The UNU-WIDER Growth and Poverty Project (GAPP)OUP Book launch at Sida, Stockholm, Sweden, April 2016

by Andy McKay and Finn Tarp

Introduction

Context• The Economist (11 May 2000): Hopeless Africa• The Independent (15 July 2009): Africa – the lost continent• The Economist (3 December 2011): The hopeful continent – Africa rising

+ Optimistic cross-country studies: Pinskovskiy and Sala-i-Martin (2014) and Young (2012)

• Recent Afrobarometer survey: ‘despite high reported growth rates, lived poverty at the grassroots remains little changed’ (Dulani et al. 2013)

• Pressing questions:– What is really happening? – What is going on at country level?– Can one make sense of all this?– Policy implications?

• The UNU-WIDER Growth and Poverty in Sub-Saharan Africa (GAPP) project

https://www.wider.unu.edu/publication/growth-and-poverty-sub-saharan-africa

Open access:

GAPP• Carried out 16 carefully designed country case studies from the 24 most populous

countries in sub-Saharan Africa:– Ethiopia, Ghana, Malawi, Rwanda, Uganda, Burkina Faso, Mozambique, Nigeria, Tanzania, Zambia,

Cameroon, Côte d’Ivoire, Kenya, Madagascar, South Africa and the DRC – Represent almost 75% of the African population and 9 of the largest 10 countries – Exact time period varies: generally the last two decades– 16 high quality teams consisting of leading local and international experts, focused on identifying and

explaining trends in monetary and non-monetary poverty and their links to growth and inequality– Synthesis and interpretation by the editors

• Key result: there is a lot to celebrate in African development (two cheers)

– But not every where, and major challenges remain (not three cheers)

Background

The African Growth Turn-Around

Child Mortality – Nearly Halved Since 1995

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 201315

35

55

75

95

115

135

155

175

195

Europe & Central Asia (developing only) East Asia & Pacific (developing only)Middle East & North Africa (developing only) Sub-Saharan Africa (developing only)Latin America & Caribbean (developing only)

Mortality rate, under-5 (per 1,000 live births)

Child Malnutrition Reduced

Europe & Central Asia (develop-ing only)

East Asia & Pacific (developing only)

Middle East & North Africa (developing only)

Sub-Saharan Africa (developing only)

Latin America & Caribbean (developing only)

0

5

10

15

20

25

30

1990 1995 2000 2005 2013

Malnutrition (weight for age) prevalence (% of children under 5)

Access to Clean Water in Rural Areas – Improved Significantly Since 1990…

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 201230

40

50

60

70

80

90

Europe & Central Asia (developing only) East Asia & Pacific (developing only) Middle East & North Africa (developing only)Sub-Saharan Africa (developing only) Latin America & Caribbean (developing only)

Improved water source, rural (% of rural population with access)

Girls’ Secondary School Enrolment – Doubled Since 1995

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 201320

30

40

50

60

70

80

90

100

Europe & Central Asia (developing only) East Asia & Pacific (developing only) Middle East & North Africa (developing only)Sub-Saharan Africa (developing only) Latin America & Caribbean (developing only)

School enrolment, secondary, female (% gross)

Substantial Variability: The 16 GAPP Country Cases

The GAPP Analytical Approach

The GAPP Approach (i)• Simple cross-country extrapolations from WDI data will not reveal underlying

dynamics or lack thereof in specific country contexts • The poverty, growth and inequality triangle

– Poverty, growth and inequality tend to be related in developing countries: helps inform the GAPP analytical approach

– Keeping in mind that the triangle is not an iron rule – It can break down: recall Abs = C+ I + G = GDP + M – X

• Poor people may not cross the poverty line even if there is growth• Growth may not be proportional across components of absorption• Terms of trade effects may imply imports fall relative to exports (similar effects due to reduction in

foreign aid, or debt service payments)• Mismeasurement

The GAPP Approach (ii)• Bring together:

– Available macroeconomic data– Comparable household budget surveys (at least two surveys from each country)– Demographic and Health Survey data, and – A host of other information (prices)

• Critically examine validity and consistency of existing data (triangulation)• Develop coherent country case stories• Presentation and discussion of drafts at WIDER meetings and conferences

Conference on Inclusive Growth in Africa

Country Categorisation and Selected Cases

Four Categories of Countries• Relatively rapid economic growth and corresponding poverty

reduction: Ethiopia, Ghana, Malawi, Rwanda, and Uganda • Relatively rapid economic growth and limited poverty reduction:

Burkina Faso, Mozambique, Nigeria, Tanzania, and Zambia• Uninspiring or negative economic growth with corresponding

stagnation or increasing poverty: Cameroon, Côte d’Ivoire, Kenya, Madagascar, and South Africa

• Low-information countries: DRC

Contextualising Four Selected Countries

0

200

400

600

800

1,000

1,200

Ethiopia Mozambique Tanzania Cameroon SSA developing average

GDP per capita (2005 USD), 2014

Note: significant differences

Contextualising Four Selected Countries (ii)• Though not all indicators have the same pattern …

Ethiopia: Agricultural Success Plus• Strong growth performance overall,

though with periodic shocks

• Despite multiple challenges: weather shocks, periods of high inflation, post-electoral instability in 2005

• Agricultural Development Led Industrialisation (ADLI) policy

• Impressive progress in agriculture modern inputs and big infrastructural improvements

• Widespread safety net

Ethiopia (ii)• Data shows poverty fell from 46.8% in 2000 to 46.0% in 2005 and 23.2% in

2011

• Both urban and rural poverty fell sharply; and both fell in each of the 11 regions; depth of poverty also fell

• Increased agricultural production a key actor: area and yields; but also growth in services sector

• And impressive improvements in many non-monetary indicators: infant mortality, malnutrition, education, access to public goods (electricity, drinking water, sanitation)

Mozambique: Off-track or Temporarily Side-lined?

• Significant growth 1994-2014 – But remember mega-projects (1%) and population growth (2.5%) -> 3.5%/year = doubling in 20 years

• Household surveys: (i) 1996/97, (ii) 2002/03, (iii) 2008/09, and (iv) 2014/15 being analysed => Results: poverty head counts (next slide) and non-monetary indicators

• Getting the story right: – Poverty estimates; GDP growth (the sub-story of agricultural production estimates); differential

impact across elements of absorption; terms of trade; and inequality estimates

• A formal assessment: head count rate would have been 45% in 2008 had it not been for combination of (i) low agriculture productivity growth (aggregate GDP growth effect and differential impact on consumption), and (ii) world price shifts (food and fuel), which also suggest rise in inequality (not captured under standard assumptions)

• Perspectives for the future (extrapolations)

Mozambique: Official Consumption Poverty Headcounts

Difference (% points)

Levels (%)

1996/97 to 2002/03 to

1996/97 2002/03 2008/09

2002/03 2008/09

National 69.4 54.1 54.7

-15.3 0.6

Urban 62.0 51.5 49.6

-10.5 -1.9

Rural 71.3 55.3 56.9

-16 1.6

Tanzania: Reconciling the Macro and Micro• Challenge of data: both macro and

micro • Reported fast growth since 2000

(stagnant before)• Drivers mostly investment

(pubic/private); government spending; trade deficit; slower growth in consumption

• Population growth• Slow growth in agriculture• And significant price increases

Tanzania (ii)• But poverty reduction much slower … national headcount 39% in

1992/3, 34% in 2007, 28% in 2011/12• Very low responsiveness to growth using macro measures, but not

using micro data – Because household real consumption increased relatively little between

surveys

• Poverty fell most in Dar es Salaam, limited elsewhere• However, progress across the country in non-monetary welfare

outcomes between 1992 and 2010

Cameroon: Inconsistent Growth and Spatial Diversity

• Cameroon is not a poor country on average (USD GDP p.c. around SSA average)

• But country faces difficult political economy tension of keeping peace in face of regional, ethnic, religious and linguistic diversity

• Public sector jobs and fiscal policy in part respond to this; less focus on service delivery

• Erratic growth record, with boom linked to exploitation of oil (1979-86), followed by crisis (1987-93) and then slow post devaluation recovery

Cameroon (ii)• Monetary poverty fell from 53% in 1996 to 40% in 2001, still 40% in 2007;

reduction in early post devaluation period, not sustained • Little progress in non-monetary outcomes (under five mortality, malnutrition,

education), 1991-2004; better between 2004 and 2011– Non-monetary outcomes also not good compared to SSA average

• Significant spatial differences in all of these indicators: much higher poverty and much worse non-monetary outcomes in north of country, followed by East– Under five mortality in three northern provinces twice the SSA average

• Monetary poverty also increased in north and east, while it fell in big cities and better connected locations

Findings and Lessons from GAPP Country Studies

Findings• Socio-economic progress in Sub-Saharan Africa has been

markedly better than almost anyone expected 20 years ago, but progress has not been even

• The development process without exception highly non-linear• The fragility of gains evident• The regional powerhouses of Kenya, Nigeria, and South Africa

not among the better performers in terms of growth or poverty reduction

Eight Lessons/Topic Areas• Peace and stability• Data issues• Volatility of monetary poverty measures• Importance of multi-dimensional assessments• The role of agriculture• Relative prices• Aid flows• The perils of existing cross-country studies

Conclusion

Outstanding Challenges and a Thorny Dilemma• Global demographic projections (2015-2050): from 7.3 to 9.7

billion, and Africa’s population is set to double to 2.5 billion (bigger than both China and India and Nigeria bigger than US)

• Structural transformation slow• Jobs and employment creation lagging• Agriculture and industrialization constrained• While the potential pitfalls are many, there is every reason to push

forward in African development over the next 15-20 years (remembering T x G = 69)

Policy Implications• While realistic expectations are required, the basic elements of a self-sustaining

development process, as opposed to a temporary growth spurt due to favourable circumstances, are starting to come into place on the sub-continent

• Forward looking investments to promote this growth process are merited• Policies to stimulate smallholder agriculture form an integral part of a coherent

growth, poverty reduction and industrialization strategy for most countries of SSA• While principal responsibility lies with country governments, the international

community should consider how to promote growth and poverty reduction across Africa

• Doing better on information systems in Africa is crucial to achieving development goals, not just tracking them

www.wider.unu.eduHelsinki, Finland