YEARS OF REFUGEE LIFE IN SERBIA - CHALLENGES FOR A NEW BEGINNING: STAY OR RETURN HOME?

Abstract

According to the data of United Nations High Commission for Refugees (UNHCR) for 1993, out of the total number of 15 million refugees in the word, more than 2.5 million people were from Bosnia and Croatia, where refugees (citizens of the same country) moved to other parts of the country. According to the international law, all these persons have the status of refugees, but according to the republican legislation of Serbia and Montenegro at that time, they were divided into three categories - refugees (43%), exiled persons (35%), and persons without status (22%). According to 2005 Census, the gender structure of the refugee population was balanced (women comprised 51.4% of population) and it is approximately the same as in domicile population. [...]this forces the Serbian authorities to help them in solving property issues and other demands they have in former domicile abodes (La?evi?

Headnote

Abstract: The paper represents project results of a multi-year long research on the problem of refugees. First, we started from the fact that although the number of refugees in Serbia is decreasing, their social status has been bad for years and the qualitative changes are slow. We tried to answer the question relating to the changes in number and quality of collective centres. Secondly, we analyzed the relationship between refugees and domestic population. We confirmed the hypothesis that refugees from Croatia and Bosnia after years of living in Serbia hope to integrate their community into the host society. Thirdly, there is no doubt that repatriation in Croatia and Bosnia is the most acceptable solution for the majority of refugees, but it is not the realistic solution due to security issues, property security and full-time employment in the places of their origin.

Keywords: refugees, socio-economic status, host society, future, Serbia

DOI: 10.3176/tr.2011.3.02

Introduction

International migrations across countries and continents are fuelled by global tragedies, globalization trends, and the search for refuge and opportunities (Pieke 2004, UNHCR 2004, Stewart et al. 2008). In the past three decades the numbers of international migrants have escalated from 75 million in 1980 to 175 million in 2000, 159 million being classified as voluntary migrants, and 16 million as refugees (United Nations 2005, Stewart et al. 2008). The civil war in the area of former Yugoslavia between 1990 and 1996 led to the largest forced migration in Europe after the Second Word War (Harvey 2006). According to the data of United Nations High Commission for Refugees (UNHCR) for 1993, out of the total number of 15 million refugees in the word, more than 2.5 million people were from Bosnia and Croatia, where refugees (citizens of the same country) moved to other parts of the country. One part of the refugee population ended up in Western Europe, North America and Australia. The majority of refugees were accepted by FR Yugoslavia (Serbia and Montenegro). A complicated political situation, poor resourcefulness and hesitancy of international factors to get involved more decisively in problem-solving, as well as international isolation of FRY, additionally complicated the situation of taking care and protecting the refugees. The situation of refugee population from SFRY in new living conditions and new surroundings was studied by demographers in Serbia (Knezevi? 1992, Kondi? 1997, Petrovi? 1997, Milosavljevi? 1997, Gre?i? 1998, Cvetkovi? 1998, Jakovljevi? 1998, Matkovic 1999, Luki? and Nikitovi? 2004, Nikitovi? 2005, Stevanovi? 2005, La?evi? and Stankovi? 2006, Nikitovi? and Luki? 2009) and in the world (Hammond 1999, Black 2002, Black and Gent 2004, 2006, Ambroso 2006, Harvey 2006, Eastmond 2006, Black et al. 2006). Their work was the methodological and scientific base for this research.

In the first section we analyzed the change in the number of refugees in Serbia, as a consequence of reintegration, repatriation and moving to the third countries in the period between 1996 and 2009. The second part of the paper refers to the analysis of spatial distribution of refugees in Serbia in different districts and regions. In the third chapter, on the basis of data from UNHCR and CRRS, we carried out the analysis of demographic and socio-economic position of refugees in Serbia. The fourth chapter refers to the integration of refugee population in Serbia, and the relationship between refugees on the one hand and local political elites on the other. Namely, local political elites manipulated the destiny of refugees, glorifying refugees as symbols of patriotism and rural purity, and then as victims and heroes. This is the reason why towns in Serbia, especially Belgrade were highly reserved in welcoming the wave of refugees. The native city way of life was threatened by newcomers from rural Croatia and Bosnia. Antagonism towards them was also present in the city media and it reflected political-economic and cultural aspects of the new situation (Martin 2002). With time, and with the political-security situation in the surroundings calming down, the mentioned tensions weakened. To confirm this fact, we found in our research that the majority of refugees wanted to continue living in Serbia. This is why the integration of refugees, as a constant challenge for state and local authorities, becomes of great significance.

Methods and data

In this paper we used methods of human geography and demography, which combine methods of scientific research with the methods of communicating the achieved results. Those two standpoints determined the methodology in our research, in defining and communicating results. The complexity of the subject matter required the application of scientific methods suitable for setting up hypotheses, proving attitudes, verifying and forming conclusions. In methodological procedures, a general and mathematical-statistical method were used, and as special analysis, synthesis, comparison, generalization and cartographic method, which enabled demographic variables to be brought in the chorological-chronological connection and their effects in refugee population. The representative nature of demographic data and statistical-demographic variables, the territorial distribution of refugee population in districts and time frame of research are the main indicators of the performance of scientifically based results presented in the paper. Based on the selection of the analyzed demographic parameters, the obtained results were used to explain the phenomenon of refugees in Serbia, their socio-economic position and status.

The problems encountered in our demographic research of refugee population in Serbia were often differing methodologies of the Commissariat for Refugees of Serbia (CRRS) and UNHCR. At the same time, an important characteristic of the refugee phenomenon in Serbia is the fact that the largest number of refugees (73.3%) were staying at the houses of their families, relations, friends and kindly people, but not in collective centres as is the practice in the world (Gre?i? 1998:94). They were not often listed in refugee registers, which led to discrepancies in statistical data of UNHCR, CRRS, international and national NGOs.

Changes in the number of refugees in Serbia 1996-2009

Wars in the area of the former Yugoslavia produced a large number of refugees from Croatia and Bosnia. The majority of the refugees in Serbia were ethnic Serbs, 93% (Matkovi? 1999, Luki? and Nikitovi? 2004). Moving to Serbia also occurred on the eve of the collapse of the former Yugoslavia. The character of those migrations in the late 1980s and at the beginning of the 1990s pointed to the future development of events. Modern refuges in the area of present Serbia (which together with Montenegro used to comprise the former Federal Republic of Yugoslavia, FRY), can be followed since 1990. That was the year which marked the beginning of the arrivals of family members of military officers, students and other people who due to the newly created situation found more secure refuge in the newly formed state.

Moving to Serbia was followed by property transfers and a regulation of citizenship. Those processes intensified in 1991 with secessions of Slovenia and Croatia and especially in 1992 and 1993 when the civil wars in Croatia and Bosnia broke out. With the escalation of wars in Croatia (1991) and Bosnia (1992), over 200,000 people came to Serbia and the status of refugees was officially given to people originating from Bosnia (Luki? and Nikitovi? 2004). The United Nations' Security Council imposed sanctions to FR Yugoslavia in 1993. Sanctions together with hyperinflation, as well as irregular and insufficient aid from international community, above all UNHCR, made the situation for domicile population and refugees more difficult. The research conducted in 1993 shows that refugees cited as the most important reasons for abandoning their homes: war danger, fear and insecurity, threats of hostile population (including hostile military formations) from their immediate surroundings, physical expulsions, destruction of property, psychological and physical harassments, death or injuries of the closest family members and similar reasons. In August 1995, Serbia accepted the biggest wave of refugees from Croatia (400,000 according to UNHCR, 1996), but they did not get the status of refugees, but the status of exiled persons, even though there was no essential difference between these two categories. Out of 200,000 Serbs before Croatian military operations only 5,000 remained in Croatia. Estimations show that in the period from 1991 to 1995 between 300,000 and 500,000 Serbs found refuge in Croatia (Gre?i? 1998). The process continued after the Dayton Accords. The total number of refugees after 1996 was 61,992 people or 16.3% of the total. According to the international criteria, persons who came to Serbia from Croatia and Bosnia after August 1995, did not get the status of refugees, but were defined as persons without status (Stevanovi? 2005). The last wave of refugees came at the end of the 20th century and the beginning of the 21st century, when 200,000 Serbs as well as other members of non-Albanian nationality, left Kosovo and Metohia. According to the international standards and Serbian legislation, those refugees are treated as internally displaced persons, because their migration occurs within the borders of the same country, that is, they do not have the status of refugees.

The first census of refugees was organized by CRRS in cooperation with UNHCR in 1996. 537,937 refugees were registered in Serbia on that occasion. This number included persons with the status of refugees or exiled person, as well as those without status. According to the international law, all these persons have the status of refugees, but according to the republican legislation of Serbia and Montenegro at that time, they were divided into three categories - refugees (43%), exiled persons (35%), and persons without status (22%). People who came to Serbia before August 1995 got the status of refugees. Persons who came to Serbia from Croatia in August 1995 got the status of exiled persons, and the persons who came after August 1995 were not registered and are marked as persons without status. The census in 1996 also registered 79,791 war-affected persons: those are people who came to Serbia for the same reasons as refugees, but they already had Serbian citizenship, so they could not get the status of refugees. Thus, in 1996, the total number of registered refugees and war-affected people in Serbia was 617,728. The census established not only the exact number but also the types of the refugee population. In 1996 in Serbia there were 290,667 refugees from Croatia or 54% of the total number of refugees; 232,974 refugees from Bosnia (185,065 refugees from Muslim-Croatian Federation and 47,909 from the Republic of Srpska) or 43.3% of the total refugee population; 3,014 from Slovenia (0.56%), 1,306 (0.24%) from Macedonia etc. Ethnic Serbs comprised 92.3% of refugee population, Montenegrins 0.3%, Muslims 0.9%, Yugoslavs 1.5%, others 3.7%, etc.

The second census of refugees was carried out in April 2001 by CRRS and UNHCR. It registered 377,131 refugees and exiled persons, and 74,249 waraffected persons. The majority of registered refugees originate from Croatia (64.3% or 242,624), which means that their share in refugee population increased by 10.3% (1996-2001). These data point to the existing problems of refugees from Croatia that are connected to their return or integration in Serbia. Refugees from Bosnia comprised 35.5% (133,853 persons), with their share in the total number of refugees decreased by 7.8% (1996-2001), thanks to either their return or getting Serbian citizenship. The number of refugees continued to decrease. Certain number of people applied and got documents of Serbia, because since 2001 the legislation created more simplified granting of citizenship for refugees. Some of them entered programs of local integration, a certain number went to third countries and one part returned to the country they took refuge in. The third census was carried out in the period from November 2004 to January 2005 by CRRS and UNHCR. The revision of the refugee status was carried out and the status of refugees was granted to 104,246 persons. According to the 2005 Census the majority of refugees (73.4%) or 76,546 came from Croatia, and 26.4% or 27,541 refugees from Bosnia. The trend of a growing share of refugees from Croatia continued and the share of refugees from Bosnia continued to decrease. More than a half of the registered refugees in Serbia (51.7%) came during August and September 1995. Serbs were 95.7% of the registered refugees. Persons with refugee status comprised 1.45 of the total number of population in Serbia. The main reasons for the decrease in the number of refugees in Serbia should, above all, be found in the possibility of getting Serbian citizenship and identification card (from which a certain number was excluded in local integration), return to the countries of origin, but in moving to third countries as well.

The largest number of refugees (154,300) got the citizenship of Serbia and de jure integrated. However, citizenship did not significantly change the economic status of the majority of them, because granting citizenship represents a formal, but not an essential change. The second group by its number (115,700) comprises persons who were not registered in censuses in the period from 1996 to 2009. Those people were not registered either among returnees (146,500) or among persons who immigrated to the third countries with the help of UNHCR (22,400). This group represents the group of refugees who went to third countries without UNHCR mediation and those who died in the period from 1996 to 2009. The process of going to third countries went on intensively, especially between the first two censuses (1996 and 2001), but it continued later as well. A certain amount of immigration to the third countries was carried out through humanitarian programs of the governments of the third countries.

Spontaneous repatriation is more evident when we talk about refugees from Bosnia. Out of 125,500 spontaneous returnees, 65,000 originate from Bosnia. However, regarding returnees with the help of UNHCR, out of the total number of returnees (20,000), 13,900 or 69.5% is from Croatia. In addition to this, the total number of returnees includes a certain number of the so-called illusory returnees - persons who took Croatian documents because of different incentives connected to the trip, but de facto live in Serbia. The main reasons for a small number of returnees include lack of employment, security in the place of former residence, a large social gap between Croatian and Serbian population etc.

Changes in spatial distribution of refugees

On the basis of data about the number of refugees in districts in Serbia from 1996, 2001, 2005 and 2009 (CRRS and UNHCR), we can conclude that the decrease in the number of refugees happened evenly on the whole territory of Serbia and that spatial distribution of refugee population was not significantly disturbed. The whole period was marked by refugee migrations to bigger urban areas of Serbia. Since the beginning of the refugee crisis in Serbia, refugees have mostly lived in towns, so we can talk about their "deruralization" (La?evi? and Stankovi? 2004). Before war conflicts, the refugee population also lived in urban settlements in Croatia and Bosnia, but coming to Serbia made them more attached to urban settlements. In 2002, 60% of refugees lived in Serbian towns, which is more than the average number for Serbia (56%). The basic reason for this is of economic nature, because towns like Belgrade and Novi Sad offer more favourable employment conditions and livelihoods. This reality made refugees even more detached from agriculture - at the time, 9% of refugee population was employed in agriculture, while the average number for Serbia is 19% (La?evi? and Stankovi? 2004). Most of the refugees were in the tertiary sector. Small changes of spatial distribution of refugees in Serbia do not exclude the presence of local migrations within the borders of districts and regions. On the contrary, refugees move from rural settlements and smaller urban centres to bigger regional centres where employment conditions are more favourable. Despite the trend of refugees moving around and an increase in urban refugee population, during the first decade of the 21st century there is a mild proportional decrease of refugee population in Belgrade and Voivodina, and a proportional increase of refugees in Central Serbia. This trend is explained by better employment possibilities for refugees in Belgrade and the towns of Voivodina, which leads to suspension of refugee status and gaining citizenship and identification cards of Serbia. In addition, refugees' settlements were set up on the periphery of Belgrade, i.e. settlements with population comprised of refugees or people who used to have a refugee status.

Refugees from Croatia, as the biggest group of refugees were the majority in almost all parts of Serbia during the whole period, and they were especially dominant in Voivodina. Only in the part of Serbia that borders with Bosnia (districts of Loznica, Bajina Basta, Mali Zvornik, and Ljubovija) have relatively more refugees from BIH than from Croatia.

In the period of the first census in 1996, the majority of refugees found their refuge in Voivodina - 229,811 or 42.7% of the refugee population in Serbia. In Central Serbia there were 148,367 refugees or 27.6%, and on Belgrade territory 140,662 refugees or 26.1%. In 1996 refugees comprised 8.7% of the Belgrade population, where the highest concentration was in border and town districts: Barajevo, Zemun, Grocka, Zvezdara and ?ukarica. The largest number of refugees settled in Zemun (22,085), New Belgrade (21,392), ?ukarica (16,026), Zvezdara (16,002) and Vozdovac (10,506). In Voivodina, the highest share of refugees was in the regions bordering Croatia (Sremski, Juznoba?ki, Zapadnoba?ki). The highest concentration of refugees among regions had Sremski region, where on average every fifth inhabitant was a refugee (22.2% of refugees). The districts in Sremski, Juznoba?ki and Zapadnoba?ki region had the largest number of refugees. The largest number of refugees was registered in Novi Sad (40,602 or 14.6%). Distribution of refugees in Voivodina was mostly even, so a large number of refugees were not only registered in the biggest districts with important urban centres. On the contrary, with small population, non-urbanized and economically undeveloped districts also underwent a wave of refugees. The vicinity of the places they took refuge in and the distribution of colonists immediately after the Second World War had a huge influence on the distribution of refugees in Voivodina. In 1996 refugee population in Voivodina was 11.5% of the total population. On average, every tenth inhabitant of Voivodina was a person with a refugee status, so numerous contingents like this one influenced the demographic growth of population in Voivodina. This was especially reflected in the results of 2002 Census of Serbia. Thanks to the wave of refugees in the period from 1991 to 2002, Voivodina had a positive average growth rate (2.8%).

In Central Serbia, except Belgrade, more than 2,000 refugees were registered in 16 districts. Those are economically developed districts with a larger population (Sabac, Kraljevo, ?a?ak, Nis, Uzice etc.). Districts of Ma?vanski region, bordering Bosnia, had a higher share of refugees in the total population. Among them the dominant place occupies the district of Loznica with 26,790 refugees (30% of total population). In 1996, refugee population comprised only 3.5% of population in Central Serbia (without Belgrade).

According to 2001 Census, the majority of refugees (48.7%) or 183,721 settled in Voivodina, 30.1% or 113,580 on the territory of Belgrade and 21.1% or 79,443 in Central Serbia. Changes in spatial distribution of refugees were minimal, so the border regions with Croatia and Bosnia still had the highest proportional share of refugees. Sremski region had the highest share of refugee population 17.7%, even though there was a certain decrease in the number of refugees. In Belgrade Metropolitan Area refugee migrations to suburban districts took place, where they found better living and employment conditions. So, the share of refugee population in Barajevo, Zemun, Mladenovac and Sopot increased in comparison to 1996. A drastic drop of refugees in Serbia happened in the period of two censuses: 2001 and 2004/2005. This trend was influenced by a large number of factors. Among them, the most important factors are granting of the citizenship of Serbia and integration in the society, while returning to homeland and going to foreign countries are less important.

In 2005, out of the total number of refugees from Croatia and Bosnia, 48.4% or 50,363 were settled in Voivodina. Sremski region with 4.5% of refugees in population remained the part of Voivodina with the highest concentration of persons with the status of refugees. In Belgrade the concentration of refugee population was 28.7% (29,866), in Central Serbia 22.7% (23,601) and a minor number 0.2% (257) on Kosovo and Metohia. Among Belgrade districts, Zemun and Barajevo were with the highest share of refugees (4.4%-5.1%).

The next period remained with the existing spatial distribution of refugee population in Serbia, with a constant decrease of the number of refugees. According to the data from 2009, a proportional share of refugees in 99 districts out of the total of 161 was less than 1%. The largest number of districts with proportionally the least share of refugees is in Central Serbia (without Belgrade). In 2009 the share of refugee population in the total population of Central Serbia was only 0.5%. This is because a small number of refugees during the whole period is typical of mostly economically undeveloped and depopulated districts, without bigger urban settlements. The border districts of Ma?vanski region stand out as districts with the highest share of refugees in the total population - Loznica (2.1%) and Mali Zvornik (2.1%). Districts of Ma?vanski region that also have a large number of refugees are Sabac (1,936) and Loznica (1,761). The city districts of Belgrade had a higher percentage of refugees, where 1.5% of the total population had the status of refugees. The highest proportional share of refugees remained in peripheral districts of Belgrade: Barajevo (3.6%), Zemun (3.5%), Sur?in (2.6) and Obrenovac (2.0). There are numerous reasons for this kind of spatial distribution of refugees in Belgrade. The concentration of refugees in the above mentioned districts was influenced by lower costs of living in comparison to the central urban districts, land with permits for housing construction, built housing projects in the ownership of refugees or persons with former status of refugees etc. The largest number of refugees lived in the districts of Zemun (5,513), while a bit fewer refugees were in the districts of Palilula (2,625) and New Belgrade (2,625). The highest percentage of refugees was in the districts of Voivodina, where 19 districts had 2-5% of refugees in the total population. On the level of the whole Voivodina refugees comprised 2.1% of population. Zapadnoba?ki, Juznoba?ki and Sremski region stand out in the number of refugees, that is, the districts of Novi Sad (6,606 refugees) and Pazova (3,889).

Changes in demographic and socio-economic characteristics of the refugee population in Serbia

In the period from 1996 to 2010, besides the decrease of the number of refugees, there were some changes in their demographic and socio-economic characteristics that resulted in the correction of certain activities in order to deal with the phenomenon of refugees. In addition to changes in basic demographic characteristics, there is no significant difference between refugees and domicile population. This is understandable considering the fact that in home countries - Croatia and Bosnia, the refugee population has developed for five decades in similar socio-economic, cultural, educational and other conditions as the population of Central Serbia and Voivodina. This was also augmented by the massive exodus from the 1990s, when entire families were in refuge. The demographic picture of refugees was thus more complete and real, because it included all generations. A bit larger differences could be seen in socio-economic characteristics, but they also decrease with time. Due to a relatively small share of refugees in the total population of Serbia, especially in Central Serbia, the influence of refugee population on demographic and socio-economic characteristics of population in Serbia was minimal. The most important contribution of the refugee inflow reflects in alleviation of the depopulation trend in Serbia, on the level of larger spatial macro-entities, as well as on lower level of administrativeterritorial entities (districts), depending on the registered number of refugee population in them. However, population projections show that by the middle of the 21st century positive effects of basic larger population, caused by refugees, will have disappeared due to the same reproductive behaviour of refugees and domicile population, emigration of young people etc. (Nikitovi? and Luki? 2010).

In 1996 Census women comprised 52.8% of refugee population. In later censuses they had a smaller share. According to 2005 Census, the gender structure of the refugee population was balanced (women comprised 51.4% of population) and it is approximately the same as in domicile population. The basic characteristic of gender structure of refugees is a very high share of women in the oldest population, over 60. As a consequence of men having died in the war, the share of women older than 60 was 58.8% in 2001. In 2005 women comprised even 63.9% in the same contingent. During the whole analyzed period of 1995-2005 men were more numerous among the population younger than 18.

The analysis of refugees' age structure in 1996 showed that refugees, as a rule, were younger than average in Serbia. Regarding the age structure of refugees in 1996, the share of persons between 19 and 64 was higher (60.2%), while the share of other age groups was significantly lower, especially the share of persons older than 65 (19.7%). The youngest population, below 18, comprised 21.5%, while the share of children up to 4 years old was distinctly low (2.2%), which was understandable in those conditions due to fewer births. A relatively low share of the elderly is the result of weaker mobility (older categories of population were not willing to leave their home country during wars), but of repatriation as well. So, among returnees - refugees in Croatia (60,000) the number of the elderly is significant. Only the elderly live in the villages in Croatia (La?evi? and Stankovi? 2004).

After 2001, there was a decrease of the share of the youngest population (groups 0-4 and 5-18) and an increase of the share of persons older than 65. This points to the fact that the refugee population in Serbia was older, which makes their integration more complex and more difficult. The share of refugees younger than 18 thus decreased from 14.5% in 2005 to 9.6% in 2009, while the share of older than 65 increased from 25.95% to 30.1% within the same period. In comparison to the age structure of domicile population, we can conclude that domicile population has a significantly higher share of 0-9 year-old children, as well as persons in age intervals over 50. However, besides differences in certain age intervals, the synthesis indexes show that there is actually no significant difference between domicile and refugee population in Serbia. The average age of domicile population in Serbia is 40.2, while the average age of refugee population is 39.4 (La?evi? and Stankovi? 2004). On the basis of the analysis, one can say that besides gender structure, age structure of the refugee population is also very similar to domicile population, because whole families or the majority of family members took refuge.

The education structure of refugees in Serbia, according to the data from 1996 Census, shows that the largest number of refugees older than 15 is with high school level education (46.7%) and with primary school level education (28.9%). In this age group 10.6% of refugees had higher and university education, while 8.6% had no education at all. Compared to domicile population, the educational structure of refugees was more favourable, which points to the fact that persons with higher education level were more inclined to decide to take refuge, believing in their ability to adapt and to find employment more easily. Regarding the connection between gender and education, there are more women without education, as well as those with uncompleted or completed primary school. Men dominate in the category of persons with high, higher and university level education.

The category of refugees with high school education remains the most numerous after 1996. Persons with no education comprise the oldest refugee population, average age between 68 and 69 in 2001, and 72 in 2005. There is a decrease in the number of refugees with higher and university degree, which points to their ability of finding employment more easily, solving status problem by taking citizenship etc. So, in the period from 1996 to 2005 the number of persons with higher and university degree halved. The education structure among the refugee population is less favourable than the one we find in domicile population regarding both population with primary school or with uncompleted primary school, and population with completed higher and university education.

The degree of economic adaptation of refugees is a very important factor in the process of their integration and this could also be seen in the structure of refugee population according to economic activities (Luki? and Matijevi? 2005). The analysis of refugees according to the working status in the period from 1996 to 2009 shows a clear trend: decrease of the share of economically active persons at the expense of pensioners and supported persons - above all, the unemployed. In 1996 Census 47.2% of refugees belonged to the category of economically active persons. In 2005, the share of employed persons was 19.4%. This shows that a significant number of employed population from earlier censuses lost the status of refugees. Decrease in the share of employed persons in refugee population is the result of more and more unfavourable age and educational structure (the number of the elderly and uneducated among refugees is growing, because the refugee population constantly decreases) and general socio-economic conditions in Serbia as well. These also explain a constant decrease in the number of students and pupils. Anyway, the number of employed men is twice higher compared to the female population.

The share of employed persons increases with the increase of education level. The employment rates in age groups with the lowest education level according to the research from 2006 are almost twice higher in domicile population than in refugee population. In refugee population as well as in domicile population, persons with the lowest level of education are dominantly categorized in nonactive population. In domicile population the share of the unemployed is the highest among persons with high school degree (15.33%). In the refugee population the proportion of unemployed persons is much balanced and moves from 19% with university education to 29% with completed high school. The most worrying fact is the high share of unemployed persons with high school degree (28%) and university degree (19%) in the refugee population, in comparison to 9.48% and 6.59% in domicile population. Most refugees are employed in the trade and service sector, then in crafts and production occupations, and then basic occupations. Dissatisfaction with their own socio-economic position is far more distinctive among refugees than it is in domicile population, which makes this group more subjected to demagogic manipulations (Opa?i? 2007).

The high unemployment rate is the basic indicator of the situation of refugees in Serbia. The share of the unemployed constantly increased - almost 60% of refugees were unemployed in 2005. Compared to 2001 Census the situation worsened, so one part of the refugees was forced to provide their livelihood in grey economy.

The results of research from 2006 show that among the unemployed refugees persons with completed high school dominate with 61.6%, compared to 69.39% of the unemployed in domicile population. Persons with completed primary school have 20.4% of unemployed refugees compared to 19.03% in domicile population, which is the consequence of the fact that these two education levels are always most numerous in the population. Regarding regional distribution the highest unemployment rate is in Central Serbia where the difference in unemployment rates of refugees and displaced persons, and domicile population is the highest. The lowest unemployment rate of domicile and refugee population is in Belgrade, which justifies underlined concentration of refugees in Belgrade (Opa?i? 2007). In 1996, the share of pensioners in refugee population was relatively low, which could be explained by difficulties that refugees faced in achieving the rights from working status in former republics of Yugoslavia.

Procurement of long-term accommodation represents the main problem of refugees, because if it is not solved, it causes a series of new problems as a chain reaction: lack of financial sources for food, medicines, hygiene, education, clothes etc, considering the fact that the basic earned income is spent on accommodation. The majority of refugees in the first weeks and months, sometimes even years of refuge, stayed with their friends and relatives. With the extension of refugee crisis, the share of refugees accommodated at their relatives or friends decreased, while the number of those living in private accommodation (rented house or flat) increased, which made this problem financially more noticeable.

The basic characteristic of accommodation during several years after coming to Serbia was the domination of staying with friends and relatives, 55.2%. An especially high share of refugees who found this kind of accommodation was in Belgrade. In 1996, 4.7% of well-off persons solved their housing problems, while one fifth of refugees lived in rented apartments. Every tenth refugee, total 54,409 refugees, lived in collective centres, which were scarce in Belgrade in those days. Significant category (9.2%) comprised refugees who lived in spaces that were not meant for inhabiting, like attics, laundries, etc. After 1996, there is a tendency of decreasing the share of collective centres and accommodation at relatives, while more frequent types are their own or rented accommodation. This trend points to the slow process of refugee integration in local environment, while certain family members remain in the status of refugees. Donations in the form of flats, construction materials and similar become more and more important. Selling real estates in home countries allowed many families to buy land and build their own houses. This led to decreasing the pressure in collective centres which contributed to their slow closing. On the other hand, refugees' families who are forced to rent flats are under growing financial pressure.

The structure of the refugee accommodation as an indicator of socio-economic status did not significantly change after 2001. In 2005 refugees mostly lived in rented accommodation (45%). 28.6% of refugees lived with their friends and relatives and 19% in their own accommodation. If we compare these data to the data from 2001, we can notice that they are similar regarding proportional share of certain types of accommodation. Refugees from Croatia had a bit higher share of their own accommodation. Even though the share of refugees with their own accommodation in the period from 2001 to 2005 slightly changed, the decrease in absolute sum is drastic: from 67,143 refugees (2001) to 19,791 (2009). This means that the majority of refugees in rented accommodation lost the status after 2001. There is also a frequent case that in a family which got citizenship and has its own accommodation, one or two members, usually older, maintain the status of refugees. Refugees from Bosnia more often lived with their friends or relatives, especially in the districts of Ma?vanski region (Loznica, Sabac, Mali Zvornik) and Belgrade districts. Although the number of refugees in collective centres decreased, the share of collective centres as a type of accommodation remains relatively significant: 4.2% (2005), compared to 5.6% (2001). The majority of refugees in collective centres come from Croatia. At the same time, proportional share of refugees who were accommodated in social institutions increased from 0.6% in 2001 to 2.0% in 2005.

Local integration

The majority of refugees preferred integration as their choice of permanent solution during the first years after coming to Serbia. This choice was influenced by years spent in refuge, where after the initial period of adaptation, mistrust and social tensions between refugees and domicile population, ties with local community were built. Refugees in a new environment as well as other immigrants spend their first year adapting and rebuilding social and family networks, which was shown in other studies (Fawcett 1989, Gurak and Caces 1992, Küün 2008, Portes 1995, Godin 2008, Hyvönen 2008). As Godin (2008) says in his work, most found their first job during this time which does not correspond with their premigration socio-occupational ambitions, thus shortening the period of the first job and increasing the risk of transition to another job. Later, Godin continues, after they assimilated the formal and informal policies of the host work world, they could seek out the labour market to find a job consistent with their pre-migration socio-occupational ambitions. Several researchers studied the impact of labor market conditions on their employment activity (Miller 1986, McDonald and Worswick 1998, Åslund and Rooth 2003, Aydemir 2003, Aydemir and Skuterud 2005, Godin 2008) and they showed that it played only a weak role in recent immigrants' initial employment. Many refugees from Croatia and Bosnia found employment in the grey economy in Serbia.

At the time of the first census in 1996, 60% of registered refugees decided to stay in the then existing FR Yugoslavia, while 20% were indecisive about permanent solution. The second census in 2001 showed that 61% of refugees from Croatia and 60% from Bosnia were for integration as a permanent solution. According to the data from research in 2006, refugees with lower education, the elderly and women were more interested in integration. Interesting data is that among persons aged between 15 and 39, over 40% wanted to go to the third countries instead of choosing integration or return to home countries. 41% of unemployed refugees preferred the same thing, which can be explained by worse financial situation of the refugee population compared to the domicile population.

At the beginning of the 1990s the focus was on the issues of accommodation, feeding and employment of refugees, i.e. issues related to existence. The second half of the 1990s was dominated by issues that related to return, permanent providing for refugees and their integration into society. Since 1992 Serbia and international community have helped refugees by accommodating them in collective centres and by starting housing projects. As a result of the plan of gradual closing down, the number of collective centres rapidly decreased after 1996. In 1996 in Serbia there were 700 collective centres, and in January 2002, only 388 collective centres where 26,863 persons were accommodated. The majority of people in collective centres had the status of refugees (64.83% or 17,415 refugees), and the rest were internally displaced persons from Kosovo and Metohia. After 2003 there was a rapid decrease of the number of collective centres as well as refugees in them. On 1 January, 2010 there were 60 collective centres with total 4,867 persons (1,099 refugees and 3,768 internally-displaced persons) in the territory of Serbia. The largest number of collective centres is in Central Serbia and in Voivodina (43 with 4,219 persons). In 2010 there are only two collective centres in the territory of Voivodina: in the district of Kovin, near Deliblato settlement in the town of Pan?evo, with about 270 refugees and displaced persons. In the territory of Belgrade there are seven collective centres, among which the biggest one is Krnja?a in the district of Palilula (400 refugees and displaced persons). In the territory of Kosovo and Metohia there are 17 collective centres with 657 persons. Besides the official collective centres there are also 52 informal ones (unofficial collective centres), i.e. objects with unknown owners or whose owners did not sign contracts with UNHCR) where 2,000 people are accommodated. Only in Kraljevo there are 5 unofficial collective centres. Besides collective centres in Serbia there are also 150 specialized institutions which accommodate refugees from Bosnia and Croatia and internally-displaced persons from Kosovo and Metohia.

The processes of repatriation and integration of refugees in Serbia in 2010 have not finished yet. There are numerous problems, mostly on a smaller scale then they were in the second half of the 1990s. 1,099 persons have remained in collective centres. They are in a difficult financial situation, with no real chance to solve their problems. They comprise only 1.28% of the total refugee population in Serbia. If we add to this number the sum of 155,000 refugees granted Serbian citizenship in the period 1996-2009 which did not improve their economic status, then the situation of refugee population in Serbia is clear. Problems of refugees cannot be solved in humanitarian way, because it is of partial and short-term character, but in the frame of overall development and social policy in Serbia. The importance of the issue of integrating refugees in Serbia comes from the fact that the majority of this population do not want to return to Croatia and Bosnia. In a way, this forces the Serbian authorities to help them in solving property issues and other demands they have in former domicile abodes (La?evi? 2004). Their children are an important reason why refugees do not want to return to their home countries. The refugee population aged up to 20 comprises about 100,000 persons. Most of them do not even remember homeland, one part was born in Serbia and is completely assimilated in the community - they regard Serbia as their home country.

When collective centres were closed the refugees were offered different options: return to their former place of residence, accommodation in finished housings, buying up village households, help with construction materials for persons who have already started building their own houses, providing socially affected refugees and displaced persons with items needed in their social housing, encouraging families who already have had certain possibilities to become independent and leave collective centres. This implies provision of commodity - financial aid depending on the size of a family and moving to other collective centres if none of the above mentioned options were accepted.

Many of those still living in collective centres are sick, old or have other specific needs, so finding permanent solutions for them can be difficult and expensive. Considering the fact that the process of voluntary repatriation to Bosnia is mostly finished as well as the process of moving to the third countries, local integration remains the most sustainable solution for the rest of refugees from Bosnia. The same relates to the refugees from Croatia, even though UNHCR is convinced that some would return to their home countries if there was more certainty considering their legal rights to pensions, tenancy rights, property rights or compensation. Some refugees are not ready for naturalization until those issues have been solved, because they are afraid of their rights being ignored in Croatia if they lose the status of refugees. There is an estimated 20,000 refugee families lost their flats in Croatia. The issue of tenancy rights strongly influence the carrying out the permanent solutions, because regaining their former rights would enable many refugees to solve their problems without the help of the state. UNHCR supports solving this problem, suggesting that the best way is to set up a fair mechanism of compensation. Authorities on national and local level have put in a great effort to ease social and economic integration of refugees in Serbia. Inadequate living conditions, unemployment, inaccessibility of university education and lack of personal documents are the main problems that refugees face as well as those who accepted Serbian citizenship. In addition to this, there are about 1,300 endangered refugees who live in inadequate collective centres. Out of this number, 93% want to integrate in Serbia. One of the obstacles of legal integration is a limited capacity of the Serbian Ministry of Internal Affairs to process applications for Serbian citizenship and deregister those who gained citizenship. This led to significant delays in dealing with applications for citizenship and out-of-date statistical data.

Certain housing projects were realized in Serbia by 2001. The program of permanent solutions, i.e. integration of refugees began in 2002. In the period from 1997 to 2004 CRRS realized certain housing programs in co-operation with UNHCR, Swiss Agency for Development and Cooperation, Norwegian Refugee Council and local authorities. Since 2004 CRRS, in cooperation with The European Agency for Reconstruction, has been working intensively in order to realize the refugee programs from CARDS programs. As a result of those activities we can estimate that upon finishing CARDS programs, nearly 30,000 persons were provided for. From the budget of Serbia and in cooperation with international organizations, above all, UNHCR, the European Agency for Reconstruction, as well as governments of certain countries and with local authorities, CRRS conducts programs of permanent solutions for refugees through the following projects: complete construction, partial construction and self-construction of housing units, donation of pre-assembled houses, aid in the form of construction materials for finishing the already started housing units, building of objects for social living in protected conditions, adaptation and repurposing of collective centres into institutions for older persons, adaptation and extending the capacities of institutions for social protection. As many European countries have great regional variations in their housing capacity and price of housing, the effective utilisation of housing capacity is often given as an argument for placing immigrants outside the urban areas (Andersson 2003, Robinson 2003, Mulder and Krahn 2005, Cvetkovi? 2009). The Swedish experience shows that immigrants, who were initially strategically spread over the whole country, gravitated towards urban areas after some time (Ekberg 1993, Popoola 2002). In cities there are more job opportunities and at least a minimum concentration of immigrants from any given culture (Cvetkovi? 2009). As is the case for other immigrants in Europe, providing a roof over people's head is the main precondition for successful integration. In 2009 1,358 families were provided with housing, or helped economically or in some other way. Part of the refugee population sold or exchanged their previous property. In that way they got resources that enabled them to buy land, build houses, i.e. secure integration to a certain extent. In Belgrade the largest building occurred in the districts of Zemun, New Belgrade, Sur?in and Palilula. New settlements were set up, such as Busije, Grmovac, etc. A typical example of a refugee settlement is Busije. In 1997 the district of Zemun sold land for affordable prices to refugees from Croatia and Bosnia. The settlement emerged on the land which was not intended for urban construction. In 2009 this settlement had 5,000 inhabitants; most of them (80%) are refugees from Croatia and Bosnia (meanwhile the majority of inhabitants got Serbian identification cards). Another example of a refugee settlement is Grmovac. The settlement was founded in the end of 1996 when local authorities of the district of Zemun sold 2,800 plots of land where 600 homes emerged. This settlement is mostly inhabited by refugees from Bosnia and Croatia. It is impossible to find exact data stating how many refugees solved their housing issues in Serbia. The Statistical Office of the Republic of Serbia does not keep special records on refugees; it keeps records on those who got citizenship as it does for all other citizens. Even if we knew the official number of flats and houses that refugees own, that number would not describe the real situation, because statistics do not track data on illegal construction, which is especially in Belgrade Metropolitan Area quite massive.

Besides housing issues refugees also face problems of unemployment, property restitution etc. The economic situation in Serbia makes economic integration of refugees very difficult. General unemployment increases competition and decreases labour price, which also causes sporadic intolerance between refugee and domicile populations. A situation like this makes the most affected categories of refugee and domicile populations even poorer. In the following year the socioeconomic integration of refugees in Serbia should contain key actions to finally close the "story of refugee" in Serbia. Even after a long series of years the issue of refugees in Serbia is not solved. The fact is that the number of refugees decreases, but there are a relatively large number of people who did not reach permanent solutions. Though all necessary preconditions for the integration of refugees are fulfilled, the most important component of this process is housing issue which demands extremely substantial financial funds. A significant number of refugees still intend to retain their status of refugees, even though they were offered facilitated procedures for applying for citizenship. Some of the main reasons for making this decision are: Serbian citizenship does not solve social problems of refugees, and keeping the status of refugees gives them a hope that in the future there will be help and facilities in dealing with housing and all other problems; internationally recognized status of refugees offers 'security' and distinguishes them from other citizens by underlining their affliction; a small number of older refugees who are accommodated in collective centres fear losing their right to be accommodated in such centres if they become Serbian citizens.

Conclusion

The refugee population from Croatia and Bosnia moved to Serbia during the 1990s in waves. The first bigger wave came just after the outburst of the first armed conflicts in 1991 and 1992. In 1993 and 1994 it calmed down. During 1995 there was the most massive process of immigration, and it continued even after 1996. Serbia is the first country in Europe in the number of refugees, which caused UNHCR to classify Serbia among five countries of 'long-term refugees' crisis. According to the UNHCR standards, one country has a refugee crisis if it has more than 25,000 refugees living in it for more than five years. In Serbian case, the number of refugees in much higher and the crisis lasts longer. In the period from 1996 to 2009, the efforts of Serbian state and international community for the survival and better life standards of refugees and domicile population led to constant decrease in the number of refugees. The most complex and the most thorough process was the process of local integration of refugees. Even though preconditions for integration were provided, the most important segment of integration - housing issue, demands significant financial funds (Petrovi? 1997, Stevanovi? 2005).

Two decades of experience with the phenomenon of refugees in Serbia have shown that dealing with the refugee problem needs consensus and political will. The seriousness and complexity of the problem exceeded the existing resources of the state institutions in Serbia. It required an engagement of international community, international organizations, non-governmental organizations, refugee associations etc.

Acknowledgments

The paper resulted from work on the projects financially supported by the Ministry of Science and Environmental Protection of the Republic of Serbia between 2011 and 2015.

References

References

Ambroso, G. (2006) The Balkans at a crossroad: Progress and challenges in finding durable solutions for refugees and displaced persons from the wars in the former Yugoslavia. (New Issues in Refugee Research. Research Paper, 133.) Geneva: United Nations High Commissioner for Refugees.

Andersson, Roger (2003) Settlement dispersal of immigrants and refugees in Europe: policy and outcomes. (Working Paper, 03-08.) Vancouver: RIIM.

Åslund, Olof and Dan Olof Rooth (2003) Do when and where matter? Initial labour market conditions and immigrant earning. (IFAU Working Paper.) IFAU, January.

Aydemir, Abdurrahman (2003) Effects of business cycles on the labor market assimilation of immigrants. (Analytical Studies Research Paper Series.) Ottawa: Statistics Canada.

Aydemir, Abdurrahman and Mikal Skuterud (2005) "Explaining the deteriorating entry earnings of Canada's immigrant cohorts: 1966-2000". Canadian Journal of Economics 38, 2, 641-671.

Black, Richard (2002) "Conceptions of 'home' and the political geography of refugee repatriation: between assumption and contested reality in Bosnia-Herzegovina". Applied Geography 22, 123-138.

Black, Richard and Saskia Gent (2006) "Sustainable return in the post-conflict context". International Migration 44, 3, 15-38.

Black, Richard and Saskia Gent (2004) Defining, measuring and influencing sustainable return: the case of the Balkans. (Working Paper, 7.) Sussex: University of Sussex, Development Research Centre on Migration, Globalization and Poverty.

Black, Richard, Marita Eastmond and Saskia Gent (2006) "Introduction: sustainable return in the Balkans: beyond property restitution and policy". International Migration 44, 3, 5-13.

Cvetkovi?, Vladimir (1998) "Refugees in Serbia and Montenegro: Causes, Status, Structure, Consequences". In the labyrinth of crisis - preconditions for a democratic transformation in FR Yugoslavia, 465-490. Belgrade: Institute for European Studies.

Cvetkovic, Anita (2009) "The integration of immigrants in northern Sweden: a case study of the municipality of Stromsund". International Migration 47, 1, 101-130.

Eastmond, Marita (2006) "Transnational returns and reconstruction in post-war Bosnia and Herzegovina". International Migration 44, 3, 141-164.

Ekberg, Jan (1993) The geographical and socio-economic mobility of immigrants. (In Swedish) Stockholm: Fritzes.

Fawcett, J. T. (1989) "Network, linkages and migration systems". International Migration Review 23, 3, 671-680.

Gre?i?, Vladimir (1998) "Return of refugees and displaced persons in light of the Dayton accords". In Refugees - international protection, legal status and life problems of refugees, exiled and displaced persons, "International Politics", Sluzbeni list SRJ, 91-110. Belgrade: Faculty of Law, Faculty of Political Science, Institute for International Politics and Industry.

Godin, Jean Francois. (2008) "Immigrant economic integration: a prospective analysis over ten years of settlement". International Migration 46, 2, 135-165.

Gurak, Douglas T. and F. Caces (1992) "Migration networks and the shaping of migration systems". InInternational migration systems: a global approach, 150-176. M. M. Kritz, L. Lean Lim and H. Zlotnik, eds. Oxford: Clarendon Press.

Hammond, L. (1999) "Examining the discourse of repatriation: towards a more proactive theory of return migration". In The end of the refugee cycle? Refugee repatriation and reconstruction, 227-244. R. Black and K. Koser, eds. Oxford: Berghahn.

Harvey, J. (2006) "Return dynamics in Bosnia and Croatia: a comparative analysis". International Migration 44, 3, 89-112.

Hyvönen, Heli (2008) "The strength of native ties: social networks of Finnish immigrants in Estonia". Trames 12, 4, 421-440.

Jakovljevi?, Vladimir (1998) Refugees - international protection, legal status and life problems of refugees, exiled and displaced persons. Belgrade: International Politics.

Knezevi?, J. eds. (1992) Refugees in Serbia. Belgrade: CRRS and Red Cross of Serbia.

Kondi?, K. (1997) "Children in war and exile". In Psychology of Refuge, 37-40, Belgrade: Publishing House Nauka.

Küün, Elvira (2008) "The ethnic and linguistic identity of Russian-speaking young people in Estonia". Trames 12, 2, 183-203.

La?evi?, Petar and V. Stankovi? (2004) Refugees' corpus in Serbia, according to the data of 2002 Census. The Ministry of Human and Minority Rights of Serbia and Montenegro.

Luki?, Vesna and Dragana Matijevi? (2006) "Districts in Voivodina with the highest share of refugees - influence on dynamics and structural characteristics of population". Collection of Papers of Matica Srpska for Social Sciences 121, 103-110.

Luki?, Vesna and Vladimir Nikitovi? (2004) "Refugees from Bosnia and Herzegovina in Serbia: a study of refugee selectivity". International Migration 42, 4, 85-110.

Martin, Susan (2002) "Best practice options: Yugoslavia". International Migration 40, 3, 59-69.

Matkovic, Gordana (1999) "Refugees and other war- affected persons". In Population development in Serbia 1991-1997. M. Rasevi?, ed. Belgrade: Demographic Research Centre of Institute of Social Sciences.

McDonald, J. and C. Worswick (1998) "The earnings of immigrant men in Canada: job tenure, cohort, and macroeconomic conditions". Industrial and Labour Relations Review 51, 3, 465- 482.

Miller, Paul W. (1986) "Immigrant unemployment in the first year of Australian labour market activity". Economic Record 62, 176, 82-87.

Milosavljevi?, Milosav (1997) "Social aspects of exile". In Psychology of Refuge, 9-20, Belgrade: Publishing House Nauka.

Miriam, S., J. Anderson, M. Beiser, E. Mwakarimba, A. Neufeld, L. Simich, and D. Spitzer (2008) "Multicultural meanings of social support among immigrants and refugees". International Migration 46, 3, 123-159.

Mulder, Marlene and Harvey Krahn (2005) Individual and community level determinants of support for immigration and cultural diversity in Canada. (Working Paper, 01-05.) Vancouver: PCERII.

Nikitovi?, Vladimir and Vesna Luki? (2009) "Could refugees have a significant impact on the future demographic change of Serbia?" International Migration 48, 1, 106-128.

Nikitovi?, Vladimir (2005) "Could forced immigrants be of significant impact on future demographic change of Serbia?" International scientific conference Migrations, Crises and Recent Conflicts in the Balkans, 27-29 October. Belgrade.

Opa?i?, Goran (2007) Soci-economic status of refugees and displaced persons and their position on the labour market. Novi Sad: Humanitarian Centre of Novi Sad.

Petrovi?, D. (1997) "Exile and the elderly - adaptation at an advanced age". In Psychology of Refuge, 49-55. Belgrade: Publishing House Nauka.

Popoola, M. (2002) "Integration, a reflection of our times". Ajour 8, 47-65.

Portes, Alejandro (1995) "Economic sociology and the sociology of immigration: a conceptual overview". In The economic sociology of immigration: essays on networks, ethnicity and entrepreneurship, 1-41. A. Portes, ed. New York: Russell Sage Foundation.

Robinson, Vaughan (2003) "Dispersal policies in the UK". In Spreading the 'burden'? A review of policies to disperse asylum seekers and refugees. V. Robinson, R. Andersson and S. Musterd, eds. Bristol: Policy Press.

Stevanovi?, Radoslav (2005) "Refugees and demographic growth of population in Serbia". Population 44, 1-4, 43-60.