Zhiwu Wei (p), Davide Luca (p), Neil Lee

Maria Podkorytova

We study the importance of spatial inequality in protest incidence. Leveraging data on micro-estimates of relative wealth index across 2.61 million neighbourhoods, we construct novel measures on spatial inequality for 28,675 regions at Administrative Level 2 within 89 Global-South countries. The measures are linked to 0.67 million georeferenced protest instances observed at a daily frequency between 1 January 2014 and 31 December 2018. By exploiting the variation within months specific to regions at Administrative Level 1, we find that regions characterized by higher levels of spatial inequality experience more frequent instances of protests. We further match the spatial inequality and protest data to a series of country-level characteristics. Our results suggest that the effects of spatial inequality are particularly pronounced in (i) mature democracies where citizens’ freedom of expression is well protected and the rule of law is strong; (ii) countries with lower levels of economic development and higher levels of unemployment; as well as (iii) previous British colonies. The evidence suggests that spatial inequality has profound implications for political discontent, but may not be identical across all countries in the Global South.

Gloria Cicerone, Sebastian Losacker, Raquel Ortega-Argilés (p)

Davide Luca

Regions are currently facing a twin transition, involving a shift towards more environmentally friendly economies as well as the digitalization of economic activities. Against this backdrop, this paper examines the geography of green, digital, and twin economic activities in the UK, utilizing a novel dataset that captures these emerging sectors through a real-time industrial classification of firms. Additionally, the paper employs a relatedness approach to analyze local capabilities in the twin domain. The results underscore the crucial role of twin activities, encompassing both green and digital elements, in facilitating regional diversification into these respective sectors.

Tomasz Starakiewicz, Piotr Wójcik (p)

Raquel Ortega Argiles

The restaurant sector is pivotal to firm exit research, influencing economic policy and managerial strategy recommendations. Recent studies using online data are based on geographically limited datasets and have largely omitted temporal dynamics in user interactions. Additionally, these studies rely on manual labeling for text analysis, a resource-intensive approach. Our study introduces the first comprehensive, nationwide analysis of restaurant survival using Google Maps data. We analyze a unique dataset collected by us between September 2022 and June 2023 that includes all Polish restaurants that are listed on Google Maps and have at least one review. The sample includes almost 41 thousand companies and characteristics related to individual characteristics of companies, macroeconomic and regional conditions, local surroundings. In addition, we enhance the predictive performance of model by incorporating time-sensitive user interactions. Our model controls for established determinants of business exit and proves robust to data quality issues associated with user-provided business directories. We apply an efficient, label-free method for extracting semantic content from almost 6.5 million reviews, creating useful features for firm exit prediction. We also use a state-of-the-art xgboost algorithm and present an efficient feature selection strategy using hierarchical agglomerative clustering that retains predictive power while reducing model complexity. Our model has broad applications ranging from credit scoring to early-warning systems for business closures, making regional and local comparisons, and presents a viable alternative to geographically constrained Yelp data.

Ilyes Boumahdi (p), Nouzha Zaoujal

Piotr Wójcik

Sentiment analysis of newspaper has been used to assess the overall subjective well-being of countries (Carlquist et al., 2017; Hills et al., 2019). However, few works have covered the territorial aspect. (Boumahdi & Zaoujal, 2023) is one of them, except that it only covered one of the twelve regions of Morocco, making it difficult to make a spatial comparison of the evolution of subjective well-being in Morocco. Also, in this article we propose to extend the evaluation of daily and monthly regional subjective well-being (RSWB) to all regions of Morocco through the analysis of the sentiments of their media coverage.
We scraped nearly 174,000 newspaper articles published online from October 1991 to September 2020. This was followed by the construction of a lexicon of localities, for each of the twelve regions of Morocco, in order to detect articles that cover these regions. Finally, we analysed the textual data of these articles by detecting the sentimental polarity of the words using the French-speaking lexicon FEEL (Abdaoui et al., 2017). Finally, we constructed the RSWB in each region as a relative balance of feelings.
Thus, at the regional average level over the period, the highest RSWB concerns Fès-Meknes (16.7%) slightly above that of the administrative capital Rabat-Salé-Kénitra (16.6%) and the economic capital Casablanca-Settat (16.6%) against a regional average of 16.5% and the lowest index in Guelmim-Oued Noun (15.6%).
The index is relatively homogeneous at the national level with an average coefficient of variation of 4% while its variability differs at the regional level, notably in Guelmim-Oued Noun which records an average coefficient of variation of 21%. Furthermore, 107 early warnings of RSWB were triggered at the regional level between October 1992 and September 2020, the highest number of which was triggered in the Béni Mellal-Khénifra region (17).

Maria Podkorytova (p)

Ilyes Boumahdi

The research examines shifts in demand for labor in Sweden between 2006 and 2022 based on the database of published vacancies. The analysis is provided in spatial and temporal dimensions, giving a focus on regional inequality. The core of the research is constructed by the vacancies from the global advanced producer service (APS) companies working in six sectors: accountancy, advertising, banking, insurance, law, and management. These companies are supposed to ease the inclusion of regions in the global production chains and empower the global exchange of skills. The unique database of Swedish vacancies provides opportunities to investigate the assumptions. It is possible to grasp from the data diversity of approaches to localization and hiring and highlight the regional inequality within Sweden. The diversification of approaches to hiring by the sector of specialization after COVID-19 pandemic is also noted in the research. Coherence of hiring patterns between global APS companies and other companies is in the spotlight of the study. The demand for occupations, provided by the global APS companies is compared to the general distribution of vacancies in selected years and regions. Special attention is given to rapidly growing and globalizing municipalities of Northern Sweden. Overall, combination of remoteness and globalization, revealed through unequal demand for labor, is the core axis of the research.