UNITED NATIONS

Economic and Social Council

Unofficial copy for information purposes prepared from the official electronic file.

E/CN.3/AC.1/1996/R.4
24 January 1996
ORIGINAL: ENGLISH

WORKING GROUP ON INTERNATIONAL STATISTICAL PROGRAMMES AND COORDINATION

Eighteenth session

New York, 16-19 April 1996

1. At its twenty-eighth session (New York, 27 February-3 March 1995), the Statistical Commission, in its consideration of agenda item 12, "Measuring and monitoring economic and social development", requested its Working Group on International Statistical Programmes and Coordination to consider further the statistical implications of the follow-up to the World Summit for Social Development (Copenhagen, 6-12 March 1995). The Commission also established the Expert Group on the Statistical Implications of Recent Major United Nations Conferences to draw up a work programme reflecting the major action areas identified by the Summit and indicating where international statistical work in the social field should be concentrated, with due regard to the Programme of Action of the International Conference on Population and Development (Cairo, 5-13 September 1994) and the expected outcome of the Fourth World Conference on Women (Beijing, 4-15 September 1995). The Commission requested that, in the light of the policy areas identified, the Expert Group propose specific statistical activities for the period 1996-1998, and assign priorities to them based on the skills and financial resources likely to be available in international organizations and countries. 1/

________________________

96-01603 (E) /...

*9601603*

2. The Expert Group held a technical meeting in Oslo on 10 June 1995, immediately following the meeting of the Siena Group on Social Statistics. The Expert Group meeting was attended by Ms. C. Denell (Sweden), Mr. I. Ewing (New Zealand), Mr. G. Guteland (Sweden), Mr. F. Gutierrez (Mexico), Mr. L. Lundgren (Sweden), Mr. R. Madden (Australia), Ms. M. McEwin (Australia), Mr. B. Petrie (Canada) and Mr. D. Roberts (United Kingdom of Great Britain and Northern Ireland), as well as by representatives of the United Nations Statistics Division, the Economic Commission for Europe (ECE) and the Economic Commission for Latin America and the Caribbean (ECLAC), the International Labour Office and the Statistical Office of the European Community (EUROSTAT). The Expert Group held its second meeting in Geneva on 13 June 1995. In addition to the above-mentioned participants, that meeting was attended by Mr. J. Olenski (Poland), Mr. J. Pullinger (United Kingdom), Mr. S. Sathyam (India), Ms. J. Szczerbinska (Poland) and Mr. L. Thygesen (Denmark). Mr. Guteland served as Chairman and Mr. Madden as Rapporteur at both meetings. The report of the Expert Group is contained in the annex.

3. The Working Group may wish to:

(a) Endorse and recommend to the Statistical Commission for adoption the list of policy themes and main areas of social concern recommended by the Expert Group in paragraph 8 of its report as a subject-matter framework for further work in statistics to follow up recent major United Nations conferences;

(b) Recommend to the Statistical Commission the adoption of the list and menu of statistical indicators to form a minimum national data set (MNDS), as contained in paragraphs 96-98 and in the appendix of the report, following technical consultation and refinement with national and international specialists in these fields, for the purposes outlined in paragraphs 11-16 of the report;

(c) Take note of the data gaps and other needs for further work identified by the Expert Group in sections II-VI of its report, and request the Statistics Division to invite the international statistical services concerned to comment on their plans and objectives in these fields for the period 1996-1999. In this connection, the Working Group may wish to:

(i) Endorse and recommend to the Statistical Commission the organization of an expert group on poverty, with a planning meeting to be convened by ECLAC in 1996 or 1997, as recommended by the Expert Group in paragraphs 45-47 of its report;

(ii) Consider the Expert Group proposal in paragraph 54 of its report to establish an informal group on income, consumption and expenditure statistics;

(d) Request the Statistics Division, in cooperation with ECE, to report to the Statistical Commission at its next session on their suggestions for the nature and timing of an international report on social progress, initially as a pilot report, as recommended in paragraphs 111-114 of the Expert Group report;

(e) Endorse and recommend to the Statistical Commission the timetable suggested by the Expert Group, as contained in paragraph 115 of its report, subject to the availability of appropriate resources;

(f) Request the Statistics Division to submit to it at its nineteenth session a consolidated report on the above recommendations, accompanied by the report of the Expert Group.

1/ Official Records of the Economic and Social Council, 1995, Supplement No. 8 (E/1995/28), paras. 65-70.

Paragraphs Page

I. INTRODUCTION ........................................... 1 - 16 6

A. Establishment of the Expert Group ................ 1 - 3 6

B. The vital role of social statistics ..................... 4 - 7 6

C. Broad areas of social concern ......................... 8 - 10 7

D. The Expert Group proposal ........................... 11 - 16 8

II. POPULATION AND DEVELOPMENT ...................................... 17 - 30 9

A. Statistical indicators ........................... 18 - 22 9

B. Statistical and measurement issues ............... 23 - 28 10

C. Further work ..................................... 29 - 30 11

III. ERADICATION OF POVERTY ............................... 31 - 54 11

A. Statistical indicators ........................... 35 - 40 12

B. Statistical and measurement issues ............... 41 - 44 14

C. Further work ..................................... 45 - 54 15

IV. EXPANSION OF PRODUCTIVE EMPLOYMENT AND REDUCTION OF

UNEMPLOYMENT ......................................... 55 - 66 16

A. Statistical indicators ........................... 58 - 59 17

B. Statistical and measurement issues ............... 60 - 62 18

C. Further work ..................................... 63 - 66 18

V. SOCIAL INTEGRATION ................................... 67 - 79 19

A. Statistical indicators ........................... 70 - 73 20

B. Statistical and measurement issues ............... 74 - 76 21

C. Further work ..................................... 77 - 79 21

Paragraphs Page

VI. STATUS OF WOMEN AND MEN .............................. 80 - 87 22

A. Statistical indicators ........................... 81 - 82 22

B. Statistical and measurement issues ............... 83 - 85 23

C. Further work ..................................... 86 - 87 24

VII. STATISTICAL RESPONSE TO THE FIVE POLICY THEMES OF

RECENT MAJOR UNITED NATIONS CONFERENCES .............. 88 - 115 24

A. A minimum national social data set ............... 89 - 94 24

B. Suggested composition of the minimum national

social data set .................................. 95 - 99 25

C. Assessing national social statistics capacity .... 100 - 106 26

D. Building national social statistics capacity ..... 107 - 110 27

E. International reporting on social progress ....... 111 - 114 28

F. Suggested timetable .............................. 115 29

Appendix. Menu of indicators .......................................... 31

1. Since its inception, the United Nations has promoted social and economic development and world peace. In the last three years, Governments have taken a number of new initiatives to address social issues at the head-of-State level, beginning with the World Summit for Children. In the last 24 months, three major conferences on social issues have been organized by the United Nations: the International Conference on Population and Development (ICPD), which discussed population and welfare issues; the World Summit for Social Development, which covered the issues of poverty, employment and social integration; and the Fourth World Conference on Women, which concentrated on the status of women. Many themes were raised at all three conferences.

2. The agreed outcomes of the above-mentioned conferences were documented in their programmes of action, which describe issues of current world concern and propose solutions, and mention the need for adequate information to inform social policy development and monitoring. Common requirements for social statistics include: a/

(a) Improved reliability and validity;

(b) Greater disaggregation, particularly by gender;

(c) Improved international comparability;

(d) Wider dissemination and use.

3. At the twenty-eighth session of the Statistical Commission (27 February-3 March 1995), the Expert Group on the Statistical Implications of Major United Nations Conferences was established to:

(a) Consider the programmes of action adopted by ICPD, the World Summit for Social Development and the Fourth World Conference on Women;

(b) Agree on a number of critical policy domains;

(c) Identify relevant statistical issues arising from such policy domains;

(d) Propose a work programme for presentation to the Working Group at its eighteenth session.

4. A sound system of social statistics is vital to the effective development of social policy, to informed decision-making on policy issues, and to the evaluation of the impact of social and economic policy. For many countries, inadequate systems of social statistics constitute a major impediment to effective social development. Insufficient awareness of the importance of the linkages between policy development and social statistics, the need for more internationally agreed statistical standards and guidelines, and the need for improved frameworks in which to summarize policy outcomes all point to the need for greater national and international priority to be given to social statistics.

5. To acquire the necessary capability to monitor country progress in achieving the outcomes of the policy themes of major United Nations conferences, a good understanding of the linkage between sound social data systems and effective social policy development is a prerequisite. This calls for partnership between national statistical offices (NSOs) and policy makers to ensure that statistical objectives and priorities are focused on providing the data foundations necessary for effective social policy development.

6. It should be noted that since the 1940s an understanding of the linkages between economic statistics and economic policy issues has guided statistical efforts in such areas as labour, prices, national accounts, government finance and balance of payments, to the benefit of economic policy monitoring and development.

7. The international community (NSOs, international statistical bodies and funding agencies) needs to accept responsibility for supporting the development of social statistics in each country as an urgent priority in the programme to implement the recommendations of the World Summit for Social Development.

8. The Expert Group has identified the following broad areas of social concern arising from the five policy themes of ICPD, the World Summit for Social Development and the Fourth World Conference on Women:

Policy themes Main areas of social concern

Population and development Health

Material well-being

Education

Eradication of poverty Income and expenditure

Economic resources

Expansion of productive employment Work

and reduction of unemployment Working environment

Education and training

Social integration Housing

Work

Crime and criminal justice

Status of women and men Health

Education

Work

Income

9. The policy themes are themselves strongly interconnected. For example, the expansion of productive employment is a key issue affecting the status of women and men and the incidence of poverty; changes in population composition affect the demand and supply of labour; and gender issues are key determinants of population growth and development.

10. Ideally, a framework for social statistics should be developed by considering how these broad areas of social concern affect particular population groups, such as women, retired people, children and single parents. Such a framework, however, does not exist, and its articulation is unlikely in the short to medium term. Nor has the Expert Group attempted this complex task. Rather, focusing on short and medium-term action priorities, the Expert Group believes it would be feasible to identify a minimum set of statistics and indicators that would assist countries in assessing their social position and progress.

11. The Programme of Action of the World Summit for Social Development states that:

"The United Nations system's capacity for gathering and analysing information and developing indicators of social development should be strengthened, taking into account the work carried out by different countries, in particular by developing countries. The capacity of the United Nations system for providing policy and technical support and advice, upon request, in order to improve national capacities in this regard should also be strengthened". (para. 99 (e))

12. Accordingly, the Expert Group has focused on national issues at its first priority but has borne in mind the requirements for international monitoring and reporting.

13. The present report identifies a number of indicators that may be used to monitor or assess development, and discusses potential sources of data. In all the five policy theme areas, the suggestions of the Expert Group, albeit far from comprehensive, are numerous. In some instances, much work will be needed both to define and to win acceptance for a specific indicator, and still more work to actually collect and compile the data. The appendix contains a further selection, a basic list of 15 indicators making up a minimum national social data set (MNSDS).

14. The Expert Group proposes that the MNSDS be developed to provide key statistics for national and international reporting and monitoring. It is envisaged that the MNSDS would consist of regularly available statistics disaggregated by gender and broad age groups. In addition, the MNSDS must be of high integrity and reliability and must be reasonably timely.

15. The last point should be emphasized. For a great number of countries, MNSDS data are available only from censuses, which with several quinquennial exceptions are generally conducted every 10 years. There is thus a major problem for many countries in finding or creating mechanisms that will deliver the MNSDS on a reasonably timely basis.

16. The Expert Group recognizes that the dichotomy between developing and developed countries implied in the report is an oversimplification and that the indicators proposed in consequence also suffer from oversimplification. While it would be impossible to capture the full spectrum of development variations across regions, it is hoped that the proposed regional reviews (see paras. 100-106 below) will provide at least some of the required adaptation.

17. Three key objectives of ICPD are:

(a) Reduction of mortality rates, specifically infant, child and maternal mortality;

(b) Universal access to and use of high-quality reproductive health care, including family planning and sexual rights;

(c) Universal access to basic education, especially for girls and women.

18. To assess the current situation of countries relative to the goals of the ICPD Programme of Action and to monitor their progress and achievements, three sets of internationally accepted indicators on mortality, reproductive health and basic education should be used.

Mortality rates

19. The Programme of Action clearly specifies three mortality indicators that have universally accepted definitions: infant, child and maternal mortality.

Reproductive health

20. The international community has defined a number of indicators relevant to reproductive health, including:

(a) Percentage of pregnant women who have at least one ante-natal visit;

(b) Percentage of pregnant women who have a trained attendant at delivery;

(c) Percentage of pregnant women immunized against tetanus;

(d) Percentage of contraceptive prevalence rate;

(e) Percentage of infants weighing less than 2,500 grams (g) at birth.

21. The World Health Organization (WHO) is working on additional indicators for the global monitoring of reproductive health, including indicators on the incidence and prevalence of sexually transmitted diseases, the quality of family planning services, access to and the quality of maternal health services, and the incidence of female genital mutilation.

Education, particularly basic education

22. There is to date no agreed short list of educational indicators. However, the integrated programme of statistical work of the ECE Conference of European Statisticians has proposed increasing the emphasis on measures of educational outcomes, education sector activities and associated expenditures, teaching and other personnel, and socio-economic and other correlates/determinants of education outcomes.

23. The best source for mortality data and a number of other indicators is a properly functioning civil registration system. Few developing countries, however, have such a system; if they have, it is not complete. To build up such systems is a long and painstaking process and although that long-term goal should be supported, it will not yield results in time for the present purpose. Instead, the necessary data will frequently have to be derived from other administrative records or from surveys.

24. Due to measurement problems, such as misreporting or small samples, many mortality estimates at the national level are not accurate reflections of the present situation.

25. In spite of an accepted definition of reproductive health agreed at ICPD, there is no universally accepted set of components of what interventions contribute to reproductive health; hence, few statistics are available.

26. There exists a large amount of statistical information on literacy, as well as on the availability and use of educational facilities. In many developing countries, such data suffer from the same defects as the mortality data mentioned above. Again, as indicated earlier for reproductive health, there are no standard definitions or data for measuring access to education or educational facilities; furthermore, the concept of basic education needs to be more clearly defined. In addition, there are no universally accepted indicators for measuring the degree of discrimination against female participation in the educational system. Such indicators could be developed but will require international acceptance.

27. Measures of outcomes are very important for assessing the impact and effectiveness of social policies and programmes. There is also a need to recognize that process indicators are useful for effective programme management at the national and subnational levels.

28. For example, for the reduction of maternal mortality proposed indicators are based on the assumption that the most effective strategy is to increase access to (a) family planning services for raising the contraceptive prevalence rate and (b) prompt, adequate and accessible obstetric care. The indicators therefore measure progress in improving access to, utilization of and the quality of family planning and emergency obstetric care services. Using process indicators will help programme planners to identify priority interventions and areas, as well as aspects of the programme that need strengthening.

29. The United Nations Population Fund (UNFPA), as the lead agency for the ICPD Programme of Action, has identified several areas needing further work. Some pertain to research, such as the interrelationship between population and poverty, the impact of social and economic decisions and actions on the well-being of families, the status of women within families, and the interrelationships between fertility and mortality levels.

30. Other important areas include gender disaggregated data illustrating the demographic and development impact of human immunodeficiency virus (HIV) and acquired immunodeficiency syndrome (AIDS), flows and stocks of international migrants, and general data to strengthen demographic analysis.

31. Commitment 2 of the Copenhagen Declaration on Social Development states that:

"We commit ourselves to the goal of eradicating poverty in the world, through decisive national actions and international cooperation, as an ethical, social, political and economic imperative of humankind".

32. Commitment 2 calls for policies and strategies to reduce overall poverty, reduce inequalities and eradicate absolute poverty. In view of the multidimensional nature of poverty - such as lack of income and productive resources, hunger, limited access to education, employment and housing, and lack of social participation - absolute poverty depends not only on income but on access to a wide range of social services and facilities. The need for development of criteria, indicators and methods to measure all forms of poverty, especially absolute poverty, is seen as an urgent one, leading to the recommendation continued in the Programme of Action of the World Summit for Social Development to reduce overall poverty as an urgent one, by:

"Elaborating at the national level, the measurements, criteria and indicators for determining the extent and distribution of absolute poverty. Each country should develop a precise definition and assessment of absolute poverty, preferably by 1996, the International Year for the Eradication of Poverty". (para. 26 (d))

33. The meaning of poverty is dependent on the customs, standards and values of individual countries. There are many countries in which people living in poverty do not have sufficient resources to secure minimum levels of food, clothing and shelter. In most industrial countries absolute poverty is almost non-existent and the concept of poverty is a relative one, dependent on social participation relative to the average, for members of society. In terms of income, relative poverty is measured with reference to the average level of income in the nation as a whole.

34. The distinction between absolute and relative poverty, though essential, does not necessarily mean that the suggested indicators can be neatly classified as being only relevant to conditions of either absolute poverty or relative poverty. Indeed, it could be argued that certain indicators, such as families below a minimum standard of living (poverty line) and "poverty gap", which are in the present report being included among the indicators of relative poverty, really belong with the indicators of absolute poverty.

35. Absolute poverty is normally defined as an acute shortfall of income or assets, as well as by the lack of access to certain basic services. In developing countries, and especially in those with the lowest per capita income, such deficiencies are usually defined in absolute terms in relation to the thresholds of basic needs. Statistical indicators should be selected keeping in mind the need to formulate policies for combating poverty. In the case of absolute poverty, indicators could refer to:

(a) Percentage of the population in poverty (poverty or poverty line defined nationally);

(b) Possibility of entering the labour force;

(c) Money income;

(d) Monetary value of the basket of food needed for minimum nutritional requirements;

(e) Food prices;

(f) Access to productive assets, especially land and water;

(g) Geographic location;

(h) Public transfers;

(i) Access to public or private health services;

(j) Housing;

(k) Educational services.

Relative poverty

36. Within industrialized nations, the problems of measuring income inequality and poverty can be approached in two main ways. The first approach takes a broad view of economic and social circumstances that measures social participation relative to the average for members of the society. This approach reflects the multidimensional nature of poverty: it requires a selection of key indicators, such as indicators on health and access to health care, employment and working conditions, economic resources, educational resources, housing and neighbourhood facilities, family and social integration, recreational and cultural opportunities, and political resources. The measurements, criteria and indicators for determining the extent and distribution of poverty using this approach depend on quantifying and assessing the levels and equality of access to such resources and services.

37. The second approach views poverty as defined by the (non-)capability to access a minimum level of material resources, and is generally measures indirectly through access to either income or expenditure.

38. To identify the numbers of people living in poverty requires a set of indicators that permits the whole population to be ranked on a comparable basis and a criterion that separates the poor from the rest of the population in terms of such indicators; the poverty line is the clear dividing line.

39. Alternative definitions of income, units of analysis and equivalence scales will result in different measures of the population in relative poverty, and the population in poverty will have different characteristics in such areas as family composition and labour force status.

40. Possible indicators of relative poverty and income inequality include:

(a) Families below a minimum standard of income (poverty line);

(b) Poverty gap;

(c) Families with less than a specified percentage of mean income;

(d) Gini coefficients (summary measure of inequality in income distribution);

(e) Income share of lowest income quintile;

(f) Income share of highest income quintile;

(g) Percentiles of median income.

Data sources

41. Data may not cover the whole population adequately. For example, people living in remote rural areas or in areas in rapidly growing cities not delineated in available cartography, the homeless and other marginal groups, such as the institutionalized population and guest workers, might be excluded from the scope of data sources. Even if included, the quality of information obtained from particular groups, such as low-income recipients or the self-employed might be poor due to differential non-response or to high and/or differential levels of respondent error. The main data sources are:

(a) Population and housing censuses;

(b) Household income and expenditure surveys;

(c) Labour force surveys;

(d) Standard of living surveys;

(e) Administrative records;

(f) National accounts.

Income concept

42. In addition to monetary earnings, income can include the monetary value of government services, such as education, health, housing, and welfare received by households, non-cash employment benefits, the value of home production, including subsistence farming, and the imputed income accruing from home ownership and other household durables. For most developing countries, it is important to measure home production, including fetched firewood.

Unit of analysis

43. The definition of the income unit to reflect the pooling of income as well as economic and legal dependency has a substantial effect on distribution measures. Income is more evenly distributed across households than individuals. In many developing countries, inter-household transfers have the same role as government transfers in developing countries.

Population groups

44. The strong interrelationship between population groups in poverty and certain other characteristics can be made apparent only if the statistical data sets are there to support it, which requires that the statistical framework specifically incorporate a population group dimension, such as women, unemployed persons, or single parents. The interconnections between poverty and income distribution, on the one hand, and population groups on the other, pose an additional conceptual requirement for a system of social statistics designed to measure progress in the eradication of poverty.

Absolute poverty

45. Keeping in mind the needs of the developing countries, and to assist countries in developing their own definition of poverty, the Expert Group considers that the existing Task Force on the Measurement of Poverty is not an appropriate mechanism because it consists of representatives of international agencies only. Instead, an Expert Group on Poverty, serving as a standing committee and meeting on and ad hoc basis, is suggested, comprising country experts from each United Nations region and appropriate experts from international bodies and the regional commissions. It is suggested that the group be informal, and that it focus on conceptual work that has practical implications for poverty measurement in all regions.

46. The mandate of such a group would be to gather from each regional commission the concepts of poverty applying in its region, and to aim to identify preferred measurement concepts and indicators. The group would also advise on the statistical instruments required for collecting poverty indicators, such as a minimum set of technical guidelines or a technical manual that would include the description of reliable techniques for poverty measurement and solutions to the routine problems faced in such exercises.

47. As a first step and to assess the need for and feasibility of such an expert group, a special conference on poverty definition and a simple technical manual for use at the national and/or regional level is proposed for late 1996 or early 1997. Because of the work done on poverty definition in the ECLAC region, Santiago is suggested as the venue. Assistance would be required for some country experts to attend.

Relative poverty

48. The strategies of the World Summit for Social Development call for the development of agreed gender disaggregated indicators of poverty and vulnerability, including indicators of income and wealth. They also call for strengthening international data collection and statistical systems to support countries in monitoring social development goals, which requires the development of international guidelines on income distribution and related statistics.

49. International standards should be capable of supporting a wide range of applications for income and income distribution statistics, and for poverty research. In addition to intercountry comparison at a given point in time, statistics should also support comparisons over time. International standards should, at a minimum, cover data sources, the definition of income, the definition of income units, use of equivalence scales, and indicators of poverty and income inequality.

50. Other areas that need attention include procedures for standardizing for different institutional arrangements; standardizing for contributions to social insurance, particularly in the area of employee contributions; and the interaction between social security transfers, taxation and redistribution over the life cycle.

51. A range of international activity is under way on income statistics standards and other work to improve international comparability. For example, with the agreement of OECD and ECE, Eurostat has convened an Advisory Income Steering Group (AISG) to revise the 1977 United Nations provisional guidelines on statistics of the distribution of income, consumption and accumulation of households. Income distribution will provide their focus but the guidelines will also touch on poverty and social exclusion.

52. At its first meeting in June 1995, the Eurostat AISG agreed that the revision of concepts definitions and classifications should be carried out at the world level, but that data sources would vary "enormously" throughout the world. Further, AISG recognized that to build world level guidelines and standards at this stage would be overambitious; the guidelines should therefore be restricted to the ECE/OECD region but should in turn be an input for later work to build more global standards.

53. The revision will address the concepts of disposable and final income, expenditure, the informal sector, the income-receiving unit and equivalence scales, paid and unpaid work etc. The work will be discussed at the August 1996 meeting of the International Association for the Review of Income and Wealth (IARIW) in Lillehammer, Norway. A draft set of revised guidelines is envisaged for consideration by a joint Eurostat/ECE/OECD meeting on household income statistics to be held in late 1996/early 1997.

54. The Expert Group welcomes the Eurostat AISG initiative but notes that the proposed timing is restrictive. To permit the proposed guidelines to be fully debated before they are put forward for adoption, the Expert Group therefore considered the need for further broad-based discussion among experts following the Lillehammer meeting before the revised guidelines are finalized. The Expert Group suggests that the Lillehammer participants consider the value of a continuing informal group for debating and developing income, consumption and expenditure issues, including the draft revised guidelines; AISG members would form a valuable nucleus for such a group.

IV. EXPANSION OF PRODUCTIVE EMPLOYMENT AND

REDUCTION OF UNEMPLOYMENT

55. Commitment 3 of the Copenhagen Declaration states that:

"We commit ourselves to promoting the goal of full employment as a basic priority of our economic and social policies, and to enabling all men and women to attain secure and sustainable livelihoods through freely chosen productive employment and work".

56. The Programme of Action of the World Summit for Social Development identifies a need for:

"Strengthening labour market information systems, particularly the development of appropriate data and indicators on employment, unemployment, underemployment and earnings, as well as dissemination of information concerning labour markets, including, as far as possible, work situations outside formal markets". (para. 53 (i))

57. The Programme of Action of the Fourth World Conference on Women urges relevant bodies to:

"Develop a more comprehensive knowledge of all forms of work and employment by:

(i) Improving data collection on the unremunerated work which is already included in the United Nations System of National Accounts ...;

(ii) Improving measurements that at present underestimate women's unemployment and underemployment in the labour market;

(iii) Developing methods ... for assessing the value, in quantitative terms, of unremunerated work that is outside national accounts ...". (para. 206 (f))

58. Almost all labour data are relevant to social policy development and evaluation. Earnings from employment are the principal determinant of the economic welfare of individuals and families.

59. It is envisaged that the next International Labour Organization (ILO) World Employment Report will contain a statistical annex displaying the values of a set of statistical indicators on employment performance for as many countries as possible. Ten indicators are currently being considered, each designed to assess one aspect of employment performance requiring readily available data for most countries. The 10 indicators are:

(a) Gross domestic product (GDP) per capita at purchasing power parity;

(b) Unemployment rate, by sex where possible;

(c) Employment-population ratio, defined by total employment, including self-employment, by sex where possible, as a percentage of the corresponding population aged 16 to 64;

(d) Wage employment as a percentage of the population aged 16 to 64, for males and females separately where possible;

(e) Formal sector employment as a percentage of total employment;

(f) Median and average length of job tenure in years, for males and females separately where possible;

(g) Index of real wages in manufacturing and in the economy as a whole, where possible;

(h) Ratio of average wage in the formal sector to GDP (or total wage bill) per person employed in the economy;

(i) Wage dispersion in manufacturing industries, measured by the coefficient of variation, for males and females separately where possible;

(j) Ratio of average female to average male wage in manufacturing and in the economy as a whole, where possible.

60. Labour statistics are often the best developed and the best funded of national statistics. Data are collected from:

(a) Population censuses;

(b) Household surveys, particularly labour force surveys;

(c) Registers;

(d) Establishment surveys.

61. However, there are often difficulties in considering or comparing data from the various sources, and much work remains to be done to achieve both national and international comparability. In addition, special data collection systems are needed to cover the informal sector and unpaid work outside the market economy.

62. The provision of social policy relevant data for small geographic areas or population subgroups of special interest is frequently poor. With the increasing emphasis in social policy development on targeting particular geographic areas or population subgroups, and with the devolution of social policy development (or programme delivery) to subnational areas, the absence of small area data can be problematic.

63. The basic indicators listed above fall far short of portraying the complex status and dynamics of labour markets, especially in developed economies. For example, there is a vast imbalance between data describing the supply side of the labour market and data illuminating the demand side. The lack of all but the most rudimentary demand-side data has resulted in the entire job creation process remaining poorly understood. There is little knowledge about the process that translates changes in the demand for goods and services into changes in either employment levels or the kinds of jobs created or eliminated.

64. In addition there is a lack of data on non-wage compensation (fringe benefits). There is also a need for information on underemployment, but its development will require substantial further effort, building on work currently under way by the ILO in preparation for the next International Conference of Labour Statisticians, to be held in 1988. Expanding the boundaries of labour statistics to encompass unremunerated work and employment, as recommended by the Fourth World Conference on Women, will also require substantial development work, in both concepts and techniques of measurement.

65. Given its mandate, tripartite structure and expertise, the ILO was assigned a special role by the World Summit for Social Development. In performing this role, it is suggested that the ILO insist on the global nature of the employment problem (i.e., the links between the employment problems of countries at all levels of development in a global economy) and on a broad concept of employment, covering not only job creation but also workers' rights and quality of employment, and including self-employment and the frequently very precarious forms of employment in the informal and rural sectors.

66. The Expert Group endorses the current efforts of the ECE Conference of European Statisticians, the ILO, Eurostat, OECD and several NSOs to collaborate in the identification of issues and priorities for the further development of labour statistics in developed countries, and recognizes the unique role of the ILO in articulating standards and objectives for both developed and developing countries.

67. Commitment 4 of the Copenhagen Declaration states that:

"We commit ourselves to promoting social integration by fostering societies that are stable, safe and just and based on the promotion and protection of all human rights, and on non-discrimination, tolerance, respect for diversity, equality of opportunity, solidarity, security and participation of all people, including disadvantaged and vulnerable groups and persons."

68. There are no absolute standards of social integration and no internationally recognized standards against which to measure the disadvantage suffered by particular groups. Similarly, there are no universally accepted ways of defining which social groups merit monitoring in all countries; each country must determine for itself which groups within its society are potentially disadvantaged and vulnerable.

69. In doing so, countries should take account of the following factors:

(a) Disadvantaged and vulnerable groups may not necessarily be small in number: they could even form the majority of the population;

(b) Some of the most common factors leading to social disadvantage are gender, race, creed, and physical or mental impairment. Groups defined in such terms tend to change only slowly. Other groups may be disadvantaged as a result of social factors, such as homelessness or drug dependency;

(c) Some of the most disadvantaged may not feature at all in data sources;

(d) There are no value-free ways of assessing lack of social integration or progress in social integration. The World Summit of Social Development, however, provides a basis for addressing the issue, and provides a clear remit for statisticians to monitor the existence and conditions of disadvantaged and vulnerable groups and persons.

70. It is not possible to define a single set of country indicators that can be used to monitor social integration. The choice of indicators will depend on the nature of the vulnerable groups identified, which will differ between countries.

71. The sort of indicators that are likely to be required for each group are:

(a) Number of people in the group;

(b) Age/gender structure;

(c) Occupational profile;

(d) Duration of unemployment;

(e) Economic activity profile;

(f) Income levels;

(g) Position within overall income distribution;

(h) Housing standards/amenities, such as access to safe water, sanitation and floor space per person;

(i) People living alone and with little contact with close family;

(j) Health status, such as infant mortality rate, age-specific mortality rates, expectation of life and nutritional intake;

(k) Educational standards, such as adult literacy rate, number of years of formal education and participation rates (for children);

(l) Crime victimization rate;

(m) Participation in volunteer work, sport, cultural activities and travel;

(n) Proportion eligible to vote.

72. These quantitative indicators may not always fully describe the situation of disadvantaged groups; part of their disadvantage, for example, may be a lack of religious freedom. Such indicators may therefore have to be supplemented with indicators of subjective well-being.

73. The aim is to monitor progress in social integration, that is, the convergence between the situation of the identified groups with the rest of the population within the country. Therefore, indicators should be presented in time-series form with data for the rest of the population shown alongside for comparison. Countries may also find it helpful to compare both their identified population groups and the associated indicators with those of other countries within their region.

74. One of the most difficult issues in this area is the question of definition. Definitions should be sought that are internationally applicable and capable of identifying the majority as well as minority population groups within the country, as well as groups that are subject to hidden disadvantages or who face legal barriers to their full participation in society.

75. A second set of issues concern data collection. Some disadvantaged and vulnerable groups may be well covered and identifiable in conventional data sources, particularly groups defined by gender and age. Other disadvantaged and vulnerable groups may be too small to feature significantly in conventional data sources, or there may be a reluctance to ask or to provide information that would identify persons as belonging to such groups.

76. Conventional data sources may also fail to identify certain disadvantaged and vulnerable groups and persons because of the absence of such groups from the sampling frames or their reluctance to participate in surveys. For example, people without shelter will be excluded from any sampling frame based on housing, and there are likely to be difficulties in gaining sufficient trust from groups subject to discrimination or victimization. Special techniques may have to be developed either to ensure that such groups are included in data collection that purports to be national in coverage or to target them for special surveys.

77. First, statistical offices should explicitly list the groups within their country that are considered vulnerable or disadvantaged, notwithstanding all the difficulties already enumerated that they may face in so doing. These are the groups that the World Summit for Social Development lays a duty on statistical services to monitor, and obviously monitoring can only be carried out if the groups are sufficiently well-defined to be identified within statistical collection systems.

78. The nature of the groups identified, in particular the causes of their vulnerability or disadvantage, will inform the choice of indicators for monitoring progress in social integration.

79. In most cases, long-term statistical programmes will be required to monitor changes in conditions over time. It is recommended that such statistical programmes be clearly identified in the planning programmes of national statistical offices, and that such programmes include details of the groups to be monitored, planned data collection and dissemination arrangements.

80. The Beijing Platform for Action adopted at the Fourth World Conference on Women concentrates on overcoming key obstacles to reaching equality between women and men in all parts of the world. Twelve critical areas of concern are identified in paragraph 44 of the Platform:

(a) The persistent and increasing burden of poverty on women;

(b) Inequalities and inadequacies in and unequal access to education and training;

(c) Inequalities and inadequacies in and unequal access to health care and related services;

(d) Violence against women;

(e) The effects of armed or other kinds of conflict on women, including those living under foreign occupation;

(f) Inequality in economic structures and policies, in all forms of productive activities and in access to resources;

(g) Inequality between women and men in the sharing of power and decision-making at all levels;

(h) Insufficient mechanisms at all levels to promote the advancement of women;

(i) Lack of respect for and inadequate promotion and protection of the human rights of women;

(j) Stereotyping of women and inequality in women's access to and participation in all communication systems, especially in the media;

(k) Gender inequalities in the management of natural resources and in the safeguarding of the environment;

(l) Persistent discrimination against, and violation of, the rights of the girl child.

81. The goal of work with gender statistics is to ensure that statistics related to individuals are collected, compiled, analysed and presented by sex and age, and reflect issues related to women and men in society.

82. This implies that gender statistics cannot be produced and improved in isolation. A gender perspective is needed in all traditional statistical fields. Such work must be integrated into the development of the overall national statistical system, a process that is termed "mainstreaming". Improvement of content, methods, classifications and measurements should be made part of the ongoing work to improve statistical sources, namely, censuses, surveys and administrative systems.

Improve measurements, concepts, definitions and classifications

83. Traditional concepts and definitions do not always adequately reflect the situation of women and men in society, and can introduce gender-based biases. Many topics of special relevance to gender issues in society are also difficult to quantify and measure.

Improve production of statistics to fill in data gaps

84. In some areas of concern, gender statistics are particularly scarce, due to problems with concepts and definitions, difficulties in data collection and a lack of communication between users and producers of statistics. In some cases, statistics are not collected at all; in others, they are collected but not by sex (e.g., individual incomes are not registered as separate parts of household income).

Improve ways to present and disseminate gender statistics

85. Gender statistics should be presented on a regular basis both for utilization in policy and programme planning and implementation. They should also be presented in a form suitable for a wide range of non-technical users. A regular overview of the situation of women and men should be a first step. A basic set of social and economic statistics and indicators should cover information on:

(a) Population and households;

(b) Health;

(c) Education;

(d) Time-use;

(e) Child and elderly care;

(f) Gainful employment;

(g) Wage, salary and income;

(h) Violence and crime;

(i) Influence and power.

International statistical cooperation

86. The basic work to improve gender statistics must be done at the national level. However, cooperation between countries is necessary to permit international comparisons and to make effective use of scarce resources to develop new standards and methods. The Expert Group suggests that the Statistical Commission and its working groups can provide support in various ways to develop relevant concepts, definitions and measurement methods for the collection and presentation of statistics. Two important areas are the development of methods for time-use surveys and satellite accounts.

Guidelines and manuals

87. Guidelines and manuals should be developed on how to produce, present and disseminate gender statistics, including a basic set of statistics and indicators on gender issues. Development of such indicators should be integrated with the work on social and economic statistics and indicators identified in other priority areas.

VII. STATISTICAL RESPONSE TO THE FIVE POLICY THEMES

OF RECENT MAJOR UNITED NATIONS CONFERENCES

88. Each of the five policy themes identified in paragraph 8 above indicates the need for statistical action. The first priority is to identify desirable action at the national level and determine how it can be initiated quickly, after which the question of gathering information from each country to assess international progress in each of the five areas can be discussed.

89. The Programme of Action of the World Summit for Social Development appreciates the need to focus on national requirements, both in building general capacity in social statistics and in the particular matter of defining a poverty indicator. The Expert Group review of the five policy themes supports the national focus.

90. The Expert Group also recognizes the benefits of ensuring as much consistency as possible between countries. To the extent that data items, supported by the same definitions and classifications, are common across countries, less duplication will be involved in developing national social statistics capacity and the opportunity for future international comparison at regional or world level will improve.

91. The Expert Group also considers that priority should be given to ensuring that each country has the capacity to produce a small number of indicators crucial to the five policy themes in order to guide policy development and decision-making in the country itself.

92. Given these conclusions, the Expert Group proposes that each country compile an MNSDS. Each country should be encouraged and, where necessary, provided with assistance in producing the MNSDS data items as soon as feasible.

93. It is important that the MNSDS be seen as a suggested minimum data set; its adoption should not become a precondition for assistance in improving social statistics in a particular country. National circumstances and priorities differ, which must be accepted and recognized as a potential source of differing statistical priorities. The MNSDS can be viewed as a menu from which countries can select data items of highest national priority. But the suggested MNSDS has been kept as small and as basic as possible to improve its chances of adoption by as many countries as possible.

94. Many countries in all parts of the world will be able to advance or have already advanced well beyond the MNSDS; the Expert Group welcomes this and encourages such countries to develop further social indicators. Such indicators are identified by the Expert Group above and are suggested in the appendix including indicators on social integration, such as crime victimization, and on the respective roles of men and women, such as time-use. Work on the demand for labour and the nature of the labour market will also merit further attention, even in countries with well developed traditional labour-supply data.

B. Suggested composition of the minimum national

social data set

95. The suggested list of indicators for the MNSDS was compiled by the Expert Group on the basis of the following criteria:

(a) Direct relevance to the five policy themes identified in paragraph 8 above;

(b) Accepted international definition and classification;

(c) A collection instrument that is feasible in most countries;

(d) Feasibility of gender disaggregation.

96. Because the World Summit for Social Development recognized that each country should define its own poverty indicators, the suggested MNSDS does not include any specific poverty indicators. Nevertheless, many of the indicators included are relevant in assessing the extent of disadvantage in key areas of participation in the economy and society. Thus, achievement of the MNSDS in a country is likely to be an important step in assessing the extent of poverty and progress in its alleviation.

97. The 15 items suggested for the MNSDS (see also appendix) are:

(a) Population estimates by sex, age and, where appropriate and feasible, ethnic group;

(b) Life expectancy at birth, by sex;

(c) Infant mortality, by sex;

(d) Child mortality, by sex;

(e) Maternal mortality;

(f) Percentage of infants weighing less than 2,500 g at birth, by sex;

(g) Average number of years of schooling completed, by sex, and where possible by income class;

(h) GDP per capita;

(i) Household income per capita (level and distribution);

(j) Monetary value of the basket of food needed for minimum nutritional requirements;

(k) Unemployment rate, by sex;

(l) Employment-population ratio, by sex, and by formal and informal sector where appropriate;

(m) Access to safe water;

(n) Access to sanitation;

(o) Number of people per room, excluding kitchen and bathroom.

98. All the items included in the list above should be produced and presented disaggregated by urban/rural areas, where the rural population is greater than about 25 per cent of the total population.

99. The suggested MNSDS does not include process indicators, despite their importance at the national and subnational levels, as discussed above. This reflects the lack of internationally accepted definitions and the absence of collection mechanisms in many countries; their exclusion form the MNSDS does not diminish the value of collecting such information at the subnational and programme levels.

100. The Expert Group proposes a two-stage process for improving national social statistics capacity. First, the regional commissions should be requested to gather information from each country on the state of its social statistics capacity. Each region should adapt the MNSDS to allow for any known common features of the region. The MNSDS (as so adapted) should be described to each country and feedback sought on whether the items are seen as appropriate for that country and whether the MNSDS data could be compiled. Subject to those comments, the MNSDS should serve as a benchmark for establishing whether the country has the minimum social statistics infrastructure and, if there are gaps, what contributions are needed to fill them.

101. The questionnaires used by each region should be based on a common model, given the general coverage of the minimum data set. Regional variations may be required, however, if the MNSDS has been modified for the region, if some of the information is already held by the regional commission, or if additional information is sought. The Expert Group suggests that the United Nations Statistics Division draft a model questionnaire.

102. The regional commissions should collate the responses, both to identify what capacity and data are available and to assess assistance needs for capacity-building, production, presentation and dissemination of MNSDS data.

103. The regional commissions will require some funding to enable them to gather the information on capacity and assistance needs from countries in their region. In addition, resources should be allocated for questionnaire design and coordination by the Statistics Division.

104. This should not require excessive resources, and in the ECE region, the existing UNDP project should enable the information to be readily obtained from the countries in transition and the countries of the former USSR. Very tentatively, a project of $2 million would provide $400,000 for each region other than the ECE, plus $200,000 each for the United Nations Statistics Division and the ECE Statistical Division.

105. The Expert Group urges that UNDP and other donors make special efforts to release funds form the 1996 programmes and give this initial information exchange project the highest priority in their plans to follow up the World Summit for Social Development. This project is essential if other funders are to be drawn in to assist in improving national social statistics capacity in an organized and efficient fashion.

106. Regrettably, it will hardly be possible for the regional reports to be collated for consideration by the Statistical Commission in early 1997. Such a report should be prepared by the United Nations Statistics Division for the autumn 1997/spring 1998 meeting of the Working Group.

107. The aim would be to give each national statistics office the capacity to produce and disseminate MNSDS data on a regular basis as soon as possible, and then to maintain the momentum to improve capacity as gaps are identified.

108. Even before the collation of needs is completed, the Expert Group urges funders interested in assisting the implementation of the Copenhagen Declaration and Programme of Action of the World Summit for Social Development to contribute, preferably to a fund available at world level but also at a regional level or on a bilateral or subregional basis in order to allow an immediate start on social statistics capacity-building.

109. Once the country responses to the MNSDS proposals are collated and the overall assistance required has been assessed, an international agency or agencies will need to provide a pool of funds to improve national social statistics capacity. Priority should be given to countries with the greatest assessed needs. UNDP is the best placed agency to lead this funding effort, but other international, regional and country funders should be drawn in.

110. The regional commissions should be given the role and resources required to coordinate this capacity-building among the countries of their regions. The UNDP/ECE project provides a model for the assistance that can be provided by a regional commission:

(a) Assistance with the 2000 census round;

(b) Improving use of administrative data;

(c) Improving capacity to conduct household surveys;

(d) Improving information technology infrastructure;

(e) Assisting in the production of national statistical compendiums.

111. So far, the emphasis has been on improving national statistical capacity; in the short term, until the end of 1998, this should remain the priority. The Expert Group also recognizes that there will be an imperative to bring together at a regional or international level a report showing the state of each country's population. In addition, monitoring of national progress on improvements in gender statistics and in data on disadvantaged subgroups of the populations has been specifically requested by recent major United Nations conferences.

112. Ideally, such a report would show that progress has been achieved since 1994. But given the absence of sufficient MNSDS indicators in many countries at present, the first such report may well be a status report rather than a progress report.

113. To keep the momentum and iron out some of the difficulties encountered in preparing such a report, it is suggested that a pilot report be prepared, restricted to a sample of countries. Such a report could be considered by the Working Group in autumn 1997 or spring 1998, together with the information on national capability collated by the Statistics Division.

114. The Expert Group proposes that the United Nations Statistics Division propose to the Statistical Commission at its twenty-ninth session suggestions for the nature and timing of an international report, partly based on the findings of the pilot report. The scope of such a report, i.e., whether it should have a regional or global emphasis, should be carefully considered.

115. The suggested times/periods for consideration and commencement of the various activities outlined above are summarized below:

April 1996 Consideration of the Expert Group report by the Working Group

1st half 1996 Begin identification of desirable action at the national level

1st half 1996 Begin fund-raising for the regional commissions' assessment exercises

2nd half 1996 Assessment exercises by the regional commissions

1st half 1997 Collation by the United Nations Statistics Division

Autumn 1997 Consideration by the Working Group

or spring 1998

1st half 1997 Begin fund-raising for national capability-building

1st half 1997 Begin national production and dissemination of MNSDS data

1st half 1997 Begin preparation of pilot international report

Spring 1997 Discussion of nature of pilot international report by Statistical Commission

Autumn 1997 Consideration of pilot international report by Working

or spring 1998 Group

Spring 1999 Consideration of draft international report by the Statistical Commission.

a/ See Report of the International Conference on Population and Development, Cairo, 5-13 September 1994 (United Nations publication, Sales No. E.95.XIII.18), chap. I, resolution 1, annex II, chap. XII; Report of the World Summit for Social Development, Copenhagen, 6-12 March 1995 (A/CONF.166/9), chap. I, resolution 1, annex II, para. 16 (e); and Report of the Fourth World Conference on Women, Beijing, 4-15 September 1995 (A/CONF.177/20), chap. I, resolution 1, annex II, paras. 206-209.

The Expert Group has compiled a list of 15 indicators (marked with a bullet "") to form a suggested minimum national social data set (MNSDS). These indicators are listed below, together with a number of other indicators (marked with a dash "--") that are also seen as valuable and relevant for monitoring and evaluating the progress within the policy areas of recent major United Nations conferences.

Population estimates by sex, age and, where appropriate and feasible, ethnic group

Life expectancy at birth, by sex

Infant mortality, by sex

Child mortality, by sex

Maternal mortality

Percentage of infants weighing less than 2,500 g at birth, by sex

Average number of years of schooling completed, by urban/rural, sex and, where possible, by income classes

-- Percentage of pregnant women who have at least one ante-natal visit

-- Percentage of pregnant women who have a trained attendant at delivery

-- Percentage of pregnant women immunized against tetanus

-- Contraceptive prevalence rate

-- Incidence and prevalence of sexually transmitted diseases

-- Quality of family planning services

-- Access to, and quality of, maternal health services

-- Incidence of female genital mutilation

-- Physical and mental health

-- Literacy

-- Family conditions

-- Unemployment

-- Social exclusion and isolation

-- Homelessness

-- National and international causes underlying poverty

Absolute poverty

Number of people per room, excluding kitchen and bathroom

Access to safe water

Access to sanitation

Monetary value of the basket of food needed for minimum nutritional requirement

-- Percentage of the population in poverty (poverty or poverty line defined nationally)

-- Access to services related to health, nutrition, and community or environmental infrastructure

-- Income

-- Education

-- Possibility of entering the labour force

-- Food

-- Food prices

-- Access to productive assets, especially land and water

-- Geographic location

-- Public transfers

Relative poverty

-- Families below a minimum standard of income (poverty line)

-- Poverty gap

-- Families with less than 25 per cent and 40 per cent of mean income

-- Gini coefficients (summary measure of inequality in income distribution)

-- Income share of lowest income quintile

-- Income share of highest income quintile

3. Expansion of productive employment and reduction

of unemployment

GDP per capita

Household income per capita (level and distribution)

Unemployment rate, by sex

Employment-population ratio, by sex and, where appropriate, formal and informal sector

-- Wage employment as a percentage of the population aged 16-64, for males and females separately where possible

-- Formal sector employment as a percentage of total employment

-- Median and average length of job tenure in years, for males and females separately where possible

-- Index of real wages in manufacturing and in the economy as a whole where possible

-- Ratio of average wage in the formal sector to GDP (or total wage bill) per person employed in the economy

-- Wage dispersion in manufacturing industries, measured by the coefficient of variation, for males and females separately where possible

-- Ratio of average female to average male wage in manufacturing and in the economy as a whole where possible

-- Unpaid work outside of the market economy

-- Non-wage compensation (fringe benefits)

-- Time-use

-- Precariousness of employment

-- Visible underemployment

-- Invisible underemployment

-- Training data, including informal kinds of training

-- Number of people in vulnerable groups

-- Age/gender structure

-- Occupational profile

-- Economic activity profile

-- Income levels

-- Position within overall income distribution

-- Housing standards/amenities, such as access to safe water, sanitation and floor space per person

-- Health status, such as infant mortality rate, age-specific mortality rates, expectation of life and nutritional intake

-- Educational standards, such as adult literacy rate, number of years of formal education and participation rates (for children)

-- Crime victimization rate

-- Proportion eligible to vote

Data distributed by sex on:

-- Population and households

-- Health

-- Diseases and causes of death

-- Education

-- Enrolment rates

-- Drop-out rates

-- Higher level education by subject

-- Time-use

-- Child care

-- Gainful employment

-- Wage, salary and income

-- Individual and household income

-- Informal sector

-- Income control

-- Access to land and credit

-- Influence and power

-- Decision-making

-- Time-use

-- Violence and crime