European insurance statistics - Eurostat project for a statistical system in the insurance sector1

Artikelförfattare: August Götzfried
Utgåva:
1, 1995
Språk: Engelska
Kategori:

44 NFT 1/1995 by August Götzfried and Nikolaus Wurm, Luxembourg2 European insurance statistics Eurostat project for a statistical system in the insurance sector1 August Götzfried 1. Introduction Step by step, the European internal market for insurance services has been created. The first and second genera- tion of EU Insurance Directives on life and non-life insurance business were followed by the third Directives3 , which provide for considerable deregulation of the European insurance sector and thus almost complete the internal market in this sector of services. In accordance with the right of establishment and the freedom to provide services, insurance undertakings have unrestricted access to the markets of the other Member States. The consumer should also have a considerably increased choice between new products and between new companies appearing on the market. 1) Eurostat is the Statistical Office of the European Communities. It is based in Luxembourg and is part of the European Commission. 2) August Götzfried:Seconded from the Bundesaufsichtsamt für das Versicherungswesen (The Federal Insurance Supervisory Office) to Eurostat (Directorate D, Unit D 4: Distributive trades, services and transport); Nikolaus Wurm: Head of Section in Eurostat, Directorate D (Business and energy statistics, research and development, and statistical methods), Unit D 4 (Distributive trades, services and transport), services statistics section. This article sets out the authors’ personal opinions and is not to be regarded as an official publication of the European Commission. 3) Third Directive on life assurance: Council Directive 92/96/EEC of 10 November 1992 on the coordination of laws, regulations and administrative provisions relating to direct life assurance and amending Directives 79/267/EEC and 90/ 619/EEC, OJ No L 360 of 9 December 1992, p. 1. — Third non-life insurance Directive: Council Directive 92/49/EEC of 18 June 1992 on the coordination of laws, regulations and administrative provisions relating to direct insurance other than life assurance and amending Directives 73/239/EEC and 88/357/EEC, OJ No L 228 of 11 August 1992, p. 1. Given the likely developments portrayed, Eurostat (Unit D 4 — business statistics in the services sector) began as early as 1991 to create a European statistical system in the insurance sector. The main objectives of the project were as follows: In order to make increasing use of liberaliza- tion and deregulation, it is essential for the market participants to have a certain amount of information. As regards the national insu- rance markets and the broad range of potential users, this information should be as homoge- neous as possible. 45 a) for data on insurance undertakings to be submitted to Eurostat annually by the competent national authorities (insurance supervisory authorities or statistical insti- tutes); b) for Eurostat to produce annual publica- tions on the basis of these data to provide an overall view of the European insu- rance markets; c) for the data collected to be used also for the national accounts. 2. Organization of the work Since 1991 Eurostat has held regular meet- ings of working parties and task forces to work on the question of European insurance statistics. Representatives of all twelve EU Member States have taken part in the working party meetings, as have representatives of other EEA countries, whereas usually only seven EU Member States and two other EEA countries were represented at task force meet- ings. Since both these groups have met several times a year, a stable, qualified group of participants has been built up and constructive, productive work has been possible. As a rule, the participating countries are represented by one delegate from the insurance supervisory authority and one delegate from the statistical institute. The European insurance associa- tions (e.g. CEA, BIPAR and ACME4) have also shown keen interest. The composition and activity of the working parties and task forces and other bilateral contacts clearly dem- onstrate the wide attention Eurostat’s work is attracting. 3. Basis for the statistical system To determine the scale and limits of the sta- tistical system to be developed, it was first necessary, following a more theoretical ap- proach, to compare the information available at national level with the requirements of the various users. The main source of information at national level is certainly the published (i.e. external) accounts and the accounts for the insurance supervisory authority (i.e. internal or confi- dential accounts). While the external accounts have been harmonized by Directive 91/674/ EEC of 19 December 1991 (on the annual accounts and consolidated accounts of insu- rance undertakings)5 at European level, the confidential returns differ between the differ- ent Member States. Further information sources at national level include the informa- tion submitted by insurance undertakings to statistical institutes or other authorities (e.g. tax authorities). This considerable pool of data is available from the national authorities at virtually no additional cost. Access to such data does not in any case involve additional burdens on the insurance undertakings. As regards the analysis of the requirements of the various data users, Eurostat’s work is to a large extent determined by the provisions of Council Decision 92/326/EEC on the develop- ment of European statistics on services6 (in particular Article 4). In accordance with that Decision, an analysis, which covered all areas of services, was carried out on the needs of the main national users. The results were totally conflicting. On the one hand, the need to limit 4) CEA: Comité Européen des Assurances; BIPAR: Bureau International des Producteurs d’Assurances & de Réassurances; ACME: Association of European Cooperative and Mutual Insurers 5) Council Directive of 19 December 1991 on the annual accounts and consolidated accounts of insurance undertakings (91/674/EEC), OJ No L 374 of 31 December 1991, p. 7. 6) Council Decision of 18 June 1992 establishing a two-year programme (1992 to 1993) for the development of European statistics on services (92/326/EEC), OJ No L 179 of 1 July 1992, p. 131. 46 the burden on undertakings was acknowl- edged. On the other, however, detailed infor- mation was seen as necessary in an internal market in the process of development. As regards the practical work on the statistical system in the insurance sector, the results of this analysis show, among other things, that the additional data to be collected from insu- rance undertakings should be kept to a mini- mum. 4. Current situation of the project As regards the main components of the project, the current situation is as follows: 4.1. Analysis of information existing at national level In accordance with Article 5 of the above- mentioned Council Decision 92/326/EEC, the Member States are obliged to ”supply the Commission with existing statistics on servic- es and any information it may require con- cerning the methodological framework used for collecting such statistics”. This should make it possible to form a detailed assessment of the information available at national level. On the basis of this, a questionnaire on the insurance sector was sent to the insurance supervisory authorities and statistical insti- tutes in July 1993. In it, detailed questions were asked on both the insurance statistics available and the methodological aspects. The questionnaire referred to life, non-life and specialist reinsurance undertakings and pen- sion funds. Eurostat has since received replies from all the EEA countries and Switzerland. These answers have been analysed and summarized in a document that was published in autumn 1994 after being discussed in the working party/task force. At previous meetings this analysis repeatedly proved very useful, when details on the availability of certain data in individual Member States were discussed. 4.2. Activity and product classification The fundamental elements of a statistical sys- tem are, on the one hand, the classification of the economic sectors and, on the other, the classification of the goods or products con- nected with these sectors. Both classifications are to be considered in the wider sense as dependent on each other. They make it possi- ble to collect statistics that refer to specific economic sectors. For this, undertakings each have to be classified in the economic sectors to which their main activity belongs. Product statistics can also be compiled; these do not necessarily classify the product against the producing undertaking. The legal basis for the statistical classifica- tion of economic activities is Council Regula- tion (EEC) No 3037/90 of 9 October 19907 . In this Regulation, ”Insurance and pension funding except compulsory social security” is broken down into three categories: Life insu- rance (66.01), Pension funding (66.02) and Non-life insurance (66.03). Specialist rein- surance undertakings are therefore mostly clas- sified under Class 66.03, since their main activities usually fall under non-life insurance. The discussions of the working party/task force rapidly came to the conclusion that the classification contained in the Regulation is not adequate. It was felt that a special class (66.04) was necessary for specialist reinsur- ance undertakings. This proposal is likely to be taken into account in a forthcoming revi- 7) Council Regulation (EEC) No 3037/90 of 9 October 1990 on the statistical classification of economic activities in the European Community, OJ No L 293 of 24 October 1990, p. 1. 8) Council Regulation (EEC) No 3696/93 of 29 October 1993 on the statistical classification of products by activity (CPA) in the European Economic Community, OJ No L 342 of 31 December 1993, p. 1. 47 sion of the Regulation. The product classification is based on Coun- cil Regulation (EEC) No 3696/93 of 29 Octo- ber 19938 . In this Regulation, insurance and pension funding services, except compulsory social security services, are also divided into three categories: Life insurance services (66.01.1), Pension funding services (66.02.1) and Non-life insurance services (66.03.1). So far, there have been no further subdivi- sions. After lengthy discussions, in which there were conflicts of opinion, the working party/task force drew up the following pro- posal for a product classification, which, how- ever, should still not be seen as final: a) Four categories are included: life insu- rance services, direct business (66.01), pension services, direct business (66.02), non-life insurance services, direct busi- ness (66.03) and reinsurance services (66.04). b) Within these categories use was made as far as possible of the sub-divisions that appear in the relevant insurance Direc- tives. Consequently, the product break- downs of direct non-life insurance to a large extent follow Article 63 of Directive 91/674/EEC (on the annual accounts and consolidated accounts of insurance under- takings). For direct life insurance, use is made of the structure of the annex of Directive 79/267/EEC (first life-insurance Directive9). Further subdivisions of the products taken from there also appeared appropriate. As mentioned above, the proposal under dis- cussion has been carefully considered but should not yet be viewed as final. The distinc- tion between direct life insurance products and direct pension products proved particu- larly difficult, especially since the products on offer in the individual Member States differ considerably. It is to be hoped that a final proposal for a product classification in the insurance sector will be drawn up at one of the next meetings, especially since the first prod- uct-related data are already to be recorded on a harmonized basis by the Member States in 1996 (for the 1995 financial year). 4.3. List of the data to be collected The list of the statistics to be collected forms the core of the project. It contains all the data to be sent by the competent national authori- ties each year to Eurostat. A preliminary draft of this list was submitted in June 1993. Fol- lowing intensive work at a number of meet- ings the list has since reached a certain level of completion. The salient points are as follows: a) For each variable, the type of insurance undertaking (life, non-life, joint or spe- cialist reinsurance undertaking) for which the aggregate figure is to be provided was laid down. b) All the variables were allocated to two priority gradings. The types of data in the first priority grade are taken directly from the relevant insurance Directives; they are to be submitted to Eurostat for the first time in 1996 for the 1995 financial year. The variables in the second priority grade are to be submitted in a second stage (the first submission to Eurostat is to take place in 2000 for the 1999 financial year). c) The list of the data to be collected com- prises the following chapters: Chapter 1: Structural data. Alongside simple variables such as the number of undertakings (including according to le- gal form), some statistics are focused on the developments expected in the internal 9) First Council Directive of 5 March 1979 on the coordination of laws, regulations and administrative provisions relating to the taking up and pursuit of the business of direct life assurance (79/267/EEC), OJ No L 63 of 13 March 1979, p. 1. 48 market (e.g. the number and situation of branches in other Member States, the geo- graphical distribution of parent compa- nies, etc.). Chapter 2: Accounting data. Here it is mainly the profit and loss account that is portrayed (technical and non-technical part). Reference here is made to Article 34 of Directive 91/674/EEC (on the annual accounts and consolidated accounts). How- ever, in addition to that, the whole techni- cal part of the profit and loss account is included here on a gross basis or with the corresponding share of reinsurance. This central section of the list of variables to be surveyed also includes data that are prima- rily intended for the national accounts (e.g. variables that refer to specific types of cost, such as commissions or wages and salaries). Chapter 3: Product- and client-related data. Product-related data refer mainly to the most important gross values of the technical part of profit and loss account. These data should be broken down ac- cording to the classification of products mentioned above, to produce a somewhat simplified profit and loss account for the insurance sector. Client-related data are on a smaller scale, since in most Member States such data are not very developed. Gross premiums and gross claims should be surveyed with reference to resident private households and resident undertak- ings only for non-life insurance undertak- ings and the non-life insurance transac- tions of joint insurance companies. Chapter 4: Data referring to the interna- tional activities of undertakings. Along- side a general geographical breakdown of transactions, partly based on Article 63, IV of Directive 91/674/EEC (on the an- nual accounts and consolidated accounts), this chapter contains a detailed geographi- cal account of business transacted in con- nection with the right of establishment and freedom to provide services, broken down according to Member State and product. The basis for this is Article 43 of Directive 92/96/EEC (Third life insurance Direc- tive) and Article 44 of Directive 92/49/ EEC (Third Directive on non-life insu- rance). In addition to the variables con- tained in those Directives, a few data were added from the Protocol on Cooperation between Insurance Supervisory Authori- ties, which was adopted with particular reference to the application of the Direc- tives on life insurance and non-life insu- rance. The data in this chapter provide valuable information on the mutual mar- ket penetration expected in the developing internal market. Eurostat is also fully prepared to take an active role in the exchange of data between the competent national authori- ties, as laid down in the above-mentioned articles of the Third Directives. Chapter 5: Employment data. The few variables listed here (such as the number of employees or the number of wage- and salary-earners) reveal scarcely any par- ticularities specific to the insurance sec- tor. Chapter 6: Investments/liabilities. The structure of investments has to be seen as an essential element of any insurance sta- tistics. The composition of investments set out in Article 6 of Directive 91/674/ EEC (on the annual accounts and consoli- dated accounts) was for the most part adopted. A few liability headings linked to investments were added (e.g. liabilities to credit institutions). For ”land and build- ings” and the ”total of other financial investments”, where possible both bal- ance sheet values referred to in Articles 45 ff. of Directive 91/674/EEC (i.e. purchase 49 values and current values) are requested. Chapter 7: Capital and reserves/technical provisions. In addition to balance sheet assets, liabilities, with their many compo- nents, are also included in the list of data to be collected. The headings listed under capital and reserves are virtually all taken from Article 6 of Directive 91/674/EEC. As regards technical provisions, only the gross amounts were taken into account, and some variables were broken down according to direct business and business accepted. Two components of technical provisions should also be broken down by product. Chapter 8: Distribution. Although this section contains very few data, it was the subject of much dispute. Out of the view of insurance undertakings number , pre- miums and expenses of dependent and independent intermediaries should be pro- vided. As regards the definitions of the terms, reference is made to both Directive 91/674/EEC (on the annual accounts and consolidated accounts) and the Commis- sion Recommendation of 18 December 1991 on insurance intermediaries10 . Chapter 9: Remaining data. The final chapter contains additional variables like number of insurance contracts at the end of the year, the number of insured persons or the number of claims made. These variables are requested for only a limited range of products. The variables, grouped together in individual chapters, that the competent national authori- ties are to submit each year to Eurostat are undoubtedly comprehensive. However, many of these data are taken from the relevant insurance Directives (in most cases Directive 91/674/EEC), which means that they have to be compiled and submitted by the insurance undertakings in any case in connection with their external or internal reporting obliga- tions. Eurostat’s statistical system should, therefore, entail only very limited additional burden on undertakings, which will doubtless differ from Member State to Member State. The list of data to be collected has not yet been finalized. However, far-reaching changes are no longer to be expected. 4.4. Methodological manual Under the above-mentioned Council Deci- sion 92/326/EEC on the development of Eu- ropean statistics on services, Eurostat was made responsible for developing the metho- dological framework (see Article 3 of the Decision). As a result, a general methodologi- cal manual was compiled that contains the variables (with the relevant methodological information) laid down for all the service sectors. In addition, a special chapter for insu- rance has been produced. The purposes of this section are: a) to provide the methodological basis for everyone involved in collecting, aggre- gating and submitting the insurance statis- tics in question; b) to relate the undertakings’ accounts to the national accounts (e.g. the calculation of data in the national accounts, such as gross value of output in the production account that was derived from the profit and loss account); c) to contain sector-specific explanations of those variables that are to be collected for the whole of the service industry. Since the overall project made considerable progress in 1993 and 1994 as regards its main components (e.g. the list of variables to be 10) Commission Recommendation of 18 December 1991 on insurance intermediaries (92/48/EEC), OJ No L 19 of 28 January 1992, p. 32. 50 collected) and the methodological manual for the insurance sector has not been revised since April 1993, in its present form the manual is unable to fulfil the above-mentioned require- ments. For instance, it does not contain all the variables in the list of data to be collected. Consequently, it is planned that a revision, provisionally final, of the insurance chapter of the methodological manual will be carried out in 1994/1995. The revised version of the manual, which should meet all the above- mentioned requirements, will probably be available by mid-1995 at the latest. 4.5. Legal basis As in other areas of European statistics on services, Eurostat’s insurance statistics are to have a legal basis. The new statistical system is to be incorporated in the Council Regula- tion on structural business statistics , which at present is only at the draft stage11. The Regu- lation is broken down into a main part con- taining the fundamental provisions affecting all sectors, Annex 1 which contains the gen- eral module for all sectors covered by the Regulation (including the data to be collected in all areas), and Annexes 2 and 3, which contain special modules for industrial and distributive trade structural statistics (also in- cluding the variables to be collected in these areas). The module for the insurance sector is to be included as Annex 4. This module, which was first presented in June 1994, contains the following important characteristics: a) The list of data to be collected will be an integral part of this Annex. b) The data to be compiled refer to non-life, life, joint and specialist reinsurance un- dertakings and Lloyd’s. Branches in the Member States of undertakings whose head offices are outside the European Union will be given the same treatment as these undertakings. c) The reference years for the statistics to be provided are, depending on the two prior- ity gradings, the calendar years 1995 and 1999. d) The data are to be submitted to Eurostat by the competent national authorities ten months after the end of the reference pe- riod (financial year) (18 months in the case of specialist reinsurance undertakings). Initial discussions on the insurance Annex have been entirely positive. Depending on how the work on the whole Regulation progresses, the approval of this Annex by the European Council is certainly unlikely before 1995. 5. Planned publications In the normal course of events, the competent national authorities should be sending harmo- nized data to Eurostat for the first time in 1996, with reference to the 1995 financial year. Originally this meant that Eurostat’s first publication on the insurance sector could appear at the earliest at the beginning of 1997. This situation appeared unsatisfactory to both Eurostat and some of the Member States. Since information was wanted on the situa- tion on the European insurance markets be- fore the start of the Internal Market , it was agreed by the working party/task force that the authorities should start sending data on a voluntary basis in 1994. The first data collec- tion is characterized by the following: a) Eurostat has sent a working party/task force-approved questionnaire to the EEA countries (including Switzerland) in au- tumn 1994, containing an extract from the list of variables to be collected. Since this first data collection can refer only to the available, unharmonized statistics, the 11) Draft Council Regulation on structural business statistics, version 25.1 of 20 June 1994. 51 structure of the questionnaire must be adapted to the individual situation of each Member State. b) Each Member State is requested to com- pile a ”country chapter”, on the basis of this questionnaire, that contains detailed comments in addition to mere facts and figures. c) The data should refer to the 1992 and 1993 financial years (situation prior to the open- ing-up of the Internal Market). Eurostat is to publish the national contribu- tions together with summary chapters that it will draw up itself in its first comprehensive publication towards the middle of 1995. This first major Eurostat report on the insurance sector should be able to provide a basic over- view of the situation of the European insu- rance markets before the Internal Market opened up. On the basis of the data collected for the following years (as from 1995), Eurostat in- tends to publish an annual insurance review that concentrates on the figures collected rather than explanations of them. It is expected that only in exceptional cases, and with the ex- press agreement of the Member States, coun- try chapters will be produced again. Eurostat hopes that the first publication, planned for 1995, will be a useful contribution to market information and meet the needs of a wide range of users. 6. OECD12 Eurostat’s new statistical system for the insu- rance sector overlaps with an existing OECD system. The data required for the OECD sys- tem are also submitted by the competent na- tional authorities of the Member States. Be- cause of the burden on the relevant authori- ties, some degree of harmonization between the two surveys is necessary. 12) OECD: Organization for Economic Cooperation and Development, Paris. Although the two systems to some extent concentrate on different points, it appears possible to determine a common volume of data requirements and to coordinate the trans- mission of data accordingly. Both interna- tional institutions have recognized the need to cooperate. Practical procedures that will re- sult in reducing the burden on the national authorities must follow now that Eurostat’s system has reached a sufficient level of devel- opment. 7. Outlook Eurostat’s project for European insurance sta- tistics made considerable progress in 1993 and 1994. However, it has not yet reached full fruition. On the basis of the 1994 survey, Eurostat is in a position to produce its first comprehen- sive publication on the insurance sector in 1995. Annual publications will follow. Provided it has the resources, Eurostat in- tends to press on with other projects concern- ing other financial services areas. It could, for instance, set up comparable statistical sys- tems for pension funds, banks or investment funds. Demand in such sectors is doubtless just as great as in the insurance sector.