correlates, 1991 and Variation in Population, by Rural and Urban Area and Region, Household Heads Who Were Not Born in Their Current Places of Residence Decomposition of Poverty Changes, Education Coefficients, by Quintile and Year Employment Category Coefficients, by Quintile and Year Decomposition of Household Consumption Decomposition of Changes in Characteristics Decomposition of the Gini Index, by Household Characteristic, Gini Indexes in Sub Saharan Africa Relative Consumption Distribution, Median-Adjusted Relative Consumption Distribution, Median-Adjusted Relative Consumption Distribution Series, Ghana, Regional Price Indexes: Total and Nonfood, A.1 Distribution of Highest Education Level of Household Head, by Quintile and Year A.2 Distribution of Type of Employment, by Quintile and Year A.3 Distribution of Household Characteristics, by Quintile and Year iv Contents

7 Maps 3.1 Poverty Maps, 2 and Additional Indicators of Welfare, 211 and A.1 Administrative Map of Ghana...53 Tables 1 Infant and Under-5 Mortality, Vaccination, and Fertility Rates, x 2 Interquartile Consumption Ratios, by GLSS Wave, xii I.1 Ghana Living Standards Surveys 1 6, Mean and Median Real Consumption (Adult Equivalent), (C ) Poverty Rates, the Poverty Gap, and the Severity of Poverty, The Profiles of Poor and Non-poor Households, Educational Attainment, Bottom 2 and Top 2, by Gender, Infant and Under-5 Mortality, Vaccination, and Fertility Rates, GDP Growth Decomposition, by Per Capita Value Added, (%) Employment, by Major Sector, (% of Total Employment) Measures of Per Capita Household Consumption Expenditure, Interquartile Consumption Ratios, by GLSS Wave, A.1 Price deflators (25/26 212/213)...55 A.2 Poverty Rates using the Revised and the Old Poverty Lines with Different Price Deflators B.1 Probit Results...59 B.2 Quantile Regression, B.3 Quantile Regression, B.4 Quantile Regression, B.5 Quantile Regression, B.6 Oaxaca-Blinder Poverty Decomposition by 4 th and 6 th Percentiles, Variation Between Contents v

8

9 Acknowledgments This report was prepared by a core team comprising Vasco Molini (Task Team Leader, GPVDR), Pierella Paci (Lead Economist, GPVDR), Fabio Clementi, Dan Pavlesku, and Francesco Schettino (consultants), under the overall guidance of Pablo Fajnzylber (Practice Manager, GPVDR), Yusupha Crookes (Country Director, AFCW1), Kathleen Beegle (Program Leader, AFRW1) and Andrew Dabalen (Lead Economist, GPVDR). Rose Mungai, Aly Sanoh, Ayago Wambile, Marco Ranzani, and Dhiraj Sharma (GPVDR) and Michele Tuccio (consultant) provided important inputs. We would also like to thank Ghana Statistical Service for their collaboration and willingness to share their data. The report benefited from discussions with government officials, development partners, the World Bank Africa Poverty team and participants in a workshop held in Accra in October 215. Martin Buchara (Team Assistant, GPVDR) provided excellent logistical assistance in the preparation of this report. Final editing was done by Robert Zimmerman. vii

10

11 Executive Summary Ghana has posted a strong growth performance during the past two decades. After more than a decade of stable annual growth in gross domestic product (GDP) at between 4 and 5 percent, growth began to pick up in the early 2s and reached a steady rate of nearly 8 percent after 26 (figure 1). Over the last 2 years, the Ghanaian economy has almost always grown more quickly than have the economies of other Sub-Saharan African countries (Africa hereafter) and at rates similar to those of lower-middleincome countries. Rapid growth translated into accelerated poverty reduction. The poverty rate fell by more than half between 1991 and 212, from 52.6 percent to 21.4 percent (figure 2). The country seems easily on track to reduce the poverty rate in line with Millennium Development Goal 1. Its performance compares well with that of other countries in Africa. In 212, the poverty rate in Ghana was less than half the African average of 43 percent, while, in 1991, it had been only 1 percent lower than the African rate. 1 The extreme poverty rate declined even more quickly, dropping from 37.8 percent in 1991 to 9.6 percent in Between 1991 and 26 the impact of growth on poverty in Ghana was higher than the rest of Sub-Saharan Africa. Until 25 Ghana enjoyed an elasticity of poverty to growth well above 2, closer to that of other developing countries and well above the SSA average. However, since then the elasticity has declined to a more modest.7, close the SSA average but considerably below that of other developing countries. Progress has gone beyond the reduction of consumption poverty. Ghana has also substantially improved various nonmonetary indicators of poverty. For example, infant mortality declined from 57 deaths per 1, live births in 1998 to 41 in 214, and under-5 mortality declined by more than half (table 1). Fertility is also decreasing and this has led to reduction in the dependency ratio. These improvements in poverty reduction occurred during a period of rapid change in the economic and sociodemographic structure of the country. Three are the factors associated with the reduction in poverty: (1) structural transformation, (2) the growing skills of the labor force and (3) geographical mobility. Figure 1 Real GDP Growth, Percent Source: World Development Indicators (WDI) database Ghana Lower-middle income Sub-Saharan Africa (developing only) ix

12 Figure 2 Trends in Poverty and Extreme Poverty, Percent Poverty Extreme poverty Source: Calculations based on GLSS 3 6. Table 1 Infant and Under-5 Mortality, Vaccination, and Fertility Rates, Rate Infant mortality Under-5 mortality Vaccination Fertility Source: Demographic and Health Surveys : STATcompiler (DHS Program STATcompiler) (database), ICF International, Rockville, MD, The recent economic growth is associated with a shift of the economy out of agriculture. The share of agriculture in GDP declined by nearly 4 percent, from over a third of GDP in 1991 to 23 percent in 212. As a result, by 211, agriculture was the smallest sector in the economy in terms of value added, although it is still the main sector of employment, representing 43.2 percent of total employment. Agriculture also saw a stable increase in productivity and a substantial reduction in the workforce. The production of cocoa and other cash crops were the main drivers of these changes. Beginning in the late 199s, cocoa production expanded quickly, and Ghana became the world s third-largest cocoa producer. Workers leaving agriculture moved mostly into services and, to a lesser extent, industry. Employment in the service sector expanded from 28.8 percent of total employment in 1991 to 42. percent in 212, and the share of construction more than doubled over the two decades. The share of employment in industry also increased, although at a lower rate, from 1.7 percent to 14.9 percent. The main driver of x the relative growth in the service sector 23.9 percent of total GDP growth was an heterogeneous group of activities including financial and business services, public administration, education, health care, social protection, and other services. This was followed by transport, storage and communication, 18.5 percent of overall GDP growth, and wholesale, retail trade, restaurants and hotels, 13.8 percent. The type of job done is strongly correlated with poverty status. In 212, agriculture was by far the dominant sector of employment among the poorest 2 percent of the consumption distribution, but it accounted for only 14 percent of the jobs among the richest 2 percent, 42 percent of whom are wage employees (figure 3). The labor force has become better educated and higher level of education translated in better job opportunities over the past two decades. Between 1991 and 212, the share of the labor force without schooling almost halved from 41 to 24 percent and in 212 the majority of workers had completed at least junior secondary education, compared to 39 percent in Figure 4 also shows that education is an important driver Executive Summary

13 Figure 3 Distribution of Employment, by Sector and Consumption Quintile Richest Fourth Third Second Poorest Percent Wage, private Wage, public Self, nonagriculture Self, agriculture Source: Calculations based on GLSS 3 6. Figure 4 Workforce Educational Attainment, by Sector of Employment, Percent None Primary Junior secondary Senior secondary Tertiary Source: Calculations based on GLSS 3 and 6 Wage, public Wage, private Self, nonagriculture Self, agriculture of higher earnings within each type of work and it also facilitated access to more productive, and better paid, activities. It is particularly striking that, more than half of Ghana s tertiary educated workers are employed in the public wage sector; with another thirty percent in the private wage sector, and only 14 percent in non-farm selfemployment. The high concentration of highly educated workers into the public sector may act as barrier to the development of a modern private sector. Location was another major correlate of poverty. Spatial inequalities in the incidence of poverty are striking and patterns of poverty are closely linked to a divergence in employment Executive Summary opportunities. It is therefore not surprising that moving to the faster growing areas of the South and Ashanti has been seen by many as an effective way of escaping poverty. In absolute numbers, since 1991, Accra and Ashanti gained over 2.4 million inhabitants each, around half of which in the last decade. Which of these developments were the main drivers of poverty reduction and shared prosperity? Using econometric analysis and standard decomposition techniques the report concludes that, on average over the period changes in the characteristics of the household, in what they do and where they leave are just as important in explaining the positive performance as changes in xi

14 Figure 5 Decomposition of Poverty Changes Variation in poverty rate Endowments Coefficients Interaction Source: Calculations based on GLSS 3 6. Table 2 Interquartile Consumption Ratios, by GLSS Wave, Year p9/p1 p9/p5 p1/p5 p75/p25 p75/p5 p25/p5 Gini Source: Calculations based on GLSS 3 6. the rewards to these characteristics. However, before 25 the improvement were mainly driven by high returns to certain household characteristics while further improvements after 25 were driven by accumulation of productive endowments i.e. to changes in some of those characteristics (Figure 5). Among these characteristics, it is worth noting the importance of access to basic infrastructures, increased educational attainment and modification in the household structure. Despite Ghana s success in reducing poverty and promoting shared prosperity, three significant challenges remain: growing inequality and polarization in household consumption, large spatial disparities in welfare, and the deteriorating macroeconomic environment. Consolidating the progress made in poverty reduction and shared prosperity in recent decades will require addressing these challenges effectively. Inequality in household consumption increased, particularly between 1998 and 25. xii In 1991, consumption per capita was about five times greater among the top decile of the distribution than among the bottom percentile (table 2). By 212, the gap had widened to nearly seven times, and the Gini index rose 8 percent, from 37.5 to 4.8. However, Ghana still compares favorably with other African countries; its Gini is still below the median and one of the lowest compared with rapidly growing African countries. Much of the increase in inequality is the reflection of increased regional disparities although within region inequalities are also pronounced. Poverty rates have fallen below 2 percent in the large area that includes the Ashanti, Eastern, Greater Accra, and Western regions, southern Brong Ahafo Region, and coastal Volta Region. Poverty has also declined well below the 4 percent in the central belt. Recent improvements notwithstanding, the poverty rate is far above 4 percent in most districts in the north. As a result poverty has increasingly become concentrated in rural areas and in the Executive Summary

15 Figure 6 Poor Individuals by Regions and Urban Rural Areas: Number of individuals in thousands 1, 5 Western Central Greater Accra Volta Eastern Ashanti Rural Brong Ahafo Northern Upper East Upper West Western Central Greater Accra Volta Eastern Ashanti Brong Ahafo Northern Urban Upper East Upper West Source: Calculations based on GLSS 3 6. Northern part of the country: one out three poor people lives in the northern rural areas while in 1991 it was only one out of five (Figure 6). Ghana is facing deteriorating macroeconomic prospects. Since 214 GPD growth has halved and is projected to slow further to 3.4% in 215 as energy rationing, high inflation, and ongoing fiscal consolidation continue to weigh on economic activity. Simulations indicate that the decline in household purchasing power alone could cause an increase of the current poverty rate of about 5 percent, with the urban lower deciles being the most affected. Looking forward, the development challenge faced by Ghana is to consolidate its poverty reduction successes in the context of difficult internal and external economic conditions and a rapidly changing economic and social environment. The deteriorating macroeconomic outlook and the persistent inequalities threaten future progress given the strong link between economic growth and poverty reduction and the tendency of higher inequality and polarization to translate into a lower growth elasticity of poverty reduction. Preventing further deterioration in the macroeconomic environment is the most Executive Summary immediate policy priority. The country s longterm growth prospects remain positive, but, to realize its full potential, Ghana needs to succeed in the implementation of the stabilization and reform program undertaken in 214. The success of the program hinges on a sustained commitment to fiscal discipline, rapid progress in structural reforms and the reduction of inflation. Tackling inequalities in outcomes and opportunities is a longer-term development challenge, but it is key for consolidating the country s middle-income status. Ghana entered a new stage of development with its designation as a middle-income country in 211. It will now be difficult to achieve sustained progress in poverty reduction and shared prosperity without broadening the reach of the development process to those people who have so far been left behind. The main challenge is to improve access to opportunities across the entire population without stifling the energy of the economy. This will require a multifaceted, well-targeted, and fiscally sustainable package of policies that balance the needs of the poor with the needs of the most dynamic economic sectors. A small set of win-win policy areas emerge as priorities to consolidate Ghana s success in xiii

16 poverty reduction and shared prosperity. Improving the business climate is a must to enable a modern private sector to flourish and create high productivity, well-paying jobs. A continued commitment to investment in infrastructure and skills development will also be key to increase productivity in agriculture, create modern jobs and ensure that workers have the skills needed to take advantage of the new employment opportunities. Increased connectivity between rural and urban areas, combined with clearly Notes 1. The average poverty rate in Africa has been calculated using the international poverty line of $ For comparing the 4 survey rounds analyzed in the report, they were used the 1999 defined land rights and more efficient land markets will facilitate structural transformation by allowing Ghana to benefit from geographical agglomeration. Finally experimenting with innovative ways of expanding the coverage of social protection and improving its targeting can help reduce the high vulnerability to shocks of Ghanaian households and increase the productive potential of individuals by breaking the vicious circle that links inequality of income to unequal opportunities over generations. poverty line and food poverty line. See also GSS (Ghana Statistical Service) Poverty Profile in Ghana (25 213). August, GSS, Accra, Ghana. xiv Executive Summary

17 Introduction Ghana has long been at the vanguard of development in Africa. As the first country on the continent to gain independence from European rule, in 1957, Ghana continues to provide a model of stability, democracy, and prosperity to low- and middle-income countries in Africa and further afield. In addition, it has played a leading role in its neighborhood by promoting peace, economic development, and regional cooperation in West Africa. Framing Ghana s Success Progress made since the 199s have been particularly remarkable. The country has achieved dramatic gains in living standards, public health, and educational attainment, and it has enjoyed a stout increase in consumption among the bottom 4 percent of the consumption distribution (the bottom 4). These achievements have been accompanied by strong economic performance, leading to the country s landmark achievement of middle-income status in 21, a decade earlier than anticipated. Ghana s success story has been underpinned by the ability of the economy to generate an essential dividend from growth: poverty reduction. Both absolute and extreme poverty rates have dropped dramatically in the last 25 years, Since 1991, the national poverty rate has fallen by more than half, from 52.7 percent that year to 21.4 percent in 212 1, a feat few other countries can boast, and one that sets Ghana on course to meet the Millennium Development Goal Objective 1. Moreover extreme poverty rates have dropped dramatically in the last 25 years, and the share of the extreme poor in the population declined from 37.6 percent in 1991 to 9.6 percent in 212. Within such a background, this poverty assessment seeks to shed light on Ghana s milestone performance by profiling the changes and identifying their main drivers. The last 25 years have borne witness to perhaps the greatest advances in the country s history. For this reason, this study takes a medium-term perspective to identify and understand more closely the circumstances in which poverty has fallen so quickly and steadily since the 199s. The report profiles the progress made during this period in reducing poverty and increasing the consumption of households in the bottom 4 and identifies the main drivers of this success. The country s rapid economic growth has been the main engine of poverty reduction but challenges remain. Ghana has been more effective than other African countries in sharing the increased prosperity and transforming it into poverty reduction. Recent successes notwithstanding, several challenges remain and, using a combination of standard and more innovative techniques such as poverty mapping and polarization measures the report highlights three areas of concerns that are seen as particularly important: the growing inequalities of consumption and opportunities, the persistent spatial disparities, and the deteriorating macroeconomic environment. Despite the challenges, however, what follows herein is ultimately an extraordinary success story with lessons that may be learned and applied elsewhere in the developing world. In itself, the country s graduation to middle-income status is not without challenges. More Ghanaians than ever are well educated and seeking employment, while civil society is thriving. As more people are lifted from poverty, their expectations will rise, along with their living standards. Maintaining progress will therefore require laying the groundwork for the country s next stage of development. The policy agenda outlined at the end of the report is intend to guide the Ghanaian government in identify a comprehensive and fiscally sustainable policy package to Introduction 1

18 successfully address the challenges and consolidate Ghana s successes in the years to come. This report consists of 4 chapters. Chapter 1 profiles the trends in household consumption and poverty rates, and in the characteristics of the poor that occurred between 1991 and 212. Descriptive statistics of consumption and selected poverty indexes are presented and a profile of how the characteristics of the poor is given. The chapter concludes with an analysis of vulnerability. Chapter 2 uses descriptive and econometric technics to identify the drivers of Ghana s success over the last two decades. Chapter 3 examines the main challenges Ghana continues to face: widening inequalities, persistent spatial divide, and the deteriorating macroeconomic environment. Chapter 4 concludes and provides a roadmap for policy action to effectively address these challenges and consolidate Ghana s success as a middle income economy. Data The analysis in this report is based on data of the Ghana Living Standards Survey (GLSS) produced by the Ghana Statistical Service (GSS). The GLSS is a multipurpose survey that collects detailed information on individual and household characteristics and on basic indicators of living standards (table I.1). Six rounds of GLSS data have been collected since 1987, thereby providing over 2 years of comparable data. However, only the last four rounds, from GLSS (round) 3 to GLSS 6, have been based on the same questionnaire and are therefore fully comparable. The GSS collects sufficient information through the GLSS to estimate the total consumption of each household. This covers the consumption of both food and non-food items (including housing). Food and non-food consumption commodities may be explicitly purchased by households or acquired through other means (own production activities or through receipts). The household consumption measure takes into account all these sources in the various modules of the survey questionnaires. The availability of comparable survey data beginning in 1991 allows us to carry out an updated, detailed analysis of poverty in Ghana. The surveys have improved in quality over the years and involved the collection of data on both the monetary and non-monetary dimensions of welfare, thereby permitting an accurate analysis of poverty and inequality over time. The GLSS has emerged as one of the most important tools for the welfare monitoring system in Ghana. It provides the basis not only for official welfare measures and analysis, but also for detailed information on several socioeconomic and demographic characteristics, the household consumption of purchased and home-produced goods, asset ownership, and remittances. The GLSS is based on a two-stage (non- stratified) sample design; in data analysis, sampling weights are therefore used to account for the survey design. To enhance the comparability of consumption data over the four latest, comparable waves, all expenditures are deflated across both space and time, expressed in 25 constant prices, and converted, if necessary, from the Ghanaian second cedi ( ) to the Ghanaian third cedi (27 ), that is, to accommodate GLSS 3 to GLSS 5. Each of the four waves Table I.1 Ghana Living Standards Surveys 1 6, Dataset Collection period Sample size, number of households Representativeness Comparability GLSS 1 September 1987 August ,172 National, urban, and rural GLSS 1 and 2 are comparable GLSS 2 October 1988 August ,194 National, urban, and rural GLSS 1 and 2 are comparable GLSS 3 September 1991 August ,523 National, urban, and rural GLSS 3 6 are comparable GLSS 4 April 1998 March ,998 National, urban, and rural GLSS 3 6 are comparable GLSS 5 September 25 August 26 8,687 National, urban, and rural GLSS 3 6 are comparable GLSS 6 October 212 October ,772 National, urban, and rural GLSS 3 6 are comparable Source: GLSS Introduction

19 is organized into four modules, which are stored in the individual, labor force, household, and household expenditure files. The availability of comparable and extensive information covering over two decades represents a success on its own. Ghana is one of the few countries in Africa that has produced comparable, high-quality household data covering over two decades. 2 This is an important achievement because the availability of such rich and comparable information beginning in 1991 allows an updated and detailed analysis of the country s recent successes in poverty reduction, including the drivers behind the reduction. The quality improvements of the surveys over the years and the fact that they collect data on both the monetary and the non-monetary dimensions of welfare are particularly welcome because this permits the establishment of an accurate picture of poverty and inequality over time. Definitions and Technical Notes The selected measure of welfare is consumption per adult equivalent. Consumption has proven preferable to income as a measure of poverty because it is less volatile (for example, see Deaton and Zaidi 22; Haughton and Khandker 29). In agricultural economies in particular income is more volatile and more highly affected by the growing and harvest seasons, so that relying on income as an indicator of welfare might under- or overestimate living standards significantly. Consumption is a better measure of long-term welfare also because households can borrow, draw down savings, or receive public and private transfers to smooth short-run fluctuations. The GLSS collects sufficiently detailed information to facilitate estimates of the total consumption of each household. It relies on consumption per adult equivalent to capture differences in need by age and economies of scale in consumption. Scales of consumption by age and sex are computed by the GSS. The terms poverty and incidence of poverty refer to the poverty headcount index, unless otherwise specified. The headcount index measures the proportion of the population with per adult equivalent consumption below the value of a minimum basket of food and non-food items, that is, the poverty line. This population share with per adult equivalent consumption below the value of a minimum basket of food (food poverty line) only is called the extreme poor. 3 The report uses the 1999 poverty line. In 213, the GSS revised the poverty line and modified the basket to take into account the significant changes that had occurred in the consumption basket (see appendix A). 4 However, given the focus of this report on comparisons over time, we maintain the 1999 line as the poverty threshold throughout the entire period under analysis. This does not appear to have unduly affected the poverty rate for 212, which we estimate at 21.4 percent, compared with the official 24.3 percent based on the revised poverty line. The poverty gap index measures the extent to which individuals fall below the poverty line (the poverty gap) as a proportion of the poverty line. The sum of the poverty gaps so calculated yields the minimum cost of eliminating poverty if transfers were perfectly targeted. The squared poverty gap index (or the index of the severity of poverty) averages the squares of the poverty gaps relative to the poverty line. The Growth Elasticity of Poverty (GEP) is the percentage reduction in the poverty rate that is associated with a percentage change in mean (per capita) income. A numerical example clarifies the concept. In Ghana between 25 and 212 the GEP assumed value -.7 (see chapter 2). This value implies that a 1% increase in per capita income was associated with a.7% decrease in the poverty rate. The growth incidence curve plots the growth rate at each quantile of per capita consumption (or income). Graphs of growth incidence curves allow us to compare differences in the incidence of growth between poorer and richer segments of the population or with the rate of growth of mean consumption (or income) (see chapter 2). The cumulative consumption curve shows the level of welfare enjoyed by various percentiles Introduction 3

20 of the population in any particular year. Distributions to the right on the curve reflect an improvement in the overall welfare of the population, that is, they are statistically dominant and can be considered. The horizontal axis represents consumption measured as a percentage of the poverty line. The vertical axis represents the percent share of the population, and each point on the distribution function shows the share of the population below a certain percentage level of the poverty line. The point on the vertical axis that corresponds to the vertical line that indicates 1 percent of the poverty line yields the poverty rate. There are several measures of the vulnerability to poverty. In this report, we use a simple, unsophisticated measure: 14 percent and 18 percent of the poverty line The Oaxaca decomposition (Oaxaca 1973), explains the gap in the means of an outcome variable (consumption in our case) between two groups (e.g., between two survey rounds, 1991 and 212). The gap is decomposed into that part that is due to group differences in the magnitudes of the determinants (endowments effect) of consumption, on the one hand, and Notes 1. For comparing the 4 survey rounds analyzed in the report, they were used the 1999 poverty line and food poverty line. See also GSS (Ghana Statistical Service) Poverty Profile in Ghana (25 213). August, GSS, Accra, Ghana. 2. The GLSS is also part of the Survey-Based Harmonized Indicators Program, which is run by the World Bank to combine and harmonize household surveys across various African countries, including Burkina Faso, Cameroon, Ethiopia, The Gambia, Ghana, Kenya, Madagascar, Malawi, group differences in the effects of these determinants (coefficients effect), on the other. Polarization is the combination of the divergence from the global mean income and the convergence toward local mean incomes. Polarization differs from inequality because the latter is the overall dispersion of the distribution, that is, the distance of every individual from the median or mean income. In income-polarized societies, people are clustered around the group means and tend to be remote from the mean or median of the overall distribution. Within each group, there is income homogeneity and often narrowing income inequality. Thus, we may talk of increasing identification. Between the two groups, we talk, rather, about increasing alienation (Duclos, Esteban, and Ray 24). The overall impact of the forces of identification and of alienation between two groups of significant size leads to effective opposition, a situation that may give rise to social tensions and conflict (Esteban and Ray 1999, 28, 211). Also, the group at the top of the distribution possesses voice, while the other group, which is made up of those at the bottom, are voiceless in matters that affect their welfare and society at large. and Zambia. The program involves verification of internal and external consistency and harmonization across a common set of variables. 3. Alternatively, the extreme poor are those people whose standard of living is insufficient to meet basic nutritional requirements even if they were to devote their entire consumption budgets to food. 4. The old poverty line was C per adult equivalent per year in 25 prices. The food poverty line is C per adult equivalent per year in 25 prices. See also GSS (214). 4 Introduction

21 Chapter 1 A Tale of Success: Shared Prosperity, Poverty Reduction, and a Boost in Opportunity The economic development of Ghana has been a tale of success over the last two decades. The country has achieved sound economic growth and a stout increase in consumption among the bottom 4 percent of the consumption distribution (the bottom 4). It has cut poverty by half and considerably reduced vulnerability. Strong growth in income and consumption has been accompanied by substantial improvement in non-monetary indicators of living standards. Newborns in Ghana today are expected to live two years longer than newborns in 25, and children are more than twice as likely to be enrolled in secondary school. They live in houses that are more than twice as likely to have electricity and improved sanitation facilities. The striking progress made in providing better opportunities for all is more than an achievement in its own right. It has also strengthened the prospects for strong and inclusive growth in the future. Two Decades of Expanding Prosperity for All Ghana has posted a solid growth performance during the last two decades. After over a decade of stable annual growth at between 4 and 5 percent, gross domestic product (GDP) began to pick up in the early 2s and reached a steady growth rate of nearly 8 percent after 26. There was then an impressive peak in 211 mainly because of the discovery of oil and the rebasing of GDP 1 Since 28, Ghana has grown more quickly than other African economies and, since 21, more quickly than the average among lower-middle-income countries (figure 1.1). It reached the middle income status in 21. These achievements have been important especially in light of concurrent external shocks, such as the global financial crisis of 28 9, reduced trade revenues, and fluctuating oil prices. Figure 1.1 Real GDP Growth, Percent Source: World Development Indicators (WDI) database Ghana Lower-middle income Sub-Saharan Africa (developing only) A Tale of Success: Shared Prosperity, Poverty Reduction, and a Boost in Opportunity 5

22 Table 1.1 Mean and Median Real Consumption (Adult Equivalent), (C ) Indicator Mean Nationwide ,132 Urban ,281 1,397 Rural Median Nationwide Urban ,5 1,7 Rural Source: Calculations based on GLSS 5 6. Figure 1.2 Growth Incidence Curves, and a b National National Percentiles 95% confidence bounds Growth rate Growth rate in mean Growth rate at median Percentiles Source: Calculations based on GLSS 5 6. The impressive GDP growth of the last decade was associated with a substantial increase in average household consumption. Mean and median real consumption almost doubled, with the most rapid growth occurring between 1998 and 25, when consumption expanded by about 3 percent (table 1.1). The rates of growth in rural and urban areas were almost the same in , though there were substantial differences across sub-periods. In urban areas, the growth rate was greater than the national average between 1998 and 25, but slowed significantly between 25 and 212. In rural areas, growth slowed during the latter period, but was, overall, more rapid than that in urban areas. The benefits of growth were broadly shared among the population, although the top percentile received relatively more. The growth incidence curves in figure 1.2 show the variation in consumption between 1991 and A Tale of Success: Shared Prosperity, Poverty Reduction, and a Boost in Opportunity

23 (top quadrant) and between 25 and 212 (bottom quadrant) across percentiles of the distribution. Overall, the growth of consumption in was less than the growth over the whole period: an average rate of 2.5 percent a year against 3.2 percent. On the basis of their performance, three subgroups stand out: the top 2 percent of the distribution (the top 2), the bottom 25 percent (the bottom 25), and the percentiles in between. The top percentiles grew consistently more rapidly than the national average, and this contributed to a widening in inequality (see below). The bottom 25 performed less well than the rest over the whole period, but its performance improved, in relative terms, in the last decade: it continued to grow at about 2.5 percent a year, which, over the decade, coincided with average consumption growth. The middle percentiles grew at rates slightly below the mean (2.8 percent versus 2.9 percent), but still at a sufficiently high rate to justify considering their performance particularly good. This group roughly corresponds to those people who, in 1991, were poor or vulnerable to poverty and who had managed to move out of poverty (or, among some of them, out of vulnerability) by 212 (see subsequent chapters). Shared Prosperity Led to a Reduction in Poverty and Vulnerability Poverty has been cut in half over the last two decades. The estimated poverty rate fell by 31.2 percentage points, from 52.7 percent in 1991 to 21.4 percent in The country seems on track to reduce the poverty rate by half in line with Millennium Development Goal 1. Ghana s performance in reducing poverty compares well with that of other African countries. In 212, the rate of poverty in Ghana was less than half the African average of 43 percent, while, in 1991, it had been only 1 percent lower. 3 However, the highest poverty reduction, 13. percent, was recorded between 1991 and Since then, the speed of poverty reduction has declined, to 11. percent in and to 7.1 percent in This is despite increasing GDP growth (see figure 1.1). Extreme poverty declined more rapidly. The share of the population with consumption below the food poverty line declined from 37.6 percent in 1991 to 9.6 percent in 212 (figure 1.3). Rapid growth in average consumption was the driving force behind the impressive poverty reduction. Figure 1.4 decomposes the change in the incidence of poverty experienced between 1991 and 212 into a component caused by the expansion in average household consumption and a component caused by changes in inequality (Kolenikov and Shorrocks, 23). The sharp decline in poverty was clearly driven almost exclusively by the increase in average household consumption. Inequality changed little over the period, and the impact of growth was large enough to offset the potential rise in poverty associated with the small increase in inequality experienced after Figure 1.3 Trends in Poverty and Extreme Poverty, Percent Poverty Extreme poverty Source: Calculations based on GLSS 3 6. A Tale of Success: Shared Prosperity, Poverty Reduction, and a Boost in Opportunity 7

24 The important impact of economic growth on poverty reduction is reflected in the comparatively high poverty elasticity of economic growth. Overall, the poverty elasticity of growth over the last decade has been low in Africa compared with the rest of the world, but Ghana has been an exception. It experienced an average elasticity of 1.36 over (figure 1.5). However, the value declined from well above 2. to a more modest.7 after 25. The comparison with the averages of Africa (SSA in the figure) and Figure 1.4 Growth, Inequality, and Poverty Decomposition, Variation in poverty incidence Growth contribution Source: Calculations based on GLSS 3 6. Inequality contribution Poverty variation the rest of the developing world (ROW) also shows that the elasticity was higher in Ghana than in the rest of Africa and closer to the elasticity of other developing countries in , but has since fallen dramatically also in relative terms. The depth of poverty has also declined. Both the poverty gap and the severity of poverty fell quickly, in particular until 25 (table 1.2). This, in combination with the outcomes among the extreme poor, suggests that important changes occurred to the consumption of people living below the poverty line (figure 1.6). The share of people living below the poverty line by a certain fixed amount (8 percent or 6 percent) can be used to measure the depth of poverty; 8 percent of the poverty line is close to the value of the food poverty line. The reduction in the incidence of poverty was accompanied by a sharp drop in the depth of poverty because the share of people living below 6 percent or 8 percent of the poverty Table 1.2 Poverty Rates, the Poverty Gap, and the Severity of Poverty, Year Poverty rates Poverty gap Severity of poverty Source: Calculations based on GLSS 3 6. Figure 1.5 Growth Elasticity of Poverty, GHA SSA ROW Source: Calculations based on GLSS 3 6 and WDI. 8 A Tale of Success: Shared Prosperity, Poverty Reduction, and a Boost in Opportunity

25 Figure 1.6 Vulnerability as Distance from the Poverty Line, Nationwide, Share of population % of poverty line Source: Calculations based on GLSS 3 6. line narrowed considerably. That the distance from the 1991 incidence curve is greater around the poverty line suggests that the expansion in consumption was most rapid among those households with consumption levels closer to the poverty threshold. This is consistent with the conclusion that the extreme poverty rate fell more quickly than the poverty rate. Vulnerability remains widespread despite the recent in the number of Ghanaians with consumption levels close to the poverty line. Non-poor households with consumption levels slightly above the poverty line are often labeled vulnerable because even comparatively small shocks may push them into poverty. In the profile of the vulnerability of the Ghanaian population, vulnerability thresholds of 14 percent and 18 percent of the poverty lines are particularly relevant because a loss of less than $.5 a day in consumption can push households below the national poverty line of around $1.3 a day. The cumulative curves of consumption in Figure 1.6 show that the share of the population living under 14 (18) percent of the poverty line declined from around 73 (83) percent in 1991 to 39 (51) percent in 212. In both cases, the drop was slightly greater than the reduction in poverty of around 3 percent. The decrease in the number of the vulnerable was particularly significant in and Vulnerability is the norm in rural areas, and the progress in urban areas has slowed appreciably since 25. A rural-urban disaggregation shows that the increased in consumption in urban areas occurred mostly in and, especially, in , but slowed considerably in , that is, the curves for the last two periods almost overlap (figure 1.7). By contrast, in rural areas, progress was more modest, but more evenly distributed. However, more than half of rural households are still extremely vulnerable to shocks, that is, their levels of consumption are less than 14 percent of the poverty line. The Poor over Time: Better Attributes and Living Standards Poverty has always been predominantly rural. In recent years, however, poverty has become even more concentrated in rural areas. By 212, poverty rates were 38.2 percent in rural areas and 1.4 percent in urban areas, and the median A Tale of Success: Shared Prosperity, Poverty Reduction, and a Boost in Opportunity 9

26 Figure 1.7 Vulnerability and Poverty Depth as Distance from the Poverty Line, Rural and Urban Areas, Rural Urban Share of population for both rural and urban Share of population for both rural and urban % of poverty line % of poverty line Source: Calculations based on GLSS 3 6. consumption per capita was almost two times greater in urban areas than in rural areas. For every poor person in urban areas, there were nearly four poor people in rural areas. The share of the poor among rural residents did not change much between 1991 and 212, but the share of rural residents in the total population fell. Whereas, in 1991, 67 percent of the total population was living in rural areas, the share had dropped to 5 percent by 212. The incidence of poverty was especially high among people living in the rural savanna. These people accounted for more than 4 percent of the overall poverty rate. The poor live mostly in the north. The trends in the number of the poor by region confirm that their concentration has been relatively greater in the north than in the rest of the country. This is caused by the combination of less favorable climate, distance from the sea, and lack of infrastructure. However, these disadvantages appear to have grown over the last two decades to the extent that, whereas both the poverty rate and the absolute numbers of the poor have declined in the more populous southern and central regions, the number of the poor has risen in the Northern Region and Upper West since 1991 (figure 1.8). As a result of these divergent trends, nearly 4 percent of the poor were living in the north in 212, but only 17 percent of the population. The joint profile of residence in the north and residence in rural areas accounted for 34 percent of the poor in 212, but only 2 percent in Poor households tend to be larger than non- poor households. In 212, poor households had an average of 7.5 members, nearly 5 percent more than non-poor households, and, on average, they also had more children (table 1.3). In poor households half of the members are below 14 years of age, and 2 percent are under 5 years of age, while, among non-poor households, the shares are, respectively, 4 percent and 1 percent. The heads of poor households are likely to be older men who are self-employed. The average head of a poor household is approaching 5 years of age 3.5 years older than the average head of a non-poor households and is considerably less likely to be a woman (by 33 percent). 89 percent of heads of poor household are self-employed, 78 percent of them in agricultural. This compares with, respectively, 66 percent and 35 percent of heads of non-poor household. Agricultural self-employment remains the prevalent economic activity among household heads in the bottom three quintiles (figure 1.9). 1 A Tale of Success: Shared Prosperity, Poverty Reduction, and a Boost in Opportunity

27 Figure 1.8 Trends in Regional Poverty Headcounts, ,5 Number of poor individuals (thousands) 1, 5 Western Central Greater Accra Volta Eastern Ashanti Brong Ahafo Northern Upper East Upper West Source: Calculations based on GLSS 3 6. Table 1.3 The Profiles of Poor and Non-poor Households, Poor Non-poor Household characteristic, % unless indicated Age of household head, mean Sex of household head Female Male Makeup of household Mean number of members Dependency ratio, age 14 and 65 relative to Share of 14 age-group Share of 4 age-group Highest educational attainment, household head No education, other, not known Primary Secondary Higher Household head, occupation Wage, private sector Wage, public sector Non-agricultural self-employment Agricultural self-employment Other, including unemployed Region of residence Western Central table continues next page A Tale of Success: Shared Prosperity, Poverty Reduction, and a Boost in Opportunity 11

28 Table 1.3 continued Poor Non-poor Household characteristic, % unless indicated Greater Accra Volta Eastern Ashanti Brong Ahafo Northern Upper East Upper West Area of residence Rural Urban Water supply Protected Unprotected Vendor, truck Connection of electricity in dwelling None None Connected Main cooking fuel Firewood Other Main toilet facility Toilet, improved pit latrine Other Ownership of durables Radio and television Television Bicycle Motorcycle Source: Calculations based on GLSS 3 6. The poor are largely unskilled. Over 5 percent of the poor in 212 had no education, compared with only 21 percent of the non-poor. An additional 23 percent of the poor had only primary education, compared with 2 percent of the non-poor. In addition the proportion of poor without education has declined little since 1991 while it has sharply declined amongst the non-poor. In 212, the majority of the poor still had little access to basic infrastructure, though they did have some connectivity. A majority of poor households did not have access to electricity (62 percent) or adequate sanitation (69 percent), and 85 percent relied on firewood for cooking. This is in sharp contrast to the availability of services of the non- poor, which mostly enjoyed centrally provided electricity (74 percent) and adequate sanitation (63 percent) and used fuels other than firewood for cooking (59 percent). Nonetheless, even the poor enjoy some degree of connectivity with the outside world at least via radio. 12 A Tale of Success: Shared Prosperity, Poverty Reduction, and a Boost in Opportunity

29 Figure 1.9 Main Sectors of Employment, Household Heads, by Quintile, 212 Richest Fourth Third Second Poorest Percent Public wage Private wage Self-employment, non-agr Self-employment, agr Source: Calculations based on GLSS 6. BOX ? How has the typical poor household changed between 1991 and Despite the considerable socio-economic changes Ghana experienced in the period, the typical poor Ghanaian household in 1991 and the typical poor household in 212 were not dissimilar: 1. each had seven-eight members; 2. they were both headed by a man approaching 5 years of age who had no education and worked in agriculture; 3. they had no access to electricity or adequate sanitation and used firewood for cooking; and 4. they were connected to the outside world through the radio (see table 1.3). However, a more careful analysis reveals that, compared to its 1991 counterpart, the typical poor household in 212 was considerably better-off in non-monetary indicators. It had: 1. considerably better access to basic services and infrastructure; 2. higher ownership of durables; and 3. nearly twice the probability of enjoying some level of educational attainmenta. Better and More Equal Opportunities for Sustainable Progress Welfare includes more than reduced poverty and greater consumption among the less well off. Our analysis has so far focused mainly on progress measured through monetary and consumption indicators of household welfare. However, income or consumption alone do not provide a complete picture. The non-monetary dimensions of deprivation are also important not only in their own right, but also because they are associated with inequalities of opportunity that can exacerbate A Tale of Success: Shared Prosperity, Poverty Reduction, and a Boost in Opportunity 13

30 disadvantages in income or consumption. Among these dimensions, asset accumulation, in particular, is an important indicator of how economic growth is shared across a population and of the success or failure of a sustainable poverty reduction strategy. Likewise, improvement in human capital accumulation and in access to basic services are also key components of welfare. Ghana has achieved good progress on these indicators, too. Average educational attainment has improved considerably in Ghana. By 212, the youth literacy rate had reached 79 percent, up 24 percentage points since 1991, and the gender gap in this indicator had narrowed to about 7 percent, from 11 percent in 1998 (table 1.4). 4 Overall, adult literacy had also improved, and, by 212, Ghana ranked 15th in Africa, with about 75 percent of the adult population considered literate. The net enrollment rate in primary school rose from 55 percent in 1991 to 75 percent in 212. However, there are many persistent challenges in the education sector. Enrollment rates in senior-secondary school and tertiary education remain low, particularly among the bottom 2 percent of the consumption distribution (the bottom 2) and among rural residents. In 212, only 1 percent of youth living in poor households had attained tertiary education, compared with 39 percent among youth in the 5th (richest) quintile. Moreover, the average child in the richest quintile was more than twice as likely as the average child in the poorest quintile to complete primary school, 69 percent compared with 3 percent. Small but persistent gender gaps in educational attainment are also characteristic of the bottom income groups. In addition the quality of learning in both basic and post-basic education remains low and of the education system has limited capacity to create relevant skills for increasing competitiveness and productivity. About Table 1.4 Educational Attainment, Bottom 2 and Top 2, by Gender, Level Bottom 2 Top 2 Bottom 2 Top 2 Bottom 2 Top 2 Gross primary enrollment ratio Female Male Gross lower-secondary enrollment ratio Female Male Gross senior-secondary enrollment ratio Female Male Net primary enrollment rate Female Male Net lower-secondary enrollment rate Female Male Net senior-secondary enrollment rate Female Male Youth literacy rate (ages 14 25) a Female Male Source: Calculations based on GLSS 3 6. a. The total number of literate females or males (based on self-reporting) aged expressed as a percent of total females or males in the age-group. The 1998 enrollment rates are not comparable with the rates in the 1991, 25, and 212 surveys because the years of education were not captured in the 1998 survey. 14 A Tale of Success: Shared Prosperity, Poverty Reduction, and a Boost in Opportunity

31 Table 1.5 Infant and Under-5 Mortality, Vaccination, and Fertility Rates, Rate Infant mortality Under-5 mortality Vaccination a Fertility Source: Demographic and Health Surveys : STATcompiler (DHS Program STATcompiler) (database), ICF International, Rockville, MD, a. Children aged months are fully vaccinated, that is, they have received BCG, measles, and three doses of DPT and polio vaccine (excluding polio ) according to their vaccination cards or the reports of their mothers. Figure 1.1 Access to Basic Services Has Improved, Proportion of population (%) Electricity Improved and flush toilet Garbage collection Source: Calculations based on GLSS 3 6. two-thirds of the students who complete primary school do not attain proficiency in core subject areas and there are large disparities between the 1 most highly subscribed Senior High Schools (roughly 2 percent of students) and the rest of the sector. The top 1 percent of schools produce 9 percent of the students entering university while students with low WASSCE examination marks enroll in some form of training or apprenticeship program (World Bank, forthcoming). Newborns in Ghana today are expected to live two years longer than newborns in 25, and girls have higher life expectancy than boys. Over the last decade, infant mortality declined by about 3 percent, from 57 deaths per 1, live births in 1998 to 41 in 214, and under-5 mortality decreased by more than half, from 54 to 19 deaths per 1, live births (table 1.5). This is partly the result of efforts of the government to raise vaccination rates among all children. About 84 percent of children across the country have been immunized. 5 The fertility rate is declining. This has led to a reduction in the dependency ratio. However, in 28, fertility rates varied considerably across Ghana: they were higher in rural areas 4.9 births per woman, compared with 3.1 in urban areas and in the Northern and Upper West regions. There was substantial improvement in access to basic household services. The last decades have seen improvements in access to sanitation, electricity, and clean drinking water (figure 1.1). The striking expansion in the A Tale of Success: Shared Prosperity, Poverty Reduction, and a Boost in Opportunity 15

32 coverage of garbage collection after 25 from less than 1 percent to about 6 percent reflects the efforts of the government to provide such services to the urban poor through subsidies. The success of the policy has made Ghana a leader in sanitation in Africa. The positive trends in household consumption and in other dimensions of household welfare over the last decade can be summarized through an asset index. 6 The cumulative distributions of the asset index highlight the rapid pace at which households along the entire population distribution have accumulated assets over the two decades. Thus, the curve for 212 lies well to the right of the curves for previous years (figure 1.11). The progress was particularly striking after 25, reflecting the lagged effects of the substantial consumption growth between 1998 and 25 and the sustained public investment in infrastructure, including roads, electricity grids, and sanitation infrastructure and services. A comparison confirms the large divide between urban and rural areas. The urban distribution is always to the right of the rural distribution, and the probability density is always much higher above than below (see figure 1.11). By contrast, despite recent improvements, the rural probability density continues to lie mostly below. However, the considerable shift to the right of the rural distribution shows that the economic growth of the last two decades has been inclusive and has gone some way to reducing inequalities in opportunity. Progress in the non-income dimensions of household welfare strengthens the prospects for future growth and poverty reduction by improving the opportunities for all. Empirical evidence from across all the regions of the world shows that disadvantages in the accumulation of physical and human capital and in access to services inequality of opportunity are pervasive, Notes 1. In 21 Ghana changed its base year from 1993 to 26, and this led to a jump in GDP and the conclusion that, in previous estimates, about $13bn Figure 1.11 Asset Index, Rural and Urban Areas and Nationwide, Share of population Share of population Share of population a. Rural Assist index values b. Urban Assist index values c. National Assist index values Source: Calculations based on GLSS 3 6. exacerbate income disadvantages by limiting the potential of individuals from birth, and may have long-term negative impacts on the potential for growth and poverty reduction within countries. Ghana has taken important steps toward breaking this vicious circle. ( 8bn) of economic activity had been missed. As a result, Ghana was upgraded from a low-income to a lower-middle-income country. 16 A Tale of Success: Shared Prosperity, Poverty Reduction, and a Boost in Opportunity

33 2. See chapter 1, on the poverty line selected. 3. The average rate for Africa is calculated using the international poverty line of $ The youth literacy rate as used here is the percent of the population aged years that can, with understanding, read and write a short, simple statement about their everyday lives. The adult literacy rate is based on the same criterion except the population of reference is aged 15 years or above. 5. Rural areas show infant (under-5) mortality rates of 56 (9) per 1, live births, compared with 49 (75) in urban areas. Infant (under-5) mortality rates are 7 (137) and 97 (141) per 1, live births in the Northern and Upper West regions, respectively, but well below 4 (65) per 1, live births in the Ashanti, Brong Ahafo, and Greater Accra regions. Progress in immunization has been widespread, although coverage in the Northern Region was below 6 percent in The asset index, suggested by Filmer and Pritchett (21), uses principal-components analysis to calculate the weights of the index. The first principal component, the linear combination capturing the greatest variation among the set of variables, can be converted into factor scores, which serve as weights. The rationale for using this index is that it captures the household s permanent welfare dimension more effectively relative to simple consumption data and can provide more reliable rankings across households. A Tale of Success: Shared Prosperity, Poverty Reduction, and a Boost in Opportunity 17

34

35 Chapter 2 Drivers of Poverty Reduction The improvements in household consumption and poverty reduction described in chapter 2 occurred during a period of rapid change in the economic and socio-demographic structure of the economy. In this chapter, we look more closely at factors associated with poverty reduction, focusing on three key developments of the past decade: (1) structural transformation, (2) increased skills among the labor force, and (3) urbanization. Underlying the choice of these factors is a simple analytical framework according to which per capita household consumption is driven by household composition, income from labor and other sources, and prevailing consumption patterns. Non-labor income is usually derived from public and private transfers and returns to capital. With the possible exception of remittances, transfers and returns to capital are virtually nonexistent in Ghana, as they are in most other African countries. Thus, work is the main source of income, especially among the poor. Many families escape (or fall) into poverty because family members obtain (or lose) jobs or because the returns from work are augmented (or reduced) by factors within their control (investments in education, migration) or by exogenous shocks (poor or abundant rainfall). However, in Ghana having a job is often not sufficient to bring workers out of poverty because the returns to work tend to be extremely low. It is growing earnings from work that make a difference in the effort to escape poverty. Family composition and demographics are also important because they affect the dependency ratio, that is, the number of consumers relative to the number of earners in the household. Azevedo et al. (213), in a quantitative analysis, suggest that changes in labor income accounted for nearly half the reduction in poverty in Ghana between 1998 and 25, but that changes in household composition were also important. Moreover, Falco et al. (214), in an econometric study that uses panel data to trace the earnings of individuals over time, find that the main determinants of both earnings and the growth of earnings over the life span are type of job defined as in this chapter and level of education. They also find that beginning working life in a low-paying activity such as agriculture has a scarring effect by reducing earning prospects for the rest of the worker s life. The structure of the chapter is as follows. In the first section, we describe three key developments in Ghana over the past decade. The second section relies on GLSS data to examine how these developments have been associated with poverty reduction and consumption growth and, thereby, identifies the main drivers behind these phenomena. Key Economic Developments Structural Transformation Ghana s recent economic growth has been associated with a shift of the economy out of agriculture into services. The share of agriculture in GDP declined by nearly 5 percent, from over a third of GDP in 1991 to 23 percent in 212 (figure 2.1). As a result, by 211, agriculture was the smallest sector in the economy in terms of value added, and its share has continued to decline since. Meanwhile, the service sector expanded to nearly half of GDP, from an initial 34 percent in Services account for more than half of per capita GDP growth. The main driver of the relative growth of the service sector, accounting for 23.9 percent of total GDP growth, was an heterogeneous group of other activities (table 2.1). This included financial and business services, Drivers of Poverty Reduction 19

36 Figure 2.1 Sectoral Composition of GDP, (%) 55 $4,5 5 $4, 45 $3,5 Percentage of GDP $3, $2,5 $2, $1,5 2 $1, 15 $ GDP per cap, PPP (const 211 $) Services value added Agriculture value added Industry value added $ Sources: National Accounts Main Aggregates Database; World Bank data. Table 2.1 GDP Growth Decomposition, by Per Capita Value Added, (%) Sector Contribution of within-sector changes in output per worker Contribution of changes in employment Contributions of intersectoral shifts Total Agriculture Industry Mining, manufacturing, utilities Construction Services Wholesale, retail trade, restaurants and hotels Transport, storage, and communications Other activities a Subtotals Demographic component n.a. n.a. n.a Total 1. Total % change in value added per capita 92.6 Sources: National Accounts Main Aggregates Database; World Bank data. Note: n.a. = not applicable. a. Other activities include financial and business services, public administration, education, health care, social protection, and other services. public administration, education, health care, social protection, and other services. This was followed by transport, storage, and communication, accounting for 18.5 percent of overall GDP growth, and wholesale, retail trade, restaurants and hotels, accounting for 13.8 percent. The boom in the service sector derived from a number of factors. An important engine was the rapid growth in high value added services such as information and communications technology, finance and insurance, and real estate. This growth took place mainly in Accra, which saw a massive inflow of capital in the last decade, but also a surge in real estate prices. The other service component is represented by those activities borderline between formal and informal that characterize West African towns: retail activities, construction, transport, and so on. 2 Drivers of Poverty Reduction

37 Increased public sector employment in education, health care, and public administration also contributed to the service boom. By contrast, the share of the industrial sectors in GDP increased, and, in 211, for the first time since independence, it was larger than the share of agriculture. The main contributor, besides gold, was crude oil production. Construction was also an important driver of GDP growth, contributing about 14 percent thanks to an increase in productivity and a positive intersectoral shift (see table 2.1). The speed and nature of structural transformation changed after 25. The share of agriculture in value added declined more quickly after 25, while the share of services remained virtually unchanged. This resulted in an expansion in the share of industry to 27 percent of GDP in 212 (see figure 2.1). The sectoral distribution of employment adjusted somewhat in line with the changes in the structure of the economy. Although agriculture remains the main sector of employment, at 43.2 percent, workers have increasingly moved away from this sector into services and, to a lesser extent, industry over the last decade. Employment in the service sector expanded from 28.8 percent in 1991 to 42. percent in 212, and the share of construction more than doubled over the two decades. The share of employment in industry also rose, although at a lower rate, from 1.7 percent to 14.9 percent (table 2.2). The shift in the sectoral composition of employment accelerated after 25. The speed at which workers moved out of agriculture nearly doubled after 25, from 1.6 percent to 3. percent a year (see table 2.2). However, unlike the shift in GDP, workers gravitated mostly toward services (4. percent a year), while industry picked up only a small share of employment. However, the major driver of growth has been within-sector increases in value added. The total contribution of the within-sector change in value added per worker explains 47.2 percent of per capita GDP growth (see table 2.1). Value added per worker in agriculture rose in because of the outflow of workers and the development of commercial farming. It also increased in services. Agriculture and services explain, respectively, 37.2 percent and 52.6 percent of the observed change in output per worker. Structural transformation alone, that is, the intersectoral shifts, accounted for less than half of GDP growth over the two decades. This is likely linked to the relatively small expansion of employment in industry and the high concentration of workers in lower-productivity service activities, where output per worker appears to have been less than in agriculture. The limited role of increased employment in explaining growth is consistent with the fact that most Ghanaians work. In 212, nearly 4 adults in 5 aged and 9 adults in 1 aged over 25 were working (figure 2.2, chart a). As in many African economies, the open unemployment rate was low, around 2 percent of the labor force, and highly concentrated amongst the better-offs. Table 2.2 Employment, by Major Sector, (% of Total Employment) Sector Agriculture Industry Mining, manufacturing, utilities Construction Services Wholesale, retail trade, restaurants and hotels Transport, storage, and communications Other activities Source: Calculations based on GLSS 3 6. Drivers of Poverty Reduction 21

38 Figure 2.2 Poverty and Access to Employment, Adults Aged 15 64, 212 a. Labor market status Total Percent Employed Unemployed Inactive Source: Calculations based on GLSS 6. Percent b. Employed, by consumption quintile Poorest Second Third Fourth Richest Employment-to-population ratio It follows that having a job is not a guarantee of escaping poverty. Indeed, the share of employed adults in the poorest consumption quintile (8 percent) is slightly higher than the corresponding share in the richest quintile ( figure 2.2, chart b). 1 Instead, it is the types of jobs that are important in poverty reduction. Four broad types of employment are of importance: employment in agriculture, non-farm self-employment, wage employment in the private sector, and wage employment in the public sector. Agriculture is dominated by low- productivity smallholder farming and is mainly rainfed. The most competitive cash crop is cocoa beans (Teal et al, 26; Vigneri, 25); Ghana controls an average of 14.5 percent of the world market and is the third-largest producer. Nonetheless, the average yield (431. kilograms per hectare in 25 12) is low compared with the top 2 cocoa producers, Côte d Ivoire (595.7 kilograms per hectare) and Indonesia (576. kilograms per hectare). Other cash crops include cotton, rubber, and tobacco, of which Ghana produces only a small share of the world output. Since independence, raising agricultural productivity and transforming agriculture from a subsistence base to a market base have been priorities of the government. Various policies and interventions have been tried to boost cash crop production for internal and, especially, international trade. But the results have been mixed, and agricultural productivity remains limited in many parts of the country because of traditional farming methods and volatility in rainfall (Molini et al. 21). Nonetheless, the poverty rate among cocoa farmers declined from 6 percent in 1991 to about 24 percent in 25 (Breisinger et al. 28). The positive performance of staple crops also helped boost consumption and reduce poverty. Figure 2.3 shows production growth in a number of key food crops between 2 and 21 using 2 as a base year. The growth in the production of rice, maize and, to a lesser extent, millet after 27 is striking: relative to 2, the production of rice was 2.5 times greater; of millet, 1.5 times greater; and, of maize, almost 2. times greater. The rise in production was also rapid in other food crops, such as yams, groundnuts, soybeans, and cassava (Mohan and Matsuda, 213). Average staple food-crop output grew much more quickly than the population, and per capita production was more than 8 percent greater in 25 7 than in Growth in higher-value vegetables and fruit for domestic and export markets was also encouraging (Breisinger et al. 28). Non-farm self-employment is dominated by household enterprises. It is the most important economic activity in urban areas and typically provides supplementary income to rural households. Despite its generally low income-generating capacity, it represents an initial opportunity to quit agriculture among low-skilled youth and 22 Drivers of Poverty Reduction

39 among adults who may not have adequate skills or experience to fill wage jobs, especially in areas where medium or large private enterprises are limited. However, some service sector activities with a high concentration of relatively low-skilled workers grouped under wholesale, retail trade, restaurants and hotels in figure 2.4 are not necessarily a ticket out of poverty because the output per worker is less than in agriculture. Private and public sector wage jobs tend to be concentrated in more well off urban areas and to attract a more highly skilled labor force Figure 2.3 Production Indexes, Major Crops Index Maize Millet Rice Sorghum Source: Ministry of Agriculture Figure 2.4 Output per Worker, by Sector, , 1, Constant 25$ 8, 6, 4, 2, Agriculture, hunting, forestry, fishing Mining, manufacturing, utilities Construction Wholesale, retail trade, restaurants and hotels Transport, storage and communication Other activities Sources: National Accounts Main Aggregates Database; GLSS 3 and 6. Drivers of Poverty Reduction 23

40 by offering greater earnings and better working conditions. The public sector offers particularly advantageous conditions for the better educated. Even after we control for observed characteristics and selectivity, wages among men appear to be notably higher in public administration than in the private sector (Ranzani and Tuccio, 215 forthcoming). This is a concern because the high returns to public employment may increase the reservation wage for employment in the private sector, thereby inducing workers to wait in line for government jobs and crowding out employment in potentially more productive activities. The type of job done by the household head is a good predictor of poverty status. In 212, over 42 percent of the employed workforce relied on low-productivity farming for income. Another 34 percent were working for themselves or their families in household enterprises in the nonfarm sector. The wage sector remained small, accounting for about 24 percent of the workforce, mostly in the private sector. In the north and east, agriculture is the main source of income, accounting for between 54 percent of income in the Volta Region and 73 percent in the Upper West Region (figure 2.5). The type of employment greatly influences consumption levels. Agriculture was by far the dominant economic activity among the bottom 2. By contrast, agriculture accounted for only 14 percent of jobs among the heads of households in the top 2, 42 percent of whom are wage employees. The recent growth in household consumption has been associated with a considerable growth in non-agricultural private employment. Agriculture has given way to non-agricultural self-employment and, to a lesser extent, private sector wage employment. Between 1991 and 212, the share of workers in nonwage and wage work outside agriculture expanded in about equal measure: from 26 percent to 34 percent in self-employment and from 15 percent to 24 percent in wage work. There was also a significant shift from public sector to private sector wage employment. The share of private sector wage employment tripled, from 6 percent to 18 percent, while the share of public sector wage employment fell, from 9 percent to 6 percent. As a result of these trends, the share of agriculture in employment is now lower, and the share of non-farm employment employment are now higher than the average among lower-middle-income countries in Africa (where it reaches 55, 31, and 14 percent, respectively) (Filmer and Fox 214). However the magnitude of the sectoral shift in employment differed significantly across regions. In Greater Accra, the only region in which agriculture had already contributed little to employment in 1991, private sector wage employment expanded significantly at the expense of both Figure 2.5 Distribution of Employment, by Sector, Region, and Poverty Status a. By sector and region b. By sector and consumption quintile GHANA Upper West Nothern Upper East Brong Ahafo Volta Central Eastern Western Ashanti Greater Accra Richest Fourth Third Second Poorest Percent Percent Wage, private Wage, public Self, non-ag Self, ag Source: Calculations based on GLSS Drivers of Poverty Reduction

41 public sector employment and self- employment. It now accounts for 42 percent of all jobs. Other southern regions (the Ashanti, Central, Eastern, and Western regions) witnessed a considerable shift away from agriculture toward wage employment, but the shift was smaller there than in the remaining regions of the north (the Brong Ahafo, Northern, Upper East, Upper West, and Volta regions). A More Skilled Workforce Building a skilled workforce is an accumulative process that depends on many years of good-quality schooling and the creation of job-relevant skills. Good-quality basic education is a fundamental requirement of skills development. Nonetheless, skills that are relevant to labor markets are not synonymous with education. Skill is the ability to perform some function (or specific job) because of one s knowledge (which may be acquired through education), but also practice and aptitude. Yet, quality education is an important stepping-stone to the acquisition of most job skills because it develops foundational skills such as literacy and numeracy, builds non-cognitive skills, and familiarizes students with learning situations (Adams, Johansson de Silva, and Razmara 213). The labor force has become better educated over the past two decades. Between 1991 and 212, the share of the labor force without schooling fell by almost half, from 41 percent to 24 percent. In 212, the majority of workers, 52 percent, had completed at least lower-secondary education, compared with 39 percent in However, regional disparities persist ( figure 2.6). Despite some improvements, the Northern, Upper East, and Upper West regions still have a basic education completion rate under 25 percent, while, in Accra, 4 percent have senior-secondary or tertiary degrees, compared with 7 percent in The share of the workforce with tertiary education was practically nonexistent in 1991, but has now reached about 8 percent. Education is an important correlate of job opportunities. The educational attainment of the workforce has increased across all jobs types, though differences in educational attainment across categories of work have also become more accentuated. Over 5 percent of public sector wage workers have tertiary education, compared with less than 2 percent among the non-farm self-employed (figure 2.7, chart a). More than half of the workers with tertiary education are employed in the public (wage) sector; another Figure 2.6 Labor Force with at Least Basic Education, by Region, 1991 and Percent Greater Accra Ashanti Western Central Eastern Brong Volta Ahafo Upper East Upper West Nothern GHANA Source: Calculations based on GLSS 3 and 6 Drivers of Poverty Reduction 25

42 3 percent in the private (wage) sector; and only 14 percent in non-farm self-employment ( figure 2.7, chart b). Lack of education has acted as a barrier to occupational mobility, but workers with primary education or above have been increasingly entering non-farm employment, first as self-employed and then as wage workers. Urbanization and Agglomeration The spatial patterns of poverty reduction and consumption growth in recent years suggest that location of residence is a major correlate of the risk of poverty and that geographical mobility may be an effective strategy among individuals to exit poverty. Our analysis above highlights the long-established spatial patterns of poverty that the structural transformation and economic growth of recent decades seem to have reinforced. The patterns appear closely linked to a widening gap in employment opportunities. Moving to more rapidly developing areas in the south has been a strategy for escaping poverty. The urban population has increased rapidly, mostly because of internal migration. Around 1.5 million people flowed into urban areas in ; 1.9 million in ; and 4.7 million in (figure 2.8). By 212, Ghanaians were equality split between urban and rural areas. This represented a considerable shift relative to 1991, when 7 percent of the population was living in rural areas. Since 1991, Accra and Ashanti have gained over 2.4 million inhabitants each, around half of them in the last decade. Migration is not the only possible explanation for the growth of the urban population. There may also be other demographical phenomena at play, for example a fast increase of household size in urban areas. However, data on the birthplace of household heads in rounds 5 and 6 of the GLSS and in related analysis support the rural-urban migration hypothesis (Molini, Pavelesku and Ranzani, forthcoming) (figure 2.9). Urbanization is likely to have facilitated the process of structural transformation. Sustainable economic transformation entails important socioeconomic changes correlated with urbanization. For example, urbanization has enabled higher-quality education to reach a larger share of the population. Larger urban areas have an advantage over small towns and rural areas in their ability to Figure 2.7 Education and Employment Are Strong Correlates, 1991 and 212 a. Highest educational attainment, by sector b. Sector share of total, by educational level Percent Percent Wage, public Wage, private Self, non-ag Self, ag None Primary Junior sec Senior sec Tertiary None Primary Junior sec Wage, public Self, non-ag Senior sec Wage, private Self, ag Tertiary Source: Calculations based on GLSS 3 and Drivers of Poverty Reduction

43 Figure 2.8 Variation in Population, by Rural and Urban Area and Region, Rural Urban Western Central Greater Accra Volta Eastern Ashanti Brong Ahafo Northern Upper East Upper West Western Central Greater Accra Volta Eastern Ashanti Brong Ahafo Northern Upper East Upper West 1, 5 5 1, 1, Number of individuals in thousands Source: Calculations based on GLSS 3 6. Figure 2.9 Household Heads Who Were Not Born in Their Current Places of Residence 1,, 8,, Number of individuals 6,, 4,, 2,, Greater Accra Western Urban Eastern Urban Central Urban Volta Urban Ashanti Urban Brong-Ahafo Urban Northern Urban Upper West Urban Upper East Urban Source: Calculations based on GLSS 5 6. Drivers of Poverty Reduction 27

44 provide greater access to secondary and tertiary educational institutions and higher quality teaching and support services. Urban centers in Ghana have thereby improved human capital, roughly doubling and tripling the shares of their populations with secondary and tertiary education, which, in 2 1, grew from 9.3 percent to 19.3 percent and from 2.3 percent to 7.6 percent, respectively. The two regions in which Ghana s two largest cities are located, Ashanti (Kumasi) and Greater Accra (Accra), experienced the greatest improvement in secondary educational attainment in the country over the period. Thus, urbanization may have advantages beyond the increased welfare among migrants by boosting the prospects for inclusive growth and the reduction of inequalities of opportunity. Although the recent urbanization has likely contributed to the reduction in poverty, it has also raised challenges. Often, rapid population shifts from rural to urban centers are associated with the uncontrolled expansion of the urban centers (Agyei-Mensa and Owusu, 21), including in slums, if the provision of housing and basic services is inadequate. If the supply of services cannot meet the growing urban demand and if urban economies do not generate sufficient job opportunities, slums rise up, leading to declining health outcomes, growing poverty, and greater insecurity. These and other challenges will emerge if urbanization continues apace without changes in the country s current policies and institutional structures. The Main Drivers of Poverty Reduction and Consumption Growth Which of the above developments were the main drivers of Ghana s recent success? Above, we describe three major developments of the last two decades that may have played a key role in reducing poverty and increasing consumption: structural transformation, the educational achievements of the labor force, and effective urbanization. Below, we use econometric analysis to quantify the extent to which each of these factors has contributed to poverty reduction in Ghana and, thereby, to identify the main drivers. The analysis proceeds in two stages. First, we analyze the determinants of poverty status using a limited dependent variable model (probit) in which the dependent variable is binary with a value of for the non-poor and 1 for the poor and the explanatory variables are standard correlates of household consumption grouped in five broad categories: household characteristics, location, education and type of job of the household head, and an index of access to infrastructure 3 (see Appendix B for a detailed list and the detailed results). We then analyze the determinants of changes in consumption at the 2th, 4th and 6th percentiles of the consumption distribution using unconditional quintile regressions and the same explanatory variables (Fortin et al., 211). 4 In 1991 households in the 6th percentile where around the poverty line but by 212 their consumption level was well above the poverty threshold. Households in the 4th percentile were poor in 1991 but had progressively moved out of poverty by 1998 although they remained vulnerable to falling back into it. Finally the 2th percentiles includes those households who remained poor over the whole period. The educational attainment of the household head and the jobs she does are a main determinant of poverty status. The education of the household head and the job she does are important factor in determining the probability of the household been poor. 5 For example in 212, as detailed in Table B.1 in appendix B, households headed by somebody with tertiary (secondary) education were 2 (9) percent less likely to be poor than identical households with an uneducated head. Likewise, having a head that works outside agriculture reduces the probability of the households being poor significantly and, if the head works in the public sector, the household is 13 percent less likely to be poor than an identical one headed by a farmer. Urban households are 22 percent less likely to be poor than their rural counterparts and the probability of being poor is lower for smaller households 28 Drivers of Poverty Reduction

45 (4 percent) and those with better access to infrastructures (9 percent). Ghana s recent success with poverty reduction is driven by a combination of improvements in the characteristics of the households and higher returns to these characteristics. Using a standard Oaxaca-Blinder (OB) decomposition, we can decompose the change in poverty occurred between different periods into the coefficient effect (changes in returns on variables), the endowment effect (changes in mean characteristics), and a residual. The results, presented in figure 2.1, show that the rapid improvement in the characteristics of households are as important in explaining the poverty reduction occurred between 1991 and 212 as the increase in the returns to those characteristics. However, there were significant differences across sub- periods. Increases in returns were particularly important between 1991 and 25, as the growing economy was putting a high premium on skills and the returns to having access to infrastructure and to being in non-agriculture employment were high. By contrast, between 25 and 212 poverty reduction appears to have been driven mainly by investment in improved endowments while the returns to those characteristics declined. 6 This was probably due to the limited capacity of the modern urban sectors to absorb the increasing educated labor force and to take full advantage of the improved endowments. The drivers of poverty reduction were also the engines of consumption growth at different points in the distribution. The results of the unconditional quantile regressions (given in details in Table A2-5 in Appendix) confirm that the drivers of poverty reduction were the same as those that led to higher consumption across all points in the distribution. Three interesting trends emerge regarding the role of the education attainment of household heads. First, during the entire period, and for all percentile analyzed, the returns to education were positive and increasing with the level of education. Second, for the 4 th and 6 th percentiles, returns on primary and secondary education flattened over time (or declined as for primary in 6 th percentile) but they grew very fast for the 2 th percentile between 25 and 212. Finally, the gap between returns to higher education and those to the other levels grew substantially for the 4 th and 6 th percentiles. These findings suggests that investment in primary and secondary education remain key to increasing consumption of the poor (bottom 2 th in 212), who still have comparatively low educational attainments. By contrast, for the better-off only investments in higher education continue to pay as the poor Figure 2.1 Decomposition of Poverty Changes, Variation in poverty rate Endowments Coefficients Interaction Source: Calculations based on GLSS 3 6. Drivers of Poverty Reduction 29

46 quality of education, and the inability of the economy to absorb the fast growing supply of more educated, have muted the return to lower educational attainment. Employment outside agriculture had a positive impact on consumption at all levels of the distribution. In 1991, when household heads where predominantly working in agriculture, the returns to a non-agriculture job were high across all percentiles (figure 2.12). The returns started to decline between 25 and 212, and became mostly insignificant, possibly as a reflection of the inability of the economy of absorb the workers moving out of agriculture in higher productivity activities. The reduction in returns was particularly noticeable for the employment in the public sector, which declined sharply after 25 for both the bottom and the top 2 percent. However, non -agricultural self-employment was the only activity for which returns remained unchanged over the entire period, especially for the bottom 2. However, the focus of our analysis on household heads only may underestimate the full extent of the returns to sectoral transformation, as the diversification in the household portfolio of activities may affect the household head only in its final stage. The role of urbanization as a driver of poverty reduction is confirmed by the high returns to living in urban areas. These returns are significant across all percentiles (See Tables 2/5 in Appendix B). For the 4th and 6th percentiles, the coefficients follow a similar pattern: they declined between 1991 and 1998, surged rapidly between 1998 and 25, and eventually declined again in the last decade. Again, these trends may be closely linked to the rise in the number of households migrating from rural areas between 25 and 212 which progressively raised the opportunity costs of moving to urban areas, while reducing the net returns. This confirms that structural transformation, changes in the labor market, and urbanization are inextricably linked. The results on the infrastructure index and the regional dummies confirm that living in dynamic areas and having access to basic infrastructures significantly improve consumption Figure 2.11 Education Coefficients, by Quintile and Year Coefficient Survey year Source: Calculations based on GLSS 3 6 Primary education Secondary education Higher education 3 Drivers of Poverty Reduction

47 Figure 2.12 Employment Category Coefficients, by Quintile and Year Coefficient Survey year Private employee Public employee Non-agricultural self-employed Source: Calculations based on GLSS 3 6. Figure 2.13 Decomposition of Household Consumption over over over Total difference P2 P4 P6 P2 P4 P6 P2 P4 P6 Endowments Coefficients Interaction Source: Calculations based on GLSS 3 6. Drivers of Poverty Reduction 31

48 for all percentiles. Using as the baseline one of the poorest regions in the country, the Upper West Region, we find that all other regions tended to perform consistently better during the entire period (See Tables 2/5 in Appendix B). Overall, the coastal regions and Ashanti fared better than the regions of the north. The coastal regions are also the most highly urbanized, and the urban area coefficient became positive and significant in 25 across all percentiles and remained positive and (often) significant in the following survey round. The combination of improved characteristics and higher returns to these characteristics were also the main drivers of increases in consumption for all Ghanaians. The results of the OB decomposition for the entire consumption distribution, reported in Figure 2.13, confirm the findings of the analysis based on growth incidence curves (reported in Chapter 2) that consumption grew more quickly for the highest percentiles. 7 They also suggested that the difference in growth across percentiles was due mostly to the higher rates at which the better-offs accumulated endowments during the period. While the returns to characteristics changed in very similar ways for all Ghanaians, the households characteristics in the 6 th percentile increased faster. Amongst the characteristics, infrastructures, location and household composition 8 were Notes 1. Given variations in the structure and implementation of the GLSS, efforts have been made to construct comparable labor market indicators across surveys using the approach developed by the International Labour Organization. Essentially, the employed are defined as those people who worked for pay or profit during the seven days prior to the survey(s), while the unemployed are people who did not work, though they wanted to and were actively trying to find a job or start a business. The remaining population is defined as inactive. According to the approach of the International Labour Organization, unpaid household duties are not counted as employment. As shown, for instance, in Tanzania, this may lead to underreporting of economic activities, especially with respect to women who do unpaid work such Figure 2.14 Decomposition of Changes in Characteristics Total difference th percentile Household composition Infrastructure index Educational attainment Source: Calculations based on GLSS th percentile th percentile Location Employment category those that drove the changes. As illustrated by Figure 2.14, the key factor in changes in the consumption of the bottom 2 th is the movement towards richer areas. For the 4 th percentile location remains important but access to infrastructure becomes more important and change in household composition also matters. The drivers of change for the 6 th percentile are similar to those of the 4 th but changes in educational attainment play a bigger role. as cultivation, raising livestock, and so on. (See Bardasi et al. 21). 2. Agriculture in Ghana, Facts and Figures, Ministry of Food and Agriculture of Ghana, December The infrastructure index is obtained combining using principle component analysis four variables: access to protected water, access to electricity, access to protected sanitation and access to safe sources of cooking. 4. The unconditional quantile regression methodology is based on regressions in which the dependent variable is a transformation (the recentered influence function ) of the outcome variable. While conditional quantile regression allows to recover the impact of a small location shift in the distribution of a variable of interest at quantiles (different points) of the conditional distribution of 32 Drivers of Poverty Reduction

49 the dependent variable that is, given the distribution of the variable of interest, unconditional quantile regressions allow estimating the same impact for the entire (unconditional) distribution of the dependent variable. 5. Negative coefficients indicate less poverty. 6. When comparing the marginal effects (Table 1, Appendix) of education, labor, infrastructure and locality between 25 and 212, those of 212 are systematically smaller with the notable exception of secondary education. Private and public sector dummies, although with the right sign, become insignificant. 7. Results are reported in Appendix B, table B6. The interaction term is not reported. 8. The variables estimated in the quantile regression have been grouped in five groups, Household characteristics, location, infrastructure index, employment category and educational attainment. See Tables B2-B5 in Appendix for details. Drivers of Poverty Reduction 33

50

51 Chapter 3 Remaining Challenges Despite Ghana s success in reducing poverty and promoting shared prosperity, challenges remain. This chapter describes three of these challenges: the growing inequality and polarization in household consumption, the large spatial disparities, and the deteriorating macroeconomic environment. Consolidating the progress made in poverty reduction and shared prosperity over recent decades will require that these challenges be addressed promptly and effectively. Growing Inequality and Polarization in Household Consumption Inequality in household consumption widened substantially between 1998 and 25. The picture that emerges is consistent across different indicators of inequality (table 3.1). Inequality in household consumption was initially constant, but widened considerably between 1998 and 25 a jump of about 9 percent in the Gini coefficient and 2 percent in the Theil index. The consumption share of the poorest quintile of the population (the bottom 2) also declined steadily between 1991 and 25 (from 6.8 percent to 5.7 percent) while the share of the top 2 increased slightly (from 44.8 percent to 46.6 percent). Inequality has remained constant at the higher level after 25 but the trends in the share of consumption of the bottom and top quintile has continues in the same direction. What is behind the changes in inequality? Our econometric analysis in chapter 2 points to four main drivers of household consumption: educational attainment, structural transformation, spatial transformation, and demographic changes. To quantify the relative importance of these factors in determining the overall level of inequality, we decompose the Gini index into three components for each household characteristic: one reflecting cross-household variations in the characteristics (the between component, represented in Figure 3.1 by the red blocks), one measuring differences in consumption across groups of households with the same characteristics (the within component, represented in Figure 3.1 by the blue blocks), and an interaction or overlap term (represented in Figure 3.1 by the green blocks. 1 In the first chart in figure 3.1, for example, the red component of the bars (the between component) represents the share of inequality that is accounted for by differences in educational attainment among household heads. The blue component of the bars (the within component) shows the share of inequality arising because of differences in consumption across households the heads of which have the same educational attainment. The green component of the bars shows the residual of the decomposition. Differences in household characteristics generally account for only a small part of inequality. The relative large red (between) block in the relevant graphs in Figure 3.1 shows that household differences in the region of residency, the education attainment of household heads and the type of job she does (non-agriculture) are important determinants of consumption inequality. However, the fact that blue (within) component is also large suggest that a large degree of inequality persisted across groups of households. Inequalities among regions played a particularly important role, accounting for an average of 4 percent of the total Gini over the period, while within regions inequalities were responsible for only 1 percent of total inequality. By contrast, the heterogeneity in consumption was much greater within the rural and urban areas and among households headed by Remaining Challenges 35

52 Table 3.1 Measures of Per Capita Household Consumption Expenditure, Measure Gini Theil Mean Median Consumption shares Bottom 5 percent Bottom 1 percent Bottom 2 percent Top 2 percent Top 1 percent Top 5 percent Source: Calculations based on GLSS 3 6. Figure 3.1 Decomposition of the Gini Index, by Household Characteristic, Agriculture/non-agriculture Education level Percent 6 4 Percent Region Rural/urban Percent 6 4 Percent Overlap Between Within Source: Calculations based on GLSS 3 6. non- agricultural workers. A comparison of the components across periods also shows that the link between inequality, on the one hand, and education and non-agricultural employment, on the other, increased (the relative size of the red block has grown), while the link of inequality with location declined slightly. Finally, despite the increase in the 199s, inequality remains narrow in Ghana compared with other African countries. Figure 3.2 ranks several African countries according to the average Gini index over the last 2 years. In 1991, Ghana was in the bottom 2 percent of the Gini distribution in Africa, and, despite some 36 Remaining Challenges

53 deterioration, in 212, it was still below the median and among the lowest among the rapidly growing African economies. However, the narrative about inequality is more nuanced than the summary measures suggest. The summary measures of inequality analyzed above only partially capture the changes at various points of the consumption distribution. Analysis of the incidence curve provides more detailed information on the changes occurring at all points of the distribution (see chapter 2). The results of a simple interquantile analysis can complement the analysis of the incidence curve (table 3.2). They show that the ratio of average consumption among the top 1 percent of the distribution (the top 1) to the average consumption among the bottom 1 percent (the bottom 1) had risen considerably even before 1998, suggesting that the more well off had benefited more than the poorest decile from the economic growth in Over the years, the consumption levels of the top and the bottom of the distribution Figure 3.2 Africa 7 6 GHA 12 GHA 91 2 Source: Calculations based on PovcalNet. Gini Indexes in Sub Saharan Percentiles of Gini in SSA continued to diverge at a steady rate so that the gap expanded by 3 percent over the full period. The divergence was widening because the bottom 1 was being left behind, rather than because the top 1 was gaining disproportionally compared with the rest of the population. The average consumption of the 9th percentile rose little relative to the median, the 5th percentile, while the average consumption of the bottom 1 had deteriorated by nearly 2 percent by 25. The bottom 1 appears to be losing ground also compared with other households in the bottom 25, who are also losing ground to the median but only half as quickly. This findings are in line with the clear evidence indicating a rise in polarization. A detailed analysis of polarization allows us to zoom in on the dynamic changes occurring at different points of the consumption distribution. 2 The results of this analysis suggest that the distributional changes observed in hollowed out the middle of the Ghanaian household consumption distribution and increased the concentration of households around the highest and lowest deciles. Like inequality, polarization began to increase in the late 199s, but continued to grow even after inequality stabilized, although at a slower pace. We measure the degree of polarization of Ghanaian society using the relative distribution method. 3 This method involves the creation of a distribution that captures the share of households in the comparison year (212) that falls within each income decile of the reference year (1991). This distribution is then decomposed into a location effect, which tells us if there is a change in the median (or mean) of the income distribution, and a shape effect, which, representing the relative distribution, net of the location change, is useful Table 3.2 Interquartile Consumption Ratios, by GLSS Wave, Year p9/p1 p9/p5 p1/p5 p75/p25 p75/p5 p25/p Source: Calculations based on GLSS 3 6. Remaining Challenges 37

54 in isolating movements (redistribution) between the reference population and the comparison population. The generalized increase in consumption meant that, in 212, households were crowded around the top deciles of the 1991 distribution. Figure 3.3 shows the share of households in 212 that fall into each decile of the 1991 distribution. It is obvious from the figure that the share of households with consumption levels equivalent to the consumption levels of the upper deciles in 1991 rose dramatically throughout the two decades, while the shares with consumption levels equivalent to the consumption levels of the bottom and around the middle in 1991 declined. However, the crowing of the population in the top 1991 decile masks growing polarization. To gauge this, we decompose the changes in the relative density into the changes arising because of the increase in median income (the location effects) and the changes arising because of changes in the distribution (the shape effect). The latter are those changes produced by the changes in distribution, net of the impact of the increase in median consumption (figure 3.4). The polarization of households around the bottom 1 and the top 1 is striking. Without the rise in median income, the greater dispersion of consumption expenditures would have led to relatively more low-consumption households in 212, and this effect was mainly concentrated among the bottom 1. By contrast, at the top of the distribution, the effect of higher median income reinforced the polarization in the distribution, which, of itself, would have boosted the share of households in the top 1 by nearly 16 percent. In sum, once we net out changes in real median expenditure, we observe a U-shaped Figure 3.3 Relative Consumption Distribution, Relative density Reference proportion Source: Clementi, Molini, and Schettino 215a. Figure 3.4 Median-Adjusted Relative Consumption Distribution, Relative density Reference proportion Source: Clementi, Molini, and Schettino 215a. 38 Remaining Challenges

55 Figure 3.5 Median-Adjusted Relative Consumption Distribution Series, Ghana, Relative density Reference proportion Wave Source: Clementi, Molini, and Schettino 215a. relative density, indicating that polarization was hollowing out the middle of the household consumption distribution. The level of polarization has been growing. For each wave of the GLSS between 1991 and 212, figure 3.5 shows the shape effect of the household consumption relative density; 1991, as the reference sample, is constructed to be flat. 4 Following the plot through each successive wave, we find that the shares of households at both the top and the bottom tails of the consumption distribution rose consistently over the period, while the shares in the middle declined. Compared with 1991, more households had dispersed from the central deciles toward the lowest and the highest tails by 25. In 212, the hollowing out of the middle continued, but at a slower rate: both the 1st and the 1th percentiles were above their respective values in 25 and, of course, far above the 1991 reference levels. Persistent Spatial Disparities Our descriptive profile and econometric analysis suggest that geography has been an important correlate of the patterns of poverty and inequality in Ghana (see above and chapters 2 and 3). Two spatial trends are evident. As in 1991, poverty is mainly a rural phenomenon, and the lion s share of the poor are located in the north. Through the period , 8 percent of the poor lived in rural areas. The increasing concentration of the poor in the north is also quite clear; the highest poverty rate in the country is in the Upper West Region (in the north). Moreover, while, in 1991, 25 percent of the poor were living in the north, the share had increased to about 4 percent by 212, despite the stabilization in the population share of the north at around percent. However, great heterogeneity exists even within regions. The better performance of urban areas is driven mostly by the striking progress of the Ashanti and Greater Accra regions. Greater Accra has enjoyed the lowest poverty rate in the country since at least 1991, and the rate has continued to decline there. Although the number of the poor rose considerably in 25 as result of mass migration from surrounding poorer areas, the poverty rate fell by half in Greater Accra between 25 and Among the coastal and forest regions, Greater Accra is the only one that had reduced both the poverty headcount and the number of the poor by 212. Ashanti, the most heavily populated and the second most urbanized region in the country, has experienced a more stable pattern of poverty reduction over the last 2 years. There, the poverty rate continued to decline, and the decline accelerated between 1998 and 25, although the number of the poor began to grow between Remaining Challenges 39

56 25 and 212. Ashanti was the only region in which both the urban and rural populations rapidly expanded. Unlike Greater Accra, half the population of Ashanti lives in areas classified as rural. Between 1991 and 212, the Brong Ahafo and Western regions enjoyed the most rapid poverty reduction in the country. The poverty rate dropped from about 6 percent to less than 25 percent. A key driver of this outcome was the boom in cocoa production, which is highly concentrated in these regions. However, the urbanrural divide remains large: poverty rates in the urban areas of the two regions were around 1 percent, compared with 2 percent (the Western Region) and 3 percent (the Brong Ahafo Region) in rural areas. Moreover, after a long positive trend, the fall in the poverty headcount halted in 212, and the absolute number of the poor started to rise in the Western Region, despite the expected positive spillovers from the discovery of oil there. All regions bordering Greater Accra, but especially the Central and Eastern regions benefited from the growth of the capital city, where a large share of the population in these regions work. In the Central and Eastern regions, the poverty rate and the absolute number of the poor fell by more than half between 1998 and 25. In the subsequent period, poverty reduction stagnated, and, in the Eastern Region, the absolute number of the poor began to rise, albeit marginally. Both the poverty rate and the number of the poor were stubbornly high in the north until 25, but have declined somewhat since. Between 1998 and 25 poverty rates fell in the north too. Nevertheless, due to the rapidly growth in population, the absolute number of the poor rose in all three regions in the north (see Figure 1.8). In 212, both the poverty rate and the absolute number of the poor declined across these three regions. In the Upper East Region, the share of the poor was cut by almost half. Nonetheless, the poverty rate was still high. More than 4 people in 1 were poor in the Upper East Region (44.4 percent), increasing to 5 in 1 in the Northern Region (5.4 percent), and 7 in 1 in the Upper West Region (7.7 percent). In 25 12, the pace of poverty reduction slowed in those regions that had previously witnessed considerable poverty reduction. In , poverty reduction was driven by the progress made in the southern and central regions. By contrast, between 25 and 212, Greater Accra and the three regions in the north drove most of the change. In the southern and central regions, poverty rates declined slightly, but the influx of people from poorer areas offset this reduction, and the net effect was a slight increase in the number of the poor. To understand fully the extent of the spatial inequalities in Ghana, one should deepen the with-region analysis because heterogeneities exist also within regions. Poverty maps allow a better focus on the spatial distribution of poverty at the local level (Elbers et al, 23). Although the number of districts (11 and 216) differ across 2 and 212, the two years shown in map 3.1, the visual interpretation is still meaningful. Map 3.2 adds two more indicators relevant to welfare. In 2, poverty rates above 2 percent were the norm in Ghana. Only a few districts in the Ashanti and Greater Accra regions showed poverty rates below 2 percent. Districts in the Ashanti, Greater Accra, and Western region and in the Eastern Region neighboring Accra, typically had poverty rates at between 21 and 4 percent. Rates in the regions in the center of the country mostly exceeded 4 percent. There was a rather homogenous central belt stretching across almost all districts in the eastern Ashanti, Brong Ahafo, Eastern, and Volta regions in which poverty rates never fall below 4 percent. Part of the Northern Region bordering the Brong Ahafo and Volta regions showed similar characteristics, while practically all districts in the Upper East and the Upper West regions recorded poverty rates well above 4 percent. At the extreme, a group of Upper East and Upper West districts bordering, respectively, Togo and Burkina Faso showed rates above 8 percent. 4 Remaining Challenges

57 Map 3.1 Poverty Maps, 2 and 212 Source: Calculations based on GLSS 4 and 6. Map 3.2 Additional Indicators of Welfare, 211 and 212 Source: Calculations based on GLSS 6. Remaining Challenges 41

58 BOX 3.1 Poverty Maps in Ghana The GSS has produced two comparable poverty maps. One, issued in May 25, is based on data of the 2 Population and Housing Census and the GLSS (round 4). It shows 11 districts. The more recent one has been computed based on data of the 21 Population and Housing Census and the GLSS (round 6). It illustrates 215 districts. Both poverty maps have been created based on the small area estimation methodology that has been developed to allow accurate estimates of consumption-based poverty and inequality at lower levels of disaggregation by combining information from censuses and household consumption surveys. The country was radically different 13 years later. In the corner formed by the Ashanti, southern Brong Ahafo, Eastern, Greater Accra, Western, and coastal Volta regions, poverty rates had typically declined to below 2 percent in 212. There were clusters of districts with higher poverty rates, particularly in the inland parts of the Central Region, but, overall, the improvement is striking. The central belt is now much more heterogeneous, albeit most districts have poverty rates at less than 4 percent, mainly in the Brong Ahafo and Volta regions. In the north, the difference in performance between the eastern and western portions of the area is apparent. In the east, the districts belonging to the Northern Region reduced poverty to less than 4 percent, while most of the districts belonging to the Upper East region, which were among the poorest in 2, reduced the poverty rate to below 6 percent. By contrast, in the west, particularly in the Upper West Region, poverty rates remained mostly above 6 percent, although with large within-region disparities, from a low of about 36 percent in Wa Municipal District to approximately 84 percent in Wa East District and over 9 percent in Wa West District. It is noteworthy that the two poorest districts in the Upper West Region border the least poorest district, a fact that would not have been apparent without a poverty map. The poverty maps also reveal islands of poverty and prosperity. The poverty rate in Adaklu District (89.7 percent) in the Volta Region is more than two and half times the regional average (33.8 percent). Greater Accra has the lowest regional poverty rate in the country, and poverty there is concentrated in two districts, Ningo Prampram District (31.2 percent) and Shai Osudoku District (55.1 percent). In the Northern Region, the headcount rate is around 5 percent, but two districts have poverty rates below 3 percent, Tamale Metropolitan District (24.6 percent) and Sagnarigu District (29.3 percent). The incidence of poverty appears highly correlated with proximity to roads in hours and the yield of the maize crop. A comparison of the 212 poverty map (map 3.1, part b) and the map of the proximity to roads in hours (map 3.2, part a) a good proxy for market connectivity and ease of access to basic facilities such as hospitals and schools shows that proximity to roads is strongly associated with low poverty rates. The south and west of the country,where poverty is below 2 percent in most districts, also have the best road networks; there, the average distance to the nearest road is less than an hour. By contrast, in the north, the average distance to the nearest road is greater, as are the poverty rates. The map of the yield of the maize crop, the most common staple in the country, has some similarities with the poverty map, too (see map 3.2, part b). In areas of the Eastern Region 42 Remaining Challenges

59 and the inland parts of the Central Region where poverty was reduced rapidly, maize yields are the highest in the country. By contrast, in the highpoverty north, maize yields are low. These are areas that are only modestly urbanized and in which cash crops are not so diffused. There, low levels of production seem to be associated with low productivity in the cultivation of maize and, consequently, low levels of consumption. Meanwhile, in southern Ghana, high maize yields are not always associated with low levels of poverty. Often, the contrary is true. Along the coast, poverty shrank, but this is not an area suitable for maize cultivation. Likewise, highly urbanized areas such as the Ashanti Region registered rapid poverty reduction, but not because of higher crop productivity. In the Western Region, there was rapid poverty reduction and higher productivity in crops other than maize. The Deteriorating Macroeconomic Environment Ghana has suffered a number of serious external and internal macroeconomic shocks since 212. Major external shocks included the rupture of the West African natural gas pipeline in 212 and the highly volatile export price of gold. The pipeline supplies Benin, Ghana, and Togo with natural gas from Nigeria. Ghana was thus forced to increase oil imports for the generation of electricity, causing the import bill to rise dramatically, to approximately $27 million a month. 6 The increase in the cost of oil imports was partially offset by the rising export price for gold. However, by 214, global gold prices had tumbled and could no longer offset the higher cost of oil imports. Furthermore, the rising oil imports pushed the current account deficit to an average 11.5 percent of GDP in , and it remained high, at 9.2 percent, in 214. Currency exchange rates depreciated the Ghanaian cedi by 35 percent against the U.S. dollar on the interbank market and by 43 percent on the foreign exchange market. Expansionary fiscal policies increased inflationary pressures. Overall inflation rose sharply, from 8.8 percent in 212 to 17. percent in 214, and nonfood inflation shot up from 11.6 percent to 23.9 percent. The producer price index climbed by around 17.1 percent in 212 before jumping to 35.8 percent in 214. The inflationary impact of central bank financing was aggravated by rising public sector wages, the pass-through effects of rising fuel and utility prices, and the depreciation of the cedi, which effectively boosted import prices. In an effort to stabilize the economy, the government prepared and adopted a multiyear plan aimed at reducing the fiscal deficit. It took important steps to address fiscal imbalances, including the elimination of subsidies on fuel products and utilities. In 213, as crude oil prices rose and the cedi depreciated, it began to pass a larger share of the higher cost of energy production on to consumers, which severed the link between commodity price volatility and fiscal accounts. In February 213, administratively set gasoline and diesel prices increased by 2 percent, and, by December 214, gasoline and diesel prices had risen by 1 percent, while liquefied petroleum gas prices rose by 128 percent over the same period. Median electricity prices skyrocketed by 16 percent between October 213 and December 214, and water prices climbed by 8 percent. This adjustment, combined with the rapid increase in inflation and the depreciation of the cedi, had a strong impact on household consumption in urban areas other than Accra and in the rapidly growing south. Households faced a sharp fuel price rise and, as a consequence of high inflation and capital outflows, a strong cedi devaluation (Clementi et al., 215b). The fuel price rise affected the cost of many other items, including the prices of nonfood items, which increased the most (figure 3.9). These developments hit urban households particularly hard because they spend a larger share of their budgets on fuel and imported items and are typically net food consumers. Probably because of the increase in transport costs, the impact was Remaining Challenges 43

60 Figure 3.6 Regional Price Indexes: Total and Nonfood, 213 a. Total b. Non-food Accra_Sep Accra_Sep Accra_Jan=1 95 Accra_Jan= Western Central Volta Eastern Ashanti Brong Ahafo Northern Upper East Upper West Source: Calculations based on the consumer price index. Jan_213 Sep_ Western Central Volta Eastern Ashanti Brong Ahafo Northern Upper East Upper West especially severe in urban areas other than Accra. During the first nine months of 213, prices grew relatively more in the Ashanti and Western regions than in Accra, and the price differential was appreciably larger for nonfood items. By contrast, the living standards in rural areas and in the north where about 2 percent of food consumption relies on self-production Notes 1. For more details on the decomposition, see Bhattacharya and Mahalanobis (1967) and Rao (1969). 2. Whereas inequality is a measure of the overall dispersion of the distribution and refers to the distance of every individual from the median or the mean income, polarization captures the combination of divergence from the global mean income and convergence toward local mean incomes. For a detailed explanation of the concept of polarization and the techniques used to measure the extent of polarization, see Clementi et al. (215); Clementi, Molini, and Schettino (215a). 3. Techniques based on the relative distribution method developed by Handcock and Morris and the nonfood component represents around 6 percent of consumption were more resilient to these short-term macroeconomic cycles (Fox 215). The spike in prices in the regions that have been the backbone of Ghana s economic success raises concerns about the country s future growth prospects. (1998, 1999) enable the counterfactual comparison of compositionally adjusted distributions. Relative distribution methods can be applied whenever the distribution of some quantity across two populations is compared either cross-sectionally or over time. For our purposes, the relative distribution is defined as the ratio of the density in the comparison year to the density in the reference year evaluated at each decile of the consumption distribution; it can be interpreted as the share of households in the comparison year s population that fall into each decile of the reference year s distribution. This allows us to identify and locate changes that have occurred along the entire Ghanaian household consumption distribution. 44 Remaining Challenges

61 4. This analysis uses relative distribution methods that allow us also to analyze how redistribution occurred across households over the entire period. For more details, see Clementi, Molini, and Schettino (215a). The relative distribution and, therefore, the corresponding shape effect are, by definition, flat in the reference year (see Morris, Bernhardt, and Handcock 1994). 5. We must exclude from this group a large number of residents living close to the border between the Eastern Region and the Greater Accra Region that, in 26, were wrongly counted in the Eastern Region and that, in 212, were reassigned to Greater Accra. Additional research is needed to establish the exact number of people affected. 6. Oil imports rose in value from $2.9 billion in 212 to $3.7 billion in 214; see the report First Macroeconomic Stability for Competitiveness s and Growth development policy financing, World Bank (215) Remaining Challenges 45

62

63 Chapter 4 A Roadmap for Policy Action The economic development of Ghana over the last two decades has been a tale of success. The country has cut the poverty rate by half, considerably reduced vulnerability, and substantially improved nonmonetary indicators of living standards, such as measures of health care, education, and access to basic services. The driver behind this success has been sustained economic growth, which has led to a generalized increase in consumption. However, the position of the bottom 4 relative to the rest of the population changed little and inequalities in both outcomes and opportunities remain substantial. The development objective of Ghana is now to consolidate the success in the face of internal and external challenges and a rapidly changing economic and social environment. The deteriorating macroeconomic outlook, the nature and speed of structural and spatial transformation, and the persistent inequalities threaten future progress in poverty reduction and the prospects for growth. After highlighting the main achievements in recent years and the challenges ahead, this chapter outlines a roadmap for policy action. The task ahead is complex, and the sequencing and prioritization of policies need to be carefully considered. The key to success will be the identification and implementation of a fiscally sustainable policy package that balances the acute needs of the poorest regions and the requirement to tap into the most dynamic segments of the economy. A Tale of Success The poverty rate has been reduced by half over the last two decades. It fell from 52.7 percent in 1991 to 21.4 percent in 212 (chapter 2). It is now less than half the average in Africa, while, in 1991, it was only 1 percent lower. The extreme poverty rate declined even more quickly and, by 212, was only around a quarter of the 1991 rate. However, the speed of poverty reduction slowed after 1998, from 13. percent to 7.1 percent in despite the rising GDP growth rate. The increase in household consumption was accompanied by dramatic improvements in nonmonetary indicators. Today, over the lifespan, newborns are expected to live two years longer than newborns in 25; children are more than twice as likely to be enrolled in secondary school, and households are more than twice as likely to have electricity and improved sanitation facilities. In Ghana in 212, the level of child malnutrition was among the lowest in Africa, and the literacy rate and rate of access to basic services were among the highest. The striking progress in providing better opportunities for all is more than an achievement in its own right. It also strengthens the prospects for strong and inclusive growth in the future. The reduction of poverty and the expansion in shared prosperity were achieved during a period of rapid changes in the economic and sociodemographic structure of the economy. Four factors are especially relevant in this outcome: (1) the decline in the average household size and the drop in the dependency ratio, (2) structural transformation away from agriculture, (3) the increased skills among the labor force, and (4) successful urbanization. The changes in household composition led to the lower dependency ratio (Falco et al. 214). The changes in demographics played an important role in securing the position of vulnerable households. We observe important changes in household structure starting with the 6th percentile (chapter 3). The other three factors helped raise the earnings potential of the typical adult in A Roadmap for Policy Action 47

64 the household. In a country in which employment rates are high, the increase in earnings is strongly associated with growing consumption and a movement out of poverty for example it explains nearly half of the poverty reduction between 1998 and 25 (Azevedo et al. 213). In 211 industry had a larger share of value added than agriculture. The recent economic growth has been associated with a shift of the economy out of agriculture and a substantial rise in agricultural productivity. The share of agriculture in GDP declined by nearly 5 percent, and, by 211, agriculture was the smallest sector in the economy in terms of value added. Meanwhile, the service sector expanded to nearly half of GDP, from an initial 34.4 percent in Moreover, in 211, for the first time since independence, the share of industry in GDP surpassed that of agriculture. Workers are now less concentrated in agriculture. The sectoral distribution of employment adjusted in line with the changes in the structure of the economy albeit at a lower rate. Agriculture remains the main sector of employment, at 43.2 percent of total employment. However, over the last decade, workers have been shifting into services and, to a lesser extent, industry. Employment in the service sector expanded from 28.8 percent in 1991 to 42. percent in 212. The share of construction more than doubled. The share of industrial employment also increased, from 1.7 percent to 14.9 percent. The labor force has become better educated over the past two decades. Between 1991 and 212, the share of the labor force without schooling shrank by half, from 41 percent to 24 percent. By 212, the majority (52 percent) of workers had completed at least lower-secondary education, compared with 39 percent in However, large inequalities remain in access to education, especially secondary education and above. Moreover, the quality of learning in basic education and beyond appears to be declining, and the capacity of the education system to foster relevant skills to boost competitiveness and productivity is limited (Darvas et al, 215). Ghana is urbanizing quickly. Between 1991 and 212, over 8 million Ghanaians migrated to urban areas, mostly into the metropolitan areas of Accra and Ashanti, which gained over 2.4 million inhabitants each, around half in the last decade. As result of the shift, by 212, the population was equality split between urban and rural areas, while, in 1991, 7 percent had lived in rural areas. A Less Positive Outlook Despite the success in reducing poverty and promoting shared prosperity, challenges remain. We analyze three development challenges in detail in this report: growing inequality and polarization in household consumption, large spatial disparities, and the deteriorating macroeconomic environment. Consolidating the progress made over recent decades require that these challenges be addressed promptly and effectively. There are still large inequalities in opportunities and outcomes. The recent success with poverty reduction and shared prosperity has been driven by the generalized growth in living standards and has been achieved despite a small increase in inequality, which was particularly noticeable between 1998 and 25. Our analysis points to differences in educational attainment among household heads, their jobs, and their region of residency as important determinants of consumption inequality, although inequality also persisted across households with similar characteristics (chapter 4). Inequalities between regions played a particularly significant role; they explain, on average, 4 percent of the total change in the Gini. Of special concern are inequalities in the quality of education across schools and geographical areas for example, the top 1 percent of schools produce 9 percent of the students entering university (World Bank, 215). The crucial role of the spatial dimension in explaining income inequality is evident in the magnitude of disparities across regions. 48 A Roadmap for Policy Action

65 The incidence of poverty has historically been higher in the north than in the rest of the country because of the less favorable climate, the distance from the sea, and the lack of infrastructure. This disadvantage has increased over the last two decades. Poverty rates have fallen below 2 percent in the large area that includes the Ashanti, southern Brong Ahafo, Eastern, Greater Accra, coastal Volta, and Western regions, but they are still around 5 percent in the north. In 212, as a result of this gap, nearly 4 percent of the poor were living in the north, which accounted for only 17 percent of the country s population and despite the large outflows to the richer south. The number of the poor was rising in the north, while it was declining everywhere else. The urbanization process is at a crossroads. The population is moving steadily into urban areas. It has grown exponentially in Accra and Kumasi, but, outside these metropolitan areas, it has grown more quickly in towns and smaller cities than in larger cities. However, recent improvements in jobs opportunities and in access to services have been highly concentrated in the larger metropolitan areas. Combined with high inflation, this trend is likely the reason poverty rates in urban areas other than Accra are rising. Moreover, all urban areas have started to see the side effects of rapid urbanization, including congestion, unregulated expansion, and a decline in access to services and affordable housing. The macroeconomic environment is deteriorating. Since 212, Ghana has suffered a number of serious macroeconomic shocks, both external the rupture of the West African natural gas pipeline and the high volatility in the price of gold exports and domestic. GDP growth fell by half in 214 and is projected to slow to 3.4 percent in 215. Inflation has risen sharply, from 8.8 percent in 212 to 17. percent in 214, and nonfood inflation has increased from 11.6 percent to 23.9 percent. The Ghanaian cedi depreciated by percent against the U.S. dollar, and the mix of the escalating public sector wage bill, the energy-rationing regime adopted in response to the disruption in natural gas supplies, and the liquidity constraints associated with the purchase of oil, has exacerbated the imbalance. A Roadmap for Policy Action The most immediate policy priority is to prevent further deterioration in the macroeconomic environment. Our decomposition points to economic growth as the main driver of the success in reducing poverty (chapter 2). Thus, addressing the macroeconomic imbalances that have muted the country s economic growth prospects is a priority in the effort to insure that progress in poverty reduction is sustained. The long-term growth prospects are positive, but, to realize the full potential of growth, it is crucial to continue the stabilization program started in 214. Its success hinges on maintaining the commitment to fiscal discipline and rapid structural reforms. Assuming that the issue of energy rationing is resolved and that the planned fiscal adjustment remains on track, we expect the GDP growth rate to rebound to 5.9 percent in 216 and 8.2 percent in 217. Provided that the growth elasticity of poverty remains at the levels of the last decade, this can have a strong impact on additional poverty reduction. Tackling the inequalities in outcomes and opportunities is a longer-term development challenge. With its rise to middle-income-country status in 211, Ghana entered a new stage of development, one in which it will be difficult to achieve progress in poverty reduction and shared prosperity without broadening the reach of the development process to those people who have so far been left behind. The main development challenge in this new phase is to enhance the access to opportunities for all without stifling the energy of the economy. This calls for a multifaceted, well-targeted, and fiscally sustainable policy package that balances the needs of the poor with the needs of the most dynamic sectors. Reducing the inequality of opportunity requires a continued commitment to investment in improving nonmonetary indicators. A Roadmap for Policy Action 49

66 Recent years have seen considerable advances in human development indicators and in access to services. However, service quality is still low, and the inequalities are significant, particularly across geographical areas. The key to fostering shared prosperity in the new phase of development will be the capacity to reduce these disparities. This is because the inequality of opportunity such as unequal access to education, health care, and credit can persist across generations by exacerbating income inequalities, which limit the realization of the potential of individuals based only on inherent characteristics (such as gender or ethnicity) or accidents of birth (such as parental income and educational attainment). It can also have a severely negative effect on productivity. Thus, to insure the sustainability of growth and despite the tight fiscal space, it is essential for the government to maintain its commitment to investment in reducing the gaps in service delivery and in improving the provision of quality services outside the metropolitan areas, but especially in the poorest regions. An important element of a successful policy package to reduce inequality is the implementation of more and better targeted social protection. The expansion of social protection can have a positive impact in a country such as Ghana in which (1) poverty is concentrated among households with particular characteristics in relatively deprived areas, (2) the various dimensions of poverty overlap so that the poor are at a disadvantage in more than one dimension, and (3) vulnerability to poverty is substantial. If well targeted, social protection can help reduce the inefficiencies in the allocation of resources and boost the productive potential of individuals and communities by breaking the vicious circle that links income inequality and inequality of opportunity across generations. Experimenting with innovative ways to expand social insurance to cover self-employed and informal workers can also lower the vulnerability of households to shocks. The government expenditure on social protection 1.4 percent of GDP is low compared with the expenditures among Ghana s middle-income peers in Africa. The evidence also suggests that current programs are poorly targeted in Ghana and simulations indicate that extreme poverty can be eradicated with a minimum investment of.5 percent of GDP in a well-targeted transfer program (World Bank, 215b). Thus there is scope to continue to expand programs such as Livelihood Empowerment against Poverty. Given the extent to which multiple disadvantages overlap in some geographical areas or across some types of individuals, school feeding programs or cash transfer programs conditional on a particular behavior among recipients in health care, education, or the work environment may be a viable alternative to current transfer programs. A scaling-up of the public works program in the north might also be considered after a rigorous evaluation of the impact of the program so far. However, the main component of an effective program for sustainable poverty reduction are policies designed to increase earnings among the poor. Labor is the principal asset and main source of income among the poor. Changes in labor income accounted for nearly half the reduction in poverty in Ghana between 1998 and 25 (Azevedo et al. 213). Boosting earnings requires that the process of structural transformation of the country into a modern and diversified economy be accelerated, while increasing the productivity of traditional economic sectors. Achieving this objective will require three sets of policy interventions designed to: (1) increase productivity in agriculture and in low-productivity sectors, (2) raise the number of jobs in the modern private sector (in wage employment and in higher productivity self-employment), and (3) facilitate the occupational and geographical mobility of workers. The barriers to improving agricultural productivity are many and complex. Increasing agricultural productivity and transforming agriculture from a subsistence base to a market base have been government priorities since independence. Various policies and interventions have been implemented to boost cash crop production for internal and especially international trade, 5 A Roadmap for Policy Action

67 but the results have been mixed (Molini et al. 21). Cocoa beans are the main export crop, and Ghana s increasing share of the global cocoa market has been strongly associated with poverty reduction: the poverty rate among cocoa farmers declined from 6 percent in 1991 to about 24 percent in 25. However, despite targeted government interventions, the average yield per hectare 431. kilograms in is low compared with the yields among the country s main competitors, Côte d Ivoire and Indonesia. The growth in the production of staple crops such as rice, maize, and millet, and in higher-value vegetables and fruits for domestic and export markets is also encouraging (Breisinger et al. 28). The average output of staple crops grew much more quickly than the population, and per capita production was more than 8 percent higher in 25 7 than in However, the potential for gains in agricultural productivity is limited by traditional farming methods, rainfall volatility, and poor access to the expanding internal market. Interventions to raise productivity in agriculture should become focused on reducing the reliance of crops on rainfall, increasing the access to larger markets through better infrastructure, and scaling up production. However, the most effective ways of reducing poverty among farming households remain (1) reducing the dependence of incomes on volatile yields by expanding agricultural insurance and encouraging income diversification into nonfarm activities and (2) fostering the shift of underemployed workers out of the sector. The capacity of individuals to diversify their incomes and undertake more productive activities hinges crucially on the availability of jobs in more productive sectors, the extent to which the agricultural labor force possesses the skills required by more productive jobs, the degree of connectivity with markets and with the areas where nonagricultural jobs are located, and the existence of well-defined land title rights and a well-functioning land market. Reforms in these areas are critical. Further progress in poverty reduction and shared prosperity requires a concerted effort to boost the development of the modern sector. In recent years, the economy has undergone a profound structural transformation (chapter 3). However, Ghana still lacks a vibrant modern sector capable of offering good earnings opportunities to the growing nonagricultural labor force. Although, in 211, the share of the industrial sectors in GDP exceeded the corresponding share of agriculture for the first time since independence, the industrial share of employment is still below 15 percent. Most newly created jobs are concentrated in services. In Accra, a key element in the expansion of the service sector was the rapid growth of high value added services such as information and communication technology, finance and insurance, and real estate. However, elsewhere, the bulk of the increase was in low value added activities that are borderline between the formal and informal sector and that characterize West African towns: retail activities, construction, transport, and so on. Progressing to the next stage of structural transformation will hinge on the creation of a dynamic, but laborintensive private sector capable of absorbing the relatively low-skilled workers released from agriculture. Policies to boost the creation of modern jobs are essential. These will include policies aimed at creating an enabling business environment, promoting investments in skills and innovation, and enhancing the rule of law and property rights. However, a more detailed analysis of the binding constraints to modernization of the economy needs to be carried out to identify the policy priorities within these broad categories. An initial analysis indicates that poor infrastructure and poor skills among labor force outside Accra are major potential culprits. The lack of adequate skills is an especially important constraint and is exacerbated by the considerable wage premium for public employment, which is crowding out private sector employment among the more well-educated. The occupational and geographical mobility of workers must be fostered. The ongoing structural transformation requires a more highly skilled labor force clustered around the A Roadmap for Policy Action 51

68 urban and peri-urban areas where job opportunities are increasingly concentrated. The increase in educational attainment among workers experienced since 1991 has been a major factor in the successes so far. The concentration of skilled workers in the areas of the country that were expanding economically was facilitated by the mass movement of workers to urban areas. Investment in infrastructure and in the provision of public services to the growing urban population was promoted by the spatial and economic transformation. Despite the inflow of 8 million people into urban areas, average household consumption continued to increase, and poverty declined from about 3 percent in 1991 to 9 percent in 212. Nonetheless, that much of the progress has been concentrated in the metropolitan areas of Accra and Kumasi where most of the jobs are also clustered means that efforts must now be undertaken to broaden the reach of the transformation process to other areas ensure that workers can shift into higher-productivity jobs and enjoy higher earnings elsewhere, too. Several policies can foster structural transformation and help maximize the benefits of urbanization. On the demand side, effective interventions to unleash the potential of urbanization include better land use management and planning in municipal and metropolitan areas, improvement in transport to connect markets and boost factor mobility, and a more efficient credit market. On the supply side, it will be crucial to ensure that workers have the skills required by the new jobs and that the costs of migrating to urban areas are not prohibitively high. The success of the interventions will hinge on ensuring that economic transformation and migration expand beyond the large metropolitan areas into smaller cities. A small set of win-win policy areas are emerging as priorities in the effort to extend Ghana s successes in poverty reduction and shared prosperity. The most immediate priority is to restore a sound macroeconomic environment so as to enhance the prospects for growth and foster additional economic transformation. Ensuring adequate investment in infrastructure and skill development will be key to increasing productivity in agriculture, creating modern sector jobs, and ensuring that workers have the skills they need to take advantage of the new employment opportunities. Greater and better connectivity between rural and urban areas, combined with clearly defined land title rights and more efficient land markets, will facilitate structural transformation by allowing Ghana to benefit from geographical agglomeration. 52 A Roadmap for Policy Action

69 Appendix A Computing Poverty This appendix discusses the methodology to compute poverty using the Ghana Living Standard Survey round 6 data. The World Bank provided Technical Assistance (TA) to the General Statistical Service of Ghana (GSS). The GSS and World Bank worked together to construct the household consumption expenditure aggregates, adjust this measure of total expenditure into real terms for comparison across space and time (back to 25/26 to compare with the previous survey, GLSS5), and construct two poverty lines to measure the poor and the extreme poor. In a second stage of collaboration, a poverty report was produced. This effort faced two major challenges: rebasing the poverty lines and computing reliable price deflators. This brief note describes the methodology used to calculate the new poverty Map A.1 Administrative Map of Ghana lines and the methods used to construct robust price deflators. Rebasing the Poverty Line The poverty line and extreme poverty line (measured in Ghana Cedis) are designed to measure whether a household is poor or extreme poor, by comparing the total household consumption and food consumption to each line. The basis for extreme poverty line is the cost of buying a basic bundle of food which is sufficient to give adequate calories and based on typical consumption of the poor/near poor (in terms of food types and their quantities). The poverty line is computed by adding an amount to cover nonfood expenditures to the extreme poverty line. Two key decisions are made: the bundle of food (types/quantities) to reach sufficient calories and the nonfood amount (usually a fixed share of the food poverty line). The basis for both the basket of food items and the nonfood share had been selected in 1999 and had not been updated since then. Several new items had entered Ghanaian household consumption since then as well as general changes in consumption patterns had occurred (items such as DVD/VCD, MP3/MP4 players, vacuum cleaner, rice cooker, mobile phone, tablet PCs, etc.) In this regard, these aspects were deemed to be potentially outdated and warranted revision. The GSS decided to re-compute the poverty lines based on the GLSS6, to reflect changes in the food basket consumed by Ghanaian households. In line with international practice, GSS calculated the average expenditure of the food consumption basket for the bottom 5 percent of individuals ranked by consumption per adult equivalent, and derived the amount of calories in this basket. The calorie price is then Computing Poverty 53