SHARING THE GROWTH STATE OF THE AFRICA REGION WORLD BANK IMF ANNUAL MEETINGS 2013 OFFICE OF THE CHIEF ECONOMIST - AFRICA REGION WORLD BANK

OUTLINE I. Progress against poverty in Africa: an overview II. A four-part roadmap for sharing growth in Africa 1. Maintain strong macroeconomic discipline 2. Build better human and physical capital 3. Promote growth in the places and sectors where the poor are 4. and/or create social protection and promotion systems that enable them to share in growth elsewhere.

AFRICA S GDP GROWTH IN THE 2000S WAS STRONG 15% 13% GDP GROWTH IN SSA, CHINA AND OTHER DEVELOPING COUNTRIES Real GDP growth 11% 9% 7% 5% 3% 1% -1% 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Rest of Developing (excl. China) China Sub-Saharan Africa

AFRICA S GDP GROWTH IN THE 2000S WAS STRONG, ALTHOUGH NOT AS IMPRESSIVE IN PER CAPITA TERMS GDP GROWTH IN SSA, CHINA AND OTHER DEVELOPING COUNTRIES Real GDP growth 15% 13% 11% 9% 7% 5% 3% 1% -1% 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Rest of Developing (excl. China) China Sub-Saharan Africa GDP PER CAPITA GROWTH IN SSA, CHINA AND OTHER DEVELOPING COUNTRIES Real GDP per capita growth 15% 13% 11% 9% 7% 5% 3% 1% -1% 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Rest of Developing (excl. China) China Sub-Saharan Africa Source: World Development Indicators

AS A RESULT, AFRICA S POVERTY INCIDENCE FELL, BUT NOT SPECTACULARLY 70 In the last twenty years, extreme poverty fell by eight percentage points in Africa, as compared to 44 points in East Asia. $1.25 a day headcount (%) 60 50 40 30 20 10 0 56.5 59.4 58.1 58.0 South Asia 31 East Asia and Pacific 12.5 LAC 5.5 MENA 2.4 ECA 0.6 1990 1993 1996 1999 2002 2005 2008 2010 55.7 52.3 49.2 Sub-Saharan Africa 48.5 Source: PovcalNet.

THIS POVERTY REDUCTION GROWTH GAP REFLECTS THE REGION S LOW GROWTH ELASTICITY OF POVERTY 15 GROWTH ELASTICITY OF POVERTY REDUCTION, 2000-2010 Five most populous countries by region, except Poland and Sri Lanka. 10 5 0-5 -10-15 Egypt, Arab Rep. Turkey Ukraine Yemen, Rep. Kazakhstan Argentina--Urban Peru Philippines Tunisia Morocco Vietnam Colombia Nepal Bangladesh India* Indonesia* Nigeria Ethiopia China* Uganda Iran, Islamic Rep. Tanzania Thailand Brazil Mexico Romania Pakistan South Africa BIGGER IMPACT OF EACH PERCENTAGE POINT OF GROWTH ON POVERTY REDUCTION Source: estimates based on PovcalNet.

WHICH IN TURN REFLECTS HIGH INEQUALITY Most African countries have high levels of consumption or income inequality, relative to the rest of the world. Seven of the ten most unequal countries in the world today are in SSA. Gini coefficient 70 60 50 40 30 20 10 0 MLI BDI ETH NER SDN SLE GNB TZA LBR BEN CMR TGO GIN NGA TCD BFA SEN MRT GAB CIV AGO GHA MDG MWI UGA ZAR MOZ GMB COG KEN CPV RWA STP SWZ LSO CAF ZMB BWA NAM ZAF COM SYC Consumption Survey Income Survey Sub Saharan Africa Source: PovcalNet, most recent survey available.

AND A GROWTH PATTERN THAT IS OFTEN NOT INCLUSIVE In one African country, average p.c. household consumption grew by 6.5% between 2004-2010. But whereas the top 5% of the population experienced annual growth rates of almost 8%, the bottom 5% grew by between 1% and 3%. 9 Growth Incidence Curve, 2004-2010 Percent growth in consumption 8 7 6 5 4 3 2 1 0 1 8 16 23 31 38 46 53 61 68 76 83 91 98 Consumption expenditure percentile Source: estimates based on household surveys from Survey-based Harmonized Indicator Program (SHIP)

LOOKING FORWARD TO 2030, AFRICA S POVERTY WILL LIKELY* BE REDUCED FROM A HALF, TO BETWEEN A QUARTER AND A THIRD OF THE POPULATION 70 450 $1.25-day headcount (%) 60 50 40 30 20 10 0 289.7 330 349.4 376.8 390.2 394.8 399.3 413.7 29.9 (403.8) 26.4 (356.1) Headcount (Number of poor (mil.)) 1990 1993 1996 1999 2002 2005 2008 2010 2030 Scenario 1 Scenario 2 Historical Headcount (Number of poor (mil.) in bars) 400 350 300 250 200 150 100 50 0 Number of poor (mil.) * Scenario 1 projects each country s historical GDP per capita growth rate over the 2000-2010 period forward to 2030. Scenario 2 does the same using mean household survey incomes. Both scenarios assume distribution-neutral growth. Source: PovcalNet and staff estimates based on Chen, Kleineberg, Kraay, Lanjouw (2013).

LOOKING FORWARD TO 2030, AFRICA S POVERTY WILL LIKELY* BE REDUCED FROM A HALF, TO BETWEEN A QUARTER AND A THIRD OF THE POPULATION 70 450 $1.25-day headcount (%) 60 50 40 30 20 10 0 289.7 330 349.4 376.8 390.2 394.8 399.3 413.7 29.9 (403.8) 26.4 (356.1) Headcount (Number of poor (mil.)) 1990 1993 1996 1999 2002 2005 2008 2010 2030 Scenario 1 Scenario 2 Historical Headcount (Number of poor (mil.) in bars) 400 350 300 250 200 150 100 50 0 Number of poor (mil.) 63% of the world s poor in Africa 78% of the world s poor in Africa * Scenario 1 projects each country s historical GDP per capita growth rate over the 2000-2010 period forward to 2030. Scenario 2 does the same using mean household survey incomes. Both scenarios assume distribution-neutral growth. Source: PovcalNet and staff estimates based on Chen, Kleineberg, Kraay, Lanjouw (2013).

THERE WOULD ALSO BE CONSIDERABLE HETEROGENEITY WITHIN THE REGION Under this business as usual scenario (1), poverty incidence would remain above 20% in almost half of all countries, and above 40% in nine countries. 90 80 $1.25 poverty headcount (%) 70 60 50 40 30 20 10 0 $1.25 headcount = 3% Madagascar Burundi Congo, Dem. Rep. Central African Zambia Malawi Comoros Niger Guinea-Bissau Nigeria Benin Guinea Swaziland Kenya Togo Congo, Rep. Gambia, The Cote d'ivoire Mali Senegal Mozambique Tanzania Lesotho Burkina Faso Liberia Rwanda Sierra Leone Mauritania Sao Tome and Principe Ghana Uganda Chad Cameroon Namibia Sudan Angola Botswana Gabon Seychelles South Africa Cape Verde Ethiopia Source: PovcalNet and staff estimates based on Chen, Kleineberg, Kraay, Lanjouw (2013).

TO SUM UP Poverty in Africa remains unacceptably high, and the pace of reduction remains unacceptably slow Sustained economic growth in the next two decades is essential for progress But it is not sufficient: Growth must be more inclusive With falling inequality in outcomes and opportunities.

AFRICA S GROWTH BENEFITED FROM A BENIGN EXTERNAL ENVIRONMENT IN THE 2000S Good policies leading to more public and private investment were key But so were rising commodity prices and abundant international liquidity Average GDP growth, 2000-11 (%) Average GDP growth and Terms of Trade (2000-11) Slope: 0.013 p-value: 0.07 Growth in Terms of Trade, 2000-11 (%) Resource Rich Resource Poor Source: World Economic Outlook, International Monetary Fund

THESE TAILWINDS MAY WELL BE RECEDING Metal and mineral prices rose by 99.4% between 2000 and 2013. Since December 2010, they have fallen by 25%. Price index, 2005=100 250 200 150 100 Agriculture Energy Metals and minerals 50 0 Jan-00 Jul-01 Jan-03 Jul-04 Jan-06 Jul-07 Jan-09 Jul-10 Jan-12 Jul-13 Source: Development Prospects Group

WITH LIKELY CONSEQUENCES FOR GROWTH. Impact of one standard deviation decline in metal prices on GDP growth Cape Verde Malawi Kenya Togo Cote d'ivoire Senegal Mauritius Comoros Rwanda Nigeria Madagascar Congo, Dem. Rep. Namibia Lesotho South Africa Tanzania, United Rep. Mali Botswana Mozambique Ghana (% ch, GDP growth) -2.5-2.0-1.5-1.0-0.5 0.0 0.5 1.0 (Simulation based on a global macroeconomic model. Source: DECPG)

FISCAL AND CURRENT ACCOUNT BALANCES HAVE GENERALLY WORSENED IN THE LAST DECADE Change in Current Account Balance (% GDP) and Fiscal Balance (% GDP), 2000-2012 Change in Current Account Balance, % GDP 20 10 0-10 -20-300 -200-100 0 Change in Fiscal Balance, % GDP Resource Rich Resource Poor If terms of trade deteriorate at the same time as capital flows become less abundant, countries with low fiscal and current account deficits will fare much better. Source: World Economic Outlook, International Monetary Fund

GREATER (AND BETTER) INVESTMENT IN HUMAN AND PHYSICAL CAPITAL IS NEEDED Sustained growth is essential for poverty reduction As tailwinds recede, the domestic investment effort needs to become even greater Ultimately, growth comes from the build up of effective human and physical capital Regional growth in GDP per capita (1960-2000) against average test scores, after conditioning on initial GDP per capita. Conditional Growth -2-1 0 1 2 LATAM SSAFR S-EUR C-EUR N-EUR COMM MENA ASIA -50 0 50 Conditional Test Score coef =.02305259, se =.00130272, t = 17.7 Source: Figure 1 in Hanushek and Woessman, Journal of Economic Growth (2012)

AFRICA S LEARNING OUTCOMES COMPARE POORLY WITH THOSE ACHIEVED ELSEWHERE Proportions of Grade 8 students scoring at low ; intermediate / high ; and advanced benchmarks (Math, TIMSS 2011) 1.0 0.8 0.6 0.4 Proportion 0.2 0.0-0.2-0.4-0.6-0.8-1.0 Korea International Average Malaysia Chile Jordan Botswana Indonesia South Africa Ghana >625 400-625 <400

PARTLY BECAUSE OF VERY UNEQUAL SCHOOLING OPPORTUNITIES Rich-Poor Gaps in Proportion of 15-19 Year Olds who have Completed Grade 6 1 0.8 0.6 Proportion 0.4 0.2 0 Niger 2006 C.A.R. 1994 Chad 2004 Mozambique 2003 Mali 2006 Burkina Faso 2010 Guinea 2005 Comoros 1996 Togo 1998 Liberia 2007 Madagascar 2008 Cote d'ivoire 2005 Rwanda 2010 Burundi 2010 Ethiopia 2011 Mauritania 2000 Senegal 2010 Benin 2006 Sierra Leone 2008 DR Congo 2007 Uganda 2011 Malawi 2010 Congo Rep. 2005 Gabon 2000 Sao Tome and Cameroon 2011 Nigeria 2008 Zambia 2007 Ghana 2008 Swaziland 2006 Lesotho 2009 Tanzania 2010 Kenya 2008 Namibia 2006 South Africa 1998 Zimbabwe 2010 Richest quintile Poorest quintile Average Source: DHS. http://econ.worldbank.org/projects/edattain. Most recent survey available shown.

ON THE PHYSICAL CAPITAL SIDE, INFRASTRUCTURE REMAINS SCARCE AND (VERY) EXPENSIVE IN AFRICA Median prices to final user in Africa (relative to South Asia) 7.0 6.0 Ratio of prices 5.0 4.0 3.0 2.0 1.0 0.0 Power International call Water Road freight Internet dial-up Mobile telephone Causes include lack of scale economies, geographic fragmentation and lack of competition. Consequences include higher costs transmitted downstream to infrastructure users, and hence reduced competitiveness and diversification.

CHANGING THE DISTRIBUTIONAL PROFILE OF AFRICA S GROWTH CAN HAVE LARGE IMPACTS ON POVERTY An iso-growth line in poverty/inequality space Headcount 2010: 74.4% 72% $1.25-day headcount 2030 62% 52% 47% 41% 31% 21% CAG: 3.6% 10% 30 20 10 0 0% Inequality change (%) Gini 2010: 57.4-10 -20-30 Source: estimates based on household surveys from Survey-based Harmonized Indicator Program (SHIP)

CHANGING THE DISTRIBUTIONAL PROFILE OF AFRICA S GROWTH CAN HAVE LARGE IMPACTS ON POVERTY AND IT CAN BE DONE! Latin America: Declining income inequality by country: 2000-2011 Annual change of Gini in % 3.00 2.00 2.12 1.00 0.61 0.82 0.77 0.40 0.00-1.00-2.00-2.05-1.99-0.20-0.10-0.47-0.39-1.30-1.24-1.17-1.07-1.03-0.91-0.79-0.74-0.72-0.95-3.00-2.64 Nicaragua Bolivia Ecuador Argentina El Salvador Mexico Venezuela Brazil Peru Dom. Rep. Panama Chile Costa Rica Paraguay Uruguay Guatemala Honduras LAC-17 China South Africa India USA Source: Lustig (2013)

GROWTH THAT TAKES PLACE IN RURAL AREAS WHERE MOST OF THE POOR LIVE IS MOST EFFECTIVE IN REDUCING POVERTY Econometric evidence from China, India and Brazil suggests that the geographic and sector composition of growth affects the overall growth elasticity of poverty. In Africa, most of the growth has not been coming from the places and sectors where the poor are. Contribution of urban and rural areas to population, poverty reduction and consumption growth in a particular African country, 2005 09 100% 80% 60% Population share Share of poverty reduction Shares in total consumption growth So most of the poverty reduction comes from very specific growth sources 40% 20% 0% -20% Rural Urban Source: Kaminski and Christiaensen 2013

SO, BOOSTING RURAL INCOMES THROUGH GREATER AGRICULTURAL PRODUCTIVITY AND BETTER OFF-FARM JOBS IS KEY Percent growth in consumption 6 4 2 0 Simulated Growth Incidence Curves, Zambia 2010-2030 1 10 20 29 39 48 58 67 77 86 96 Consumption expenditure percentile GIC 2030 2010; annual growth 3.6%, Agricultural annual growth 2.1%, Other sectors 4.4%, Unemployed 2.6% GIC 2030 2010; annual growth 3.6%, Agricultural annual growth 3.3%, Other sectors 3.9%, Unemployed 2.6% GIC 2030 2010; annual growth 3.6%, Agricultural annual growth 3.3%, Other sectors 3.9%, Unemployed 2.6%, structural transformation (employment in agriculture decreases 20% points) Source: estimates based on household surveys from Survey-based Harmonized Indicator Program (SHIP)

GROWTH THAT TAKES PLACE AWAY FROM THE POOR CAN ALSO BE HARNESSED The natural resource sector is seldom directly pro-poor The rents it generates should typically be re-invested Building other forms of capital to replace the natural capital being depleted Cash transfers targeted to the poor have a place in that investment portfolio 836.6 Annual transfer to each poor person ($.125 a day) from 10% of natural-resource fiscal revenues in selected countries. 326.3 178.8 64.2 29.8 5.3 4.9 Equatorial Guinea* Angola Republic of Congo Mozambique Nigeria Uganda Tanzania Note: number of poor for Equatorial Guinea calculated using national poverty line. Source: Estimates of natural resource fiscal revenues from Devarajan and Giugale (2013)

THE CCT REVOLUTION IN SOCIAL PROTECTION Since the late 1990s, conditional cash transfers have shown that: (i) Good targeting is possible (ii) Transfers to increase family incomes and reduce poverty (iii) Households use the transfers to improve nutrition,

IS COMING TO AFRICA. increase investments in human capital; UCTs CCTs 1.0 8% 23% 41% 36% 24% 48% Average impact of cash transfers on the odds of school enrollment (and 95% confidence interval) Results from a systematic review of 8 unconditional and 27 conditional cash transfer interventions (Baird et al., 2013)

FAMILIES SYSTEMATICALLY INVEST PART OF WHAT THEY RECEIVE and even to save and invest in physical and financial capital as well! Five years ago when my oldest daughter was in school and we received money from PROGRESA, we saved 600 pesos to buy wood and the other materials for building a chicken coop, and with what was left we bought a few chickens. Since then, we have raised many chickens that we sometimes sell, and we collect 10 to 15 eggs per week that we eat ourselves. - Oportunidades beneficiary in rural Mexico, August 2004, cited in Gertler, Martinez and Rubio-Codina (AEJ: Applied Economics, 2012)

FAMILIES SYSTEMATICALLY INVEST PART OF WHAT THEY RECEIVE and even to save and invest in physical and financial capital as well! Community Based Conditional Cash Transfers in Tanzania Some Results (endline) 1.09 1,625 0.38 7% points Indigenous goats Local chickens INVESTMENT IN LIVESTOCK -1,593 Cigarettes, tobacco, snuff Insurance (car, medical, life) ANNUAL AVERAGE HOUSEHOLD NON-FOOD EXPENDITURE, LCU Shoes CHILDREN'S ASSETS Note: Investment in livestock refers to additional goats and chickens purchased by treatment households. Additional annual average non-food expenditure by treatment households measured in Tanzanian Shilling (TSH). Children s assets refers to the higher likelihood of treated children to own shoes in percentage points. Source: charts based on results from Evans, Hausladen, Kosec and Reese (2013)

CONCLUSIONS Sustained economic growth is essential to the fight against poverty in Africa Macroeconomic prudence reduces vulnerability to receding tailwinds Investment in human and physical capital must become more effective (quality as well as quantity) But growth is not sufficient: shared prosperity will require a reduction in inequality (of outcomes and opportunities) Boost productivity in agricultural and rural off-farm jobs Cash transfers are a real policy option

THANK YOU!