Ms Ragdad Cani Miranti (p), Ms Sri Juli Asdiyanti

A two-way interdependence linkage between food security and price stability has been a concern for decades, highlighting the importance of these two indicators in economic stability. This study aims to assess the clustering pattern, space-time inter-correlation, and direct-indirect effects of food security on price stability. Combining official data and remote sensing data from 2021 to 2024, we construct a comprehensive Food Security Index by integrating official statistical data with remote sensing approaches, aggregating 15 indicators into four dimensions: economic security, quantity security, quality security, and resource security. The resource security dimension represents a methodological innovation, utilising MODIS Land Cover data for cropland classification, Normalized Difference Vegetation Index (NDVI) and Normalized Difference Water Index (NDWI) for vegetation and water monitoring, CHIRPS rainfall data and Global Surface Water datasets for flood risk assessment which derived from Google Earth Engine. This integration accommodates critical shortcoming in traditional food security measurements, particularly the temporal lag and spatial granularity constraints of official statistics, while enabling real-time monitoring of agricultural resources and early detection of supply shocks. To address data limitations in inflation measurement, we employ a sister city approach using hierarchical clustering methods to estimate inflation rates for non-Consumer Price Index (non-CPI) districts based on economic indicators and consumption patterns. Results from Rose Diagram and Directional LISA reveal that 3 of 4 dimensions are vulnerable to the external shocks from neighbourhood. The three security dimensions (quantity, economics, and resources) exhibit significant vulnerability to external shocks from neighbouring regions, with Central Sumatra exemplifying the highest sensitivity to spatial spillovers during the time period. Quality security emerges as the most persistent dimension, demonstrating consistent positive spatial associations across regions. In addition, Dynamic Spatial Durbin Model indicates pronounced indirect effects of economic security, quality security, and resource security on the inflation level. These findings necessitate integrated inter-regional policies that recognise spatial coordination, particularly incorporating sustainable food security management to monitor inflation rate and price stability.

Prof. Stefano Usai (p), Ugo Gragnolati, Alberto Tidu, Gioele Filia

In this paper, we primarily contribute to assessing the spatial concentration of economic activities in Italy since the Great Recession. To do so, we rely on an especially detailed and complete data set that features information on location, employment, and other economic variables for every plant in Italy across a wide range of 3-digit ATECO sectors. In order to capture the behaviour of interest at different spatial scales, we turn to the M function as defined by Marcon
and Puech (2010). While allowing for a thorough evaluation of statistical significance, this approach also permits decomposing the aggregate measure of spatial concentration into its local components, as described by Marcon and Puech (2023). Hence, by computing the M function for 2007 and 2021, we can trace how the spatial concentration of economic activities has evolved across various geographic scales and how the different local economies have contributed to such change. In parallel, we use the M function also to provide a fine-grained measure of local productive variety, thus allowing us to evaluate in detail where and how variety has evolved in Italy after the Great Recession.
One particular novelty of this paper is the analysis of spatial concentration using the M function, extended to the inter-sectoral level. That is, instead of measuring how many other plants in the same sector are proportionately present within some distance from an establishment, we focus on plants in other sectors from the one in which the establishment operates. The resulting measure then allows for evaluating the co-location of industries, as described by Marcon and Puech (2010, pp. 751–52). When evaluated at the local level, this application allows identification of the spatial interaction locus between any two sectors.

Dr. Dávid Bilicz (p)

This paper examines how interregional collaboration networks relate to regional knowledge production when collaboration is conceptualized as a multilayered phenomenon. Building on the regional knowledge production function framework, the study integrates three complementary collaboration channels (Framework Programme partnerships, scientific co-publication networks, and co-patenting relations) within an overlapping multiplex network structure. The empirical analysis is conducted at the NUTS3 level for European Union regions using panel data and spatial econometric models to jointly account for relational and geographical mechanisms of knowledge diffusion. By combining multiplex network representations with spatial error specifications, the framework captures both network-mediated knowledge inflows and spatially correlated innovation processes.
The results show that the relationship between network position and regional innovation is neither uniform nor layer-invariant. Network effects differ across collaboration channels and across centrality dimensions. Connectivity- and embeddedness-oriented positions (particularly degree and closeness centrality, and alfa-coreness in some cases) are positively associated with regional patenting performance, suggesting that broad access to collaboration partners, the short network distances and membership to the tightly connected core facilitate knowledge diffusion. In contrast, brokerage-oriented roles measured by betweenness centrality are not consistently beneficial and may even be negatively related to innovation outcomes in technological collaboration networks, indicating potential coordination burdens or fragmented collaboration structures. Accounting for spatial dependence materially affects the magnitude and statistical significance of network coefficients, demonstrating that relational and spatial processes are intertwined. Moreover, multiplex aggregation reveals network-innovation relationships that are not detectable in single-layer analyses, highlighting that cross-layer consistency in regional embeddedness can matter even when individual collaboration layers show weak effects.
Overall, the findings contribute to a more comprehensive empirical understanding of how interregional collaboration structures shape regional innovation performance. Methodologically, the study demonstrates the added value of combining multiplex network analysis with spatial econometrics at a fine territorial scale. Substantively, it refines the common assumption that “more central is better” by showing that different network roles carry distinct implications for regional innovativity. The results underscore the importance of jointly accounting for relational and geographical dimensions in regional innovation research and offer a more nuanced basis for the design and evaluation of collaboration-oriented innovation policies.