EUROSTAT
Directorate E: Social and regional statistics and geographical information system
Unit E-2: Living conditions /
Doc. E2/HBS/135/01/EN
DOC.E2/CC/05/2001
Working Group
HOUSEHOLD BUDGET SURVEYS
01-02 OCTOBER 2001
Seminar
Co-operation with Candidate Countries on StatistcIs
On Income, Poverty and Social Exclusion
03 October 2001
EU-SILC
1.BACKGROUND
At the meeting of Directors of Social Statistics of June 1999 it was decided the replacement of the ECHP after 2002 in order to adapt the production of EU statistics on income and living conditions to the high political demand arising from the Amsterdam treaty. A Task Force (TF) was set up having met already four times - in November 1999, February 2000, September 2000 and March 2001. Its mandate is to carefully look at all options on the content and structure for an EU-harmonised source on income and living conditions with a specific focus on the multidimensional phenomenon of social exclusion. A fully developed integrated solution (with tested design and procedures) is to be ready by the end of 2002.
At the last meeting of the ECHP Working Party held in Luxembourg on 23-24/03/00, the first achievements of the TF were discussed:
- the identification and detailed description of the basic principles to be met by the new instrument (hereafter referred to as EU-SILC, i.e. EU Statistics on Income and Living Conditions),
- the list of topics to be included in it,
- and the main future steps to be followed by both, the TF and WP.
Some members of the WP raised the question of the need to enforce EU-SILC through a legal act. This would provide a firm legal basis for future funding, which is likely to significantly improve the quality and timeliness of EU-SILC data, both within National Statistical Institutes (NSIs) and Eurostat.
This issue was submitted to NSIs at the annual meeting of Directors of Social Statistics on 13-14/06/00. Directors concluded as follows regarding the introduction of a legal act for EU-SILC (Doc. Eurostat/E0/00/DSS/2/9/EN):
“Some Member States would need a legal framework; some would have no problem in accepting one; the others preferred not to have one since it might reduce flexibility, but would accept one if it were necessary and could be done in a way that would achieve the objectives for the project.”
As a consequence, Eurostat decided to start investigating further the procedure to be followed for the preparation of such a legal act, as well as investigating concrete examples from similar projects. The present document is the first step in this direction.
It builds on the various requirements highlighted by NSIs' Directors of Social Statistics in their June meeting as well as the work of the TF and various consultations with the Commission services.
2.Policy relevance
The need for a statistical tool on income, living conditions and social exclusion, capable to produce comparable indicators in these domains at EU level, is justified in several legal texts produced by the Council and the Commission. A brief summary of the high political demand existing for such statistics is given next.
2.1.The Amsterdam Treaty
According to art. 285 of the Amsterdam Treaty, “the Council, acting in accordance with the procedure referred to in Article 251 (i.e. the co-decision procedure[1];), shall adopt measures for the production of statistics where necessary for the performance of the activities of the Community”.
Articles 136, 137 and 284 of the Treaty, show that the statistics on income, living conditions and social exclusion that are to be collected in the context of EU-SILC are indeed “necessary for the performance of the activities of the Community”.
2.2.The Lisbon, Feira and Nice European Councils
The conclusions of the Lisbon European Council (23-24/03/00) demanded to make a decisive impact on the eradication of poverty by setting adequate targets to be agreed by the Council by the end of the year 2000, and by inviting the Council and the Commission to promote a better understanding of social exclusion through continued dialogue and exchanges of information and best practice, on the basis of commonly agreed indicators. Three months later, the Feira European Council (19-20/06/00) concluded along the same lines on the need to modernise social protection and promote social inclusion.
The High Level Working Party on Social Protection (HLWP) has been involved in this work preparing the objectives of combating poverty and social exclusion, which were approved by the Nice European Council. Member States have been invited to develop their priorities in relation to these objectives, and to submit by June 2001 a national action plan covering a two-year period and defining indicators and monitoring mechanisms capable of measuring progress.”
The Social Protection Committee, which took over the activities of the HLWP on Social Protection from January 2001, is now in charge of the implementation of the measures to fight against poverty and social exclusion as approved in the European Councils.
2.3.Commission Initiatives
The “Programme of Community action to encourage co-operation between Member States to combat social exclusion”, which the Commission has submitted in the year 2000 to the European Parliament and the Council, is directly linked to the Lisbon and Feira conclusions. The first suggested strand for the implementation of this programme is indeed the “analysis of characteristics, processes, causes and trends in social exclusion”. The “collection and dissemination of comparable statistics in Member States and at Community level” is one of the measures considered to promote this.
The European Commission’s Communication COM(2000) 594 on Structural Indicators, adopted by the Commission on 27/09/00, is another major step in this direction. Some of the social cohesion indicators included refer to the inequalities of the income distribution, poverty rates before and after social transfers and the persistence of poverty, that again, need to have a high degree of comparability, in order to focus on progress among the EU countries in this policy domain.
3.Timetable for the preparation and adoption of the framework EU-SILC Regulation
A target timetable is as follows:
05-06 March 2001Opinion of Task Force on the long-term future of the ECHP
25-27 April 2001Opinion of the ECHP Working Party
30-31 May 200141st SPC: first presentation of a draft framework Regulation proposal to be adopted by both the European Council and European Parliament (see section 4.1 below)
June 2001Analysis of technical remarks from 41st SPC by NSIs’ Directors of Social Statistics
End of June 2001Inter-services consultation within the Commission services (DGs: Employment, Research, Legal Service, Secretariat General)
Mid July 2001Finalise first draft of framework Regulation proposal to be presented at the 2001 SPC meeting.
20 Sept 200142nd SPC: approval of a draft framework Regulation proposal
Beginning of Nov 2001Adoption of the proposal by the Commission
ThereafterDocument transmitted to the Council and the European Parliament
ThereafterFirst reading of the Council (agenda in the hands of the Presidency) and of the European Parliament
ThereafterIf needed: amendments, second reading, conciliation procedure and adoption of the Regulation.
Entry into force of this framework Regulation: Twenty days after publication in the Official Journal of the European Communities. Target date for entry into force: first January 2003.
Commission's Regulations as described under 4.1 will be adopted under this framework Regulation.
4.Principles for a EU-SILC Regulation – Legal basis and coverage
4.1Legal basis for EU-SILC
The legal basis for the EU-SILC is proposed to be a Regulation of the European Parliament and the Council. The basic (framework) Regulation will be supplemented by Commission Regulations implementing the methodology and other technical aspects pertaining to the basic Regulation. Within the limits of the power delegated to the Commission, subsequent amendments to specifically identified elements of the basic Regulation will also take the form of Commission Regulations. As from now, most urgent Commission's Regulations can be drafted and consultations about these can be launched. However, approval of these is of course subject to prior approval of framework Regulation.
The objective of the basic Regulation will therefore be to produce a document clearly defining the general terms of the methodology and procedures, leaving to the Commission (by delegation, art. 202 of the Treaty) the possibility, after discussion with the Statistical Programme Committee (SPC), of amending some of the criteria.
4.2Requirements for EU-SILC
Based on the conclusions of NSIs’ Directors of Social Statistics (June 2000) and on subsequent work of the TF, Eurostat proposes that EU-SILC meets the following requirements:
a) General statements
The aim of EU-SILC is to generate information to support Community (i.e. Commission andmember states) social policies in their broadest sense, which clearly point to the need to cover diverse areas. However, only income information and main indicators of social exclusion should be covered in detail in EU-SILC. In particular, EU-SILC is to become the reference source of comparative income distribution and social exclusion statistics at EU level.
The domains and variables to be covered by EU-SILC in its two major dimensions (cross-sectional and longitudinal) as identified to date are given in annex 1 and annex 2. They are based on the work of the TF. The final coverage of EU-SILC is consolidated in such a way that the general requirement on flexibility and timely national cross-sectional data are not put in danger. Apart from income, topics such as socio-demographics, labour, social participation, housing, health… are all included. However they are limited to all(but no more than) key variables required for the full understanding of difference in income distribution and of social exclusion in its complexity and multidimensionality.
The priority of EU-SILC is to deliver high quality data at the national (not subnational) level. Comparability and timeliness is of key concern in that respect.
Data are required for the monitoring of income and social exclusion in both cross-sectional (pertaining to a given time of time period) and longitudinal (pertaining to individual level changes over time) dimensions. However, the first and clear priority is to be given to the delivery of high quality cross-sectional data.
Requirements for longitudinal data are more restricted than those for cross-sectional data - in terms of both coverage and sample size. One of the primary objectives of this dimension is to identify the incidence and dynamic processes of persistence of (monetary) poverty and social exclusion among subgroups in the population.
In order to be in a position to carry out multi-dimensional analysis at the level of households and persons -whether directly or indirectly (i.e. in the context of specific EU-SILC research contracts (see section 4.6))- and in particular to investigate major issues of social concern that are new and require specific research, it is essential that the cross-sectional (resp. the longitudinal) information be linkable at the household and personal level.
The longitudinal data set does not need to be “linkable” with the cross-sectional data set at the micro-level, though such linkage is not precluded as would normally be the case when the two types of data come from the same source (see below: the “flexibility” requirement).
EU-SILC has to be flexible in terms of data sources. Eurostat strongly encourage the use of existing data sources, whether surveys or registers.
Given that EU-SILC will also have to collect some qualitative/subjective data (on perception, self-assessment…), even those member states that plan to use administrative registers will have to complete this information by survey-based data. This may involve the adaptation of existing national surveys for the purpose where possible, or, perhaps more typically, the development of new survey instruments. The latter choice will have the advantage of facilitating the adoption of more closely ('input') harmonised concepts and definitions (such as the household concept, see section 4.7.1).
While encouraging the use of national sources, Eurostat will recommend an integrated design for EU-SILC to those countries planning to launch a new operation. This design aims at meeting at best costs/efficiency both cross-sectional and longitudinal requirements.
The cross-sectional and longitudinal micro-data sets will be updated on a yearly basis.
All the above does not preclude the possibility of adding ad-hoc modules to EU-SILC to investigate particular areas of policy interest in more detail if and when required.
b) Cross-sectional data requirements
It is envisaged that the cross-sectional micro-data set will contain statistical information on all the members of a given household and the relationship between those members, with full information on members aged 15 and above[2]. However, for flexibility reasons, Nordic countries might not interview all household members, provided that they prove that comparability is not affected.
The sample-based cross-sectional data will come from nationally representative probability sample of sufficient size to permit detailed analysis by population subgroups (such as demographic, socio-economic and other 'non-geographic' classes). Specific sample size requirements will be developed and agreed upon for the purpose.
It is neither possible nor necessary to standardise the sample design and procedures, whether the data come from registers or surveys. For instance, depending on the sampling frame available in each country, the sampling unit may be either the person or the household. However, the sample selection and implementation procedures must follow certain common standards so as to ensure that a representative sample of households and persons is obtained in each country.
c) Longitudinal data requirements
Longitudinal data will be restricted to income information, labour and a limited set of critical qualitative, non-monetary variables of deprivation.
The longitudinal component will also be more limited in sample size compared to the primary, cross-sectional component. Furthermore, for any given set of individuals, micro-level changes will be followed up only for a limited duration, such as a period of four years. Detailed tracing rules will be developed later.
For register countries, a key issue is to implement longitudinally the concept of household (for the construction of household-based longitudinal indicators such as persistent poverty).
ANNEX 1: List of SILC Target Variables
1
Table 1: Primary Topics to be Included in EU-SILC
Unit (Individual or Household) / Mode of Collection / Domain / Topic or Area / ReferencePeriod / Cross-sectional (X) or
Longitudinal (L)
Household / Household Respondent[3] or extract from registers / BASIC DATA / Basic household data / Current / X, L
INCOME / Total household income (gross and disposable) / Income reference period[4] / X, L
Gross income components at household level / Income reference period / X, L
SOCIAL EXCLUSION / Arrears / Last 12 months / X, L
Non-monetary household deprivation indicators / Current / X, L
Physical and social environment / Current / X
HOUSING / Basic housing conditions / Current / X, L
Amenities in the dwelling / Current / X
Housing-costs / Current / X
All persons under 16 / BASIC DATA / Demographic data / Current / X, L
Former Household Members / Demographic Data / Income reference period / L
Unit (Individual or Household) / Mode of Collection / Domain / Topic or Area / Reference
Period / Cross-sectional (X) or
Longitudinal (L)
All persons aged 16 and over in the household / Personal interview of all household members aged 16 and over[5] or extraction from registers / INCOME / Gross personal income, total and components at personal level / Income reference period / X, L
BASIC DATA / Basic personal data / Current / X, L
Demographic data / Current / X, L
EDUCATION / Education / Current / X, L
LABOUR
INFORMATION / Basic labour information / Current/ Income reference period / X, L
Second job / Current / X
At least one household member aged 16 + (the selected respondent) / Personal interview of the household member(s) (proxy an exception) or extraction from registers / HEALTH / Health / Current / X, L
Access to health care / Past 12 months / X
LABOUR
INFORMATION / Detailed labour information / Current / X, L
Activity History / Working Life / L
Calendar of Activities / Income reference period / L
1
Table 2: Target variables in EU-SILC
Domains / Topics or Areas / Qr. Block / Target Variable / Reference Period / Cross-sectional (X) or Longitudinal (L) / Unit (Individual or Household) / Mode of CollectionBASIC DATA / Basic household data / HB1 / Country / Constant / X,L / Household / Frame
HB2 / Year of survey / Current / X,L / Interviewer record
HB3 / Household Identification Number / Current / X,L / Frame or register or interviewer record
HB4 / Household Split Number / Current / L
HB5 / Region ( NUTS 2) / Current / X,L
HB6 / Primary Strata / At selection / X,L*[4] / Frame or register, sample design
HB7 / PSU / At selection / X,L*
HB8 / Order of selection of PSU / At selection / X,L*
HB9 / Degree of urbanisation ( LFS concept) / Current / X,L / Constructed
HB10 / Household cross-sectional weight / Current / X,L
HB11 / Household design weight / At the selection / X,L
HB12 / Change of address / Current / L / Interviewer record
HB13 / Address incidence / Current / X,L
HB14 / Household interview result / Current / X,L
HB15 / Day of household interview / Current / X,L
HB16 / Month of household interview / Current / X,L
HB17 / Year of household interview / Current / X,L
HB18 / Identification number of person responding the household Questionnaire / Current / X,L
HB19 / Identification number of person 1 responsible for the
Accommodation / Current / X,L
Domains / Topics or Areas / Qr. Block / Target Variable / Reference Period / Cross-sectional (X) or Longitudinal (L) / Unit (Individual or Household) / Mode of Collection
BASIC DATA / Basic household data / HB20 / Identification number of person 2 responsible for the
Accommodation / Current / X,L / Household / Interviewer record
HB21 / Number of minutes to complete the household
Questionnaire / X,L
INCOME / Total household income (gross and disposable) / HY1 / Total gross income[5] / Income reference
period / X,L / Constructed
HY2 / Total disposable income[6] / X,L
Gross income components at
household level / HY3 / Imputed rent[7] / X,L
HY4 / Income from rental of a property or land / X,L / Household respondent or registers
HY5 / Family–related allowances / X,L
HY6 / Social assistance / X,L
HY7 / Housing allowances / X,L
HY8 / Regular inter-household cash transfers received / X,L
HY9 / Interest, dividends, profit from capital investment in
unincorporated business received / X,L
HY10 / Interest repayments on mortgage[7'] / X,L
Domains / Topics or Areas / Qr. Block / Target Variable / Reference Period / Cross-sectional (X) or Longitudinal (L) / Unit (Individual or Household) / Mode of Collection
INCOME / Gross income components at household level / HY11 / Income received by people aged under 16 / Income reference