Political Science

Isabella Bellezza is a sixth-year Ph.D. candidate.  Her research is at the intersection of international security and political economy, with a focus on global policing. Isabella’s dissertation, Blurring the Line: The Global Transformation of Border Control in the Fight against Illicit Trade, and related papers investigate the causes of international cooperation between border agencies to police global supply chains—a stark and understudied shift away from previous traditional, territorial approaches to border control. The dissertation draws on new cross-national and longitudinal data to map this post-Cold War transformation in how borders are governed, and it addresses three questions: (i) How did a new “intelligence-driven” approach to border control based on the inter-state exchange of sensitive information and risk-based profiling emerge? (ii) Why did that policy model rapidly diffuse around the world despite opposition from privacy and sovereignty advocates? And (iii) Why do states ultimately share sensitive policing information – a critical input for intelligence-driven border governance – with some states but not others? In a related stream of research, Isabella investigates the consequences of this global transformation for trade and geopolitical rivalry. She uses a range of quantitative and qualitative tools to answer these questions, including large-N observational studies, inferential and descriptive network analysis, elite interviews, process tracing, and event ethnography. Findings contribute to research on change and continuity in global governance, international cooperation, transgovernmental networks, and networks in IR.

Previously, Isabella was a Fulbright recipient and NSEP Boren scholar in Brazil, where she researched transnational organized crime. Isabella received a B.A. in political science with highest honors from Swarthmore College in 2017.

Job Market Paper Title: 

 “Network Effects and the Globalization of Customs Policing"

Job Market Paper Abstract:  

Why do states exchange sensitive law enforcement information with some states but not others? Customs agencies increasingly exchange policing information with their foreign counterparts to govern inbound flows of goods and people before they arrive at territorial borders. Agreements facilitating this exchange of sensitive information have proliferated twenty-fold since 1990, from 37 cumulative agreements in 1990 to nearly 800 as of 2022. This rapid proliferation is puzzling because, although states have a common interest in cooperating against mutual non-state threats, sharing sensitive information is risky. Partners could exploit or mishandle shared secrets by, for example, revealing sources and methods of intelligence collection, providing politicized information and jeopardizing ongoing investigations. Thus, notwithstanding demand for cooperation rooted in the prospect of joint gains, states need assurances that they will not be exploited. Where do states turn for such assurances?

I argue that states look to their social network to assess the reliability of potential partners. Specifically, I hypothesize that two network signals – information deduced from the structure of relationships in which an actor is embedded – reduce the fear of being exploited: the number of friends in common with the potential partner and their relative popularity. I evaluate this network theory of international cooperation with a longitudinal inferential network model (a TERGM) and find evidence of a robust association between the hypothesized network effects and the initiation of customs police cooperation. Once network effects are modeled, I find surprisingly little evidence that the shared threat environment, domestic institutional similarities, or geopolitical affinity drive states to initiate cooperation on customs policing. The quantitative results are bolstered by a set of cross-national interviews with customs practitioners obtained by attending international practitioner forums. Findings show how networks can help endogenously solve collective action problems. In so doing, the paper contributes to a longstanding research agenda on international cooperation and to an emerging body of research that uses network (rather than dyadic) theoretical and methodological approaches to study IR.