Understanding Poverty: Definitions, Measurement, and Approaches, Exams of Nutrition

Download Understanding Poverty: Definitions, Measurement, and Approaches and more Exams Nutrition in PDF only on Docsity!

CHAPTER ONE

INTRODUCTION TO POVERTY

1.1 INTRODUCTION

The 2006 Nobel Peace laureate Muhammad Yunus argues that poverty should not be part of any civilized society. In a speech delivered on the occasion of receiving an Honorary Doctorate from the University of Venda in May 2006, he commented:

Poverty deserves only to be in the museums, where small children can see it in the future and be shocked how we allowed such an inhuman condition to exist for so many people for so long (Yunus 2006:1).

He indicated that he was very excited when the United Nations announced the Millennium Development Goals at the United Nations Summit in New York in 2000. The central objective of the Millennium Goals, which 149 countries agreed to, is halving poverty by 2015. Yunus (2006) believes that creating a world free from poverty is possible, as many poor people can get themselves out of poverty if they are given the same opportunities as those afforded to people who are not poor.

In theory, a healthy human being born in South Africa is fully equipped, not only to take care of himself or herself, but also to contribute to the development and well-being of the country as a whole. Unfortunately, in reality, 40% of the population in South Africa are classified as being in extreme poverty and over 60% of black South Africans live in underdeveloped rural areas (UNDP 2000).

The first democratically elected government in South Africa in 1994 inherited one of the most unequal societies in the world. One of the consequences of the legislation of the past, especially of laws such as the Group Areas Act, was that it created townships located on the outskirts of the cities, far from white residential and business areas. The Job Reservation Act restricted black people’s income, irrespective of their ability and educational qualifications. As a result, developed white areas enjoyed a per capita income comparable to that of an upper middle income in a developed country, while the majority of the population experienced extreme poverty in terms of their income and

inequality high on its development agenda. South Africa is one of 23 countries that have had their national anti-poverty plans assessed by the UNDP to identify the obstacles to reaching the target and to highlight successful actions. South Africa has set the target date of 2020 for reducing extreme poverty to 0%.

Unfortunately, South Africa, unlike some other countries, has no viable poverty monitoring system yet (UNDP 2000). Morocco is an example of a country that has used a sophisticated system of indicators to determine the poorest provinces and then to identify the most deprived communities within these provinces (UNDP 2000).

The measurement and analysis of poverty, deprivation, inequality and vulnerability are crucial for several reasons. Firstly, for cognitive purposes, it is vital to know what the situation is, in other words, who is poor and where are the poor located in the country? Secondly, for analytical purposes, it is helpful to understand the underlying factors contributing to poverty. Thirdly, for policy-making purposes, it is important to be able to measure and analyse the situation in order to be able to assist the relevant parties in introducing interventions to improve the quality of life of the individuals and households that are affected by poverty. Finally, for monitoring and evaluation purposes, measurement and analysis are needed to assess the effectiveness of the chosen policies in eradicating poverty.

The aim of this study is to develop multi-dimensional techniques to identify the most deprived households and communities. The methods that countries use to determine income poverty tend to differ across countries and this makes comparisons difficult. The method of comparison developed in this study will help to measure the effectiveness of poverty alleviation programmes and strategies in poor communities.

1.2 DEFINITION OF POVERTY

The definition of poverty is very complex. A definition is difficult to formulate because poverty means different things to different people. Some people may define poverty as a lack of income resulting in the absence of a car or refrigerator, while others may describe it as a lack of formal housing, basic services or opportunities for training and employment. According to the Oxford English Dictionary (1989), the adjective “poor” means “lacking adequate money or means to live comfortably”. The noun “poverty” is defined as “the state of being poor” and as a “want of the necessities of life”. Other definitions for poverty and being poor include expressions such as having a “deficiency in”, “lacking of”, “scantiness”, “inferiority”, “want of”, “leanness or feebleness”, and many more.

Historically, the idea that some people are trapped in poverty while others have short spells in poverty was a central element of poverty analysis. Social commentators in eighteenth-century France distinguished between the pauvre and the indigent (Hulme and Mckay 2005). The pauvre experienced spells in poverty, such as seasonal poverty when crops failed or the demand for casual agricultural labour was low. The indigent was trapped in poverty and continued to remain permanently poor because of ill health (physical and mental), the results of an accident, age or alcoholism. The central aim of the policy was to support the pauvre in ways that would stop a person from becoming indigent.

From the above it is clear, firstly, that poverty and the poor are associated with a state of want and deprivation and, secondly, that such deprivation is related to the necessities of life. Thus, the term “poverty”, in its daily use, implies a comparison between the condition of a household or person on the one hand and the perception of the person who speaks or writes about what is necessary to sustain life on the other.

Poverty is hunger. Poverty is lack of shelter. Poverty is being sick and not being able to see a doctor. Poverty is not being able to go to school and not knowing how to read. Poverty is not having a job, is fear of the future, living one day at a time. Poverty is losing a child to illness brought about by unclean water. Poverty is powerlessness, lack of representation and freedom.

The World Bank definition of poverty has not changed much from the definition of poverty by Godard (1892) in the nineteenth century.

In the current study, poverty is regarded as the measurement of well-being and deprivation, that is, the more deprived a household is, the poorer the household.

1.2.1 Horizontal and Vertical Vagueness of Poverty

In multi-dimensional poverty studies, there is no consensus as to what the dimensions of poverty should be or how many dimensions are adequate. The following are some examples of dimensions of poverty: a lack of nutrition, housing, safety, clothing and health, income, education, literacy, sanitation and clean drinking water.

Some dimensions contribute more to poverty than others, depending on the time and place, and this is referred to as the horizontal vagueness of poverty (Qizilbash 2002).

There is no consensus on where or how to distinguish between the poor and those who are not poor in each dimension. So, for example, individuals differ in their nutritional requirements, depending on their age, sex, height and weight. This implies that there is no clear threshold where nutritional poverty starts or where it ends.

There is also no consensus as to which level of education is acceptable, since the requirements of society may differ from place to place. Qizilbash (2002) refers to this as

the vertical vagueness of poverty. This vagueness of poverty contributed to a large extent to the debate on and difficulty of measuring poverty.

1.2.2 Income Poverty and Human Poverty

The poverty report of the United Nations Development Programme (UNDP 2000) distinguishes between income poverty and human poverty. Income poverty can be divided further into extreme poverty and overall poverty. Extreme poverty or absolute poverty is the lack of income necessary to satisfy basic food needs, usually defined on the basis of minimum calorie requirements.

Figure 1.2.1: Income inequality in South Africa in 2007

Source: Sunday Times (2007)

The monetary approach is the one most frequently used to define and measure poverty. A poverty line is defined in terms of the monetary income sufficient for a person to attain a minimal standard of living. A person whose income falls below the poverty line is considered to be poor. The World Bank estimate for the poverty line is $2 per person per day for developing countries. In South Africa, the poverty line for households was set at R800 per household per month in 1996 prices.

The capabilities approach pioneered by Amartya Sen emphasizes that income is only valuable in so far as it increases the capabilities of individuals and thereby permits them to function in their society. According to the capabilities approach, poverty is pronounced to be a deprivation in well-being. Many poor people in South Africa live without the fundamental freedoms of action and choice that more affluent people take for granted. The poor often lack adequate food and shelter, education and health, deprivations that keep them from leading the kind of life that everyone values. They are extremely vulnerable to ill health and natural disasters such as floods and fires. Many of them are ill-treated by institutions of state and society and are powerless to influence key decisions affecting their lives. These issues are all dimensions of poverty. The experience of multiple deprivations is intense and painful and many believe that it is impossible to escape poverty.

All these forms of deprivation severely restrict what Sen (1976) calls the “capabilities that a person has, that is, the substantive freedoms he or she enjoys to lead the kind of life he or she values”. The ultimate objective is to have capabilities such as the ability to lead a long life, to function without chronic morbidity, to be capable of reading, writing and performing numerical tasks and to be able to move from place to place. According to this approach, a person whose capabilities or functioning falls below a minimum acceptable standard is poor. The resources required to achieve the same capabilities can vary from person to person. A capability poor person may not necessarily be income poor. The capability approach is much broader and addresses the neglect of social goods in the monetary approach.

The social exclusion approach emphasizes relations between individuals. Social exclusion occurs when individuals or groups are unable to participate fully in the society in which they live. As a result of exclusion, the income capabilities or other characteristics of the poor become unacceptably distant from the norms of their community. In terms of the exclusion approach, poverty is a social construct and has little to do with the fulfilment of the individual’s minimum needs. This is often a characteristic of groups rather than individuals, for example, of women, the aged, and the handicapped or particular racial or ethnic categories. In South Africa the cities and towns are well developed with the local municipalities able to provide basic services to the majority of households. Unfortunately, in the rural communities several households are severely deprived of educational opportunities, housing and basic services.

A participatory approach takes into account the views of poor people themselves. The people themselves decide what it means to be poor and that determines the magnitude of poverty. Van Praag (1978) introduced this approach to the measurement of poverty based on sample surveys about the perception of poverty of the people interviewed for his study. The study conducted by the South African Participatory Poverty Assessment is a good example of a participatory approach. This approach leads to perceived relative personal welfare rather than to a perceived poverty index and is therefore not discussed in more detail in this study.

1.3 LITERATURE REVIEW ON POVERTY

Several studies on poverty have been conducted in nearly every country in the world. For the sake of brevity, in this section, only the studies that promote the development of the different poverty measures are discussed. The relevant developments are listed chronologically.

The concept of poverty was first introduced, over a century ago, by Booth (1892) and Rowntree (1901).

particular aspect of poverty. This approach is applicable to dichotomous variables, polytomous variables and continuous variables.

Schokkaert and Van Ootegem (1990) were the first to operationalize the capability approach using micro data. They applied the capability approach on 1979 data on the unemployed in Belgium. They showed that material factors are almost irrelevant in the determination of the well-being of the unemployed, thus providing support for a broad concept of well-being.

Slottje (1991) used 20 indicators to compute a well-being index for 126 countries. The study showed that the world rankings of the quality of life index vary when the information from several economic well-being indicators is aggregated into one summary index.

Smeeding et al. (1993) compared the incidence of poverty among Organization for Economic Cooperation and Development (OECD) countries by assigning a monetary value for each of the welfare attributes of housing, education and healthcare. Estimating the per capita cost of primary, secondary and university education and allocating these costs to each individual in a household that completed a certain level of education allowed them to obtain the distribution across households of education services.

Ellman (1994) studied the sharp decline in living standards after the collapse of the USSR and argued that there were severe negative effects on mortality and morbidity over the period from 1987 to 1993.

Cheli and Lemmi (1995) modified the Totally Fuzzy Approach suggested by Cerioli and Zani (1990) and named their proposed method the Totally Fuzzy and Relative Approach. This approach has the advantage of taking a relative approach to poverty according to some dimension, where one is usually poor in respect of some other individual.

Balestrino (1996) analysed whether a sample of officially poor people indicated that they were functioning poor, income poor or both. Of the 281 Italian households in his sample, 73 households were pure functioning poor (in other words, they lacked education, nutrition or suffered some health failure), 71 were pure income poor and 137 were both. The analysis suggested that a sizeable portion of the poor in affluent societies is actually not income poor.

Ruggeri-Laderchi (1997) tested to what extent an income indicator can capture some of the most essential functioning (education, health and child nutrition). He used 1992 Chilean data. The test concluded that the income variable appears to be an insignificant determinant for the shortfall in the three selected functioning areas. Hence, poverty analysis is highly dependent on the indicators chosen and thus “the approach should be kept as broad as possible in order to more fully capture the multi-dimensional nature of such a complex phenomenon” (Ruggeri-Laderchi 1997:345).

Vero and Werquin (1997) suggested a further fuzzy approach to poverty measurement. Their method adjusts for certain indicators that may be highly correlated in the multi- dimensional measure of poverty.

The UNDP (1997) introduced the Human Poverty Index (HPI) as an example of a multi- dimensional index of poverty in terms of functioning failure. The HPI aggregates the country level deprivations into the living standards of a population for the basic dimensions of life, namely, decent living standards, educational attainment rate and life expectancy at birth.

Brandolini and D’Alessio (1998) used the Bank of Italy’s 1995 household survey, which covered six functioning areas (health, education, employment, housing, social relationships and economic resources). This exercise provided an interesting picture of the distribution of the achievements and deprivation of functioning. They also

Adams and Page (2001) compared the performances recorded for each welfare indicator for several countries in the Middle East and North America using aggregate data from the World Bank. The comparison observed no clear relationship between a reduction in monetary poverty and an improvement in other welfare indicators.

Balestrino and Sciclone (2001) tested the strength of the correlation between income and functioning on a regional comparison of well-being in Italy. Their study showed that the functioning-based ranking and income-based rankings are strongly positively correlated.

Lelli (2001) did an empirical test on the Panel Study of Belgian Households, and found that an analysis with fuzzy sets or factor analysis makes little difference if the same variables are selected.

Robeyns (2003) assessed gender inequality in Western societies in terms of functioning and capabilities using the British Household Panel Study to make a quantitative empirical application. This study found that women are disadvantaged on more dimensions than men, but enjoy better social relations than men.

Qizilbash (2002) used fuzzy set theoretic measures to rank South African provinces in terms of financial and human poverty. The human poverty criterion contained some capability-like dimensions, and some resources that served as proxies for capabilities. He showed that the provinces’ ranking changed considerably, depending on whether one focuses on household expenditure or on the capability-related multi-dimensional poverty measure. The study concluded that the picture obtained from looking at household expenditures alone can be highly misleading.

Table 1.3.1 summarizes the poverty research conducted in South Africa since the first democratically elected government came into power in 1994. The techniques and data sets used in the different studies are listed.

Table 1.3.1: Poverty studies in South Africa (1994-2006)

Author Techniques Data sets used Comments Klasen (1997) Income based analysis SALDRU 1993 Kwa Zulu Natal McIntyre et al. (2000)

General index of deprivation using principal component analysis

Census 1996 Magisterial level

Hirschowitz et al. (2000)

Household infrastructure index using factor analysis

Census 1996, using 11 indicators

South African provinces

Klasen (2000) Composite deprivation index

SALDRU 1993,

using 14 indicators

9000 households in South Africa Ngwane et al. (2001)

CHAID Analysis OHS 1995 South African provinces Qizilbash (2002) Borda score, Human poverty index

Census 1996 South African provinces Ngwane et al. (2003)

Head count index, Watts measure

OHS 1995 South African provinces UNDP (2003) Service deprivation index

Census 2001 Nationally by province, race and gender Bhorat et al. (2004)

Asset and service deprivation

Census 1996, Census 2001

South African provinces Van der Walt (2004)

TFA,

TFR

Census 1996 Districts of the Eastern Cape Oosthuizen and Nieuwould (2002)

FGT poverty indices OHS 1995 Western Cape

Keller (2004) Heckprobit analysis OHS 1995, IES 1995

African males in South Africa May et al. (2004) Micawber threshold, Asset index

KIDS 1998 Kwa Zulu Natal

Leibbrandt and Woolard (1999)

FGT poverty indices LSDS 1993, IES 1995

South African provinces Naidoo et al. (2005)

TFR Census 1996, Census 2001

South African provinces Woolard (2005) Shorrocks rigidity index, Gini index

PSLSD 1993,

KIDS

Kwa Zulu Natal

Noble et al. (2006)

Provincial indices of multiple deprivation

Census 2001 South African provinces

1.5 MULTI-DIMENSIONAL MEASURES OF POVERTY

The study of poverty is commonly oversimplified, because the manifestation of poverty is perceived as dichotomous. Poverty is conventionally analysed by splitting the households in a population into two groups: poor and non-poor, defined in relation to the poverty line.

Poverty should be regarded as a multi-dimensional phenomenon of which income is only one aspect. The study of poverty should be supplemented by a number of sets of non-monetary indicators of deprivation which can then be used to understand the different types of hardship experienced by households. The multi-dimensionality of poverty is now internationally recognized, as is clear from the World Bank’s (2001) report on poverty and the adoption of social indicators by the European Union.

Deutsch and Silber (2005) detail a systemic comparison of the following four approaches to multi-dimensional poverty analysis:

• a fuzzy set approach,
• a distance function approach,
• an information theory approach, and
• axiomatic derivations of multi-dimensional poverty indices.

The current study introduces the neural network approach to poverty measurement using self-organising maps. The Kohonen vector quantization method, the Kohonen self- organizing maps and the Batch self-organizing maps are discussed.

1.5.1 The Fuzzy Set Approach to Poverty Analysis

Zadeh (1965) introduced the theory of fuzzy sets on the basis of the idea that certain classes of objects may not be defined by precise criteria of membership, such as cases where one is unable to determine which elements belong to a given set and which do not. He characterized a fuzzy set as a class with a continuum of grades of membership.

The fuzzy set approach may be easily applied to the concept of poverty. Some households are in such a state of deprivation that they should certainly be considered poor, while others have such a level of welfare that they should certainly not be classified as poor. There are some households where it is not clear whether the household is poor or not. This is especially true when one takes a multi-dimensional approach to poverty measurement, where, according to some criteria, one would define the household as poor, whereas, according to other criteria, one should not regard the household as poor. Such a fuzzy approach to the study of poverty has taken various forms in the literature.

Cerioli and Zani (1990) applied the concept of fuzzy sets to the measurement of poverty. Their approach is called the Totally Fuzzy Approach. The idea is to take into account a whole series of variables that are supposed to measure a particular aspect of poverty. In the analysis of poverty there are several qualitative variables that may take more than two values. In such cases, the first step is to assume that one may rearrange these values by increasing order, where higher values denote a higher risk of poverty. They defined membership functions for three categories of variables: dichotomous variables, polytomous variables and continuous variables. When the membership function takes the value of one, it indicates a condition of absolute deprivation, while a membership value of zero indicates the absence of deprivation.

Cheli and Lemmi (1995) suggested the Totally Fuzzy and Relative Approach as a modification of the Totally Fuzzy Approach. This method takes a relative approach to