EUROPEAN COMMISSION EUROSTAT Directorate F: Social statistics Unit F-5: Education, health and social protection Luxembourg, 14 April 2016 DOC SP-2016-07-Annex 7 https://circabc.europa.eu/w/browse/70400e55-173f-433f-93adc8315904a11e ESSPROS Expert Group (1486): https://circabc.europa.eu/w/browse/96adef83-8ee4-4c76-8c37-83502f932ec9 Working Group Social Protection 27-28 April 2016 Comparing ESSPROS and MISSOC qualitative data Item 7 of the Agenda 1
Comparing ESSPROS and MISSOC qualitative data 1. Introduction The ESSPROS qualitative information (referred to as QI in this document) is collected annually by regulation 1 according to the specifications laid out in Appendix II of the ESSPROS manual and user guidelines 2. As noted in 1 of Appendix II, the QI is intended to provide: 1. in-depth information on social protection schemes 2. the means to evaluate the classification of schemes and benefits applied 3. a clear basis for footnotes in publications and for the ESSPROS datasets in the Eurostat database 4. a means to respond to questions from users on the data by scheme 5. the basis for the validation of ESSPROS data (Core System and modules on pension beneficiaries and net social protection benefits) The importance of the QI is reinforced by the recent agreement to openly publish data by scheme, which can only be understood through reference to the detailed descriptive information available in the QI. It is therefore crucial that the QI is clear, comprehensive and of consistent quality. However, at the current time the information provided by countries varies in quality and there are no mechanisms in place to conduct a meaningful validation of the QI. The Task Force meeting held in November 2015 acknowledged that the qualitative information needs to be improved but noted that this implies a major burden on the providers who are mainly statisticians unfamiliar with legislation and thus find it difficult to compile the required information. The Task Force also recommended efforts to consolidate information from similar data collections such as MISSOC in order to avoid duplication of effort. The Mutual Information System on Social Protection (MISSOC) database collects detailed qualitative information about national social protection systems and the benefits they provide. Although the objectives of ESSPROS and MISSOC differ, there are significant overlaps between the qualitative information collected in both. The database is managed by the MISSOC-Secretariat on behalf of DG EMPL and comprises data collected from official representatives in either national ministries or other institutions responsible for social protection. Establishing links between ESSPROS and MISSOC could have several benefits. In the short-term, MISSOC could be used as a source for validating and complementing the ESSPROS qualitative information. In the longer-term, the possibility of a joint exercise with MISSOC could be investigated in order to avoid duplication of effort and reduce the burden on data providers. In order to understand how these two databases relate, it is essential to first identify their similarities and differences. This document presents a general overview of the qualitative information in ESSPROS and MISSOC in order to identify the broad links between the two. This represents a first step towards a more in-depth assessment focusing on the family/children function, and possibly the housing function, that will be conducted and presented to the ESSPROS Task Force later in the year. This more detailed exercise will complement ongoing work to improve the 1 Regulation (EC) No 458/2007 of the European Parliament and of the Council (http://eur-lex.europa.eu/legalcontent/en/txt/?qid=1417170413043&uri=celex:32007r0458) and Commission regulation (EC) No 1322/2007 (http://eurlex.europa.eu/legal-content/en/txt/?qid=1459947580539&uri=celex:32007r1322) 2 http://ec.europa.eu/eurostat/web/products-manuals-and-guidelines/-/ks-ra-12-014 2
dissemination of ESSPROS data, including the development of a Statistics Explained article on the family/children function (see DOC SP-2016-08-Annex). 2. Qualitative information in ESSPROS and MISSOC The general comparison of ESSPROS and MISSOC reviewed the following characteristics of each dataset: - Geographical coverage - Scope of social protection covered - Structure and content of the data on social protection - Reference period and update schedule Analysis is based on the 2012 ESSPROS Manual and user guidelines 2, MISSOC Correspondents Guide 3 and the cross-cutting introductions 4 (published by the MISSOC-Secretariat) which are all publicly available. 2.1 Geographical coverage ESSPROS covers a total of 33 countries: the 28 EU Member States plus Iceland, Norway, Switzerland, Serbia and Turkey. The MISSOC database covers all of these except Serbia and Turkey (see Table 1). However, the MISSCEO database 5, managed by the Council of Europe, collects data according to the MISSOC methodology for a number of countries that are not part of the MISSOC network, including both Serbia and Turkey. MISSOC/MISSCEO data are therefore available for all countries covered in ESSPROS. Table 1 Geographical coverage of databases Countries ESSPROS MISSOC MISSCEO EU Member States EFTA countries: IS, NO and CH Candidate countries: RS and TR 2.2 Scope of social protection covered The scope of ESSPROS is broadly delimited by the definition of social protection given in 16 of part 1 of the ESSPROS manual 2 : all interventions from public or private bodies intended to relieve households and individuals of the burden of a defined set of risks or needs, provided that there is neither a simultaneous reciprocal nor an individual arrangement involved. The risks or needs that may give rise to social protection are referred to in ESSPROS as functions and are listed in 16 as sickness/health care, disability, old age, survivors, family/children, unemployment, housing, and social exclusion not elsewhere classified. These functions are defined in detail in part 2 of the ESSPROS manual which outlines the types of benefits that can be included under each function. MISSOC does not provide an explicit definition of social protection so that its scope is implicitly delimited by the different types of social protection covered by eleven of the twelve comparative 3 http://missoc.org/missoc/informationbase/comparativetables/correspondentsguide_en.pdf 4 http://missoc.org/missoc/informationbase/comparativetables/crosscuttingintro/introduction.htm 5 http://www.missceo.coe.int/ 3
tables 6 that form the basis for the MISSOC data. The final table describes the financing of social protection in each country. The broad types of social protection covered by MISSOC are health care, sickness cash benefits, maternity/paternity, invalidity, old-age, survivors, accidents at work and occupational diseases, family benefits, unemployment, guaranteed minimum resources and longterm care. The MISSOC Correspondents Guide provides detailed guidance on the types of benefits to be included in each table. The scope of social protection covered in both data collections can thus be compared by establishing a correspondence between the comparative tables in MISSOC and the ESSPROS functions (see Table 2 with further details in Annex 1). Table 2 Correspondence between ESSPROS functions and MISSOC comparative tables ESSPROS Function Sickness/Health care Disability Old age Survivors Family/children Unemployment Housing Social exclusion not elsewhere classified MISSOC Comparative table Sickness Cash benefits Health Care Long-term care Accidents at work and occupational diseases Invalidity Old age Survivors Family Benefits Maternity/paternity Unemployment Guaranteed minimum resources Overall, the mapping shows that there are no major differences in the overall scope of social protection covered by ESSPROS and MISSOC. However, this is only a first level mapping and in-depth comparisons of specific tables and functions are required in order to identify possible differences at a more detailed level (i.e. whether specific types of social protection are covered in one but not in the other). 2.3 Structure and content of data on social protection The qualitative information in ESSPROS is organised in a two dimensional system which breaks the information down by scheme and by type of information (about the scheme or the benefits it provides). The specifications of the information required are laid out in Appendix II of the ESSPROS manual. Ten items of information are collected about each scheme and five for each benefit provided by each scheme. There is one set of information per scheme with the information on each benefit reported according to the classification used in the Core system. Further, as in the ESSPROS quantitative data, if a scheme provides two separate benefits of the same type then the qualitative information for both is reported together under the same classification. The following information is required for each benefit: 1. English and original name of benefit 2. Year of introducing or abolishing the benefit 6 http://missoc.org/missoc/informationbase/comparativetables/missocdatabase/comparativetablesearch.jsp 4
3. General conditions for granting the benefit 4. The categories of beneficiaries 5. Taxation of benefit As this information is all reported together in a single cell of the data collection template it is difficult to automatically check that all items have been completed and manual validation is not undertaken sistematically. At the same time, it is recognised that in some countries it is not feasible to provide all this information for all benefits. Consequently, the completeness and quality of the information varies between benefits and between countries. In contrast, the qualitative information in MISSOC are organised according to a one dimensional system which breaks the information down by type of information but not by scheme. Each comparative table has a number of chapters, which may be further divided into different topics (see Figure 1). Although there are some elements common to most tables, the subjects of the chapters and topics are table specific and tailored to the type of social protection covered. Further, these may cover either general information pertaining to a particular type of social protection or information on specific benefits. Figure 1 Structure of the MISSOC data Comparative tables Chapters Topics Note: See http://missoc.org/missoc/informationbase/comparativetables/missocdatabase/comparativetablesearch.jsp Table 3 lists the 33 different chapters used across the eleven comparative tables focusing on specific forms of social protection. Applicable statutory basis is the only chapter to be used across all tables and there are only five others that are used in more than two thirds of tables: Basic principles, Taxation and social contributions/taxation (used in 10 tables), Conditions, Field of Application (used in 9 tables) and Benefits (used in 8 tables). Note, however, that in some cases differences arise only because of the level at which information is collected. For example, nine comparative tables include the chapter Conditions in the table on unemployment, Conditions appears as topic under the separate chapters on Total unemployment, Partial/temporary unemployment and Benefits for older unemployed. 5
Chapter Sickness Cash benefits Health Care Long-term care Accidents at work and occupational diseases Invalidity Old age Survivors Family Benefits Maternity/ Paternity Unemployment Guaranteed minimum resources Total Table 3 Use of chapters in MISSOC comparative tables 1 Applicable statutory basis x x x x x x x x x x x 11 2 Assessment of claims x 1 3 Basic principles x x x x x x x x x x 10 4 Benefits x x x x x x x x 8 5 Benefits for informal carers x 1 6 Benefits for older x 1 unemployed 7 Benefits for young x 1 unemployed 8 Benefits promoting labour x 1 market integration 9 Cash benefits x 1 10 Child benefit x 1 11 Child care allowances x 1 12 Child-raising allowances x 1 13 Conditions x x x x x x x x x 9 14 Cumulation with earnings x x x x 4 from work 15 Cumulation with other x x x 3 social security benefits 16 Entitled persons x 1 17 Entitled persons / x 1 beneficiaries 18 Exemptions from x x x 3 compulsory insurance 19 Field of application x x x x x x x x x 9 20 Housing and heating x 1 allowances 21 Indexation x x x x x x 6 22 Organisation x x 2 23 Other benefits x 1 24 Partial retirement x 1 25 Partial/temporary x 1 unemployment 26 Recovery of benefits x 1 27 Return to active life x 1 28 Risk covered x x x 3 29 Sanctions x 1 30 Special rights in health care x 1 31 Taxation and social x x x x x x x x x x 10 contributions / Taxation 32 Total unemployment x 1 33 Waiting period x 1 Total 7 6 12 8 12 10 8 7 6 13 10 99 6
The MISSOC structure means that information about different benefits of a particular type should always appear in the same comparative table. In ESSPROS, however, the information could be split between schemes. The differences in the structure of the qualitative data collected in ESSPROS and MISSOC mean that a universal correspondence between the different items of information collected for all benefits cannot be established. It should, however, be possible, on a case by case basis, to map each ESSPROS detailed benefit type to a MISSOC comparative table and then identify a correspondence between the information available for that benefit in both datasets. Nevertheless, comparing the items of information that the ESSPROS manual requires to describe each benefit with the most commonly used chapters in the MISSOC comparative tables (identified in Table 3) suggests that all the information collected in ESSPROS should also be available in MISSOC (see Table 4). Further, MISSOC includes more information about specific benefits (amounts, calculation methods, etc.) than the ESSPROS QI. Table 4 Comparison of benefit-specific information required in ESSPROS QI and the most commonly used chapters in MISSOC ESSPROS - Information in description of benefit English and original name of benefit Year of introducing or abolishing the benefit General conditions for granting the benefit The categories of beneficiaries Taxation of benefit (Only requested for the scheme) (Not requested) (Only requested for the scheme) MISSOC - Commonly used chapters (Names of benefits are reported in several different chapters whenever they refer to a specific benefit) (Date of introduction/abolition may be identified using the time series of MISSOC data) Conditions Field of Application Taxation and social contributions/taxation Basic principles Benefits (amounts, calculation methods, etc.) Applicable statutory basis While MISSOC does not specifically collect information about each scheme, which is a concept unique to ESSPROS, some information of similar nature is collected. For example, information collected in the MISSOC chapter on Basic principles includes elements covered by the items on Classifications of schemes and Scope in ESSPROS. If correspondence with MISSOC can be established for all the benefits provided by a scheme in ESSPROS then it should be possible to compare some of the scheme level information. Also relevant here is the separate MISSOC table on financing. There is clearly a considerable overlap in the content of the information collected in relation to social protection. MISSOC (in theory) contains all the information about specific benefits that is collected in ESSPROS. Information about schemes that is gathered in ESSPROS is partially available in MISSOC but in a way that is not directly linked to the scheme concept used in ESSPROS. Part of this is collected at the general level (e.g. information on financing is collected in a separate comparative table) and part of it is collected at the level of specific types of social protection (e.g. information on basic principles and applicable statutory basis). This is liable to complicate, but not exclude, comparison of this data. 7
2.4 Reference period and update schedule The ESSPROS QI is updated annually and disseminated with a 22-month delay while the MISSOC data are updated biannually with a 6-month delay and the MISSCEO data are updated annually with a 12- month delay (see Table 5). This means that the MISSOC/MISSCEO data are always available before the QI. Indeed, they are available before the transmission deadline for QI (30 June of the year N + 2) so they can potentially be used as a means to validate ESSPROS data before release. Table 5 Database update schedule Database Latest Data Update Frequency Delay ESSPROS QI 2013 Annual 22 months MISSOC 1 July 2015 Biannual 6 months MISSCEO 1 January 2015 Annual 12 months Note: Delay refers to the time between the reference period and the dissemination of the data. However, there is one very important difference between ESSPROS and MISSOC/MISSCEO: the reference period. ESSPROS only publishes one set of QI per country which covers a number of reference years (specified in the introduction page of the qualitative questionnaire) that may vary between countries. Item 9 of the general description of each scheme should provide information on the history of the scheme (i.e. changes through time) but this is not always completed with any detail. MISSOC/MISSCEO data, on the other hand, provide a snapshot of the social protection system at a specific point of time and a full time series of biannual data from 2004 to 2015 is available online 7. Note that the structure of the MISSOC data does vary over time and could impact the possibility to compare historical data. A decision would therefore have to be made in relation to which reference period in MISSOC will be used when making a comparison (either referring to the 1 January or 1 July). This issue does not extend to the MISSCEO database (containing data for Serbia and Turkey) as it is updated on an annual rather than biannual basis. Another implication of this is that the ability to compare MISSOC with ESSPROS will be dependent on whether information on when benefits were introduced and abolished has been provided in the description of benefits in ESSPROS. 3. Issues for consideration This section focuses on issues specific to family/children and housing benefits that will need to be considered when attempting to compare data between ESSPROS and MISSOC. 3.1 Family/children benefits There are two key issues specific to family/children benefits. The first, as mentioned in section 2.2, is that while ESSPROS reports all such benefits under one function, MISSOC reports them in two separate comparative tables. The ESSPROS Manual defines the family/children function under 52 in part 2 of the ESSPROS manual as one that includes benefits that: provide financial support to households for bringing up children; provide financial assistance to people who support relatives other than children; 7 Note that data for 2004 and 2005 are only available for one reference point in the year and this point may deviate from those available in other years. 8
provide social services specifically designed to assist and protect the family, particularly children. In MISSOC family/child benefits are split between two comparative tables. One on family benefits and another on maternity/paternity (see Annex 1). That on family benefits includes: Child benefits: benefits provided to meet the additional expenses incurred by households bringing up children Paid parental leave or child-raising allowances: paid parental leave or child-raising allowances provided as a replacement of the income for parents who take time off work to care for their children these benefits start after maternity/paternity entitlements end. Child-care allowances: provided to cover the cost of childcare (i.e. paid to working parents whose children are care for while they are at work). Other benefits: birth/adoption grants, allowances for lone parents, special allowances for children with disabilities, periodic payments in cases where alimony is not paid. Meanwhile, that on maternity/paternity includes only benefits provided in respect to maternity/paternity. The two tables are intended to be mutually exclusive. A second key issue is the way in which family supplements are reported. A wide range of cash benefits include supplementary payments for beneficiaries with dependants. The treatment of these is described in 43F, 74D and 85D in part 2 of the ESSPROS manual which note that while supplements for dependants should, in principle, be separated out, this is difficult in practice so supplements should be reported along with the main benefit (details are collected as part of the optional data). The qualitative information for the main benefit should be consistent with the quantitative data and therefore provide details of any such supplementary payments. However, this is not always the case and there is no guidance in the ESSPROS manual on this point. The comparative tables in MISSOC, on the other hand, include dedicated items related to supplementary payments. Such items can be found in the tables on invalidity, old-age, accidents at work and occupational diseases, and unemployment. For example, the MISSOC correspondents guide 8 gives the following guidance on the family supplements item in the unemployment table: It is very important to note that this section refers to supplements i.e. additional payments on top of those mentioned above, granted to unemployed persons with dependants: If there are no such supplements, please write No family supplements. If there are no separate supplements but the family composition is taken into account in the main benefit calculation, please write No family supplements. Family composition taken into account in Rates of the benefits. If you have separate family supplements, please describe them in full. MISSOC could, therefore, be a useful source of information to complement ESSPROS QI in relation to family supplements. 8 http://missoc.org/missoc/informationbase/comparativetables/correspondentsguide_en.pdf 9
3.2 Housing benefits In ESSPROS housing benefits are reported under a dedicated housing function which the ESSPROS manual defines under 75 in part 2 of the manual as including interventions by public authorities to help households meet the cost of housing. All benefits reported within this function must be meanstested and are all classified, by convention, as benefits in kind even if they are paid in cash. Further, housing supplements, paid in cash to top up to other benefits, should, as per 76 in part 2 be reported separately from the main benefit and recorded in the housing function (i.e. opposite treatment to that recommended for family supplements described above). In MISSOC there is no comparative table specifically for housing. The possibility of including a separate comparative table has been discussed 9, but a table has not yet been agreed or added. For most countries information on housing benefits is reported under the housing and heating allowances chapter of the guaranteed minimum resources table (see Annex 1). Further, unlike family supplements, MISSOC does not have specific items to collect information on housing supplements. It is known that some information appears under the tables on invalidity, old age, survivors, accidents at work and occupational diseases, family benefits, and unemployment but there does not appear to be any consistent treatment across countries 9. 4. Conclusions Comparison of the coverage and organisation of qualitative information in ESSPROS and MISSOC shows that there are significant overlaps in the data collected by the two databases but also structural differences that present difficulties for directly comparing their content: 2.1 Geographical coverage: MISSOC data are available for all countries covered by ESSPROS except Serbia and Turkey, but these two countries are covered by the MISSCEO database which collects qualitative data on social protection systems according to the MISSOC methodology. Scope of social protection covered: A mapping of ESSPROS functions and MISSOC comparative tables suggests no major differences in the scope of social protection covered in both databases. However, further in-depth analysis is needed in order to identify differences at a more detailed level. Structure and content of data on social protection: The way in which data are organised in the two databases is quite different so that detailed comparison is not straightforward. ESSPROS uses a two dimensional system which breaks down the information by scheme and by type of information whilst MISSOC applies a one dimensional system by type of information with no breakdown by scheme. This means that whilst scheme specific information in ESSPROS may be available in MISSOC, it may not be possible to link it directly to the scheme concept used in ESSPROS (which may result in different classifications between countries). On the other hand, the information collected to describe each benefit in the ESSPROS QI should, in theory, be available in MISSOC and it is at this level that possible links might be established. Reference period and update schedule: MISSOC/MISSCEO data are always published with a shorter delay than ESSPROS (6/12-months compared to 22-months) and MISSOC is updated more regularly (biannually rather than annually) which means that MISSOC data could 9 What do we know about housing and related benefits in the European Union Prepared by Jonathan Bradshaw, Professor of Social Policy University of York and Durham University, with a contribution from Terry Ward, MISSOC Secretariat, and Stefán Ólafsson, Professor of Sociology at the University of Iceland 10
potentially be used to validate ESSPROS data prior to dissemination. However, there is an important difference in the reference periods used. MISSOC/MISSCEO provides a snapshot a specific point in time, ESSPROS QI covers a period of time spanning several years. The next phase of work planned for this exercise will involve a more in-depth comparison of the qualitative information available in both databases for a specific sub-set of benefits for a selection of countries. This will focus initially on family/children benefits, and possibly later on housing benefits, with the aim of providing a more in-depth illustration of how links between the two data sets can be established and any limitations that may exist (i.e. whether or not both data collections cover the same detailed benefits). Issues identified in this document will need to be taken into account. The Working Group is invited to comment on the document and in particular on its conclusions. Does the WG agree on further work on this comparison with particular reference to the two functions identified (family/children and housing)? 11
Annex 1 Correspondence between ESSPROS functions and MISSOC comparative tables ESSPROS Manual MISSOC Guide Comments Function Brief description 10 Category Brief description 11 Income maintenance and support in cash in connection with Sickness Cash Benefits provided in cash in case of sickness (excluding benefits physical or mental illness, excluding disability. Health care benefits in case of maternity). intended to maintain, restore or improve the health of the Health Care Health care systems (benefits in kind in case of sickness) people protected irrespective of the origin of the disorder. Sickness/ Health care Disability Old age Survivors Family/ Children Unemployment Housing Social exclusion n.e.c Income maintenance and support in cash or kind (except health care) in connection with the inability of physically or mentally disabled people to engage in economic and social activities. Income maintenance and support in cash or kind (except health care) in connection with old age. Income maintenance and support in cash or kind in connection with the death of a family member. Support in cash or kind (except health care) in connection with the costs of pregnancy, childbirth and adoption, bringing up children and caring for other family members. Long-term care Accidents at work and occupational diseases Invalidity Old age Survivors Family Benefits Maternity/ Paternity Income maintenance and support in cash or kind in connection Unemployment with unemployment. Help towards the cost of housing. Guaranteed Benefits in cash or kind (except health care) specifically minimum intended to combat social exclusion where they are not resources covered by one of the other functions. Benefits provided (in cash or kind) to people who, as a result of frailty related to old age or mental or physical disability, have a reduced degree of functional capacity and are dependent on help over an extended period of time. Benefits provided in relation to injuries and diseases caused by the employment environment; sickness, invalidity, and death. Invalidity benefits to support those who through long-term sickness or disability are unable to undertake paid employment. Old-age includes basic first-pillar pension systems (countries to include statutory second pillar or supplementary pension schemes if they are also mandatory). Benefit paid by a pension plan/fund to the designated beneficiary of an employee (usually a spouse or partner), upon the death of that employee. Benefits (in kind or cash) provided to families in connection with raising children. Social protection systems in respect of maternity/paternity. The benefits included under the Sickness/Health care function in ESSPROS are broken down into four tables in MISSOC. Note: The Accidents at work and occupational diseases does not apply to countries which do not make a distinction between work non-work related benefits. This table also covers invalidity and death benefits whereas in ESSPROS the corresponding function only covers benefits related to health care and sickness. No major differences. No major differences. No major differences. Benefits provided in cash or kind in connection with No major differences. unemployment. Table includes benefits related to help towards housing costs and other types of benefits relating to social exclusion. The benefits included under Family/children function in ESSPROS are divided into two tables in MISSOC. Both tables apply mutually exclusive approach i.e. exclude benefits reported in the other. The benefits reported under the two functions in ESSPROS are included in a single table in MISSOC. The table also includes other types of benefits considered as a minimum source i.e. those related to unemployment and old-age. 10 110 of section 1 in the ESSPROS Manual. 11 Based on information provided in the MISSOC correspondents Guide and cross-cutting introductions on each table. 12
13