Abstract: We introduce the inclusive synthetic control method (iSCM), a modification of synthetic control type methods that allows the inclusion of units potentially affected directly or indirectly by an intervention in the donor pool. This method is well suited for applications with either multiple treated units or in which some of the units in the donor pool might be affected by spillover effects. Our iSCM is very easy to implement using most synthetic control type estimators. As an illustrative empirical example, we re-estimate the causal effect of German reunification on GDP per capita allowing for spillover effects from West Germany to Austria.

Abstract: Many treatments are non-randomly assigned, continuous in nature, and exhibit heterogeneous effects even at identical treatment intensities. Taken together, these characteristics pose significant challenges for identifying causal effects, as no existing estimator can provide an unbiased estimate of the average causal dose-response function. To address this gap, we introduce the Clustered Dose-Response Function (Cl-DRF), a novel estimator designed to discern the continuous causal relationships between treatment intensity and the dependent variable across different subgroups. This approach leverages both theoretical and data-driven sources of heterogeneity and operates under relaxed versions of the conditional independence and positivity assumptions, which are required to be met only within each identified subgroup. To demonstrate the capabilities of the Cl-DRF estimator, we present both simulation evidence and an empirical application examining the impact of European Cohesion funds on economic growth.

Abstract: The European Investment Bank (EIB) is the lending arm of the European Union. It provides loans and other financial products to support investments. One of its main objectives is to support convergence and the growth of the less developed regions. We investigate to what extent EIB attains these goals by estimating the impact of EIB lending on regional economic growth via a counterfactual approach in the continuous treatment framework of both peripheral and core regions. We use an extensive dataset encompassing a rich set of regional economic and demographic variables (NUTS-2 level) and original data on the whole lending portfolio of the EIB between 2012 and 2016 that allows us to estimate whether more transfers generated a stronger impact on economic growth. We then further the analysis, obtaining a more sophisticated set of estimates in which we consider that lending in one region could indirectly affect growth in other regions (the indirect, or spill-over, effect) We find that the growth impact of lending (the “EIB multiplier”) is always positive and greater in the lagging-behind regions. Therefore, absent EIB lending, the observed divergence across EU regions in this period would have been even larger. Furthermore, we find that the spill-over effect is consistently positive, implying that the EIB contributes to the growth of EU regions even beyond where it directly intervenes.

Abstract: The measurement of insecurity in relation to populism has been centered around the operationalization of the risk of unemployment, leaving other dimensions of insecurity largely overlooked. In this working paper, we use advanced statistical data matching with computational techniques to create a synthetic dataset of EWCS and ESS in 23 European countries based on demographics, region, industry sector, and educational attainment of respondents. By combining representative data on high work pressure and job dissatisfaction in the EWCS with the information on the political preferences of respondents and on financial insecurity in the ESS, we demonstrate the presence of a gendered association between insecurity and populist voting. We find that higher work pressure and higher job dissatisfaction are associated with a higher probability of voting for the populist right populist among men, while higher financial insecurity is associated with a higher probability of total populist party voting among women. In addition to demonstrating the advantages of statistical matching over average-based imputation, the paper argues that the debate on the socioeconomic triggers of populist voting needs to account for the role of gender in mediating the effects of insecurity in populist voting. The paper is part of the Working Package “The socio-economic and cultural drivers of populist vote” of the Volkswagen-funded project PRECEDE.

Abstract: An increasing number of studies make a conceptual and empirical link between micro-level experiences of insecurity/precarity and support populist attitudes, voting for populist parties or disengagement from politics (non-voting). The usual operationalization of insecurity in political science examines the political behaviours of insecure individuals by identifying them via precarious contracts or as individuals who face a higher risk of unemployment (see Rovny and Rovny, 2017; Abou-Chadi and Kurer, 2021). To expand this limited understanding of precarity in politics, recent studies in political sociology have offered an enlarged operationalization of precarity in relation to populism that relates to the insecurity in the conditions of work and to financial insecurity (Antonucci et al, 2021; Zhirnov et al, 2023). This article expands this extended approach to investigate the political effects of the insecurity shocks occurred pre and during Covid-19.

Covid-19 scholarship stressed that the pandemic represented a temporary shock to work quality and financial insecurity (Pereira and Oliveira, 2020). In particular, scholars have found a ‘rise and fall of job insecurity’ during and after the pandemic (El Khawli et al, 2022). Hence, we use Covid-19 as a natural experiment to examine the effects of a temporary shock to work and financial insecurity on a longitudinal panel of voters. We analyse a longitudinal panel of 13,590 individuals (potential voters) surveyed before (2018) and during Covid-19 (2022) in six “most different” countries within Europe: the Netherlands, Germany, Italy, France, Spain and Sweden. 

We use a longitudinal panel that is generated from VAA users. While VAA-recruited data may be non-representative of the general population, this issue is minimised by our longitudinal use of data and VAA-generated survey are generally considered reliable sources of data (see Toshkov and Romeijn, 2021). Furthermore, to limit issues of representativeness, we use sampling weights on four demographic variables: age, education, gender, region, and their marginal distributions among vote-eligible populations in each country.

We ran a principal component factor analysis on our indicators of work insecurity to extract two factors that have an eigenvalue higher than one. We find two factors: “precarity of tenure” (a proxy of subjective insecurity regarding the tenure of work) and “precarity at work” (which captures the insecurity in the condition of work in terms of salary, work-life balance etc). We also construct two factors of financial insecurity based on a set of variables on capacity to afford routine expenses, cover for exceptional expenses and make savings: extreme insecurity (which captures difficulty in affording basic needs) and mainstream financial insecurity (a proxy of the financial insecurity faced by ordinary citizens, such as those from the squeezed middle).  

First, we calculate the probability of transition of the voters from and to mainstream/populist vote and from and to vote/non-vote before and after Covid-19. Our models investigate: (1) the relationship between the increase/decrease in work precarity and the probability to vote during Covid-19; (2) the relationship between the increase/decrease in work precarity and the probability to vote during Covid-19 given that individuals were mainstream voters before Covid-19; (3) the relationship between the increase/decrease in work precarity and the probability to vote during Covid-19 given that individuals were populist voters before Covid-19; (4) the relationship between the increase/decrease in work precarity and the probability to vote during Covid-19 given that individuals did not vote before Covid-19.

The findings show that for most types of work insecurity – and for all conditions considered – having higher work precarity is associated with lower mainstream party support, higher right-wing and general populist support and higher probability of non-vote. This confirms the previously found relationship between precarity and populist voting using longitudinal data, thereby establishing a stronger causal link between insecurity and populist voting than found in previous studies using cross-sectional data. We also find separate associations between financial insecurity and voting for populist parties during Covid-19.

While the effect of precarity of tenure and at work moved in the same direction for those who were voting for populist and mainstream parties before Covid-19, the findings suggest the presence of different effects of precarity for those who were non-voters before Covid-19. Precarity of tenure keeps previous non-voters in the non-vote camp, while precarity at work moves them from non-vote to populist voting. We explain this finding through the presence of a “precarity tipping point”: a more extreme form of precarity (precarity of tenure) pushes individuals to “exit” voting (non-vote), while everyday precarity pushes them towards right-wing populist voting. These findings contribute to the understanding of insecurity in relation to populism by showing a continuum between exiting from voting and populist voting as protest vote.

Abstract: The relationship between technology use, innovation, and economic performance is a crucial topic in the economic literature. Our work proposes different integration strategies of two business surveys targeting the same enterprises population, one aiming at the use of ICT the other at innovation (CIS). Firstly, a simple record linkage is performed which uses CIS as pivot survey, retrieving the respondents present in the same three-yearly period in the corresponding ICTSs (harmonisation of reference period). Secondly, a Statistical Matching approach is used, to impute the missing information. The results of the matching are evaluated in terms of the preserved marginal distribution of the variables imputed in the synthetic dataset. All alternative proposals represent zero-burden solutions to provide an integrated dataset of microdata, which constitutes an important source of information for research/policy purposes and to support official statistics in exploring complex causal effects and gain new insights into these fundamental economic phenomena.

Abstract: Depopulation processes of rural and fragile areas are strongly connected with ageing, driven by a lower birth rate compared to the death rate and negative net migration due to limited economic and social opportunities (Reynaud and Miccoli, 2018). An alternative to migration is commuting. However, commuting significantly impact subjective well-being, especially for residents in inner areas. This paper aims to study the relationships between depopulation, mobility choices, daily activities and individual well-being by accounting for the spatial heterogeneity between those living in small towns and disadvantaged areas versus those living in core areas. Previous research (Wheatley, 2014) highlighted relevant differences in subjective well-being between women and men in relation to travel-to-work. Our analysis develops on data from the Italian Time-Use Survey (IT-TUS). This analysis allows to understand how commuting from inner areas has a different impact on well-being in relation to individual characteristics such as gender, age, and resource availability.