1. The Seminar on Poverty Statistics,
organized by the secretariat of the Economic and
Social Commission for Asia and the Pacific (ESCAP)
, was held from 21 to 23 June 1999 in Bangkok.
Financial support for the Seminar was provided
by the Government of the Netherlands, while the
United Nations Development Programme, through
its country offices, provided support for the
attendance of several participants.
A. Attendance
2. The Seminar was attended
by 30 participants from the following 18 members
of ESCAP: Bangladesh, Cambodia, China, Fiji, India,
Indonesia, Kazakhstan, Lao People=s
Democratic Republic, Malaysia, Mongolia, Myanmar,
Nepal, Pakistan, Philippines, Singapore, Sri Lanka,
Thailand, and Viet Nam.
3. The Seminar was also attended
by representatives of the United Nations Development
Programme (UNDP), United Nations Population Fund
(UNFPA), United Nations Children=s
Fund (UNICEF), Food and Agriculture Organization
of the United Nations (FAO), International Labour
Organization (ILO), Asian Development Bank (ADB),
International Council on Social Welfare (ICSW),
and the Statistical Institute for Asia and the
Pacific (SIAP).
B. Opening of the Seminar
4. The Seminar was inaugurated
by Mr Adrianus Mooy, Executive Secretary of ESCAP.
In his opening statement, the Executive Secretary
expressed his gratitude to the Government of the
Netherlands for its generous financial support
which enabled the secretariat to organize the
Seminar and to the United Nations Development
Programme (UNDP) for providing financial support
for several of the participants. He also acknowledged
the excellent cooperation of the United Nations
Economic Commission for Latin America and the
Caribbean (ECLAC), ADB, FAO and the ILO in providing
technical papers and reference material for the
Seminar.
5. The Executive Secretary noted
that over the years the number of poor persons
had been increasing steadily, particularly in
the developing countries throughout the world.
The largest proportion of the world's poor was
concentrated in the ESCAP region where the recent
economic crisis in Asia had resulted in deterioration
of the poverty situation.
6. During the past several decades,
governments had been according high priority to
alleviating poverty and had therefore been allocating
more resources for that purpose. However, in many
countries the formulation and implementation of
appropriate poverty alleviation programmes had
to a large extent been hampered by the lack of
reliable relevant data. The inadequate database
had also inhibited the development of appropriate
methodologies for obtaining reasonable estimates
of poverty incidence.
7. The Executive Secretary noted
that with increasing demand for quality data and
information the statistical systems in most countries
had come under growing pressure to generate more
reliable and up-to-date information for accurately
measuring the incidence of poverty and for formulating
and monitoring poverty alleviation programmes.
He emphasized that in order to respond to those
needs the national statistical offices should
focus on, among other things, improving the scope
and content of existing data sources and tapping
potential sources to obtain more direct data in
sufficient detail. He also highlighted the need
for a continuing dialogue between data producers
and data users to strengthen national poverty
databases. In that connection, the Executive Secretary
also referred to the growing involvement of the
international community in the realm of poverty
statistics, notably in regard to poverty concepts
and measures.
C. Election of officers
8. The Seminar elected Mr Faizullah
Khilji (Pakistan) as Chairperson, Mr P.V. Thomas
(India) as Vice-Chairperson, and Ms Carmelita
Ericta (Philippines) as Rapporteur.
D. Adoption of the agenda
9. The Seminar adopted the following
agenda:
Opening of the Seminar.
Election of officers.
Adoption of the agenda.
Review of poverty concepts
and measurement.
Poverty measurement
in the context of policy, plan and programme
formulation at the national and subnational
levels in ESCAP region.
Data requirements for
formulating poverty alleviation programmes
and for monitoring their implementation.
Strengthening statistics
on poverty.
Recommendations.
Adoption of the recommendations.
E. Documentation
10. The documents presented
at the Seminar are listed in the Annex to the
report.
II.
REVIEW OF POVERTY CONCEPTS AND MEASUREMENT
11. The Seminar had before it
the following documents: "Evolution in the Nineties
and Structural Factors Behind Poverty and Income
Distribution in Latin America" (STAT/POV/1) prepared
by ECLAC, a secretariat note entitled "Poverty
Concepts and Measurement in Developing ESCAP Countries"
(STAT/POV/2), and documents submitted by FAO entitled
"Note on FAO's Work Relating to the Measurement
of Poverty" (STAT/POV/3); and "Guidelines on Socio-Economic
Indicators for Monitoring and Evaluating Agrarian
Reform and Rural Development" (STAT/POV/4). Presentations
were also made based on 20 country papers as listed
in the annex.
12. The Seminar noted the remarkable
homogeneity across most countries of ESCAP region
in the conceptualization of poverty as a state
of deprivation with reference to socially accepted
norms of basic human needs. Practically all developing
countries in the region had adopted absolute poverty
as the relevant poverty measure, with a poverty
line based on the basket of goods and services
approach being the most widely used measurement
indicator of poverty.
13. Legal recognition was also
accorded to the definition of the poor. For example,
in the Philippines through a Republic Act known
as the Social Reform and Poverty Alleviation Act,
the poor had been defined as those individuals
and families whose incomes fell below the poverty
threshold as defined by the National Economic
and Development Authority and/or those who could
not afford in sustained manner to provide their
minimum basic needs of food, health, education,
housing and other essential amenities of life.
14. The Seminar noted that although
there was general agreement on conceptual definition,
there was less agreement on the statistical measurement
of poverty. The diversity in measurement methodologies
between as well as within countries had largely
resulted from difficulties in specifying the basket
of minimum needs (food and non-food) and from
lack of relevant data and limitations of available
data and information.
15. While the minimum food needs
had in most cases been determined on the basis
of a recommended minimum nutritionally adequate
diet, the Seminar noted that the per capita food
intake for survival assumed for deriving the food
poverty line varied across countries as well as
within countries from 2100 calories to 2750 calories
per capita per day. There were also problems in
determining the items that constituted the basket
of non-food needs because those needs varied not
only across countries but also between regions
and population groups within the same country.
Whereas there was a scientific basis for determining
an individual's minimum need for food, there was
hardly any method to determine accurately the
norms for non-food items. Consequently, researchers
had perforce to adopt tailor-made but crude techniques
for estimating the non-food component of the poverty
line. The Seminar, therefore, underscored the
importance of developing normative yardsticks
for measuring not only the nutritional requirements
but also other basic needs such as housing, clothing,
education, and transport to obtain reliable estimates
of poverty incidence.
16. The Seminar also highlighted
the need for establishing separate poverty lines
to allow for significant variations in the minimum
needs baskets, the consumption patterns, and in
prices of goods and services between various areas
within a country. In that connection, the Seminar
noted that in large countries, poverty lines were
being derived separately in respect of each geographically/administratively
distinct region and for each socio-economically
divergent population group. In Thailand, for example,
separate poverty lines had been established for
urban and rural areas in four administrative regions
and for Metropolitan Bangkok region. In India,
separate poverty lines had been derived for each
of the constituent States and for urban and rural
areas within each State.
17. The Seminar was of the view
that poverty measures should go beyond the traditional
income/expenditure - based assessments to capture
other dimensions of the poor. The choice of the
right tools was considered to be critical not
only for determining the number of poor people
but also for understanding the causes and correlates
of poverty. The Seminar noted the efforts made
in many countries of the region to expand the
analysis of poverty in a multi-dimensional manner
and additionally to examine the welfare aspects
of the poor including their access to specific
goods, services and functions on the basis of
carefully selected indicators.
18. The Seminar was also briefed
on the activities of the Expert Group on Poverty
Statistics (also known as the Rio Group) established
by the United Nations Statistical Commission.
The attention of the Seminar was particularly
drawn to three basic considerations which had
influenced the activities of the Rio Group:
Poverty being a multi-dimensional
social phenomenon, data and information needed
for its analysis should be developed and used
on the premises that the meaning of indicators
and their value would always be conditioned
to a particular social and political phenomenon.
Hence, attempts to give a world-wide meaning
to poverty indicators were unlikely to succeed;
In the area of poverty,
as in many other areas, statistics served
at least two different objectives: to meet
the demands of the political system for synthetic
indicators, and often for one synthetic indicator;
and to satisfy the information needs of planners
and policy-makers, for comprehensive data
on the different socio-economic groups affected
by poverty, to enable them to determine the
interventions best suited to the problems
specific to each group;
Given the current inchoate
stage in the development of poverty statistics,
it would be more practical to identify and
adopt best practices than to indulge in prolonged
discussions aimed at deriving one or more
standards related to poverty indicators. In
the same vein, it would be perfectly possible
to define different indicators of poverty
and to use them in a complementary fashion
instead of starting arguments about the supremacy
of any one of them. In practice, at least
in Latin America, a good combination of a
well defined indicator and the possibilities
of measuring it systematically and frequently
had proved more important than going in depth
into the definition of indicators that had
little chance of being estimated.
The Seminar was also informed
that the concept of absolute poverty had been
adopted as the relevant poverty measure in developing
Latin American countries. The two most frequently
used methods to measure absolute poverty were:
(i) income or consumption insufficiency method
(poverty income lines), and (ii) unmet basic needs
method. Since most of the calculations of unmet
basic needs were based on data from the decennial
population and housing censuses, poverty lines
could in most cases be constructed only once in
ten years.
20. The Seminar also noted that
governments that had adopted an official measurement
of poverty or were working towards that goal normally
had an objective closely related to policy. In
most of those cases, there was a prior decision
about the range of resources that would be made
available to establish policies in that field.
In such situations, the estimates needed to be
viewed from an empirical approach and not on the
basis of any theoretical background.
III.
POVERTY MEASUREMENT IN THE CONTEXT OF POLICY,
PLAN AND PROGRAMME FORMULATION AT THE NATIONAL
AND SUBNATIONAL LEVELS IN ESCAP REGION
21. For discussions under the
agenda item, the Seminar had before it documents
STAT/POV/5, entitled "Conceptual and Estimation
Issues in the Incidence and Alleviation of Poverty",
prepared by a consultant to the secretariat, and
STAT/POV/INF.6, entitled "Poverty Incidence in
the Asia-Pacific Region: Data Situation and Measurement
Issues" submitted by ADB. The Seminar also considered
relevant sections of the country papers.
22. The Seminar was provided
with a historical perspective to the concerns
about poverty measurement and poverty alleviation
in developing countries. The government in many
of those countries had, over the years, promised
to reduce or even eradicate poverty, malnutrition,
diseases and illiteracy upon gaining political
independence from colonial powers. However, in
most countries, those promises remained unfulfilled,
and in some the poverty situation had deteriorated
after independence. Although the benefits of the
growth-oriented development strategies of the
1960s were expected to trickle down to the vulnerable
and poorer sections of the population, that hope
had seldom been realized.
23. The disillusionment with
"growthmanship" and failure of accompanying
foreign aid policies led to a shifting of focus
towards poverty at the national and international
levels during the subsequent decades. The impact
of the "oil crisis" of the 1970s on the political
economy of international economic cooperation
resulted in a demand for a New International Economic
Order by the developing countries and in an emphasis
on the importance of domestic policy reforms by
developed countries. Those developments culminated
not only in the emergence of poverty as a prominent
agenda item on the international development scene,
but also in efforts to bring poverty measurement
and poverty alleviation into the mainstream of
policy planning and programme formulation in many
of the developing countries.
24. The Seminar also noted that
in recent decades the concept of income poverty
had been broadened to embrace a wider set of basic
needs including those in the social sphere. Thus,
poverty had come to be defined not just as lack
of income but also as lack of access to health,
education, and other services. The concept of
basic needs had spawned a number of policy initiatives
such as integrated rural development, basic health
services, universal literacy, and access to clean
water and housing.
25. The Seminar agreed that
given the broader perception of poverty and the
commitments of national governments to poverty
alleviation, measurements of poverty needed to
be expanded to provide not only reliable estimates
of poverty incidence, but also comprehensive data
on the causes and correlates of poverty. While
information regarding the number of poor persons
was no doubt of political and administrative interest,
the Seminar underscored the importance of providing
planners and policy-makers with up-to-date information
on the basic characteristics of the poor in terms
of their geographic distribution, socio-economic
grouping, household size and demographic composition,
employment, occupation and living conditions of
household members as well as their access to and
control over resources. Besides improving the
understanding of the complex combination of factors
responsible for the incidence of poverty, such
information would enable the formulation and implementation
of the types of interventions best suited to the
various poverty groups.
26. While acknowledging that
the desired information on the basic characteristics
of the poor was not readily available in most
developing countries, the Seminar noted the efforts
made in several of those countries to gather the
required data for assembling profiles of the poor
in terms of their dominant characteristics. For
example, the Seminar was informed that in India
a Below Poverty Line (BPL) census was being regularly
undertaken by various state governments to identify
the poor households and the causes of their impoverishment
so as to design anti-poverty programmes specific
to their needs. In Cambodia, data collected through
the Socio-Economic Survey had been used to compile
poverty profiles along a variety of descriptive
characteristics of the poor.
27. In underscoring the importance
of expanding the analysis of poverty in a multi-dimensional
manner, the Seminar emphasized the need for developing
a minimum set of social indicators to obtain a
clear picture of the various dimensions of poverty.
It noted that social indicators served as potentially
important tools for creating wider public and
political awareness of the poverty syndrome and
for helping national planners and international
donors in redirecting efforts towards enabling
of the poor to meet their own essential needs.
28. The Seminar also discussed
the question of national versus international
requirements of poverty estimates and measures.
It agreed that the first priority in the collection
and processing of data was to meet the demands
of national planners and policy-makers for reliable
information and measures on the incidence, causes
and consequences of poverty at the national and
sub-national levels. However, it was necessary
to develop an objective basis and mechanism to
comply with some unavoidable international needs.
The Seminar emphasized the importance of balancing
the information needs of donors for making inter-country
comparisons and those of the national governments
that had to follow several diverse, often conflicting,
national objectives.
29. The Seminar also noted that
in addition to the baseline information needed
for programme formulation, data collection and
analysis on a continuing basis was essential for
assessing the efficiency of programme implementation
and for evaluating the programmes' impact on the
poverty situation. The data and indicators required
for those purposes would relate to changes in
poverty incidence and in the basic characteristics
of the poor and were usually obtained as a by-product
of programme implementation or through special
surveys. They would help to determine the appropriateness
of the poverty alleviation programmes to the intended
beneficiaries, the problems encountered in the
implementation process, and to ascertain the extent
to which the programmes had contributed to improve
the conditions of the beneficiaries. The Seminar
also acknowledged that those data and indicators
would be needed not only by national authorities
but also by international donors to assess the
impact of their interventions on the poor.
30. The Seminar agreed that
the strategy to address poverty data requirements
should aim at meeting the needs of national authorities
for adequate information for formulation and implementation
of effective poverty alleviation programmes as
well the reasonable demands of donors for comparable
data and indicators for inter-country comparisons.
The Seminar therefore underscored the importance
of a programme of synchronized poverty measurement
activities with national authorities assuming
responsibility for the collection and analysis
of primary data and with donors providing assistance
to build and/or strengthen national capacities
and to support studies for enhancing the understanding
of the poverty syndrome.
31. The Seminar also discussed
the frequency at which data needed to be collected
and analysed and measurements undertaken to monitor
poverty reduction programmes. It agreed that most
national statistical systems would find it expensive
to carry out annual poverty measurements, especially
in times of severe economic crisis when, ironically,
rapid assessments would be most needed. It therefore
suggested that updating of poverty data and indicators
for smaller areas within countries could be undertaken
at longer intervals of five or more years, and
that depending on individual country situations,
monitoring in between quinquennial large-scale
synchronized surveys should be carried out at
least once in respect of larger subnational areas
and for the country as a whole.
IV.
DATA REQUIREMENTS FOR FORMULATING POVERTY ALLEVIATION
PROGRAMMES AND FOR MONITORING THEIR IMPLEMENTATION
32. The Seminar discussed the
topic on the basis of documents STAT/POV/5, STAT/POV/8
entitled "Poverty Data Requirements" prepared
by the secretariat, and relevant sections of the
country papers.
33. The Seminar noted that in
addition to continuing national political commitment
to reduce poverty, the emphasis on human development
placed by various international summits had brought
into focus the need for accurate, consistent and
timely statistics on the incidence of poverty
as well as on the socio-economic characteristics
of the poor. Further, the recent economic crisis
and the resultant heightened priority accorded
to poverty reduction had also drawn attention
to the need for improving the understanding of
the inter-relationships of economic events, good
governance, and the dynamics between the poor
and the non-poor.
34. However, the Seminar noted
that most developing ESCAP countries lacked the
knowledge base essential for adequately addressing
poverty issues and for formulating and monitoring
effective poverty alleviation programmes. That
situation had largely been due to the fact that
measurement of poverty had been overly preoccupied
with the estimation of poverty incidence while
other important dimensions such as understanding
of the deprivation of the poor, the causes and
consequences of poverty, and the monitoring of
poverty alleviation programmes had received less
attention. The Seminar stressed the need for redressing
that imbalance.
35. The Seminar also noted that
the substantial gaps and serious defects in available
poverty-related statistics had led to compromises
in the use of data and rendered the task of poverty
analysis more arduous. It observed that poverty
studies undertaken in respect of most developing
countries lacked the depth that a historical wealth
of information could provide. In order to bridge
the increasing gulf between the development of
conceptual analysis and statistical implementation
of poverty-related issues, the Seminar underscored
the urgent need for a poverty statistics information
policy or strategy to address the inadequacies
and limitations of statistical data on poverty.
36. The Seminar observed that
the first logical step in designing poverty alleviation
strategies was the undertaking of a situation
analysis to determine the extent of poverty, its
possible causes, and the characteristics of the
poverty groups. That analysis, based on available
data and studies, would help in understanding
the problems and needs specific to various poverty
groups and to formulate the interventions best
suited to each of those groups. The Seminar agreed
that the minimum statistical data required for
the situation analysis should cover such areas
as demographic composition, socio-economic characteristics
of the population (e.g., education, health and
nutrition, housing, employment, occupation, income)
and access to and control of resources including
land, credit facilities etc., with disaggregation
at subnational levels.
37. The Seminar also agreed
that the planning and formulation of effective
anti-poverty programmes required the use of a
range of statistical indicators to ascertain who
the poor were, where they lived, the social groups
for which special alleviation programmes would
be implemented, and households and individuals
that needed to be assisted. Such indicators would
help to identify the prospective target groups
for programme intervention, to understand the
conditions of poverty that needed to be changed,
and to determine the most appropriate interventions.
Those indicators, which should be measurable,
factual, valid, verifiable and sensitive, could
be based on micro-level statistics from household
surveys and special case studies. However, the
Seminar noted that in most countries of the region,
very little effort had been made to measure poverty
at the micro-level to identify the target beneficiaries.
V.
STRENGTHENING STATISTICS ON POVERTY
38. The Seminar had before it
secretariat documents STAT/POV/9, entitled "Sources
and Limitations of Poverty Statistics in Developing
ESCAP countries" and STAT/POV/10, entitled "Improvement
of Household Incomes and Expenditure Surveys for
Collecting Data for Poverty Measurement", and
background information presented to the Seminar
through country reports.
39. The Seminar noted the great
diversity among developing countries of the region
in regard to the availability of poverty-related
statistics. In several of those countries, the
available data were considered to be inadequate
for accurately estimating the incidence of poverty
and for comprehensively analysing the socio-economic
characteristics of the poor. Further, the practice
in poverty analysis of utilizing data obtained
for other purposes had restricted the choice of
statistical indicators and given rise to misleading
interpretations.
40. The Seminar also noted that
data on income, expenditure and other aspects
gathered through household sample surveys had
mostly been used to derive poverty lines and to
address public concerns about poverty in developing
countries of the region. However, the Seminar
observed that as a main source of statistical
information for poverty analysis, those surveys
were of questionable reliability for various reasons:
The surveys might not
be frequent enough, being carried out at
intervals of five years in some countries
and three years in several others;
Generally the samples
were not large enough to obtain reasonably
valid estimates of poverty incidence in
respect of small regions and socio-economic
groups;
Since those surveys were
not specifically designed to collect poverty
data, information available for poverty
analysis was usually limited, thus necessitating
the combination of information from one
survey with that from another which might
differ in several respects including sample
coverage and timing;
The quality and reliability
of the data gathered were, in many instances,
considered to be doubtful due to non-sampling
errors such as timing of the survey, duration
of field investigations, and mis-reporting
of income and consumption by respondents.
It was generally acknowledged that there
were serious errors in survey data on incomes
arising from tendencies to deliberately
under-report incomes, and non-reporting
of income in kind and transfer incomes.
Often surveys did not contain sufficient
data on expenditure on individual food items,
quantities of different items consumed and
food consumed outside the household.
41. While the Seminar noted
that the quality of household surveys had improved
markedly in recent years, it nevertheless felt
that there were still serious problems of standardization
which needed to be resolved in order to render
the comparability of the data collected over time
and across countries reliable with a high degree
of confidence. The Seminar also noted that in
most countries the results of even officially
conducted surveys were seldom used for building
up macro-economic aggregates, a large part of
which continued to rely on dubious interpolation
or time trends for intercensal or inter-survey
years. Consequently, macro-economic analysis in
many countries was often confronted with reconciling
trends on poverty based on household survey (cross-section)
data and growth based on national income (time
series) data generated by different agencies.
The Seminar underscored the need for creating
institutional mechanisms for reconciling such
conflicting evidence through transparent research
and discussions to make optimum use of available
data and information.
42. The Seminar also noted that
population and housing censuses constituted another
important source of data for constructing a number
of poverty-related indicators such as family size,
literacy, educational participation and attainment
levels, employment and occupation etc. In some
countries, census data on income and consumption
were utilized to derive poverty lines. By providing
detailed data and information for smaller areas
and population groups, censuses also promoted
development of analytical studies of greater geographical
disaggregation.
43. However, the Seminar observed
that since censuses were conducted at ten year
intervals in most developing countries, census
data were not useful for frequent assessment and
monitoring of the socio-economic characteristics
of the poor.
44. The Seminar also noted that,
although a considerable amount of statistical
information continued to be gathered routinely
as part of the administrative process, very few
countries in the region used administrative records
for analysing the incidence of poverty and the
characteristics of the poor.
45. The Seminar recognized that
with increasing commitments of governments to
poverty alleviation, there was a growing demand
for more relevant and reliable data to derive
meaningful estimates of poverty as well as to
design appropriate poverty alleviation strategies
and to evaluate the impact of those strategies.
Consequently, the national statistical systems
were under heavy pressure to improve the poverty
databases in terms of their scope, coverage, reliability,
and, more importantly, relevance for policy-making.
46. The Seminar was of the view
that national poverty databases could be adequately
strengthened by adopting two broad, but complementary,
measures. The first measure involved improving
the scope and content of existing data sources
such as surveys and censuses to obtain more direct
data in sufficient detail needed for development
of statistics and indicators relevant to a comprehensive
understanding of the poverty situation at the
national and sub-national levels. Those efforts
included expanding the content of the questionnaires
to elicit detailed data on the composition of
consumer expenditure in terms of those goods and
services that constituted the basis for deriving
poverty lines, and on the household characteristics
considered to be correlates of the poor households.
47. Since sample household surveys
were the most frequently used sources of income
and expenditure data used for deriving the poverty
line, the Seminar emphasized that improving the
quality and reliability of income and consumption
data should be a major goal for those in charge
of such surveys. In that connection, the Seminar
recognized that incorporating social transfers
in kind would improve household surveys as an
instrument to measure poverty and monitor poverty
alleviation interventions, and would harmonize
the concepts and practice at the micro and macro
level statistics. However, the Seminar was conscious
of the enormous problems in imputing the value
of the services, particularly in the health sector
where there was a wide variation in the quantity
and type of services availed of by the households.
48. The Seminar noted that although
a considerable amount of statistical data and
information relevant to poverty analysis was being
collected routinely by various agencies in many
countries, efforts were not being made to exploit
those data in a systematic manner to obtain quantitative
information on a wide range of social, demographic
and economic aspects relating to the poor. The
Seminar underscored the need to process those
data on a continuing basis to provide much needed
additional information for estimating poverty
incidence and for the formulation and monitoring
of poverty alleviation programmes.
49. The Seminar recognized that
the types of data employed in poverty measurements
were not only closely related to the statistical
capabilities of each country but also depended
on the needs voiced by the main users of such
data. It therefore underscored the importance
of a continuous dialogue between data producers
and data users to identify data-cum-conceptual
gaps.
VI.
CONCLUSIONS AND RECOMMENDATIONS
50. The Seminar on Poverty Statistics
reviewed the current situation relating to the
incidence and alleviation of poverty in the ESCAP
region in the context of the prevailing global
economic situation and domestic policy environment.
The main concern of the Seminar centered on the
availability and development of statistical data
and methodology for the understanding and measurement
of poverty. In making that review the Seminar
was guided by the opening statement of the Executive
Secretary of ESCAP and the rich and varied documentation
presented by the country participants and the
ESCAP secretariat and its consultant, along with
that contributed by other international agencies
and organizations, especially the Asian Development
Bank, ECLAC, FAO, and ILO.
51. The Seminar noted with satisfaction
that the overall poverty situation as indicated
by the number of poor as a proportion of the total
population in the developing ESCAP region had
shown steady, if slow, improvement over the past
few decades. However, the total number of poor
people remained disquietingly high in view of
the high rates of population growth still prevailing
in many of the developing countries of the ESCAP
region. It was also noted that much of the improvement
in the poverty situation of the region in the
last two decades in quantitative terms was attributable
to significantly lower poverty incidence in the
most populous country of the region, China. The
Seminar also noted with concern the reverses in
the poverty situation in the East Asian economies,
which had previously succeeded in reducing the
high rates of poverty incidence during the last
two decades through their successful export-led
development strategy, as a result of the East
Asian financial crisis, which coincided in some
of those countries with the El Nino phenomenon.
Although precise statistical evidence was lacking,
the poverty situation in the transition economies
of the Central Asian republics was presumed to
have worsened severely in the 1990s in view of
the continued declines in per capita incomes and
the withdrawal of access to a variety of goods
and services previously available at subsidized
rates.
52. The country papers presented
at the Seminar gave a vivid account of the efforts
being made by the governments and statistical
authorities to quantify the poverty situation
and to provide assistance in the implementation
of the poverty alleviation programmes being launched
by those countries. The country papers reflected
the diversity of the situation both in terms of
the objective reality and the extent to which
it had been possible to capture it through various
statistical measures . Some countries had been
monitoring the poverty situation in a systematic
manner, but many others, especially the least
developed and transition economies, had launched
those efforts only recently. While most countries
of the region had received technical and financial
assistance from donor agencies, especially the
World Bank and UNDP, the overall financial situation
in the face of the need for structural adjustment
and reduction in government budgets had in many
instances seriously constrained such efforts from
becoming institutionalized on a sustainable basis.
53. The Seminar also devoted
considerable attention to the objectives of poverty
studies and the complexity of the concepts which
had emerged in recent research. While the initial
interest in poverty research was focused on the
proportion of people falling below the poverty
line, more recent work had stressed the need for
treating poverty as a complex issue which could
not be captured by a single statistic, notwithstanding
the usefulness of such a summary measure for purposes
of arousing public concern and cultivating donor
support for alleviating poverty. In that context
the Seminar also considered the competing demands
on poverty research made, on the one hand, by
national planners and policy makers for the purpose
of formulating, implementing and innovating strategies
for the country's long-term development and, on
the other, by the pressures for inter-country
comparability of poverty statistics for the purpose
of mobilizing donor support. The Seminar felt
that since the primary responsibility for poverty
alleviation was that of the national governments,
the former consideration should receive precedence,
given the restricted availability of resources.
Nevertheless, the Seminar fully appreciated and
endorsed the need for continued efforts for meaningful
comparisons of the poverty situation among different
countries, especially on a regional basis. In
that context, the Seminar noted with great appreciation
the studies undertaken by ECLAC for comparative
inter-country research on the poverty situation
in that region. The Seminar expressed the need
to launch similar research studies in the ESCAP
region, subject to the availability of funds and
in cooperation with other agencies and institutions,
especially the ADB and the World Bank.
54. As a result of the discussions
held during its deliberations, the main features
of which have been highlighted above, the following
principal operational recommendations emerged
from the Seminar.
The Seminar endorsed the
need for continuing and accelerated in-depth
research on the poverty situation at the national,
regional and international levels. It was
recognized that poverty alleviation was increasingly
regarded as a principal objective of development
policy. In that regard, the Seminar stressed
the need for greater clarity and objectivity
in specifying the goals of poverty alleviation
and its time frame in order to lend credibility
to the objective and dispel cynicism about
the possibility of its achievement among the
public at large, especially the poor.
In view of the complex and
intransigent nature of the poverty problem,
the Seminar endorsed the need for a differentiated
approach to its study, measurement and formulation
of strategies in the context of a country's
specific situation, development stage and
medium-term objective, national capability
and available resources. The Seminar stressed
the need to consider the goal of poverty alleviation
as part of a strategy of equitable growth,
rather than in either a static or inequitable
growth setting. The Seminar considered that
the essential objective of poverty research
should be to provide a deeper understanding
of the extent and nature of deprivation and
the causes and consequences of poverty in
the countries of the region. In that regard
national statistical offices should accommodate
in their collection and dissemination the
need to develop social indicators to complement
income-based monitoring.
The Seminar recommended
that while maximum efforts should be made
to choose indicators of poverty and its correlates
on a broadly comparable basis, the countries
should make the choice that best suits their
national needs at the given time. While most
medium-sized countries with broadly homogeneous
population and regional characteristics might
choose a single, national poverty line, it
was felt that more disaggregated poverty lines
would be required for most countries with
heterogeneous population groups and ecological
differences. There was a need to focus attention
on the problems of "hard core" or "acute poverty"
situations. Studies also needed to be conducted
on the seasonal and cyclical aspects of the
poverty problem.
There was a need for a participatory approach
in the choice of methodology and data; the
methodology chosen should be fully documented
and disseminated. That approach would also
help in reducing the confusion that often
accompanied the multiplicity of poverty
estimates. Expert groups, NGOs and other
relevant groups should openly debate and
discuss the appropriateness of each methodology.
There was a need for more
concerted, well thought out and better designed
methods for inter-country comparison of poverty
estimates. The existing methods such as those
using the World Bank-sponsored $1 a day poverty
line needed to be viewed with scepticism and
should be employed with caution. Nevertheless,
the search for more appropriate methodologies
or international comparisons should continue
with greater transparency, debate and additional
resources.
The exchange of experience
and expertise at the regional level for both
estimation and alleviation of poverty should
be promoted. ESCAP might consider mobilizing
donor support to launch such a regional project
and convene a regional expert group meeting
for that purpose. Such a meeting could also
discuss the possibility of adopting more comparable
definitions, procedures and methodologies
for poverty measurement.
The Seminar emphasized that
the burden of poverty measurement could not
be carried by statisticians alone and required
the active collaboration not only of economists,
development planners, and policy makers, but
also of those engaged in related disciplines
such as anthropologists and sociologists.
In addition, the NGOs active in the field
of poverty alleviation, many of which were
stakeholders and produced poverty related
statistics, could provide useful insights
into the underlying nature and dimensions
of the poverty problem, and play a catalytic
role in the design of statistical indicators.
The Seminar recommended that statisticians
and researchers consult those diverse groups
actively and involve them as appropriate in
the various stages of poverty measurement
and of the monitoring and evaluation of poverty
alleviation programmes.
If poverty alleviation as
a major objective of development policy of
the government was to be realized, considerably
larger resources would have to be devoted
to the task of building and strengthening
adequate national capacity for the development
of statistical methodologies and databases
for the measurement of poverty and the monitoring
of the performance of poverty alleviation
programmes. The resources available at present
for undertaking those tasks had largely been
provided by donor agencies on an ad hoc basis,
sometimes with little regard for the sustainability
of those efforts. The Seminar recommended
that donor assistance for poverty research
should contribute towards the building of
such a sustained capability and the institutionalization
of the needed research and data collection.
The Seminar recognized that it included the
use of information technology in data processing
and dissemination, as well as specialized
training on monitoring and evaluation.
The tasks of poverty measurement
and monitoring had grown in complexity and
scope in recent years due to the increasing
integration of developing economies with the
global economy and their increased vulnerability
to external shocks. That required that the
statistical systems of the countries should
stand ready to undertake ad hoc studies, with
the help of truncated surveys, panel data,
modelling and other methods to provide policy
makers with the required information. Similar
efforts might be required in the event of
natural disasters such as floods, earthquakes,
crop failure and other serious disruption
of economic activities.
The Seminar recommended
that countries might consider undertaking
more frequent household income and expenditure
surveys where required, preferably with larger
samples and more intensive coverage of the
vulnerable groups of the population. In that
regard, the Seminar felt that the activities
of the informal sector needed to be covered
as adequately as possible by suitable changes
in the design of the present surveys or by
undertaking separate surveys for the informal
sector. It was also recommended that suitable
modifications be made in other data collection
instruments such as censuses and administrative
reporting systems. The Seminar also emphasized
the need to exploit and utilize data already
available in relevant administrative records.
Annex
LIST OF DOCUMENTS
Symbol
Title of document
Relevant agenda item(s)
STAT/POV/L.1
Provisional Agenda
3
STAT/POV/L.2
Annotated Provisional
Agenda
3
Seminar documents
STAT/POV/1
Evolution in the
Nineties and Structural Factors Behind Poverty
and Income Distribution in Latin America
4
STAT/POV/2
Poverty Concepts
and Measurement in Developing ESCAP Countries
4
STAT/POV/3
Note on FAO's Work
Relating to the Measurement of Poverty
4
STAT/POV/4
Guidelines on Socio-Economic
Indicators for Monitoring and Evaluating Agrarian
Reform and Rural Development
4
STAT/POV/5
Conceptual and Estimation
Issues in the Incidence and Alleviation of Poverty
5, 6
STAT/POV/8
Poverty Data Requirements
6
STAT/POV/9
Sources and Limitations
of Poverty Statistics in Developing ESCAP Countries
7
STAT/POV/10
Improvement of Household
Income and Expenditure Survey for Collecting Data
on Poverty Measurement
7
STAT/POV/CRP.1
Recommendations
8, 9
Country papers
STAT/POV/7[Bangladesh]
Country paper
4,5,6,7
STAT/POV/7[Cambodia]
Poverty Measurement
in Cambodia
STAT/POV/7[China]
Poverty Statistics
in China
STAT/POV/7[Fiji]
Poverty Measurement
Discuss
STAT/POV/7[India]
India Country Paper
on Poverty Measurement
STAT/POV/7[Indonesia]
Poverty Measurement,
the Case of Indonesia
STAT/POV/7[Kazakhstan]
Measurement of Poverty
in Kazakhstan
STAT/POV/7 [Kazakhstan]Add.1
Country Paper, Kazakhstan
STAT/POV/7[Lao PDR]
Poverty Measurement
in Lao People's Democratic Republic
STAT/POV/7[Malaysia]
Malaysia: Poverty
STAT/POV/7[Mongolia]
Mongolia: Poverty
STAT/POV/7[Myanmar]
Country Paper, Myanmar
STAT/POV/7[Nepal]
Country Paper: Nepal
STAT/POV/7[Pakistan]
Pakistan Country
Paper
STAT/POV/7[Philippines]
Country Paper, Philippines
STAT/POV/7[Singapore]
Country Paper, Singapore
STAT/POV/7[Sri Lanka]
Measurement of Poverty
in Sri Lanka
STAT/POV/7[Thailand]
Poverty Measurement
in Thailand
STAT/POV/7[Viet
Nam]A
Concepts, Contents
and Measurement of Poverty in Viet Nam
STAT/POV/7[Viet Nam]B
Concepts, Contents
and Measurement of Poverty in Viet Nam
Background
documents
STAT/POV/INF.1
Poverty Measurement=s
Present Status of Concepts and Methods
4
STAT/POV/INF.2
Summary of the Debates
4
STAT/POV/INF.3
Role of SIAP in Providing
Training on Poverty Statistics
4, 5, 6, 7
STAT/POV/INF.4
The Statistical Measurement
of Poverty
4
STAT/POV/INF.5
Poverty Statistics
- Some Useful References
7
STAT/POV/INF.6
Poverty Incidence
in the Asia - Pacific Region: data situation and
measurement issues
