We develop a key-player model by allowing for link formation so that when a person is removed from a network the other individuals can form new links while still optimally providing crime effort. We then put our model to the test, using data on adolescent delinquents in the United States, and provide new results regarding the identification of peer effects. This is done by a structural estimation and simulation of our model. Compared to a policy that removes randomly delinquents from the network, a key player policy engenders a crime reduction that can be as large as 35 percent. We discuss how to implement the key-player policy in the real world, primarily within criminal networks, but also within financial, R&D, development, political and tax-evasion networks.

We propose a novel mechanism to facilitate understanding of systemic risk in …nan-cial markets. The literature on systemic risk has focused on two mechanisms, common shocks and domino-like sequential default. We provide a new model that draws on the games-on-networks literature. Transmission in our model is not based on default. In-stead, we provide a simple microfoundation of bankspro…tability based on competition incentives and the outcome of a strategic game. As competitors loans change, both for closely connected ones and the whole market, banks adjust their own decisions as a result, generating a transmissionof shocks through the system. Our approach permits us to measure both the degree that shocks are ampli…ed by the network structure and

The centrality of an agent in a network has been shown to be crucial in explaining different behaviors and outcomes. In this paper, we propose an axiomatic approach to characterize centrality measures for which the centrality of an agent is recursively related to the centralities of the agents she is connected to. This includes the Katz-Bonacich and the eigenvector centrality. The core of our argument hinges on the power of the con-sistency axiom, which relates the properties of the measure for a given network to its properties for a reduced problem. In our case, the reduced problem only keeps track of local and parsimonious information. Our axiomatic characterization highlights the conceptual similarities among those measures.

We propose a novel mechanism to facilitate understanding of systemic risk in financial markets. The literature on systemic risk has focused on two mechanisms, common shocks and domino-like sequential default. We provide a new model that draws on the games-on-networks literature. Transmission in our model is not based on default. Instead, we provide a simple microfoundation of banks’profitability based on competition incentives and the outcome of a strategic game. As competitors’loans change, both for closely connected ones and the whole market, banks adjust their own decisions as a result, generating a ‘transmission’of shocks through the system. Our approach permits us to measure both the degree that shocks are amplified by the network structure and

This paper proposes a model of network interactions in the interbank market. Our innovation is to model systemic risk in the interbank network as the propagation of incentives or strategic behavior rather than the propagation of losses after default. Transmission in our model is not based on default. Instead, we explain bank prof-itability based on competition incentives and the outcome of a strategic game. As competitors ’ lending decisions change, banks adjust their own decisions as a result: generating a ‘transmission ’ of shocks through the system. We provide a unique equi-librium characterization of a static model, and embed this model into a full dynamic model of network formation. We also determine the key bank, which is the bank that is crucial for the stability of the financial network.

The paper studies how the institutional organization of the metro, the working of labor and land markets, and the process of tax competition between central and suburban jurisdictions shape the economic structure of the metro when consumers and firms are free to choose their locations. Contrary to general belief, we show that institutional fragmentation is socially desirable when suburban jurisdictions are not allowed to free ride on the public goods provided by the central city. Specifically, jobs are more agglomerated than the supply of public goods. We also show that the institutional and economic structure of the central city never coincide. Last, fiscal competition leads to an excessive decentralization of jobs, whereas corporate tax harmonization sustains the optimal employment distribution..

We examine how interaction choices depend on the interplay of social and physical distance, and show that agents who are more central in the social network, or are located closer to the geographic center of interaction, choose higher levels of interactions in equilibrium. As a result, the level of interactivity in the economy as a whole will rise with the density of links in the social network and with the degree to which agents are clustered in physical space. When agents can choose geographic locations, there is a tendency for those who are more central in the social network to locate closer to the interaction center, leading to a form of endogenous geographic separation based on social distance. We also show that the market equilibrium is not optimal because of social externalities. We determine the value of the subsidy to interactions that could support the first-best allocation as an equilibrium and show that interaction effort and the incentives for clustering are higher under the subsidy program. Finally, we interpret our model in terms of labor-market networks and show that the lack of good job contacts would be here a structural consequence of the social isolation of inner-city neighborhoods.

We study the relationship between network centrality and educational outcomes using a field experiment in primary schools in Bangladesh. After obtaining information on friendship networks, we randomly allocate students into groups and give them individual and group assignments. We find that the groups that perform the best are those whose members have high Katz-Bonacich and key-player centralities. Leaders are mostly responsible for this effect, while bad apples have little influence. Group members’ network centrality is also important in shaping individual performance. We show that network centrality captures non-cognitive skills, especially patience and competitiveness

Abstract. We study the class of potential games that are also graphical games with respect to a given graph G of connections between the players. We show that, up to strategic equivalence, this class of games can be identified with the set of Markov random fields on G. From this characterization, and from the Hammersley-Clifford theorem, it follows that the potentials of such games can be decom-posed to local potentials. We use this decomposition to strongly bound the lengths of strict better-response paths. This result ex-tends to generalized graphical potential games, which are played

We study the causal impact of network centrality on educational outcomes using field experiments in primary schools in Bangladesh. After obtaining information on friendship networks, we randomly allocate students into groups and give them individual and group assignments. We find that groups that perform best are those whose members have high Katz-Bonacich and key-player centralities. Leaders are mostly responsible for this effect, while bad apples have little influence. Own Katz-Bonacich centrality is associated with better individual performance only if it is above the aver-age centrality of the group. Further experiments reveal that leadership, as measured by network centrality, mainly captures non-cognitive skills, especially patience and competitiveness.