Jesse Tweedle
I am an PhD student in economics at the University of Calgary and a research affiliate in CDER at Statistics Canada. I work on social and financial and production networks.
Presenting
Ownership Incentives and Network Evolution
How does a network evolve? A network based on speed has a tendency to become centralized---think of the Internet or the WWW, or the production network of the US, or even airline routes---the optimal ownership structure results in a centralized network. Economics can help us understand the use and ownership incentives that shape the evolution and formation of networks.
I will present real world examples of the interaction between network ownership and performance---think of Facebook's investments in Internet.org, or why and how Turkey frequently blocks Twitter. For example, if performance in a network is defined by speed, the network infrastructure tends to come from a small number of sources and can be easily controlled by them or co-opted by government agencies. If those governments are less than democratic, citizens may not have any control over sources of communication. However, if networks require reliability, networks should be decentralized, which reduces incentives for infrastructure investment. Some of the solutions to these problems are technological, some political, but a serious study of the incentives surrounding network infrastructure and ownership is crucial to understanding the potential of networks.
These real world examples tie directly to network research, drawing from statistical random graphs (Albert and Barabasi (2002)), vulnerability of communication networks (Albert, Jeong and Barabasi (2000)), and financial contagion in economics (Acemoglu et al. (2015), Elliot et al. (2014)). These theories provide useful measures for measuring network structures (Bonacich centrality, concentration centrality, degree sequences, integration and diversification, and more), and how the properties of network integration and diversification interact with performance.
The takeaway from this talk is that economic perspectives are useful for studying networks. You can study the incentives of network use and formation to explain and predict network evolution.
