STUDY Afghan Exodus Afghan Asylum Seekers in Europe and Germany THOMAS RUTTIG February 2017 n In 2016, Afghans were the second largest group both of migrants seeking protection in Europe and of those formally applying for asylum. Since the peak of the refugee crisis in late 2015, the number of arrivals – both in general and in terms of Afghans – have dropped significantly. European countries have made traveling to, staying, and integrating into the society increasingly complicated. Numbers of asylum applications widely differed between European countries. Furthermore, the EU and individual member states have put agreements in place with the Afghan government that allow »voluntary« and»enforced« returns of large numbers of rejected asylum seekers. n In this comprehensive three-part dispatch, AAN’s co-director Thomas Ruttig looks at the latest figures and trends as well as changes in policy and the social climate that have im-pacted the situation for Afghan asylum seekers in Europe. This will be followed by an overview of the situation in a number of individual European countries and a case study on Germany, the largest recipient country in Europe for refugees. The last part will also draw some conclusions.

THOMAS RUTTIG  |  AFGHAN EXODUS Contents 1. Europe – The Changing Situation.......................................... 3 1.1 Overall 1.2 Afghan 1.2.1 Arrivals in 1.2.2 Asylum Applications in 1.2.3 Decisions in 1.3 Policy Changes: Sealing Borders 1.4 Policy Changes: Turning the Trend from Influx to 1.5 The Changing Climate in Recipient Countries 10 2. The North-South Divide................................................ 12 2.1 The Situation for Afghans in Mediterranean Countries......................... 13 2.1.1 13 2.1.2 13 2.1.3 Italy 14 2.1.4 15 2.1.5 Bulgaria 15 2.1.6 16 2.1.7 17 2.2 The Central Divide: Austria and 17 2.2.1 17 2.2.2 17 2.3 Afghans in Select Northern EU Countries 18 2.3.1 Sweden 18 2.3.2 2.3.3 United 21 2.3.4 France 21 2.4 European Developments Put Into Context – A Brief 3. Case Study Germany 22 3.1 The 3.2 Slowing Down Afghan Cases 24 3.3 The German-Afghan Migration 3.4 Afghanistan: Safe for 3.5 Afghans As»Second-class« Asylum 27 3.6 A Climate of Fear and 3.7 Will Germany Treat Afghans Fairly 4. Conclusions 29 5. Figures and Sources 30

THOMAS RUTTIG  |  AFGHAN EXODUS 1. Europe – The Changing Situation 1.1 Overall Figures The overall number of arriving migrants in Europe has dropped sharply in 2016. Arrivals from non-European countries of origin to Europe – i. e. the 28 EU membercountries(including the brexiting UK) plus the four nonmembers(Norway, Switzerland, Iceland, and Liechtenstein) – decreased by two thirds from 1,015,000 in the peak year of 2015 to close to over 362,000 in 2016. These UNHCR figures 1 only count those arriving across the Mediterranean, which is by far the most important entry route. There are no statistics about other routes where much smaller numbers of migrants can be assumed, for example through Russia. Of these first-time applicants from all countries of origin, 63 percent registered in Germany, almost the same percentage as in 2015(more detail in part 3). Migrants also registered in Sweden(11.8 %), Italy(8.8 %), and France (5.2 %). Austria, Greece, and the UK each had over three percent; Hungary over two percent; Bulgaria, the Netherlands, Belgium, and Spain each over one percent. Other countries registered under one percent, with Estonia and Slovakia(both 0.01 percent) at the absolute bottom. In the first three quarters of 2016, Germany also had the highest rate of asylum seekers per capita of the population(2,155; 2,273; 2,945). With one exception(Austria in the first quarter), this was more than double all runnerup countries. The next highest rates were, in descending sequence, in Austria, Malta, and Luxemburg; Hungary, Austria, and Greece; Malta, Greece, and Austria). This is still different from 2015 when, amidst the highest absolute number of incoming migrants, Germany registered only a comparatively low percentage of them as asylum applicants. Then, Germany ranked fifth in Europe – although per capita rates were far higher. Germany had 5,441, trailing Hungary(17,699), Sweden(16,016), Austria(9,970), and Finland(5,876). The overall number of people applying for asylum or other forms of protection in Europe, after dropping by one third between the last quarter of the peak year of 2015 (with 426,000 applicants) and the first quarter of 2016 1. See a daily update here: http://data.unhcr.org/mediterranean/regional. php#_ga=1.139236342.473982958.1465761939. (less than 290,000), again started to rise in 2016. A total of over 951,000 was reached by the end of the third quarter, according to the most recent published data from the EU(full 2016 figures are expected in March 2017). If the trend continues, the 2015 level of 1.26 million applicants (more than double 2014) might be reached again. Incoming but still incomplete national data for the full 2016 year reviewed by the Asylum Information Database (AIDA) 2 indicates contradictory trends among European countries. While an increase in asylum applications compared to 2015 was reported from Germany, Italy, France, and Greece 3 ,»most other countries remain far behind Germany and reported a decrease in the number of asylum applications registered last year.« The seeming contradiction between the drop-in arrival figures and continued high levels of asylum applications reflects a situation where, in 2016, many of those who had arrived in 2015 but had not been able to formally register an asylum claim, due to their large numbers, or had avoided doing so, finally registered. Moreover, migrants who had arrived before 2015 and lived illegally in Europe may have used the opportunity to register. 1.2 Afghan Figures 1.2.1 Arrivals in Europe Looking at Afghan in-migration, 43,400 individuals arrived across the Mediterranean in 2016. In the peak year of 2015, almost five times that many, some 200,000 (find an analysis of 2015 trends in this AAN dispatch 4 ). The percentage of Afghans among all arrivals across the Mediterranean Sea dropped from 20 percent in 2015 5 to twelve percent in 2016. This drop by almost 80 percent in their absolute figures is even steeper than the average from all countries. In 2016, almost all Afghan migrants to Europe continued to arrive in Greece. Only 349 Afghans came to Italy(0.2 2. See http://www.asylumineurope.org/news/23-05-2017/asylum-statistics-2016-sharper-inequalities-and-persisting-asylum-lottery. 3. http://asylo.gov.gr/wp-content/uploads/2017/01/Greek-Asylum-Servi ce-statistical-data_December2016_gr.pdf. 4. https://www.afghanistan-analysts.org/an-afghan-exodus-facts-figurestrends/. 5. http://www.unhcr.org/news/latest/2015/12/5683d0b56/million-sea-ar rivals-reach-europe-2015.html. 3

THOMAS RUTTIG  |  AFGHAN EXODUS percent of all arrivals) and none to Spain. The majority of the over 39,000 Afghans that arrived in Greece came before mid-March 2016 when the updated EU-Turkey migration deal kicked in(officially it is called the Joint Action Plan, the first version of which had come into force in November 2015). 6 After March 2016, Afghan arrival figures in Greece dropped drastically, to 1,590 between April and September 2016, i. e. 265 per month on average. Relatively smaller numbers of Afghans entered Finland and northern Norway through Arctic Russia, mainly in 2015 and early 2016. The figures for Finland were 720 for 2015, compared to 28 in 2014 and 14 in 2013, according to this government website 7 . In January and February of 2016 the numbers increased again to 1,000 until Russia and Finland agreed to close their border for third-nation citizens. Norway and Afghanistan agreed in December 2016 that Kabul would take back 90 percent of its 4,000 citizens who had crossed the temporarily permeable Russian-Norwegian border close to the polar circle in the same period.(For more on this, see part two of this dispatch; see also this AAN dispatch 8 ). 9 1.2.2 Asylum Applications in Europe The trend found above for all countries of origin – that the drop in the number of incoming new migrants in 2016 did not result in a drop of asylum applications over the same period – is also true for Afghans. After the quarterly figure fell by more than half between the last quarter of 2015 and the first quarter of 2016 10 (from 79,255 to 34,800), figures began rising again from quarter to quarter in 2016. They reached 50,300 in the second 11 and 62,100 in the third quarter 12 . By then, the 6. http://www.consilium.europa.eu/en/press/press-releases/2016/03/18eu-turkey-statement/. 7. http://www.raja.fi/download/65156_tphakijat_itarajalla.pdf?1dc9c56e 963fd488. 8. https://www.afghanistan-analysts.org/afghan-exodus-can-the-afghan -government-deal-with-more-returnees-from-europe/. 9. Finnish journalists and analysts told AAN they saw Russian steering behind this part of the migration movement; it ended as abruptly as it had started. 10. http://ec.europa.eu/eurostat/documents/2995521/7494855/3-1606 2016-BP-EN.pdf/4ff50bf8-82fc-4af0-9907-9c8546feb130. 11. http://ec.europa.eu/eurostat/documents/2995521/7662180/3-2209 2016-AP-EN.pdf/22f5de3b-b5a8-4195-82fe-3072a4a08146. 12. http://ec.europa.eu/eurostat/documents/2995521/7773598/3-1512 2016-BP-EN.pdf/30f7b06b-1634-44dd-964c-13a5f9c436eb. total amount was 147,200 or 15.5 percent of the over 951,000 first time applicants from all countries. The number reached by the end of the third quarter 2016 indicates that, if the trend continues, the overall figure for 2015(178,200, i. e. 14.2 percent of all applicants and four times more than 2014) might have been reached again in 2016. By the end of the third quarter of 2016, the largest number of Afghan asylum applications was registered in Germany(102,900) 13 – more than two thirds of their total, followed by Austria(10,100), Hungary(9,800), Bulgaria (6,500), France(4,500), Italy(under 3,900), Switzerland (3,000), Sweden, the UK(2,600 each), and Belgium (2,000). In the third quarter of 2016, Afghanistan featured among the top five countries of origin in 16 EU countries plus in Norway and Switzerland. In four countries, Afghanistan was the most important country of origin, although with comparatively low numbers(Austria 2,185, Hungary 1,610, Bulgaria 100, and Slovenia 70). 14 Given all the figures for 2016 above, Afghans were the second largest nationality of both arriving migrants and asylum applicants. 1.2.3 Decisions in Europe The number of Afghan asylum cases that have been decided upon by authorities in member countries, even in the first instance(wherein there is the right to appeal), has remained much lower than the amount of people 13. As in this case, the quarterly Europe-wide figures published by Euro­ stat deviate from national figures. The report of the German asylum authority for the period from January to September 2016 gives 115,342 asylum applications from Afghans. As Eurostat publishes its quarterly figures later(in the third months of the following quarter), they might be more accurate here as they seem to incorporate adjustments. In other cases, as EU sources working on asylum issues told AAN, adjusted national figures are not communicated to Brussels, creating other gaps in the data. Another example of inconsistent data is Germany’s overall figure for the incoming migrants in 2015: on September 30 th , 2016, the German government had to correct down this figure from 1.1 million to 890,000, by circa 20 percent(see media report and video of original statement by the interior minister here). Surprisingly, Germany continued to use the unadjusted figures, even in key documents published after the correction such as its December 2016 asylum statistics report that also doubles as the annual 2016 report and its 2015 Migration Report published in December 2016(that covers all aspects of migration). 14. http://ec.europa.eu/eurostat/statistics-explained/images/2/2e/Five_ main_citizenships_of_first_time_asylum_applicants%2C_3rd_quarter_ 2016.png. 4

THOMAS RUTTIG  |  AFGHAN EXODUS who’ve applied. In the first and second quarters of 2016, decisions were reached on fewer than 20,000 Afghan cases. Processing picked up in the third quarter, with 27,300 decided cases. 15 These cases still represent only around 20 percent of the 240,000 Afghan asylum cases reportedly pending with the EU by mid-November 2016 – not counting the unknown number of Afghans who had not yet had a chance or decided not to file an application. The Europe-wide protection rate for Afghan asylum applicants was above 50 percent throughout the three first quarters of 2016. In the first quarter, 4,215 of the 7,415 decided cases(56.8 percent) ended positively; they received protection status. There were 3,200 negative decisions. In the second quarter, the rate sank slightly to 53.1 percent, based on a growing number of cases decided(12,840); 6,820 Afghans received protection while 6,020 cases were rejected protection. That gives an overall protection quota of 54.5 percent for the first half of 2016. In the third quarter, the rate dropped to 50.7 percent, with more than twice as many cases decided(27,300) than in the previous quarter. Large numbers of rejected asylum applications does not mean that similar numbers of people have been forcibly deported to their country of origin. In fact, countries such as Germany(until 2015) and Sweden(for some of 2016) generally categorized Afghans as»protected from deportation« for humanitarian reasons, due to the on-going war. However, this is now changing(see more below). The German daily Frankfurter Allgemeine reported 16 in December 2016, that»for no other country of origin has the recognition quota in the individual EU membercountries differed so widely« as for the Afghans –»from 14 to 96 percent.« On the other hand, as a UK government figure shows 17 , Afghans were the nationality with the third highest number of positive decisions(6,820 or 53 percent) in the EU as a whole in the second quarter of 2016. 15. http://ec.europa.eu/eurostat/documents/6049358/7005580/Rates+ of+Recognition+-+EU28+Q3+2016.pdf/5588cb63-21f6-4981-b806-51 4eef99ea45. 16. http://www.faz.net/aktuell/wirtschaft/teure-afghanen-analyse-zur-sam melabschiebung-14579268.html?printPagedArticle=true#pageIndex_2. 17. https://www.gov.uk/government/publications/immigration-statisticsjuly-to-september-2016/asylum. The AIDA database 18 , with incomplete all-2016 statistics, also reported general»protection disparities« as well as specific disparities in protection for Afghans. The protection rate ranges from 30 percent in Norway to 59 percent in Belgium. Finland’s protection rate was 42.4, Sweden 45, Greece 48.8, Germany 55.8, and Austria 56 percent. 19 1.3 Policy Changes: Sealing Borders The drop in overall arrivals as well as Afghan arrivals reflects changes in European policies.»Temporary« border controls, even between EU member states, were re-introduced in 2015 and are still in place. In September 2015, Germany implemented an increasing number of checks at its Austrian border. At the same time, Hungary closed and started fencing its borders with non-EU Serbia and Croatia. Moreover, Slovenia fenced its border with Croatia. Croatia did not close its Serbian border, as a large part is formed by the Sava River and therefore is difficult to cross. This was followed by similar measures in the Czech Republic, Austria, Denmark, Sweden, France, and in the non-EU states of Norway and Switzerland. At the end of November 2015, authorities in the most affected countries on the Balkan route decided to allow only Syrian, Afghan, and Iraqi nationals to cross their borders. This changed on 18 February 2016, when heads of the national police in Austria, Slovenia, Croatia, Serbia and Macedonia decided that Afghans could not pass their borders anymore. An AP journalist wrote at the time:»Suddenly, Afghans appear to be the new pariahs of Europe.« Although Germany, as the greatest recipient of Afghan arrivals, profited most from the decision, chancellor Merkel condemned the move at the time, as she realized that this would put a large burden on Greece and might undermine attempts to set up distribution quotas in the EU – which it did. 20 A few weeks later, on 9 March 2016, Serbia, Macedonia, Slovenia, and Croatia fully closed their borders to 18. http://www.asylumineurope.org/news/17-01-2017/asylum-statistics2016-sharper-inequalities-and-persisting-asylum-lottery. 19. The AIDA figures need to be taken with a pinch of salt. 2015 protection rates for Afghans, for example for Germany, seem too high. Officially, the Afghan protection rate was below 50 percent there. AIDA has probably used adjusted(excluding Dublin cases) figures for 2015 and unadjusted figures for 2016. 20. http://www.bbc.com/news/world-europe-35772206. 5

THOMAS RUTTIG  |  AFGHAN EXODUS any new migrants with the implicit backing of the European Union 21 , which announced the Turkey deal at the same time. Slovenia’s and Croatia’s announcements to return to full implementation of the Schengen Border Code had a domino effect among other countries in the region who adopted daily quotas and sought to re-establish greater border control. As AAN reported at the time 22 , thousands of people became stuck in Greece as well as at various locations along the route north, with many more on the way from Syria, Afghanistan, and other places. In Serbia, which as a result of these measures became an EU antechamber, approximately 800 migrants were stuck in in Preševo (near the Serbian-Macedonian border) and 600 people in Šid(near the Serbian-Croatian border). On 20 March 2016, the EU-Turkey Action Plan came into force. It stipulated that the legitimacy of asylum claims of all new irregular migrants crossing the Aegean Sea from Turkey to the Greek islands would be checked there and those found illegitimate returned to Turkey.(Read more detail in this AAN analysis 23 and in this German media report 24 .) However, this plan did not work out, as numerous EU countries refused to agree to accept a quota of those legitimate asylum seekers. The EU also did not fully live up to its commitments to send additional migration experts to Greece 25 and even refused to send some to the Greek islands, as the situation was»too dangerous« there. 26 Furthermore, in August 2016 the Turkish government decided to withdraw its liaison officers from the Greek islands, making the practical implementation of the deal even more complicated. 27 Turkey has repeatedly threatened to cancel the deal with the EU as a result of deteriorating EU-Turkey relations after the crackdown following the July 2016 coup attempts. 21. https://www.nytimes.com/2016/03/10/world/europe/europe-refugeecrisis.html. 22. https://www.afghanistan-analysts.org/afghan-exodus-the-opening-andclosing-of-the-balkan-corridor/. 23. https://www.afghanistan-analysts.org/afghan-exodus-the-opening-andclosing-of-the-balkan-corridor/. 24. http://www.zeit.de/politik/ausland/2016-09/fluechtlinge-aegaeis-inselngriechenland-unterstuetzung-eu. 25. http://aa.com.tr/en/europe/lack-of-asylum-experts-keeps-migrants-st uck-in-greece/658655. 26. http://apokoronasnews.gr/eu-refuses-to-send-asylum-experts-to-greekislands/. 27. https://www.theguardian.com/world/2016/aug/31/turkish-police-with drawal-greece-stalls-eu-migration-pact-unhcr. Although some EU member countries stuck to their commitment under the deal, only 5,875 asylum seekers entering Greece had been relocated to other EU countries by 28 November 2016, according to the European Stability Initiative, a Berlin-based think tank that reportedly designed the EU-Turkey deal. 28 The same applies for Italy(see more below), from where only 1,802 asylum seekers have been relocated(specific numbers about how many Afghans were among them are not available). The combined figures for Greece and Italy only reach around five percent of the original relocation target. 29 The UN Special Rapporteur on the human rights of migrants criticized in mid-2016 that»the EU and the overwhelming majority of EU member States have abandoned Greece – a country that is fighting to implement austerity measures – leaving it to deal with an issue that requires efforts from all«. 30 Since then, there has been no major change in this situation. Bulgaria had already started building a fence along most of its border with Turkey in 2014. 31 Greece fenced parts of its Turkish land border, near Bulgaria. In early December 2015, Austria began building a fence along its border with Slovenia, the first to be set up between two Schengen countries. 32 Another fence was erected at the border crossing between Norway and Russia. (The Economist has an interactive map on this subject. 33 ) For Afghans and others seeking protection, this blocked the way into Europe at the outer EU border, or at least made access to Europe more risky, costly, and dangerous. 34 A number of refugees are trying to wait out the situation in Turkey while still others have changed their minds and are staying in Turkey for good(as this AAN 28. http://www.esiweb.org/pdf/ESI- Eleven facts about the EU-Turkey Agreement- December 2016.pdf. 29. https://www.theguardian.com/world/2016/dec/08/eu-met-only-5-oftarget-for-relocating-refugees-from-greece-and-italy. 30. https://refugeesmigrants.un.org/eu-has-largely-%E2%80%98aban doned%E2%80%99-greece-deal-migrant-crisis-its-own-un-expertwarns. 31. http://www.novinite.com/articles/160538/Bulgarian+Defense+Minist er%3A+Border+Fence+to+be+Finished+in+June. 32. https://www.afghanistan-analysts.org/afghan-exodus-the-openingand-closing-of-the-balkan-corridor/. 33. http://www.economist.com/news/europe/21664902-fight-betweentwo-eastern-european-union-members-may-be-more-about-politics-about. 34. https://www.afghanistan-analysts.org/afghan-exodus-the-re-emerge nce-of-smugglers-along-the-balkan-route/. 6

THOMAS RUTTIG  |  AFGHAN EXODUS dispatch 35 showed). Those who had made it into Greece but were unable to travel on via the closed Balkan route, experienced the Greek government’s increasing pressure to file an asylum application there(also a prerequisite for redistribution in the EU, demanded by Greece, which so far has not happened in any significant numbers). 36 The number of applicants in Greece rose from around 1,000 a month(up to February 2016) to over 7,500 in November 2016, reaching almost 47,000 by that month. Among the total were 3,295 Afghans, but their percentage in this group(7 %) is very likely way below their actual proportion of the total number of migrants currently in the country.(Here is an amazing NPR radio show about refugees in Greece broadcast in July 2016. 37 ) A few months after the closure of the Balkan route, in summer 2016, a number of migrants – including Afghans – used what a local newspaper described as»Europe’s last needle’s eye to the North«: the mountainous and unsealed Italian-Swiss border into Switzerland or further into Germany. 38 According to the Swiss authorities, 4,833 incoming migrants left the country via this route again in 2016, with 3,385 of them arriving in Germany. 39 This appears as though Switzerland was making sure that most incoming migrants would leave the country again. Over the same period, between January and October 2016, Switzerland itself received 3,035 Afghan asylum applications 40 . 41 Later, according to Swiss media reports, the country’s border police started to reject migrants at the southern border with Italy, even if they tried to request asylum. NGOs also collected cases on the Italian side of the Swiss border of asylum seekers who were rejected even though 35. https://www.afghanistan-analysts.org/afghan-exodus-smuggling-net works-migration-and-settlement-patterns-in-turkey/. 36. By mid-December 2016, only 6,461 refugees – instead of the 66,400 envisaged – had been redistributed from Greece to other EU member states, according to MSF Germany(quoted here). 37. https://www.thisamericanlife.org/greece/. 38. https://www.woz.ch/-7062. 39. https://www.woz.ch/1634/como/das-versteckspiel-der-behoerden. 40. https://de.statista.com/statistik/daten/studie/293542/umfrage/asylge suche-in-der-schweiz-nach-herkunftslaendern/. 41. In 2015, Switzerland had 7,831 Afghans apply for asylum(5,902 in December alone), making Afghanistan the second largest country of origin for that year(up more than tenfold from the mere 747 applications in 2014; see here) and representing 19.8 percent of all applicants for 2015. Switzerland also has a significant Afghan community. Currently 1,194 accepted Afghan asylum seekers are living in Switzerland, 4,074 have been granted temporary protection and 12,194 others are still in the process (see here, all figures third quarter 2016). they had family members in Switzerland which. According to regulations, migrants with family in the country should have been given them entry. 42 Dublin cases – migrants, even if they are under-age, whose entry has previously been registered in another EU county can, according to EU law, be returned there to process their asylum application – are often not processed according to the official procedures, says Schweizer Flüchtlingshilfe (Swiss Refugee Help), a leading local support organization for migrants. 43 After the temporary opening and subsequent closure of the route through Arctic Russia into northern Norway and Finland in late 2015, other»exotic« routes came up during 2016. The Washington Post reported 44 that Afghan asylum applications in India had»doubled« by early 2016, compared with the year before. In January 2016, the UNHCR New Delhi Factsheet said 45 that India was hosting 13,381 Afghan refugees and asylum seekers, mostly settled in and around the capital, New Delhi. Other Afghans reportedly tried to cross into the US or Canada by obtaining visas for Cuba, Mexico, or other Latin American countries. 46 A German official statistic 47 included asylum request figures as of October 2016 from other leading western countries, the US(almost 100,500), Canada(almost 37,000), Australia(over 12,200), and New Zealand(319), but did not specify countries of origin. 1.4 Policy Changes: Turning the Trend from Influx to Return Following border enforcement measures, the European countries sought to reverse migration patterns from influx to return. Afghans were one of the groups that received special attention as they are the second larg42. http://www.taz.de/!5332876/. 43. https://www.woz.ch/-70a0. 44. https://www.washingtonpost.com/world/asia_pacific/kabul-libre-onenew-afghan-trail-to-the-west-goes-through-cuba/2016/04/16/da2149260188-11e6-8bb1-f124a43f84dc_story.html?utm_term=.9bf9276abb1a. 45. http://www.unhcr.org/protection/operations/50001ec69/india-factsheet.html. 46. https://www.washingtonpost.com/world/asia_pacific/kabul-libre-onenew-afghan-trail-to-the-west-goes-through-cuba/2016/04/16/da2149260188-11e6-8bb1-f124a43f84dc_story.html?utm_term=.9bf9276abb1a. 47. http://www.bamf.de/SharedDocs/Anlagen/DE/Downloads/Infothek/ Statistik/Asyl/201612-statistik-anlage-asyl-geschaeftsbericht.pdf;jsession id=90490C1DD17015A6FD367DE4BEC55CC0.1_cid359?__blob=publi cationFile. 7

THOMAS RUTTIG  |  AFGHAN EXODUS est group in Europe – while a number of governments claimed that the Afghan war was far less destructive than the one in Syria or Iraq and therefore Afghans were mainly»economic migrants«. EU and individual member states concluded a number of multi- and bilateral cooperation agreements on migration with the Afghan government. A framework was set with the finalization of a re-admission agreement, titled the EU-Afghan»Joint Way Forward on Migration,« that was rushed to be signed despite some last-minute hurdles in Kabul before the October 2016 International Afghanistan Conference in Brussels(see detail 48 ; text 49 ). The conference agenda included donor countries’ reconfirmation of financial pledges for the next phase of Afghanistan’s 2014–2024 »transformation« period, providing an opportunity for donor countries to pressure Kabul to agree to take back rejected asylum seekers. As AAN reported 50 at that time, »the organizers of the Brussels conference(…) feared that failure to negotiate a readmission agreement with Afghanistan(…) would leave member countries reluctant to publicly commit to future funding«. 51 While European governments have denied using aid conditionality to achieve this aim, Afghan officials have understood it that way and told various media sources so. 52 Germany, Finland, Sweden, and other countries signed (or renewed) bilateral agreements at the same time. These agreements are designed to create conditions that will allow the repatriation of large numbers of Afghans. Although the EU and German agreements, for example, state that signatories see»voluntary returns« as the priority, they also strongly emphasize the option of»nonvoluntary returns«. 53, 54 . The EU-Afghan»Joint Way For48. https://www.afghanistan-analysts.org/eu-and-afghanistan-get-dealon-migrants-disagreements-pressure-and-last-minute-politics/. 49. https://eeas.europa.eu/sites/eeas/files/eu_afghanistan_joint_way_for ward_on_migration_issues.pdf/. 50. https://www.afghanistan-analysts.org/eu-and-afghanistan-get-dealon-migrants-disagreements-pressure-and-last-minute-politics/. 51. See also this AAN dossier: https://www.afghanistan-analysts.org/thebrussels-conference-on-afghanistan-between-aid-and-migration/. 52. Read one report from Germany’s main TV network here: http://www. tagesschau.de/ausland/afghanistan-fluechtlinge-111.html. 53. More detail about the agreements can be found in this AAN analysis: https://www.afghanistan-analysts.org/eu-and-afghanistan-get-deal-onmigrants-disagreements-pressure-and-last-minute-politics/. 54. The German agreement has not been published; brief official information about the Swedish agreement can be found here: http://www. government.se/articles/2016/10/agreement-between-sweden-and-afghanistan-on-readmission/; A description of the Finnish agreement can be found here: http://www.migri.fi/our_services/customer_bulletins/bulletins_asylum/1/0/the_agreement_with_afghanistan_only_applies_to_ those_who_have_received_a_negative_decision_70331. ward« even includes an option to create the logistical infrastructure to process large numbers of returning Afghans when they arrive in Afghanistan:»Both sides will explore the possibility to build a dedicated terminal for return in Kabul airport.« The figures for rejected Afghan asylum seekers who are legally required to leave Europe are in the tens of thousands. A draft EU paper 55 prepared for the October 2016 Brussels conference on Afghanistan, leaked in March 2016, mentioned that 80,000 Afghans»could potentially need to be returned in the near future« from all member countries. Germany, the largest recipient country, including for Afghans, officially had 12,539 Afghans who were»ausreisepflichtig«(required to leave) in midNovember 2016. 56 Given the over 240,000 Afghan asylum cases pending all over Europe and the average protection rate of slightly over 50 percent, 120,000 more potential»returnees« could emerge. That would bring the EU-wide number up to around 200,000. Based on the memoranda of understanding on returns and readmissions with several EU/Schengen member states, some EU and non-EU countries have been sending back rejected Afghan asylum seekers for some years already. 57 For example, there were»return« flights from 55. http://statewatch.org/news/2016/mar/eu-council-afghanistan-673816.pdf. 56. https://thruttig.files.wordpress.com/2016/11/20161114bureg-antwauf-anfrage-afg-abschiebungen.pdf. 57. These countries were: France(2002), UK(2002), Netherlands(2002), Denmark(2004), Switzerland(2005), Norway(2005), and Sweden (2006, valid until 2009)(see AAN analysis here). In the UK, the Court of Appeal ruled in March 2016 that»removals« to Afghanistan could be resumed after a temporary halt. Between 2007 and 2015, the UK»re-moved« 2,018 formerly unaccompanied Afghan minors after their asylum applica-tions were rejected and after they had turned 18, as this 2016 media report had revealed. Sweden had also temporarily halted deportation for some months in 2016. According to a 2015 Masters paper at a Norwegian university(»Unintended Conse-quences of Deportations to Afghanistan«; not available online, hard copy with the author), Norway started increasing involuntary returns to Afghanistan from 2006 onwards, also including families with children since 2013. Between 2006 and 2014, Norway carried out 762 (37 %)»assisted« and 1,299(63 %) involuntary returns. In July 2016, the Swiss Federal Administrative Court decided that Afghan refugees could not be returned to their country involuntarily. According to December 2016 Swiss media reports, however, an Afghan family with three small children that had been returned to Norway by Switzerland, based on the Dublin regulation, ending up being notified that they would be involuntarily returned to Afghanistan if they did not leave voluntarily. According to the head of Frontex, the European border management agency, alto-gether 42 percent of rejected asylum seekers from all countries of origin are deported from the EU. Frontex personnel were also on board the flight with rejected Afghan asylum seekers when they were returned from Germany in January 2017. 8

THOMAS RUTTIG  |  AFGHAN EXODUS Number of Afghan Deportees – IOM data Returning from Australia Belgium Denmark Finland France Germany Indonesia Netherlands Norway Oman Portugal Sweden Switzerland UK Grand Total 2003–2009 – 1 62 – 39 224 1 71 284 466 1 4 1 2.989 4.143 2010 – – 7 – – – – – 41 – – – – 733 781 2011 – – – – – – – 6 74 – – – – 1.023 1.103 2012 2 9 – – – – – 78 196 – – 5 – 527 817 2013 – 11 – – – – – 61 250 – – 74 – 513 909 2014 3 2 – – – – – 14 437 – – 94 – 404 954 2015 4 2 – – – 1 – 2 88 – – 26 – 89 212 2016 1 3 – 3 – 34 – – 12 – – 26 – 87 166 Gesamt 10 28 69 3 39 259 1 232 1.382 466 1 229 1 6.365 9.085 Germany(34 deportees), Finland(three), and jointly Sweden and Norway(13 or 14, according to IOM all Afghans from Iran) in 2016. Between 2003 and 2016, in total 8,608 Afghans were deported from Belgium, Denmark, Finland, France, Germany, the Netherlands, Norway, Portugal, Sweden, Switzerland, and the UK. 6,365 of them were deported from the UK and 1,382 from Norway. (There were also three non-European countries that have deported Afghans back to their country over the same period: Australia 10, Indonesia 1, and Oman 466.) There is also an increasing number of voluntary returns of Afghan asylum seekers. An IOM official told AAN that 6,864 persons returned voluntarily to Afghanistan in 2016 through IOM-run Afghanistan programs. From 2003 to 2016, there were 22,436 voluntary returns of Afghans from all countries according to IOM, so that the 2016 figure – which is almost one third of all cases – represents a serious increase. Almost half of those returns in 2016 – 3,159 persons – came from Germany. Most returns occurred in the first three quarters of the year, when on average 200 persons returned a week; between September and December this rate drop to less than 100 returns a week. In mid-December 2016, Sweden and Germany started to put their new agreements with Afghanistan into practice. On December 13 th , 2016, some twenty Afghans were returned in a joint Swedish-Norwegian operation. 58 This happened despite the Afghan-Swedish agreement having run into trouble two weeks earlier, when the lower house of the Afghan parliament(the Wolesi Jirga) voted against it on November 30 th , 2016. 59 According to Abdul Qayum Sajjadi, a member of the house’s International Relations Commission, most MPs considered the agreement to be against the Afghan constitution and international human rights conventions as, in their view, its content emphasized deportation rather than voluntary return; the vote was 117 against 6(no abstentions). The Swedish government rejected this view, and the Afghan government, in the person of Deputy Foreign Minister Hekmat Karzai who travelled to Stockholm in early December 2016, ensured Sweden that Kabul would uphold 58. Different figures have been published. The Guardian reported that »13 Afghans were forcibly returned from Sweden(…). That flight also carried nine Afghan citizens from Norway.« The German interior minister, in a letter dated January 9 th , 2017, mentioned that altogether there were 27 Afghans on board(not online, quotes here). 59. A short report here: http://sverigesradio.se/sida/artikel.aspx?program id=2054&artikel=6576737. 9

THOMAS RUTTIG  |  AFGHAN EXODUS the agreement, 60 de facto overruling the parliament. Despite the parliament’s objection, the Afghan authorities authorized the December»return« flight. A few days later, on December 15 th , 2016, Germany repatriated 34 rejected asylum seekers by charter flight to Kabul 61 – all men, about one third of them convicted for crimes. On the same day, according to an official letter from the German interior minister 62 dated January 9 th , 2017, the Netherlands also carried out some »return action.«(IOM data, seen by AAN, however, cannot confirm any deportation from the Netherlands to Afghanistan in 2016 – but there were 110 voluntary returns. The minister’s letter also did not give specific numbers.) In Germany, the forcible return was met by public protests and intra-party controversies, even in the ruling German coalition. A number of MPs from the smaller coalition partner, the Social Democrats, the German parliament’s commissioner for the armed forces 63 and the government’s commissioner for migration 64 – not to mention the opposition and human rights groups – all challenged the government’s claim that Afghanistan was»sufficiently safe« to forcibly return rejected Afghan asylum seekers. These doubts are particularly strong in some governments of Germany’s federal states that consequently refused to put rejected Afghans under their jurisdiction(deportations are in states’ jurisdiction) on the December 15 th flight. The Conference of the States’ Interior Ministers, held in early December 2016, had tasked the government to update its assessment of the Afghan situation, with the support of UNHCR and the International Organisation for Migration. The UNHCR’s official reply, sent to the German states on January 9 th , 2017, diplomatically but firmly contradicted the government’s assessment, stating that it was not in a position to distinguish between safe(r) and unsafe areas. 65 IOM’s 60. http://www.svd.se/atertagandet-av-asylsokande-fortsatter. 61. http://www.dw.com/en/refugee-deportation-from-germany-sparkscontroversy/av-36779177. 62. https://thruttig.files.wordpress.com/2016/10/20161006brief-bmi-anlc3a4nder-abschiebgn-nach-afg.pdf. 63. http://www.zeit.de/politik/deutschland/2016-12/hans-peter-bartelswehrbeauftragter-afghanistan-abschiebungen. 64. http://www.morgenpost.de/politik/article209165461/Oezoguz-nenntAbschiebungen-nach-Afghanistan-zynisch.html. 65. https://thruttig.files.wordpress.com/2016/10/20161006brief-bmi-anlc3a4nder-abschiebgn-nach-afg.pdf. director general, in an interview 66 with a German daily in December 2016, supported the government by saying that some Afghan areas were»sufficiently safe« for returnees. 1.5 The Changing Climate in Recipient Countries Throughout 2015, the growing numbers of arriving asylum seekers put a strain on local social services, particularly in countries with a high per capita rate of arrivals. Local institutions were at times unprepared or unable to cope with the influx. The mood seemed to change to a large degree, from initially generally welcoming to refugees to one of rejection. A recent poll 67 by the Friedrich Ebert Foundation, published in November 2016, showed, however, that 55.5 percent of Germans continued to welcome the fact that Germany had received many refugees, while 86.1 percent still agreed with the statement»People who flee from war should be received in Germany.« At the same time, 52.9 percent supported a capping of refugees allowed into the country(in September, with a different methodology, weekly magazine Focus had 60 percent 68 ). The changes in the general mood and the problems local authorities faced were picked up by anti-immigrant parties throughout Europe, which were already strong or growing in a number of parliaments. Extra-parliamentarian nationalist groups, often with a violent fringe, became more vocal. These two camps partly overlap in various countries, although in different degrees. In Germany, for example, 120 arson attacks were made on asylum seeker accommodations in 2015, 69 increasing to 141 in 2016, according to research by Berlin daily taz. 70 In contrast, the German police(occasionally accused of turning a blind eye to right-wing terrorism) counted 66 arson and four explosives attacks for 2016. Only in 20 cases, the daily writes, did information show that the case was still being investigated. 66. https://www.welt.de/politik/deutschland/article160281515/Afghanistan-in-einigen-Regionen-ausreichend-sicher.html. 67. https://www.fes.de/de/gespaltene-mitte-rechtsextreme-einstellunge n-2016/. 68. http://www.focus.de/politik/ausland/fluechtlingskrise-im-news-ticker -ausschreitungen-zwischen-rechten-und-fluechtlingen-in-bautzen_id_59 46861.html. 69. http://www.taz.de/!5260711/. 70. https://www.taz.de/Archiv-Suche/!5367296&s=brandanschl%C3% A4ge/. 10

THOMAS RUTTIG  |  AFGHAN EXODUS In order to counteract voter losses, some governing mainstream parties changed their rhetoric and tightened their policies and laws on migration. German legislation on asylum, residence, and integration has been amended twice since October 2015(more detail in the case study in part three of this dispatch). A third legislation package that included plans to further reduce in-cash support for individual asylum seekers was rejected by the upper house of parliament on 16 December 2016. 71 Sweden tightened its asylum process to reach what Prime Minister Stefan Löfven, a social democrat, has termed the»EU minimum level« in asylum and migration policy. 72 This means, for instance, that fewer applicants get full asylum rights and only those who do have an unconditional right to family reunification. The measures are meant to be provisional, and the intention is to revert to a more generous approach as soon as the reception situation is deemed to be stabilized. 73 According to Swedish migration lawyers 74 the Swedish Migration Board(SMB) is also using its own version of»safe(r) zones« in Afghanistan, here termed regions»less influenced by war.« Finland is the first EU country where the government practically declared all of Afghanistan, as well as Somalia and Iraq, safe for return. It did so in May 2016, stopping short of literally calling it a»safe country.« The statement from the Finnish immigration service on these three countries 75 says: »In the past few months, the security situation has gradually improved in all three countries, although it may have got[sic] worse at times for certain specific areas locally. Due to the improved security situation, it will be more difficult for applicants from these countries to be granted a residence permit on the basis of subsidiary protection.(…) According to the Finnish Immigration Service, it is currently possible 71. http://www.mdr.de/nachrichten/politik/inland/bundesrat-asylbewerberleistungsgesetz-gestoppt-100.html. 72. http://www.thelocal.se/20151124/sweden-set-to-tighten-asylum-rulesfor-refugees. 73. See more detail in this AAN analysis: https://www.afghanistan-analysts. org/an-afghan-exodus-2-unaccompanied-minors-in-sweden/. 74. https://www.newsdeeply.com/refugees/articles/2016/07/11/afghanrefugees-feel-a-sudden-chill-in-sweden. 75. http://www.migri.fi/for_the_media/bulletins/press_releases/press_releases/1/0/humanitarian_protection_no_longer_granted_new_guidelines_issued_for_afghanistan_iraq_and_somalia_67594. for asylum seekers to return to all areas in Afghani­ stan, Iraq and Somalia without the ongoing armed conflicts as such presenting a danger to them only because they are staying in the country.« By late 2015, Finland had already stopped giving subsidiary protection to Afghan asylum seekers from the provinces of Helmand, Khost, Paktika, Uruzgan, and (additional) parts of Ghazni. 76 It also seriously tightened asylum policies, abolishing the law that allowed for providing refugees status on the grounds of»humanitarian protection« and made family reunification more difficult. For family reunification to be possible, the whole family now needs to be legally in Finland at the time the application is filed. In September 2016, Afghan refugees in Finland 77 demonstrated against what they perceived as an unfair asylum process and demanded that their cases be heard and processed according to international norms. Denmark’s much tightened asylum laws 78 have even been criticized by the UN, as they now include provisions for detaining asylum seekers without a court order. It also makes family reunion more difficult, involuntary return easier, and allows the confiscation of asylum seekers’ money and jewelery worth more than 1,350 Euros. The country is reportedly 79 planning even more radical regulations, in a so-called»general plan for a stronger Denmark.« According to this plan, in a»crisis situation« the government could close the border for all asylum seekers; the granting of permanent residence would be delayed(taking place after eight instead of six years) and only granted after the refugee had not claimed social welfare benefits for four years; family reunification would only be granted after eleven years and child benefits after five years; and the threshold for withholding permanent residence because of a conviction was lowered(from twelve to six months detention). A case from Denmark that technically was a voluntary return and was recounted in a 2015 Guardian article 80 ; 76. http://www.migri.fi/for_the_media/bulletins/press_releases/press_releases/1/0/subsidiary_protection_no_longer_granted_to_all_asylum_ seekers_from_south_and_east_afghanistan_64229. 77. https://kabulblogs.wordpress.com/sample-page__trashed/picture/. 78. http://www.taz.de/Fluechtlinge-in-Daenemark/!5319625/. 79. https://www.taz.de/!5332780/. 80. https://www.theguardian.com/global-development/2015/oct/06/tragic-tale-afghan-brothers-sent-home-from-denmark-to-an-uncertain-fate. 11

THOMAS RUTTIG  |  AFGHAN EXODUS this case demonstrated how such a practice can go wrong. In this case, two Hazara brothers from MaidanWardak province(one adult, one minor) had their asylum applications rejected in 2012 and agreed, under some prompting, to voluntarily return to their country in June 2015. The Danish authorities argued that the elder could act as the younger’s guardian. Both ended up sleeping in the streets of Kabul. The younger one disappeared when they tried to obtain ID cards in their native province and was later reported killed. After that, the older brother moved to Iran and from there, as it is assumed in the article, possibly back to Europe. In Austria, the parliament decided in June 2016 that the government could request that no new asylum applications be accepted after an annual ceiling of 37,500 was reached. 81 A UNHCR spokesman called this»breaking a taboo,« as the legislation 82 also summarily equated migrants with a»threat«. After the threshold is reached, only asylum requests by refugees with close relatives already living in the country or who are threatened by torture or other inhuman treatment upon return, will be accepted. With 42,073 asylum requests in 2016, figures went down by more than half, compared to the 88,900 cases in 2015. But since less than two thirds of these applicants were admitted only for the asylum procedure, numbers remained under the ceiling(even with 8,800 pending cases from 2015 added) and did not trigger the new measures. In mid-2016, the Austrian foreign minister proposed an»Australian solution« for migrants entering the EU: keep them on the Greek islands until their cases have been decided. 83 But this might have been part of the hard fought presidential run-off election campaign, with a right-wing populist as one of the candidates(he narrowly lost in the end). Hungary, which in 2015 was the European country with the second highest number of overall(174,435) and Afghan(45,650) asylum applicants as well as the country with the highest per capita number of all asylum seek81. http://www.focus.de/politik/deutschland/noch-restriktivere-regeln-mitnotstand-massnahmen-schottet-sich-oesterreich-weiter-gegen-asylbewerber-ab_id_5589708.html. 82. http://derstandard.at/2000043891373/Die-Gefahren-der-Notverordnung. 83. http://derstandard.at/2000042318038/EU-Tuerkei-Deal-Es-gibt-keinen-Plan-B. ers(17,699), took the most draconian measures to bring down the amount of migrants in the country. In September 2015, it rigorously closed its border with the main influx country, Serbia. This left only two official border crossings open, through which small but even further decreasing numbers of migrants were allowed in(mainly families). In October 2015, the border with Croatia followed. It also was the first EU country to start entirely fencing the vulnerable parts of its border. 84 Overall numbers of asylum seekers in Hungary dropped to 28,803 in 2016, 38 percent of them Afghans(almost 11,000). 85 On January 13 th , 2017, the government additionally introduced mandatory detention for all asylumseekers with pending cases in the country in so-called transit zones. Prime Minister Viktor Orban said on the radio,»we have reinstated alien police detention in the cases of those whose applications to enter Europe have not yet been legally judged.« This is against EU law, which allows such a measure only in»exceptional cases.« In early October, however, the government failed to secure a referendum vote for its proposal to close the country for all refugees; although 98 percent of participants were in favor, voter participation fell short of the legally required 50 percent threshold. 86 In November, parliament voted narrowly against the move. 87 2. The North-South Divide The situation and number of Afghan migrants in Europe differed from country to country in 2016. The division lay, roughly, along the Alps. To the south, the number of incoming migrants, though still high, dropped while requests for asylum continued to rise in some countries. Living conditions, meanwhile, deteriorated sharply. To the north, significantly fewer new Afghan migrants arrived – particularly after the March 2016 EU-Turkey deal on migration – while the number of asylum requests grew in certain countries and 84. See more detail here: https://www.afghanistan-analysts.org/afghanexodus-the-opening-and-closing-of-the-balkan-corridor/; and in this AAN dispatch: https://www.afghanistan-analysts.org/afghan-exodus-notesfrom-a-belgrade-squat/. 85. http://www.helsinki.hu/wp-content/uploads/HHC-Hungary-asylumfigures-1-December-2016.pdf. 86. https://www.theguardian.com/world/2016/oct/02/hungarian-voteon-refugees-will-not-take-place-suggest-first-poll-results. 87. https://www.theguardian.com/world/2016/nov/08/hungarian-parliament-narrowly-rejects-pms-migrant-resettlement-ban. 12

THOMAS RUTTIG  |  AFGHAN EXODUS fell sharply in yet others. The general treatment of and sentiment towards migrants became less generous. Among those Afghans stuck along borders in the south or threatened with deportation in the north, hopelessness has been growing. 2.1 The Situation for Afghans in Mediterranean Countries As a result of tighter border controls and stricter migration policies, many refugees are now stuck between the almost hermetically closed outer borders of the EU or between individual EU countries. A significant number of them are Afghans; most of them are now stranded in Turkey, Greece, and Serbia. The countries on the Iberian Peninsula do not play a substantial role here, as they are too far from the main entry route across the Aegean Sea in the eastern Mediterranean region. 2.1.1 Turkey Turkey hosted between 111,000 and 160,000 Afghan migrants in the summer of 2016. 88 As AAN reported 89 in September 2016, they came with different strategies and aims. Many thousands of them have stayed in Turkey and built an expatriate community that both aids and exploits those passing through. Some of them diversified their»business« following the EU-Turkey deal, branching out into a broad array of activities, from renting out accommodation, arranging jobs for their compatriots, to drug-running. Others have opted for legal resettlement in Turkey. The country operates several»deportation centers«, including in Pehlivanköy in the European part, Erzurum on the north coast, 90 as well as in the extreme east, near the borders with Syria and Iran. Access for UNHCR, journalists, and volunteers is limited. 88. An AAN dispatch by guest author Noah Arjomand in September 2016 pointed to UNHCR statistics according to which there were 3,109 Afghan refugees and 107,655 Afghan asylum seekers in Turkey at the end of July 2016. A July 2016 report by Am-nesty International(AI) report said that»Turkey hosts more than 400,000 non-Syrian refugees« while a European Parliament document from December 2016 estimated that 40 percent of non-Syrian refugees in Turkey were Afghans. Putting these two figures together, that would bring the number of Afghans in Turkey to over 160,000. 89. https://www.afghanistan-analysts.org/afghan-exodus-smuggling-net works-migration-and-settlement-patterns-in-turkey/. 90. http://www.woz.ch/-6ce2. 2.1.2 Greece Greece has become one of the main victims of the EU’s failure to develop a distribution system for arriving asylum seekers among its member states. While more asylum seekers arrived in 2016(although in lower numbers than in 2015), only a small number of them were relocated to other EU countries. 91 As a result of this failure – as well as the fences built along parts of its Turkish land border near the triangle with Bulgaria – there were 63,000 migrants stuck in Greece 92 as of December 2016, 49,000 of them on its mainland in over forty camps. Around 3,000 of the total were children. 93 Accommodation facilities are overcrowded, 94 with people sleeping outdoors and many without access to drinking water. 95 This is particularly the case on the Greek islands near the Turkish coast from where most of the Afghan migrants in 2015 crossed over into the EU, but also in the capital, Athens. 96 It has been repeatedly reported that under-age refugees in these camps, among them Afghan boys, have been forced into the sex trade. 97 The EU does not want Greece to ferry any migrants to its mainland, as this could be interpreted as a reopening of the Aegean route. 98 The overcrowded conditions have led to several riots in camps, 99 growing tensions with parts of the local population, 100 and attacks by anti-immigrant groups. 101 According to a media report in December 2016, 13,000 91. For more figures, see part one of this dispatch: http://aan.af/2kMGO8U. 92. http://www.wsj.com/articles/migrants-in-greece-drop-off-grid-1481 106606. 93. https://www.theguardian.com/world/2016/sep/10/child-refugeesgreece-camps. 94. http://www.usatoday.com/story/news/world/2016/04/08/afghan-re fugees-trapped-limbo-greece-nowhere-go/82745148/. 95. http://www.taz.de/!5326485/. 96. This website has some vivid visual impressions about the situation on the Greek mainland: https://www.thisamericanlife.org/greece/#baseballstadium. 97. http://edition.cnn.com/2016/11/29/europe/refugees-prostitution-teen agers-athens-greece/. 98. http://www.spiegel.de/politik/ausland/fluechtlinge-in-griechenland-eustaaten-fuerchten-um-sicherheit-ihrer-beamten-a-1118701.html. 99. http://www.dw.com/en/refugees-clash-with-authorities-in-bulgariangreek-camps/a-36518439. 100. http://www.latimes.com/world/europe/la-fg-greece-chios-refugees2016-story.html. 101. http://www.independent.co.uk/news/world/europe/refugee-crisislatest-greek-islands-chios-camp-attack-far-right-firebombed-molotovcocktails-rocks-a7425386.html. 13

THOMAS RUTTIG  |  AFGHAN EXODUS of those registered in Greek refugee camps are unaccounted for and could have slipped further north into Europe, according to European immigration officials. 102 At the same time, there is still a wide array of volunteer support for the migrants. 103, 104 Again, there is no official data on how many Afghans are among the migrants in Greece. 105, 106 The number of Afghan asylum seekers was relatively low, the 2015 figure was 1,545. 107 Given what is known, Afghans make up a more significant number of those staying in Greece. According to the UNHCR, 108 one measure by the Greek asylum authority was important»for Afghans in particular«: a re-registration campaign that was started on June 8 th , 2016 open to those who entered Greece between January 1 st , 2015 and March 20 th , 2016. As a result, over 15,500 asylum-seekers on the Greek mainland received temporary cards, valid for one year, that allowed them to reside legally in Greece while awaiting a final decision on their asylum applications. It also gives them the right to access services and should help identify those eligible for family reunification or relocation. The particular importance for Afghans point to their significant number, but also to their dire situation as, according to the UNHCR, the initial entry documents of most of them, known as»police notes,« had expired. As a result, their presence in Greece had technically become illegal, which could have resulted in arrest and possible deportation. A likely result of this was that Greece had the second largest number of Afghans voluntarily returning to their country in 2016 after Germany; this number rose from 152 in 2015 to 1,257 in 2016, according to IOM figures. 102. http://www.wsj.com/articles/migrants-in-greece-drop-off-grid-148 1106606. 103. https://www.theguardian.com/society/2016/dec/09/help-refugeeswe-will-never-abandon-them. 104. http://www.irishtimes.com/news/world/europe/migrants-remainstranded-in-greece-in-spite-of-turkey-deal-1.2878051. 105. https://www.theguardian.com/world/2016/sep/10/child-refugeesgreece-camps. 106. http://www.euronews.com/2016/09/09/greece-afghan-migrants-inlimbo. 107. Eurostat only publishes the top three countries of origin for each EU member-state per quarter. There, Afghanistan was in the top three for Greece in the third quarter of the year only(with 670 applications); in the first and second quarter, Afghanistan had less than 480 applications. 108. http://www.unhcr.org/news/latest/2016/7/577687af4/15500-asylumseekers-pre-registered-mainland-greece.html. Many Afghans are thought to have applied for these cards mainly to avoid possible deportation to Turkey, as many still aim to travel onwards if the chance arises. Deportations from Greece to Turkey have, however, not happened – apart from a few exceptions 109 – as Greece does not consider Turkey a safe third country. 2.1.3 Italy In Italy, Afghans have not even been among the top ten nationalities of asylum seekers since 2012. 110 Their numbers have grown steadily, however, over the last few years, peaking in 2015 with 3,975 applicants. The closure of the Balkan route in early 2016 halted that trend again. Asylum requests by Afghans per month fell from 665 in January 2016 to 118 in August. Although their number started to grow again in later months, altogether fewer Afghans are likely to have applied for asylum at the end of 2016, compared to 2015. Numbers of Afghan asylum seekers may be relatively low, however the recognition rate for them in Italy is high(over 97 percent in 2015, with 3,280 Afghans granted protection). Most of the Afghans arrive and apply for asylum in north-eastern Italy. Trieste, and on a smaller scale Udine and Gorizia, on the eastern border with Slovenia, host a comparative majority of Afghan refugees. 111 Afghan asylum applicants usually wait around six to nine months before their asylum hearing. After the recognition, the duration of state support can vary from a few days to more than a year, depending on the area and the type of reception facility in which the refugees are hosted. 112 Some Italian prefectures allow them to remain in the reception system with the same benefits granted before the hearing for up to six months after recognition, while others urge them to become fully independent the very same day they are issued their asylum documents. Only a fraction of those who receive protection can, once they exit this primary reception sys109. https://www.theguardian.com/world/2016/nov/01/syrian-refugeestricked-into-returning-to-turkey-greece-eu. 110. http://www.asylumineurope.org/sites/default/files/resources/anci_rap porto2016.pdf(p. 89). 111. One unlucky Afghan asylum seeker was killed in the summer 2016 earthquake in Amatrice, in central Italy 112. In Italy, there is no uniform reception system. State first contact reception centers can be managed by public local entities, consortia of municipalities, and other public or private bodies specializing in the assistance of asylum applicants(more detail here). 14

THOMAS RUTTIG  |  AFGHAN EXODUS tem, access nation-wide asylum-specific projects(such as SPRAR), that provide refugees with additional state support for up to one year. In addition to the Afghans who travelled directly to Italy, there is a sizeable back-flow of»Dublin cases« 113 from central and northern European countries. The BBC reported in September 2016 that in the northern province of Udine alone, about 5,000 migrants entered from Austria since the start of 2016 alone;»about 90 % of them (…) from Pakistan or Afghanistan« and»the overwhelming majority« were young men. 114 Most of these Dublin cases eventually obtain protection in Italy, at the price of longer waiting times and considerable stress over the fear of being sent back yet again if there is another country of first entry, from where they would often face a further deportation to Afghanistan. For the most part, Afghans asylum seekers in Italy were, until now, transitory refugees. Many of whom, even after they had obtained their asylum documents, continued to try to reach Scandinavian countries, Germany, or the UK. Apart from some early Hazara refugees who came in the 1990s, Italy does not have a large Afghan diaspora into which substantial numbers of newcomers could easily integrate and access the job market. Although this may slowly be changing, especially in big cities such as Rome or Milan, these communities’ capacity may not be sufficient to accommodate the growing number of Afghans with Italian asylum documents who have returned in the last two years after facing increasing difficulties in finding residence and work – even informally – in other European countries and who, in Italy, are now quickly being exited from the reception system. 2.1.4 Serbia Serbia hosted between 6,200 and 10,000 migrants by the end of November 2016, as more continued to arrive despite the closure of the Balkan route in early 2016. 115 113. This term refers to asylum seekers in the EU who, according to an EU regulation, can be sent back to their first EU country of entry(if registered there) if they apply for asylum elsewhere. This regulation was adopted in Dublin in 2003(see more here). 114. http://www.bbc.com/news/world-europe-37475086. 115. See AAN reporting here: https://www.afghanistan-analysts.org/afgh an-exodus-notes-from-a-belgrade-squat/. By October 31 st , 2016, the Serbian Asylum Office had registered 10,201 individuals who expressed their intention to seek asylum, of whom 4,447 were Afghans. According to Serbian policy, a foreigner can express »the intention to seek asylum«; s/he is then»recorded« (rather than registered). The asylum seeker then needs to report to an asylum official or asylum center within 72 hours to register the actual request. 116 A recent media report from Belgrade said that, according to the local branch of Save The Children, on average 100 additional refugees entered the country per day throughout December 2016, many of them Afghans. 117 In total, 40 percent of the refugees were children and one quarter of these children were unaccompanied; an estimated 75 percent of the unaccompanied children came from Afghanistan. The newspaper reported on a group of children who were respectively three, nine, ten, and eleven years old. In August of 2016, it was reported that»a hunter« in Serbia shot a 20-year old Afghan refugee who had illegally crossed the Bulgarian border. 118 UNHCR Serbia, in its updated report from December 2016, 119 said that it and its partners had»encountered« around 6,900 refugees, asylum-seekers, and migrants in the country. Over 5,500(i. e. 80 %) were accommodated in thirteen governmental facilities, while the remainder were sleeping rough in Belgrade’s city center or at the border with Hungary. UNHCR Serbia estimates that 25 percent of the former(ca. 1,500) are Afghans, while they constitute a majority of the latter(i. e. at least 700). 2.1.5 Bulgaria Bulgaria, where a small part of the border with Turkey has not yet been fenced, 120 has become one of the last entry points into the EU from Turkey. It is not an easy ac116. https://www.afghanistan-analysts.org/afghan-exodus-in-transitthrough-serbia/. 117. http://www.taz.de/Kaeltewelle-in-Suedosteuropa/!5371809/. 118. http://www.dw.com/en/afghanistan-refugee-shot-dead-in-serbia/a19497602. 119. http://reliefweb.int/report/serbia/unhcr-serbia-update-12-14-decem ber-2016. 120. http://www.independent.co.uk/news/world/europe/europe-refugeesfreeze-to-death-hypothermia-bulgaria-athens-cold-weather-serbia-sleeping-rough-a7520106.html. 15

THOMAS RUTTIG  |  AFGHAN EXODUS cess route. The Politico blog called the country»Europe’s most hostile port of entry.« 121 By November 2016, Bulgaria reportedly had 13,000 migrants within its territory,»most of them Afghans.« 122 There is a growing number of reports about sub-standard government facilities for them as well as maltreatment by security forces. Human Rights Watch 123 , Oxfam 124 , and other organizations have reported on how Bulgarian law enforcement officials subject asylum seekers to violence at the Turkish, Romanian, and Serbian borders. Refugees regularly report beatings, dog bites, having their money and personal belongings stolen, and a»lack of adequate food and unsanitary conditions« in detention facilities. A number of migrants AAN encountered in Belgrade in June and November 2016 reported similar incidents. 125 The Bulgarian government, like the Hungarian government 126 , further condones paramilitary vigilante groups 127 that hunt illegal migrants. Some of them are self-employed and yet others funded by the government. These groups even attract activists from other EU countries’ right-wing nationalist groups 128 and are regularly accused of violence against migrants. In November, riots broke out at Bulgaria’s largest camp, Harmanli, near the Turkish and Greek border. 129 It was inhabited, at that point, by 3,000 people, most of them reportedly Afghan. 130 The place had been beset by anti-immigrant groups, and the authorities had reacted by curbing the migrants’ right of movement. Following 121. http://www.politico.eu/article/bulgaria-threat-to-refugees-migrantshuman-rights-dangerous/. 122. http://www.aljazeera.com/news/2016/11/bulgaria-refugee-proteststreatment-turn-violent-161124203041244.html. 123. https://www.hrw.org/news/2016/01/20/bulgaria-pushbacks-abuseborders. 124. http://www.oxfam.org.uk/media-centre/press-releases/2015/11/refugees-crossing-into-europe-tell-of-abuse-at-hands-of-bulgarian-police. 125. See report here: https://www.afghanistan-analysts.org/afghan-exodus-the-re-emergence-of-smugglers-along-the-balkan-route/; and also here: https://www.afghanistan-analysts.org/afghan-exodus-notes-froma-belgrade-squat/. 126. http://www.ibtimes.co.uk/road-hungarys-far-right-vigilante-migranthunters-1581483. 127. http://bulgaria.bordermonitoring.eu/2016/04/13/bulgarian-statesupports-racist-groups-which-are-hunting-for-migrants/. 128. http://www.taz.de/!5323792/. 129. http://bulgaria.bordermonitoring.eu/2016/11/24/the-provoked-riotin-harmanlis-refugee-camp/. 130. http://www.bbc.com/news/world-europe-38090753. the riots, the Bulgarian government took a number of measures to lower the number of migrants. Similarly to Greece, Bulgaria started urging incoming migrants to apply for asylum upon arrival. As a result, applications increased by 82 percent from the second to the third quarters of 2016, to an absolute figure of 6,365 – almost half of the new applicants(3,145) were from Afghanistan. 131 It also started pushing for a bilateral readmission agreement with Afghanistan that would allow it to send back rejected asylum seekers. 132 According to media reports, Bulgaria cooperates closely with Turkey: Turkey takes back refugees who pass the bilateral border illegally, are picked up on the Bulgarian side, and immediately returned. It is unclear which refugees are allowed in to request asylum and which are immediately returned. 2.1.6 Romania EU member-country Romania did not play much of a role as a transit country while the Balkan route was open. Reaching Romania via Bulgaria would require crossing the River Danube. Throughout 2015, 96 Afghans filed an asylum application in Romania, out of a total of 1267 applicants. 133, 134 Figures dropped even further in 2016, with only ten Afghans applying in each of the first and the second quarters and 30 in the third quarter of 2016. Romania could potentially become part of a secondary route, due to the daily changes in the movement strategies in the Balkan countries, as it has not yet closed its borders. Romania – in line with Hungary, Slovakia, and the Czech Republic – voted against compulsory EU reception quotas for asylum seekers in September 2015. 135 On a national level, however, measures were taken to raise the capacity on the border and to offer basic supplies as well as medical and humanitarian aid, in case more migrants come. 131. http://www.balkaneu.com/first-time-asylum-applicants-bulgariaincrease-82-q3-2016-6365/. 132. http://www.bta.bg/en/c/DF/id/1471667. 133. http://www.unhcr-centraleurope.org/_assets/files/content/resourc es/_pdf_en/statistics/2015/asylum-trends-romania-2015.pdf. 134. http://www.hrl.sk/sites/default/files/files_downloads/country_presen tation_romania.pdf. 135. See here for political background: https://www.iedonline.eu/download/2016/IED-Mihai-Sebe-Working-Paper-2016-update.pdf. 16

THOMAS RUTTIG  |  AFGHAN EXODUS 2.1.7 Croatia Croatia was a major transit country along the Balkan corridor, but has become relatively quiet again following the closure of the Balkan route. There is some permeability at the border between Serbia and Croatia, 136 but only a small number of Afghans applied for asylum in Croatia in 2015. Of the six who did so in 2015, four were rejected, one was given refugee status, and one case seems still to be pending. 137 In 2016, the number of applications rose to 370(first to third quarter). 2.2 The Central Divide: Austria and Hungary Austria and Hungary constitute the divide between the south and the north, but at the same time the Balkan route extends into both countries. Both countries were among the top receivers in 2015 and were still processing large numbers of that year’s asylum seekers in 2016. Hungary adopted a very harsh attitude to prevent new arrivals from coming in, while Austria took a comparatively more moderate stance. 2.2.1 Austria In 2015, Austria received the fourth-highest number of asylum applicants(88,900) from all countries of origin, 138 following Germany, Sweden, and Hungary. In 2016, although numbers of all asylum seekers, including Afghans, dropped, it remained the fifth largest recipient country, and the second largest for Afghans in terms of new asylum requests. It was unclear how many of these individuals ended up staying in the country. In 2015, half of all migrants entering Austria subsequently left the country again, according to official government figures. 139 Numbers of Afghan asylum applicants went down by more than half, from 24,480 in 2015(among them 4,000 unaccompanied minors in the first half of the 136. See AAN reporting here: https://www.afghanistan-analysts.org/afgh an-exodus-notes-from-a-belgrade-squat/. 137. http://www.asylumineurope.org/reports/country/cROATIA/statistics. 138. https://www.caritas.at/aktuell/news/detail/news/75470-caritas-bilanz1-jahr-hilfe-fuer-menschen-auf-der-flucht/. 139. http://derstandard.at/2000042820469/Ueber-200-000-Zuwanderer-im-Jahr-2015. year alone 140 ) to 11,289 by the end of November 2016. The number of unaccompanied minors dropped particularly sharply to 287. 141 Afghans, of whom there is a 35,000-strong community in Austria – are relatively well integrated. In mid-2016, 3,800 of them(11 %) had taxable jobs. 2.2.2 Hungary Hungary, in 2015, received a total of 174,400 asylum applications – the second-most of any European country. Of the total asylum applications, 45,600 came from Afghanistan. Most of those who had entered Hungary in 2015 never intended to stay, transiting Hungary on their way to Western Europe, without ever registering. In 2016, Hungary dropped out of Europe’s top ten, receiving 28,803 asylum applications, 38 percent of them from Afghans(almost 11,000). 142 This decrease was largely a result of Hungary’s decision to fence its entire southern borders(with Serbia and Croatia) to implement toughened laws that, in essence, violate EU legislation. In July 2016, a new law came into force that allows the Hungarian police to automatically»push back« anyone who is caught within eight kilometers of the border – without registering their data or allowing them to submit an asylum claim. In early 2017, the government introduced mandatory detention for all migrants that began the asylum procedure. Of the 2015 applicants, only 146 were granted asylum, according to government statistics. 143 Another 362 were permitted to stay, but unlike recognized asylum seekers, they did not receive state subsidies. The comparatively low number of asylum applicants in Hungary is very likely limited to those who managed to enter illegally, were caught, and then asked for asylum. With the border closed on the Serbian side, Hungary still has a small»opening« through which migrants can 140. http://diepresse.com/home/politik/aussenpolitik/4832008/Asyl_Afghanen-die-verdraengten-Fluechtlinge-. 141. http://www.bmi.gv.at/cms/BMI_Asylwesen/statistik/files/2016/Asylstatistik_November_2016.pdf. 142. http://www.helsinki.hu/wp-content/uploads/HHC-Hungary-asylumfigures-1-December-2016.pdf. 143. http://www.aljazeera.com/indepth/features/2016/03/refugee-centresfill-hungary-tightens-borders-160310201559275.html. 17

THOMAS RUTTIG  |  AFGHAN EXODUS enter to apply for asylum – but in very limited numbers and under extremely harsh conditions. Since October 2016, a decreasing number of migrants – currently 20 migrants per working day(a maximum of 100 per week, down from originally around 700 per week) – are allowed to register an asylum request at the Horgoš and Kelebija border crossings. The process prioritizes families with children and unaccompanied women, as they have the greatest chance of success, while largely overlooking single men(who constitute the majority of Afghans in Serbia). The number of overall asylum cases registered in Hungary amounted to 1,610 cases in the third quarter of 2016, dropping significantly from the first(6,830) and second quarters(14,915). These asylum claims at the border can, according to Human Rights Watch, be dismissed under Hungarian law without any consideration of the merits of the case, and often are dismissed within the space of a single day, since Hungary has declared Serbia a safe third country. So far, Hungary is the only EU country to do so. In early 2017, reports emerged that migrants were being kept in the»no-man’s land« right next to the border fence in freezing temperatures. 144 2.3 Afghans in Select Northern EU Countries After the closure of the Balkan route and the implementation of the EU-Turkey deal in early 2016, the number of asylum seekers dropped significantly in the EU countries north of the Alps. This includes the three Nordic EU member-countries, the three Benelux countries, and Austria, all of which had registered particularly high numbers in 2015 when large numbers of migrants were ferried through the Balkans and into the EU. In the northern and some north-western EU countries, the numbers of Afghan asylum seekers dropped significantly between the fourth quarter of 2015 and the first quarter of 2016. This seems to have been a result of both border controls reinstated in late 2015 and, possibly, a quicker registration process than in 2015. Further south, mainly in Germany and Austria, migrants continued to arrive in large numbers. Between 2015 and 2016, numbers in Finland decreased by 89 percent, from 4,300 to 490; 144. http://www.reuters.com/article/us-europe-migrants-serbia-hungaryidUSKBN14T1PV. in the Netherlands by 69 percent, from 1,950 to 600; in Denmark by 58 percent, from 1,680 to 620; and in nonEU Norway by 97 percent, from 4,905 to 150. Over the second and third quarters, these figures dropped even further: in Finland to 60 and 80; in Norway to 80 and 85; in Denmark to 280 and 130; in the Netherlands to 170 (third quarter figures were not available, as Afghanistan was no longer in the top-five countries of origin). The same was true in Sweden, although on a higher level, where numbers dropped by over 90 percent, from over 41,500 in total in 2015 to 2,969 in 2016. 2.3.1 Sweden Sweden closed its borders and tightened its asylum laws in general, leading to a general drop in asylum applications, including from Afghans. The 2016 figure of Afghan asylum seekers was closer to the 2014 level when 3,104 Afghans lodged such an application. In January 2017, this figure was at 193 applications, suggesting that the 2016 level has stabilized. In the peak year of 2015, Sweden had a particularly high number of Afghan minors who applied for asylum. These 23,480 cases represented more than half of all Afghan cases. 145 This figure dropped to 665 in 2016. 146 Between January and October 2016, residency permits were granted in 44 percent of Afghan asylum cases. For January 2017(with 669 cases decided), this rate was almost unchanged at 45 percent. 147 In the same months, the acceptance rate for minors was 82 percent(with 191 cases, Afghanistan was the country of origin for almost half of all 439 asylum cases of minors). By February 1 st , 2017, the country had altogether 36,895 Afghans living in the migration authorities’ reception centers, among them 17,195 unaccompanied minors. 148 A reassessment by the Swedish government of the security situation in Afghanistan(in the form of a directive 145. See an earlier AAN analysis here: https://www.afghanistan-analysts. org/an-afghan-exodus-2-unaccompanied-minors-in-sweden/. 146. https://www.migrationsverket.se/Om-Migrationsverket/Pressrum/ Fokusomraden/Afghanistan-och-ensamkommande-barn.html. 147. https://www.migrationsverket.se/download/18.4100dc0b159d67d c6146d1/1485958543871/Avgjorda+asyl%C3%A4renden+2017+-+As ylum+decisions+2017.pdf. 148. A table showing the number of Afghan asylum applicants in the country between 2000 and 2015 can be found here: https://www.migrationsverket.se/download/18.2d998ffc151ac3871598615/1485556064255/ Asyls%C3%B6kande+2000-2015+samtliga+medborgarskap.pdf. 18

Asylum Seekers in Sweden from 2010 to 2017 THOMAS RUTTIG  |  AFGHAN EXODUS from the migration authority 149 ), however, concluded that security had deteriorated overall, but that the conflict affected different parts of the country and different population groups in different ways. At the same time, the Swedish public perception about and compassion for Afghans in Sweden deteriorated due to the involvement of Afghan asylum seekers in some highly publicized crimes, including battering and sexual offences. 150 In 2016, the government created some specific programs to improve the life of some of those who arrived as unaccompanied minors. It suggested that the minors whose asylum applications had been rejected and who would be deported when they reached 18 years of age could stay to finish their secondary schooling. 151 They would also be granted residency if they were able to find employment. By the end of November 2016, around 1,600 asylum applications by minors were approved, while around 500 were rejected. The government’s suggestion would only apply to those who were already in secondary school. The suggestion needs parliamentary 149. http://www.migrationsverket.se/download/18.2d998ffc151ac3871 59194c1/1481191096023/RCSR5916.pdf. 150. See media reports here: http://www.expressen.se/nyheter/fem-tonaringar-doms-for-grov-gruppvaldtakt/; http://www.dn.se/nyheter/sverige/15arige-namir-slogs-ned-och-valdtogs-pa-en-aker/; http://www.expressen.se/ kvallsposten/under-tio-minuter-var-det-katastrof/. 151. https://www.afghanistan-analysts.org/an-afghan-exodus-2-unaccom panied-minors-in-sweden/. approval. By the end of November 2016, around 1,600 asylum applications by Afghan minors were approved, while around 500 were rejected. This indicates a substantive backlog of such cases that still need to be processed. The tighter asylum laws in general, together with an increasing tendency by the asylum authorities to carry out age reviews resulting in»minors« being re-defined as»adults« and thereby eligible for deportation, has put increasing pressure on young Afghan asylum seekers. 152 The fact that some of the Afghans resided in Iran before attempting to seek asylum in Sweden, but will be deported to Afghanistan if their asylum claims are rejected, adds to the pressure. 153 Groups working with asylum seekers, including the non-profit organization Ensamkommandes förbund and the network of Vi står inte ut have warned against depression, suicide attempts, and suicides among especially young male Afghan asylum seekers. Reuters reported on the suicide of a young Afghan already last year. In April 2016, Mustafa Ansari committed suicide in the center for young asylum-seekers in the southern Swedish village of Svangsta. The report said:»Ansari, who had no papers[…] was 152. http://www.svt.se/nyheter/lokalt/vast/flera-sjalvmord-bland-ensamkomman. 153. http://www.bbc.com/persian/afghanistan-38913284. 19

THOMAS RUTTIG  |  AFGHAN EXODUS described in the autopsy as 17« and that»he was suffering from depression and bipolar disorder. Friends say he desperately missed his family. He waited months for a meeting to process his claim, but the agency cancelled one meeting and messed up the venue for the other«. 154 Later in 2016, one of the main Swedish newspapers, Da­ gens Nyheter reported that close to 40 percent of the unaccompanied minors(many of whom are Afghans) seeking psychiatric support with health services in Stockholm had suicidal thoughts. 155 Reuters quoted Swedish migration agency records that showed asylum-seekers threatened or attempted suicide at least 500 times between January 2014 and end-August 2016. 156 The trend has continued in 2017. Ahmad Zaki Khalil, an Afghan working with asylum seekers in Sweden told the BBC’s Farsi service on February 8 th , 2017 that the three last suicides happened in January and on the 4 th and 7 th of February. 157 He was quoted as saying that he believed the lack of papers that proof they were minors might have been the reason for the three youth’s suicides. On February 9 th , 2017, the website Norway Today quoted Mahboba Madadi from Ensamkommandes förbund;»in recent weeks, seven people attempted to commit suicide and three of them succeeded. They were all from Afghanistan, all boys[…].« 158 The migrants were all under 18 and were at different housing centers across Sweden. In early February, the Swedish mainstream daily Göteborgs Posten raised an alarm that the suicides were not only planned individually, but that»group suicides« among »refugee children« were planned over social media. 159 2.3.2 Netherlands The Netherlands has had a relatively low number of Afghan asylum seekers. During 2015, a total of 2,680 Afghans requested asylum(6.0 % of all 45,035 cases) ac154. http://www.reuters.com/article/us-europe-migrants-sweden-specialreport-idUSKCN11S0W4. 155. http://www.dn.se/nyheter/sverige/sjalvmordstankar-vanligt-blandunga-ensamkommande/. 156. http://www.reuters.com/investigates/special-report/europe-migrantssweden/. 157. http://www.bbc.com/persian/afghanistan-38913284. 158. http://norwaytoday.info/news/suicide-wave-amongst-young-afghanasylum-seekers-sweden/. 159. http://www.gp.se/nyheter/sverige/larm-om-sj%C3%A4lvmord-villplanera-tillsammans-1.4150520. cording to government figures. 160 This number includes first-time asylum requests(2,550), repeated requests (310), and requests for family reunification(85). In 2016, until November 30 th , the total number of asylum requests had dropped by more than half, to 18,695, while Afghan cases decreased slightly less by percentage – to 1,345(7.2 %). Of these, 1,010 were first requests, 335 were repeated requests, and 50 were family reunifications. Afghan asylum seekers were, for a brief while, in the Dutch top three countries of origin in the first quarter 2016(with 600 applicants), whereas they were not even among the top five throughout 2015. With around 44,000 people, the Netherlands hosts one of the largest Afghan communities in Europe. There are 33,058(76 %) first generation arrivals, while 10,674 (24 %) are second generation, meaning they were born in the Netherlands(this figure is from January 1 st. , 2015). The Netherlands(together with Germany) hosts a relatively high proportion of the PDPA elite, many of whom left Afghanistan in the 1990s. Due to a strict implementation of article 1F of the Refugee Charter, 161 all Afghans who worked for KhAD, the intelligence service under the communist government, or who are otherwise suspected of having been part of a chain of command responsible for torture, have been blocked from receiving asylum. The Dutch government has, over the years, tried to deport several of these Afghans. There have been several cases of trials for alleged war crimes. 162 The Netherlands has a specific policy in place for»westernized girls« who come from countries like Afghanistan: girls over ten years of age who have not been given a protection status but who have spent at least the last eight years in the country can be considered to be so westernized they would face problems if they returned. These girls can be allowed to stay, together with their families(this is, however, not a given rule; decisions are taken on a case-by-case basis). This policy came into being in 2011, after an upheaval over the intended deportation of a teenage Afghan girl. The Dutch minister 160. http://statline.cbs.nl/Statweb/publication/?DM=SLNL&PA=83102NED &D1=a&D2=0&D3=0&D4=0-1&D5=16,33,50,54,58,61-64&HDR=T,G4& STB=G1,G2,G3&VW=T. 161. http://pilpnjcm.nl/files/afghans-the-1f-procedure/. 162. See AAN analysis here: https://www.afghanistan-analysts.org/a-36year-wait-for-justice-dutch-arrest-suspected-afghan-war-criminal/. 20

THOMAS RUTTIG  |  AFGHAN EXODUS responsible for asylum policies estimated in April 2011 that, at the time, there were around 400 girls who might match these criteria. 2.3.3 United Kingdom The United Kingdom remained relatively untouched by the massive 2015/16 influx of migrants due to its geographical position and earlier efforts to deter migrants (made after peak numbers of asylum applications, circa 84,000, in 2002 163 ). However, for many refugees, the UK was their destination of choice. Thousands of them gathered at the mainland entrance to the Eurotunnel, near the French city of Calais, seeking to illegally board lorries and trains. 164 This included many Afghans, of whom a proportionally large number were minors. In October 2016, the UK took in 750 children, including many Afghans, from an unofficial camp near Calais, known as the»Jungle«, when it was closed by French police amidst violent protests. 165 This was highly unusual. The UK normally only accepts claims for asylum from people who have reached Britain. Between January and September 2016, the UK had registered the highest number of asylum applications for all nationalities in the first three-quarters of the year since 2004, with a total of 33,960. This is a reflection of the Europe-wide developments since 2015. Although relatively few migrants reached Britain compared with other countries, there was still a noticeable boost in UK numbers. In the fourth quarter of 2016, this trend ceased, though, with numbers lower by more than 25 percent compared to the second quarter of the year (from 10,231 to 7,146). Those seeking asylum in the UK are encouraged to make a claim as soon as they arrive. Decisions usually should take a matter of weeks. 166 While waiting for an asylum decision, there is no automatic state support. Those whose bids are successful are given»refugee status« or, if the application is on human rights grounds,»humani163. http://researchbriefings.parliament.uk/ResearchBriefing/Summary/ SN01403. 164. See here for a media report where this resulted in an Afghan fatality: http://www.thelocal.fr/20160528/afghan-migrant-killed-by-lorry-in-calais. 165. https://www.buzzfeed.com/rosebuchanan/calais-child-refugees-mountlegal-challenge-after-being-refu?utm_term=.ch1aJl6va#.bxZwNkr2w. 166. http://www.asylumaid.org.uk/the-asylum-process-made-simple/. tarian protection«: they have the right to work and claim state benefits as well as to seek family reunion(not available for applicants under the age of 18). After five years, if it is still considered unsafe for applicants to return to their country of origin, they can apply for»Indefinite Leave to Remain« in the UK. Those whose claims are rejected can appeal in a hearing before an immigration judge. If that is rejected, they can usually only make a second appeal if they can present fresh evidence. If a claim is rejected, people are expected to make arrangements to leave the country, or they may be forcibly deported. The number of Afghans among asylum seekers in 2016 was low compared to other countries, with 2,567 applications, around 7.5 percent of all applications, but this still made them the fourth largest group(they had ranked only sixth a year earlier, in September 2015). Among unaccompanied, asylum-seeking children, however, as in Sweden, Afghans represented the largest national group, with 783 cases registered by September 2016(circa 25 %), out of a total of 3,144. Many, perhaps most, of those came from the»Jungle« in Calais. 2.3.4 France France is also an outlier from the 2016 trend, as the number of asylum applicants did not drop as in most other European countries. Throughout the year, it consistently had the third-highest number of overall asylum applications per month, between 6,120 and 7,655. As a result of Europe-wide events, Afghanistan was back in the top ten of France’s main countries of origin in 2015 (ranked number ten) with 2,122 registered Afghan»requests for international protection« in total; the protection rate was high, with 80.3 percent. In 2014, Afghanistan was still at rank 31, with only 472 Afghans claiming protection. 167 In 2015, Afghanistan also was the most important country of origin for asylum seeking minors (14.6 %) in France. In its 2015 annual report, the French asylum authority OFPRA accredited the increase in asylum applications from Afghans to the influx into the Calais»Jungle« and 167. https://ofpra.gouv.fr/sites/default/files/atoms/files/rapport_dactivite_ ofpra_2015_hd.pdf(pp. 6, 37, 42, 54). 21

THOMAS RUTTIG  |  AFGHAN EXODUS Paris. When these camps were shut down in 2016, those inmates not allowed in by the UK were forced to apply for asylum in France. Also, as AAN heard in Italy, many Afghans prefer France over Italy as a destination, also due to relatively high recognition rates for Afghans. This contributed to the increase of Afghans applying for asylum. The 4,455 Afghans applying by the end of the third quarter in 2016 already surpassed the 2015 total. According to IOM figures, 118 Afghan asylum seekers returned voluntarily to their country from France in 2016 (2015: 9). There have not been any forced returns from France since 2009. 2.4 European Developments Put Into Context – A Brief Outlook With numbers of incoming migrants having dropped significantly, many European countries have begun to make a dent in processing the large backlog of asylum requests(1.2 million in total), while requiring those still not registered to do so. It can be expected, therefore, that the overall number of rejected asylum seekers will continue to grow. All applicants have the right to appeal, which, if exercised, would extend the duration of their stay considerably. Should many applicants repeal, the number of Afghans with a last instance rejection will grow; as a result, the number of returns – voluntary or not – is likely to grow throughout 2017. For those stuck between closed borders in southern and south-eastern Europe, it has become almost impossible to reach their favored destinations north of the Alps, mainly Germany, northern countries, or the UK. If the EU remains unable to agree on a distribution quota for all countries, and if the Dublin regulation continues to be increasingly applied, the danger of so-called»chain deportation« will arise once again. If some countries chose to deport asylum seekers across outer EU borders(as Hungary does to case of Serbia), they might once again end up in the country they had tried to flee from. The German Institute for Human Rights had already warned this might happen in a position paper published after the EU-Turkey deal was concluded. 168 This would impact not only Syrians(pushed back by Turkey to Syria), but also Afghans(working translation by AAN): 168. http://www.institut-fuer-menschenrechte.de/fileadmin/user_upload/ Publikationen/Stellungnahmen/DIMR_Stellungnahme_Menschenrechtliche_ Bewertung_EU-Tuerkei-Vereinbarung_in_ihrer_Umsetzung_20_06_2016.pdf. »For non-Syrian asylum seekers, who, for example, had fled from Afghanistan or Iraq, there also is the danger that they might be deported from Turkey back to their countries of origin, in breach of the Ge­ neva Refugee Convention and the European Human Rights Convention.« Italy and Greece, both in economic crisis, will continue to carry the largest share of the burden of accommodating asylum seekers. This might further strain their social systems and possibly result in a negative change of attitude among larger parts of the population vis-à-vis the migrants, with relatively small but vocal xenophobic movements already active. 3. Case Study Germany Germany led in Europe in almost all categories of incoming refugees and asylum applications in 2015 and 2016, both in absolute and relative figures. Roughly six out of ten migrants who came to Europe ended up in Germany. Afghans were strongly represented in all categories. This prompted the German government to change its 2015 asylum policy, which was widely seen as generous, to become more rigid. It even applied specific measures to make the country less attractive for Afghan refugees, with the specific aim of decreasing their number. It also took the lead in pushing the Afghan government to readmit rejected asylum seekers. This made Afghans – in contrast to Syrians, Iraqis, and Eritreans –»second class asylum seekers«. 3.1 The Figures The number of Afghans who came to Germany to seek asylum in 2015 and 2016 officially stands at 202,668 (154,046 or 14.0 percent of all arrivals for 2015 and 48,622 or 15.1 percent for 2016). These figures, however, are based on data from a special computing system called EASY, introduced during the 2015 peak influx (and known to be prone to shortcomings, such as double registrations), and had to be corrected in September 2016. Then the German government revised the figure for all asylum seekers that arrived in 2015 from 1.1 million to 890,000, downwards by around 20 percent. 169 169. https://www.welt.de/politik/deutschland/article158465433/Deutsch land-korrigiert-Fluechtlingszahl-fuer-2015.html. 22

THOMAS RUTTIG  |  AFGHAN EXODUS Replying to an AAN query, the spokeswoman of the Ministry of the Interior Ministry explained by e-mail that on January 16 th , 2017,»It is impossible, unfortunately, to specify the figure of the 890,000 according to individual nationalities.« This means there is no reliable data on how many Afghans truly came to Germany in 2015. What can be assumed is that it is likely lower than the 202,668 calculated above; if the 20 percent error margin was used, this would put the number closer to 160,000. 170 The government’s 2015 Migration Report states 94,902 Afghans »settled« in Germany that year, a category that includes asylum seekers. The net increase was even lower, at 89,931, indicating a registered outmigration of almost 5,000 Afghans in the same year. For the other groups that fall into that category – labor migrants, family reunions, students, and others, including so-called local employees who had worked for»German authorities« in Afghanistan(this includes the army, intelligence, and civilian authorities) 171 – only 1,423 visas were granted by the German embassy in Kabul in 2016(by mid-November). From this, it can be concluded that an overwhelming majority of the almost 95,000 Afghans that »settled« in Germany in 2015 were asylum seekers. Nevertheless, this number differs from the figures published by the EASY system and BAMF. The number of Afghans who formally requested asylum probably reflects the extent of the rise of the Afghan diaspora in 2015/16 more correctly: the total for both years is 158,394. It is likely, though, that this includes some who had come into the country before the 2015 migration peak and used this cover to legalize their stay. Still others might not have applied yet. While the number of Afghans applying for asylum was relatively low in the years immediately before 2015(2013: 8,240 cases; 2014: 9,673 cases), a large Afghan community of 126,334 persons(figure from 2009) already existed in the country. The community included many who had fled from war and a series of repressive regimes in the 1980s and 170. As already stated in part one of this dispatch series, the German government nevertheless continues to use the uncorrected figure in the annual asylum report for 2016, published in January 2017(see here). The most recent migration report, published in December 2016, contains the correction. 171. According to the report, 771 visas for Afghan local employees were granted in 2015. This brings the sum to almost 2,000 persons, including family members. 1990s. 172 Around 40 percent of them(49,081) have already received German passports, showing the degree of integration into German society. Remarkably, these Afghan immigrants did not have to face the high degree of anti-immigrant hatred exhibited in 2015/2016. According to recent official figures 173 , Afghan citizens living in Germany numbered 246,954 by September 30 th , 2016 – this includes everyone with a permanent residence permit(6.6 %) or any type of temporary residence permit (71.0 %). 174 More than one third(91,359) were under 18 years of age in 2016. From these figures, it can be concluded that between 2009 and 2016 the net growth of the Afghan community in Germany(including all asylum seekers) was 120,000 people. Throughout the year 2015, around one fifth of the Afghans arriving(31,382) were able to formally apply for asylum. This led to a large backlog of Afghans(and other nationalities) who were registered in the EASY system but not yet officially recognized as asylum seekers. The German authorities started processing their number in 2016. Therefore, while 48,622 more Afghan asylum seekers entered Germany in 2016, the number of Afghans applying for asylum rose to 127,012 in the same year(17.6 percent of all countries of origin), including newcomers from 2015 and 2016 and older cases. Already some 364,664 asylum cases were pending from all countries of origin by 31 December 2015; the accessible Eurostat data does not sort pending cases by country of origin. The number of Afghan first asylum applicants stood at 158,394 at the end of 2016 – the second largest group among all countries of origin after the Syrians. Among those Afghan applications, 4,744 were unaccompanied minors in 2015 and 7,509 in the first half year of 2016. This was the highest number among all countries of origin. 175 172. http://www.bpb.de/internationales/asien/afghanistan-das-zweitegesicht/153504/afghanische-diaspora-und-brain-drain?p=all. 173. http://dip21.bundestag.de/dip21/btd/18/103/1810336.pdf. 174. The category for the remaining 22.5 percent(»others«) is not explained but it can be assumed that these are asylum seekers with pending cases. Afghans living in Germany illegally are obviously not covered here; AAN also has not come across data or even estimates. 175. There is no country-specific data about this category in the reports for the entire year of 2016, see also here: https://www.bamf.de/SharedDocs/Anlagen/DE/Downloads/Infothek/Asyl/um-zahlen-entwicklung.html. 23

THOMAS RUTTIG  |  AFGHAN EXODUS Of the 2015 and 2016 Afghan asylum applications, altogether 74,212 were decided upon in the first instance in both years. In 2015, 5,966 Afghan asylum cases were decided in the first instance, 176 while this number rose more than tenfold to 68,246 in 2016. Despite this rise, more than half of all applications were still pending. In 2015, 2,842 of those cases ended positively: 48 Afghans received full political asylum(according to the Geneva Refugee Convention), 1,660 refugee status, 325 subsidiary protections, and 809 were granted Abschie­ beverbot(temporary leave to stay; a literal translation from German sounds even stronger:»ban from deportation«). Although Abschiebeverbot legally is not a protection title and therefore can be revoked on short notice, it is officially counted under the»protection rate«(Schutz­ quote). This added up to a protection rate for Afghans of 47.6 percent in 2015.(This rate had been 47.9 percent in 2013 and 46.7 percent in 2014.) In 2016, absolute figures as well as the protection rate rose significantly. Of the 68,246 Afghan cases that came to a decision, 38,090 ended positively: 80 with full political asylum, 13,733 with refugee status, 5,836 with subsidiary protection, and 18,441 with Abschiebeverbot. This lead to a protection rate for Afghans of 55.8 percent for the year 2016. 177 Even this increased rate is still well under the average rate for all countries of origin: 62.4 percent. Altogether, the cases of 25,636 Afghan asylum applicants were refused in 2015 and 2016 while those of 7,644 others were decided»in other ways.«(This legal category mainly refers to»Dublin cases.«) 178 Human rights NGOs in Germany argue that those cases should not be counted in the»pro176. Surprisingly enough, the federal government gave completely different figures for 2015 in an answer to a parliamentarian query in November 2016; it reported 31,902 decided cases, i. e. five times more than in the BAMF annual report. Moreover, the data for 2015 in the related table on page 21 does not add up at all. 177. The figures for under-age Afghans in the first half of 2016 were as follows. Decisions were made about only 331 cases: 98 cases were granted refugee status; 25 were granted subsidiary protection; and in 112 cases Abschiebungsverbot was granted. This results in a protection ratio of 71 percent. 38 applications were rejected and 58 closed, based on the Dublin provision or other regulations. Processing those cases took 10.6 months on average – while the average for all under-age cases stood at 7.4 months. 178. Those NGOs particularly challenge the inclusion of the so-called Dublin cases, as this pushes the protection rate down. These cases do not involve a»substantial« decision(i. e. whether the applicant is entitled to any form of protection or not) and, when included, strengthen the percentage of cases that did not end with a form of protection. They suggest that an»adjusted protection rate« is used instead that excludes the Dublin cases altogether. This would have brought the protection rate for Afghan over the 50 percent threshold in 2015 and put them in the category of good»chances to stay« instead of having a bad or slim chance, with all repercussions for integration. tection rate,« as they are not decided upon on substantial reasons but finalized procedurally. Doing so, this rate would raise for Afghans from the official 55.2 to 61.5 percent. We shall see below why this is relevant. At the same time, as a result of the reintroduced border controls, the number of migrants who were refused entry into the country at its borders rose in 2016. The Federal Police reported 21,200 cases, 179 but no countries of origin were given. An earlier report covering the first half of 2016, with 13,324 such cases, said that every fourth person was an Afghan. 180 This included, according to another media report, 458 under-age persons,»most of them« from Afghanistan. 181 3.2 Slowing Down Afghan Cases At the end of 2016, 417,076 asylum applications from all countries of origin were pending in Germany, a rise compared to a year earlier. Many of these cases are likely remaining from before the peak influx in the second half of 2015; there were already 150,257 pending cases – from all countries of origin – at the end of 2014. Almost one quarter of all first-instance pending cases (101,382) were Afghans, the largest number among all countries of origin – and clearly(by 40 %) above their percentage among all asylum seekers. Afghans also had a relatively high number of pending follow-up cases(1,474 in the second and third instances) at that time. This is the result of a government decision to prioritize cases that were easy to decide(from countries that either were declared safe – with a likely rejection – or those with a high acceptance quota such as Syria and Eritrea). Afghanistan fell between, leading to a situation where relatively few Afghans’ cases were decided and where the duration of an Afghan case was almost three times that of the overall average in March 2016: 15 versus 5.2 months. 182 That led to a series of 560 legal actions for failure to act submitted by Afghan asylum applicants in the first quarter of 2016. 179. http://www.bild.de/politik/inland/fluechtlingskrise/asylbewerber-zahl en-2016-49694132.bild.html#fromWall. 180. http://www.tagesschau.de/inland/fluechtlinge-deutschland-abschie bungen-103.html. 181. http://www.tagesspiegel.de/politik/fluechtlingsabkommen-mit-dertuerkei-gebremste-migration/13990518.html. 182. http://www.n-tv.de/politik/Asyl-Entscheidung-dauert-oft-Jahre-article 17723576.html. 24

THOMAS RUTTIG  |  AFGHAN EXODUS Afghans were also the largest national group in 2016 among asylum seekers who ended up with a form of »lower quality« protection, i. e. subsidiary protection (5,836 cases) and Abschiebungsverbot(18,441), indicating that they found it more difficult to receive full political asylum or refugee status according to the Geneva Convention. Full political asylum was awarded in only 80 cases(around 0.1 percent) and refugee status in 13,733 (around 15 percent) of the 68,562 Afghan cases that were decided in 2016. 183 3.3 The German-Afghan Migration»Agreement« To decrease numbers of asylum seekers in general, the German government not only took steps to keep refugees away from its own borders – for example, by initiating the EU-Turkey deal – but it also adopted various measures to make it less attractive to come to Germany. Legislation governing asylum, residence, and integration has been tightened twice since October 2015. Several states reintroduced compulsory residence for asylum seekers. As a result, they are restricted to looking for accommodation in a limited area on an already tight market and they lose social welfare if they violate the new regulations. This provision also prevents them from moving in with relatives outside their»area of residence«. A third legislation package that included plans to further reduce in-cash support for individual asylum seekers was rejected by the upper house of parliament on December 16 th , 2016. 184 In Afghanistan itself, the government(like the Australian and the Austrian ones) launched campaigns to inform about perceived rumors associated with the flight to Germany and to discourage more Afghans from leaving the country. 185 183. At the same time, the number of Afghan asylum applications started dropping significantly in the second half of the year. From the peak in August 2016(19,840 applications) to 14,434(September), 5,351(October), 2,937(November) and finally 1,822(December); the number went down to less than ten percent within half a year. It is not clear whether this was due to a new procedural approach or because the number of Afghans without an asylum application was exhausted. Refugee legal activists, however, told AAN that German authorities had prioritized the cases of families, which also explains the increased protection rate and that they now were preparing to process the(larger number of) cases of single men in what is expected to result in a much lower protection rate again. 184. http://www.mdr.de/nachrichten/politik/inland/bundesrat-asylbewer berleistungsgesetz-gestoppt-100.html. 185. http://www.aljazeera.com/news/2015/11/germany-campaign-refugees-leaving-afghanistan-151124131156428.html. With its bilateral»readmission« agreement with the Afghan government – signed on the same day as the EU-Afghan»Joint Way Forward« framework document, October 3 rd , 2016 – the government in Berlin opened the way to more»returns« of rejected Afghan asylum seekers. The agreement is called»Joint Declaration of Intent on Cooperation in the Field of Migration,« but its text has not been published(AAN has seen a copy in English). AAN also has learned in Kabul that the Afghan government had asked that the document not be called an»agreement,« as in that case it would have to be submitted to the Afghan parliament where approval was far from sure – see the events around the Swedish agreement mentioned in Chapter one»Europe – The Changing Situation«. 186 In their agreement, both governments stress that they see»voluntary return« as the»preferred way of fulfilling the obligation[of rejected asylum seekers] to leave the country« and commit to protect»asylum seekers and refugees rights« according to international law. Germany further guarantees that it»will continue to grant protection to those(…) who are entitled to it under German law.« At the same time, both sides agree»that effective enforcement measures need to be taken in a timely fashion, if the voluntary return does not take place within the given time limit.« The agreement also states that voluntary returnees»will receive all available benefits from current programs« and»will have the right to apply for any future return program.« The agreement indeed sounds as if the authorities want to push Afghan asylum seekers to opt for voluntary return(and the not too generous 187 attached financial incentives) even before their cases are definitively decided, including the appeal option that is open to each applicant when rejected in the first instance. The message behind it is clear: leave and take the assistance offered before you are rejected and returned without 186. https://www.afghanistan-analysts.org/eu-and-afghanistan-get-dealon-migrants-disagreements-pressure-and-last-minute-politics/. 187. Under the IOM-managed REAG / GARP program(see here), transportation costs by plane can be covered. Or, travel assistance of 200 Euro per adult / youth and 100 Euro per child under twelve years of age can be paid. An IOM official spoke of providing about 700 Euros in cash at the airport before departing Germany in conversation with an AAN representative. Additionally, BAMF compiles a list of persons eligible for additional financial support in Afghanistan through the IOM office in the range of an equivalent of 800–2,500 Euros, s/he told AAN. According to REAP/GARP, voluntary returnees can also apply for start-up cash of 500 Euros per adult/youth and 250 Euros per child under the age of 12, with a maximum amount of 1,500 Euros to families, for so-called»Dublin cases,« i. e. those are»required to leave« not to their country of origin but to the EU country where they had been first registered upon entry. 25

THOMAS RUTTIG  |  AFGHAN EXODUS any assistance. This approach is heavily criticized by nongovernmental organizations in Germany. In 2016, 3,159 Afghans in Germany took this option. 3.4 Afghanistan: Safe for Deportations? Starting in late 2015, Afghans were singled out for a particular treatment. This has to do with widespread fatigue from Afghanistan among Germany’s political decision makers, parts of the media, and the public. Federal Interior Minister Thomas de Maizière, who is from chancellor Angela Merkel’s party and has the lead in her cabinet on migration-related issues, put it into words when he stated on October 28 th , 2015(my transcript and translation from a video 188 ): »German soldiers and police contribute to make Af­ ghanistan secure. Much development aid went to Afghanistan. So, one can expect that the Afghans stay in their country.« In the same press conference, he called it»unaccept­able« that Afghans were the second largest group among incoming migrants(at that point this was true for all of Europe, but not for Germany, where they were still only the fourth largest group). At a meeting of EU interior ministers on November 9 th , 2015, he announced,»We want that the signal gets to Afghanistan: ›Stay there! We will send you out of Europe(…) directly back to Afghanistan!‹« Later in the same month he stated:»When you deport, more people also depart voluntarily.« 189 This approach is based on a combination of two assessments by the government of the situation in Afghanistan that are, however, challenged by many in Germany. This includes members of Merkel’s own government and MPs who belong to the smaller social-democrat coalition partner and governments of some of Germany’s federated states(the Länder), not to mention NGOs and the opposition. The government’s first claim is that Afghans have a »slim chance to stay«(geringe Bleibechance) in Germa188. http://www.bmi.bund.de/SharedDocs/Kurzmeldungen/DE/2015/10/ pressestatement-zur-migrationslage.html?nn=3314802. 189. http://www.tagesspiegel.de/politik/innenminister-in-saarbrueckende-maiziere-fordert-nationale-kraftanstrengung-bei-abschiebungen/149 14568.html. ny, based on its decision that such a definition applies for any nationality of asylum seekers with a»protection rate« of under 50 percent in the previous period of six months(formerly one year). Afghans, between 2013 and 2015 had a rate of(just) below 50 percent(but above it in 2016). This method is sharply criticized by non-governmental organizations working with asylum seekers. The biggest of them, Pro Asyl, says that»sociological« grouping of immigrants prejudices the outcomes of their cases and, as a consequence, undermines the constitutional principle that asylum cases need to be decided on an individual basis. The government vehemently denies this accusation but plenty of cases prove it to be true (see remarks about the»Memorandum for a Fair and Accurate Asylum Procedure« below). Secondly, the German government projects Afghanistan, at least in part, as a country safe enough to receive rejected asylum seekers. This is doubted, significantly, by a number of governments of the German federated states (the Länder) and challenged in the»Memorandum for a Fair and Accurate Asylum Procedure in Germany« 190 published by Pro Asyl and eleven other social, human rights, and legal organizations in late 2016. The authors state, among other things, that»changes in legislation and political directives influence decision making by the [BAMF], while the situation in the particular countries of origin did not significantly change.« In their view, it should be the situation in the country of origin and individual circumstances in each case on which decisions about asylum are taken, not domestic political considerations. Meanwhile, the December 2016 terrorist attack in Berlin, the October 2016 rape and murder of a young woman in Freiburg by an Afghan immigrant, and the July 2016 amok run by a young Afghan in a train increased pressure to deport the»criminals« from the asylum seekers. 191 Now, the threshold for deportation is rather low, as shown in the case of an Afghan who was scheduled for deportation after he was fined for throwing a beer 190. https://www.proasyl.de/wp-content/uploads/2015/12/Memorandum -f%C3%BCr-faire-und-sorgf%C3%A4ltige-Asylverfahren-in-Deutschland -2016.pdf. 191. Current law already provides that those asylum seekers are exempted from protec-tion from deportation whose presence is a threat either »to the security of the country« or»to the general public.« Now, some politicians(not only those from the neo-populist right wing) suggest allowing deportation for minor offenses, including repeatedly riding public transport without a ticket or breaching limitations of residence. 26

THOMAS RUTTIG  |  AFGHAN EXODUS mug during a brawl at a local festival(without injuring anyone) 192 ; it is not clear whether he has already been returned. Among the 60 Afghans on the first two charter flights, in December 2016 and January 2017, of involuntary returnees were a number of criminals, according to the German government, but it gave no further details. As German media reported, the Afghan authorities were not made aware of this fact. 193 3.5 Afghans As»Second-class« Asylum Seekers This policy has contributed to a political climate in which Afghans are increasingly seen as economic migrants, not refugees fleeing from war, and thus not fully entitled to protection. Susan Fratzke of the Migration Policy Institute Europe, headquartered in Brussels, said,»There’s definitely a distinction being made, at least in the public mind,[about] nationalities who are considered to be ›legitimate,‹ as refugees.« 194 This is reflected by the treatment of Afghans by local administrations that, under the German federal system, deal with the accommodation of asylum seekers and their inclusion in measures furthering their integration, such as language courses which, in turn, are key prerequisites for gaining access to the job market if granted protection. 195 There are plenty of reports in the German media, by NGOs, and the still-large number of volunteers, that Afghans are increasingly excluded from such measures. The northern port city of Stralsund, for example, has moved Afghan refugees»for organizational reasons« out of apartments back into mass accommodation facilities. 196 In the state of Bavaria, the interior ministry issued a regulation in late 2016 that limits the hand-out of work permits for asylum seekers from countries with a»low chance to stay.« Volunteers from the town of Garmisch-Partenkirchen and from areas around Munich have even told the press that some local authorities have 192. http://www.pnp.de/mobile/afghane_wird_abgeschoben_ich_will_ nicht_weg_2341162/. 193. http://www.taz.de/Abschiebungen-nach-Afghanistan/!5378336/. 194. https://www.buzzfeed.com/jinamoore/this-is-how-europes-rules-dis criminate-against-afghan-refuge?utm_term=.jclmo7ELq9#.mdK650rlJx. 195. A Swiss study has found(based on quantitative research of cases between 1994 and 2004 in this country, quoted here) that a long duration between the asylum application and the decision about it»significantly reduces« the likelihood of finally-accepted asylum seekers to find a job. 196. http://www.ostsee-zeitung.de/Extra/Meinung/Leserbriefe/Politik-Leser briefe/Familien-sind-gut-integriert. actively started withdrawing already granted work permits»particularly from Afghans and Pakistanis.« 197 Inequality in the treatment of Afghans(and others) also extends to the BAMF, the governmental authority under the federal ministry of the interior that evaluates and takes decisions about asylum applications. The memorandum already quoted above lists a number of such examples: n asylum seekers being inadequately informed about their rights and duties; n asylum seekers having insufficient access to legal advice and representation; n speeded-up procedures blocking effective access to due process and negatively impacting accuracy and fairness in the process; n BAMF officials not fulfilling their fiduciary duty vis-àvis applicants; n hearings being held in an»interrogation-like« atmosphere; n personal circumstances of applicants being insufficiently considered; n officials»not approach[ing] the hearing objectively and without bias«; n translations being inexact, harming the applicant’s chances of success; and n the»use of text modules« in hearing verdicts, showing that decision-takers did not sufficiently consider individual cases. There were cases, the authors further wrote, with»improper deliberations in rulings that led to the suspicion that responsible BAMF officials were aiming from the beginning to reject the asylum application.« 198 197. http://www.sueddeutsche.de/bayern/asylbewerber-fuer-fluechtlingewird-es-schwieriger-eine-arbeitserlaubnis-zu-bekommen-1.3317005. 198. Many of the shortcomings pointed out in the NGOs’ memorandum were already mentioned in an earlier version in 2005 and, as the authors stated, have therefore to be considered»structural deficiencies in the German asylum procedure.« They add that»over-long« procedures in individual asylum cases did»not only exist since the rise in the number of asylum applications« in 2015. 27

THOMAS RUTTIG  |  AFGHAN EXODUS Furthermore, Pro Asyl calls the German authorities’ behaviour an»unofficial but systematic strategy of discouragement.« It also accuses them of deliberate»duping and wrongly advising« asylum applicants in order to make them take the»voluntary« return option. 199 This seems to work: Now that almost 3,200 Afghans have »voluntarily« returned in 2016(tenfold the 2015 figure 200 ), one fourth of those 12,539 Afghans»required to leave« 201 – the official term for finally rejected asylum seekers – could already be outside the country.(The categories, however, do not fully overlap; various media reports speak of individual Afghans who gave up before they had fully exhausted the legal asylum procedure and particularly the option to appeal.) NGOs such as Pro Asyl warn asylum seekers not to prematurely throw away the chance to be granted protection. Even media leaning to the conservative side, such as Fo­ cus magazine, called the result of the German asylum policy a»two-class society among refugees«. 202 3.6 A Climate of Fear and Doubts All this has created an atmosphere of fear among Afghan asylum seekers. Volunteers in Hamburg, for example, told AAN that young Afghan men, particularly, were leaving jobs and vocational training»and going underground, saying ›this does not make sense anymore if we are deported anyway.‹« There are recurrent reports about suicide attempts, linked with the December 2016 deportations but also during previous individual, then abolished deportation attempts. 203 The government’s approach needs to be put into the context of the domestic policy debate before the upcoming general elections in September 2017 and a situa199. http://www.faz.net/aktuell/politik/fluechtlingskrise/fluechtlingemehr-afghanen-kehren-zurueck-14372049.html; see also: http://www. deutschlandradiokultur.de/rueckfuehrungen-in-hessen-schicksale-hinterunfreiwilliger.976.de.html?dram:article_id=355654. 200. http://de.reuters.com/article/deutschland-fl-chtlinge-idDEKBN14H0JU. 201. https://thruttig.files.wordpress.com/2016/11/20161114bureg-antwauf-anfrage-afg-abschiebungen.pdf. 202. http://www.focus.de/politik/ausland/gute-gegen-schlechte-bleibechancen-zwei-klassen-gesellschaft-warum-deutschlands-umgang-mitfluechtlingen-gefaehrlich-ist_id_5370373.html. 203. See media reports here: http://www.sueddeutsche.de/bayern/asylpolitik-er-sagte-er-fuerchte-in-afghanistan-um-sein-leben-1.3296031; and also here: https://www.welt.de/regionales/bayern/article152872371/ Mit-Omar-Arasch-und-Oezguer-in-die-Kreisklasse.html. tion in which, for the first time in decades, an anti-immigrant party looks prone to enter the federal parliament, the Bundestag. Particularly because the German-Afghan »readmission« agreement will unfold simultaneously with the election campaign. Its text stipulates that in an »initial phase of six months(…), it is necessary to limit the number of returnees per flight for involuntary return operations(…) to 50.« This covers the period from the date of signature, 2 October 2016, to 2 April 2017. Although no frequency for the flights is explicitly stipulated for the initial phase, the agreement allows the number of flights and returnees to increase from April 2017 onwards. Immediately after the agreement with Afghanistan was signed, Minister de Maizière sent a letter to the interior ministers of the Länder(the states in Germany’s federal system) 204 demanding its implementation»without delay(…), now that we have a considerably better base to work on.« But it took till December for the first flight to depart, as some Länder refused to cooperate. One of the main reasons is that they are sceptical about the federal government’s assessment of the security situation in Afghanistan and, as a consequence, of the feasibility of enforced returns. The first return flights, however, were preceded by the revocation of an informal Abschiebestopp(leave to remain) that had been in force for rejected Afghan asylum seekers for over ten years, agreed upon by the Interior Minister Conference of the Länder. 205 In late 2015, under pressure from the federal interior ministry – and its description of parts of Afghanistan as »sufficiently secure« – individual states declared the Ab­ schiebestopp over, starting with social democrat-governed Hamburg. 206 (Hamburg, a port city, traditionally has the largest Afghan community in Germany.) Other states, however, publicly stated that they would stick to their decision to not support»involuntary returns« of Afghans. These reportedly include Berlin, Rhineland-Palati­ num, Lower Saxony, Bremen and Schleswig-Holstein. 207 204. https://thruttig.files.wordpress.com/2016/10/20161006brief-bmi-anlc3a4nder-abschiebgn-nach-afg.pdf. 205. http://www.fluechtlingsinfo-berlin.de/fr/pdf/Bleiberecht_IMK_2006. pdf. 206. http://www.zeit.de/2016/20/fluechtlinge-aus-afghanistan-abschie bung-hamburg. 207. http://www.spiegel.de/politik/deutschland/fluechtlinge-mehrerebundeslaender-stellen-abschiebungen-nach-afghanistan-zurueck-a-11332 81.html. 28

THOMAS RUTTIG  |  AFGHAN EXODUS Anne Spiegel, Rhineland-Palatinum’s minister responsible for integration, said in November 2016,»I continue to look at deportations to Afghanistan with extreme scepticism because of the security situation there« – indicating a lack of trust in the federal government’s assessment. 208 In November already, the German Länder Interior Minister Conference had urged the federal government to update its Afghanistan assessment»in cooperation with UNHCR and IOM.« The resulting UNHCR report 209 contradicted the government’s assessment in»diplomatic but nevertheless unambiguous words,« as one German newspaper wrote. 210 IOM’s official answer is not known yet, but its director general, in an interview with a German newspaper in December 2016, supported the government’s stance. 211 Following this ambiguous outcome, Schleswig-Holstein’s interior minister urged his Länder colleagues to temporarily renew the Ab­ schiebestopp until the assessment of the Afghan situation has been clarified. 3.7 Will Germany Treat Afghans Fairly Again? Now it remains to be seen whether the latest development, particularly the rise of the Afghans’ 2016 protection quota to over 50 percent, will have political repercussions and restore fair access to integration and language courses, work permits, and access to jobs and vocational training for Afghan asylum seekers. An analyst from a large NGO working in the field who asked AAN not to be named quoted contacts in BAMF as saying that the agency had so far processed Afghan families’ applications who have a higher chance of recognition but that currently, since December 2016, hearings for single Afghan men are being held. The went on to say that the BAMF therefore expects the»protection« rate to drop steeply again. Therefore, he was told,»it was not worthwhile« to change the»slim chance to stay« assessment for Afghans. The federal interior ministry also 208. http://www.swr.de/swraktuell/rp/integrationsministerin-zu-bundesbe schluss-afghanen-abschieben/-/id=1682/did=18503220/nid=1682/1lzr340/. 209. Only available in German: https://thruttig.files.wordpress.com/2017/ 01/20170111-unhcr-zu-silage-afg-reaktion-de-maziere.pdf. 210. http://www.zeit.de/politik/deutschland/2017-01/afghanistan-asylabschiebung-unhcr-abschiebestopp. 211. https://www.welt.de/politik/deutschland/article160281515/Afghan istan-in-einigen-Regionen-ausreichend-sicher.html. told the Dari program of Deutsche Welle radio in early December 2016 that the 2016 increase in the»protection rate« would not result in a reopening of the doors of integration courses for Afghans. 212 Publicly, in answer to an oral question in parliament on January 17 th , 2017, the government’s spokesman was more diplomatic and said that this question is currently»being considered.« 213 4. Conclusions The dropping number of asylum seekers arriving in Europe after the peak in the second half of 2015 reflects that the combination of closed and reinforced borders, tightened laws, lowering standards of humanitarian and integration measures, their treatment as»second-class asylum seekers« in some countries, a general atmosphere of discouragement for incoming Afghan migrants, and the system of readmission agreements has worked, from governments’ point of view. The EU-Turkey deal has particularly affected Afghans, while it did not have a strong impact on Syrian migration. This is demonstrated by the over proportional drop in the number of incoming Afghans in 2016, particularly from March onwards. But these measures have not stopped Afghans fleeing their country in general; the difference is that Afghans can no longer reach Europe easily. Many might have been discouraged from starting the long, dangerous journey with its uncertain outcome while others are stuck at the closed borders with no safe way forward and no willingness to go back. For those migrants in particular, conditions have deteriorated considerably. AAN colleagues Martine van Bijlert and Jelena Bjelica have described this in detail in their dispatch from Serbia. 214 The general drop in incoming asylum seekers in 2016 in Germany has had a side effect. According to Günter Burkhardt, the chief executive of Pro Asyl, the largest German NGO working on migration,»in Germany now accommodation facilities are standing empty, while in Greece refugees live in the streets and often even do not get the chance to register their asylum request.« 215 212. http://www.dw.com/fa-af/ برای- پناهجویان-افغان-هنوزهم-زمینه-شرکت-بهکورس-های-ادغام-داده-نمی-شود /a-36607514. 213. http://www.ulla-jelpke.de/wp-content/uploads/2017/01/18214_AFGIntegration_rechtmAufenth_Jelpke.pdf. 214. https://www.afghanistan-analysts.org/afghan-exodus-notes-froma-belgrade-squat/. 215. http://www.taz.de/!5317159/. 29

THOMAS RUTTIG  |  AFGHAN EXODUS This also shows that Germany fails to show solidarity with other EU countries, despite its better general performance compared to most others. In Afghanistan’s case, the on-going exodus(not only) to Europe reflects an unchanged – and partially even worsening – general security situation in the country. The combination of a lack of security and fears of an uncertain future, after four decades of war, continue to motivate people to leave. 216 Under these circumstances, deportations to Afghanistan are highly problematic. This is indirectly reflected in the policies of European governments who have, in 2015 and 2016, carried out fewer deportations to Afghanistan annually than in the previous years. Nevertheless, the support for voluntary returnees, which covers a short transition period at best, and the even lower support for the forcibly deported, show that the multi- and bilateral readmission agreements that declare that integration programs should be set up have not yet been translated into visible action at this point. They also cannot substitute for still necessary comprehensive, long-term political and financial investments to remove the main trigger of the exodus, the on-going war, and to address the underlying socio-economic causes of it. Not least, this will require the Afghan state to drastically improve its own performance vis-à-vis its populace. tries of origin are published. This leaves out Afghans, for example, in data from the Netherlands, the UK, and Italy in some or all quarters of 2016. The following colleagues provided details, mainly about their home countries: Kaisa Pylkkanen(Finland); Fabrizio Foschini(Italy); and the Guardian’s Sune Engel Rasmussen(Denmark). AAN colleagues Martine van Bijlert (Netherlands); Kate Clark(UK); Jelena Bjelica(Serbia, Romania, Croatia, and Hungary); as well as Ann Wilkens from the AAN advisory board(Sweden) also provided crucial information. The research for this dispatch is funded by the Kabul office of the German foundation Friedrich-Ebert-Stiftung (FES) and is a part of a dispatch series for a joint publication with FES. For more information, the paper»We Knew They Had No Future in Kabul: Why and How Afghan Families Decide to Leave« on this subject was part of an earlier project with FES. 218 5. Figures and Sources Unless stated otherwise, all statistical data on the EU in this dispatch is from Eurostat, 217 in order to maintain compatibility. The term»asylum applicant« refers to first-time applicants. Applicants have the right to file a follow-up application if personal circumstances relevant to their claim have changed; this leads to a higher number of overall applications. No full set of data on Afghan migrants for all European countries is available in the Eurostat statistics. For individual member states, only the top three or five coun216. See AAN analysis from our project with the FES: https://www.afghan istan-analysts.org/publication/other-publications/we-knew-that-they-hadno-future-in-kabul-why-and-how-afghan-families-decide-to-leave/. 217. http://ec.europa.eu/eurostat/statistics-explained/index.php/Asylum_ quarterly_report; http://ec.europa.eu/eurostat/web/asylum-and-managedmigration/publications. 218. https://www.afghanistan-analysts.org/publication/other-publications/ we-knew-that-they-had-no-future-in-kabul-why-and-how-afghan-families-decide-to-leave/. 30

About the author Thomas Ruttig is the co-director and co-founder of the Afghanistan Analysts Network. He has a degree in Asian Studies (Afghan Studies) from Humboldt University, Berlin(Germany). His worked for twelfe years(1989–2000) as a foreign news editor and freelance journalist specializing in Afghan and Central Asian development affairs and has spent over ten years working in Afghanistan and Pakistan. In 1988/89 and 2000–2006 Thomas was based locally in the region as a diplomat at the GDR Embassy in Kabul(1988–1989) and later as a political affairs officer for two UN missions in Afghanistan(2000–2003). This included assignments as UNSMA head of office in Kabul, adviser to the Afghan Independent Emergency Loya Jirga Commission, and the UNAMA head of office in Islamabad and Gardez. He then worked as the Deputy to the EU Special Representative for Afghanistan(2003–2004) and as a Political Adviser to the German Embassy in Kabul (2004–2006). From 2006 to 2008, he joined the German think-tank Stiftung Wissenschaft and Politik(SWP) as a Visiting Fellow, and since 2008 has been an independent political analyst, author, and consultant. As a co-founder, he has worked with AAN since its founding. Thomas travels to Afghanistan several times per year. He blogs about Afghanistan in German as well, under »Afghanistan Zhaghdablai.« Imprint Friedrich-Ebert-Stiftung| Department for Asia and the Pacific Hiroshimastr. 28| 10785 Berlin| Germany Responsible: Marc Saxer, Head, Department for Asia and the Pacific Phone:+49-30-26935-7451| Fax:+49-30-26935-9250 https://www.fes.de/referat-asien-und-pazifik/ To order publications: Julia.Schultz@fes.de Commercial use of all media published by the Friedrich-EbertStiftung(FES) is not permitted without the written consent of the FES. The views expressed in this publication are not necessarily those of the Friedrich-Ebert-Stiftung. This publication is printed on paper from sustainable forestry. ISBN 978-3-96250-262-1