No abstract

The level of fertility in the majority of European countries is deeply below the replacement level. On the other hand, in the case of some ethnic groups, fertility continues to be on an above-average level. One such example is the Romany population of Slovakia, especially those in very poor, segregated settlements. Very likely, there are strong interconnections between regional disparities, social and cultural segregation and reproduction of Romany. The study tries to discover them. We perform factor and cluster analyses in order to identify major factors which differentiate and create clusters in the set of 79 Slovak local administrative unit (LAU-1) districts from the view point of above-replacement fertility levels. Using the transversal and cohort approaches, another aim is to identify major specificities of Romany reproduction and the range of lag behind the majority population in terms of demographic transition. The concept of 'high fertility trap' is presented in this respect.

We analyse Roma fertility in four neighbouring countries in Central and Eastern Europe with a large Roma minority: in Hungary, Slovakia, Romania and Serbia. The sources of data are the respective national population censuses from 2011. Fertility is measured at the birth cohort level as the average number of children ever born. We make an international comparison of the fertility of Roma and non-Roma majority population women on the basis of completed education. In the case of Hungary, we also explore how the correlation between fertility and ethnic identity is modified when completed education and ethnic residential segregation are controlled. The fertility of Roma women is far above the majority population average in all birth cohorts and in each country. Educational attainment modifies this relationship. The fertility of highly educated Roma and majority population women is converging. The exposure to majority behaviour also has an effect. The lower the level of ethnic residential segregation, the smaller the difference between the fertility of Roma and majority population women. Completed education and residential segregation may exert different forces at the two ends of the educational hierarchy when their joint effect is explored. At the upper end of the social hierarchy, neither segregation nor ethnicity matters; at the lower end, however, both exposure to ethnic majority behaviour and ethnicity matter.