Regional per capita income differences: Spatial and hierarchical dependencies

Abstract

Relevance. Regional differences in per capita income are a matter of concern for many countries for many reasons, including the threat that such regional disparities pose to national security. Multiple tools and methods are used to investigate these disparities and fix them. The use of lower level aggregated data and the analysis that takes into account spatial interactions thus become particularly relevant because it allows us to reveal the diversity of interactions at the micro-level. Research objective. This study aims to determine the significance of spatial relationships at different levels of data aggregation and hierarchical dependencies in per capita income and highlight the level of administrative division (regional or municipal) that has the greatest impact on per capita income. Methods and data. The analysis relies on the data from 2,270 municipalities in 85 Russian regions. The Hierarchical Spatial Autoregressive Model (HSAR) was used to distinguish both spatial and hierarchical effects. We used three specifications of the model: with estimates of the spatial interaction on the higher level (spatial error at the regional level), on the lower level (spatial lag at the municipal level), and on both levels. Results. Spatial interactions explain the observed variation of per capita income at the municipal level data at both the higher (regional) and lower (municipal) levels but the model with the estimated spatial interaction on the higher level was better. Conclusion. Despite the importance of spatial interactions at the lower level, models that take into account spatial interactions only at the upper level may better explain the observed differences in some cases. Our findings contribute to the rather scarce research literature on spatial relationships on several levels of administrative division. We have shown that for each specific case it is important to identify not only the factors but also the spatial effects in relation to this or that level of the territorial hierarchy.