5.2 Data on Dublin indicators

It is important to note that the data presented in this section are exchanged by EU+ countries with EASO to provide timely information but they are provisional and not validated. The official statistics on the Dublin procedure are collected by Eurostat on an annual basis.xxiv As EU regulations foresee a three-month time limit for data transmission, the Eurostat data were incomplete at the time of writing to adequately describe developments in 2019. Therefore, EASO data were used in this section, but they may differ from validated data subsequently submitted to Eurostat.xxv The conclusions drawn from the dataset can also be considered partial, as EASO data cover only three Dublin indicators: decisions received in response to outgoing Dublin requests, decisions to apply the discretionary clause based on Article 17(1)xxvi  and implemented outgoing transfers.

Decisions in response to outgoing Dublin requests

This indicator includes all persons covered by decisions received by a reporting country in response to outgoing Dublin requests sent to another (partner) country. The data are regularly exchanged between EASO and 30 EU+ countries.xxvii The number of decisions on Dublin requests increased slightly, with 3 % more decisions taken in 2019 than in 2018. This represented almost 145 000 decisions in total in 2019 (the figure includes both requests and re-examination requests). The ratio of received Dublin decisions to lodged asylum applications was 20 % in 2019, which may imply that a high number of applicants for international protection continued with secondary movements across EU+ countries.

As seen in previous years, France and Germany received the most decisions in response to their Dublin requests, each representing close to one-third of the total decisions. However, Germany received fewer decisions than in 2019 as a result of an overall drop in national asylum cases. At the same time, responses for France increased due to a bilateral arrangement between the two countries. Decisions on Dublin requests also increased for Belgium, which received two-thirds more decisions in 2019 than in 2018, as well as in Greece, Ireland and the Netherlands.

Turning to countries issuing decisions on Dublin requests, Italy remained the main partner country, issuing one-quarter of all decisions taken, despite a decrease in the overall number of decisions issued. Germany accounted for 15 % of all decisions on requests in Europe, followed by Spain, Greece and France. Each of these three countries took significantly more decisions than in 2018, more than one-half in Spain and Greece and almost one-third in France.

The acceptance rate for decisions on Dublin requests measures the proportion of decisions accepting responsibility (explicitly or implicitly) for an application out of all decisions issued in 2019. The overall acceptance rate dropped for the second year in a row, to 62 % in 2019. But the total share masks great variations across countries. About 90 % of all decisions issued by Latvia, Lithuania and Portugal accepted responsibility for the case, whereas this was just 6 % in Greece. Among countries deciding on the most requests in 2019, acceptance rates were the highest in France, Italy, Spain and Sweden.

Most countries responding to Dublin requests tended to consistently issue positive or negative decisions, irrespective of the country of origin of the third country national (see Figure 5.1). Citizens of Afghanistan, Nigeria, Iraq and Syria were issued the greatest share of decisions overall in 2019, receiving 9 %, 8 %,
6 % and 5 % of the total decisions respectively (citizenship was not reported in about 11 % of all cases). Within this group, the number of decisions for Nigerian and Afghan citizens rose compared to 2018, whereas there was a drop in decisions for Syrian and Iraqi nationals. 

Figure 5.1 Number of decisions issued in response to Dublin requests and acceptance rates, by partner country and Top 20 citizenships, 2019


Notes: The size of the bubble corresponds to the total number of decisions on Dublin requests.  The colour of each bubble indicates the acceptance rate, with green representing a high rate and red representing a low rate. Data from France are not disaggregated by citizenship of the third country national when a partner country has rejected a request from France. Therefore, data from France are not included in the total calculation as they would significantly bias the acceptance rates. 
Source: EASO.

Turning to absolute terms, the number of decisions increased for citizens of Turkey and Palestine by more than 900 each compared to 2018. In contrast, decisions fell significantly for Sudanese (by almost 1 600) and Eritrean (by close to 1 300) nationals.

Take charge and take back requests

In accordance with the Dublin III Regulation, Articles 8-16 and 17(2), take charge requests refer to Member State A (reporting country) requesting Member State B (partner country) to take responsibility for an application for international protection although the applicant submitted an application in Member State A and has not submitted an application in Member State B, but the Dublin criteria indicate that Member State B is responsible. Reasons can be due to, for example, family reunion (in particular for unaccompanied minors), documentation (visas, residence permits), entry or stay reasons (e.g. using Eurodac proof) and humanitarian reasons.

Under the Dublin III Regulation, Articles 18(1)b-d and 20(5), take back requests refer to Member State A (reporting country) requesting Member State B (partner country) to take responsibility for an applicant who applied for international protection within the reporting country because:

  The person has already previously made an application for international protection in Member State B (and afterwards left that Member State); or
Member State B has already previously accepted its responsibility following a take charge request from some other Member State.

Of the decisions takenxxix in 2019, two-thirds were in response to take back requests, meaning that most decisions on Dublin requests concerned applications which had already been lodged in another EU+ country. Nevertheless, the data should be interpreted with caution due to the high number of cases in which the legal basis could not be verified. The acceptance rate for take back requests was 60 %, similar to the 61 % rate for take charge requests. The acceptance rate for both types of legal basis (as well as those of unknown legal basis) dropped due to decreases in the share of positive decisions.

Use of the discretionary clause

The evocation of the Dublin III Regulation, Article 17(1), known as the discretionary or sovereignty clause, largely decreased in 2019xxx compared to 2018. It was evoked over 6 900 times in 2019,  compared to about 12 300 times in 2018. This decrease is mainly explained by the fact that there were fewer cases in which Germany, and to a lesser extent the Netherlands, made use of this legal provision. In contrast, France evoked Article 17(1) much more often than in 2018. Germany still continued to use the sovereignty clause the most often, followed by France, Belgium and Switzerland. 

In just over one-quarter of the cases, Italy was identified as the partner country to which a request could have been sent. In another quarter of cases, the identified partner was Greece. Other countries included Hungary (9 %), Spain and Germany (5 % each). However, in 17 % of the decisions to apply the sovereignty clause, the potential partner country was not reported.

The citizenship of the applicant was not identified in 36 % of all cases involving the evocation of the discretionary clause. Of the cases with reported citizenship, 12 % concerned nationals of Nigeria, 11 % of Turkey, 9 % of Afghanistan and 6 % of Syria. 


Under the Dublin procedure, a transfer occurs when Member State B (partner country) accepts to take responsibility for an application for international protection from Member State A (reporting country) in line with the conditions set out in the Dublin III Regulation. 

Reporting countries implemented just under 27 200 transfers in 2019,xxxi a 3 % decrease compared to 2018, which is in line with the small decrease in accepted requests. About 30 % of the transfers were implemented by Germany, followed by France (20 %), the Netherlands (11 %), Greece (9 %), Poland and Austria (5 % each). While France and the Netherlands (and to a lesser extent Poland) carried out considerably more transfers than in 2018, there was a large drop for Greece and smaller decreases for Austria and Germany, even though the latter could actually raise its ratio of successfully implemented transfers in regard to the total number of national Dublin cases. Nearly one-half of all Dublin transfers were to Italy and Germany. Other countries receiving significant numbers of transfers included France, Spain, Sweden, the Netherlands and Switzerland. 

The persons who were transferred originated from diverse countries. The majority were nationals of nine countries: Afghanistan, Algeria, Guinea, Iran, Iraq, Nigeria, Pakistan, Russia and Syria. More transfers of Nigerians were realised in 2019 compared to 2018, consistent with an increase in the number of accepted requests for this nationality. In contrast, there were markedly fewer transfers of Syrians compared to the previous year, in line with a considerable drop in their number of accepted requests.  

Figure 5.2 illustrates the Top 10 combinations of sending country, citizenship and receiving country for implemented Dublin transfers in 2019, representing 13 % of all transfers. The top flow featured Nigerians sent from Germany to Italy, accounting for 3 % of all transfers and becoming even more prominent than in 2018. Another major flow comprised Syrians transferred from Greece to Germany, although the total number of transferees dropped significantly. Other important flows included Russians (in particular Chechens) sent from Germany to Poland and Iraqis sent from Greece to the United Kingdom. 

Turning to the characteristics of transferees, almost two-thirds of all transferred persons were adults aged 18 years and older (see Figure 5.3). There were almost 3.5 times as many male transferees than female ones. Among the cases with reported data on age and sex, about two-thirds of all transferees were adult males.

While Dublin Member States generally do not transfer unaccompanied minors, the number of girls and boys younger than 14 years who were transferred was relatively similar, potentially implying that minors in this age group in Dublin transfers were largely involved in asylum applications with their families. This seems to be case for Syrian families, as nearly one-half of all transferred Syrians were under 18 years of age. 

Figure 5.2 Top 10 combinations of sending country, citizenship and receiving country for implemented Dublin transfers, 2019

Sending country                                  Citizenship                               Receiving country

Source: EASO.

Figure 5.3 Transferees in the Dublin procedure, by sex and age group, 2019

Source: EASO.

Existing challenges reported by civil society organisations

Input from civil society organisations complements this picture. ECRE published two reports on the implementation of the Dublin III Regulation in 2018 (covering 23 countries) and in the first half of 2019 (covering 21 countries), based on publicly-available information and input from other organisations. A third report was commissioned by the European Parliament Research Service to analyse the implementation of the Dublin system. The ECRE reports concluded that the aim of the Dublin III Regulation – to ensure that the Member State responsible is rapidly determined in order to guarantee effective access to procedures and swift processing of applications – in practice is not functioning as envisaged and gaps in its methodology should be addressed to take account of the rights of applicants.

The Dutch Advisory Committee on Migration Affairs also concluded that the Dublin system is not functioning efficiently to combat secondary movements and a fundamental reform should include positive incentives for both applicants and Member States to follow the procedures of the system, instead of increasing restrictions.186

In 2019, the Hungarian Helsinki Committee facilitated a Dublin procedure for several unaccompanied minors from Hungary to Germany.187 

Safe Passage International was concerned that the authorities in the United Kingdom used the CJEU Joined Cases C‑47/17 and C‑48/17 to circumvent responsibility for Dublin requests by failing to respond within two weeks to a re-examination request sent by another state. This resulted in the responsibility for unaccompanied children reverting back to the requesting country. The organisation underlined that this practice denies the right to family life and the best interests of the child.188

Swiss NGOs, which were united under the Dublin Appell coalition in 2017, 189 continued to advocate for a more lenient approach on the use of the discretionary clauses by authorities.190 In particular, the coalition brought attention to the family criteria which were applied narrowly191 and to the short timeframe of five days for Dublin appeals.192 

Caritas Austria noted issues related to the legal standing of Dublin applicants and the scope of the legal remedy193  in view of denials of family reunion, where the Supreme Administrative Court had decided that no legal remedy was available.

The 2019 update of the AIDA report on Bulgaria highlighted that communication between local reception centres and the Dublin Unit of the State Agency for Refugees was unduly lengthy to gather the necessary documentation for a Dublin request. To address this, a new draft proposal for the Law on Asylum and Refugees aims to accelerate the process by removing some of the administrative burdens.

In Poland, the length of the appeal procedure before the Refugee Board in Dublin cases was reported to have increased. This was linked to the expiry of term of the former Board and the appointment of a new composition of the Board in February 2019 and the related transitional period of new members taking on their tasks.  

The Spanish Fundación Cepaim observed instances when Dublin applicants had to cover the costs of travel for a transfer194  and access to reception facilities for applicants returned to Spain under the Dublin III Regulation remained a challenge.

Regarding the operation of Regional Dublin Procedural Hubs in France, the civil society organisation, Forum réfugiés – Cosi, signalled that applicants often face difficulties travelling to the appointments and missing an appointment can likely result in the withdrawal of their reception conditions.195 


xxiv Based on Article 4.4 of Regulation (EC) 862/2007 of 11 July 2007 on Community statistics on migration and international protection and repealing Council Regulation (EEC) No 311/76 on the compilation of statistics on foreign workers.
xxv Iceland and Liechtenstein do not participate in EASO data exchange.
xxvi Through the discretionary clauses, the Dublin system makes it possible for the Member States to take fully into account the legitimate concerns of applicants for international protection and to derogate from both the mechanical application of the responsibility criteria and the one-chance-only principle. The first one is the ‘sovereignty clause’ in Article 17(1) of Dublin III Regulation. This clause authorises any Member State with which an application for international protection is lodged to examine it, by derogation from the responsibility criteria and/or the readmission rules; the second one is the ‘humanitarian clause’ in Article 17(2) of the Dublin III Regulation. This clause authorises and encourages Member States to bring family relations together in cases where the strict application of the criteria would keep them apart. ​​​​​
xxvii In addition to Iceland and Liechtenstein, data were not available for several months for Cyprus. France generally provides data with a one-month delay. Thus, data for France for 2019 cover the period December 2018 to November 2019.

xxix EASO data do not contain information on the specific article of the Dublin III Regulation used as a basis for sending a request. However, 69 % of all decisions in 2019 distinguished between responses to take charge and take back requests.
xxx Data on the use of the discretionary clause were shared by 27 reporting countries, but one country did not report every month. In addition to Iceland and Liechtenstein, which do not participate in the EASO data exchange, data for 2019 were completely missing for Bulgaria, Cyprus and Greece, and partially for Croatia.

xxxi Data were partially missing for Cyprus.


186 Dutch Advisory Committee on Migration Affairs. (2020). Input to “EASO Asylum Report 2020: Annual Report on the Situation of Asylum in the European Union". https://www.easo.europa.eu/sites/default/files/easo-annual-report-2019-Dutch-Advisory-Committee-on-Migration-Affairs-contribution.pdf
187 Hungarian Helsinki Committee. (2020). Input to “EASO Asylum Report 2020: Annual Report on the Situation of Asylum in the European Union". https://www.easo.europa.eu/sites/default/files/easo-annual-report-2019-Hungarian-Helsinki-Committee-contribution.pdf; AIDA Hungary. (2020). Country Report: Hungary - 2019 Update. Edited by ECRE. Written by Hungarian Helsinki Committee. https://www.asylumineurope.org/sites/default/files/report-download/aida_hu_2019update.pdf
188 Safe Passage International. (2020). Input to “EASO Asylum Report 2020: Annual Report on the Situation of Asylum in the European Union". https://www.easo.europa.eu/sites/default/files/easo-annual-report-2019-Safe-Passage-International-contribution.pdf
189 Dublin-Apell. (n.d.). Dublin-Appell. Retrieved 20 May 2020, from https://www.dublin-appell.ch/de/
190 AIDA Switzerland. (2020). Country Report: Switzerland - 2019 Update. Edited by ECRE. Written by Swiss Refugee Council. https://www.asylumineurope.org/sites/default/files/report-download/aida_ch_2019update.pdf
191 AIDA Switzerland. (2020). Country Report: Switzerland - 2019 Update. Edited by ECRE. Written by Swiss Refugee Council. https://www.asylumineurope.org/sites/default/files/report-download/aida_ch_2019update.pdf
192 AsyLex. (2020). Input to “EASO Asylum Report 2020: Annual Report on the Situation of Asylum in the European Union". https://www.easo.europa.eu/sites/default/files/easo-annual-report-2019-AsyLex-Switzerland-contribution.pdf
193 Caritas Austria. (2020). Input to “EASO Asylum Report 2020: Annual Report on the Situation of Asylum in the European Union". https://www.easo.europa.eu/sites/default/files/easo-annual-report-2019-Caritas-Austria-contribution.pdf
194 AIDA Bulgaria. (2020). Country Report: Bulgaria - 2019 Update. Edited by ECRE. Written by Bulgarian Helsinki Committee. https://www.asylumineurope.org/sites/default/files/report-download/aida_bg_2019update.pdf
195 Fundación Cepaim. (2020). Input to “EASO Asylum Report 2020: Annual Report on the Situation of Asylum in the European Union". https://www.easo.europa.eu/sites/default/files/easo-annual-report-2019-Fundacion-Cepaim-contribution.pdf
196 Forum réfugiés – Cosi. (2020). Input to “EASO Asylum Report 2020: Annual Report on the Situation of Asylum in the European Union". https://www.easo.europa.eu/sites/default/files/easo-annual-report-2019-Forum-refugies-Cosi-contribution.pdf; AIDA France. (2020). Country Report: France - 2019 Update. Edited by ECRE. Written by Forum réfugiés – Cosi. https://www.asylumineurope.org/sites/default/files/report-download/aida_fr_2019update.pdf




Download PDF