EUROPEAN COMMISSION EUROSTAT Directorate F: Social statistics Unit F-5: Education, health and social protection Luxembourg, 07/11/2017 DOC SP-TF-2017-06.3 https://circabc.europa.eu/w/browse/5010d8a2-7c57-4e6c-9766-40a46329e281 ESSPROS Task Force on Methodology 14-15 November 2017 Methodological seminar on multi-country database on benefit recipients Overview and conclusions Meeting of the ESSPROS Task Force on Methodology Luxembourg, 14-15 November 2017 BECH Building (Eurostat) B2/464
Methodological seminar on multi-country database on benefit recipients Overview and conclusions Brussels 26 th April 2017 Overview On 26 th April, a methodological seminar on social benefits recipients was co-organised by the European Commission (DG EMPL) and the OECD. The main objectives were to (i) provide an overview of the current status of the international database on social benefit recipients (SOCR) (ii) make progress on selected remaining methodological issues; and (iii) inform on-going discussions about whether and how the benefit recipients module in the European system of integrated social protection statistics (ESSPROS) can be extended to include (non-pensions) working-age benefits. Representatives from the Eurostat team in charge of the ESSPROS as well as delegates representing 16 EU countries were present: Austria, Belgium, Bulgaria, Czech Republic, Denmark, Finland, France, Greece, Ireland, Italy, Latvia, Lithuania, Slovak Republic, Spain, Sweden and the United Kingdom. Representatives from the International Labour Office, the Council of the European Union and Alphametrics also attended the seminar. Based on the results from the ongoing European Commission OECD project, this technical workshop recalled the objectives of the SOCR database, informed about country data sources and procedures and clarified type of data needed and the appropriate methodology to be followed. Opening remarks from DG EMPL and OECD focused on the motivations for the project and for collecting data on benefit recipients more generally. They highlighted the role of the common methodological framework developed as part of the EC/OECD agreement in compiling consistent and comparable data on benefit receipt as input into national and European policy processes and debates. The morning session consisted of four presentations on different institutions experiences with collecting and using data on recipients of working-age benefits: 1. DG EMPL summarised the main accomplishments since the start of the EC/OECD cooperation in 2010 (development of clear scope and a common data collection methodology and implementation of a fully functional database). They then provided an overview of the relevance and use of the results for DG EMPL (to complement tax-benefit indicators and of aggregate spending figures; programme participation and pseudo-coverage rates and better understanding of receipt trends during economic downturns). The presentation concluded by recalling the advantages of administrative data vis-à-vis surveys (more complete information, absence of response selection biases) and outlined main challenges and priorities in the short-to-medium-term (build links with other databases, 2
improve timeliness and address, where possible, remaining issues affecting comparability across countries). 2. The Chair of the Indicator sub-group of the Social Protection Committee provided details about recipient data for a range of working-age benefits collected by ISG and illustrated uses these data for policy analysis. He emphasized the importance of timeliness but also recalled that, because of the lack of a methodological framework and its scope as an ad-hoc exercise initiated in response to the crisis, the ISG database is subject to comparability/consistency issues when used in the context of monitoring coverage levels or trends across countries. 3. OECD presented progress on the EC/OECD project. In a first part, he provided details on the latest update of the SOCR database (in 2016) including information for 2013 and 2014, response rates by type of benefit and by country as well as illustrations of analytical uses of SOCR data. The second part was devoted to methodological issues, recalling the links that were established between SOCR data on one side, and ESSPROS and ISG databases on the other, and explaining that these links helped to further validate the SOCR data and facilitate synergies between the data collection exercises, as well as the future integration of SOCR data into the European Statistical system. The presentation finished with an outline of next steps and future priorities in two specific areas: (i) harmonisation of reporting units (especially regarding minimum income benefits awarded at household level), and (ii) agreement on a methodology to account for double counting issues, including two short studies (with voluntary country participation) in 2017, quantifying the extent of double counting among a selected groups of programmes as part of the ongoing EC/OECD agreement. 4. Eurostat started his intervention with a brief summary of the structure and modules of the ESSPROS framework, followed by a detailed analysis of the possibilities and limitations of identifying working-age benefits and working-age recipients of pensions in the current version of ESSPROS. The main conclusion of this study is that working-age benefits can be identified in ESSPROS for the Unemployment and Family branches and, to a lesser extent for Sickness and Disability categories 1 ; regarding Social Assistance and Housing branches, as long as benefits are awarded at household level, further methodological clarifications would be needed in order define what a working-age household is. 2 Regarding working-age recipients, the Pension module does not include age-breakdowns; based on eligibility rules, it was concluded that it is safe to assume that recipients of disability and early retirement pensions are of working-age. The first afternoon session was dedicated to sharing country administrations experiences on benefit recipient data collection and, more specifically on the technical feasibility of detecting double counting and performing relevant adjustments for working-age benefits. The French delegate presented the various minimum income benefits existing in France (minima sociaux), their eligibility conditions (in particular those preventing claimants to benefit from more than one at the same time) and the different institutions who provide 1 It should be noted that those benefits that are identifiable as working-age in ESSPROS are precisely those included in SOCR, i.e. the earning replacement ones. 2 The most common definitions of working-age household are: (a) a household with at least one working-age individual, (b) a household without retirement-age individuals. 3
and administrate them. Finally, he presented a unified database allowing detecting multiple recipiency and therefore to produce recipients figures adjusted for double counting. The Lithuanian delegate described the two main information systems containing information about recipients of cash benefits existing in Lithuania: SODRA IS and SPIS. The former covers pensions, maternity/paternity benefits, disability benefits and unemployment benefits whereas the latter covers social support delivered at local level (municipalities). Exchanges of information about individual claimants are possible between both systems but a full integration (that would be a solution to the double counting issue) has not been undertaken yet. In a tour de table, country delegates shortly presented, for their respective countries, which data are available, as well as the technical feasibility of producing benefit recipients stocks adjusted for double counting issues. Most countries (among those who attended to the seminar) are either able to detect individuals who receive more than one benefit or are implementing integrated IT systems that will allow them to do so. The ILO representative informed the participants of an ongoing data collection on beneficiaries of employment injury benefits, inviting them to contribute to this exercise. In the concluding session OECD presented two OECD policy analyses that draw in SOCR data, namely the Faces of Joblessness project (www.oecd.org/els/soc/faces-of-joblessness.htm) which uses pseudo-coverage rates and links between recipients and expenditure data, and a project on the Cyclicality of Social Spending, which seeks to determine the responsiveness of social protection programmes to the economic cycle, both in terms of expenditure and recipients stocks. The cyclicality project requires longer-term uninterrupted time series, illustrating the necessity of maintaining a continuous SOCR data collection, at least until permanent and stable data collection of working-age benefits recipients at European level is firmly established. Conclusions and next steps Conclusions were discussed by OECD and Eurostat, who also outlined key priorities for future work. SOCR, which was initially a feasibility study followed by a pilot project, is now fully operational. After four rounds of questionnaires, recipient data exist for 8 years, 39 countries, and more than 1100 programmes. Most of the initial conceptual, technical and data-related challenges have been resolved, for instance, o The database scope is identified consistently across countries: earnings replacement programmes for all age groups; o Information is organised in a structured database following a common methodology, covering recipient stocks and sometimes inflows and average benefit amounts; o The remaining open issues are also well identified: (i) double counting; (ii) units for measuring recipient numbers. Regarding a possible extension of ESSPROS recipients module to working-age benefits o Links between SOCR and ESSPROS programmes have been established, permitting a matching of SOCR data on benefit recipients, with ESSPROS functions and schemes, and hence, with ESSPROS spending and qualitative data. 4
o For future work on integrating data on recipients of working-age benefits as part of ESSPROS, priority should be given initially to data on unemployment and minimum income benefits (social assistance), which are both relevant for policy analysis and technically identifiable in the ESSPROS framework. Programmes included in the ISG database are also included in SOCR and, in principle and if desired, SOCR could be used as the basis for recipient trends feeding into SPC and ISG work. Most participating countries are either already well prepared to provide data adjusted for double counting, or are in the process of adjusting the data collection technologies in ways that will facilitate a better identification of double-counting issues that was the case in the past (e.g., in 2013, when the previous SOCR methodological seminar took place). Most participating countries are now also in position to provide flow data for unemployment benefits (and, in some cases, also for social assistance benefits). Regarding the next steps of the project, the OECD and DG EMPL outlined the following points: The outcome of this Seminar will be distributed for information to delegates of the relevant working groups under the responsibility of the two organisations (EU Commission and OECD). The OECD will send to potentially interested countries a short document describing the purpose and specifications of the short study on double counting among working-age benefits. ESSPROS delegates and representatives who attended to the Seminar will be contacted. The OECD will send to DG EMPL and to Eurostat an updated version of the tables matching SOCR and ESSPROS entries based on qualitative information and programmes descriptions. Eurostat will consider the possibility of including an extension of ESSPROS to working-age benefits in the agenda of next meeting of the ESSPROS task force (currently planned for November 2017), and to invite the OECD to participate in these discussions as relevant, and with a view to build on data and methodological work undertaken as part of the SOCR project. 5