The changing geographies of fertility in Spain (1981-2018)

The objective of this article is to investigate the variation of fertility across Spain's geographic areas between 1981 and 2018, to highlight spatial change over three decades of major fertility transformations. During the last decades, Spanish fertility decreased considerably to below replacement levels. Although total fertility remains below replacement level in Spain, there are important differences in subnational trends that seem to concentrate around certain areas. Starting from the assumption that there is fertility diversity across the country, which persists over time and such variation is not random but rather spatially driven, we aim to describe the divergence from national trends and analyse the dynamics of spatial patterns of fertility over time with spatial analysis tools. Using from Spanish municipality data, we use 910 territorial units that ensure spatial contiguity and construct yearly fertility indicators derived from census and register data, encompassing fertility by age, birth order, and age at childbirth. We investigate the spatial patterns of fertility and their changes over time, by means of spatial and correlogram analysis, exploring the effects of neighborhood definitions. Results confirm the presence of spatial autocorrelation for all variables throughout the considered timeframe, both at global and local scale. The considered time frame depicts substantial changes in the distribution of low and high fertility clusters, reshaping the geographical distribution of fertility in Spain, with big metropolitan areas as leaders in high fertility, as childbearing is deeply impacted by labor market covariates. The fertility decline in Spain has driven total fertility to below replacement levels in a short period of time, shifting the classical North-South divide of low-high fertility into an East-West clustering, with economic centres such as cities becoming the new focal points of higher fertility. The descriptive and econometric spatial approaches adopted in this article, together with the detailed data available for this study, make it possible to appreciate the scale of fertility changes across the country, its heterogeneity across regions, and the evolution of fertility determinants over time