September 20, 2017

Still in Limbo: About a Million Asylum Seekers Await Word on Whether They Can Call Europe Home


The estimates presented in this report are based on data from Eurostat, Europe’s statistical agency. Eurostat is a central repository of the European Union’s economic, environmental and population data. Its goal is to provide high-quality statistics that enable comparisons across countries. Eurostat also provides nationality and country of application data for asylum applications. It contains data on the number of asylum applications submitted, the number of applications withdrawn, application decisions (first-time and final decisions) and data on those ordered to leave, also known as returns. These publicly available data are not linked together. Consequently, Pew Research Center arranged these data to produce estimates for the status as of the end of 2016 of Europe’s asylum seekers who applied in 2015 and 2016.

The method used to estimate the possible status of asylum seekers as of the end of 2016 is based on this formula:

TOTAL       =         Approved          +         Rejected          +         Waiting 

Public data on Europe’s asylum seekers are anonymized, meaning that analysts cannot track individuals across various stages of their asylum application process. This anonymization does not allow analysts to determine quickly the legal status of asylum seekers in a specific migration wave. Consequently, estimation methods must be used to calculate the number of people in each possible phase of the application process (approval, rejection and waiting for a decision as of the end of 2016).

The estimates in this report are based on calculations using both country of application and asylum seeker nationality information. For example, the total number of Syrian asylum applicants across Europe is based on calculations for the number of Syrian applicants in Germany plus the number of Syrian applicants in France plus the number of Syrian applicants in Sweden and so on. In the same way, the total number of asylum applicants in Germany is based on individual estimates by nationality, or, for example, the number of Afghan asylum applicants in Germany plus the number of Iraqi asylum applicants in Germany and so on.

All estimates were conservatively rounded to the nearest 5,000 to avoid overstating the level of precision associated with the estimates. Detailed estimates by nationality and country of application are provided only for asylum seeker groups of 20,000 or more in this analysis.

Complete data from Eurostat were used through the end of 2016. Consequently, the figures in this report on the status of asylum seekers applying for refugee status in 2015 and 2016 reflect applicants’ status as of Dec. 31, 2016.

Estimating the total number of asylum seekers in 2015 and 2016

Across the European Union’s 28 member countries plus Norway and Switzerland, some 2.5 million first-time asylum applications were filed in 2015 and 2016. With the sudden movement of hundreds of thousands of asylum seekers into Europe during these years, however, it is likely that this total for 2015 and 2016 includes applications filed by the same people in multiple countries. It is also possible that some asylum seekers returned to their home countries or were remaining in Europe in some other legal status before proceeding further in the asylum seeker process. Consequently, it is important to obtain an estimate of the unique (or “net”) number of applicants across Europe during this time period.

Applications are not the same as arrivals. It can take several months for newly arrived asylum seekers to make formal applications for asylum and get included in monthly statistics. Applications represent individuals, however, not households. Regardless of age or family relationship, each person claiming refugee status must file an asylum application. It is possible some people in refugee situations entering Europe never submit an asylum application. This report does not include those people.

EU countries as well as Norway and Switzerland lawfully agree to the Dublin Regulation, which states that asylum seekers are to apply for refugee status in the first European country they enter. Germany temporarily waived their participation in the regulation during the refugee surge of late summer and early fall of 2015. Consequently, many refugees traveled through Greece, making their way north and west to Germany and other destinations using a route through several Balkan countries as well as EU member countries such as Hungary and Austria. Some asylum seekers applied for refugee status in transit countries before making it to their desired destination of Germany or other countries. Many asylum seekers who applied in Hungary and Austria, for example, later had their applications withdrawn.

To avoid double counting, the total number of asylum seekers used in this report is reduced by the number of withdrawn applications from April 2015 through March 2017 – a total of 350,000 applications. This adjustment allows for a three-month buffer for earliest arriving applicants in early 2015 and applicants arriving in late 2016 to see their application withdrawn.9 Withdrawn applications are either implicit (applicant did not appear for meetings with officials and thus had their application withdrawn by the country of application, probably moving onto a different country to apply) or explicit (applicant requested their application be removed).10

Estimating the number of approved asylum seekers

Once asylum seekers submit their applications, they are given food, medicine and shelter as they wait for their case to be reviewed for the first time by immigration personnel. This wait time can range from a month to more than a year, depending on the nationality of the asylum seeker as well as the country of application.

Eurostat provides quarterly decision data (acceptance or rejection) for first-time asylum applications. But, since it can take some time for those applying in early 2015 to hear a decision on their case, quarterly decision data used in these estimates needed to be delayed by the application wait time in each country. For example, if a European country’s average wait time for application review was six months, then decision data for the third and fourth quarters of 2015 (allowing for six months for asylum seekers filing applications in early 2015) and all of 2016 were used.

Researchers primarily used estimated wait times from AIDA (Asylum Information Database), a database service that collects data on the asylum seeker process for selected European countries. Most wait times are in months, and the nearest quarter of asylum decision data from Eurostat (usually the following quarter after the reported average waiting period) was assigned to the country of application.11 In some countries of application, wait times are specific to certain nationalities.

All positive decisions were recorded as approvals, regardless of the kind or length of stay for refugee status granted to the applicant (e.g., Geneva Convention, humanitarian, subsidiary).12 For the Pew Research Center estimate, the number of approvals was recorded by country of application and nationality of the applicant.13



Estimating the number of rejected asylum seekers

Decision data from Eurostat also provide the number of asylum seekers whose applications were rejected. The estimated average wait times in calculating the number of rejected applications were used in the same way as approved applications to estimate the number of rejected applications from 2015-16. As with approved applications, only data on applications yet to receive a first decision were used.

Once an application is rejected, there are three possible directions the applicant’s status can take: (1) appeal the decision – a process that can take months or years to pursue while they remain in Europe on a temporary, yet legal basis; (2) be returned to their home country or some other non-EU country; or (3) continue to reside in Europe, despite being unauthorized to do so. Pew Research Center estimated the likely number of applicants in each category.

Number of rejected asylum seekers that were likely appealing their first application decision

Annual counts of rejected asylum seekers with appeal decisions are available from Eurostat. However, the submission date of their application appeal is unknown. Consequently, data for the number of appeals are comprised of rejected, first-time applicants from as little as a month prior to several years before. An estimation procedure is needed to calculate the share of rejected, first-time applications during 2015 and 2016 that were likely in the appeal process as of the end of 2016.

Historical appeal rates were calculated using the number of applicants with an appeal decision (positive or negative) in 2016 divided by the number of rejected decisions a year earlier in 2015 for each country of application and nationality combination. The appeal rate expresses the likelihood that an asylum seeker of a given nationality group applying for asylum in a particular country of application would appeal a rejection. This appeal rate was applied to rejected applications for each country of application and nationality group in 2015 and 2016.14

If the appeal rate exceeded one (in other words, the number of appeals in 2016 for an application country and nationality group was greater than the number of rejected first-time applications in 2015), then it is assumed that all applicants in that country of application/nationality group appealed their negative, first-time decisions.15

Number of rejected asylum seekers that were likely returned to their home countries or other non-EU country

A portion of asylum seekers each year are returned to their home countries or another non-EU country after having their application rejected. The annual number of foreign nationals who were ordered to leave and have effectively left EU member countries, Norway and Switzerland is available from Eurostat. These data, however, are not specific to asylum seekers. The data represent any person (asylum seeker, migrant worker or visitor) found to be living illegally in Europe.

Estimates for the number of rejected asylum seekers in 2015-16 who have been returned is based on the assumption that any returns in 2016 from a specific country of application and for a particular nationality group not first estimated to have been in the application appeal process are asylum seekers that have been returned to their home or other non-EU country.16

Number of rejected asylum seekers that were likely unauthorized to live in Europe

Remaining, rejected applicants not estimated to have appealed the rejection decision or who have not been returned to their home or other non-EU country were deemed to be asylum seekers residing in Europe without authorization. In some countries it is possible that a portion of this group may have some kind of temporary, legal status, such as humanitarian need, to temporarily remain in Europe. Others may not be in Europe and had left before the end of 2016.

Estimating the number of asylum seeker applications waiting for a first decision

The remaining number of asylum applications that have been neither approved nor rejected using the estimation techniques earlier explained are considered to have been under review for the first time as of the end of 2016. In other words, these applicants are waiting for their initial application decision. This number of applications is calculated by subtracting the total number of decisions (approved and rejected) from the total number of applications (or “net” after the removal of withdrawn applications) received in 2015 and 2016.

Eurostat provides a monthly tally of the number of pending applications yet to be processed, but the actual submission date of these applications is unknown. Consequently, the total number of pending applications includes all previous submission dates, as little as a month prior to several years before. And, Eurostat’s number of pending applications can also include duplicate applications that would later be found to exist across two or more countries. Nonetheless, the Center’s estimates of net applications received in 2015 and 2016 still under review as of the end of 2016 were compared with Eurostat’s number of pending applications also at the end of 2016. In most countries of application, the number of pending applications reported by Eurostat was the close to the number of pending applications estimated in this study.

For the purposes of results presented in this report, “waiting” applicants include those both waiting for a result in their initial application (the estimate calculated in this last step) as well as those waiting for an appeal decision (rejected applicants who were likely in the appeal process) explained earlier.

  1. Eurostat provides statistics on requests from a country of application to move applications to the first country asylum seekers entered. However, the number of transfer requests in 2015 and 2016 were far fewer than actual withdrawals stated by application countries. It is likely that Dublin procedures are ongoing and take additional time to process than actual withdrawals. In order to provide a more accurate number of total applications, removing withdrawn applications from the total number of asylum applications was a more favorable approach for this study.
  2. A small number of nationality/country of application groups had more applications withdrawn than total applications during the period. This amounted to only 0.3% of total withdrawn applications. The net number of applications for this small number of nationality/country of application groups was assumed to be zero.
  3. Countries of application with no reported wait times were assigned the third quarter of 2015, the most frequently used decision quarter among countries with reported wait times.
  4. Number of approvals does not include rejected applicants applying in 2015-16 that may have gone through an appeal process and won that appeal before the end of 2016. Given the backlog in applications from the refugee surge, it is assumed this number of approved applicants after appeal is small.
  5. In some country of application and nationality pairings, the number of decisions was larger than the net number of applications for the 2015-16 period, possibly because some decisions were based on applications in earlier years. In these cases, all applications were assumed to be decided and were separated proportionally by the total breakdown of positive and negative decisions found in the data. The number of decisions exceeded the number of applications in 355 country-of-application/nationality cases.
  6. For many countries, appeal decisions are generally granted within a year. Thus, this appeal rate is based on the assumption that all those applicants who had a first-time, rejected application in 2015 had a final appeal decision (positive or negative) the following year. This assumption may not always be true as some applicants may have been rejected at the end of 2015, but did not get an appeal hearing until early 2017, or later. Nonetheless, without monthly appeal data, this is the best rate measure that can be applied to the study.
  7. The estimates in this study may tend toward a greater number of rejected applicants filing for appeals as the appeals were considered first before returns. However, court dockets, at least in Germany, suggest that the most common step among rejected, first-time applicants is to file an appeal, not wait for deportation or remain in Europe unauthorized. Thus, it makes sense for the appeal rates to first be applied to rejected, first-time applicants rather than data on returns.
  8. It is likely that some of the rejected, first-time asylum seekers estimated to have returned to their home or another non-EU country might be foreign nationals that entered Europe before the recent refugee surge of 2015-16.