Demographic Change in the Balkans Since the End of the 1980s {Population, 4-5, 2000) In: Population, 13e année, n°2, 2001 pp. 49-70. Citer ce document / Cite this document : Demographic Change in the Balkans Since the End of the 1980s {Population, 4-5, 2000). In: Population, 13e année, n°2, 2001 pp. 49-70. http://www.persee.fr/web/revues/home/prescript/article/pop_0032-4663_2001_hos_13_2_7190 Abstract Sardon Jean-Paul.-Demographic Change in the Balkans Since the End of the 1980s Throughout Eastern Europe, the 90s witnessed the collapse of the communist regimes and the far- reaching economic and social changes this set in train, and in the Balkans the decade was marked by a series of wars resulting from the breakup of the former Yugoslavia. As well as heavy loss of life, these conflicts have caused massive movements of population, both voluntary and forced, the effects of which are still being felt. Refugees from the former Yugoslavia numbered 917,000 at the end of 1999. The deterioration in living standards consequent on the transition to the market economy has caused large numbers of people to migrate from Bulgaria and, especially, Albania, where 16% of the population left the country between 1990 and 1998. Until 1997, it was also responsible for a stagnation and even fall in life expectancy at birth in all countries of the region with the exception of Croatia and Slovenia. Since then, however, rapid improvements have occurred and the level of longevity observed in 1990 has been exceeded. At the same time, the indicators of nuptiality and fertility have registered a sharp fall, though to a somewhat lesser degree in the former Yugoslavia. In the light of the changes described, the specific features of the Balkans as regards demographic behaviour are not easy to identify. The impact on these countries of longstanding membership of the socialist bloc and its sudden demise, for the moment masks any other differences, at least at the national level. While the specific characteristic of the Balkans lies in the diversity of regional situations originating in the heterogeneity of the population, this has been singularly reduced as a result of the operations of "ethnic cleansing" carried out in the former Yugoslavia over the last ten years. Resumen Sardon Jean-Paul.-La evolución demográfica de los Balcanes desde el final de la década de los ochenta La década de los noventa, caracterizada en toda la Europa del Este por la caida de los regímenes comunistas y las consecuentes transformaciones económicas y sociales profundas que la acompaňaron, también estuvo marcada en los Balcanes por la sucesión de guerras provocadas por la fragmentación de la antigua Yugoslavia. Además de un elevado numero de muertes, estas guerras causaron desplaza- mientos masivos de la población, voluntarios о forzados; los efectos de taies desplazamientos todavia son visibles en la actualidad. A finales de 1999, el numero de refugiados procedentes de la antigua Yugoslavia ascendía a 917,000. La degradación de las condiciones de vida causada por la transition hacia una economía de mer- cado résulté en numerosas salidas de Bulgaria y especialmente de Albania, donde el 1 6% de la población emigró entre 1990 y 1998. También résulté en un estancamiento o, en ciertos casos, disminución, de la esperanza de vida al nacer en todos los paises de la región, excepto Croacia y Eslovenia, hasta 1997. A partir de 1997 se han observado progresos importantes y se ha superado la esperanza de vida observada en 1990. Al mismo tiempo, tanto la nupcialidad como la fecundidad han disminuido fuerte- mente en toda la región de los Balcanes a excepción de la antigua Yugoslavia, donde la disminución es moderada. A la luz de las evoluciones descritas, hallar caracteristicas especificas de la región balcánica no es una tarea simple. La fuerte influencia de la larga etapa socialista y de su desaparición brutal anula, de momento, cualquier otra distinción, al menos a nivel nacionál. La especificidad del mundo balcánico se halla sin duda en la diversidad regional producida por la heterogeneidad de la población, aunque tal diversidad se ha reducido significativamente debido a las operaciones de «limpieza étnica» que afectan a la antigua Yugoslavia desde hace quince aňos. Résumé Sardon Jean-Paul.-L'évolution démographique des Balkans depuis la fin de la décennie 1980 La décennie 1990, celle de la chute, comme dans toute l'Europe de l'Est, des régimes communistes et des profondes transformations économiques et sociales qui l'ont accompagnée, a été marquée dans les Balkans par la succession de guerres qui ont résulté de l'éclatement de l'ancienne Yougoslavie. Outre de lourdes pertes en vies humaines, ces guerres ont entraîné des déplacements massifs de populations, volontaires ou forcés, dont les effets sont toujours visibles aujourd'hui. Fin 1999, on compte encore 917 000 réfugiés issus de l'ancienne Yougoslavie. La dégradation des conditions de vie engendrée par la transition vers l'économie de marché a provoqué de nombreux départs en Bulgarie et surtout en Albanie, où 16 % de la population a quitté le pays entre 1990 et 1998. Elle est aussi à l'origine d'une stagnation, voire d'une régression de l'espérance de vie à la naissance dans tous les pays de la région, à l'exception de la Croatie et de la Slovénie, jusque vers 1997. Depuis cette date, des progrès rapides ont toutefois permis de dépasser le niveau de la longévité observé en 1990. Dans le même temps, les indicateurs de nuptialité et de fécondité ont fortement baissé, bien que de façon un peu plus modérée dans l'ancienne Yougoslavie. À la lumière des évolutions décrites, il n'est pas aisé de dégager des spécificités du monde balkanique en matière de comportement démographique. L'influence de la longue appartenance de ces pays au monde socialiste et de sa disparition brutale est telle qu'elle écrase, pour l'instant, toutes les autres distinctions, au niveau national du moins. Si la spécificité du monde balkanique réside sans doute dans la diversité des situations d'une région à l'autre sous l'effet de l'hétérogénéité du peuplement, celle-ci recule singulièrement du fait des opérations de «nettoyage ethnique» qui ont touché le territoire de l'ancienne Yougoslavie depuis une dizaine d'années. Demographic Change in the Balkans Since the End of the 1980s Jean-Paul SARDON* In the Balkan countries belonging to the Communist bloc, the fall of the Berlin wall had considerable demographic as well as other repercussions. The former Yugoslavia was in a particular position in that it did not really belong to this bloc, but the Yugoslav Federation also broke up at the time and entered upon a series of internal wars. With the exception of migration, the demo graphic impact of these wars has apparently been less acute, howe ver, than the changes that have occurred in the other former Communist countries. Jean-Paul Sardon analyses here these developments in the Balkans and investigates whether "demo graphic unity" exists in the region. The second conference^1) on demography in the Balkans was held in Sarajevo from 10 to 13 May 2000. This conference on "Changes in the 1990s and the Demographic Future of the Balkans" was attended by demog raphers, economists and statisticians to discuss population dynamics and the socio-economic context of the 1990s. Following its findings it seems appropriate to sum up the changes the region has undergone since the col lapse of the Communist system at the end of the 1990s. In the 1990s, the region was deeply affected by the fall of the Comm unist regimes and important economic and social changes which occur- * Institut national d'études démographiques. ** Translated by Paul Belle. ("This conference on the demography of the Balkans in the second half of the 20th cen tury, like the first one, was organized upon the initiative of the Balkan Demographic Network (Demobalk), the international association of French-speaking demographers (AIDELF) and the Association of Balkan Statisticians (ABS), with the financial support of Cesd-Communautaire, the European Free Trade Association (EFTA) and Eurostat. The first conference, "Demographic Developments in the Balkans in the Second Half of the 20th Century", was held in June 1996 in Thessalonica. The proceedings featured in three publ ications: B. Kotzamanis, A. Parant and J.-P. Sardon (eds.), La démographie des Balkans. Sources des données, Aidelf, 1998; La démographie des Balkans. Mouvements migratoires et répartition spatiale de la population, "Series on Transition in the Balkans", University of Thessalonica, Volos, Greece, 2000; La démographie des Balkans. Composantes de Г évolution, "Series on Trans ition in the Balkans" , University of Thessalonica, Volos, Greece, 2000. Population: An English Selection, 13 (2), 2001 , 49-70 50 J.-P. Sardon red at the same time(2^, as well as the succession of wars which resulted from the break-up of the former Yugoslavia. This context had a heavy im pact on demographic changes in the region, the effects of which are still felt today. In addition, the wars in Slovenia, Croatia, Bosnia, and more re cently the operations in Kosovo not only caused huge numbers of deaths but also massive movements of populations, both voluntary and forced. The wars have not been the only reasons behind the huge migrations of re cent decades: the considerable deterioration in conditions and the unrest that some countries experienced during this period of transition towards a market economy led to a deep economic crisis characterized by high unemployment and inflation, a deterioration of health and social services and a fall in living standards. This resulted in a sharp decline in nuptiality and fertility, and a significant increase in mortality. To understand the current situation in the Balkans one has to refer to the characteristics of the Communist regimes and the planned economies which governed the lives of the populations in the region, except for Greece, for half a century. The current demographic situation results from the interaction between a context of brutal interruption of the policies im plemented by fifty years of Communism followed by economic collapse, and war in the case of some of the States of the former Yugoslavia, as well as a much older cultural substrate, the trace of which may be found for example, in the area of early and universal marriage pointed out by Hajnal to the east of an imaginary line running from Trieste to Saint Petersburg. We will examine, each in turn, the changes in the different pheno mena which have marked population dynamics in the region. We will start by the one which characterised a great part of the 1990s in the Balkans, the forced displacements of populations which must definitely be disti nguished from other forms of migration. We will then deal with migrations, internal and external, and then with mortality and fertility, where the chan gesseem to bear the marks of the political, economic and social upheavals affecting the region. Lastly we will seek to determine whether there is a true Balkan identity from a demographic point of view or whether the ap parent unity is only due to the shared history of the last half-century and the Communist heritage, unless this unity stems from the fact that more than half the States which make up the region belonged to the former Yugoslav Federation. (2> Even though in the present Yugoslav Federation maintenance in power of the current group of leaders has deferred any change until its replacement, the new political conditions which are currently emerging in Yugoslavia will undoubtedly have repercussions, but these will not be of the same magnitude as those observed elsewhere in Eastern Europe because of the special as pects of Communist Yugoslavia. Demographic Change in the Balkans Since the End of the 1 980s 51 I. The 1990s in the Balkans 1. The displacements of populations and the refugee problem The displacement of populations has undoubtedly been a major phe nomenon in the Balkans in recent years. Although it has only concerned a small number of countries all belonging to the former Yugoslavia it has also, through the huge numbers of refugees, indirectly and profoundly af fected many more countries, notably during the latest episode in Kosovo. It should be noted, however, before going any further into the study of these recent cases and without seeking to consider such unacceptable practices as commonplace, that population displacements, far from being exceptional, are part of a long tradition in the region. The Balkans, a crossroads and area of contact between several empires and armies, were forced to suffer because of brutal confrontations from many population displacements, like the Great Migration of the Serbs to Vojvodina in 1690. Deportation was also a means of controlling frontiers and a policy in cons tant practice, for reasons of security or population, of the multinational Byzantine and Ottoman empires which successively dominated the region. Among the most recent forced displacements was the arrival of a million and a half Greeks from Asia Minor, repatriated following the Treaty of Lausanne in 1923, of 500,000 Turks deported from Greece at the same time and the deportation of Germans from Romanian Transylvania and from the Serb Banat and Backa regions in 1944-1945; not forgetting the Greeks who were driven out of the Vardar region of Macedonia between 1913 and 1918 and the Macedonian Slavs deported from Greece during the Greek civil war (1943-1949). However, the displacements of populations at the end of the 20th century seem to be of another nature. They appear to be the expression of the ratification by the political elites and the populations of the Western concept of the Nation State, which was a relatively unfamiliar concept in the region as a whole, although exchanges in populations following the war between Greece and Turkey in 1923 constituted a first attempt at ho mogenising a population. The wars that accompanied the break-up of the former Yugoslavia, in Slovenia (June-July 1991), Croatia (June 1991-January 1992 and MayAugust 1995), Bosnia and Herzegovina (March 1992-December 1995) and, more recently, the events in Kosovo (in the spring of 1998-June 1999) provoked hundreds of thousands of deaths (200,000 to 300,000 deaths in Bosnia and Herzegovina representing between 5 and 7% of the population) and resulted in several million people being forced onto the roads as they moved to other parts of their countries or to other countries. Thus, during the air strikes from March to June 1999(3\ close to a million Kosovars, (1)The air strikes started 24 March 1999 and ended 10 June. 52 J.-P. Sardon most of them of Albanian origin, but some of Serb and Romany origins, fled to neighbouring countries or to other regions of Yugoslavia. The numb erof refugees or displaced persons reached somewhere between 2.4 and 2.7 million during the war in Bosnia and Herzegovina, representing 55 to 62% of the country's total population. The current situation described by using the définition employed by the United Nations High Commission for Refugees (UNHCR) which di stinguishes between internally displaced people (IDPs), refugees in other countries and repatriated persons, is presented in Table 1. At the end of 1999, four years after the Daton peace agreement^), there were still 917,000 refugees from the former Yugoslavia. Among the people from Bosnia and Herzegovina, 448,700 were refugees living abroad and 809,500 were internally displaced persons within Bosnia, of which 59% were within the Croat-Muslim Federation. There are 340,400 refugees from Croatia and 234,900 displaced persons in Yugoslavia. The Federal Republic of Yugoslavia^ is still sheltering the largest refugee population in Europe, representing half a million people from Bosnia and Croatia. During the last two years 50,000 more refugees have come in from the Croatian Danube region and tens of thousands of displaced persons arrived from Kosovo in 1999. Table 1.- Estimation of the number of persons concerned by hcr activities at the end of 1999 Country of residence Albania Bosnia & Herzegovina 65,600 20 161,000 809,500 Croatia Refugees 3,900 28,400 Asylum seekers 20 30 Returnees"' 35,500 Displaced 52,400 Displaced returnees"' 73,000 63,600 Various Total 3,920 1,109,120 179,930 Total population 3,695,500(2) 4,527,459(3' at 01/01/2000 3,401,198 О Given assistance no more than 2 years after their return. (2) 1998 (3) 1999. Source: United Nations High Commission for Refugees. Macedonia Slovenia Yugoslavia 21,200 4,400 610 500,700 30 755,500 234,900 21,200 2,012,705<3> 11,300 16,310 1,987,755 168,900 1,660,030 10,613,708<2> Following the return of the Kosovars of Albanian origin, whether r efugees or internally displaced persons, the number of Kosovars of nonAlbanian origin (Serb or Romany) displaced in Serbia or in Montenegro exceeded 220,000, 130,000 of which were displaced after the deployment <4> 21 November 1995. (5> Composed of the Republic of Serbia (Central Serbia, Vojvodina and Kosovo) and of the Republic of Montenegro. Demographic Change in the Balkans Since the End of the 1 980s 53 of KFOR. In October 1999, more than 850,000 Kosovars had returned from Albania, Bosnia and Herzegovina, Macedonia and Montenegro, as well as some 80,000 people who had left Macedonia within the framework of the humanitarian evacuation plan. The population displaced inside Kosovo includes 12,000 people of Serb origin and 8,000 people of Albanian origin as well as 1,650 Romanies. Among the neighbouring countries granting asylum, Albania received over three months close to 500,000 refugees from Kosovo; Macedonia had received 255,000 by midJune 1999 (15,000 in October 1999) and Montenegro 70,000. 2. The migrations International migrations The fall of the Communist regimes, the collapse of the economic market (COMECON) to which these countries belonged, and the crisis brought about by the transition towards a market economy, aggravated by the opening of the borders, have pushed part of the working population to migrate towards countries where it can find employment or towards re gions where the consequences of the crisis can be more easily overcome. Following these upheavals, the contrast between standards of living in Eastern and Western Europe led to fears among many experts that large population transfers would occur: an estimated 1 million people per year for fifteen years. However, this was far from what happened. It is true that emigration towards Western Europe was significant, especially towards Germany, but it did not reach the levels predicted and has significantly d iminished since 1993, notably with the implementation of restrictive poli cies concerning the reception of migrants. In certain countries outside the Balkans (such as the Czech Republic and Hungary) the migration balance is even positive. These migrations are characterised by substantial inter-regional movements which partly correspond to ethnic migrations. For the most part in this region these concerned German minorities (some of which came from Romania), Hungarian minorities (also from Romania), Bulgarian minorities (from the former USSR), Greek minorities (from the shores of the Black Sea and Albania) and Turkish minorities (from Bulgaria). The other migrations result, as we have just seen, from ethnic conflicts and wars which have led to large population displacements in the territory of the former Yugoslavia. The development of these movements inside the new migration area formed by the former Communist countries of Central and Eastern Europe (CEEC), which reaches far beyond the Balkans, is mainly due to the histo rical, economic, cultural and social experiences shared by these countries, even though these displacements partly correspond to transit migrations 54 J.-P. Sardon towards Western European countries. These transit migrations encourage the development of illegal immigration and employment which are spread ing, especially given the fact that in the former Eastern bloc countries the informal economy is blossoming. Albania is the country which has been most affected by emigration: according to estimates by H. van der PoK6), updated by A. Gjonça^7), ap proximately 500,000 people, representing close to 16% of the population, left the territory between 1990 and 1998. 95% of emigrants were aged between 14 and 45 and 71% were men. For Albania, this represented a drain in terms of the country's labour force: this emigration has led to the number of men aged between 20 and 35 being reduced by 50%. Although migration may at one time have been considered as a way of regulating the labour market and of increasing the income of house holds, this migration flow, which so greatly affects the youngest part of the adult population, will undoubtedly have considerable long-term repercus sions, not only concerning future demographic trends within the country but also in terms of its economic and social development. Indeed, the first to leave are the most qualified and well trained people. The scale of these departures comes as a natural result of unprece dentedeconomic problems, political and social instability and increasing insecurity within the country, but it is also undoubtedly a response to the fact that Albania has been totally closed to the outside world for half a century. In Albania, as in most Eastern European countries, the first sign of a return to democratic freedom for many people was the freedom to leave the country. Departures, especially in 1990-1991 and in 1997, often took the form of an exodus towards countries where it was possible to find work (Greece and Italy mainly), even for immigrants without visas. Today close to four in five Albanian emigrants live illegally in their receiving countries. Bulgaria has also been affected by massive migrations. Whereas for several decades only Bulgarians of Turkish origin left (more than 500,000 of them left Bulgaria for Turkey between 1965 and 1995), departures now affect, albeit to a lesser degree, the population as a whole. It was in 1989 that the emigration of Bulgarians of Turkish origin had reached its highest point with the migration of 218,000 people. Between 1990 and 1992 a to tal of 193,000 people left Bulgaria, 65% of whom left for Turkey, and between 1993 and 1995, 174,000 other people emigrated^ <9i Bulgaria <6) Cf. INSTAT, UNFPA, Projeksionet e Popullsise se Shqiperise 1995-2010, Tirana, 1996. (7)A.Gjonça, Albanian Emigration in the 1990s, paper presented at the conference on "Changes in the 1990s and the Demographic Future of the Balkans", 10-13 May 2000, Sarajevo. <8'Cf. V. Zlatanova and S. Zlatanov, Current Trends in Migratory Movements in Bulgaria, paper presented at the conference on "Changes in the 1990s and the Demographic Future of the Balkans", 10-13 May 2000, Sarajevo. *9)It is not easy to get a precise idea of changes in population movements (as with mortali ty rates) since the different publications available differ greatly in this area. The Council of Europe Yearbook reports a migration balance of + 300 in 1989, -470,600 between 1990 and 1992 and +40 between 1993 and 1995. Demographic Change in the Balkans Since the End of the 1 980s 55 has lost more than 710,000 inhabitants since the fall of Communism and the beginning of the transition towards a market economy, which repre sents close to 8% of the country's population of 1988, just before these massive migrations occurred. This emigration, initially linked to ethnic reasons which were soon supplanted by economic motives, has (and above all will have), although to a lesser degree than in Albania, important consequences for the develo pment of the country. During the first three years of the transition the num ber of people employed in scientific research was reduced by half, indicating the importance of the brain drain, even if this was also due to the reduction in funding for research and of the transfer of some resear chersto other sectors of the economy. In 1998 the number of people em ployed in this sector was only a quarter of what it was in 1990. This crisis has especially affected engineering sciences, whereas biology and medi cine have been spared. Although the number of confirmed researchers has dropped by a quarter, it is the younger researchers who are leaving in the greatest numbers: between 1990 and 1998, almost two thirds left. These departures are all the more worrying since potential emigration would still appear to be considerable, judging by surveys conducted among people who travel abroad. In the other countries of the region, population movements were pro portionally smaller; in Romania, for example, the total number of departur es did not reach 300,000 between 1990 and 1998, representing no more than 1% of the population of 1990. Internal migrations In certain countries, especially in Albania and Romania, economic problems have prompted people to migrate inside their country towards re gions where living conditions appear to be less difficult. In Albania there was a dual migration from the country to urban areas and from the provin ces towards the main cities, such as Tirana and Durres, adding the problem of urban overpopulation to that of political and economic transition. The population of Tirana has increased by 30% since 1990. In the same way, in Romania internal migrations which had conside rably declined during the 1980s have substantially increased since, espe cially in the 1990s, due to an easing of restrictions for people moving to the big cities. However, from 1992, internal migrations ceased to be mainly rural-urban migrations, which represented more than half of the overall exchanges, and moved towards a balancing between the different types of migrations to the benefit of rural areas: since 1997 urban-rural migration has been predominant^0). There are several reasons for this re versal: the marked increase in unemployment in urban areas is one, but (l0'Cf. M.-E. Trica, Internal Migration of Romania during 1990-1998, paper presented at the conference on "Changes in the 1990s and the Demographic Future of the Balkans", 1013 May 2000, Sarajevo. 56 J.-P. Sardon there is also the restitution^1 ^ to their rightful owners of the lands brought under State control when the Communists came to power. The reasons given justifying migration have also changed: although family reasons still prevail, work problems are much less often given as reasons than was the case in the first years of the transition. This new dis tribution of the population naturally favours the most dynamic and the richest regions. 3. Mortality In many countries belonging to the Communist bloc, such as Bulgaria, the deterioration of living conditions and the inability of health services to adapt has, since the beginning of the 1960s, led to life expec tancy at birth levelling off and in many cases even decreasing, although this trend is much less marked for females. There is insufficient informat ion to be able to precisely measure the consequences on mortality of the transition towards a market economy. For certain countries (Albania, Bosnia and Macedonia) few life tables are available, especially for the 1990s. Moreover, the values of the indicators often greatly differ from one source to the next (Council of Europe, Eurostat, etc.), which makes the changing trends observed relatively uncertain. However, based on the information available (cf. Tables 2 and 7 of the Annual Demographic Report^12)), it would appear that recent events have undoubtedly made their mark on changing mortality trends, espe cially that of male mortality. Malnutrition has sometimes appeared and use of medical services has reduced. In addition, increased poverty and social tensions, the disappearance of traditional values and increasing crime have fostered higher risk behaviours, limiting improvements in life expectancy. Therefore, even in Slovenia, which has suffered least from these ra dical changes, male life expectancy at birth, which had greatly progressed in the 1970s and 1980s, levelled off between 1990 and 1993 before resu ming its progress in 1994. Overall gains in life expectancy will no doubt be very similar for the last three decades (respectively 2.4, 2.0 and 2.0 years). In spite of such fragmented data it seems that Yugoslavia, within the framework of its current borders, has experienced similar chan ges, although the improvements in life expectancy have been three times lower and have remained stationary since the middle of the 1990s. This situation is the opposite of that in countries such as Romania and Bulgaria which, although spared from war, have seen their life expec tancy definitely reduced; it is true that these two countries had not expe rienced any progress during the 1980s. On the contrary, in Bulgaria male life expectancy fell by 2 years between 1964 and 1994 and by 1.0 years (">By the law 18/1991. 02) In Population: An English Selection, 12, 2000, 293-331. Demographic Change in the Balkans Since the End of the 1980s 57 from 1990 to 1997 (0.33 between 1993 and 1994). In Romania changes have been much more marked: after reducing in the 1980s, life expectancy at birth underwent a slight recovery, subsequently halted by the transition to the market economy which was at the origin of a decline of 1.7 years between 1991 and 1996. Since this date there has been a marked recovery in life expectancy, which has recovered the level it had reached at the end of the 1970s. Elsewhere, as far as can be judged from the fragmented data availa ble, deterioration of conditions of survival seems to have been limited to the year 1991 in Croatia, which witnessed the deadliest military opera tions on its territory that year; however, from 1994 to 1998 the crude mort ality rate increased considerably. In Macedonia, spared by the war, economic problems due to the transition, exacerbated by the sanctions im posed by Greece and by the consequences of the international embargo imposed on Yugoslavia, led to a reduction of one year in life expectancy in 1992, which only returned to its 1991 level in 1997. In Bosnia, mortality increased during the period from 1992 to 1995 because of the war, without our being able to estimate the impact on life expectancy. In the same way, in Albania the deterioration in living conditions and political and social unrest, especially in 1997, which caused many deaths, could only have a negative effect on life expectancy, which apparently dropped by one year between 1994 and 1997. Women seem to have resisted better than men in these difficult ci rcumstances. The maximum reduction of their life expectancy was gene rally more limited than that of the men: 0.7 compared with 1.7 in Romania, 0.6 compared with 1.0 in Macedonia, 0.5 compared with 0.4 in Slovenia. Bulgaria is an exception, the reduction reached 1.1 years for wo men and 1.0 years for men. This greater "resistance" among females which must be linked to their greater longevity, reflects the differences in behaviours between the sexes which was also the reason for the diversity of changes among mortality indicators for men and women during the pre vious decades. In particular, female life expectancy was increasing while male longevity was falling in Bulgaria and Romania. Analysis by age shows that newborn children are relatively spared from increases in mortality. Apart from the exceptional case of Bulgaria where child mortality in 1998 had not yet recovered its 1988 level, and no doubt that of Bosnia, child mortality has very rarely undergone any dete rioration during the transition. This was nevertheless the case in Croatia in 1991-1992, in Romania in 1992-1994 and 1996-1997, in Slovenia in 19901992, in Macedonia in 1992, in Yugoslavia in 1992-1993, and in Albania in 1991-1993. 58 J.-P. Sardon Whereas in Macedonia^13) the deterioration of the population's state of health resulted in an increase in male mortality at all ages, except for early childhood and especially between the ages of 10 and 55, in Bulgaria*44) it was limited to the 35-60-year-old bracket and was due to an increase in deaths caused by diseases of the circulatory system, infectious and parasitic diseases and deaths due to "other" causes. In Romania*45), also at the origin of the decline in life expectancy were circulatory dis eases, associated with tumours. The general impoverishment of the population and budget restric tionsin the public health system during the transition towards the market economy have no doubt contributed to an increased deterioration in the health of the population which had already started in certain countries due to inadequate health services. 4. Couple formation and fertility In the People's Republics of Eastern Europe, including in the Balkans, family policies were especially important in achieving State ob jectives of equal levels of education and employment for men and women whilst maintaining a sufficient fertility rate to ensure abundant labour. Everything was done to enable women to reconcile having a family and being in employment. Although women did not have more children than in the West, they did get married much earlier and had their first child sooner. The fall of the Communist regimes led to an end of policies inciting people to have children and to cutbacks in State aid, resulting in greater problems recon ciling having a family and working, as well as to a questioning of social values which until then had been unanimously accepted, so that people were discouraged from having children very early. It seems that the cur rent general deferment of family formation might have been less marked if there had not been a deep recession caused by the total breakdown of the former economic and social system. In most of the Balkan countries the disappearance of the former reg imes has not led, as in other Eastern European countries, to a profound de cline in nuptiality indicators and a very marked increase in age at first marriage (Figures 1 and 2). Only Bulgaria has followed this pattern: mean age of women at first marriage increased by 1.8 years between 1991 and <13) Cf. M. Mijovska and D. Mitkovska, General characteristics of mortality in the Republ ic of Macedonia, paper presented at the conference on "Changes in the 1990s and the Demog raphic Future of the Balkans" 10-13 May 2000, Sarajevo. (l4>Cf. M. Dimitova and H. Maleshkov, Analysis of Mortality Change in Bulgaria during 1981-1997, paper presented at the conference on "Changes in the 1990s and the Demographic Future of the Balkans" 10-13 May 2000, Sarajevo. (l5)Cf. E. Dinu, Mortality Change in Romania in the 1990s, paper presented at the con ference on "Changes in the 1990s and the Demographic Future of the Balkans" 10-13 May 2000, Sarajevo. Demographic Change in the Balkans Since the End of the 1 980s 59 1998, the total period first marriage rate falling by 42% between 1990 and 1997. This does not mean, however, that the number of unions is reducing in similar proportions, since cohabitation out of wedlock is developing. In Romania, marriage rates are falling but to a lesser degree; however, obser vation of the impact of the transition is hindered by distortion from nuptiality indicators under the effect of an imbalance between the sexes on the marriage market, brought about by the arrival at age of marriage of the large cohorts born in 1967-1968, those of the "people of the decree*46)". The whole of the former Yugoslavia has not followed this pattern, ex cept probably for Bosnia for which we do not have any statistics. The change in the regime did not lead to a reduction in nuptiality, except mo mentarily in Croatia because of the war: the reduction was limited to 1991 and the ensuing recovery in the nuptiality rate enabled it to return to its 1990 level in 1996. In Yugoslavia, relative stability since the beginning of the 1970s in the total period first marriage rate only ended in 1994. In Slovenia, which seems to belong to another cultural area closer to Austria, the fall in the nuptiality rate dates back to 1980 and the events that accom panied the country's independence do not appear to have affected changes in marriage, except for a slight fall in the total rate in 1990-1991. Fertility change during the transition towards democracy and the market economy very much resembles that which we described regarding female nuptiality, with the same distinctions between countries. Actually, it could not be otherwise given the fact that in societies where cohabitat ion and births out of wedlock are not very frequent (this is the case for the whole of the region except for Slovenia) and where fertility is rather low and early, a reduction in the number of marriages usually results very soon afterwards by a reduction in the number of births; moreover, when a de cline in nuptiality is brought about by new problems, this causes couples to defer their plans to have children. Although the total fertility rate has decreased by a third since the fall of Communism in Bulgaria and Romania, the 1990s appear to be similar in this regard to the previous decade in Yugoslavia and Slovenia (Figure 3). It should be noted, however, that in the former Yugoslav territories, it is now Slovenia which has the lowest fertility rate with 1.23 children per woman in 1998, representing a reduction of 0.9 children per woman since 1980; in the region only Bulgaria has a lower fertility rate (1.11). In Croatia the war has, in the same way as it affected nuptiality, led to a reduction in fer tility; this reduction in fertility was made up for between 1993 and 1997 but since then it has declined once again, falling to its 1992 level in 1999. In Yugoslavia, the 1990s were marked by a slight increase in the pace of fertility decline whereas in Macedonia fertility resumed an upward trend until 1994. (|6»This expression is used in Romania to describe people belonging to the cohorts born in 1967-1968 which are twice as numerous because of the sudden prohibition on abortions by the famous Decree 770 of 1996. 60 J.-P. Sardon First marriages for 100 women 1 Г First marriages for 100 women - - Romania .Lithuania /Ц Bulgaria \ \ Latvia Hungary \*~ - Slovakia Щ Щ. ' __ Estonia'A \ - Czech Republic Former GDR 1960 1970 1980 1990 2000 Years of observation First marriages for 100 women med 295 oo 1950 1960 I 1950 1970 1980 1990 2000 Years of observation First marriages for 1 00 women Yugoslavia Macedonia '•"•' '•..•■'• Croatia - \ / : - ■c^ : Azerbaijan Bosnia Slovenia»,' *-, _ - 1960 I 1950 — 1970 1980 1990 2000 Years of observation Armenia ' 1950 1960 1970 1980 1990 2000 Years of observation Figure 1.- Female total period first marriage rate in the Balkans and the former communist countries of Eastern Europe Demographic Change in the Balkans Since the End of the 1980s Mean age at first marriage (years) ined 299 oo 61 26 Mean age at first marriage (years) Lithuania Latvia -» Estonia \ j; - 4^ Slovakia Czech w_ Republic'••*••*„ **■*» 1960 Ii 1950 ••,•*••••. ....• , ', , ''Hungary 'J^LJ L 1970 1980 1990 2000 Years of observation Romania -...•••, ..■•„••" Bulgaria- Л— ■ •'.<■•' ^ 1950 1960 1970 J 1980 L 1990 2000 Years of observation 1950 J 1960 L 1970 1980 1990 2000 Years of observation Mean age at first marriage (years) Slovenia i Yugoslavia у7 Macedonia ; Bosnia Croatia 1950 1960 1970 J 1980 L 1990 2000 Years of observation Figure 2.- Mean age of women at first marriage in the Balkans and the former communist countries of Eastern Europe 62 J.-P. Sardon inec 303 00 Number of children for 100 women : - • Romania 1 350 Number of children for 100 women \ Slovakia 300 i •\ h и 250 r>. '""••„ Poland \ r. A*' '• •-• "TV: """V Hungary*» \ ft / 150 — i Latvia ' I Czech Republic / Bulgaria Estonia // Former GDR 1 \ 1 1I 100 Number of children о г 100 women ined3O5oo 200 250 300 350 1001950 150 1960 1970 1980 Years1990 of observation 2000 Bosnia \ •. Macedonia i '' \ Albania " V ь \ ц л -Л '"■■••. \ !\ »\ >\ i \\ 1 200 .Lithuania: \'; '• .„1950Number of1960 children for1970 1 00 women 1980 Years 1990 of observation ined зоб2000 оо Armenia ■ \ Azerbaijan Moldova . ~ *г Georgia'1. Belarus \ ■ Yugoslavia Ukraine Croatia**Slovenia', ' / • v* ■• - 1960 I 1950 1970 1980 1990 2000 Years of observation 1950 1960 1970 1980 1990 2000 Years of observation Figure 3.- Total fertility rate in the Balkans and the former communist countries of Eastern Europe Demographic Change in the Balkans Since the End of the 1 980s 63 In Croatia and Slovenia recent fertility declines were accompanied by a marked reduction in age at childbirth of approximately 2 years between 1990 and 1998, but in Bulgaria and Romania, countries where the total fertility rate has declined the most, age at childbirth underwent a mo derate increase, as it did in Yugoslavia and in Macedonia (Figure 4). The fertility decline therefore affects women of all ages more or less equally in Bulgaria and Romania, whereas in Croatia and Slovenia, where patterns in family formation and therefore birth timing change considerably, women want to have their children later than before. Globally, although recent events are at the origin of current changes in fertility in Bulgaria and Ro mania, the reproductive behaviour of couples was already changing in Croatia and Slovenia at the time of the break-up of the former Yugoslavia. Whereas the transition has led in a certain number of former Comm unist countries, especially among the Baltic states, to an acceleration in the frequency of births out of wedlock, in the Balkans only Bulgaria has experienced such an acceleration, the proportion of such births having been multiplied by 2.5 between 1990 and 1998 (Figure 5). Yet it is in Slovenia, a country where the increase dates back furthest, that the pro portion of births outside wedlock is the highest (35.4% in 1999), just ahead of Bulgaria (35.1%). In the region, as everywhere in Central and Western Europe, abor tionhas always played a central role in family planning due to generally very liberal legislation, very wide acceptance on the part of the populat ion, and very limited availability of modern contraception. At one time in Bulgaria, Romania and Yugoslavia there were more abortions than live births. Since the beginning of the 1990s, the frequency with which people resort to abortion has declined everywhere except in Albania, where abor tion is not frequent, in Bulgaria and especially in Romania where its libe ralisation immediately after the fall of the Ceaucescu regime resulted in a considerable temporary increase in the number of abortions (315 abortions for 100 births in 1990 and 1991). II. In search of Balkan demographic unity After a brief review of the dominant characteristics of demographic change in the Balkans during the last decade, one is tempted to ask the fo llowing question: Do the Balkans present a special case concerning demog raphic behaviour? It is a difficult question to answer, because there are many statistical methods which make it possible to measure similarities and differences, the degree of resemblance of behaviours, as there are a great many variables which may be taken into account for such an analys is.However, based on the changes described here it is possible to make the following observations. 64 J.-P. Sardon ined зо? oo I Mean age at maternity (years) Q131 Mean age at maternity (years) 1 Г .Slovakia 4 Poland \V Hungary^' / ;'•'' _ Czech Republic Former GDP 1970 1980 1990 2000 Years of observation Mean age at maternity (years) l\ Л чч Albania 1960 1970 1980 1990 2000 Years of observation 31 Mean age at maternity (years) I I med 309 oo 1950 I 1960 I 1950 - Belarus - \ \\ Macedonia If, Georgia Russia «»s ч^ч\»'• "V% Yugoslavia ;» .:'/'' Croatia 4... Slovenia - \ Azerbaijan Ukraine » . ^*> \ % -^4 Moldova V _ %%-хл X Armenia 1980 1990 2000 Years of observation 1950 1960 1970 I 1970 |I 1960 |I 1950 1980 i*i^' 1990 2000 Years of observation Figure 4.- Mean age at childbirth in the Balkans and the former communist countries of Eastern Europe Demographic Change in the Balkans Since the End of the 1980s 1 Proportion (%) 1 Proportion (%) 65 / J— Former GDR / - - 40 — / - 1Czech Republic / »* Slovakia \ — Poland Hungary/ / •''/- Estonia ..--■■' Latvia // V. -—' Bulgaria Л ^-^__ _/ Lithuania 1950 Proportion1960 (%) 1970 J\^~~ 1980 Years »—1990 'Romania of observation 2000 1 1980 Years 1990 of observation 2000 1 1970 1I l1! 1950 Proportion1960 (%) - Slovenia S Georgia / Moldova / Y Russia 1970 1980 Years 1990 of observation 2000 , 1960 I1 1950 Croatia * Ukraine -* '■'"\ : '•■■' / Belarus Armenia.... a' 1950 1960 1970 ^-^1980 Years Azerbaijan 1990 of observation 2000 I Bosnia I( Figure 5.- Proportion of births out of wedlock in the Balkans and the former communist countries of Eastern Europe 66 J.-P. Sardon Regarding nuptiality, Balkan Europe is distinguished by a young age at first marriage, although this is increasing: it is approximately 2324 years for women, which contrasts with the 27-28 years encountered in Western Europe. However, the countries of the region share this particular ity with former Eastern European countries as a whole. It is true that all the States which belonged to the former Communist bloc shared a concern for promoting a certain type of family based on early marriage fostered by a certain number of measures (priority allocation of accommodation to newlyweds, taxation of unmarried people, etc.). Bulgaria is a prime examp le in terms of the application of such a policy, which was indeed suppor ted by the region's long history of early and almost universal marriage. Right up until the end of the 1980s the region, like the other Eastern European countries, distinguished itself by a greater frequency of mar riage; the difference with Western countries, which had increased at the end of the 1970s, has practically disappeared. This reduction in the num berof marriages, which points to a spread in cohabitation out of wedlock, is accompanied by a marked increase in births out of wedlock in some countries, and especially in Bulgaria, which seems to mark an end to a particularity shared by some countries in Eastern and Mediterranean Europe. Currently, only Greece, Croatia and Macedonia, and no doubt Albania and Bosnia and Herzegovina, still have birth rates out of wedlock below 10%, like Switzerland and Italy in the West. The Balkans are not a homogeneous region concerning births out of wedlock and two groups are definitely opposed: Albania, Bosnia and Herzegovina, Croatia and Macedonia, with low fertility out of wedlock, and the other countries (Slovenia, Bulgaria, Romania, and Yugoslavia) where it is rapidly deve loping. This situation, characterised by relatively early marriage, even though age at marriage is increasing rapidly, and few births out of wedl ock, in spite of the postponement of nuptiality, seems to be an indication that traditional practices regulating matrimonial behaviour are being maintained, no doubt bolstered by a certain form of social pressure and the problems of the time. The healthy state of marriage in the Balkans seems to be confirmed by a much lower divorce rate than in Europe, the other former Communist countries included (Figure 6). In the region, the number of broken marria ges only exceeds the 20% threshold in Romania (24%); and only Italy and Poland are, like the Balkan countries (including Greece, Cyprus and Turkey), below this level. However, there is no information available on the countries of the former USSR. Concerning fertility, both the total fertility rate and mean age at childbirth lead to the same observations concerning first marriage rate: childbirths continue to be early, except in Croatia and Slovenia. Mean age at childbirth is 3-4 years less than in Western Europe. Bulgaria is different in that it has the earliest age at childbirth (24.5 years) associated with a Demographic Change in the Balkans Since the End of the 1980s ined 31500 Number of divorces for 100 marriages ř ined3i6 i■ \ 4* f» \ \ ~ ! ' 1 Number of divorces for 100 marriages 67 i - ! #-< Estonia \ V Latvia Lithuania\\/ b- ' \ S" Bulgaria •' Romania 1960 1970 1980 Years1990 of observation 2000 med 317 00 V\ 1960 1970 1980 Years1990 of observation 2000 60 Number of divorces for 100 marriages Ined3i8oo \i I i1 60 Numberof divorces for 100 marriages 1 1950 1 1950 Russiayv - 4-< —л/ _ * J Moldova S ~~ Л Slovenia Albania 1970 1980 Years 1990 of observation 2000 1950 1960 1970 1980 Years1990 of observation 2000 1 1960 |1 1950 Figure 6.- Total divorce rate in the Balkans and the former communist countries of Eastern Europe 68 J.-P. Sardon total fertility rate which is lower (1.11 children per woman). On the other hand, during the last decade, marked by deep political, economic and so cial changes, the decline in fertility in all the Eastern European countries, including the Balkans, has been such that these countries have reached le vels observed elsewhere in Europe. A last characteristic of fertility in the region, which joins our pre vious observations, is that, at equal fertility level, the proportion of births before the age of 20 is always much higher in the Eastern European count ries and especially in the Balkans, the United Kingdom being the only ex ception in the West. Concerning male mortality, the Balkan countries occupy, along with former Eastern bloc countries, a half-way position between Western Euro pean countries and most of the countries of the former USSR. Concerning female life expectancy at birth, the group formed by the Western count ries, joined by Slovenia, the Czech Republic and Armenia, with a life ex pectancy of more than 78 years, contrasts with all the others. The child mortality rate in the Balkans, except for Croatia and above all Greece and Slovenia, is relatively high — especially in Romania — and is close to that of the former Republics of the Soviet Union. It would appear in the light of these observations that it is not easy to point at any specific demographic characteristics concerning the Balkans. As we have seen earlier, the current situation is the result of an interaction between a context of brutal interruption of the policies implemented in a region dominated by planned economies, from which it is impossible to dissociate the upheavals brought about by the fall of the Communist reg imes and by the wars, and an older anthropological and cultural substra tum. In order to understand the situation more clearly it would be necessary to dissociate these two levels. During the 1990s, the Balkan countries experienced changes which were very similar and apparently had more to do with their belonging to the former Communist bloc, apart from Greece, than with their geographic al proximity. Indeed, these changes have occurred everywhere in the for mer Communist countries of Eastern Europe. The problem in distinguishing specific characteristics inherent to the Balkans may be due to the fact that points shared by all the former People's Republics — and in particular those linked to their fall — tend to prevail, for the time being, over all of the other distinctions. It will there foreonly be after the transition period that any particularities concerning the Balkan region will emerge, if indeed they do exist. The apparent unity that sometimes seems to emerge from the region could also be due to the important place occupied within it by countries of the former Yugoslavia (5 countries out of 8, except for Greece). This un doubtedly explains the impression that the region gives of having suffered less from problems due to the transition than other countries. It is true that Demographic Change in the Balkans Since the End of the 1 980s 69 the support owed to the extended family in Yugoslav society makes it easier to resist economic hardship. However, the level of a general indicator for the whole country may hide strong regional differences, as witnessed in the Federal Republic of Yugoslavia. And perhaps this is where one should look for a particularity in the Balkan region, in the diversity of situations from one region to an other resulting from the population heterogeneity. Such heterogeneity has, however, diminished during the last decade due to the "ethnic cleansing" operations which have affected the territory of the former Yugoslavia. Jean-Paul Sardon, Institut national d'études démographiques, 133, boulevard Davout, 75980 PARIS, Cedex 20, tél. : 01 56 06 20 40, fax : 01 56 06 21 99, email : odeurope@francenet.fr 70 J.-P. Sardon Sardon Jean-Paul.-Demographic Change in the Balkans Since the End of the 1980s Throughout Eastern Europe, the 90s witnessed the collapse of the communist regimes and the farreaching economic and social changes this set in train, and in the Balkans the decade was marked by a series of wars resulting from the breakup of the former Yugoslavia. As well as heavy loss of life, these conflicts have caused massive movements of population, both voluntary and forced, the effects of which are still being felt. Refugees from the former Yugoslavia numbered 917,000 at the end of 1999. The deterioration in living standards consequent on the transition to the market economy has cau sed large numbers of people to migrate from Bulgaria and, especially, Albania, where 16% of the po pulation left the country between 1990 and 1998. Until 1997, it was also responsible for a stagnation and even fall in life expectancy at birth in all countries of the region with the exception of Croatia and Slovenia. Since then, however, rapid improvements have occurred and the level of longevity observed in 1990 has been exceeded. At the same time, the indicators of nuptiality and fertility have registered a sharp fall, though to a somewhat lesser degree in the former Yugoslavia. In the light of the changes described, the specific features of the Balkans as regards demographic behaviour are not easy to identify. The impact on these countries of longstanding membership of the socialist bloc and its sudden demise, for the moment masks any other differences, at least at the national level. While the specific characteristic of the Balkans lies in the diversity of regional situations origi nating in the heterogeneity of the population, this has been singularly reduced as a result of the opera tions of "ethnic cleansing" carried out in the former Yugoslavia over the last ten years. Sardon Jean-Paul.-L'évolution démographique des Balkans depuis la fin de la décennie 1980 La décennie 1990, celle de la chute, comme dans toute l'Europe de l'Est, des régimes communistes et des profondes transformations économiques et sociales qui l'ont accompagnée, a été marquée dans les Balkans par la succession de guerres qui ont résulté de l'éclatement de l'ancienne Yougoslavie. Outre de lourdes pertes en vies humaines, ces guerres ont entraîné des déplacements massifs de popul ations, volontaires ou forcés, dont les effets sont toujours visibles aujourd'hui. Fin 1999, on compte encore 917 000 réfugiés issus de l'ancienne Yougoslavie. La dégradation des conditions de vie engendrée par la transition vers l'économie de marché a pro voqué de nombreux départs en Bulgarie et surtout en Albanie, où 16 % de la population a quitté le pays entre 1990 et 1998. Elle est aussi à l'origine d'une stagnation, voire d'une régression de l'espérance de vie à la naissance dans tous les pays de la région, à l'exception de la Croatie et de la Slovénie, jusque vers 1997. Depuis cette date, des progrès rapides ont toutefois permis de dépasser le niveau de la lon gévité observé en 1990. Dans le même temps, les indicateurs de nuptialité et de fécondité ont fortement baissé, bien que de façon un peu plus modérée dans l'ancienne Yougoslavie. À la lumière des évolutions décrites, il n'est pas aisé de dégager des spécificités du monde balka nique en matière de comportement démographique. L'influence de la longue appartenance de ces pays au monde socialiste et de sa disparition brutale est telle qu'elle écrase, pour l'instant, toutes les autres distinctions, au niveau national du moins. Si la spécificité du monde balkanique réside sans doute dans la diversité des situations d'une région à l'autre sous l'effet de l'hétérogénéité du peuplement, celle-ci recule singulièrement du fait des opérations de «nettoyage ethnique» qui ont touché le territoire de l'ancienne Yougoslavie depuis une dizaine d'années. Sardon Jean-Paul.-La evolución demográfica de los Balcanes desde el final de la década de los ochenta La década de los noventa, caracterizada en toda la Europa del Este por la caida de los regímenes comunistas y las consecuentes transformaciones económicas y sociales profundas que la acompaňaron, también estuvo marcada en los Balcanes por la sucesión de guerras provocadas por la fragmentación de la antigua Yugoslavia. Además de un elevado numero de muertes, estas guerras causaron desplazamientos masivos de la población, voluntarios о forzados; los efectos de taies desplazamientos todavia son visibles en la actualidad. A finales de 1999, el numero de refugiados procedentes de la antigua Yu goslavia ascendía a 917,000. La degradación de las condiciones de vida causada por la transition hacia una economía de mercado résulté en numerosas salidas de Bulgaria y especialmente de Albania, donde el 1 6% de la pobla ciónemigró entre 1990 y 1998. También résulté en un estancamiento o, en ciertos casos, disminución, de la esperanza de vida al nacer en todos los paises de la región, excepto Croacia y Eslovenia, hasta 1997. A partir de 1997 se han observado progresos importantes y se ha superado la esperanza de vida observada en 1990. Al mismo tiempo, tanto la nupcialidad como la fecundidad han disminuido fuertemente en toda la región de los Balcanes a excepción de la antigua Yugoslavia, donde la disminución es moderada. A la luz de las evoluciones descritas, hallar caracteristicas especificas de la región balcánica no es una tarea simple. La fuerte influencia de la larga etapa socialista y de su desaparición brutal anula, de momento, cualquier otra distinción, al menos a nivel nacionál. La especificidad del mundo balcánico se halla sin duda en la diversidad regional producida por la heterogeneidad de la población, aunque tal diversidad se ha reducido significativamente debido a las operaciones de «limpieza étnica» que afectan a la antigua Yugoslavia desde hace quince aňos.