Migrant arrivals and deaths in the Mediterranean: what do the data really tell us?

The policy and media gaze focuses on numbers of migrant arrivals and deaths. There are problems in the data for both categories.

When looking at numbers of arrivals and deaths in the Mediterranean, it is important to ask how robust the data are and what they represent. If we are to frame and inform more accurately the policy challenges facing Europe, it is crucial that data are of good quality and as complete as possible and that their limitations are fully understood.

Arrivals data

Headline figures suggesting that over 900,000 migrants arrived on the shores of Mediterranean countries in 2015 have captured the attention of media and policymakers. These arrivals data are calculated from official sources and are usually collected at borders. They can include estimates of illegal border crossings, attempts to cross a border, arrests, refusals of entry, initial registrations of intention to seek asylum, and actual asylum applications. The data are sometimes a mix of flow and stock data, and can refer to different time periods. Furthermore, the use of administrative sources means that data may reflect changes in migration patterns but also changes in border patrol practices. Data may relate to numbers of events, rather than numbers of individuals, thus possibly double-counting when, for example, an individual is apprehended twice. In other instances, under-counting may occur when migrants cross a border undetected. Even in just one country, combining data from different sources produces only an approximation of a measure. When the numbers for different countries are combined, the uncertainties are compounded.

However, the increasing use of the same sources and definitions over the last few years means that there is some consistency and there is no doubting the magnitude of the increase in arrivals compared to recent years. The figures more than quadrupled from 220,000 in 2014 to nearly 900,000 during the first eleven months of 2015.[1] Regularly collected data on the socio-economic profile of migrants arriving in Europe are necessary to design sensible policies. The arrivals data give us a picture of how many arrive and where they have come from, but there are few media reports on the skills and education of the migrants – which perpetuates the impression that is often given in the media that Europe has to absorb a rising number of people fleeing conflict and poverty, who have relatively little to offer the continent. Without minimising the need for protection, the reality is that many of the migrants are more skilled and educated than is commonly perceived.

Migrants who die are rarely identified

As in 2014, the Mediterranean crossing has proved extremely dangerous, taking more than 3,550 lives in the first eleven months of 2015, already several hundred higher than the total for 2014.[2] Despite the increase in flows on the Eastern Mediterranean route, the Central Mediterranean route remains by far the more dangerous of the two. In 2015, over 80% of deaths in the Mediterranean occurred on this route, with about two deaths per 100 migrants attempting the crossing.[3] In contrast, at the time of writing, the Eastern Mediterranean has seen over 590 deaths, or around two deaths per 2,500 travellers.[4]

While the data we have show a devastating loss of life, the numbers themselves, and methodologies used in their collection, are riddled with holes and challenges. There are a number of reasons for this; several are inherent to the nature of irregular migration, while others relate to the methods of data collection and sources of information utilised. Those inherent to the nature of clandestine movement include the lack of detailed passenger lists on migrant boats, the tendency for migrants to dispose of identity documents, and the convoluted journeys migrants take, often through multiple countries. Still other challenges are faced when tracking deaths along migratory routes in other parts of the globe, mainly due to the remoteness of terrain travelled.

Data on deaths are not collected by Frontex nor by national governments in a systematic way, and data collected by coast guards tend to present almost no demographic information. Therefore, a variety of sources must be used to try to piece together a more complete picture. These can include the media and those international organisations and NGOs involved in receiving survivors. In other areas of the world, data sources on migrant deaths are far scarcer, with almost all information coming from the media, NGOs and, at times, coroners’ offices. There is no standard approach to collecting data on deaths, nor a common definition of what constitutes a migration-related or border-related death. Because of all these challenges, figures can differ between organisations like the International Organization for Migration (IOM) and the United Nations High Commissioner for Refugees, and also between NGOs in Europe.

What data do exist are highly incomplete, lacking much basic demographic information for each individual, including nationality, age, sex, cause of death, and whether the body is identified or not. This is in part due to the lack of information – the majority of bodies are lost at sea with only rough estimates available concerning where they are from and their genders. The fact that numbers of the missing are usually estimates based on the testimonies of survivors who often do not know how many people were on board a boat also adds to discrepancies between data published by different organisations.

When entire groups die at sea, families back home may not know if their relatives made it to the coast, perished in the sea crossing or lost touch for other reasons. It is not known how many deaths go unreported. Even when estimates of the missing are available following shipwrecks in the Mediterranean, bodies are more than often not found. Even when considering only the bodies recovered from the sea, a recently launched database tracking deaths in the Mediterranean since 1990 has found that of the bodies brought to southern Europe, almost two thirds had not been identified.[5] There has been little discussion of how to improve identification and little coverage of the implications for the families left behind, who often do not know if their relative is dead or alive. [6]

In conclusion…

Data on migration are increasingly cited in the media and used to frame migration policy discussions in Europe. Because the way in which these data are presented is likely to have an influence on public perceptions of migration in Europe and on policymakers´ responses, it is important that they are presented clearly and accurately.


Frank Laczko flaczko@iom.int

Head, Global Migration Data Analysis Centre, International Organization for Migration


Ann Singleton asingleton@iom.int

Senior Research Fellow, School for Policy Studies, University of Bristol

on secondment to the Global Migration Data Analysis Centre, International Organization for Migration


Tara Brian tbrian@iom.int

Research Officer, Global Migration Data Analysis Centre, International Organization for Migration


Marzia Rango mrango@iom.int

Research Officer, Global Migration Data Analysis Centre, International Organization for Migration


[1] Data for 2014 are from Frontex; data for 2015 are an IOM estimate based on data from relevant governments.

[2] IOM Missing Migrants Project http://missingmigrants.iom.int

[3] IOM Missing Migrants Project; arrivals data are an IOM estimate based on data from relevant governments.

[4] IOM Missing Migrants Project 

[5] Vrije Universiteit Amsterdam Human Costs of Border Control project www.borderdeaths.org

[6] See Grant S (2011) ‘’Identity unknown’: migrant deaths at sea’, Forced Migration Review issue 38 www.fmreview.org/technology/grant.html



Opinions in FMR do not necessarily reflect the views of the Editors, the Refugee Studies Centre or the University of Oxford.
FMR is an Open Access publication. Users are free to read, download, copy, distribute, print or link to the full texts of articles published in FMR and on the FMR website, as long as the use is for non-commercial purposes and the author and FMR are attributed. Unless otherwise indicated, all articles published in FMR in print and online, and FMR itself, are licensed under a Creative Commons Attribution-NonCommercial-NoDerivs (CC BY-NC-ND) licence. Details at www.fmreview.org/copyright.