Essays on the Economics of Networked Competition and Conflict
Open AccessThis dissertation presents three essays on competition and conflict, the first develops a bargaining game between coalition forming agents to explain industrial mergers and market entering start-ups, the second models how states form defensive alliances under private information about commitments, and the third presents a formal multilayer network approach to understand endogenous alliance formation. Conflicts and competitions are interactions between agents with limited resources. These are fundamentally economic questions. Additionally, just as recognizing that competitors influence mutual best responses drove the development of game theory, understanding that relationships between agents acting in concert and competition drive the importance of network theory to study conflict and competition. This dissertation attempts to inspire new ideas about how private information in competitive interactions within and between networks of agents in firms, conflict, and alliances.Chapter 1 develops a novel model that describes how firms are born, vertically integrate, and splinter into new firms in high-innovation sectors. Understanding why firms need to integrate vertically has become vital to understanding the nature of firms themselves. For many years economists assumed vertical mergers create heterogeneous agents. These divisions complicate analysis and can be arbitrary. Chapter 1 presents a theory of start-up firms that endogenizes market entry using sequential bidirectional signaling games of private information. When a selling agent offers its idea for a contract to a buyer, they reveal some portion of their private information. Similarly, in negotiations the buyer reveals some of its private information, driving the seller’s beliefs about their joint probability of success. Additionally, the selling agent knows whether it will accept or reject and has a market information advantage over the buyer, leading sellers to defect, with some statistical certainty, forming new start-up firms. Agents merge, become firms, and continue to vertically integrate to increase their joint probability of market success by adding agents in exchange for contracts. This process repeats until any agent in the sequence recognizes their amassed knowledge or novel idea has an expected success greater than the expected value of the contract and defects, creating a new start-up. This novel analysis allows for an endogenous model of start-up market equilibrium and its relationship with the business cycle.Chapter 2 examines the endogenous formation of alliance networks in conflict games with private information about types where outcomes pivot on the ability to build a stable coalition rather than individual strength. This paper develops a multi-stage game where a hawk-dove game begins the first stage of conflict between an aggressive State and a target State. The target then turns to adjacent States and attempts to form a defensive coalition through a simultaneous contribution game. Chapter 2 then discusses an extension where the target develops its set of adjacent states pre-conflict, where states invest in diplomatic, economic, and military links focusing on Katz-Bonacich centrality. But any links formed are not considered secure until conflict actually ensues, allowing coalition members to renege on peacetime statements. Private information about link reliability determines stable equilibria and dominates individual military strength. The results explain the dynamics of alliances like NATO and the EU, their level of support for Ukraine, and maneuvering for coalitions in Asia.Chapter 3, coauthored with Professors Sumit Joshi and Ahmed Saber Mahmud, proposes a multilayer network approach to examine alliance formation. In a signed affinity layer, agents are partitioned into clusters, with friendly relations within and hostile connections across clusters. Agents then form defensive collaborations in an alliance layer as follows: Agents in the same cluster form a nested split graph with degree inversely correlated to the level of hostility, and agents from disparate clusters with high-degree and low-hostility form cliques. Within cliques, agents from a cluster that is “intermediate” in terms of discord serve as a bridge to interconnect agents from more “extreme” clusters.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.