Understanding and Reducing Persistent Poverty in Africa

Christopher B. Barrett, Peter Little, Michael Carter, Christopher B. Barrett, Peter Little, Michael Carter

• Publisher: Routledge
• Year: 2013
• ISBN:
  9781317997467
• Edition: 1
• Topic: Betriebswirtschaft
• Subtopic: Business allgemein

Prior work has shown that there is a significant amount of turnover amongst the African poor as households exit and enter poverty. Some of this mobility can be attributed to regular movement back and forth in response to exogenous variability in climate, prices, health, etc. ('churning'). Other crossings of the poverty line reflect permanent shifts in long-term well-being associated with gains or losses of productive assets or permanent changes in asset productivity due, for example, to adoption of improved technologies or access to new, higher-value markets. Distinguishing true structural mobility from simple churning is important because it clarifies the factors that facilitate such important structural change. Conversely, it also helps identify the constraints that may leave other households caught in a trap of persistent, structural poverty.

The papers in this book help to distinguish the types of poverty and to deepen understanding of the structural features and constraints that create poverty traps. Such an understanding allows communities, local governments and donors to take proactive, effective steps to combat persistent poverty in Africa.

This book was previously published as a special issue of the Journal of Development Studies.

Understanding and Reducing Persistent Poverty in Africa: Introduction to a Special Issue

CHRISTOPHER B. BARRETT, MICHAEL R. CARTER & PETER D. LITTLE

*Cornell University, Ithaca, NY, USA, **University of Wisconsin, Madison, WI, USA, ^†University of Kentucky, Lexington, KY, USA

ABSTRACT This paper introduces a special issue exploring persistent poverty in sub-Saharan Africa. As a set, these papers break new ground in exploring the dynamics of structural poverty, integrating qualitative and quantitative methods of analysis and adopting an asset-based approach to the study of changes in well-being, especially in response to a wide range of different (climatic, health, political, and other) shocks. In this introductory essay, we frame these studies, building directly on evolving conceptualisations of poverty in Africa.

I. The Magnitude of Poverty in Africa and the Need for New Approaches

That the majority of Africans are materially poor is hardly disputable, nor very surprising. After all, the continent has been dealt a very unfavourable historical hand. A devastating and cruel global slave trade, long periods of colonial occupation, and a series of European-backed commercial ventures to exploit Africa's considerable natural wealth provided little institutional, infrastructural, and human capital when African countries began to achieve independence during the past century (Illife, 1987). More recently, cold war and post-cold war politics, prolonged conflicts, a series of structural adjustment experiments, and the HIV/AIDS pandemic have left large parts of the region poorer than even 20 years ago. Unlike East Asia, which has enjoyed a dramatic reduction in the absolute number of people living in poverty over the last 15 years, or South Asia, which has seen a sharp decline in the percentage of its population that is poor, sub-Saharan Africa has seen dramatic increases in both the total number of poor people and the fraction of its population that is poor (World Bank, 2000).

This reality has not gone unnoticed, and world leaders have rhetorically at least placed poverty reduction in Africa at the centre of global development efforts, as embodied in the United Nations Millennium Task Force report (UN, 2005). Yet the task is daunting. The most up-to-date figures available from the World Bank (2005) imply a population-weighted poverty gap of 42 per cent for sub-Saharan Africa as a whole (including the Republic of South Africa) relative to the $2 per-day per capita international poverty line.¹ Summed together, a poverty gap of this magnitude adds up to a shortfall of more than $200 billion per year to bring all sub-Saharan Africans up to the modest $2 per day standard of living.² Equal to two-thirds of the region's gross annual income and nearly ten times current global aid flows to Africa, this staggering sum underscores the need for a rigorous, strategic focus on how to propel self-reinforcing growth among the poor.

There is a general sense that past approaches, perhaps especially those predicated on simply getting the macro economy and prices ‘right’ – the preoccupation of donor agencies in the 1980s and 1990s – have failed to generate the broadly based economic growth needed for sustainable poverty reduction (for example, see Williamson, 2003). This recognition has in turn motivated a search for better understanding of the micro-and meso-level constraints that limit economic growth and poverty reduction.

This special issue represents an interdisciplinary attempt at advancing such an understanding, at helping shed light on how and why some African households have managed to escape or avoid persistent poverty, while others have not. This focus necessarily highlights the heterogeneity of the poor, distinguishing between different categories of poverty, the variety of structural constraints faced by distinct groups of poor peoples, and alternative growth trajectories. Such a disaggregated, dynamic approach is necessary in order to establish how donors, governments and nongovernmental organisations (NGOs) might most effectively stimulate self-reinforcing growth among the poor.

The remainder of this introductory essay is organised as follows. Section II highlights novel perspectives and methods offered by the contributions to this special issue. Section III places the contributions to this issue in the context of the evolving understanding of poverty in Africa. Section IV summarises the studies’ primary empirical findings on persistent poverty. Section V concludes with reflections on the nature and design of persistent poverty reduction strategies.

II. Novel Perspectives and Methods

Several features distinguish the set of studies in this issue from previous work in this area. First, each of the papers employs longitudinal data that permit the authors to identify those households and individuals who are getting ahead from those who are not, distinguishing transitory poverty caused by predictable life cycle effects and external events (for example, drought) from more enduring impoverishment related to the structural constraints of economies that do not work for their poorest members. While prior work has used longitudinal data to establish that a subset of the poor enter and exit ‘transitory’ poverty with some frequency,³ the papers in this special issue focus on the non-transitory poverty of structurally poor households. The goal is to distinguish who amongst the structurally poor is positioned to move ahead over time within the existing structure, who is not, and what structural modifications or other interventions are needed to nudge the latter subpopulation onto a pathway out of poverty. In the terminology of the Carter and Barrett paper, this set of papers tries to move beyond familiar, first and second generation poverty analysis, to third and fourth generation approaches that bring into focus the structural determinants of poverty and the dynamics of those underlying structural positions.

A second novel feature of the studies presented here is that they share an asset-based approach to the study of poverty dynamics. Economists typically analyse poverty with reference to flow variables, income or expenditures reflecting budget constraints and consumption choices, respectively. However, flow measures tend to be more subject to considerable measurement error than stock variables, even in well-run surveys, because they can only rarely be directly observed and verified. Moreover, productive assets are the durable inputs used to generate income and offer the collateral base for expenditures based on credit rather than income. The stock of productive financial, physical, natural, social and human assets that households and individuals control largely determines their structural position in a society, and their ability to avoid poverty, or to escape from it if they find themselves falling backwards in the face of adverse shocks. As the Barrett et al. contribution demonstrates, the random noise inherent to flow-based measures can mask important features of the dynamics of household-level well-being. Understanding the dynamics of assets is thus fundamental to understanding persistent poverty and longer-term socio-economic dynamics.

The challenges posed by this asset-based approach to understanding poverty dynamics lead to a third distinguishing feature of the studies in this issue: they employ multiple methods and disciplinary approaches, mixing qualitative and quantitative information to better understand the problems of persistent poverty. The existing literature on African poverty has typically emphasised one or the other: the interpretive and historical (see Broch-Due and Anderson, 1999, Illife, 1987) or the quantitative, especially econometric dimensions (Grootaert and Kanbur, 1995; Baulch and Hoddinott, 2000). Recent work on poverty analysis has nonetheless underscored inherent methodological complementarities between qualitative and quantitative approaches in (cross-sectional) poverty analysis (Kanbur, 2003).

The papers in this issue try to take this integration to the next level, exploring the synergies between different ways of studying the dynamics of poverty and well-being across communities and countries. Measuring human well-being is inherently problematic, and becomes further complicated when trying to measure intertemporal changes in well-being. In this context, triangulation using multiple methods offers important advantages over single methodology approaches. That said, incorporating both qualitative and quantitative methods without strongly privileging one over the other is difficult to achieve. The Adato et al. article on poverty dynamics in South Africa gives one example of how to achieve this by comparing a poverty transition matrix based on econometric analysis with one constructed using qualitative methods.

A fourth and final distinguishing feature of the work in this issue is that it offers an in-depth look at a wide range of countries from across Africa, encompassing periods of political, climatic, and economic policy shocks that have had the potential to sharply impact poverty. These include successive and devastating droughts (Ethiopia, Kenya, Malawi and Zimbabwe), major political transitions (Ethiopia, Ghana, Madagascar, Malawi, South Africa and Uganda), and a range of different economic reform programmes (Ghana, Kenya, Madagascar and South Africa). Each of these macro-level processes have been previously linked with changes in poverty (Grootaert and Kanbur, 1995; Collier and Gunning, 1999; Carter and May, 2001; Elbers et al., 2002; Hulme and Shepherd, 2003; Dercon, 2004; Lybbert et al., 2004; Sachs et al., 2004). The explicit linkage of poverty dynamics to a wide range of different shocks (climatic, health, political, and other) represents an important step forward in linking the typically disparate literatures on risk and growth. In this sense we try to build directly on evolving conceptualisations of persistent poverty in Africa.

III. Evolving Conceptualisations of Poverty in Africa

The conceptualisation and measurement of poverty have been substantially rethought since most African states became independent in the early 1960s. Earlier anthropological assessments of poverty and rural differentiation highlighted labour and its organisation (‘wealth in people’) as the main measure of wealth, and control over people as the key to accumulating wealth (Guyer and Belinga, 1995; Meillassoux, 1981). The so-called ‘wealth in people’ argument remains important in relatively land abundant/low population savannah farming areas, as the Whitehead and Peters contributions emphasise. The converse problem, that the loss of labour power can precipitate a collapse into poverty, has become especially widespread in recent years with the rise of the HIV/AIDS pandemic and of increasingly drug-resistant malaria. As Krishna et al. document, health shocks are, in many places, the leading reason why previously non-poor households suffer a structural decline into persistent poverty. The ‘wealth in people’ perspective is therefore resurgent in much contemporary analysis of African poverty.

However, as population pressure has grown in important sections of Africa, and as rural economies have become more diversified, control over land, non-farm employment, and other key resources have become increasingly significant measures of wealth status. Ownership of livestock assets has been (and still is) a key wealth indicator in many parts of Africa, especially in the arid and semi-arid lands, as Barrett et al., Hoddinott and Little et al. document. Land and livestock have long been especially important in determining household food security and marketable surpluses.

One reason endowments of assets such as labour, land and livestock matter is that they condition households’ ability and willingness to take advantage of emerging opportunities (see the Carter and Barrett paper). Technological change has played a central role in improvements in all measures of human well-being – income, life expectancy, health and nutritional indicators – throughout recorded human history (Fogel, 2004). Technological change in agriculture and natural resource management has played an especially important role in rural poverty reduction, not least of which in recent decades in Asia and Latin America (David and Otsuka, 1994; Datt and Ravallion, 1998; de Janvry and Sadoulet, 2002; Ravallion and Datt, 2002; Evenson and Gollin, 2003). Yet, the Green Revolution largely bypassed sub-Saharan Africa, with patterns of adoption of improved agronomic and natural resources management practices and higher yielding technologies closely associated with households’ endowments of land, labour and livestock (Barrett et al., 2002). The opportunities afforded by liberalised domestic and international trade can likewise increase well-being through specialisation of production according to patterns of comparative advantage, helping reduce poverty where the micro-foundations of local factor and product markets permit poor households to seize the new opportunities (de Janvry and Sadoulet, 1993; Barrett and Carter, 1999). Where technologies remain rudimentary and terms of trade do not improve, households commonly remain poor.

The attention paid to markets and technologies as prospective handmaidens of rural poverty reduction has spawned increased attention to the geographic and socio-political determinants of poverty that condition market access and uptake of improved technologies. The relevant literature on Africa has focused especially on biophysical characteristics, such as how humidity and temperature affect agricultural productivity and health, and how population density, road infrastructure and distance to ocean ports affect commerce at the individual firm level (Sachs and Warner, 1997; Gallup and Sachs, 1998; Bloom and Sachs, 1998), but also on ethnic divisions, histories of political violence and patrimonial rule, and the complex, long-term effects of colonialism (Bates, 1981; Easterly and Levine, 1997; Collier and Gunning, 1999; Herbst, 2000; Acemoglu et al., 2001, 2002). The conclusion of much of this literature is that areas less favoured by nature and by states (both colonial and modern) have commonly become geographic poverty traps plagued by widespread destitution with limited opportunities for households to escape. Several papers in this volume offer evidence in support of this hypothesis of geographic poverty traps (Barrett et al., Little et al. and Whitehead).

The observation of geographically-based poverty pockets has motivated increased attention to the more general problem of persistent poverty in Africa. While much of the recent empirical work has been done by economists (Grootaert and Kanbur, 1995; Dercon, 1998; Baulch and Hoddinott, 2000; Carter and May, 2001; Elbers et al., 2002; Deininger and Okidi, 2003 Dercon, 2004; Lybbert et al., 2004), contributions have also come from anthropologists (Anderson and Broch-Due, 1999), political scientists (Chambers, 1997; Krishna et al., 2004), sociologists (Hulme and Shepherd, 2003), geographers (Watts, 1991) and historians (McCann, 1999).

Like many concepts in development studies, the term ‘persistent poverty’ – and synonymous terms such as ‘chron...