Partnership on Public Health Statistics

Transcription

1 EUROPEAN COMMISSION EUROSTAT Directorate F: Social Statistics and Information Society Unit F-5: Health and food safety statistics Doc. ESTAT/F5/06/HEA/02-2 WORKING GROUP "PUBLIC HEALTH STATISTICS" Luxembourg November 2006 Item 3 of the agenda Partnership on Public Health Statistics Draft minutes of the meeting of the European Directors of Social Statistics 18-19/09/2006 Commission européenne, L-2920 Luxembourg. Telephone: (352) Office: BECH D2/728. Telephone: direct line (352) Fax: (352)

2 EUROPEAN COMMISSION EUROSTAT Directorate F: Social Statistics and Information Society Doc. Eurostat/F/06/DSS/10/14/EN DRAFT MINUTES MEETING OF THE EUROPEAN DIRECTORS OF SOCIAL STATISTICS LUXEMBOURG, SEPTEMBER 2006 BECH BUILDING, ROOM QUETELET

3 1. Introduction 1.1. Welcome by M. Glaude: Eurostat Director Draft minutes In its introductory words, M. Glaude recalled the last modification in Eurostat's organigramme, in particular the unfortunate demise of M. Hanreich, the Director-General, at the beginning of the year, his replacement by M. Herve Carré, former deputy Director General in DG ECFIN and the arrival of the missing director of business statistics, M. Peter Bekx. The senior management team is nowadays complete. Another important event is the elaboration, together with the policy DGs and the MS of the new 5 year statistical programme Adoption of the agenda The agenda was adopted 1.3 Adoption of the minutes of the meeting on September 2005 The minutes were adopted 2. Report on the work done by the SDG M. Glaude introduced the item on the basis of the prepared document (Doc.Eurostat/F/06/DSS/10/2/EN). Two meetings of the SDG took place so far, one concentrating on some strategic issues, and leading to the creation by the DSS of two TF (core social variables and new household flexible approach), and the other collecting views of users of the Commission and producers in the members states on the future 5 year statistical programme. M. Glaude noted that at this stage, no "bureau of the DSS" function was played by the SDG. The SDG however was active on the international scene, with the organisation back to back of one of its meeting with the UNECE directors of social statistics meeting, who created a TF on emerging issues, in which 2 SDG members are represented. For the future, it was proposed to concentrate the SDG work on strategic issues and on relations with the users (priority settings for instance). The rotation of the membership should also be envisaged. Ms Wismer, member of the SDG, made further comments on the UNECE task force and follow up. The TF on emerging issues identified voluntary participation as a statistical issue.. A TF is being created on the topic. Participation is welcome. New family structures are also considered as an emerging issue at the international level, and a TF on the topic is also created. Finally, the TF would prepare a seminar, during which the TF on voluntary participation and on new family structure should present their report. During this seminar, the broader issue of the framework for social statistics would also be addressed. In their comments, the DSS underlined the importance of the SDG, proposed to maintain a rhythm of 2 meetings per year, recalled Eurostat that the role of preparation of the DSS meeting, in an enlarged Europe context and with the existence of the other sectoral groups, is important and was seen as central in the mandate of the SDG. The necessity for rotating membership was also underlined.

4 It was concluded that the SDG would try to meet twice per year, in January and June (subject to practical constraints dates would be chosen well in advance) The SDG has no "decision powers": it will continue to play an active role in identifying strategic issues, favouring the discussions between users and providers, and will also take a more active role in DSS organisational issues. M. Glaude call for additional strategic topics to discuss at the DSS and asked for volunteer in the context of renewed membership. It was agreed as well that Eurostat should be proactive in this renewal, in order to keep an adequate balance of participant. Dates for the next DSS meeting will be communicated in the next few weeks. 3. Final report from the task force on core social variables The final report of the task force has been presented by Martin Zeleny, member of the task force. The presentation referred to the key elements of the document Eurostat/F/06/DSS/10/3/EN. He underlined the main guiding principles of the TF work, the difficult issues encountered (like the income variables, the self defined labour status, the interest of a socio economic classification), as well as the implementation plans, surveys by surveys, and the conclusions and next steps. Many comments were made by the DSS. In general, the report was welcome and overall very well appreciated. The bringing to fruition of the project is seen by some participants as an achievement, in an international context. On the content of the report, many general and detailed comments were made: questions about de facto marital status, about the household (HH) definition as an economic entity, about the HH type (in particular the definition of dependant child, the classification used, the problems linked with the proxy answers, the age groups retained), cases of double citizenships, adding variables on housing conditions, etc. Whilst appreciating the work done, the needs for more detailed breakdowns have also been expressed by the users. More input harmonisation was seen as a way to increase comparability. DG EMPL called for the compatibility of the definitions to be tested against the definitions used for the main multi-dimensional lists of indicators developed in the context of EU level policy coordination. These lists typically draw on several surveys and the core variables would enhance the analytical possibilities derived from the indicators: these lists are in particular, the Laeken list (indicators for social inclusion), the EMCO list and the Structural Indicators list. Main comments were on the income variable: the choices made by the task force on collecting net monthly income limited to wages and social transfers were questioned in several directions. It was recognised that the key element is to provide a ranking of the observations by income levels. Tests and refinement were seen as essential. The self declared labour status is also a source of questions. At the same time, it is recognised that it differs significantly from the ILO definition (and that it could make a real difference in some countries), but that the implementation of the ILO definition in all surveys is way to expensive. On the implementation, some delegates underlined that the core variables should be implemented only in the survey in which their relevance is established. However, it has been reaffirmed that the intention of the whole project is to collect the core variables in all the EU social surveys, many of which already comply to varying degrees. Whilst paying tribute to the fact that the TF it self asked for a dissolution due to the termination of its work, some delegates saw however that further work should be monitored in that field.

5 On the socio economic classification, and following a paper prepared by the French delegate, it was generally agreed that this item is important in order to better describe, categorise and compare internationally socio economic groups. The link between this project and the national implementations and international revision of the ISCO was underlined. Some delegates suggested the ESS should have some ambition in addressing the issue. The next steps should include a more active role of the ESS and Eurostat, in order to test, analyse and operationalise the socio economic classification in the context of official statistics. To conclude, the DSS o Endorsed the report of the task force on core social variables (subject to the next comments), and thanked the TF members for their work, o Agreed on the proposed list of variables, o Broadly agreed with the definitions proposed, details of definitions need to be further tested. o Agreed on the principle that all core variable should be introduce in all EU social surveys subject to detailed discussions that should take place in the relevant working groups, having in mind the onjective of including all core variableq in all surveys by 2010, o Noted that income cannot at this stage be included in the LFS and confirmed that the introduction of core variables in the ICT household survey cannot be done at the expense of the substance variables of this survey: when core variables would be collected, it should be in addition to the substance variables, o Will transmit their written comments on the report by 20 October 2006 (to jeanlouis.mercy@ec.europa.eu) o Asked Eurostat to provide a consolidated version of the report at the next SDG meeting, with the help of the TF if needed, o Mandated the SDG for taking actions concerning the next steps in the work, o Welcome the proposal to further continue methodological developments and tests on a revised version of the self declared labour status and on the income variable (in the context of the EU-SILC methodological task force), o Dissolved the task force, as soon as the consolidated manual will be issued, o Created a new task force on socio economic classification (volunteers are asked to declare their interest at michel.glaude@ec.europa.eu). 4. European Household Survey (EHS): background and first proposal in relation to the content of the new EHS Eurostat presented document F/06/DSS/10/4/EN highlighting the modified concept now foreseen for this project, change from a single survey to be carried out in all Member States (EHS) to a system of household survey modules (E4SM) which is flexible both for the demand side (with the inclusion of both short term and long term modules to answer to new Commission demands) and in terms of implementation (the different modules can either be integrated into a single or a dual operation or can be integrated in existing national instruments under constraints in terms of sampling, collection period, sample size, timeliness). It was also stressed that this was only the result of one Task Force meeting and that an important volume of work would have to be done by the TF before implementation. Eurostat insisted on the fact that for modules answering new Commission demands, input harmonisation would be an essential requirement.

6 The DSS agreed on the need to take active steps to develop a system of household survey modules and on the change of concept from a single survey to a flexible system of household survey modules. Consequently, the Task Force should go on with its work: a meeting will now be organised in December. Although the change of concept was welcome, the new name (E4SM) was found too complicated by some delegates. The draft revised mandate has been accepted with a minor change proposed by UK in task 4 (Based on the foregoing (i.e. Task 3), the TF would also detail financial and regulatory implications of the different proposals/scenarios, where possible setting them in the context of a global burden). The long term perspective of this project was highlighted by several countries. The discussion further identified an over-optimistic timescale for the implementation of the project in view for example of the SILC experience (2011 would seem much more realistic); several countries expressed views/preference on the input/output approach. Some countries asked for a more precise estimation of the total cost of the project and insisted on the importance of EU funding. On the procedure for selecting themes of modules responding to new DG demands, the DSS recommended for the SPC to have the final word. There was also a request to clarify the regulatory framework. DG EMPL pointed out that the procedure and the timing for selecting modules need to take the constraints of the EU financing procedures into account since several Commission services may be involved. DG EMPL also needs to have an estimate of the financing needed fairly soon in order to be able to plan for it. In terms of the specific modules, issues such as the need to include on an annual basis the EHIS indicators and the ICT component, the integration of the LFS and SILC modules into the system, the link between the proposed system and the existing operations (ICT, AES), the problems linked to the proposed split of the current AES operation into three parts were raised. Eurostat confirmed the need for annual data on ICT and main EHIS indicators, in the context of the i2010 package of policies for ICT, for the need of European Community Health Indicators (the EHIS is only carried out every five years and there is a need for indicators with potential annual variations to be computed). It was also clarified that no duplication will be done: according to this model, the ICT and AES operations would in the long run either be implemented as currently done or be integrated in a single/dual survey depending on the national choice. The split of the AES survey would need to be further elaborated together with the AES Working Group (also in view of the results of the first data collections). It was concluded that the Task Force would tackle most of the discussed issues (cost, procedure for prioritising the themes, input/output approach, and timescale) and come back with a progress report in the next DSS meeting. 5. Draft Commission Regulation (EC) implementing European Parliament and Council Regulation No 1177/2003 on Community Statistics on Income and Living Conditions (EU-SILC) concerning the 2008 list of target secondary variables on financial exclusion and over indebtedness Eurostat presented the draft Commission regulation concerning the module on over indebtedness and financial exclusion to be included in 2008 in the cross-sectional component of the EU-SILC project. This draft Commission regulation is planned to be approved by the

7 SPC on 16 November 2006 and adopted by the Commission at the end of 2006 or beginning of The discussion mainly focussed on the need for a new round of written consultation of the Working Group given the fact that the format and the content of the module has somewhat changed since the last consultation of the Working Group (during summer) mainly with the drop out of type of debts and an alternative approach for the quantification of arrears. The change from a "blue print" questionnaire to a list of variables between the summer written consultation and the DSS meeting, the link with the ECB project, the definition of commercial credit and the collecting unit for this module (household or individual level) were also raised. On this last issue, the quality of collecting household information through household respondent only was especially questioned but difficulty to individual data collection given current definition (for instance use of household scale) was also recognised. Eurostat confirmed that the current document containing the list of target variables needed to be included in the Commission regulation will be complemented by a blue print questionnaire to be available at the end of this year. It was also confirmed that there is no duplication with the Eurosystem initiative to investigate the feasibility of a survey on household finance and consumption. The focus of the possible Eurosystem survey (the realisation of which is not yet decided) would be on gathering detailed information on household assets and liabilities. An oversampliong of wealthy households would be needed to get reliable results on households portfolios which is in contrast to the EU SILC module focusing on low income households. The majority of the Member States gave a favourable opinion regarding the draft Commission regulation on the 2008 EU-SILC module, but it was agreed to launch a quick written consultation of the WG (with a deadline of 22 September for answer) to produce the final version to be discussed by the SPC in November. The definition of commercial credit and the modalities of answer to the 'reasons why the household does not have commercial credit' would be slightly amended. 6. Draft Commission Regulation on LFS module 2008 on labour market situation of migrants and their immediate descendants The draft Commission Regulation adopting the specifications of the 2008 ad hoc module on the labour market situation of migrants and their immediate descendents was presented to the DSS group. The main aspects underlined were that the migration issues are having more and more political importance and therefore this module will be very likely repeated in 3 or 4 years. Consequently, this module will be a pilot experience in the area. As some variables were considered as difficult (subjectivity, low quality of proxy answers, etc), Eurostat takes the engagement of: Making an ex post evaluation of this module during 2009 (TF with Member States). This task force will also conduct the comparative analysis of the results. To conduct a careful analysis of the quality data before publication To include warnings for data considered of less quality Not publishing data for individual countries where this data is under the reliability limits and use it only for the EU totals. Use the data from this module more for structures than for levels. Apart from the derogation granted to countries having less than 1500 migrants in the LFS sample to do a light version of the module (only 4 variables), 2 other derogations will be

8 granted to France (variable 214 optional) and Germany (collect information on nationality/former nationality instead of country of birth). For this last country a legal text will agreed bilaterally. Under these conditions the DSS have a given a general support to this module. 7. Draft Commission Regulation on the scope and the breakdown extension of the quarterly Labour Cost Index (LCI) Draft Commission Regulation extending the Labour Cost Index to cover NACE rev 1 sections L, M, N, and O The draft Commission Regulation extending the Labour Cost Index (LCI) to cover NACE sections L-O was presented to the DSS group. It was explained that the results of the feasibility studies carried out in accordance with the provisions of Regulation (EC) 450/2003 were positive with respect to user needs, feasibility, and costs, although some Member States would need to make changes and improvements in their data collection. The implementation of NACE 2 by 2009 was also mentioned. The draft Regulation foresaw an implementation by 2007q1 and, if justified, a delay until 2009q1, although Eurostat underlined the need for an early implementation and a harmonised and comparable breakdown of NACE sections. DG EMPL welcomed the draft proposal and emphasised the need for back data until year 2000 in the context of the Lisbon Agenda. The ECB stressed the need for back data as well. Several Member States said that they would only be able to implement the extended coverage by 2009q1. Both DG EMPL and the ECB were in favour of reduce as much as possible the number of countries that will delay the extension of coverage until 2009q1 in order to produce EU aggregates of good quality already from 2007q1. Even though there is no legal link between the Labour Cost Survey (LCS) and the LCI, it was recognised that the LCS does not yet cover NACE section L in some Member States, which may have a negative impact on the estimation of social contributions within the LCI. The feasibility studies also covered the compilation of an index excluding irregular payments (bonuses). Since the results of those studies were not fully conclusive, the immediate priority had been given to the extension of the LCI to cover NACE sections L-O. Nevertheless, Eurostat is committed to review this issued in The ECB stressed the need for an index excluding irregular payments. The DSS gave its support to the draft Regulation and the proposal to review the issue of an index excluding irregular payments in EU-SILC: decision on the introduction of employers' social insurance contributions from 2007 Eurostat presented document F/06/DSS/10/8/EN explaining that this point is on the agenda of the DSS because according to SILC Commission regulation N 1980/2003 on definitions, it is planned for the income component 'employers' social insurance contributions' to be possibly included in SILC from 2007 based on the results of the feasibility studies. On this basis, Eurostat proposal is to include this component in SILC from 2007, but limiting its content to compulsory (legal) employers contributions if the latter represent at least 90% of total social insurance contributions (compulsory and voluntary). Technical aspects still to be solved

9 (imputation of a fictitious value for public sector employees) would be discussed in the SILC methodological TF meeting in December In the discussion, Poland expressed difficulties because the information is currently collected via interview which has proved to be non reliable. Germany made it clear that at present they can only provide the data on legal employer social insurance contributions, independently of what portion this constitutes of the total employer social insurance contributions The Eurostat proposal was found acceptable by other MS It was highlighted that this short/medium term proposal would need to be rethought in future together with the increasing role of the voluntary component. The DSS agreed with Eurostat proposal to introduce in SILC from 2007the employers' social insurance contributions variable, limited to compulsory components if it represents more than 90% of the total contributions. The level of the threshold should be monitored in function of the results of feasibility studies yet to come and the evolution of structure of social insurance. Bilateral contacts/exchange of experience with other countries would be done for Poland to evaluate possibility of developing imputation/modelling using auxiliary information collected by interview or register.. 9. The household survey on ICT usage M. Glaude introduced the item by pointing out that one of the reasons for the success of the ICT usage surveys was the fact that Eurostat had actively anticipated the policy needs. In her presentation of the paper, B. Knauth drew particular attention to Eurostat's desire to receive micro-data, regionally broken down data and data on household income on a voluntary basis. Additionally, Member States were encouraged to send a representative to the forthcoming meetings of the Task Force that will be devoted to defining the questionnaire for the 2008 survey (with a special module on advanced services). DG EAC emphasised the importance of this survey for issues related to education (e-learning, e-skills). Several Member States considered that the provision of micro-data would not constitute any problem. Eurostat noted the desire of these Member States that the transmission of aggregate data be not supplemented but replaced by a transmission of micro-data. However, as one of the important strengths of the survey is the fast release of data, it has to be ensured that the use of micro-data does not slow down the production speed. Once the respective resource issues in Eurostat will have been solved and the treatment of micro-data will have been tested successfully. the transmission of aggregate data can be given up. As regards regional data, Eurostat clarified that the purpose of the demand was not to increase the survey costs, but to receive data that might already be available on a national basis and to encourage stratification by region, which for many Member States would allow the publication of reliable basic indicators without increasing the sample size. Eurostat acknowledged the link between educational level and income and therefore the possibility to use the variable on educational level as a substitute for income. However, as studies by Statistics Canada (using micro-data) have shown the importance of income as an explanatory factor, complementary to the educational level, Eurostat maintains its interest in receiving data broken down by broad income level.

10 Although it is difficult to predict, whether a survey on ICT usage in households and by individuals will be necessary on an annual basis after 2010, Eurostat drew the attention of the participants to the fact that the phenomena under consideration are subject to very fast annual changes and that the interest of policy makers is constantly shifting to the adoption by individuals of new technologies or behaviours (e.g. from internet penetration to broadband penetration). 10. Main developments in Demographic and Migration Statistics Mr Skaliotis, Head of Unit for Demographic and Migration Statistics, briefly described the main developments and challenges in demographic statistics and replied collectively to the written comments delivered by Member States. Member States commented on the progress made in finalising the proposed Regulation on Migration Statistics. Some technical difficulties remained that would need to be addressed in finalising the Regulation, and during its implementation. The particular difficulty of producing good statistics on emigration was mentioned. It was felt by several Member States that the compromise version of the text that had been negotiated in the Council Working Group represented the very maximum that was feasible in terms of the variables and disaggregations. The European Parliament appeared not to accept some of the amendments that had been negotiated between the Council and the Commission, and this would make it difficult for Member States to continue to support the Regulation. Member States pointed out that the migration and asylum statistics data were often supplied by Ministries of Justice and Interior, not by National Statistical Institutes. It was important that these other parts of the national administration should continue to be involved during the implementation of the Regulation. Eurostat would consider how best to ensure the involvement of these Ministries in the ongoing work on migration and asylum statistics. The meeting asked for clarification of details of the planned Regulation on Population and Household Censuses in the EU. Eurostat confirmed that the first reference year would be The obligatory and recommended topics to be reported on would correspond as closely as possible to the topics listed in the recently adopted Conference of European Statisticians Recommendations for the 2010 Censuses of Population and Housing. Regarding the transfer on samples of anonymised microdata to Eurostat, which was envisaged as a recommended rather than compulsory part of the 2011 censuses programme, it was noted that detailed technical discussion would be necessary to agree the methods for the protection and secure transfer of data. A copy of the current draft of the planned Regulation on Population and Housing Censuses was provided to interested participants. Eurostat presented plans for a new round of national level population projections to be published early in Work was also well advanced on regional level projections. Member States urged Eurostat to take full account of national views when developing the assumptions for the new projections. A time plan for the exercise was required, ensuring adequate time for national comment and review. Eurostat responded that the EUROPOP2004 projections had been the subject of extensive consultations with national experts and that this would be repeated for the next projections round. Eurostat would welcome a greater involvement of National Statistical Institutes in the next round. A Working Group meeting in November would focus on national methodologies and national expertise. It was also proposed that in the next round of national releases of population projections, Member States include the EUROPOP2007 baseline variant among the published variants.

11 There were questions about possibility of updating the EUROPOP2004 projections in the light of additional information on migration available in some Member States. However, as this projections round was now formally closed, it would not be possible to make updates. 11. Partnership Health: evaluation and possibilities for an adapted structure M.Glaude introduced the point mentioning that a more general discussion on CENEX which is one of the options to continue work under Partnership Health (PH) as described in document Eurostat/F06/DSS/10/11/EN - is still ongoing at the level of the Statistical Programme Committee (SPC). On behalf of ONS-UK, who is in charge of the ongoing grant agreement 'on 'Partnership Health' (as well as for the previous grant agreements since 2003) Madhavi Bajekal presented the evaluation of the work carried out (since 2003) in the context of PH, the advantages and lessons learned and the proposed options for continuation. All participants agreed that 'Partnership Health' should continue and expressed to support it. The Portuguese delegate raised the concern about the need for a greater coordination for the work on System of Health Accounts (SHA), in particular for the revision of the SHA Manual such as by setting up a separate group on it. Several countries raised the issue of very limited (human) resources to contribute to the work. The UK delegate suggested that the resource implications (financial and technical) should be carefully worked up by Eurostat and should be given attention in further discussions, such as on CENEX, at the SPC. In reply to a question of the Finnish delegate on why some health questions would now also be included in the proposed E4SM, Eurostat replied that, for a few indicators out of the ECHI list, there is a need to have annual data, which the 5-yearly EHIS would be unable to provide. As a result of the discussions the DSS agreed that - in terms of the future PH structure 1) a fourth pillar (on morbidity statistics) will be established with only a Core Group (CG) for 2007, the establishment of a Technical Meeting will follow later; 2) the (3) Technical Meetings will be directly organised by Eurostat (so not by the Partnership Health coordinator), although there will be a close involvement of the leading partners in the preparation of and discussions at the Technical Meetings. In view of the ongoing discussions on CENEX, the DSS agreed that - for the time being - option 2 would be followed, with a possibility to progress to a CENEX structure (option 3) at a later stage, when the general discussion on CENEX would become clearer. The meeting also agreed with the proposal from the Irish delegate to set up a 'users'-panel, which would inter alia allow Commission DGs to provide their input into the PH work. Marleen De Smedt, Head of Eurostat Unit F/5 'Health and food safety statistics' stressed that the idea itself of 'partnership' requires a contribution of all Member States (MS) in the work, with a more direct involvement of the leading partners. The proposed PH strives for a balance between assuring that all MS could contribute, while keeping the number of groups and meetings within reasonable limits.

12 M.De Smedt agreed to give special attention to the cooperation with MS on the revision of SHA. The ongoing PH already envisages to establish a special Task Force (TF) on the SHA revision and the issue will be discussed with all MS at the 2007 annual Technical Meeting on Health Care data. Finally, M.De Smedt announced that - with the establishment of a new grant agreement (call for proposals will be launched early 2007) - there will be a possibility to re-look at the composition of the different core groups. 12. Statistics on crime and criminal justice Geof Thomas (from Eurostat Unit F/5) Eurostat outlined the work already done on statistics on crime and criminal justice since 2004 within the context of a task force; Eurostat suggested that a working group should now be created to carry out the mandate set out in the document Eurostat/F/06/DSS/10/12/EN. Fabio Marini of DG Justice Freedom and Security (DG JLS) described the scope of the EU Action Plan on developing a strategy to measure crime and criminal justice, which had been the subject of a communication from the Commission in August The Commission Decision to create an Expert Group to establish the policy needs in this area, developed by DG JLS in parallel with the Action Plan, would be presented to an interdisciplinary group in the Council in October. DG JLS would then ask the permanent representatives for nominations for members of the Expert Group. The comments of the participants indicated general agreement with the suggestion to set up a Eurostat Working Group to complement the activities of the Expert Group in implementing the Action Plan. Following a proposal from Italy, it was agreed to change the wording of the third objective of the Working Group, to discuss guidelines and ways of implementing the collection of statistical data The Czech delegate stressed the importance of using administrative sources in preference to survey data. Several countries expressed strong support for improving the comparability of the existing information, while also underlining the complexity of the topic and the difficulty of co-ordinating the activities of the various government bodies working on crime statistics. The UNECE informed the meeting that they had compiled with the UNODC an inventory of over 80 victimisation surveys world-wide and were now developing a manual on such surveys which would provide valuable input for the development of the EU module. The issue of violence against women was also considered of high importance and it would be examined in the context of the EU Action Plan (point 5.12). It was agreed to set up a Working Group which would initially meet once a year, beginning in March Eurostat would contact DSS members to ask them to nominate a national representative in the Working Group. 13. Statistics on the gender Pay Gap Improving the measurement of the Gender Pay Gap (GPG) indicator A majority of the DSS members were in favour of the solution of using SES as data source for the GPG, on the basis of a common coverage for all countries (i.e. NACE C-K for 2002 data) with yearly updates with national sources. In SES 2002 some countries already cover the

13 whole economy (all sizes of enterprises and LMNO sectors). Eurostat could do simulations to assess the effect on the GPG of both extensions. The SES 2006 will extend coverage to MNO sectors. The LFS was not considered as a viable solution since the earnings variable will be given only in bands. The present situation for calculating the GPG will be applied as a short-term solution only for one additional year, in order to obtain the GPG data for 2005 and should not be continued afterwards. Other discussed issues relating to the definition of the GPG (medians vs means, part-time/full-time, hourly vs monthly wages...) and to the breakdowns needed by data users (age, NACE, education, occupation...) should be finally decided at EMCO meeting, nevertheless, proposals from Eurostat (and the ESS) could be presented. The DSS agreed on the proposal of INE Spain to set up a Task Force for improving the GPG calculation. The work of the TF will include methods and simulation for producing yearly updates of the GPG for the intermediary years 2003, 2004, 2005, using the GPG from SES 2002 and national sources such as yearly earnings surveys or administrative sources. Other work of the TF may concern the assessment of trends and recommendation on the methodological issues mentioned above. Eurostat will propose to the DSS a mandate for the TF, as well as a request for participants. The work of the task force should also serve as input for preparing the Communication from the Commission in 2007 on the GPG 14. Other business next meeting No additional topics were proposed under this item. The next meeting of the DSS is foreseen for next year, at the same period.