Measuring Vulnerability in Developing Countries

New Analytical Approaches

Wim Naude, Amelia Santos-Paulino, Mark McGillivray

• Publisher: Routledge
• Year: 2014
• Print ISBN: 9780415849494
• eBook ISBN: 9781317984504
• Topic: Economics
• Subtopic: Development Economics
• Index: Economics

In all of the major challenges facing the world currently, whether it be climate change, terrorism and conflict, or urbanization and demographic change, no progress is possible without the alleviation of poverty. New approaches in development economics have in recent years started from the premise that we cannot successfully deal with poverty unless we also deal with vulnerability—but not only vulnerability to income poverty but also vulnerability to various others hazards—such as climate, conflict, macroeconomic shocks and natural disasters.

This book provide insights into new approaches in conceptualising and measuring vulnerability. It includes chapters dealing with advanced issues such as the compilation of economic vulnerability indices (EVIs) on a macro-level, of conceptualizing and measuring local vulnerability across regions in a country, and of measuring the flip-side of vulnerability, namely resilience. The book also explores the sensitivities of the various measurements of vulnerability to vulnerability lines, poverty lines, and permanent income, with consideration to some of the most vulnerable groups in developing countries. Overall, the contributions in the book consolidate new approaches as far as the concept and measurement of vulnerability on different levels and outcomes are concerned, and note directions for future research.

This book was published as a special issue of Oxford Development Studies.

Measuring Vulnerability: An Overview and Introduction

WIM NAUDÉ, AMELIA U. SANTOS-PAULINO & MARK MCGILLIVRAY

ABSTRACT This paper provides an introduction to this special issue of Oxford Development Studies. It starts by contextualizing the measurement of vulnerability, pointing to the need to take risks on the level of households, regions and countries into account in designing poverty-reduction strategies. It then summarizes the papers in this special issue, highlighting the ways in which they advance the conceptualization and measurement of vulnerability, and noting directions for future research.

1. Introduction

In all of the major challenges currently facing the world, whether they are climate change, terrorism and conflict, or urbanization and demographic change, no progress is possible without the alleviation of poverty. In recent times, development studies have advanced to the stage where it is clear that we cannot successfully deal with poverty unless we also deal with vulnerability. Where poverty was initially associated exclusively with inadequate income (or consumption) in a static manner as an ex post measure of development, it is now understood as a multidimensional and dynamic concept. A strong case can be made that proper policies and strategies to deal with poverty need to be forward looking (ex ante) and be concerned with the various risks that affect whether individuals or households are in poverty, or are likely to fall into poverty¹ (e.g. Holzmann & Jorgensen, 2000). Moreover, it is not just vulnerability to poverty that matters: it is often claimed that not only vulnerability to income poverty but also vulnerability to various other hazards—such as climate, conflict, macroeconomic shocks and others—has increased.² As a result, there have been important advances in the literature, specifically in moving towards a broader but also more operational concept of vulnerability.

In addition to broadening the concept of vulnerability, the methods and scope of measurement of vulnerability towards various hazards have also seen advances in recent years. In particular, the concept of vulnerability is recognized as being relevant on the level of socio-economic groups, places and across time (Turvey, 2007). Various efforts are now underway attempting to measure vulnerability not only on a household level, but also on the level of countries, regions and local areas (Naudé et al., 2008, 2009).

Recognition of the importance of taking into account vulnerability in addressing the problem of poverty is therefore a recognition that various risks exist on various levels of group, place and period, which hinder progress in development. Conceptualizing and measuring these are prerequisites for strengthening the ability of households, regions and countries to cope with risk, and prosper even in spite of being vulnerable.

The papers in this special issue provide insights into advances in conceptualizing and measuring vulnerability, in particular household vulnerability to poverty and country and regional vulnerability to external shocks. The aim of the papers collected here is therefore to consolidate the current state of the art as far as the concept and measurement of vulnerability on different levels and outcomes are concerned, and to note directions for future research. The remainder of this overview paper is structured as follows. In the next two sections we offer, as background to the special issue, short overviews of the concept and the measurement of vulnerability in development studies. Thereafter, we provide an introduction to the five papers contained in this issue.

2. The Concept of Vulnerability³

Given the recognition that vulnerability is relevant across various outcomes and levels, a general definition sees vulnerability as the risk that a “system”, such as a household, region or country, would be negatively affected by “specific perturbations that impinge on the system” or to the probability of a “system” undergoing a negative change due to a perturbation (Gallopin, 2006, p. 294).

Different scientific disciplines have different specific definitions of vulnerability because they focus on different components of risk (Alwang et al., 2001). For instance, the International Strategy for Disaster Reduction (ISDR) defines vulnerability as “the set of conditions and processes resulting from physical, social, economic, and environmental factors, which increase the susceptibility of a community to the impact of hazards” (ISDR, 2004, p. 16). In economics, vulnerability has often been defined as the risk of households falling into or remaining in poverty because of either idiosyncratic hazards (due to characteristics of the individual household) or covariate/aggregate hazards (external to the household) (e.g. Dercon, 2005, p. 10). By focusing on hazards, and not just transient poverty but the probability of remaining in poverty (chronic poverty), it takes into account “both exposure to serious risks and defenselessness against deprivation” (Kamanou & Morduch, 2004, p. 155).

From the common definitional elements it is clear that vulnerability relates to an undesirable outcome (e.g. vulnerability to poverty, vulnerability to food insecurity or vulnerability to natural hazards) and that such vulnerability is due to “exposure to hazards”, which cause “perturbations” (Alwang et al., 2001, p. 6). These hazards can have many origins: environmental, socio-economic, physical and political. It is also clear that the “system” can imply different spatial levels of analysis that exhibit vulnerability, from micro (household), to meso (regional) and macro levels (countries, the globe).

Given that vulnerability can exist on different spatial levels and in reference to a wide variety of potential hazards, it is no surprise that there are many ways in which to measure vulnerability. The papers in this issue discuss some of the best-known measures, in particular those now pioneered in economics. We shall not therefore go into the detail of these measures here. However, given that the number of vulnerability measures may multiply in the future (especially as better data become available), it may be useful to point out here that there are several criteria that a sound measure of vulnerability should ideally satisfy.

The first is that vulnerability is an ex ante notion, so that any measure of vulnerability should have a “predictive quality” (Cannon et al., 2003). Second, measures of vulnerability should define vulnerability in relation to a socially acceptable level of outcome (Alwang et al., 2001, p. 33). Third, vulnerability indicators should ideally contain information on the causes of vulnerability and the relative importance of idiosyncratic and covariate risk (Günther & Harttgen, 2006). Fourth, a good measure of vulnerability should refer to a particular cause of vulnerability, i.e. be hazard-specific (Cannon, 2007). Fifth, to measure vulnerability appropriately, one needs to consider the dynamics of vulnerability not only before a hazard occurs, but also during and after (Birkmann, 2007). Finally, sixth, vulnerability cannot be properly assessed without assessing a system’s ways and means of coping with risk. The term “resilience” is often used to denote a system’s response to hazards/coping mechanisms (see the paper by Briguglio et al. in this issue).

3. Measuring Vulnerability

In the literature on the economics of poverty, three main methods of measuring vulnerability to poverty are to see vulnerability as (1) uninsured exposure to risk, (2) expected poverty, or (3) low expected utility (Günther & Harttgen, 2006, pp. 3–4). What these methods have in common is that they express vulnerability as being determined by the expected mean and variance of a household’s income or consumption. Recent overviews of these methods are contained in Hoddinott & Quisumbing (2003), Ligon & Schechter (2003), Dercon (2005) and Gunther & Harttgen (2006).

Economists have also been concerned to measure vulnerability from the perspective of resilience. Here ex ante and ex post coping strategies have been distinguished. Ex ante households often attempt to diversify their sources of incomes, and ex post a negative event often relies on various forms of insurance (see e.g. Fafchamps, 2003; Dercon, 2005). Generally, household assets or endowments play an important role in coping strategies or resilience and therefore much effort has gone into measuring these. These assets include natural assets (e.g. land), physical assets (e.g. infrastructure), financial assets (e.g. insurance, savings), human assets (e.g. know-how, health) and social assets (e.g. networks). The role of assets in coping has also been studied in other disciplines, and is prominent in the sustainable livelihoods approach (SLA) (see e.g. Moser, 1998). In fact, the analysis of assets as broadly defined here in the SLA approach generally starts out from the influential definition of Chambers & Conway (1992, p. 7) of sustainable livelihoods as living that can “cope with and recover from stress and shocks”. Policies and programmes to promote sustainable livelihoods therefore have much in common with policies and programmes to reduce vulnerability (Cannon et al., 2003).

Outside economics, much advance has been made in measuring vulnerability towards natural hazards (see e.g. Birkmann, 2006). In this literature, a common approach is to measure vulnerability as the degree of risk a particular household/population/region/country faces. Thus, risk is seen as a function of hazard and vulnerability. Various indicators are used to measure hazard potential (such as the occurrence of droughts, fires, earthquakes, floods) and vulnerability (such as GDP, population density, sensitive environments). Often, indicators of community resilience are added, such as levels of education, infrastructure and assets. The selection of appropriate indicators depends on the spatial level under study as well as the availability of appropriate data. Briguglio (2001) discusses a number of methods to compile a vulnerability “index”: these range from normalizing variables and taking their averages, to mapping variables on a categorical scale, to using regression methods to estimate predicted values for an index. Various vulnerability indices on the country level have been proposed since UN-DESA initiated work on the vulnerability of small island states in the early 1990s. For instance, the Commonwealth Vulnerability Index (CVI) consists of three indicators: export dependency, export diversification and susceptibility to natural disasters (Easter, 1998). The Inter-American Bank developed a Prevalent Vulnerability Index (PVI) consisting of the averages of three composite indicators for exposure or physical susceptibility, fragility and resilience. One of the most extensive vulnerability indices is the Environmental Vulnerability Index developed by UNEP and the South Pacific Applied Geoscience Commission (SOCAP), which uses over 50 indicators covering a large number of dimensions of vulnerability and resilience.

4. Overview of the Special Issue

The papers in this special issue proceed by focusing on vulnerability from macro-level to micro-level perspectives. The first paper, “An Economic Vulnerability Index: Its Design and Use for International Development Policy” by Patrick Guillaumont, starts out by noting that there has been a renewed interest in what he terms “macro vulnerability” during the past decade or so. Reasons for this interest are to be found in rising concerns about fragile states, the persistence of poverty in Africa and the Asian crisis in the late 1990s as well as the recognition that certain groups of countries, in particular Small Island Development States (SIDS) (see also McGillivray et al., 2008a, b), are inherently more vulnerable to external shocks. Interest has also been fuelled by the fact that household-level vulnerability to poverty results “to a large extent from macro vulnerability”. Recent events such as increases in fuel and food prices and global financial instability add to the concerns about macro vulnerability.

Guillaumont proceeds to discuss the historical background to one of the earliest and perhaps best-known macro-level vulnerability indices, namely the United Nations Economic Vulnerability Index (EVI). This index is described as an attempt to derive an internationally comparable measure of vulnerability to inform international development policy. According to Guillaumont, the “economic vulnerability of a country can be defined as the risk of a (poor) country seeing its development hampered by the natural or external shocks it faces”. This indicates that in this view there are two main sources of vulnerability faced by countries: (1) environmental or natural shocks such as natural hazards; and (2) external shocks related to trade and international prices. How vulnerable a country is to these would depend on: the size and frequency of these shocks; the degree of exposure to these shocks; and the capacity of the country to react to these shocks. From this he suggests that one should distinguish between structural economic vulnerability (which is exogenous) and state fragility (which is vulnerability due to inappropriate policies, institutions and weak governance).

Guillaumont discusses further the components of the UN EVI, which is a composite index calculated from seven component indices. Four of these are used to construct a “shock” index and three to construct an “exposure” index. The EVI is an equal weighting of the shock and exposure indices. The shock index consists of measures of homelessness due to natural disasters, instability of agricultural production and instability of exports. The exposure index consists of measures of population size, remoteness and specialization. Finally, Guillaumont discuss ways in which this EVI can be used for international development policy, in particular to improve aid effectiveness. He notes that further research is required on the relationship between structural vulnerability and state fragility, and suggests that measures of structural vulnerability can be used to inform aid allocation, while measures of state fragility may determine the modalities of aid provision.

The second paper in this special issue is by Lino Briguglio, Gordon Cordina, Nadia Farrugia and Stephanié Vella, and is titled “Economic Vulnerability and Resilience: Concepts and Measurements”. The aim of this paper is to extend work on measuring macro vulnerability by proposing a definition and measure of “economic resilience”. In line with developments in the vulnerability to natural hazards...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Notes on Contributors
• 1. Measuring Vulnerability: An Overview and Introduction
• 2. An Economic Vulnerability Index: Its Design and Use for International Development Policy
• 3. Economic Vulnerability and Resilience: Concepts and Measurements
• 4. Measuring the Vulnerability of Subnational Regions in South Africa
• 5. How Precisely Can We Estimate Vulnerability to Poverty?
• 6. The Impact of Violent Conflicts on Households: What Do We Know and What Should We Know about War Widows?
• Index