DEMOGRAPHIC CHARACTERISTICS OF POPULATION OF SLOVENIAN CITIES IN THE FIRST DECADE OF 21 ST CENTURY

Demographic characteristics of population of Slovenian cities in the first decade of 21 st century We analyzed available statistical data to establish present demographic characteristics of the population of Slovenian cities and their recent changes. According to the statistical definition of cities the share of urban population in Slovenia represents half of its total population. The number of population in Slovenian cities is generally decreasing or stagnating, the coefficient of masculinity was growing in the last decade, but not much more than in Slovenia in general, the ageing of population was also a bit faster in urban areas and their ageing index is considerably higher than the Slovenian average. Urban areas also had a considerably larger share of immigrants from abroad in comparison with the national average while employed and unemployed persons were relatively equally distributed among urban and other areas. The level of education was considerably above the national average. We also compared natural growth data which showed that urban areas in 2002 experienced the decrease that was below national average and the increase in 2011 that was also lower than in Slovenia in general.


Introduction
Population of Slovenian cities and its demographic characteristics have not been a research topic in Slovenian geography for quite some time.There were some studies about Slovenian cities that had to take into consideration some basic demographic data and some that are taking into consideration only Ljubljana or Maribor (Rebernik 2000;2005;Dolenc 2000;Počkaj Horvat 1997;Pak 1994), but Slovenian urban population itself was seldom if ever the main subject of the study at least not since the 1 st Slovenian demographic symposium where Vrišer discussed "the populational development of Slovenian cities" (Vrišer 1974).This paper therefore tends to fill up this gap at least for the period of the first decade of the 21 st century.
The task is not easy at all.The first problem we face is the definition of urban population and we could easily devote the whole paper to this question alone.However, no matter how important this question is, that was not our aim.Therefore we simply used the statistical definition from 2003 (Pavlin 2003) with all of its deficiencies.That also has a practical reason.The statistical office of the Republic of Slovenia publishes the data about urban population aggregated according to the above mentioned definition of so called urban areas.There are 104 such areas with 156 settlements included.Maribor urban area consists of Maribor and 21 adjacent settlements, while Ljubljana urban area encompasses only 3 adjacent settlements.The vast majority of urban areas are in fact single cities and their main characteristic is that they are small and their urban character is dubious.
Tab. 1: Number of settlements in urban areas as defined by Statistical Office of the Republic of Slovenia (Pavlin 2003).Slovenia with roughly 2 million of inhabitants cannot have large cities.Otherwise it would have to have only few cities and extremely high level of urbanization.That is not the case.Slovenia has about 6.000 settlements, most of them with very low numbers of inhabitants and we can state something similar for the urban areas as defined by Statistical Office of the Republic of Slovenia.Even Slovenian capital Ljubljana with a bit more than a quarter of a million inhabitants cannot be considered a big city.Other cities or, to use statistical definition, urban areas are even smaller.Even according to the rank size rule most of them are well above the line of this rule ( Fig. 1).The numbers of inhabitants fall under 50.000already at the 3 rd ranked urban area, under 30.000 at the 5 th and under 20.000 at the 8 th .Only 18 urban areas had more than 10.000 inhabitants in 2011 (1 st January).Most of our analysis includes this group of urban areas as they can undoubtedly be considered as urban while the rest of them are due to low numbers of inhabitants less appropriate for demographic analysis and many of them also lack urban character.In our analysis we used the data of Statistical Office of the Republic of Slovenia (SI-STAT Data Portal, data provided by D. Dolenc).
We expected that the population of Slovenian urban areas is older than the national average and that the major cities are losing population due to suburbanization (emigration of urban population to suburban and rural areas within commuting distance from the central city).We also expected that the population is better educated while considering employment and unemployment that there are considerable differences among more and less economically prosperous urban areas.

Growth of urban population and urban areas
At the beginning of the nineties in the past century the growth of Slovenian cities already stopped (Jakoš 1993).And before that since sixties they had grown mainly because of immigration from other Yugoslav republics.At the beginning of the new millennium the total number of urban population was slightly decreasing while at the same time the number of population in Slovenia moderately grew.
The period 2003-2011 includes a break in the series due to the introduction of the new statistical definition of population, harmonized with the definition of population and migrants in the Regulation on Community Statistics on Migration and International Protection (Methodological explanation of the Statistical Office of the Republic of Slovenia).For data after January 1 st , 2008 the same criteria are applied for citizens of the Republic of Slovenia and for foreigners in preparing statistics on the number of population.The basis of the concept is the so-called usual residence, which in the case of Slovenia includes permanent or temporary residence (one year).The foreigners according to this new definition became inhabitants of Slovenian settlements and most of them lived in the cities.We can therefore observe a sudden leap in growth of urban population in 2008 ( Fig. 2).Population of urban areas continued to grow in the following years, but in 2011 the number was lower again than in 2010.Both largest cities/urban areas were losing population before the break in the series.After the leap in 2008 Ljubljana continued to grow, but less and less from year to year while Maribor grew only one year after 2008 and in 2010 and 2011 was again faced with a slight decrease.
How different the urban areas are is best shown with the fact that in 2008 (break in the series) only 35 of them increased in the number of inhabitants while the rest 69 were faced with a decrease.Nevertheless the total number of urban population in that year went up by almost 50.000due to urban areas with a higher number of foreigners such as Celje, Velenje, Koper, Izola, Novo mesto.Because of its size Ljubljana's number of inhabitants jumped up by more than 14.000.Sežana (13,7 %) had the highest relative change, a city with a lot of construction workers at that time.Among 18 urban areas with more than 10.000 inhabitants in 2011 only 8 had a higher share of total urban population in 2011 than in 2003 and 10 of them lower.Ljubljana increased its share from 25,8 % to 26,7 %.The growth was mainly reserved for urban areas in central Slovenia around Ljubljana, but not at all exclusively, there were also some urban areas with considerable positive change in the number of inhabitants in other parts of Slovenia.On the other side the top loser among the above mentioned 18 urban areas is Murska Sobota, the center of a less developed Slovenian region Prekmurje, followed by Trbovlje, the center of Zasavje, the region economically depressed because of a coal mining decline.
Fig. 3: Changes in the number of inhabitants in the period 2003-2011 in 16 urban areas (break in series because of the new statistical definition of population is marked grey; urban areas with the higher share of total urban population at the end of period -bold, with lower share -italic).

Population of Slovenian urban areas by age and sex
As expected the population of urban areas in average is older than in other settlements.Ageing index went up considerably in the 9 year period 2002-2011 (census years).Urban areas average grew more (for 26,7) than the national one (for 22,4).There are of course important differences among different urban areas.
Considering the 18 largest, 8 of them had ageing index below the national average.Ljubljana's satelite cities Domžale, Kamnik and Vrhnika even below 100.
The worst ratio between the old and the young population in 2002 as well as in 2011 was in Maribor with ageing indexes 141,7 and 174,7 among 18 largest urban areas and in Portorož (169,6 and 228,1) among all urban areas.
We analyzed the distribution of 3 basic age groups of urban population among urban areas by computing locational quotients and coefficients of localization.Coefficient of localization 0 means equal distribution while 1 means total concentration at one single location.In our case the coefficient of localization of the young population of 1 would mean that all 0-14 years old urban residents live in one single urban area while all the other urban areas should have only population older than 14 years.Of course that is not possible and in our case the coefficients of localization for all 3 age groups were close to 0. The age group of the population from 15 to 64 years was the most equally distributed among all urban areas and had coefficients of localization only slightly above 0 (0,01) for both years.Old population on the other hand was the most unevenly distributed, but still with a coefficient of localization 0,06 in 2002 and 0,04 in 2011 which is the same as the values of this coefficient for young population in both years.Locational quotients were computed for every single urban area.The value 1,00 indicates that the share of an age group of certain urban area in all the urban population of that age is equal to the share of the urban area's population in total urban population.Value above 1,00 indicates an above average share and a value below 1,00 a below average share.
In 2002 locational quotients for young population ranged from 0,80 (Spodnje Hoče) to 1,44 (Ribnica) with a standard deviation of 0,10.At the lower extreme there was almost every littoral urban area (Portorož, Lucija, Koper Izola -all below 0,90).Urban areas with high locational quotients were mainly smaller industrial centers, some from the Ljubljana urban region (Brezovica pri Ljubljani, Mengeš) and some from other parts of Slovenia (Logatec, Železniki, Žiri etc.).In 2011 the quotients ranged from 0,69 (Portorož) to 1,36 (Vipava).Beside Portorož which became the urban area with absolutely the lowest locational quotient for young population, two more urban areas from the same municipality were close to the lowest extreme (Piran and Lucija) and 3 others from this group were Bovec, a remote town in the Soča valley in the northwestern Alpine borderland, while Lendava and Radenci are from the opposite side of the country (Pomurje region at the extreme north-east).
High above average share of young population in both years had Brezovica pri Ljubljani and Logatec, while Trzin (also a neighbouring town of Ljubljana) and Naklo as well as Vipava had the highest values in 2011.
For 2011 we classified all urban areas into several groups according to their locational quotients, for all 3 age groups.The first group was the group of urban areas with a relative concentration of young population.Locational quotients for young population of these urban areas were more than one standard deviation above 1,00.It encompassed 11 urban areas and most of them were satelite cities/towns of Ljubljana (Domžale, Grosuplje, Kamnik, Mengeš, Trzin, Vir, Vrhnika).
Even the majority of the rest are urban areas located in the commuting distance from Ljubljana (Novo mesto, Železniki, Žiri, Slovenska Bistrica).
Next is a group of urban areas with relative concentration of young and deconcentration of old population (Brezovica pri Ljubljani, Logatec, Šenčur, Vipava and Zreče).Again the first 3 are located in near vicinity of Ljubljana (less than half an hour driving distance).
Another group worth mentioning is the one with more unfavourable characteristics.It is the group with a relative deconcentration of young population and it consists of 9 urban areas.Bovec, Kranjska Gora, Maribor and Trbovlje all have only considerable a below average share of young population while Lucija/Lucia and Piran/Pirano also have a considerable above average share of population aged 15 to 65. Ilirska Bistrica, Lendava/Lendva and Portorož/Portorose form another subgroup with a considerable above average share of old population.
Relative concentration of old population is significant for Štore and Tolmin with a considerable above average share of old population while Radenci beside that also have a considerable below shares of young and 15 to 64 years aged population.
Relative concentration of 15 to 64 years old population can be observed in 10 settlements, most of them also have a considerable below average share of old population.The largest urban area in this group is Velenje.
Relative deconcentration of 15 to 64 years old population was present in a group of 14 urban areas.Again most of them also had relative concentration of old population at the same time.They were located in all parts of the country and no larger urban area was a part of this group.
The group with moderate concentration or deconcentration consists of 47 urban areas and is further divided into 6 subgroups.Most of the largest urban areas (11) were in this group (Ljubljana, Kranj, Celje etc.).Sex ratio of Slovenian urban population was analyzed by the computation of the coefficient of masculinity (the ratio between the number of men and the number of women multiplied by 1000).In 2002 this coefficient was 925 for urban population and 957 for Slovenian population in general.In the period 2002 to 2011 the coefficient of masculinity grew as age specific mortality for men decreased more in all 5 year age groups than age specific mortality for women (Life expectancy at birth in Slovenia grew by 5.5% for men and only for 3.2% for women).The growth of coefficient of masculinity was higher in urban areas than in general so in 2011 it was 951 in urban areas while national average was 980 men per 1000 women.
As shown on Fig. 5 vast majority of all urban areas experienced the growth of the coefficient of masculinity (80) and also major part of them grew in terms of the number of population (66).The cross at the extreme right of the graph represents a rather small industrial town Šoštanj with 22% growth of the coefficient of masculinity (from 952 to 1159).This must be the consequence of a new statistical definition of population.The number of population in this town with a stagnating number of inhabitants just a bit above 4000 jumped up by 205 between 2007 and 2008.That means that mainly male (industrial) workers of foreign origin became a part of town's population and therefore radically changed sex ratio.This was also the case in some other urban areas with industry that used workers without Slovenian citizenship (Solkan, Trebnje etc.).

Natural increase/decrease of Slovenian urban population
In 2002 Slovenian urban areas had birth and death rates above national average.In that year Slovenia was experiencing a period of negative natural growth (decrease) as death rate was higher than birth rate.Natural decrease at the national level was -0,60 per 1000 while urban areas' natural decrease was -0,28 per 1000 inhabitants.
Over the 9 year period birth rates grew and the decrease was substituted with a natural growth of 1,58 per 1000 at national level and 0,91 per 1000 inhabitants for urban population.The death rate in urban areas also grew from a relatively small value of 8,70 to a relatively high 9,49 while death rate for Slovenia in general went down.
Individual values of natural increase for 18 largest Slovenian urban areas are shown in Fig. 6.Average values for whole Slovenian population and urban population are relatively close together while the values of urban areas are quite dispersed around them.At the lowest positions are Trbovlje and Ptuj, both with considerable natural decrease in both years.Maribor, Izola as well as Koper and Nova Gorica are all in a similar situation only their values are closer to zero.
Jesenice and Murska Sobota are also in an unfavourable position.These two urban areas still had a natural increase in 2002, but not any more in 2011.
On the other side the top position is taken by Kamnik followed by the group of urban areas consisting of regional centers Kranj, Novo mesto, Velenje and Ljubljana's neighbors Domžale, Škofja Loka and Vrhnika.Ljubljana itself and Celje as well are on the other hand urban areas that switched from a natural decrease in 2002 to a natural increase in 2011.Their natural growth was above the Slovenian average.
Fig. 6: Birth rates and death rates for 18 largest Slovenian urban areas in 2011.

Migrational characteristics of Slovenian urban population
Cities and urban areas should normally have more immigrants than non-urban settlements.In 2002 urban areas in total had only a slightly smaller share of nonmigrant population than Slovenia in average.The share of immigrants from other countries was, however considerably above the national average.Considering only the 18 largest urban areas there were some exceptions such as Domžale, Ptuj and Murska Sobota.Littoral cities and mining and industrial center Velenje, as well as Jesenice (city known by its steelwork) all had a lot of inhabitants that immigrated from other countries.
Change from 2002 to 2011 was characterised by the growth of share of foreign immigrants and the decrease of the share of population that lived in the same settlement since birth.Among the 18 largest Trbovlje remained to be the urban area with the highest share of non-migrant population while Domžale and Murska Sobota were urban areas with the lowest share of foreign immigrants.Considering interregional migrations ( Fig. 7) Ljubljana was undoubtedly in "the winning position" while its neighboring urban areas were not having many immigrants from other Slovenian statistical regions.Murska Sobota had the most considerable change from 72 to 116 per 1000 which is still less than half of the Ljubljana's number.Coefficients of localization (Fig. 8) are showing fairly equal dispersion of population that lived in the same place since birth among urban areas (low coefficients).Less equally dispersed are immigrants from abroad as well as immigrants from other statistical regions.Coefficients in 2011 were either lower than or equal to those in 2002.Therefore we cannot claim that the differences among urban areas were growing -obviously they were not.

Unemployment and education
In 2011 the rate of employed persons per 1000 inhabitants in urban areas was almost equal to national average (407 and 408) the values for unemployed were also very close together (55 and 53).
As shown on the graph (Fig. 9) old industrial centers Maribor, Kočevje and Trbovlje were characterized by moderate deconcentration of employed persons, while the center of least developed region in Slovenia Murska Sobota as well as an important industrial center Velenje were facing concentration of unemployed.Urban areas with moderate deconcentration of unemployed and moderate concentration of employed persons were mainly from central and western Slovenia.These two groups as well as the group of urban areas with moderate deconcentration of inactive (15 and older) population (3 Ljubljana's satellites) were obviously better economically situated than those from the upper side of the graph.To analyze differences in education we computed the so called "Educational attainment index" as a ratio between the number of persons (15 years and older) with high education and the number of people with basic or no education multiplied by 100.The first thing to be pointed out is a relatively high above national average index of urban areas in total.
The level of education in urban areas is therefore still considerably higher than elsewhere, but the growth of index in urban areas in the period 2002-2011 was below the national average.The difference is therefore slowly decreasing.Trzin had the highest educational attainment index in 2002 as well as in 2011, a small city at the northeast outskirts of Ljubljana with relatively recent urban development with attractive housing in a suburban setting with relatively good accessibility to the center of Ljubljana.In 2002 net migration in this small city was high above all other urban areas (31,9 per 1000) while in 2011 it already had negative net-migration.Nevertheless it attracted many highly educated and economically well situated residents and remains to be their residential area.
Among the 18 largest Slovenian urban areas, as we can see on the graph (Fig. 9: Unemployed and employed persons per 1000 inhabitants in 18 largest Slovenian urban areas in 2011 grouped according to locational quotients for employed, unemployed and inactive population. Fig. 10), Ljubljana is positioned high above all others, followed surprisingly by Murska Sobota.Index of this center of least developed Slovenian region grew from 68,8 to 104,5, partly because of the growth of the number of persons with postsecondary education and even more because of decrease of number of persons with basic or no education.
Urban areas below the national average are or were industrial centers with industry largely based on low educated labor force.
We classified all 104 urban areas according to locational quotients for 3 levels of education in 2011 into several groups and the most numerous were the group with deconcentration of persons with higher education ( 29) and the group with concentration of persons with basic or no education (15).One third of the largest 18 urban areas was also in these two groups that include urban areas with relatively lower educated population.In 2011 40 urban areas had educational attainment index below the national average among them 5 from the group of 18 largest (Fig. 9: Unemployed and employed persons per 1000 inhabitants in 18 largest Slovenian urban areas in 2011 grouped according to locational quotients for employed, unemployed and inactive population. Fig. 10).

Conclusion
We can conclude that the population of Slovenian urban areas in the first decade of 21 st century is older than the national average, but the difference is not at all very significant.There are many urban areas with ageing index below the national average even among the 18 largest.
The expected loss of population due to suburbanization is not something that can be generalized for all major urban areas in Slovenia.Some Slovenian cities had a decreasing number of population in the first decade of 21 st century, but others did not.For urban areas around Ljubljana (within commuting distance up to 30 or even 45 minutes by car) we can contest that many of them were gaining on behalf of Ljubljana.Ljubljana on the other side gained on behalf of many rural and urban settlements from all over Slovenia and in large part on behalf of international migrations.Its' population grew since 2008 (the new definition of population), not radically, but at least the number of inhabitants stopped decreasing.
Not only in urban areas, but in Slovenia in general, a natural component is a rather unreliable factor of population growth.With ageing population and modern reproduction behavior we cannot expect a high natural increase.As shown on the graph, the natural increase is in negative correlation with the ageing index.Higher ageing index generally means a smaller increase (or a bigger decrease).The Pearson coefficient of correlation for 2002 was -0,50 and for 2011, the correlation is rather stronger: -0,66.Urban areas with a relatively large share of old population tend to have a natural decrease.However, this is not the case just in the cities, many other settlements have similar problems as average values for urban areas and for Slovenia are not that much apart from each other.
Another important outcome of our investigation is that Slovenian urban areas differ the most from other settlements in Slovenia by their migration characteristics.In 2011 only 6 out of 104 had locational quotient of population that lived in the same settlement since birth above 1 (computed for all population).More than two thirds of all persons that immigrated to their place of residence from another statistical region lived in urban areas and 60% of those that immigrated from a foreign country (50% of total Slovenian population in 2011 lived in urban areas).
Considering only urban areas we acknowledged that according to coefficients of localization immigrants from another statistical region and immigrants from foreign countries are the most unevenly distributed among urban areas.
We expected that urban areas would have population with a higher level of education and the data is showing that this really is the case.Actually the 50/50 line is between vocational upper secondary and technical upper secondary education and from there up the share of persons with a certain degree of education that live in urban areas is growing up to ¾ for 3 rd cycle of higher education (in 2011).
The employed and unemployed are relatively equally distributed among urban and rural areas, but the distribution of unemployed among different urban areas is relatively uneven (coefficient of localization for 2011: 0,10).Unemployment as expected is not the matter of urban or rural environment.It is more a matter of economic situation in employment centers (usually urban areas).However, due to commuter based working force in Slovenia the consequences of economic decline in certain employment center are rather equally influencing urban center itself as well as the surrounding suburban and rural areas.
At the end we would like to express our dilemma about the urban character of Slovenian urban areas as they are defined by Statistical office.We believe that they consist in large part of settlements with only partly urban character.Many of them are very similar to their neighboring settlements that are not defined as urban.As most of Slovenian urban areas also have a very small number of inhabitants, demographic indicators may not be very accurate and tend to vary from year to year on the basis of coincidental factors.Finally we may contest that Slovenian urban population is rather differentiated.We may assume that differences in the demographic characteristics of population within urban settlements are rather important and deserve to be analyzed; wich we believe is an important task for further investigation of urban population in Slovenia.

Fig. 1 :
Fig. 1: Population of Slovenian urban areas in 2011 and the line of Rank size rule according to the size of Ljubljana (only 25 largest urban areas are shown on the graph).

Fig. 2 :
Fig. 2: Changes in the number of inhabitants in the period 2003-2011 in Slovenia, urban areas, Ljubljana and Maribor (break in series because of the new statistical definition of population is marked grey; left scale for Ljubljana and Maribor, right scale for Slovenia and Urban areas).

Fig. 4 :
Fig. 4: Index of ageing for 18 largest Slovenian urban areas in 2002 and 2011.Locational quotients were computed for every single urban area.The value 1,00 indicates that the share of an age group of certain urban area in all the urban population of that age is equal to the share of the urban area's population in total urban population.Value above 1,00 indicates an above average share and a value below 1,00 a below average share.

Fig. 5 :
Fig. 5: Relative change of the coefficient of masculinity and number of population in the period 2002 to 2011 in 104 urban areas.

Fig. 8 :
Fig. 8: Coefficients of localization for urban population according to migration status.

Fig. 9 :
Fig. 9: Unemployed and employed persons per 1000 inhabitants in 18 largest Slovenian urban areas in 2011 grouped according to locational quotients for employed, unemployed and inactive population.

Fig. 11 :
Fig. 11: The distribution of pairs of values for 104 urban areas for ageing index (x) and natural increase per 1000 (y) in 2002 (left) and 2011 (right)