|
Executive Summary This
report has two parts.
The first part presents a brief summary of the new International Comparison
Programme (ICP) strategic framework that has been developed by the World
Bank, in close collaboration with international and regional agencies.
The second part gives an account of the progress of the preparatory work
that has been underway to launch a global ICP with 2003 as a base year.
This includes progress made thus far on the fund-raising front, research
papers commissioned to address outstanding substantive issues related
to sampling design and regional linkage procedures, and the coordination
of regional and global proposals and work plans.
I.
The International Comparison Programme
1. The ICP is a global programme established to produce purchasing power
parity estimates, with the objective of facilitating cross-country comparisons
of GDP and its components. By establishing purchasing power equivalence,
where one dollar purchases the same quantity of goods and services in
all countries, PPP conversions allow cross-country comparisons of economic
aggregates on the basis of volumes, free of price and exchange rate distortions.
Since the ICP has been launched more than three decades ago, it has grown
from a pilot initiative including 10 countries to a global statistical
exercise covering 118 countries from all regions of the world. The
ICP is a truly cooperative international statistical undertaking involving
global, regional, and national agencies. It generates its results on the
basis of hundreds of item price comparisons for many detailed headings
of expenditure on GDP.
2. Demand
for PPP data has grown steadily and rapidly. The increasing demand has
taken on a higher profile since the early 1990s due to increased volatility
of exchange rates, and as a result of the more explicit use of PPP statistics
for global goal-setting and monitoring of progress towards achieving the
Millennium Development Goals (MDGs). Rapid globalization and the increasing
integration of international markets and financial institutions have also
imposed a new urgency to the need for internationally comparable data.
PPP data are widely used in comparative research on economic and social
development. The data are used, among other things, to establish international
poverty threshold and monitor progress towards achieving MDG's poverty
reduction target (World Bank); to construct Human Development Index (UNDP);
to compare health expenditure per capita (WHO); to assess per capita expenditures
in education (UNESCO); to monitor the welfare of children (UNICF);. and
to compare the relative sizes of economies and to estimate weighted averages
of regional growth rates (IMF).
3. The growing demand has brought with it closer
scrutiny of the data, particularly as related
to data quality and timeliness. Two recent
reviews of the ICP conducted by Jacob Ryten
and Ian Castles[1]
concluded that the ICP suffers from poor data
quality, limited country coverage, and lack
of timeliness. The results from the 1993/96
round bear witness to the deteriorating quality
of PPP data. Though the 1993/96 round
brought 118 countries to the ICP fold, the programme
was grossly under-funded, and this led to the
documented problem of poor data quality.
The results were counter-intuitive, especially
at the detailed level, so much so that they
have not been released officially.
[1]
Ryten, Jacob (1999), Report on the evaluation
of the International Comparison Programme,
E/CN.3/ 99/8, United Nations. Castles (1997),
Rieview of the OECD-Eurostat PPP Program, STD/PPP
(95),5, OECD, Paris
4. The Ryten report, which was commissioned by the United Nations
Statistical Commission (UNSC) and sponsored by the World Bank and IMF to
review the global exercise with particular emphasis on non-OECD countries,
highlighted that the ICP suffers from the so called public goods problem.
It stressed that without concerted remedial action by the stakeholders,
particularly international and regional organizations, the programme would
languish and face a slow death in developing regions where a reliable information
base for measuring progress towards achieving the MDGs is needed most urgently.
5. The UNSC, at its thirty-first session held in March 2000, discussed
the Ryten report and adopted its recommendations. Subsequently, it requested
the international community to develop a comprehensive reform plan and
a fund-raising strategy, with the ultimate objective of improving the
efficiency and methodology of future ICP surveys and ensuring the robustness
and reliability of their outputs, PPP data.
A. New
Strategic Framework
6. The World Bank, in close collaboration
with a number of international, regional and
national agencies, has prepared a comprehensive
strategic framework. The new framework incorporates
the widely accepted recommendations of the Ryten
report, and aligns the program with the MDGs.
The overall strategy is based on the premise
that there is an immediate need to (a) rebuild
the credibility of the ICP, and (b) promote
the wider acceptance and use of PPP data.
The approach is to carry out a well planned
and managed new round of ICP in the 2002-2005
timeframe, paying particular attention to data
quality, through improvement of price and national
accounts statistics in the participating countries.
Accordingly, it aims at:
- establishing a well coordinated
and effective national, regional, and global
management structure;
- forming a cost-sharing
arrangement among international, regional
and national institutions to establish a sustainable
funding base;
- developing a comprehensive
survey framework for data collection and validation,
with particular emphasis on harmonization
of methodologies, concepts and definitions
for cross-country consistency and comparability;
- strengthening national
capacity and maximizing the synergy between
the ICP and national statistical programs;
- integrating poverty-specific
PPPs into the mainstream of ICP work.
- supporting research in
PPP aggregation methods and regional linking
procedures; and
- promoting a wider use
of the ICP data for policy formulation and
monitoring of progress both at the national
and international levels.
7. The plans set out in the new strategy are
ambitious and their success depends in part
on the commitment of the international community
to rise up to the challenge of overcoming the
public goods problem, and in part on the willingness
of national statistical agencies to be active
partners. Because of the continuing demand
to compare levels of economic output as well
as the need to monitor global economic growth
as a key factor in poverty reduction, the strategy
provides for estimation of PPPs for GDP both
at regional and global levels. It recognizes,
however, that in some countries where
statistical systems have limited capacity, it
may be necessary to concentrate resources on
getting good quality price data for the consumption
items only.
II.A.
Global and Regional Management and Accountability
8. If the programme is to be successful, it requires wide participation
and ownership by all stakeholders. The new framework provides for effective
management, and maximum cooperation at the global, regional and national
levels. Overall coordination and accountability would be established through
an international governing body, consisting of representatives of the
UNSC, regional and international agencies, and major financial sponsors.
The responsibility of the international governing body would include setting
up objectives, drawing future directions, and approving the work plans
of regional and global coordinators. This is an important innovation not
only because it would establish accountability, but also because it would
accord regional stakeholders an opportunity to influence both the overall
strategy and the day-to-day management of the global program. Technical
advice and overall quality control would be established through a technical
advisory group, consisting of recognized experts in the areas of ICP,
CPI and national accounts. The advisory group would be responsible for
setting up a research agenda, reviewing the impact of new research on
the program as a whole, and setting quality standards and monitoring
quality.
9. Day-to-day
management of the global program would be achieved through an international
secretariat. The need for an adequately resourced global coordinator (secretariat)
was an important recommendation of the Ryten report. The secretariat
would be responsible for (a) preparing work plans to be approved by the
international governing body, (b) mobilizing resources, (c) implementing
global programmes and monitoring progress, (d) linking regional results
and generating global PPPs, (e) analyzing and disseminating benchmark results,
and (f) reporting to the stakeholders.
10. In
keeping with previous rounds of the ICP and even with enhanced global coordination
capacity, implementation of the programme will depend on the involvement
and support of regional organizations. Overall regional coordination and
accountability would be the responsibility of a regional governing body
to include representatives of countries, regional institutions, and international
organizations. The responsibility of the regional governing body would
include setting up regional objectives, goals and priorities, and approving
regional work plans.
11. National statistical agencies would be responsible for country level
data collection and management of the programme. Each participating
country would be required to appoint a coordinator who will take responsibility
for organizing the data collection process and liaising with the regional
coordinator. National statistical agencies are encouraged to participate
in the program at all levels, including (a) in the preparation of regional
item specifications (through participation in regional workshops, (b) in
the development of ICP survey framework (through discussion with the authors
of the papers under preparation), and (c) in the coordination and management
of the regional program (through representation in the regional governing
body).
12. The reform plan and the underlying strategy have been discussed widely
and have been endorsed by international organizations, regional agencies,
and prominent scholars and experts. The UNSC, at its thirty-second session
held on 6-9 March 2001, discussed the proposed strategy and affirmed its
support and welcomed the efforts of the World Bank in developing the new
strategic framework. The Commission requested the international agencies
to raise the visibility of the ICP and impressed upon them the need to
exercise collective leadership to mobilize adequate resources for the
next round of ICP surveys.
II.B.
ICP 2003 Work Plan and Progress
13. This section presents the work plans for the proposed 2003 ICP round,
and highlights the progress made thus far. A great deal has been achieved
since the UNSC meeting in March 2001. On the fund-raising front the results
have been encouraging. On the research front, a number of papers have
been commissioned and work is progressing well. Regional proposals and
work plans, highlighting regional priorities within the framework of the
global strategy, have been prepared.
III.
A. Fund-raising
14. The plan has been to raise US$14 million to finance
the proposed 2003 ICP round in five regions, including Asia and the Pacific,
Africa, Latin America and the Caribbean, the Middle East, and the Commonwealth
of Independent States. The fund-raising strategy envisages that about
one third of the total budget would be provided by the main international
sponsors (World Bank, IMF, WHO and UNDP). About 20% of the total, needed
to cover the cost of regional coordination, would be provided by the regional
agencies. The remainder, which is estimated at US$7 million needed to
cover the cost of county level data collection, training, and hardware,
would be raised primarily from the donor community. As of the middle of
November 2001, nearly 45% of the $14 million target has been pledged from
international organizations, regional agencies and bilateral partners.
15. From
the ESCAP region both the ESCAP secretariat and ADB have officially committed
themselves to help cover part of the regional cost. ESCAP has committed
in-kind contribution in the amount of US$170,000 and plans to seek additional
resources from other sources. ADB has indicated its intentions to support
the Asian and the Pacific regional component of the ICP initiative and
is ready to discuss the details of its contributions, financial and otherwise.
Equally encouraging is that the Australian Bureau of Statistics has shown
keen interest in playing a key role in the ESCAP regional comparison. It
has also expressed willingness to provide financial assistance.
III.B.1. Research
16. Following the recommendations of
Jacob Ryten and other leading experts, the ICP
proposal identifies two broad areas of research:
(a) sampling design, compilation of expenditure
weights, and regional linking procedures, and
(b) PPP aggregation methods. The
questions of comparability of quality across
regions and chaining of regional results are
widely considered as extremely important areas
to the future of ICP. A number of papers have
been commissioned to conduct thorough and wide
ranging reviews of issues related to sampling
design, expenditure weights, and regional linking
as part of the planning program. It is
hoped that the findings of these studies would
help plan and implement the next round of ICP
surveys. Research related to methods of aggregation
would be commissioned at a later stage.
17. Mr
Jacob Ryten has been engaged in research to develop a comprehensive survey
framework addressing both conceptual and practical issues related to sampling
design. His report will (a) provide explicit guidelines with respect to
coverage, selection of regions, and the number and types of outlets,
timing and frequency of collection, issues related to treatment of quality
differences, and procedures for calculating average prices, (b) discuss
the tradeoff between the comparability and representativity of items selected
for pricing and develop objective criteria to establish a balance between
exercising expert judgment and applying a standard survey framework;
and (c) investigate alternative approaches and recommend how best to collect
prices for comparison resistant services, including government expenditure.
Mr. Ryten has visited and collected information from the ESCAP secretariat.
He is scheduled to visit ECLAC, OECD and CIS to assess the statistical
capabilities of participating countries.
18. Two
studies are being conducted by World Bank staff to address issues related
to expenditure breakdowns, with the objective of improving the compilation
of expenditure weights at the basic heading level. Building on the
procedures outlined in the current Handbook of ICP, they will (a)
develop expenditure classifications to be used in the various regional
comparisons, taking into account the minimum requirements of the global
comparison (b) identify secondary sources of data for compiling and/or
validating expenditure estimates at the basic heading level, and (c) investigate
major departures from SNA and provide specific and practical correction
procedures to establish harmonized statistical concepts according to international
norms and standards.
19. A contract is under preparation
to hire the services of Mr. György Szilágyi
of the Hungarian Statistical Office. Mr. Szilágyi
will focus on comparison resistant services,
with special reference to education and government
expenditures. A proposal focusing on capital
goods has been prepared and submitted for funding
to the World Bank Research Support Committee.
The proposed research, to be conducted by Professor
Bart van Ark, would examine the possibility
of using the ICOP (International Comparisons
of Output and Productivity project) procedures
as the basis for a new approach to estimation
of PPPs for capital goods.[2]Another
proposal focusing on construction projects is
under preparation. It is hoped that all
the noted studies would be finalized before
May 2001 and would serve as background papers
for a comprehensive ICP manual. The manual will
provide a set of principles and operational
procedures that can guide data collection and
processing, document best practices, and serve
as a comprehensive reference material and a
step-by-step compilation guidebook.
[2]
The ICOP is an industry-based approach,
measuring relative prices, output and productivity
by industry origin, whereas ICP concentrates on
measurement of PPPs by expenditure category. During
the past two decades ICOP has been centered at
the University of Groningen.
20. Since the regionalization of ICP in 1980, the operational
unit for comparisons has been groups of countries, for example OECD, Africa,
and Asia and the Pacific. It has been necessary to link these regional
comparisons and generate a globally consistent set of PPP estimates to
carry out world comparison. This requires holding the quality of items
reasonably constant across regions, while maintaining the characteristicity
of regional comparisons. Two papers are in progress to shed light
on this issue. A paper by Professors Heston and Aten will consider two
alternative approaches. The first approach is a modified version of the
chain linking (spanning tree) method. This approach builds on binary
comparisons of countries that have similar expenditure patterns and comparable
price structures. The second approach is based on the so-called generalized
CPD method (an approach that has been advocated by Professor Prasada Rao.)
A second paper by professor Erwin Diewert will provide a comprehensive
review of linkage methods used in the past listing their properties, advantages
and disadvantages. The paper will explore the possibility of developing
a robust and optimal chain linking method. Both papers will demonstrate
and test their approaches using either existing data from the ICP database
or simulation techniques.
III.B.2.
Pilot Studies
21. ESCAP has been involved in the research effort by launching
two pilot studies in a total of eight countries.
The first pilot project aims to investigate
the feasibility of integrating ICP into the
larger context of routine national price data
collection work (CPI). Four countries,
China; Fiji; Malaysia; and Hong Kong, China
are involved as pilots. The principal investigator
of the study is Mr. Sultan Ahmad. Building
on two recent studies independently conducted
at the World Bank and IMF[3]by
Professor Rao and Zieschang et. Al., Mr. Ahmad
begins his study with the assessment of consumer
price statistics in terms of geographic and
item coverage through visits and discussions
with national statistical offices and other
relevant agencies. In light of the level of
statistical capacities and the price collection
methods observed in the selected countries,
his report will propose steps to strengthen
national capacity, specify what can be achieved
in the short term to strengthen deficient areas
directly connected with the data required for
the ICP, and describe what needs to be done
concurrently in the long term to maximize the
link between ICP and regular national statistical
programs.
[3]
Rao, Prasada (2001) Integration of CPI
and PPP: Methodological issues, feasibility and
recommendations, paper prepared at the request
of World Bank (DECDG). Zieschang, K, P.
Armknecht and D. Smith, Integrating Inter-area
and International Price Comparisons with Consumer
Price Index Compilation Paper prepared for OECD-World
Bank Conference on PPP.
22. The second project, which includes India, Indonesia,
Philippines and Thailand, has two objectives and is conducted by professor
Prasada Rao. The primary objective is to develop an appropriate survey
framework for measuring differences in purchasing power of income across
regions and urban and rural areas. The project aims, as its long-term
objective, to regularize poverty data collection by integrating region-specific
price data collection into the larger context of SNA. A secondary objective
is to begin the integration of the ICP techniques with work on cross-country
poverty comparisons, with a view to develop an appropriate method for
constructing "PPPs for the poor". To this end, the proposal aims to lay
a foundation for bringing poverty-specific PPPs into the mainstream of
ICP in the long run. A plan is underway to extend the two
pilot studies to Africa.
III.B.3.
Price Collection Software
23. Within the ICP, the most difficult tasks are acknowledged
to be data collection, editing, and national average prices calculation.
Given the criticisms of the Ryten Report regarding the lack of the coherence
and consistency of the ICP price data, the new strategy puts emphasis
on these issues. Price collection software has been under
development at the World Bank. The primary objective of the software is
to improve national and international data collection and processing capacity.
24. The
software aim is to strengthen key areas of data development, including
supporting national price collection efforts; standardizing price collection
and gap filling methods; permitting data validation; ensuring data collection
consistency over time and space; improving data quality and timeliness;
facilitating data archiving, dissemination, and analysis. The software
will be available in six languages: Arabic, Chinese, English, French, Russian
and Spanish. The English version of the software should be completed by
mid February for testing and presentation at the upcoming ICP conference
in March.
- The software will allow
data collection, including regional stratification,
and geographical weighting pattern (including
urban and rural).
- It will facilitate data
validation, and checking for outliers, editing
and gap filling.
- It will have capability
to store and display pictures of items, particularly
capital good items, to help ensure comparing
likes with likes.
- It will allow all the
data processing up to the calculation of national
price indexes such as CPI and provide an opportunity
to undertake analysis of data to fit national
requirements.
- It will facilitate the
calculation of national average prices for
ICP purposes. This will significantly reduce
the data processing time at the regional coordination
center by facilitating primary data processing
work to be done at the country level. It will
also shorten the duration of the regional
comparison, and help alleviate the lack of
timeliness problem of ICP.
- Data collected for CPI
and which are also suitable for ICP can be
imported directly with minimum alterations
in the format of the original data.
This would facilitate the integration of CPI
and ICP.
- In
a large number of countries there is generally
a high mobility of staff and therefore it
is difficult to ensure continuity of methods;
especially for ICP work, which is mostly not
integrated with national statistical programs.
It is hoped that with the PC based software
institutional memory can be maintained.
IV.
Coordination of regional and global strategies,
work plans and management
25. It is widely acknowledged that effective management and coordination
of the ICP is crucial. The new strategic framework provides for effective
management and maximum cooperation at the global, regional and national
levels. It balances the need to have a consistent global program with
a strong coordinating center, and regional comparisons that reflect and
maintain the characteristicity of the regions in question.
26.
In
order to achieve optimal utilization of resources and efficient coordination
of regional and global work plans, the sphere of responsibilities of regional
and global institutions should be clearly defined and the schedule for
the different activities of the programs need to be established in advance.
Regional proposal and work plan have been drafted by the respective regional
agencies. While there are some areas of differences between the global
and regional proposals, particularly in the area of budget estimates, the
World Bank will work closely with regional agencies to narrow the differences
and develop a more integrated regional and global plan before March 2002.
V.
International Conference and an expert group
meeting (11-15 March, 2002)
27. An international conference and an expert
group meeting to be organized, 11-15 March 2002,
in Washington, immediately following the UNSC's
meeting in New York. The results of the papers
and the two pilot studies will be discussed
in the first 3 days of the conference, 11-13
March. The objective of the conference is to
stimulate discussion, with a view to shed light
on critical issues and develop consensus.
The conference will be followed by an expert
group meeting (14-15 March) to be attended by
regional and international agencies, and recognized
experts. The objectives are: (a) to establish
clarity and consensus on the issues discussed
at the conference, with the view to plan the
implementation of the proposed round, and (b)
to discuss organizational issues, including
the coordination of regional and global work
plans, time tables and budget allocation. |
