ECONOMIC AND SOCIAL COUNCIL

Distr.

GENERAL

E/CN.3/1997/2

25 October 1996

ORIGINAL:

ENGLISH

Twenty-ninth session

10-14 February 1997

Item 3 of the provisional agenda*

The Secretary-General has the honour to transmit to the Statistical Commission the report of the Expert Group on Critical Problems in Economic Statistics on its meeting at United Nations Headquarters on 24 and 25 October 1995, which is contained in the annex to the present note. The report was prepared in accordance with a request of the Statistical Commission at its twenty-eighth session. 1/ It is transmitted to the Statistical Commission in accordance with a request of the Working Group on International Statistical Programmes and Coordination at its eighteenth session 2/ and includes decisions agreed upon by the Working Group at that session.

1/ Official Records of the Economic and Social Council, 1995, Supplement No. 8 (E/1995/28), chap. XVI.

2/ E/CN.3/1997/19, para. 22.

__________________

96-30827 (E) 191196 /...

*9630827*

Paragraphs Page

INTRODUCTION ..........................................	1	4
I. USER CONFIDENCE IN OFFICIAL ECONOMIC STATISTICS ......	2 - 7	4
II. FUNDAMENTAL PRINCIPLES OF OFFICIAL STATISTICS......	8 - 9	5
III. ADDRESSING USER CONFIDENCE IN THE COVERAGE, PERIODICITY AND TIMELINESS AND IN THE QUALITY AND RELEVANCE OF OFFICIAL ECONOMIC STATISTICS.........	10 - 39	5
A. Good behaviour ...................................	13 - 14	10
B. Effective communication with stakeholders ........	15 - 16	10
C. Devolution and deregulation ......................	17 - 18	11
D. Regional statistics ..............................	19	11
E. Measurement of the informal sector ...............	20 - 21	11
F. Globalization ....................................	22 - 23	12
G. Capital ..........................................	24 - 25	12
H. Resource accounting ..............................	26 - 27	13
I. Measuring price change ...........................	28 - 29	13
J. Quality of labour ................................	30 - 31	14
K. Labour and compensation ..........................	32 - 33	14
L. Economics of intangibles .........................	34	15
M. International price comparison ...................	35 - 37	15
N. Short-term economic indicators ...................	38 - 39	16
IV. PRIORITIES......................................	40	16
V. OTHER MATTERS ...................................	41 - 42	16
VI. POINTS FOR DISCUSSION......................	43	17

I.LETTER DATED 15 DECEMBER 1995 FROM THE CHAIRMAN OF THE EXPERT GROUP ON CRITICAL PROBLEMS IN ECONOMIC STATISTICS TO THE CHAIRMAN OF THE WORKING GROUP ON INTERNATIONAL STATISTICAL PROGRAMMES AND COORDINATION.................	18
II. LIST OF PARTICIPANTS..............................	20

1. At its twenty-eighth session, the Statistical Commission considered the item "Critical problems in economic statistics", which had been added to the agenda because of the current importance of the issues involved. 1/ The Commission recognized that dealing with problems related to the production and dissemination of timely, relevant and accurate economic indicators and their interpretation and use was crucial to the continued integrity of statistics. The Commission requested a group of countries and international organizations to systematize the various ingredients in the discussion and to report back to its Working Group on International Statistical Programmes and Coordination. After written exchanges of views among Working Group members, the United Nations Statistics Division convened a meeting of the Expert Group on Critical Problems in Economic Statistics at United Nations Headquarters on 24 and 25 October 1995. The present report details the findings of that meeting and incorporates decisions agreed upon at the April 1996 meeting of the Working Group. A letter from the Chairman of the Expert Group to the Chairman of the Working Group, in which the Chairman commented on the October 1995 meeting and noted, inter alia, four issues beyond the scope of the Expert Group that nevertheless called for the consideration of the Working Group, is contained in appendix I. The list of participants on the Expert Group is contained in appendix II.

2. The Expert Group focused on problems that affected the confidence of users in official economic statistics. While it acknowledged that there was an increasing number of non-official statistics, such as stock-market-related data, the Expert Group agreed that problems related to non-official statistics were beyond its mandate as defined by the Statistical Commission. Similarly, the Expert Group agreed that problems involving international statistics, such as the International Comparison Programme (ICP), were also beyond its mandate. 2/

3. The following three problems were identified:

(a) Absence of statistics;

(b) Distrust of statistics by users;

(c) Mismatch between statistics and user demand.

4. Absence of statistics implied that information was not available, was not available in sufficient detail or could not be provided when it was needed. Thus, winning user confidence in that case meant not only providing information but also ensuring sufficient coverage, regular periodicity and adequate timeliness.

5. As to user distrust of statistics, the institutional and organizational frameworks surrounding official statistics had an important effect on user confidence. The integrity of the statistical institution (i.e., whether it operated under a published statistical law and independently of political interference), the transparency of the statistical production and dissemination process, and public and equal access to information were three important concerns for winning the trust of users in official economic statistics.

6. Mismatch between statistics and user demand was a problem that raised a key concern: the quality of statistics. In order to satisfy user demand, statistical data must reflect economic realities accurately, and the statistical concepts being measured must be relevant to economic analysis and policy decision-making.

7. In short, the Expert Group found that the problems identified in paragraph 3 above might be approached in terms of:

(a) Coverage, periodicity and timeliness;

(b) Integrity, transparency and public access;

(c) Quality and relevance.

8. The Expert Group recognized the importance of the fundamental principles of official statistics, which were unanimously adopted by the Statistical Commission at its special session (11-15 April 1994). 3/ The 10 principles provided guidelines for establishing and maintaining a credible national statistical system, emphasizing the necessity to publish statistical laws, compile and make available official statistics on an impartial basis, and decide according to strictly professional considerations on the methods and procedures for the collection, processing, storage and presentation of statistical data.

9. The Expert Group agreed that the implementation of the fundamental principles - a national responsibility - was a necessary condition of user confidence in official economic statistics. In particular, adherence to the principles would help to ensure integrity, transparency and public access. The Expert Group thus decided to focus on user confidence in the coverage, periodicity and timeliness and in the quality and relevance of official economic statistics, and to propose initiatives for dealing with the issues that arose.

10. In approaching its task, the Expert Group attempted to answer three questions:

(a) What issues are raised by considering user confidence in the coverage, periodicity and timeliness and in the quality and relevance of official economic statistics?

(b) What activities can be initiated to address the issues thus identified?

(c) What mechanisms are appropriate for taking the lead in each such activity?

The Expert Group then identified 13 issues and proposed a number of activities and mechanisms for addressing them. A fourteenth issue, related to the "economics of intangibles", was added by the Working Group. For each activity and the mechanism assigned to it, a distinction was made between existing and proposed processes. In matching activities with mechanisms, the Expert Group made use of the following "taxonomy of mechanisms":

Activity Mechanism

Promulgation of standards Statistical Commission

Updating the international agenda Working Group

Coordination and implementation Intersecretariat groups

Updating of methods and standards Consultants/expert groups

Research into specific problems National statistical offices/ consultants

Sustained major research Informal ad hoc groups, constituted along the lines of the Voorburg Group on Service Statistics

Audit/programme evaluation Intersecretariat groups/consultants

11. The Expert Group felt that international institutions were the appropriate agents for issues involving coordination, development, implementation and the promulgation of standards. However, since detailed technical expertise - particularly concerning questions of data compilation - resided with national statistical offices, such offices must take the lead in technical research activities. The Expert Group recognized the need to create forums in which national statistical agency experts could address technical issues, and emphasized that informal ad hoc groups comprised of representatives of a small number of countries, constituted along the lines of the Voorburg Group on Service Statistics, could play an effective role in addressing such technical issues due to their greater flexibility compared to intersecretariat groups.

12. For each issue identified, the Expert Group reviewed ongoing research and existing coordination processes at the international level. In the light of its findings, it then elaborated proposals for further activities and recommended appropriate mechanisms for carrying them out. The 14 issues and the processes associated with them, as agreed by the Working Group, are listed in the table and are discussed below.

Concern	Issue	Existing Process		Proposed processes		Document for Statistical Commission
		Activity	By whom	Activities	By whom
Coverage, periodicity, timeliness
Which activities are proper for a statistical office to undertake and which of them increase its credibility?	Good behaviour Effective communication with stakeholders			Develop draft code of best practice for official statistics for consideration by the Statistical Commission	United Kingdom, United States of America, Mexico, IMF and others	Draft code of best practice
What problems of coverage are created by changes in the policy environment?	Devolution and deregulation			Countries share research/ experience	Informal ad hoc group. Convener (to be determined), Eurostat and others to participate	Progress report
	Regional statistics	Development of handbook on regional accounts	Eurostat	Promulgate handbook on regional accounts	ISWGNA
What problems of coverage are created by changes in economic organization?	Informal sector	Consultancies on Africa/Asia. Work in transition countries. Methodological work	UNSD, Eurostat/ OECD, ILO	Status report and proposals for future work	UNSD in collaboration with ILO, Eurostat, regional commissions; country input as appropriate a/	Status report and proposals for future work
	Globalization	Research on trade not passing through customs barriers; foreign affiliates trade; subcontracting across countries; attribution of production of multinational enterprises to countries	OECD/ Eurostat	Improved coordination, especially with WTO, WCO, UNSD, IMF, UNCTAD	OECD, Eurostat
				Prepare paper on Canadian experience in developing globalization measures	Canada, in collaboration with OECD	Information paper
Quality, relevance
How should we measure wealth, output and productivity?	Capital			Countries share research/ experience	Informal ad hoc group. Convener (Australia), France, World Bank and others to participate	Progress report
	Resource accounting	Countries share research/ experience	London group, in collaboration with Nairobi group, as appropriate	Prepare progress report	London group	Progress report
	Prices	Countries share research/ experience	Ottawa group	Prepare progress report	Ottawa group	Progress report
	Quality of labour	Countries share research/ experience	Voorburg Group	Prepare progress report	Voorburg Group	Progress report
	Labour and compensation			Countries share research/ experience	Informal ad hoc group Convener (France), Australia, ILO and others to participate	Progress report
	Economics of intangibles			Prepare scoping paper and recommendations for further work	United Kingdom, United States, Canada, Mexico, Eurostat	Scoping paper
Is it necessary to compare and aggregate across countries and, if so, how should it be done?	ICP	Extensive ongoing programme	Statistical Commission	Develop draft terms of reference and list of possible consultants for programme evaluation of ICP. If programme evaluation to proceed, steering committee to be appointed by chair of the Statistical Commission	UNSD, World Bank, United States of America and others	Discussion paper
What is required to signal the need for short-term intervention?	Quality short-term economic indicators	Surveillance system	IMF	Country implementation	National statistical offices, central banks, ministries

Key:

OECD Organisation for Economic Cooperation and Development

ISWGNA Inter-Secretariat Working Group on National Accounts

Commission Statistical Commission, twenty-ninth session (1997)

UNCTAD United Nations Conference on Trade and Development

IMF International Monetary Fund

WTO World Tourism Organization

WCO World Customs Organization

ILO International Labour Organization

ICP International Comparison Programme

Eurostat Statistical Office of the European Union

UNSD United Nations Statistical Division

a/ India is forming a group on informal sector statistics.

13. The credibility of national statistical agencies was largely determined by their behaviour, in particular by their relationship with the users of economic information. The timing and mode of data release, access facilities, and the extent to which statistical work was extended into the area of analysis were discussed by the Expert Group. As to the latter, no agreement was reached. Although some participants stressed the need to respond in a flexible manner to the demand for economic information for decision-making, including forecasting, others raised the potential conflict of interest in national statistical offices, first making projections and then publishing official statistics that either corroborated or invalidated their own projections. There was agreement, however, that statistical work should extend to include analytical presentations of economic information.

14. It was felt that a dialogue among countries on the issue of good behaviour should be initiated. Since the issue was relevant to all national statistical offices, the Statistical Commission would be the appropriate forum for discussion. The Expert Group proposed that a code of best practices be developed to complement the fundamental principles of official statistics. The Working Group agreed with that recommendation and invited the United Kingdom of Great Britain and Northern Ireland, the United States, Mexico, the International Monetary Fund (IMF) and other volunteers to develop such a draft code for discussion by the Statistical Commission at its twenty-ninth session.

15. Effective communication was vital for statistical institutions in their contact with both the providers of basic information and the users of their statistical products. Many countries were experiencing difficulty in collecting information, due to a low response rate to surveys and also to the poor quality of information obtained. The promulgation of standards and effective use of modern information technology were being explored as ways to alleviate those problems.

16. The problem of communicating how information should be used involved three challenges: how to measure the uncertainty underlying estimates of economic aggregates, such as gross domestic product (GDP); how to communicate that degree of uncertainty to users; and how to communicate the conceptual relevance of alternative statistical measures, such as matching available price measures to administrative and policy uses, to different users. It was in the best interests of a statistical agency to assume responsibility for the statistical literacy of the public at large: strategic education of important elements in the media, for instance, would have a multiplier effect on the understanding of economic data, and would thus directly enhance the perception of its quality. The Working Group accepted that, again, countries would benefit from a code of best practices.

17. Devolution and deregulation were recent examples of changes in the policy environment that caused coverage problems for the statistical system. The tendency in some countries to decentralize governmental functions to subnational governmental units, such as federal states, regions or municipalities, had two consequences for the measurement of national aggregates. First, the number of data sources increased, making relevant information more difficult to access. Second, there was a potential problem of comparability in the information provided by the smaller units. The deregulation of certain industries created a similar challenge for statistical agencies: the regulatory bodies that had formerly acted as centralized providers of information no longer had that role and must be replaced by a number of new sources.

18. Such problems mainly affected countries that pursued devolution and deregulation policies. Consequently the appropriate way to address them at the international level would be for the countries concerned to share their research and practical experiences, which could be most effectively achieved in an informal, ad hoc setting like that of the Voorburg Group. The Working Group agreed with the formation of such a group; the United Nations Statistical Division would seek interest in a country convening the group. The Statistical Office of the European Community (Eurostat) expressed interest in participating in the group. The group should prepare a progress report on its deliberations for the Statistical Commission at its twenty-ninth session.

19. For large countries with a federal structure, the issue of balancing the need for data among the regional, state and national levels was increasingly a matter of concern in view of the limited resources available for statistical activities. For example, regions might be defined independently of particular governmental jurisdictions, and different levels of government had their own specific demands for information for the purposes of economic policy-making. In regional accounting, inconsistencies could arise between regional estimates - of, for example gross regional product - and the nationally estimated aggregate, GDP. The Working Group noted that Eurostat, under the auspices of the Inter-Secretariat Working Group on National Accounts (ISWGNA), was preparing a handbook on regional (subnational) accounting which would also be of interest to non-European countries.

20. The Expert Group acknowledged that measuring the unorganized, or informal, sector of the economy was a most pressing need if the reliability of macroeconomic data were to be improved for policy decision-making, particularly - although not exclusively - for developing countries and countries in transition. The problem was challenging because it had social and environmental implications as well as economic ones: dealing effectively with the informal sector would require an interdisciplinary approach.

21. The Expert Group noted that there were already some ongoing activities at the international level dealing with the informal sector. The United Nations Statistics Division had commissioned consultants to carry out a study in a number of African countries to compare the experience of different national statistical offices, the results of which had been presented at a regional seminar (Addis Ababa, 17-21 June 1996). A similar study would be carried out for Asia. The Organisation for Economic Cooperation and Development (OECD) and Eurostat had done extensive work on measuring the informal sector in European countries in transition. The International Labour Organisation (ILO) had also undertaken work of relevance to the issue. Relevant experiences might be exchanged. The Working Group decided that the United Nations Statistics Division, in collaboration with Eurostat, ILO, OECD and the regional commissions, and with country input as appropriate, should prepare a status report for the Statistical Commission at its twenty-ninth session on activities dealing with the informal sector. Furthermore, India informed the Division that it was prepared to convene a group on the issue.

22. The globalization of the world economy had statistical consequences. The emergence of trade areas was removing conventional trade barriers and hampering the collection of trade statistics: multinational enterprises operated and produced in many countries. There was thus a need for new legal and practical arrangements for the collection and preservation of information in the new environment.

23. OECD and Eurostat were undertaking extensive research into measuring trade where no customs barriers existed; measuring establishment trade; subcontracting across countries and assigning the production of multinational enterprises to countries. The Expert Group suggested that the findings of such research needed to be more widely disseminated, in particular among developing countries concerned with such problems. At the April 1996 meeting of the Working Group, OECD and Eurostat agreed with that proposition. In addition, the Expert Group emphasized the increasing need for effective cooperation among the principal agencies involved in collecting international trade data, in particular the United Nations Statistics Division, the United Nations Conference for Trade and Development (UNCTAD), the World Bank, the World Trade Organization and the World Customs Organization. The Working Group was informed of the work being undertaken in Statistics Canada on developing a database on globalization trends in the Canadian economy. Statistics Canada was invited, in collaboration with OECD, to provide an information paper to the Statistical Commission at its twenty-ninth session.

24. Current measures of the economy were largely oriented towards capturing economic transactions - i.e., measuring flow variables. However, in the context of the ongoing debate on measuring human well-being, more attention was being paid to stock variables, such as produced and non-produced capital and wealth. The increasing importance of productivity studies were also concentrating attention on the quality of capital stock estimates. Few countries currently included balance sheets in the scope of their national accounts.

25. The Expert Group felt that, in addition to theoretical problems, there were many problems related to measuring capital stocks of a practical nature. Research results and experience could usefully be exchanged among countries attempting to compile balance sheets, and again an informal ad hoc group along the lines of the Voorburg Group might play a useful role. The Working Group accepted that proposal. Australia volunteered to convene an ad hoc group. France and the World Bank agreed to participate, and other interested countries would be invited. A progress report on the group's activities was to be provided to the Statistical Commission at its twenty-ninth session.

26. The general heightened awareness of the interaction between the economic and environmental processes had increased the demand for information on the stock of natural resources and the changes therein. A first step to taking stock of the environment was to develop measurements of environmental assets in physical units. However, in order to relate environmental assets to macroeconomic aggregates, it was necessary to assign an economic value to such assets, giving rise to both conceptual and practical problems.

27. A variety of country experiences in the practical aspects of natural resource accounting was emerging. The London Group was a forum for countries interested in developing environmental statistics. Items for consideration ranged from valuation and the physical extent of natural resources to energy use by industry, pollution abatement and control expenditures, and emissions in production and consumption activities. Participants so far had mainly been OECD members, but OECD itself, the United Nations Statistics Division and Eurostat also participated. The Expert Group felt that the London Group appeared to be an appropriate forum for debating issues in the area, but that its proceedings needed to be more widely disseminated. The Working Group noted also the work of the Nairobi Group on environmental matters. The Working Group invited the two groups to collaborate, as appropriate, and invited the London Group to prepare a progress report on its activities for the Statistical Commission at its twenty-ninth session.

28. There were various dimensions to the problem of measuring price changes adequately. Consumer price indices and producer price indices were well-known statistical concepts; however, neither measurement was designed to provide an economy-wide picture of inflation. In addition, price measures might suffer from a number of practical problems: the changing nature of output and the increased heterogeneity of products made it very difficult to compile consistent price measures over time; service industry outputs were notoriously difficult to define and to price; and measuring the price of governmental outputs presented major conceptual problems.

29. Although all countries faced similar problems concerning price measurement, they took different approaches to solving those problems, with implications for the quality and credibility of different price measures which needed to be analysed. Statistics Canada convened a meeting of a number of countries in Ottawa in 1994 in order to conduct informal consultations on those issues; a further meeting was held in 1995. The Expert Group took note of the meetings and invited the Ottawa Group to initiate a discussion with the Working Group on its focus and agenda. The Ottawa Group was also encouraged to disseminate its proceedings more widely. The Working Group agreed that the Ottawa Group was an appropriate forum in which countries could share their experiences and research on price issues. The Ottawa Group was invited to prepare a report on its activities for the Statistical Commission at its twenty-ninth session.

30. Various themes were identified under the heading - in particular, the increasing importance of productivity studies, given that the reliability of the labour input denominator in the productivity equation was being questioned in a number of countries. In addition, it was argued that some health and education expenditures of the general population were more appropriately treated as investment than as consumption. Indeed, such expenditures could be reanalysed through adjustments to national accounts aggregates, using satellite accounts.

31. At its 1995 meeting, the Voorburg Group on Service Statistics recognized the importance of the quality of labour to the performance of service industries and agreed to devote part of its agenda for 1996 to a thorough discussion of the problem. Several Scandinavian countries volunteered to do the necessary research work. The Expert Group suggested that the Voorburg Group was the appropriate forum for taking those issues further, the lessons of which could then be generalized beyond the service sector. The Working Group agreed and invited the Voorburg Group to provide a progress report for the Statistical Commission at its twenty-ninth session.

32. Employment was typically measured using both household and business surveys. Reconciling sources to obtain consistent and coherent views of labour inputs and incomes often required the resolution of complex problems, sometimes through the construction of labour accounts. The deregulation of labour markets in some countries meant that the traditional measures of labour compensation - average earnings and labour award rates of pay indices - were no longer appropriate. Some countries were in the process of constructing labour cost indices that raised many complex conceptual issues, and the phenomenon of structural unemployment was also attracting increasing attention.

33. In addition to individual countries, a number of international and supranational organizations, such as OECD, ILO and Eurostat, started work in the field. The Expert Group suggested that there was scope for a concerted effort to share research and experience in the area of labour market statistics, and the creation of an informal ad hoc group, along the lines of the Voorburg Group and under the leadership of one of the more developed countries, was recommended by the Expert Group. At the Working Group meeting, France volunteered to convene such a group. Australia and the ILO expressed interest in participating, and other countries would be invited. The group would prepare a progress report for the Statistical Commission at its twenty-ninth session.

34. The Working Group felt that the statistical measurement of "intangible" outputs was sufficiently problematic for work to be commissioned on the topic. While it was noted that the new North American Industrial Classification System (NAICS) had defined an information industry, further statistical initiatives to better understand the issues involved should be envisaged. The Working Group invited the United Kingdom, the United States, Canada, Mexico and Eurostat to prepare, for the Statistical Commission, at its twenty-ninth session, a scoping paper on the issues, together with recommendations for future work.

35. Although the ICP was outside its mandate, the Expert Group did consider a number of problems associated with the implementation of the ICP. One such controversy had arisen from the use of varying exchange rates and purchasing power parity approaches to country comparisons by international organizations. Since that use implied the use of different weights in the calculation of the world growth rate, different results were derived, depending on the conversion factor used.

36. The Expert Group felt that a thorough programme evaluation of the ICP programme was warranted, and recommended updating the documentation on its methodology and making the entire implementation process of the ICP more transparent - for example, by explaining who was participating and who was not, and why. The Expert Group expressed the view that, instead of moving immediately towards global implementation, a pilot study should be executed for a single global region. Since the current programme was being carried out under the auspices of the Statistical Commission, the Commission was seen by the Expert Group to be the appropriate forum for reviewing a formal evaluation.

37. There was extensive discussion on the proposal for an evaluation of the ICP at the Working Group meeting. It was agreed that the United Nations Statistics Division, the World Bank, the United States and other co-opted countries should prepare a discussion paper for the Statistical Commission at its twenty-ninth session. The paper should contain draft terms of reference for a possible formal evaluation of the ICP and a list of possible consultants. It was further agreed that, if the formal evaluation was to proceed, a steering committee should be appointed by the chair of the Statistical Commission.

38. The Expert Group considered the proposed role of the International Monetary Fund (IMF) in monitoring key economic indicators. 4/ IMF surveillance was expected to encourage countries to follow the standards established for such indicators. The emphasis of the IMF initiative was to provide users - in particular, financial markets - with the high-quality information necessary to allow them to better understand and respond to current economic events.

39. The Expert Group emphasized the importance of the IMF initiative for improving the reliability, comparability and credibility of economic information worldwide, noting that national statistical offices would be responsible for its implementation. The Working Group endorsed those views.

40. After listing and reviewing the most critical issues described above, the Expert Group decided not to set any priorities in its report to the Working Group, since mechanisms had been identified for addressing each of the various issues and there was no apparent conflict of priorities. Moreover, in a number of cases, the Expert Group had recommended the formation of informal ad hoc groups constituted along the lines of the Voorburg Group on Service Statistics. It was noted that it was ultimately up to the countries concerned to express their interest in such groups and the extent to which they would participate. It was noted that it was critical for one interested country to take responsibility for convening each of the proposed new groups.

41. While generally supportive of the proposals of the Expert Group to initiate informal consultation groups in order to address some of the issues identified, the Working Group requested those groups to have a clearly specified focus and to make their workings open and transparent. The Working Group also decided that their legitimacy and recognition by the international statistical community would depend upon decisions of the Statistical Commission. The United Nations Statistics Division offered its services to promote knowledge of those groups and to make their findings widely accessible. The Chairman of the Working Group agreed to write to countries in order to inform them about, and to solicit wider participation in, the informal groups. The newly formed groups, and some of the existing groups, were invited to report back to the Statistical Commission at its twenty-ninth session (see cols. 5, 6 and 7 of the table for details).

42. The Working Group noted that the Expert Group's two proposals - namely, systematically to analyse the implications for economic statistics of the recent world summit conferences, and to articulate better the research agenda of the Inter-Secretariat Working Group on National Accounts - had been taken up under the Working Group's respective agenda items 6, "Statistical implications of the United Nations summit meetings", and 3, "Implementation of the System of National Accounts, 1993 (SNA)".

43. The Statistical Commission is asked:

(a) To endorse the decisions of the Working Group on International Statistical Programmes and Coordination;

(b) To refer to document E/CN.3/1997/3 for further points for discussion.

1/ See Official Records of the Economic and Social Council, 1995, Supplement No. 8 (E/1995/28), chap. XVI.

2/ See the letter of the Chairman contained in appendix I below for further observations of the Expert Group on ICP.

3/ See Official Records of the Economic and Social Council, 1994, Supplement No. 9 (E/1994/29), chap. V.

4/ See IMF papers, "Report of the Executive Board to the Interim Committee on Data Provision on surveillance and standards to guide members in the provision of data to the public" (29 September 1995) and "Standards for the dissemination by countries of economic and financial statistics: the special data dissemination standard" (April 1996).

The Expert Group on Critical Problems in Economic Statistics, which was established by the Statistical Commission at its twenty-eighth session, met once in New York on 24 and 25 October 1995 and finalized its report through bilateral consultations with Group members. The Expert Group appreciated the efforts of the United Nations Statistics Division in facilitating its work.

The Expert Group was conscious of the potential breadth of issues arising from the discussion at the twenty-eighth session of the Statistical Commission and took the view that it should narrow the scope of its deliberations to a few key issues that could reasonably be dealt with in the short-to-medium term within the international statistical community. The Expert Group therefore concentrated on problems that affect user confidence in national programmes of economic statistics. The issues identified in the report are considered to be relevant to a wide spectrum of countries at varying stages of statistical development.

In addition, four issues arose that the Expert Group considered to be outside its scope but nevertheless important issues for the consideration of the Working Group. First, there was a concern that the research agenda of the Intersecretariat Working Group on National Accounts (ISWGNA) needs to be better articulated and prioritized to ensure that it meets the needs of countries. The Expert Group identified as a specific example the need for a handbook on regional accounts, which is in fact currently being prepared by the Statistical Office of the European Communities (Eurostat). Changing economic, social and institutional arrangements in many countries are resulting in the need for country guidance in various aspects of the System of National Accounts. In particular, the Expert Group mentioned the issue of measuring inventories and changes thereto as one that needs to be included on the ISWGNA research agenda.

Secondly, the Expert Group was very conscious of the national and international resources that are being expended on the International Comparison Programme (ICP). The Group was also aware that there is not yet a consensus among international agencies about whether world growth rates are best measured using exchange rate or purchasing power parities. There was also concern that the uses to which purchasing power parity statistics are being put may be inappropriate, given the many conceptual and practical difficulties underpinning their compilation. In short, the Expert Group felt that there should be a formal evaluation of ICP, which would probably be best undertaken by a suitably qualified consultant. Issues that the Expert Group would like to see included in the terms of reference of such an evaluation are:

(a) The overall transparency of the programme;

(b) The suitability of the methodologies being adopted vis-à-vis the uses to which the data are being put;

(c) The operational experience of a number of countries in supplying the data for ICP.

Thirdly, the Expert Group felt that it would be very useful for an international agency and/or a country or countries systematically to study recent major international conferences to assess their implications for economic statistics. The Expert Group was impressed with the thoroughness with which social statistics issues arising from the World Summit for Social Development had been summarized for the attention of the Statistical Commission at its twenty-eighth session.

Fourthly, as is made clear in the report, informal ad hoc groups, such as the Voorburg Group on Service Statistics, in which countries share their experiences and research agendas in defined statistical fields, were considered by the Expert Group to be very important for addressing some of the critical problems in economic statistics, and four new groups were proposed. It was felt that such groups must satisfy three essential criteria in order to succeed:

(a) There must be a host country willing to take the initiative;

(b) In order to maintain its focus, each group must have a specific mandate to pursue, such as the mandate of the Voorburg Group to develop the services component of the Central Product Classification;

(c) The results of the work of each group must be disseminated.

The Working Group may wish to consider those criteria in assessing proposals of the Expert Group to create four new such groups.

Subject to the deliberations of the Working Group, the Expert Group is willing to undertake further work in preparation for the consideration of the item "Critical problems in economic statistics" by the Statistical Commission at its twenty-ninth session.

Australia	Mr. Rob W. EDWARDS (Chairman)
Canada	Mr. J. RYTEN
India	Mr. S. SATHYAM
United States of America	Mr. Paul BUGG
United Nations	Mr. Hermann HABERMANN Mr. Stefan SCHWEINFEST
Organisation for Economic Cooperation and Development	Mr. C. L. KINCANNON