Reporting on a UN meeting, a recent New York Times article lamented the poor state of international migration data. The story ("UN Coaxes Out the Wheres and Whys of Global Immigration," July 7, 2002) points out a lack of official data and harmonized definitions, weak analytical tools, and abuse by officials as the underlying reasons for this deplorable situation.
However, the above assessment does not apply to all aspects of international population movements. The picture for asylum and refugee statistics, for instance, appears less bleak. Statistics on international forced migration are available for virtually all countries, are frequently updated, and are comparable among countries. What is the main reason for this qualitative difference between asylum and refugee statistics on the one hand, and those on international migration on the other?
A Wealth of Statistics
Four critical factors contribute to the relative wealth of refugee and asylum statistics. First, there is the global refugee convention. Most countries have signed the 1951 UN Convention Relating to the Status of Refugees and its 1967 Protocol, and have implemented the provisions in their national laws.
Second, the convention clearly defines who a refugee is: someone who is living outside his or her country of origin and who has a well-founded fear of persecution for reasons of race, religion, nationality, membership in a particular social group, or political opinion.
Third, together with the convention, a global international agency was created to oversee its implementation. Both the convention and the mandate of the Office of the United Nations High Commissioner for Refugees (UNHCR) refer explicitly to the obligation of countries to provide statistical data. Clearly, the drafters of the convention realized that statistics from governments were a key element in protecting and assisting refugees.
Fourth, refugees form a special group of people crossing international borders. While countries have the sovereign right to determine which foreigner has the right to enter and stay, refugees are an exception. States are obliged to admit refugees as long as they are in need of protection. This unique right ensures that refugees are carefully screened, registered, documented, and thus... counted.
Continuous Data Collection
The means to implement the laws are as important as the laws themselves. With more than 4,000 staff in 120 countries, the UNHCR is in daily contact with refugees, governments of host countries, and countries of origin. The office works with over 500 partners to implement refugee programs. Its global presence and infrastructure ensure a continuous gathering of information on refugees, including numbers and profiles. Obviously, it is much easier to collect data through the intermediary of UNHCR country offices than to wait for national bureaucracies to respond to a questionnaire from a faraway international body. Being "refugee professionals," UNHCR officials are not simply experts in collecting information. They also provide, through their daily contact with officials, training on standards, including the ways to count and record refugees.
In order to count properly, you need very strong motivations. In the case of asylum seekers and refugees, providing protection and rendering assistance appear to be very good ones. However, although asylum and refugee statistics are widely and increasingly available, they are not free from problems. There are four main weak points:
First, the very fact that the amount of data is rapidly growing has implications for their comparability. In Germany, for instance, the number of data elements reported by asylum authorities grew from five at a point 10 years ago to 21 today. Such increases in detail make it harder to compare country experiences as long as the data are based on national asylum law. Indeed, the growing complexity of asylum statistics reported by industrialized countries poses a challenge to analyzing international trends. Areas in which the asylum data have become increasingly complex include appeals processes; persons allowed to remain for reasons other than convention grounds; reasons for formally rejecting claims; and processes for not admitting claims to the asylum procedure. Clearly, an important factor in the rapid increase in availability of refugee and asylum data is technology, which allows for a much more efficient sharing of data from administrative systems.
Second, the New York Times article made specific reference to the tendency on the part of officials to inflate or obscure migration trends. This danger is relatively strong in the area of asylum and refugees. The UNHCR data are widely used by officials and the media to defend or criticize positions and policies. Yet, statisticians do not have much say in the use of "their" data beyond the presentation and basic analysis they provide. Indeed, the statistical goals of objectivity and transparency allow users to draw their own conclusions. Providing data in an impartial and balanced manner is the best data producers can do. It is up to analysts to make good use of the data.
Third, although the UN refugee convention spells out the definition of a refugee, its interpretation may be subject to debate. Whereas the objective factors causing a refugee to flee are relatively easy to document (for example, war, persecution of a particular ethnic group, etc.), individual motives are more subject to interpretation. Moreover, refugee definitions may change over time: groups considered refugees one year may be dropped from the national statistics the next, and vice versa. UNHCR statistics generally reflect the government's position, which may sometimes differ from the interpretation of non-governmental organizations. In short, defining who is a refugee may not be as easy at it appears.
Fourth, data coverage is uneven and depends on the purpose and method of data collection. As a general rule, refugees benefit from being counted because it increases their access to safety and services. This is evident in refugee camps, where official registration data often overestimate the size of the resident population: it is easier to register than to de-register people. The opposite is true for those refugees who are not dependent on international assistance or whose stay is relatively safe. These refugees tend to be underreported by international agencies.
Internally displaced persons (IDPs), although they are by definition not international migrants, merit some consideration here because they move for the same reasons as refugees. However, while their reasons for displacement may be similar, UNHCR's experience shows a great gap in data quality between refugees and internally displaced. IDP estimates are often very rough, and they tend to differ greatly in terms of the source (governments, international agencies, non-governmental organizations).
What are the main reasons for this gap in data quality? First, the internally displaced do not benefit from the same level of international protection and assistance as refugees. There is no specific international convention protecting IDPs. Also, there is no single international body entrusted with their protection and assistance. Second, IDP movements typically involve short distances, and often, short time frames. Third, as in the case of voluntary migratory patterns, internal movements are much less recorded than international movements. The inherent interest of a receiving country in who is entering is absent in the case of internal movements, which are free of restrictions and subject to fewer administrative hurdles. Fourth, considering that they are still living in the country where they have been persecuted, the internally displaced may be less willing to register than those who enjoy the protection of their asylum country.
This article was contributed by Bela Hovy, Head, Population Data Unit, United Nations High Commissioner for Refugees (UNHCR), Geneva. The views and opinions expressed are those of the author and do not necessarily represent the views of the Office of the United Nations High Commissioner for Refugees.