Draft, 29 December 2015 Annex IV A PROPOSAL FOR INTERNATIONAL RECOMMENDATIONS ON REFUGEE STATISTICS (IRRS) 1
INTRODUCTION At the 46 th session of the UN Statistical Commission (New York, 3-6 March, 2015), Statistics Norway and the Office of the United Nations High Commissioner for Refugees (UNHCR) presented a joint report on statistics on refugees and internally displaced persons (E/CN.3/2015/9) 1 which included a number of concrete suggestions. One was the organisation of an international conference on refugee statistics, bringing together international organizations and experts from national statistical offices. Its purpose would be to explore how best to incorporate statistics on refugees and displaced persons into national statistical systems and to develop a set of recommendations for improving these statistics. These could eventually be incorporated into a handbook on statistics on refugees and internally displaced persons as a practical guide for the collection, analysis and dissemination of these statistics, taking into consideration existing methods and initiatives. Recognizing the importance of the topic, the Government of the Republic of Turkey, through its Turkish Statistical Institute (TurkStat), offered to host such a conference, in collaboration with Statistics Norway, Eurostat, the Statistical Office of the European Free Trade Association (EFTA) and the UNHCR. The Statistical Commission requested the organizers of the international conference to report back to its 47 th session (March 2016) on the outcome of the conference and on the next steps to improve statistics on refugees and internally displaced persons. Given the complexities of dealing with internally displaced persons, it was decided that the international conference shall focus on refugee statistics, and that the handbook would deal exclusively with refugee statistics. UNHCR, in collaboration with the other four organizing agencies developed a draft proposal for an outline of the envisaged handbook 2, which participants were invited to study before the conference. The conference was held from 7 to 9 October in Antalya, Turkey. It was attended by representatives of national statistical offices, offices dealing with the admission of foreigners and refugees, research institutions, independent experts, United Nations agencies, the World Bank and the European Asylum Support Office. During the conference the proposal for a handbook on refugee statistics, its scope and content were discussed. Participants fully supported the development and production of such a handbook but suggested to change its title to International Recommendations on 1 See: http://unstats.un.org/unsd/statcom/doc15/2015-9-refugeestats-e.pdf 2 See: http://www.efta.int/sites/default/files/documents/statistics/training/outline%20handbook%20on%20refugee %20statistics_for%20circulation_v0_final.pdf 2
Refugee Statistics (IRRS). Deliberations at the conference led to a number of suggestions for modifications with regard to the handbook s content and scope. The UN Statistical Commission in 1976 had adopted a set of recommendations on international migration which were revised in 1998 entitled Recommendations on Statistics of International Migration, Revision 1. 3 This document provides methodological and technical recommendations on dealing with international migration and how the stock and flow of migrants should be tabulated. It includes references to asylum seekers, refugees and related categories under the heading of foreigners admitted into a country for humanitarian reasons. The proposed IRRS will take these recommendations into consideration, as deemed applicable and relevant. This document is an adapted version of the initial document A proposal for an international handbook on refugee statistics which was subject of discussion at the international conference on refugee statistics in Antalya, Turkey (7-9 October, 2015). PRELIMINARY CONSIDERATIONS The United Nations High Commissioner for Refugees has been mandated by the UN General Assembly to provide protection to all persons outside their country of origin for reasons of feared persecution, conflict, generalized violence, or other circumstances that have seriously disturbed public order. 4 In reality, the mandate of the High Commissioner and his Office includes refugees, asylum-seekers, returnees, stateless persons, and internally displaced persons. This proposal will deal exclusively with the collection and analysis of statistics on refugees. Given the state of knowledge about the collection of statistical data on refugees, it is proposed to prepare a set of recommendations that contains advice, through examples, about how to collect statistical information on refugees, entitled International Recommendations on Refugee Statistics (IRRS). ISSUES TO CONSIDER 1. Legal framework Refugee status is an internationally recognized status according to the provisions of the 1951 Convention relating to the Status of Refugees, the 1967 Protocol relating to the Status of Refugees and subsequent legislation and regulation. UNHCR has issued several 3 See: http://unstats.un.org/unsd/publication/seriesm/seriesm_58rev1e.pdf 4 See: Note on the Mandate of the High Commissioner for Refugees and his Office, p. 3, referring to the United Nations document A/AC.96/830, 7 September 1994, paras. 8, 10-11, 31-32. See: http://www.unhcr.org/526a22cb6.html. 3
handbooks and guidelines to assist countries in determining the status of refugees. These principles and guidelines form the basis of an overall frame of reference in which the statistical collection of data on refugees will take place. In addition, regional regulations in Africa 5 and the Americas 6 may need to be considered. As the collection of data on refugees generally takes place within the territory of a nation, consideration will also need to be given to national or supranational regulations and categories of refugees, based either on the classification used by an institution that deals with migration, asylum, or the visa regime, or on the visa categories that apply to refugees. To enable international comparability it is necessary to transpose national categories into standard international, hence UNHCR categories. 2. Population categories to be considered The objective of this exercise is to enhance the collection of data on refugees. Refugee status is obtained following an established administrative process that asylum seekers undergo in a particular country. Obtaining refugee status is only one of the possible outcomes of the individual asylum application process; other outcomes being the granting of complementary forms of protection (e.g. subsidiary protection, humanitarian status etc.), the rejection of such application or the administrative closure of an asylum application. 7 This categorization of outcomes may not be the same for all countries. For purposes of the IRRS, the term refugee shall include asylum seekers who enjoy refugee status or a complementary form of protection as well as those that have been granted protection on humanitarian grounds following the lodging of a formal asylum application. It is recommended that these categories are clearly distinguished within the refugee statistics. A number of countries also grant non-asylum seekers leave to enter on the basis of humanitarian grounds. These persons should be clearly distinguished from refugees, who are persons who have been granted leave to enter under the provisions of the 1951 Convention relating to the Status of Refugees, the 1967 Protocol relating to the Status of Refugees and subsequent legislation and regulation. Given that the application process is country-specific, one may wish to consider obtaining statistics on potential asylum-seekers the number of irregular migrants 8 or illegal arrivals 9 in a country. In addition, some countries may wish to identify among their foreign-born 5 1969 OAU Convention Governing the Specific Aspects of Refugee Problems in Africa. See: http://www.unhcr.org/45dc1a682.html 6 Cartagena Declaration 1984; See: http://www.unhcr.org/45dc19084.html 7 This may result from the death of the applicant, no-show for interview, withdrawal of the application, abandonment of the claim, or the determination that another country is responsible for the claim ( Dublin III procedure), among other factors. 8 See: Vollmer, B., Briefing: Irregular Migration in the UK: Definitions, pathways and scale. University of Oxford, Oxford, 2011. See: http://migrationobservatory.ox.ac.uk/briefings/irregular-migration-ukdefinitions-pathways-and-scale. 9 Australia uses the term Illegal Maritime Arrivals (IMAs) for a special category of individuals who try to enter the country illegally. See: https://www.border.gov.au/trav/refu/illegal-maritime-arrivals. 4
citizens those who entered the country as refugees and to compare their socio-economic characteristics with those of other citizens and residents. 3. Sources of statistical information There are three standard sources of statistical information of relevance: administrative records, censuses, and surveys. Each has strengths and limitations. It should be noted that in some contexts, data sources of different types may be combined to meet particular statistical needs. For example, data taken from administrative sources may be supplemented with modelled data taken from sample surveys. Administrative records are the by-product of administrative procedures and may need extensive re-working to yield information that can be used for statistical purposes. A growing number of countries are aware of this limitation and try to take into account the statistical needs/reporting at the stage of designing new administrative systems. The advantage of these records is that they could provide information required for current (stock) and continuous (flow) statistical information. A major limitation is however missing information on the status of the actual stay of individuals in the absence of a necessity to de-register in case of emigration or departure. Population censuses are the only of these sources that, ideally, covers the total population of a specific territory. Such an exercise is carried out periodically but could still be used to assess the completeness of statistics derived from administrative records. A population census can also be used to measure the number of specific population categories for whom no current or continuous information is available. In case of refugees, it could be used to estimate their numbers at the census reference period or to measure certain relevant characteristics of refugees that could be compared with those of other population groups (for instance, migrants) and the total population. The UN Economic Commission for Europe (UNECE) has been promoting the use of censuses to measure migration flows, especially during the 2010 round of population censuses. Their dedicated website 10 contains training materials that could be of use for countries that want to use a census to collect data on migration; with very few modifications, these can also be used to obtain data on refugees. The practical guide that UNECE prepared for the countries of Eastern Europe and Central Asia 11 may also be beneficial for countries of other regions. With regard to population censuses, countries may wish to consult the Conference of European Statisticians Recommendations on population and housing censuses 12, or the 10 See: http://www.unece.org/stats/video/migration.html. 11 See: http://www.unece.org/fileadmin/dam/stats/publications/international_migration_practical_guide_eng.pdf. 12 See: http://www.unece.org/fileadmin/dam/stats/publications/ces_2010_census_recommendations_english.pdf Edition for the 2020 round of censuses forthcoming. 5
guidelines and handbooks published by the UN Statistics Division as part of its World Population and Housing Census Programme. 13 Population or household surveys, ideally, use scientific sample procedures (probability sampling) to select part of the population within a territory. The aim is to obtain detailed information about the socio-economic and cultural characteristics of the population in general and, occasionally, about specific sub-populations, including migrants and refugees. Censuses and surveys have long-established and universally accepted standards, although recently alternatives to the census methodology (now called traditional census ) have been proposed. Of these three sources of statistical information, only administrative records are reliable sources for both current and continuous information about the number of refugees and related categories. Censuses and household surveys are reliable sources for stock information, but, with additional efforts, could also be used to estimate flow information. Recently a number of new developments on the basis of the availability of large bits of information through the electronic social media have entered into the statistical discourse. The concept of Big Data is now being considered by some as an additional source of statistical information. However, a number of methodological, technical and confidentiality issues still need to be determined with regard to what the status of the so-called Big Data is. The IRRS is flexible and if and when Big Data comply with the internationally accepted scientific criteria, articles or studies dealing with refugee issues based on Big Data could be accommodated in the appropriate chapter of the IRRS. 4. National population and refugee registers On an international basis, dedicated refugee registers do not seem to exist. Given that countries determine the status of a person as a refugee, information on the characteristics of those individuals under consideration is available with the government agencies that determine their status or, in the absence of efficient national asylum procedures, with UNHCR. Ideally, there should be a register in the public administration that contains information about individuals who have been accorded refugee status. Ideally, countries should have a comprehensive integrated population register that contains information about all individuals residing in their territory and which could be used for administrative purposes as well as policy development and assessment. These registers should include information for all categories of the resident population, including migrants and refugees. A number of European countries have such a register, including Norway. 14 In Norway, the Directorate of Taxes maintains a Central Population Register, the most 13 14 See: http://unstats.un.org/unsd/demographic/sources/census/census3.htm For more information, see: http://www.ssb.no/en/befolkning/statistikker/innvgrunn. 6
comprehensive population database in the country. This is shared with Statistics Norway, which maintains it as a national population database under the name BeReg. Statistics Norway receives data on migration including regarding asylum-seekers, refugees, and other relevant categories as well as the reasons for immigration from the Directorate of Immigration. In turn, these data are integrated into the BeReg database, allowing for the production of statistics of all population categories resident in Norway. Denmark and Sweden have similar integrated national population registers but the treatment of migrants and refugees is not the same as in Norway. A number of other countries in Europe, and perhaps some developing countries, also have population registers, but these do not have the option to identify migrants and refugees as separate categories. Hence, it is recommended that these countries adjust their national population systems to include migrants and refugees. Countries that do not have population registers could create a refugee register, either as a stand-alone database of refugees only or as an integrated migration register that includes all categories of migrants and refugees. The use of unique identifiers in relevant national population registers would allow for a more efficient identification of persons in case of non-integrated systems. The majority of developing countries do not have a national population register. For those countries with considerable numbers of refugees, it is recommended that they create a refugee register. Again, this could be either a stand-alone register consisting solely of refugees or an integrated migration register that includes all categories of migrants and refugees. All countries, meanwhile, could take advantage of the experiences of Denmark, Norway, and Sweden in the construction of population registers and the way they deal with data on migrants and refugees in the registers. 15 5. The nature of refugee statistics Refugee statistics are part of population statistics and as such are part of the official statistics compiled and disseminated by competent national authorities. The production, storage and dissemination of refugee statistics should therefore abide with internationally established principles, conceptual frames, methodologies and techniques. 6. Resource requirements Not every number is a statistic. Statistical data collection is always a resource-intensive activity, requiring commitments of human, time, and financial resources. In most societies 15 See: http://research.icmpd.org/fileadmin/research- Website/Project_material/PROMINSTAT_File_Exchange/PROMINSTAT_Denmark.pdf, http://research.icmpd.org/fileadmin/research- Website/Project_material/PROMINSTAT_File_Exchange/PROMINSTAT_Norway.pdf, and http://research.icmpd.org/fileadmin/research- Website/Project_material/PROMINSTAT_File_Exchange/PROMINSTAT_Sweden.pdf. 7
refugees and related categories are relative small numbers and may be dispersed over large areas. The resource requirements for the productions of valid, reliable and timely statistics will therefore be more demanding and severe. Specific measures shall be promoted at the national level to improve the quality of the statistical output: 1. Automation of data processing speeds up the production process but also ensures the continuity and consistence of reporting over time; 2. Use of quality frameworks for the statistical production and inclusion of validation procedures at various validation levels; 3. To the extent possible, use of standard classifications and code lists allowing for international data exchange; and 4. Use of standard formats for data and metadata transmission. Departure from established methodology will cause unknown errors and biases, which data users should be aware of and be prepared to accept. Statistical data collection is a technical exercise and should be carried out by technically competent institutions and personnel. THE INTERNATIONAL RECOMMENDATIONS ON REFUGEE STATISTICS The main objective of the International Recommendations on Refugee Statistics is to contribute to the enhancement of the statistical data collection on refugees. This will be achieved by emphasising the adherence to and the presentation of international standards that are applicable and by presenting concrete examples of successful data collection of refugee data. In the appropriate chapters of the IRRS dealing with technical and operation issues due attention will be given to ensure that in the introductory part of the chapter due attention is provided to the international standards and guidelines. Outline of the International Recommendations on Refugee Statistics (IRRS) Taking into account the results of the deliberation at the International Conference on Refugee Statistics in Antalya, Turkey 7 9 October 2015, this section presents an outline for International Recommendations on Refugee Statistics. The IRRS will be a publication consisting of seven chapters: the first chapter will be an introductory chapter providing background information; the second chapter will deal with international and regional legislation on asylum and refugees; chapters three through five will deal with methodological and technical issues of data collection of refugee statistics; the sixth chapter will deal with issues of coordination and cooperation while the seventh chapter will deal with future developments of refugee statistics. Each chapter could consist of one or more essays written by competent individual authors, or could present an official statement of position by a country, institution or organization. 8
The essays may be existing ones that comply with the quality standards, or could be specifically commissioned studies to deal with a specific aspect of experience of data collection on refugees. It may be necessary to provide authors with guidelines on the methodological issues that their chapters will need to comply with. The methodological, technical chapters as well as the chapter on coordination will consist of five sections: 1. Introduction: introducing the themes and detailing existing international standards; 2. Articles or studies on the subject matter that comply with the international standards; 3. Articles or studies that propose new methodologies and techniques; 4. Review articles that assess studies (the experience) in countries, specific groups of countries or regions; and 5. Articles or studies on constraints that affect the development of refugee statistics, or articles or studies suggesting how to overcome present constraints. The following is a brief description of the content of the chapters of the IRRS. I. Introduction and objectives of the IRRS This chapter provides basic information about the IRRS. It deals with: The purpose of the IRRS The scope of the IRRS (i.e. which populations does it cover and which does it not) Background to the development of the IRRS The target audience How to use the IRRS The national and international agencies involved with statistical data collection of refugees II. Legal framework and refugee definition This chapter should present details about international law governing refugees and asylumseekers, including the relevant definitions and procedures to determine their status. Relevant regional legislation and procedures in Africa and the Americas should also be included and, if their definitions and procedures are at variance with the international legislation, this should be indicated and explained. International law: 1951 Convention relating to the Status of Refugees and 1967 Protocol relating to the Status of Refugees. Regional variants: Africa: 1969 Organization of African Unity (OAU) Convention Governing the Specific Aspects of Refugee Problems in Africa. The Americas: 1984 Cartagena Declaration on Refugees as well as more recent initiatives. 9
III. Standardization of concepts and techniques In many countries, concepts and definitions of refugees and related categories are used that do not seem to be based on internationally accepted definitions, instead using terms derived from and used in national legislation on admission, residency and migration. This chapter could consider studies dealing with the standardization of relevant concepts, definitions, and techniques in the collection and analysis of statistics on refugees. This may, for instance, include reference to the legal, operational and statistical differences in defining the term refugee. It could also consider studies dealing with issues arising from the absence of an international depository of data-collection instruments dealing with refugee statistics, as well as proposals for improving this situation. IV. Measuring the number of refugees There is a need to have valid, reliable, and timely information on the number of refugees in countries, both for national administrative and programme purposes as well as to facilitate international comparisons. A considerable number of countries do publish data on refugees, asylum-seekers, and other related categories of individuals in need of protection. However, in view of the divergences in concepts and definitions between the national and international levels, this published data may not reflect the true situation of the number of refugees in some countries. This chapter should include contributions providing concrete examples of experiences in estimating the number of refugees using available statistical instruments. Estimating the number of refugees with a number of basic demographic, social, and economic characteristics is relatively simple, and for the most part the information is readily available in all statistical sources. The statistical options include the following: Administrative records: Some countries produce statistics on refugees using the records of those agencies that deal with the determination of admission to the country. National population registers in countries that have such registers. Periodic national censuses: Many countries already include some questions on migration and sometimes on refugees in their questionnaire. In the case of refugees, however, these seldom seem to be processed and published. National household survey programme: This is a viable option only if the country has a relatively large number of refugees that can be clearly identified in the sample frame and if the size of the household survey is large. It could be possible to have an additional refugee module attached to the routine household survey in which specific areas for instance, where refugees are known to be established are oversampled to obtain sufficient numbers of refugee respondents. 10
It is possible that in certain countries an ad-hoc registration of refugees has taken place to estimate the number of refugees and their basic characteristics. Such studies could also be included. Further, depending on the availability of material, separate sub-chapters could focus on type of statistical instruments and regions. V. Measuring characteristics and integration of refugees and related categories There are studies dealing with the socio-economic and other characteristics of refugees, either purely descriptive with sole regard to refugees or in comparisons with other population groups and the general population. Themes that have been used in recent years include comparative studies of the integration of refugees within a society as well as sociopsychological attitudes and opinions, deviant behaviour among refugees (for instance, drug use or alcoholism), family- or gender-based violence, and appreciation of certain aspects of life, such as housing. This chapter should include contributions of concrete examples of statistical data collection on refugee characteristics for policy and programme development and assessment. For these types of studies, the sources of statistical information are mainly the periodic national censuses and household surveys. With very few additions to the standard population census questionnaire, many countries could obtain an indication of the level of integration of refugees through their national society. Special household surveys among migrants and refugees are also at times undertaken, but care should be exercised in such situations to ensure that only cases that follow the standard methodology are presented. In this respect, studies using a mixed-method approach could be considered, including those that make use of non-probability and even convenience sampling, if these approaches are properly documented and justified. Explorative studies using the case-study approach and non-quantitative methods may also be considered. Finally, depending on the availability of material, separate sub-chapters could be prepared to focus on type of statistical instruments, objectives of the studies being presented, and region. VI. Coordination at the international, national, and regional level A considerable number of national statistical offices are publishing statistics on refugees in their standard publication programmes. However, these offices tend to base their tabulations on data provided by other government or international agencies, notably those that are responsible for migration and the refugee status determination process. It is not clear, however, if national statistical offices redefine the concepts used by source agencies to comply with international standards. 11
This chapter could present descriptions of national, regional, and international coordination arrangements among agencies dealing with population and refugee statistics. This could include the presentation of studies dealing with issues on the coordination at the national, regional, and international levels among statistical and non-statistical agencies, including NGOs, involved in the collection of data about refugees. If relevant, this chapter should specify the type of statistical output that would be envisaged by these coordination efforts for instance, refugee registers or substantive and technical reports of estimations. To achieve coordination, data producers at the country level (national statistical offices and the statistical units of other governmental agencies) will need to be aware of the activities of those agencies responsible for the determination of refugee status. Ideally, these agencies should be able to coordinate their activities to produce valid, reliable, and timely national statistical information on refugees. In turn, this information can be used to estimate refugee numbers and to develop or assess national policies and programmes dealing with refugees. This chapter could also deal with constraints, experienced, observed, or anticipated on statistical data collection and analysis of refugee data. These issues may include purely technical concerns such as a lack of standard definitions, model questionnaires, sampling frame construction, or avoiding non-response. Or, they may include more substantive concerns on inter-agency collaboration and coordination imposed by the different mandates or by national provisions of the legal system and administrative structure. Depending on the availability of material, separate sub-chapters could focus on type of statistical instruments, objectives of the type of studies (descriptive or comparative), and region. VII. Future developments In this chapter future developments in the collection and analysis of refugee statistics may be considered. These could be proposals for new or innovative developments in data sources, data collection activities or types of analysis. These could be linked to existing regional or international statistical activities like the forthcoming 2020 World Population Census Programme or could be refugee specific initiatives. 12