D. Del Prete, L. Forastiere, V. Leone Sciabolazza, Causal Inference on Networks under Continuous Treatment Interference: an application to trade distortions in agricultural markets, F.R.E.I.T. Working Paper, 1532.

This paper presents a methodology to draw causal inference in a non-experimental setting subject to network interference. Specifically, we develop a generalized propensity score-based estimator that allows us to estimate both direct and spillover effects of a continuous treatment, which spreads through a network with weighted and directed edges. To showcase this methodology, we investigate whether and how spillover effects shape the optimal level of producers’ support in agricultural markets. Our results show that, in this context, neglecting interference may lead to a downward bias when assessing policy effectiveness.

Download paper (PDF)

D. Del Prete, V. Leone Sciabolazza, G. Santoni, Trade Policy and the Network of Global Value Chains, in preparation.

The international fragmentation of production processes is dramatically deepening the structural interdependence of the world economy. Recent literature has shown that global value chains are modifying countries’ incentives to impose import protection. However the complex structure of their connections entails the existence of specific direct and indirect effects that affect the price domestic suppliers receive. The aim of this paper is to show that final goods tariffs tend to decrease in the domestic content of foreign-produced final goods but at a different pace when distinguishing the direct partner country from third countries. To get the two separate contributions, we decompose the Leontief inverse matrix into its direct and indirect connections and recompute the domestic and foreign valued added content embodied in final goods. Our results show that both direct and indirect flows play a crucial role in shaping trade policy.

Download Presentation (ESCOS 2018)

V. Leone Sciabolazza, R. Vacca, C. McCarty (2020), Connecting the dots: A network intervention to foster scientific collaboration and productivity, Social Networks, 61, 181-195

This article presents the design and implementation of a network intervention to foster scientific collaboration at a research university, and describes an experimental framework for rigorous evaluation of the intervention’s impact. Based on social network analysis of publication and grant data, an innovative type of research funding program was developed as a form of alteration of the university’s collaboration network. The intervention consisted in identifying research communities in the network and creating a new collaborative relation between pairs of unconnected researchers in selected communities. The new collaboration was created to maximally increase the overall cohesion of the target research community. In order to evaluate the impact of the program, we designed a randomized experiment with treatment and control communities based on the Rubin Causal Model approach. The paper describes the intervention design, reports findings from the program implementation, and discusses the statistical framework for future evaluation of the intervention.

V. Leone Sciabolazza, R. Vacca, T. Kennelly Okraku, C. McCarty (2017), Detecting and analyzing research communities in longitudinal scientific networks, Plos One, 12(8), e0182516

A growing body of evidence shows that collaborative teams and communities tend to produce the highest-impact scientific work. This paper proposes a new method to (1) Identify collaborative communities in longitudinal scientific networks, and (2) Evaluate the impact of specific research institutes, services or policies on the interdisciplinary collaboration between these communities. First, we apply community-detection algorithms to cross-sectional scientific collaboration networks and analyze different types of co-membership in the resulting subgroups over time. This analysis summarizes large amounts of longitudinal network data to extract sets of research communities whose members have consistently collaborated or shared collaborators over time. Second, we construct networks of cross-community interactions and estimate Exponential Random Graph Models to predict the formation of interdisciplinary collaborations between different communities. The method is applied to longitudinal data on publication and grant collaborations at the University of Florida. Results show that similar institutional affiliation, spatial proximity, transitivity effects, and use of the same research services predict higher degree of interdisciplinary collaboration between research communities. Our application also illustrates how the identification of research communities in longitudinal data and the analysis of cross-community network formation can be used to measure the growth of interdisciplinary team science at a research university, and to evaluate its association with research policies, services or institutes.

Read Article (link)

V. Leone Sciabolazza (2018), A Net of Moving People: Network Analysis of International Migration Flows, In A. Amighini, S. Gorgoni, M. Smith, Networks of International Trade and Investment, Vernon Press

This paper delves into the determinants of global migration flows by applying a social network methodology. In order to explicitly address the multidimensional aspects of this phenomenon, I employ data on bilateral flows between 169 countries from 1990 through 2010 in combination with information about language and colonial history of each population, and bilateral distance between countries. With respect to traditional approaches, the underlying mechanism characterizing the international migration network (IMN) is described using a data driven approach which takes into account both bilateral and multilateral resistances to migration. To enhance the understanding of the relationships occurring between countries, I analyze contextual effects at different granularity levels and use graph theory to explore the topological properties of the IMN. Overall, complementing existing literature, the results suggest that the IMN is characterized by a strongly persistent hierarchical architecture regardless the spatial granularity of the data. Few countries, both at the local and at the global level, control the whole connectivity of the network, with a prevalence of star-like structures. In terms of network analysis, this means that human international migration patterns can be described by a modular structure displaying high geographical clustering, where the preferential attachment mechanism is driven by cultural homophily (i.e. same language or colonial past) and economic disassortativity (e.g. migration, conditional on distance, take place between countries which differ in terms of GDP). The persistence of this topological structure over time is consistent with a well-established finding in literature, that is the presence of a self-reinforcing process on the intensive margin of the IMN. In conclusion, this paper demonstrates that the level of embeddedness of one country in the IMN is a powerful source of information to improve and further unfold the dynamics underlying migratory processes and its multilateral resistances. With this respect, SNA provides an appropriate theoretical framework of analysis that can be used to extend this study in many innovative ways.

Download Paper (PDF)