Impacts of Policies on Poverty

The Definition of Poverty

Module 004
CONCEPTUAL AND TECHNICAL MATERIALS

Impacts of Policies on
Poverty
The Definition of Poverty
Impacts of Policies on
Poverty
The Definition of Poverty

by
Lorenzo Giovanni Bellù, Agricultural Policy Support Service, Policy Assistance
Division, FAO, Rome, Italy
Paolo Liberati, University of Urbino, "Carlo Bo", Institute of Economics, Urbino, Italy

for the

Food and Agriculture Organization of the United Nations, FAO

About EASYPol
EASYPol is a an on-line, interactive multilingual repository of downloadable resource materials for
capacity development in policy making for food, agriculture and rural development. The EASYPol home
page is available at: www.fao.org/tc/easypol.
EASYPol has been developed and is maintained by the Agricultural Policy Support Service, Policy
Assistance Division, FAO.

The designations employed and the presentation of the material in this information product do not imply
the expression of any opinion whatsoever on the part of the Food and Agriculture Organization of the
United Nations concerning the legal status of any country, territory, city or area or of its authorities, or
concerning the delimitation of its frontiers or boundaries.

© FAO November 2005: All rights reserved. Reproduction and dissemination of material contained on
FAO's Web site for educational or other non-commercial purposes are authorized without any prior
written permission from the copyright holders provided the source is fully acknowledged. Reproduction of
material for resale or other commercial purposes is prohibited without the written permission of the
copyright holders. Applications for such permission should be addressed to: copyright@fao.org.
Impacts of Policies on Poverty
The Definition of Poverty

Table of Contents

1 Summary ................................................................................... 1

2 Introduction ............................................................................... 1

3 Conceptual background ................................................................ 1

3.1 Basic definition of poverty............................................................. 2
3.2 Uni-dimensional versus multi-dimensional poverty ........................... 3
3.3 Absolute versus relative poverty .................................................... 4
3.4 Poverty versus food security ......................................................... 5

4 A step-by-step procedure to define poverty..................................... 7

5 Examples on poverty definition...................................................... 9

5.1 Jamaica’s Food Stamp Program.....................................................9
5.2 Poverty in the Russian Federation ...............................................11
5.3 Income poverty in advanced countries .........................................12
5.4 Mono versus multi-dimensional definition of poverty ...................12
5.5 Variability of the poverty concept across space..............................13

6 Conclusions ...............................................................................13

7 Readers’ notes ...........................................................................13

7.1 Time requirements.................................................................... 13
7.2 EASYPol links ........................................................................... 13
7.3 Frequently asked questions ........................................................ 14

8 References and further readings ...................................................14

Module metadata ..............................................................................16

Impacts of Policies on Poverty 1
The Definition of Poverty

1 SUMMARY
This module illustrates how poverty can be defined in the context of policy impact
analysis. After reporting and discussing the definition of poverty as “the lack of, or the
inability to achieve, a socially acceptable standard of living”, it discusses the mono-
dimensional and multi-dimensional approaches to the definition of poverty.
Furthermore, the module focuses on the absolute and the relative concept of poverty,
also drawing some analogies and differences with the concept of food security. A step-
by-step procedure, illustrated real case examples, are then provided to guide the reader
through the process of poverty definition for policy impact analysis.

2 INTRODUCTION

Objectives
The aim of this module is to briefly illustrate the main approaches to define poverty,
distinguishing relative and absolute concepts. It will give the user the possibility to learn
the basic differences between these two concepts as well as the appropriate information
to discriminate between them. In addition, the user will learn the main logical steps to
follow when defining poverty in the context of policy impact analysis.

Target audience
The module targets current or future policy analysts who want to work on poverty
issues.

Required background
In order to capture the essential parts of this module, the audience should be familiar
with some basic mathematics, statistics and economics and with the basic concepts of
income and of income distribution.

It belongs to a set of modules that discuss how to define poverty for policy analysis,
how to identify poor and how to measure poverty. To find relevant materials in these
areas, the reader can follow the links included in the text to other EASYPol modules or
references1. A set of useful links to related EASYPol modules is provided in a section at
the end of the document.

3 CONCEPTUAL BACKGROUND
Many public policies impact poor people. Furthermore, some policies are specifically
designed to target the poor in order to fight poverty. In policy work it is therefore
important to be able to identify the poor, to simulate the impact of alternative policies

1
EASYPol hyperlinks are shown in blue, as follows:
a) training paths are shown in underlined bold font;
b) other EASYPol modules or complementary EASYPol materials are in bold underlined italics;
c) links to the glossary are in bold; and
d) external links are in italics
2 EASYPol Module 004
Conceptual and Technical Materials

on poverty, and to rank policy options according to some poverty impact indicators in
order to select the most preferred option. Some tools used for inequality analysis also
prove to be useful for poverty analysis. To this respect, the poverty issue may be seen as
focusing attention at the bottom end of the income distribution, i.e. as a part of the more
general problem of income inequality.

3.1 Basic definition of poverty

Poverty is not a self-defining concept. Experts and academics have suggested many
definitions over time. For example, poverty could be the lack of command over
commodities in general2; alternatively, it could be the lack of command over some basic
goods (e.g., food and housing). More generally, Sen, 1985, argued that poverty is the
lack of «capability» to function in a given society. All these definitions point to poverty
as a status in which a reasonable standard of living is not achieved. A synthesis of the
various positions has been made by the World Bank3:

Poverty is the lack of, or the inability to achieve, a socially

acceptable standard of living.

It is worth discussing some keywords contained in this definition.

LACK Æ The base case situation for the definition of poverty is that where individuals
lack command over economic resources. For example, an individual may be
considered poor if he/she lacks basic food or shelter or, equivalently, if he/she lacks
income to buy these basic needs.

INABILITY Æ This is best associated with the capability failure to participate in a

society, a concept developed by Sen (1985). Roughly, the idea is that the individual has
a space of «functionings», where these latter are either what a person is actually able to
do (realized functionings) or the set of alternatives he/she has (real functionings).
This space may be very basic (food, shelter) or complex (freedom, self-respect, social
inclusion, etc.). Inability to achieve these functionings makes the individual poor. For
example, disability not only reduces ability to earn income (which means lack of
command over resources) but also makes it harder to convert income into functionings
(even though, in terms of income, that achievement is potentially feasible). According
to this view, poverty is a state characterised by levels of capabilities that are, in the view
of society, unacceptably low.

STANDARD OF LIVING Æ Poverty depends on:

a) What is deemed to constitute a socially acceptable standard of living by a given

society at a given time. In a society where most people own cars, the use of public
transport may be a signal of poverty. Not having a TV in a technologically advanced
society might again be an indicator of poverty, while in other countries it may be a
luxury good;

2
Watts, 1968.
3
See, World Bank, 2001, Poverty Manual, (Chapter 1: The Concept of Poverty and Well-Being).
Impacts of Policies on Poverty 3
The Definition of Poverty

b) How this standard is measured, i.e, what is the variable or the set of variables used to
“capture” the standard of living.

3.2 Uni-dimensional versus multi-dimensional poverty

The standard of living, and therefore poverty, may be represented by a UNI-
DIMENSIONAL indicator (e.g. income) or a MULTI-DIMENSIONAL approach (e.g. income,
health conditions, family status, etc.). In the first case, poverty is defined by income
poverty and the standard of living is defined in the space of economic welfare, a
narrower concept than well-being. In the second case, the concept of poverty is closer to
well-being, where other welfare indicators support income in defining poverty. Any
choice entails additional problems.

ƒ In the case of the uni-dimensional approach, you have to define what the
appropriate single monetary indicator for standard of living is. There are two natural
candidates here, income and expenditures. Total expenditures are often used as an
indicator of poverty, as they better reflect the concept of permanent income of an
individual.
ƒ In the case of the multi-dimensional approach, you have to define what the
appropriate list of poverty indicators is and how to weigh them in order to get a
comprehensive vision of poverty. For example, if you have low incomes and good
health, are you richer than an ill individual with more income? If you are illiterate,
yet in good health and with enough food, are you poor? An alternative view of the
multi-dimensional approach could be that of «explaining» poverty with a set of
indicators, leaving the task of defining how poverty is explained by which factor4 to
statistical techniques. This raises the objection that simple correlation is not a causal
relation: are you poor because you are in bad health? Or are you in bad health
because you arepoor? The answers to these questions may entail quite different anti-
poverty policies.

Because of the various possibilities to define poverty, it is worth stressing that:

It is important to tailor the concept of poverty on the appropriate

context, as there is not a general concept that we can safely assume to
hold for all countries at all times.

In what follows, the focus will be on «lack of command over economic resources» and
on a uni-dimensional approach. The «capability» approach, although stimulating, has
given rise to an enormous strand of literature, most of which rather technical.
Furthermore, this latter approach is harder to make operational in terms of identifying
poor and measuring poverty and is left for more advanced material.

Finally, an issue that will not be pursued any further in the text, a slightly different
perspective is taken by UNDP, where poverty is taken to be a denial of human rights,
i.e. those rights that are inherent to the person and belong equally to all human beings.
4
An example of this approach is the principal component analysis. See, for example, Cavatassi et al.
2004. One of the most popular approaches to measure welfare in a multi-dimensional perspective is the
Human Development Index (HDI) developed by the United Nations.
4 EASYPol Module 004
Conceptual and Technical Materials

According to UNDP, the human rights-based approach to poverty reduction espouses

the principles of universality and indivisibility, empowerment and transparency,
accountability and participation. It addresses the multi-dimensional nature of poverty
beyond the lack of income5. UNDP (2003) states that the respect for human rights is a
necessary condition for socio-economic outcomes. This human rights-based definition
challenges, to some extent, the opportunity that poverty is to be measured by a uni-
dimensional criterion based on income and/or expenditure levels6.

3.3 Absolute versus relative poverty

Poverty may be thought of as either an ABSOLUTE or RELATIVE concept.

The ABSOLUTE concept of poverty refers to a standard of living defined in absolute

terms. In this case, poverty is usually measured by the value, in real terms, of a given
level of goods ensuring some form of minimum subsistence (e.g., the value of basic
food or the minimum income required to have decent lives). The first attempts to define
poverty as an absolute concept have taken into account the minimum diet cost, i.e. the
minimum cost, for each household, to achieve a given energy intake7. This approach has
two fundamental shortcomings: a) minimum diet costs may vary among households as
they do not all share the same preferences of nutrition patterns; b) non-food items are
not considered8.

The RELATIVE concept of poverty, on the other hand, refers to a standard of living
defined in relation to the position of other people in the income/expenditure distribution.
In this sense, poverty is basically a phenomenon of inequality. For example, one could
define as poor those individuals that have incomes below 50 per cent of the average
income of the society. Therefore, if average income grows because richer people gain
more, people in relative poverty might increase. This concept automatically reflects
changing social and economic conditions in a given country. The main shortcomings of
this approach is that if poverty is defined as a fixed percentage of some synthetic
indicator of the income distribution (e.g., mean or median), there will be no possibility
to eradicate poverty, unless the income distribution becomes perfectly egalitarian9.

Absolute and relative concepts of poverty have been widely debated in specialised
literature. The main result is that «poverty is neither a strictly absolute nor a strictly

5
See, for example, UNDP, 2003.
6
The multi-dimensional nature of poverty is developed by UNDP through the use of the Human
Development Index (HDI), the Gender-related Development Index (GDI) and the Human Poverty Index
(HPI).
7
This approach has been pioneered by Rowntree, 1901, and Booth, 1902, in Great Britain. The most
popular application of an absolute concept of poverty in industrialised countries is by Orshansky, 1965,
for the US. For a basic and descriptive review of the concepts of absolute and relative poverty, see Seidl,
1998. A critique of the absolutist view of poverty has been developed by Rein, 1970. Poverty concepts
are also discussed for the UNDP by Lok-Desallien, 1997. This latter paper is one of a series of Technical
Support Documents (TSD) developed by UNDP in 1997.
8
The main aspects of these methods are discussed in the EASYPol Module 005: Impacts of Policies on
Poverty: Absolute Poverty Lines.
9
This concept has been developed by Fieghen et al., 1977. who argued that with a relative definition of
poverty, «the poor are always with us».
Impacts of Policies on Poverty 5
The Definition of Poverty

relative concept»10. On the other hand, Sen, 1983, attempts to put in a hierarchy the two
approaches. He defines poverty as “absolute deprivation”, that may be interpreted as a
preference for the absolute concept of poverty when a contradiction arises between
absolute and relative concepts of poverty11. Sen basically rejects the relative concept of
poverty, arguing that there is an irreducible absolutist core in the idea of poverty,
regardless of relative positions (e.g., hunger and starvation may unequivocally be
indicators of poverty).

In addition, Sen, 1983, argues that absolute and relative concepts should not be
confused with VARIABILITY OVER TIME, although variability over time introduces, in
the poverty concept, some elements of relativity. He argues that there is a difference
between achieving relatively less than others and achieving absolutely less than others.
On this ground, he rejects relativity. However, he recognizes that achieving absolutely
less may change over time, according to major changes of society and the economy.
Absolute poverty concepts may therefore be subject to variations over time, as poverty
is always a function of variables reflecting social and economic conditions. For
example, non-food items thought of as non-essential at one stage of the development,
may become essential some years later, calling for an integration of the list of goods
ensuring minimum subsistence. This updating of the absolute concept of poverty,
however, is less automatic in nature than the updating of the relative concept, where the
link with the income distribution prevailing in a given year is parametric (e.g.
50 per cent of mean income).

Absolute poverty concepts could also be adjusted for variability over space, when
comparing poverty across countries. For example, meat may be included in a minimum
subsistence basket in industrialised economies and not in less industrialised countries.
Using the same basket in different circumstances might be quite misleading about the
relative level of poverty across countries. In this sense, absolute concepts of poverty, at
a given time, are relative to the prevailing economic conditions.

This leads us to state that:

Both the absolute and the relative concepts of poverty may be variable
in time and space.

3.4 Poverty versus food security

Of interest is also the issue of how to relate the definition of poverty to that of food
security. By food security, we mean the physical and economic access at all times to
sufficient, safe and nutritious food to meet the dietary needs and food preferences for an
active and healthy life12. Strictly related to this definition is also that of vulnerability,

10
Seidl , 1988, p. 79.
11
Seidl, 1988, p. 79, defines this concept as a lexicographic preference for absolute concepts of poverty.
12
See, World Food Summit, 1996. See also, the IFAD, 1996. The paper for the World Food Summit
suggests that increased agricultural production and rural incomes be complemented with a strategy to
move rural investment projects further towards improved nutrition.
6 EASYPol Module 004
Conceptual and Technical Materials

which refers to people’s propensity to fall, or stay, below a pre-determined food security
threshold13.

Food insecurity may therefore fall closely related to an absolute concept of poverty, in
particular to those absolute concepts emphasising the lack of a given “level” of food as
a primary source of poverty. On the other hand, food security depends on three main
factors: availability, stability and accessibility. Availability and stability may be
thought of propaedeutical to guarantee accessibility. Access can be organised if food is
available and stable. To make food available, a given country should be able to either
grow or import sufficient food. Furthermore, this process should be relatively stable
over time. But the final step is that households should have sufficient income (or other
entitlements) to purchase food. Absolute concepts of poverty, therefore, mainly focus on
the ability of households to purchase food. They are therefore most related to
accessibility, rather than availability and stability.

In this sense, food security, in terms of access, and absolute concepts of poverty are
closely related. On the other hand, food security may be thought of as a wider concept
than those underlined by absolute poverty, as it requires policies for availability and
stability of food. The former (availability) is related to the way in which food is
produced, imported and stored; the latter (stability) is related to the way in which food is
made available (market functioning and integration) and to the stock management.
Access, instead, is mainly related to how households can buy food (purchasing power,
resources, etc.)14.

The discussion so far carried out shows that the concepts of absolute and relative
poverty, even though logically distinct, have common features. The above definition by
the World Bank refers to the inability to achieve a “socially acceptable standard of
living”. Note that this socially acceptable standard of living can be defined only by
making recourse to a value judgement on what it is considered socially acceptable at a
given time and in a given space. In communities where hunting is the primary source of
food, a socially acceptable standard of living could be assured by having enough meat
or fish to eat at the end of the day. In industrialised societies the socially acceptable
standard of living could not be achieved unless everybody owns a TV or goes to school.
Therefore, the inability to achieve any absolute standard (whatever defined) often
implies some relative view of poverty.

Analogously, the concept of food insecurity, i.e. the situation where people fall below a
pre-determined food security threshold, requires a value judgement (i.e. to pre-
determine) on what is deemed to be the appropriate threshold, e.g. what is considered an
“active and healthy life”. There is no doubt that the appropriate threshold of food may
vary enormously across different societies. Romer Løvendal et al. (2004; 3), explicitly
recognize that “the nature of this minimum welfare threshold depends on the outcome in
which one is concerned”.

13
See, for example, Romer Løvendal et al., 2004. On food security, see also the material released by the
Food and Insecurity Vulnerability Information and Mapping System (FIVIMS), FAO.
14
Extensive work on food security can be found also in eleven technical guides released by IFPRI, 1999.
Impacts of Policies on Poverty 7
The Definition of Poverty

Finally, the definition of poverty given by Sen, 1992, in terms of “unacceptably low
levels of capabilities” (see above) leaves us with the question of what “unacceptably
low” means. The most likely answer is that “unacceptably low” is related to what a
society, at a given point in time, considers as an unacceptably low level, which involves
a value judgment that contains some elements of relativeness.

The following paragraph will help you to have a comprehensive view of the two
approaches.

4 A STEP-BY-STEP PROCEDURE TO DEFINE POVERTY

In order to define poverty the following steps must be taken in mind:

ƒ Analyse the context in which poverty analysis has to be carried out and the
specific goals of the poverty analysis. Poverty analysis should start with a macro-
view of the area or society to be analysed. In particular, the geographic distribution
of the main socio-economic variables should be spelled out and their variability over
time and space assessed. In the meantime, the goals of the poverty analysis have to
be clarified: what type of policy is under investigation? Is the policy under
investigation likely to have different impacts in different areas? Are the policy
impacts likely to be different in different points in time?

ƒ Choose the time and space dimension of the analysis. On the basis of the
preliminary considerations carried out in step 1, the analyst chooses the dimension
of the analysis, i.e. whether it is more meaningful to assess poverty over time, across
space or at a given point in time. For example, the stability of socio-economic
indicators across different areas of a given country may suggest a nation-wide
analysis. Alternatively, their variability among regions may suggest a poverty
assessment by areas 15 . In the same vein, stability or variability over time may
suggest a static or dynamic analysis, respectively 16 . On the other hand, policies
aimed at, or likely to affect in different ways various areas may suggest a detailed
area analysis.

ƒ Identify likely welfare-sensitive items. This step amounts to identifying a set

economic resources the lack of which are likely to cause the inability to achieve a
“socially acceptable standard of living”. On the basis of the analysis carried out in
Step 1, here you have to define the appropriate list of variables that are likely to
cause inability to achieve an acceptable standard of livings (e.g. income, food,

15
To this purpose, the technique of poverty mapping is found extremely useful when significant
geographic variation of the standard of living is expected in a given country. Poverty mapping involves
techniques that permit sufficient disaggregation of a poverty measure to local administrative levels or
small geographical units. The methods used vary from participatory poverty profiles to sophisticated
econometric techniques, but are more related to poverty measurement than to the definition of poverty.
On this topic, see Davis, 2003, and Petrucci et al., 2003. See also Deichmann, 1999.
16
An example of poverty assessment over time and space for Costa Rica can be found in Cavatassi et al.,
2004.
8 EASYPol Module 004
Conceptual and Technical Materials

shelter, etc.). In doing this, consider that over a significant time span, the
characteristics of goods may change (from necessity to luxury goods, etc.). Also
consider that across space, some goods may be more important in, say, rural areas
than in urban areas. Therefore, as far as the required information is available, it is
important to select those goods that have less variability over time and across space
in order to get meaningful results in terms of comparable standard of living. On the
other hand, different goods may enter the list in some years or in some regions. For
example, if electricity is available to all urban households but only to a fraction of
rural households, it may be taken as an appropriate variable to measure the standard
of living of rural households, much less to measure the standard of living of urban
households. This task is usually accomplished by browsing household expenditures,
if available. Data may reveal the consumption pattern of different households in
different areas and with different size and composition. This may help selecting
necessity goods. Alternatively, you can resort to interviewing households and/or
expert judgement. In any case, you should take into account that obtaining
information on which specific goods are needed may be much more difficult than
obtaining information on total income and/or expenditures.

ƒ Select crucially welfare-sensitive items, within those identified in Step 3, which

define the standard of living. The list of items obtained by Step 3 can be very long.
You may have to shorten it, in order to provide a clearer picture of poverty. The
easiest way is to use just one indicator, e.g. income (a uni-dimensional approach).
But in particular cases, it may be worthwhile to add some other indicators to income
to define poverty, like health status, educational level, availability of some services,
etc., as they may give contradictory indications, at least at an aggregate level (a
multi-dimensional approach – see example below)17. Note, however, that indicators
other than income, at a household level, are often correlated with income.
Households with higher incomes are usually in better health, relatively more
educated and dispose of basic services (water, electricity, etc.). When data are
available at household level, it is therefore strongly suggested to look at income first
and, possibly, to complement the analysis with a correlation between income and
other non-income indicators. This may help in depicting poverty more clearly.

ƒ Choose whether to adhere to an absolute or relative concept of poverty. There is

no a definite solution to this issue. The proper choice may be taken considering what
has been developed at point 1. Note that this choice is aimed at setting a benchmark
against which poverty indicators are then assessed. For example, assuming that,
income is set as the proper indicator, if the absolute poverty concept is chosen. The
benchmark may be, say, US$2/day. On the other hand, if the relative poverty
concept is chosen, the benchmark may be, say, 30 per cent of mean income. It is
apparent that relative poverty is automatically updated to income growth and
therefore more suitable for time comparisons. Absolute poverty is instead relatively
more useful when poverty is compared across space, as it provides a common
threshold against which measuring the situation of different countries or regions.

17
Useful references for a multi-dimensional approach are Filmer and Pritchett, 1998, and Cavatassi et al.,
2004, both using a principal component analysis. The principal component analysis is a factor analysis
aimed at reducing a given number of variables by extracting a linear combination which best describes
them and at transforming them into one index.
Impacts of Policies on Poverty 9
The Definition of Poverty

Figure 1, below, gives a synoptic view of these steps to define poverty.

Figure 1 - A step-by-step procedure to define the concept of poverty

STEP OPERATIONAL CONTENT

1 Analyse the environment in which

poverty is to be measured. Tailor the
concept of poverty to the context by
using a macro-analysis fo the relevant
socio-economic variables

2 Choose the dimension of the analysis. A

single point in time? Over time? Over
space? Refer to the results of Step 1

3 Define the economic resources for which

lack of command is causing inability to
achieve acceptable standard of living.
Income? Goods? Take into account the
results of Step 1

4 Select what are judged the best

indicators within those listed in Step 2.
Define whether the approach should be
unidimensional (e.g. income) or
associated to other social indicators (e.g.
health status, education level, etc)

5 Choose the concept of poverty. Absolute

or relative? Consider Step 1

5 EXAMPLES ON POVERTY DEFINITION

Let us now describe some real-world studies where the steps identified above can be
either explicitly or implicitly followed.

5.1 Jamaica’s Food Stamp Program18

This is a World Bank paper where the authors assess the anti-poverty effect of the
Jamaican Food Stamp Program, implemented by the Jamaican Government in 1984.

Step 1 – Context and goals of the poverty analysis. In order to assess the anti-poverty
effect of the Jamaican Food Stamp Program, the authors start by investigating the policy
context (p.2). They say that: «Jamaica’s policy reform efforts cover over a fifteen year
span and have yielded mixed results. The oil shocks of the early 1970s coupled with the
decline in prices of bauxite – the major export crop – resulted in a decline in real

18
Ezemenari and Subbarao, 1998.
10 EASYPol Module 004
Conceptual and Technical Materials

GDP and labor income in 1973 and 1980. In order to minimize the adverse
consequences of a decline in income, the government increased public expenditures on
social services, expanded employment in the public sector, increased the money
supply and wages. As a result, government expenditure as a percent of GDP increased
from 5 percent to 24 percent of GDP. The share of subsidies in national disposable
income increased from 2 percent to 6 percent between 1976 and 1977. By 1980, the
fiscal deficit as a percent of GDP was 18 per cent. In 1984, the government
implemented strict fiscal austerity programs and cut public employment, labor costs,
and social services. Minimum wages fell by 11.8 percent from September 1983 to July
1985. Fifty of 330 public entities were phased out. Price controls and subsidies were
removed, and minimum farm-gate prices for key export crops were introduced.
Generalized food subsidies were replaced by a targeted food assistance to the poor.
Fiscal and monetary policies continued to be expansive, fueling inflation. The
exchange rate was fully liberalized in October 1991, which was followed by a
substantial devaluation of the Jamaican dollar. The Government of Jamaica
implemented the Food Stamp Programme in 1984.»

Step 2 – Time and space dimension of the analysis. Since the study concentrates on
the impact of a specific policy on poverty in Jamaica, the spatial dimension of analysis
is the whole country. As data on food stamp programs and households are available
from 1989 to 1991, the time dimension of the analysis is three years. The authors justify
this, by saying that «the period 1989-1991 was chosen because there was a gradual
devaluation of the Jamaican dollar during this period ending in the major devaluation of
1991» (p.15). Note that, ideally, the time dimension of this analysis could have been
extended from 1984 to more recent years. However, because of lack of data, the time
span was limited to three years. This suggests that both time and space dimension may
sometimes be constrained by the availability of information.

Step 3 – Likely welfare-determining items. The kind of economic resource in which

authors are interested, is the consumption expenditure before and after the Food
Stamp Program. A level of consumption expenditure below a given threshold is taken as
an indicator of poverty. The authors explicitly argue that «household expenditures are
used as the measure of welfare».

Step 4 – Crucial welfare-determining items. In step 3, the authors do not provide a

list of items without which cause an “unacceptable standard of living”. However, they
implicitly agree on the fact that insufficient household expenditures causes the inability
to achieve an acceptable standard of living. In this simpler case, therefore, there is no
problem of selection. From the presence of only one indicator, it follows that the
approach is uni-dimensional.

Step 5 – Absolute versus relative poverty. The concept of poverty chosen by the
authors is implicitly absolute and defined in terms of expenditure. They experiment two
absolute expenditure levels as thresholds for acceptable standard of living: J$ 6,198 and
J$ 3,657 (both at 1991 prices). The fact that they choose absolute levels of expenditures
means that thresholds for acceptable standard of living would not be automatically
updated with a growth in household expenditure. This choice may be due to various
reasons. One may be that other economic parameters linked to standard of living in
Impacts of Policies on Poverty 11
The Definition of Poverty

Jamaica are defined in terms of absolute income levels, as it is the minimum wage (see
Step 1).

5.2 Poverty in the Russian Federation19

This paper, prepared within the University of Amsterdam (Netherlands) analyses
household poverty in Russia.

Step 1 – Context and goals of the poverty analysis. The analysis of the socio-
economic framework in which the poverty analysis has to be carried out is not directly
accomplished by the authors. Nevertheless, they refer to a series of studies on the
Russian Federation where a macro-view of the socio-economic context is offered.

Step 2 – Time and space dimension of the analysis. The spatial dimension of the
analysis is confined to a nation-wide view of the Russian Federation. The time
dimension is dictated by available data and confined to two years, 1997 and 1998.

Step 3 – Likely welfare-determining items. The authors try to take into account
different sets of economic resources. In two cases, the indicator is mainly expressed in
monetary terms: the first is the total financial situation of the family; the second is
monthly income net of taxes. In another case, in order to find an overall index of
satisfaction with one’s own life, they complement income indicators with “all aspects
of life”, such as: “marital happiness”, “number of friends”, “having a partner”, “having
work”, “family size” and “age”.

Step 4 – Crucial welfare-determining items. In the two cases where only the
monetary indicator is judged the best indicator of standard of living, the approach is uni-
dimensional. In the case where other aspects of life are included as determinants of the
standard of living, the authors end up to select just one of them, family size. In this case,
the approach is bi-dimensional in the sense that it takes into account income and family
size at the same time.

Step 5 – Absolute versus relative poverty. The authors note that the official poverty
line, in Russia, is defined as the cost of achieving a subsistence food basket multiplied
by 1.3. This official measure of standard of living somehow embodies both absolute and
relative elements. The absolute element is cost of a “minimum” food basket, which can
be considered a sort of subsistence food requirement.. The relative element is that an
“acceptable standard of living” should take into account also non-food items, the reason
why the cost of the food basket is multiplied by 1.3. This means that if the cost of food
basket increases, the acceptable standard of living is automatically updated. The authors
also estimate by their own acceptable standard of living in absolute terms with an
econometric procedure.

19
Ferrer-i-Carbonell and Van Praag, 2000.
12 EASYPol Module 004
Conceptual and Technical Materials

5.3 Income poverty in advanced countries20

In this “Luxembourg Income Study (LIS)” working paper, the authors analyse income
poverty in more industrialised countries. Also in this case, some of the steps of the
analysis.are clearly spelled out.

Step 1 – Context and goals of the poverty analysis. The focus on advanced countries
leaves the authors with some space to assume that these countries are relatively
homogenous in terms of social and macroeconomic variables. This may be seen as a
shortcut for Step 1. However, pooling some countries for poverty analysis always
requires that you understand the basic analogies and/or differences among them. This
should never be a random mix of countries.

Step 2 – Time and space dimension of the analysis. The spatial dimension of the
analysis is defined by the set of “advanced countries”. The time dimension of the
analysis covers all years available from the Luxembourg Income Study, on which the
study is based.

Step 3 – Likely welfare-determining items. The variables defining the absence of an

“acceptable level of living standards” are clearly listed by the authors, when they refer
to general approaches to poverty analysis. They list: a) disposable income; b) consumer
expenditures; c) consumption; d) earnings capacity; e) wealth; f) choice sets; g) utility;
h) capabilities. The authors argue that the ideal variables might differ from what one
can empirically use (due, for example, to availability of data). For example, lifetime
“utility” could be a good proxy for investigating poverty; capabilities may better reflect
what people can do and not only what people have.

Step 4 – Crucial welfare-determining items. The complexity of most items listed

above is explicitly recognized by the authors. Therefore, they end up with using
income as a relevant economic resource the lack of which may undermine the standard
of living. The main justification is that data are more easily available and that income is
one of the most “common” indicators. With income as the only indicator, they adhere to
a uni-dimensional approach to poverty analysis.

Step 5 – Absolute versus relative poverty. Most of the authors’ analysis is carried out
by using 60 percent of the current median income as a threshold to discriminate poor
and non-poor. Therefore, they adopt the concept of relative poverty. Income growth
would automatically update the threshold.

5.4 Mono versus multi-dimensional definition of poverty

The following example is taken from Gillespie, 1990, and is aimed at illustrating how
income and other social indicators may give contradictory prescriptions in a low-income
country. Pakistan’s per capita income, at that time, was about US$ 350. The average
income per capita of low income countries was US$ 290. According to this indicator,
Pakistan should not be considered as a poor country. However: a) primary school
enrolment rates were 61 per cent for boys and 32 per cent for girls, below the

20
Jäntti and Danziger, 1999.
Impacts of Policies on Poverty 13
The Definition of Poverty

corresponding figures for low income countries; b) Pakistan had a ranking of four
among the highest percentages of low birth weight babies; c) maternal mortality was
among the highest in the world; d) the female to male ratio was the lowest in the world.

5.5 Variability of the poverty concept across space

An interesting example of how the definition of poverty may vary across countries may
be taken from the experience of the European Union (EU). In assessing poverty, as
reported in de Vos and Zaidi, 1998. Two alternatives are discussed: a) the European
Union-wide concept of poverty; b) the country-specific concept of poverty. They argue
that if the EU is to be considered as one society, the level of resources necessary to
reach the minimum level of living in the society is the same throughout the Union.
However, in reality there are serious doubts as to whether the EU can really be
considered as one society with one minimal level of living. An inhabitant of
Luxembourg with the average resources of a Portuguese would definitely be considered
poor according to the standards of her society, while someone living in the countryside
of Portugal with the resources of the average Luxembourger would certainly be
considered well-off.

6 CONCLUSIONS
The discussion about absolute versus relative poverty, the variability of the poverty
concept over time and across space and the link between poverty and food security have
been discussed in qualitative terms. This discussion is preliminary to quantitative
poverty analysis. The accurate analyst will possibly experiment with different concepts
of poverty, in order to make his/her analysis robust to changes in the way poverty is
defined.

7 READERS’ NOTES

7.1 Time requirements

Time required to deliver this module is estimated at about two hours.

7.2 EASYPol links

This module is propaedeutical to the investigation of other poverty issues. The modules
included herebelow follow a logical order:
ƒ EASYPol Module 005: Impacts of Policies on Poverty: Absolute Poverty Lines
ƒ EASYPol Module 006: Impacts of Policies on Poverty: Relative Poverty Lines
ƒ EASYPol Module 007: Impacts of Policies on Poverty: Basic Poverty Measures
ƒ EASYPol Module 009: Impacts of Policies on Poverty: Distributional Poverty
Gap Measures
ƒ EASYPol Module 010: Impacts of Policies on Poverty: Generalised Poverty Gap
Measures
ƒ EASYPol Module 035: Impacts of Policies on Poverty: Dominance and Poverty
14 EASYPol Module 004
Conceptual and Technical Materials

7.3 Frequently asked questions

9 How should poverty be defined? Poverty can be defined as the lack or the inability
to achieve a socially acceptable standard of living. See paragraph 3.1 for different
facets of this phenomenon.
9 Is poverty merely a lack of income or can it be a more general lack of command
over other economic and social resources? This is related to uni-dimensional or
multi-dimensional notions of poverty. Income may command more control on many
economic resources; nevertheless, poverty may be related to the absence of other
fundamental aspects of life (see, for example, the human rights-based approach by
UNDP or the capability approach by Sen). See paragraph 3.1.
9 Is relative poverty a more satisfactory concept than absolute poverty? A
consensus has not yet been found on this question. Note that also absolute concepts
contains some elements of relativity in the way the acceptable standard of living is
defined. See paragraph 3.2.
9 What is the relation between poverty and food security? Food security may be
thought of as a component in an absolute concept of poverty. On the other hand,
poverty, as an individual or household feature, is is mostly related to the “access to
food” dimension of food security, rather than to availability and stability.
9 Can the concept of poverty take into account non-food items? Non-food items
may well constitute the set of necessary choice to achieve an acceptable standard of
living (e.g. shelter). The main problem is how to choose these elements. See
paragraph 4 and some examples in paragraph 5.

8 REFERENCES AND FURTHER READINGS

Booth A. L., 1902. Life and Labour of the People of London, MacMillan, London, UK.
Cavatassi R., Davis B., Lipper L., 2004. Estimating Poverty Over Time and Space:
Construction of a Time-Variant Poverty Index for Costa Rica, ESA Working
Paper n. 04-21, FAO, Rome, Italy.
Davis B., 2003. Choosing a Poverty Mapping Method, FAO, Rome, Italy.
De Vos K., Zaidi M. A. 1998. Poverty measurement in the European Union: Country-
Specific or Union-Wide Poverty Lines?, Journal of Income Distribution, 8(1),
pp. 77-92.
Deichmann U., 1999. Geographic aspects of inequality and poverty, text for the World
Bank’s Web Site on Inequality, Poverty, and Socio-Economic Performance,
(also available at: http://www.worldbank.org/poverty/inequal/index.htm)
Ezemenari K., Subbarao K., 1998. Jamaica’s Food Stamp Program: Impacts on Poverty
and Welfare, Poverty Group, PREM Network, paper presented at the EDI
Workshop, The World Bank, Washingon DC, USA, December.
Ferrer-i-Carbonell A., Van Praag B. M. S., 2000. Poverty in the Russian Federation,
mimeo, University of Amsterdam, Amsterdam, The Netherlands.
Fieghen G. C., Lansley P. S., Smith A. D., 1977. Poverty and Progress in Britain 1953-
1973, Cambridge University Press, Cambridge, UK.
Impacts of Policies on Poverty 15
The Definition of Poverty

Filmer D., Pritchett L., 1998. Estimating Wealth Effects Without Expenditure Data – or
Tears: An Application to Educational Enrollments in States of India, mimeo, The
World Bank, Washington DC, USA.
FIVIMS, Tools &Tips (available at:
http://www.fivims.net/static.jspx?lang=en&page=toolsandtips)

Gillespie N., 1990. Selected World Bank Poverty Studies: A Summary of Approaches,
Coverage and Findings, Poverty Reduction Handbook, The World Bank,
Washington DC, USA
IFAD, 1996. Paper for the World Food Summit, November, Rome, Italy
IFPRI, 1999. Technical Guides, Washington DC, USA (available at:
http://www.ifpri.org/training/material_tech.htm)

Jäntti M., Danziger S., 1999. Income poverty in advanced countries, LIS Working Paper
n. 193, Maxwell School of Citizenship and Public Affairs, Syracuse University,
New York, USA. (also available at: http://www.lisproject.org/publications/liswps/193.pdf)
Lok-Desallien R., 1997. Review of Poverty Concepts and Indicators, UNDP, New
York, USA.
Orshansky M., 1965. Counting the Poor: Another Look at the Poverty Profile, Social
Security Bulletin, 28/1, pp. 3-29.
Petrucci A., Salvati N., Seghieri C., 2003. The Application of a Spatial Regression
Model to the Analysis and Mapping of Poverty, FAO, Rome, Italy.
Rein M., 1970. Problems in the Definition and Measurement of Poverty, in Townsend
P. (ed.), The Concept of Poverty, Heinemann, Oxford, UK, pp. 46-63.
Romer Løvendal C., Knowles M., Horii N., 2004. Understanding Vulenrability to Food
Insecurity Lessons from Vulnerable Livelihood Profiling, ESA Working Paper
n. 04-18, FAO, Rome, Italy.
Rowntree B. S., 1901. Poverty: A Study of Town Life, MacMillan, London, UK.
Seidl C., 1988. Poverty Measurement: A Survey, in Bös D., Rose M., Seidl C. (eds),
Welfare and efficiency in public economics, Springer, New York, USA. pp. 71-
147.
Sen A., 1983. Poor, Relatively Speaking, Oxford Economic Papers, 35, Oxford
University Press, UK, pp. 153-169.
Sen A., 1985. Commodities and capabilities, North-Holland, Amsterdam, The
Netherlands.
Sen A., 1992. Inequality re-examined, Clarendon Press, Oxford, UK.
UNDP, 2003. Poverty reduction and human rights – A practice note, New York, USA.
Watts H. W., 1968. An Economic Definition of Poverty, in Moynihan D. P. (ed.), On
Understanding Poverty, Basic Books, New York, USA.
World Bank (2001), Poverty Manual, Washington DC, USA (also available at:
http://web.worldbank.org/WBSITE/EXTERNAL/WBI/WBIPROGRAMS/PGLP/0,,contentMDK:20284
296~menuPK:461269~pagePK:64156158~piPK:64152884~theSitePK:461246,00.html )
World Food Summit, 1996. Declaration on World Food Security Plan of Action, Rome,
FAO/World Food Summit, Rome, Italy.
16 EASYPol Module 004
Conceptual and Technical Materials

Module metadata

1. EASYPol module 004

2. Title in original language

English Impacts of Policies on Poverty
French Impacts des politiques sur la pauvreté
Spanish
Other language

3. Subtitle in original language

English The Definition of Poverty
French Définition de la pauvreté
Spanish
Other language

4. Summary
This module illustrates how poverty can be defined in the context of policy impact analysis.
After reporting and discussing the definition of poverty as “the lack of, or the inability to
achieve, a socially acceptable standard of living”, it discusses the mono-dimensional and
multi-dimensional approaches to the definition of poverty. Furthermore, the module focuses
on the absolute and the relative concept of poverty, also drawing some analogies and
differences with the concept of food security. A step-by-step procedure, illustrated real case
examples, are then provided to guide the reader through the process of poverty definition
for policy impact analysis.

5. Date
November 2005

6. Author(s)
Lorenzo Giovanni Bellù, Agricultural Policy Support Service, Policy Assistance Division, FAO,
Rome, Italy
Paolo Liberati, University of Urbino, "Carlo Bo", Institute of Economics, Urbino, Italy

7. Module type Thematic overview

Conceptual and technical materials
Analytical tools
Applied materials
Complementary resources

8. Topic covered by Agriculture in the macroeconomic context

the module Agricultural and sub-sectoral policies
Agro-industry and food chain policies
Environment and sustainability
Institutional and organizational development
Investment planning and policies
Poverty and food security
Regional integration and international trade
Rural Development
Impacts of Policies on Poverty 17
The Definition of Poverty

9. Subtopics
covered by the
module

10. Training path Analysis and monitoring of socio-economic impacts of policies

11. Keywords