Recommended
More Related Content
What's hot
What's hot (17)
Similar to Poverty in Africa Goes Beyond Income
Similar to Poverty in Africa Goes Beyond Income (20)
More from StatsCommunications
More from StatsCommunications (20)
Recently uploaded
Recently uploaded (20)
Poverty in Africa Goes Beyond Income
- 1. POVERTY IN A RISING AFRICA – A NON-MONETARY PERSPECTIVE Luc Christiaensen, Presentation at Measurement of Well Being and Development in Africa Conference, Durban, South Africa, 12-14 November, 2015 1 http://www.worldbank.org/africa/povertyreport
- 2. • Robust economic growth (4.5 % per year 1995-2013) …doubts @ progress in extreme poverty ($1.90-day), in part driven by data concerns • And a monetary account of poverty does not tell the full story • many aspects of well-being cannot be monetized • it does not have an intrinsic (only an instrumental) value (initial conditions) • Income is at household level (ignoring intra-household distribution) • Poverty in a Rising Africa • Scrutinizes the data for measuring consumption (and poverty, inequality) • Revisits evolution of poverty and inequality taking data considerations into account • Considers non-monetary dimensions Africa is Rising! Are People Better Off? 2
- 3. A Nonmonetary Perspective • Inspired by capability approach (functionings & capabilities) • Human Development Indicators and Multidimensional Poverty Index • Different emphasis in three areas Conceptually: dimensions? • TRADITIONALLY: Being able to read and write & being well nourished and healthy • WE ADD: being free from violence, but also being free to decide (to proxy capability) Practically : indicators? • Data abundance; but functionings not covered (mobility/capacity to aspire/social inclusion) • Remaining data issues (e.g. adult literacy) • Focus on individual and achievement (not income) Jointness in deprivation: dashboard versus single index ? • Middle ground: share of individuals poor in 1, 2, .. K dimensions (~M0) 3
- 4. The Non-Monetary Perspective – Take-Aways 4 Health, nutrition, education, and empowerment have improved, and violence has diminished. The challenges remain enormous: more than two in five adults are still illiterate and quality of schooling is often low; conflict is on the rise. Nonmonetary welfare indicators are systematically weaker in resource- rich countries, pointing to the unmet potential of natural resource wealth. Higher female educational achievement remains the game changer for improving human development outcomes Refocus on education
- 6. Adult literacy rates increased by 4 % points; But two in five adults are still illiterate 6 83 93 62 79 50 67 54 58 0 10 20 30 40 50 60 70 80 90 100 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Africa South Asia East Asia MENA LAC Adult literacy rates (%) (1995-2012) - Literacy rates are lowest in W-Africa - Gender parity gap: 25 %points - Especially high in western Africa - Much lower in southern Africa - Illiteracy higher among - Poorer, older, rural people - Resource-rich & landlocked countries Heterogeneity across Africa
- 7. Adult literacy rates remained low, despite substantial increase in primary school enrollment 7 76 106 55 74 22 47 0 10 20 30 40 50 60 70 80 90 100 110 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Percent Evolution of enrollment in SSA (1995-2012) Gross enrollment primary school Net enrollment primary school Gross enrollment secondary school male female
- 8. But, very low quality of schooling– many sixth graders lack even basic reading skill 8 -36 -73 -100 -50 0 50 100 SACMEQ Malawi Lesotho Uganda Namibia Botswana Zanzibar Kenya Swaziland Reading test scores 6th graders (SACMEQ) Pre, Emergent and Basic reading (1,2,3) Reading for meaning (level 4) Interpretive and above (levels 5,6,7,8) 40 -100 -80 -60 -40 -20 0 20 40 60 80 100 Comoros Benin Chad Cote d'Ivoire DR Congo Madagascar Burkina Faso Senegal Burundi Cameroon Gabon Reading Scores (PASEC) Level 1: students perform at or below the level expected for random guessing (score of less than 25%) Level 2: students score between 25% and 40% Level 3: Students perform at or above a level determined to be "basic knowledge" Note: 2004-2009: SACMEQ = Southern Africa Consortium for Measuring Educational Quality. PASEC = Programme d’Analyse des Systèmes Educatifs de la CONFEMEN.
- 9. Living long and well nourished 9
- 10. African newborns can expect to live 7 years longer since 2000, but progress is leveling off 10 49.9 56.9 60.7 66.9 45 50 55 60 65 70 75 80 Age(years) East Asia & Pacific Latin America & Caribbean Middle East & North Africa Sub-Saharan Africa North America Europe & Central Asia Life expectancy (1995-2013) 7 years • Follows sharp decline in under five child mortality (173 to 93 – 1995-2012) • Due to an increase in vaccination since 2000s and a drop in malaria deaths since mid 2000s • But progress leveling off • And HIV prevalence still high in southern Africa
- 11. … and malnutrition remains high, with new health issues on the horizon 11 Stunting among children(≤ 5): • dropped by 6 % points (1995-2012) • 2012: still 38.6 percent stunted 13 5 0 2 4 6 8 10 12 14 BMI<18,5 BMI>=30 % One African woman in 8 is underweight; one in 20 is obese Boys are more stunted than girls (by 5 %points); Women can expect to live in good health 1.6 years longer.
- 12. Being free from violence 13
- 13. After a decade of relative peace, violence is rising again 14 Number of violent events against civilians
- 14. Tolerance of domestic violence in Africa declined, but remains twice as high as in other developing countries 15 22% 14% 41% 30% 2000-2006 2007-2013 Other developing countries Sub-Saharan Africa Acceptance of domestic violence Note: Figures are population-weighted averages of 32 African and 28 non-African developing countries Incidence is about 60 percent of the acceptance rate. 1316 7777 0% 20% 40% 60% 80% Malawi Benin SaoTomeandPrincipe Mozambique Swaziland Madagascar Nigeria Namibia Ghana Lesotho Zimbabwe Comoros Liberia BurkinaFaso Cameroon Coted'Ivoire Gabon Kenya Tanzania Rwanda Uganda Senegal Niger Congo,Rep. Zambia SierraLeone Ethiopia Burundi Congo,Dem.Rep. Uganda Mali Very high in some countries
- 15. Higher tolerance in fragile and resource rich states. Lower tolerance among more educated & older NOT younger women. 16 -3.3 -7 -9.1 -10.8 -12.1 -13.7 -8.4 -15.8 -31.2 9.2 16 6.1 -1 -7.6 -40 -30 -20 -10 0 10 20 20-24 25-29 30-34 35-39 40-44 45-49 Age group/ compared with age group 15-19 Attended primary education Attended secondary education Attended higher education Education/ compared with No education Fragile/ compared with Non-Fragile Resource Rich / compared with Resource Poor Landlocked/ compared with Coastal Lower Middle Income/ compared with Low Income Upper Middle & High Income/ compared with Low Income Acceptance of domestic violence Declines with education Younger women are more tolerant!!! Note: only limited effects of no real effect of povOLS. Also controls for poverty, marital & employment status, annual trend, rural/urban. All limited effect
- 16. Being free to decide 17
- 17. Voice and accountability levels in Africa remain low Voice and accountability “Do not listen to the radio, watch television, or read a newspaper at least once a week “ • Africa: almost 40 % • RoW (excl. China): 25 % • Lower in Sahel, coastal West Africa, populous countries (DRC, Ethiopia) • 15% points lower among women • Poverty, rural residence, and lack of education are key differentiators. • Lower in resource-rich & fragile states (6 and 5%points respectively) Exposure to media 18 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 1996 1998 2000 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 EAP ECA LAC MENA North America South Asia Source: Worldwide Governance Indicators. Africa East Asia and PacificMiddle East and North Africa
- 18. Trend toward greater participation of women in household decision making processes. • “Husband has final say in decisions regarding their wives’ health care” • Africa: 46 % • South Asia: 39 % • Middle East and North Africa: 21 % Lower among women in poor and rural households, in resource-rich and landlocked countries; also among younger women!!! • ”Husband has final say on whether a married woman can visit friends or family” • Africa: 40 % • RoW : 33 % • “Control over a women’s earnings lies fully with someone else” • Africa: only 10 % Trend is toward greater empowerment 19
- 20. Living in resource rich countries comes with a penalty for your human development 21OLS estimates are conditional on country categories, residence and other household characteristics -3.1 -4.5 3.7 2.1 9 -6 -4 -2 0 2 4 6 8 10 Literacy (% points) Life expectancy (years) Women's malnutrition (% points) Children's malnutrition (% points) Incidence of domestic violence (% points) Resource-rich countries perform systematically worse
- 21. 22 Human development outcomes systematically higher among and in households with better educated women OLS estimates are conditional on country categories, residence and other household characteristics -19.0 -24.7 -7.1 9.4 -9.9 -4.1 -30 -25 -20 -15 -10 -5 0 5 10 15 Final decision on own health care Access to media Tolerance of domestic violence Incidence of domestic violence Chronic child malnutrition Women's malnutrition % points Attended Secondary Education
- 22. Vulnerable groups (ignored in standard poverty analyses): Numerous people are disabled, displaced, or deprived from one or both parents 23 In 2012, 32.1 African children were orphaned (of which 3.5 million having lost both parents). In 2013, two percent of Africa’s population was refugee (3.7 million) or internally displaced (12.5 million). In a sample of seven African countries, almost 1 in 10 working-age adults were disabled. Pastoralists: lack of systematic information
- 24. The vast majority of women suffers in at least one dimension; 30 percent suffer in three or more 25 26 58 82 97 4 22 58 92 14 38 69 95 0 10 20 30 40 50 60 70 80 90 100 0 1 2 3 4 Cumulativeshareofadultfemales(%) Number of dimensions deprived Non monetary poor women Monetary poor women All women 4 dimensions : - Illiterate - BMI <18.5 - Tolerant of domestic violence - Limited freedom to decide
- 25. Multiple deprivation is substantial in West African Sahel and Africa’s populous countries 26
- 26. Take Aways 27
- 27. Big Picture Take Aways (1) Findings Health, nutrition, education, and empowerment have improved, and violence has diminished. The challenges remain enormous more than 2 in 5 adults are still illiterate; quality of schooling often low almost 2 in 5 children are chronically malnourished, a burden for life conflict is on the rise Nonmonetary welfare indicators are systematically weaker in resource-rich countries better among or in households with better educated women 28
- 28. Take-Aways Methodology and data Expand beyond education and health Important practical and conceptual issues remain (Standard measures (literacy); other functionings (mobility, aspirations)) Include vulnerable groups Implications - Remove human development resource penalty - Refocus attention on education, especially for girls; adult literacy campaigns? 29
- 29. Thank you for your attention! 30
Editor's Notes
- A monetary account of poverty has appeal it accounts for preferences; it provides an objective threshold (minimum caloric diet) It allows for aggregation across domains (food & nonfood)
- Functionings is what people are and do Capabilities the capacity of people to freely choose and achieve what they have reason to be and do (i.e. the functionings) Which functionings In capability approach one determines what one values so which functionings to track? Dashboard vs single index Listing achievement by achievement vs explicit trade-offs Middle ground: share of individuals poor in 1, 2, .. K dimensions (~M0) Considers jointness, but no complete ranking (less emphasis) Indicator choice Expansion of data, though not for all functionings (mobility, capacity to aspire, social inclusion)- Remaining data concerns (availability, comparability, quality), also for common measures (literacy) -Measured at individual level Achievement if possible, inputs & conversion factors otherwise
- -Average masks substantial difference across countries. - Still high literacy rate gender gap, ddespite much progress in primary enrollment rate gender gap
- Last wave of SACMEQ and PASEC (wave 4) Pasec average of performing at or above basic knowledge: 46.82 %
- Decline in child mortality Two mortality indicators are significant drivers of changes in life expectancy in Africa: under-five mortality rates and HIV prevalence rates. For every 10 additional children per 1,000 live births surviving to the age of five, life expectancy increased by 0.7 years; for every percentage point increase in HIV prevalence, life expectancy decreased by 1 year. These two factors alone explain more than three-quarters of the variation in life expectancy in the region (under-five mortality explains 50 percentage points and HIV prevalence explains 28 percentage points). Country GDP levels and the number of deaths from conflict in previous years do not have important effects on life expectancy beyond their effects on child mortality or HIV prevalence. HIV-AIDS Southern Africa has been especially hard hit by HIV/AIDS. At least 10 percent of 15- to 49-year-olds there are HIV-positive (10.3 percent in Malawi, 10.8 in Mozambique, 19.1 in South Africa, 21.9 percent in Botswana, 22.9 percent in Lesotho, and 27.4 percent in Swaziland). Prevalence rates of 5–7 percent are observed in eastern Africa (Kenya, Tanzania, and Uganda) (map 3.1). Despite substantial progress and the increased availability of better treatment options, HIV/AIDs will continue to hold back life expectancy in a number of countries, especially in southern but also in eastern Africa
- Decline in child mortality (160 till 80 Two specific mortality indicators are significant drivers of changes in life expectancy in Africa: under-5 mortality rates and HIV prevalence rates. For every 10 additional children per 1,000 live births surviving to the age of 5, life expectancy increased by 0.7 years; for every percentage point increase in HIV prevalence, life expectancy decreased by a year. These two factors alone explain more than three-quarters of the variation in life expectancy in the region; 50.4 percentage points are explained by under-5 mortality, and 28.2 percentage points are explained by HIV prevalence. Country GDP levels and the number of deaths from conflict during the previous years did not prove quantitatively important in understanding life expectancy beyond their effects on child mortality (or HIV prevalence). The results derive from a country-fixed effect regression analysis of the life expectancy for years 2000–12 among 39 countries on the under-5 mortality rate, the HIV prevalence rate, an indicator variable taking the value of 1 if the average annual number of deaths over the five years preceding the year of recorded life expectancy exceeded 100, and GDP (constant 2005 U.S. dollars per capita) and its square term. Southern Africa has been especially hard hit by HIV/AIDS. At least 10 percent of 15- to 49-year-olds there are HIV-positive (10.3 percent in Malawi, 10.8 in Mozambique, 19.1 in South Africa, 21.9 percent in Botswana, 22.9 percent in Lesotho, and 27.4 percent in Swaziland). Prevalence rates of 5–7 percent are observed in eastern Africa (Kenya, Tanzania, and Uganda) (map 3.1). Despite substantial progress and the increased availability of better treatment options, HIV/AIDs will continue to hold back life expectancy in a number of countries, especially in southern but also in eastern Africa
- Adult women : 13 % underweight 5 % obese
- The number of deaths of violent events from declined from 20 (in the late 1990s) to 4, but the number of violent events against civilians has increased fourfold to more than 4000 in 2014 (concentrated in the central Africa and increasingly also in the Greater Horn).
- Incidence of domestic violence also declined
- Incidence of domestic violence also declined
- Africans are slightly more empowered Scores on voice and accountability rose slightly and there was a trend toward greater participation of women in household decision making processes.
- Worse off: rural and poorer populations. Not so clear about female headed households and women
- Conditional on other features, citizens in resource rich countries are less literate (by 3.1 percentage points); have shorter life expectancy (by 4.5 years), and higher rates of malnutrition among women (by 3.7 percentage points) and children (by 2.1 percentage points), and suffer more from domestic violence (by 9 percentage points
- Conditional on other features, female individuals who attended secondary education are less deprived in access to media (by 24.7 percentage points) compared to those who did not attend school. Female individuals are who attended secondary or higher education are more free to decide on their own health compared to those who did not attend school. Regarding tolerance of domestic violence, female who attended higher education are far less tolerant (by 24.3 percentage points) to to accept domestic violence compared to those who did not attend school.
- Limited freedom to decide: No media exposure or husband has final say on decisions regarding own health care, family visits, and spending) Box 3.1: Measuring Multiple deprivations with the DHS To measure multiple deprivations, we use data form the DHS on 25 countries, covering 72 percent of the population of Africa. We assess our four main dimensions as follows. We focus on women as information on these four dimensions is only available for the same individuals for reproductive age women. Deprivation in cognitive ability is illiteracy (illiterate if unable to read a full sentence, blind, or with no reading card for the required language). Over half (56 percent) of women in the sample countries are illiterate (similar to the ratio for Africa reported above). Women are classified as deprived in health if they are undernourished (BMI below 18.5). There is no direct information on life expectancy, but the correlation coefficient between country life expectancy and the proportion of undernourished adult women is 0.3. To reflect physical security, women’s attitudes toward domestic violence are used as an indicator. Across countries, social norms toward domestic violence and the incidence of casualties from political violence are correlated (a correlation coefficient of 0.4). Finally, freedom to decide is captured through an indicator for not having frequent media (not using at least one media channel—newspaper, television, radio—once a week) or not being involved in decisions regarding own health care, family visits, and spending (not involved in decisions regarding any of these three domains). Both indicators are correlated with the WGI indicator of voice and accountability (correlation coefficient of 0.4). We further augment these dimensions by adding a fifth aspect: monetary wealth. The DHS asset index is used to classify adult women as poor or nonpoor (Christiaensen and Stifel 2007; Filmer and Scott 2012; Sahn and Stifel 2000 establish correlations with expenditures). Country cutoffs are defined based on the share of the population living below $1.85 poverty headcount for the corresponding survey year. With a correlation coefficient of 0.33, on average, the correlation of this indicator with the other dimensions is low, underscoring the observation that income can mask deprivation in many basic functionings and capabilities.