Database and Development Indicators

Strengthening and Collection of Purchasing Power Parity Data in Selected Developing Member Countries

• Introduction
• Issues
• The Technical Assistance
• The President's Decision
• Contact

Introduction

The International Comparison Program (ICP) - established in 1968 as a joint venture of the United Nations, the University of Pennsylvania, and the World Bank for calculating purchasing power parities (PPPs) - expanded to the status of almost a global program by 1993. Since then a large number of developing countries have opted out of the ICP, partly due to financial and technical reasons. However, the industrialized countries have continued collecting ICP data on a regular basis. The Organisation for Economic Cooperation and Development (OECD), in collaboration with the European Commission Statistics Office (EUROSTAT), is responsible for coordinating the collection of price data to estimate PPPs for its respective member states. In Asia, the Asian Development Bank (ADB) developing member countries (DMCs) have realized that the ICP is a useful tool for comparing the income and poverty data between different countries of the region. However, limited statistical capabilities preclude many countries from carrying out full ICP computations. In the recent past the World Bank has assumed the responsibility for global coordination of the ICP in non-OECD countries, with a view to improving data collection. In September 2001, the World Bank invited ADB to participate in the program and take responsibility for the regional coordination in the Asia and Pacific region. ADB has accordingly agreed to participate in this important global initiative for the next round scheduled for 2003.¹

Top

Issues

The shortcomings associated with the use of exchange rate-based conversions are well known. Because market exchange rates are based on short-term factors and are subject to substantial distortions because of speculative movements and government interventions, comparisons of activity and income levels based on exchange rates, even when averaged over a period of time, yield unreliable and misleading results. By establishing purchasing power equivalence, which measures the price in local currency of the same quantity of goods and services able to be purchased for a unit of currency (typically a $US) in a numeraire country, PPP conversions allow cross-country comparisons of the levels of economic aggregates, free of price and exchange rate distortions.

However, PPP-based estimates have been the subject of major criticism in the past. These criticisms have been directed mainly at the process of collecting, compiling, and disseminating the statistics produced by the ICP. In the past, a lot of the work on the ICP was aimed at identifying improved methodologies for computing ICP aggregates, while the practicalities of the actual data collection were given insufficient priority. More recently, questions have nevertheless been raised on the item selection procedure, pricing of items, universal coverage and, particularly, the data quality. Weighting was based on household income and expenditure surveys (HIESs), which were often not consistent between countries, and were not always compatible with the ICP concepts. One outcome was that the same weights were arbitrarily supplied by some countries for use in several detailed ICP categories. Weaknesses in such surveys have also contributed to less than full data availability.

There are also problems of aggregation within basic headings. One set of problems arises from the fact that for a particular commodity, a country's prices may vary markedly by subregion and outlet type, particularly between rural and urban outlets. Problems associated with such variation may lead to unwanted influences on parities. There are also estimation problems, including those of weighting, originating in the data that are available and in the procedures used to aggregate basic heading parities up to the level of gross domestic product (GDP) and its major aggregates. The expenditure weights and other basic data needed for a correct aggregation are sometimes missing or not comparable across countries. In some instances, data on expenditure by specific basic headings are missing, because the expenditure data are available only for aggregates larger than the basic headings. There is now broadly based acceptance of the fact that these limitations need to be corrected.

A critical issue in setting up the item specifications is to ensure that the products priced are both comparable between countries and representative of the expenditures within each country. In the past, countries were asked to provide prices of certain specified commodities that were not necessarily representative in terms of consumption and final use in the country in question. The World Bank did most of the estimation, but there was lack of transparency in providing the countries with the methodology for computing PPP estimates, and so the implications of the types of prices the national statistics offices (NSOs) supplied were not fully understood. As a result the NSOs have not fully comprehended the ICP estimates, and have not strongly advocated their use at the country level. Hence, the countries have not had a real sense of ownership of the project.

An independent evaluation commissioned by the United Nations Statistical Commission concluded that there is no alternative to PPP data for cross-country economic comparisons or for poverty measurement at the national, regional, and global level.² As noted earlier, the ICP has become a regular and integral part of national statistical work in OECD countries and in the United Nations Economic Commission of Europe region, but there has been very little progress in the Asia and Pacific DMCs due to lack of adequate funding, technical expertise, and serious ownership on the part of the governments. For some countries these impediments will continue to influence their participation in the ICP. Therefore, there is a need to assist the DMCs in devising a survey framework for integrating ICP data collection into the larger context of routine national price collection work for the consumer price indexes (CPIs) as well as to devise a survey framework to integrate the ICP with poverty measurement and monitoring in the region. It is also important that countries improve their national accounts and adopt greater disaggregation of the accounts. Country capacities will need enhancement through a program of technical cooperation.

Top

The Technical Assistance

A. Purpose and Output

This regional technical assistance (TA) will strengthen and build sustainable capabilities of NSOs for collecting the data necessary to compile PPP-based comparisons for GDP and its main sub-aggregates. It will aim to help the NSOs of the participating DMCs to (i) mainstream ICP with national statistical work by integrating it as far as possible with the national accounts and price collection programs, (ii) support the harmonization of ICP with data collection systems for the CPI and other prices, (iii) assist NSOs in conducting specific surveys for collecting additional data not covered by existing surveys, (iv) assist NSOs in poverty analysis by encouraging the use of ICP data, (v) assist countries in identifying a comparable and representative list of items across the countries, (vi) provide advisory services to NSOs as required in carrying out the above activities, (vii) organize regional workshops and training as and when necessary, and (viii) prepare manuals and handbooks where necessary to institutionalize ICP methodologies in ADB DMCs.

Harmonization of the ICP with CPI-related work will contribute to a reduction of the costs of data collection, and will improve coordination among data collecting agencies. An integrated price collection mechanism will reduce the burden and maximize benefits. Most of the ADB DMCs, except some Pacific Island countries, collect some price data regularly to assess consumer prices and the rate of inflation. Hence, the infrastructure that already exists for the collection of price data should be extended to the collection of prices required for the calculation of PPPs.

The HIES estimates should underlie national accounts and provide weighting patterns for the CPI. Poverty analysis within a country requires detailed data, which can be classified by different income groups. Another important dimension is the integration of detailed price data from ICP with detailed expenditure data from HIES. There is a need to establish the greatest possible synergy between the ICP and HIESs. Characteristically, the HIES provides detailed expenditure data, but the price data generated are restricted in scope and coverage. The ICP, on the other hand, provides detailed price statistics. The integration of the ICP and the HIES could potentially provide cross-benefits to each program. To compute PPPs, the ICP requires detailed subcategory expenditure weights that are normally available only from the HIES. The detailed data that ICP makes available, besides being useful for international comparison of macroaggregates and cross-country regressions of the policy-poverty nexus, provide a coherent basis for assessing poverty and the impact of structural adjustment programs, the primary concerns of governments and international agencies.

The TA will contribute to a regional and national structure for implementing the ICP on an ongoing basis; calculation of PPPs for measuring economic and social progress, including poverty; preparation of methodological publications, including manuals and handbooks for institutionalizing ICP methodologies; and an improved capacity for the collection and use of price and expenditure data within participating countries.

B. Methodology and Key Activities

The regional program will cover 23 ADB DMCs.³ The Central Asian republics will not be included, as they have already been covered in the program by EUROSTAT and OECD. Among the Pacific countries, only Fiji Islands and Papua New Guinea will be included, as the others do not yet have credible price survey systems in place. Hong Kong,China; Republic of Korea; Singapore; and Taipei,China will be invited to participate in the full-round of ICP comparisons but will be asked to fund their own programs. The remaining 19 countries will be divided into two groups. Some limited financial assistance will be provided to both groups. Group 1 countries⁴ will be invited to participate in the full round ICP, whereas group 2 countries⁵ will be included in the harmonized ICP-CPI program (based on a limited coverage approach).

The strategy of the TA will bring together the continuing efforts and experiences to produce PPPs, drawing upon OECD/EUROSTAT experience. The TA effort will contribute to more timely and reliable data. The data will help improve the allocation of international resources for poverty reduction, better understanding of the process of economic change and growth, and the means to monitor progress towards meeting the Millennium Development Goals.

The TA will support the following activities:

1. A review of the current status of national accounts and price statistics in the participating DMCs will be undertaken. The review will identify appropriate institutional mechanisms and country-specific tasks required to achieve the objectives of the TA. TA implementation will be guided by the regional and global management coordination teams.⁶
   
   
2. At an inception workshop, the DMC representatives will discuss the status of national accounts and price statistics and prepare their work plans for implementing this round of the ICP. Each participating country will prepare a detailed work plan, which should include the survey framework, questionnaires, specification and selection of items, selection of outlets, initial processing and editing of data, needed software, as well as methods for computing price ratios. A set of harmonized survey guidelines and list of core commodities for regional linking will be prepared.
   
   
3. A review of the available national accounts data will be undertaken. Disaggregated GDP aggregates will be estimated by the NSOs to provide appropriate weights for estimation of PPPs. Each participating country will submit price and other related data to ADB and the regional management coordinating team for computing PPP-adjusted GDP and aggregates at the regional and global level.
   
   
4. A training program on the ICP will be conducted jointly with the United Nations Statistical Institute for Asia and Pacific (SIAP) for statisticians, economists, and policy and decision makers to deepen their understanding of the relevance of the ICP.
   
   
5. A concluding workshop will be held before the completion of the TA, at which the participating DMCs will present their results, including the price data sets they have produced and their future course of action.

C. Cost and Financing

The TA is estimated at approximately $1,200,000 over a period of 3 years, but the budget required from the 2002 Indicative Planning Figure will amount to $400,000. The TA will be financed on a grant basis from ADB's TA funding program and will provide some seed money to Group 1 and Group 2 countries based on the letters of agreement with each DMC. The participating DMCs are also expected to make substantial contributions to the TA, mostly in kind, as it will require that most of the in-country expenses be shouldered by the country's statistical offices. Moreover, the participating countries will need to sustain the TA activities and continue producing the required price and national accounts data, on a regular basis, after TA completion. The detailed cost estimates are presented in Appendix 2.

D. Implementation Arrangements

The TA will be implemented over a period of 3 years beginning in December 2002. ADB, with technical inputs from consultants and the regional and global management coordinating teams, will be the Executing Agency for the TA. The Development Indicators and Policy Research Division of the Economics and Research Department, in close collaboration with the other regional and knowledge departments, will be the lead unit for implementation. Letters of agreement will be executed between ADB and each participating DMC specifying the nature and scope of ADB assistance and the obligations of the DMC. The letters of agreement will specify the maximum amount allocated to the DMC and will outline the activities that can be funded under the TA. The activities to be undertaken by the implementing agencies, which will be the NSOs of the participating countries, and the expected outputs of the TA will be discussed and finalized at the inception workshop.

The TA will require recruitment of international consultants⁷ for 12 person-months and a domestic research assistant for about 30 person-months, on an individual basis under the TA over a period of 3 years. The international consultants will assist especially in the review, and assessment, and preparation of technical background papers for the workshop; preparation of a comprehensive and standardized survey framework; organization and conduct of workshops and training; monitoring work progress in the participating DMCs; and provision of technical advice to the DMCs in the collection and compilation of price data and preparation of disaggregated GDP estimates. In addition, there may be a need to hire national consultants, as appropriate and subject to ADB approval, not exceeding 2 person-months, by each participating country. The national consultants will assist the respective NSOs of the participating DMCs in conducting special price surveys. Outline terms of reference for the international consultants and research assistant are given in Appendix 3. The consultants will be recruited in accordance with ADB's Guidelines on the Use of Consultants and other arrangements satisfactory to ADB.

The data collected from the participating countries will be provided to the regional and global management and coordination teams for both regional and global comparisons and will also be used by ADB in its operations. The TA will assist the participating DMCs in disseminating the outputs emerging from the TA.

Top

The President's Decision

The President, acting under the authority delegated by the Board, has approved the provision of technical assistance, on a grant basis, not exceeding the equivalent of $400,000 from the 2002 Indicative Planning Figure subject to the availability of TA resources, for Strengthening and Collection of Purchasing Power Parity Data in Selected developing member countries, and hereby reports this action to the Board. Since this is a multiyear TA with a total budget of $1,200,000 over 3-year implementation period, the President has approved, in principle, the balance of $400,000 in 2003 and $400,000 in 2004 from the Poverty Reduction Cooperative Fund, subject to the availability of TA resources.

Top

Contact

Mr. Bishnu D. Pant
bpant@adb.org

Top

Top