nep-net New Economics Papers
on Network Economics
Issue of 2020‒05‒25
twelve papers chosen by
Alfonso Rosa García
Universidad de Murcia

  1. Social Networks with Misclassified or Unobserved Links By Arthur Lewbel; Xi Qu; Xun Tang
  2. A Dynamic Structural Model of Virus Diffusion and Network Production: A First Report By Victor Aguirregabiria; Jiaying Gu; Yao Luo; Pedro Mira
  3. Costly agreement-based transfers and targeting on networks with synergies By Mohamed Belhaj; Frédéric Deroïan; Shahir Safi
  4. Smooth marginalized particle filters for dynamic network effect models By Dieter Wang; Julia Schaumburg
  5. The probabilities of node-to-node diffusion in fixed networks By King, Maia
  6. Anti-conformism in the Threshold Model of Collective Behavior By Michel Grabisch; Fen Li
  7. Detecting Latent Communities in Network Formation Models By Shujie Ma; Liangjun Su; Yichong Zhang
  8. Determinants of occupational mobility within the social stratification structure in India By Vinay Reddy Venumuddala
  9. Contagion of Pro- and Anti-Social Behavior among Peers and the Role of Social Proximity By Eugen Dimant
  10. Optimal Bailouts and the Doom Loop with a Financial Network By Agostino Capponi; Felix C. Corell; Joseph E. Stiglitz
  11. International Trade and Social Connectedness By Michael Bailey; Abhinav Gupta; Sebastian Hillenbrand; Theresa Kuchler; Robert Richmond; Johannes Stroebel
  12. Optimization of an Information Diffusion Model of Influencer Marketing -Evaluation of Speed- and Cost-oriented Marketing based on Influencer-Market Elasticity- By Taisuke Ehara; Kaoru Kuramoto; Yosuke Kurihara; Toshiyuki Matsumoto; Satoshi Kumagai

  1. By: Arthur Lewbel (Boston College); Xi Qu (Shanghai Jiao Tong University); Xun Tang (Rice University)
    Abstract: We identify and estimate social network models when network links are either misclassified or unobserved in the data. First, we derive conditions under which some misclassification of links does not interfere with the consistency or asymptotic properties of standard instrumental variable estimators of social effects. Second, we construct a consistent estimator of social effects in a model where network links are not observed in the data at all. Our method does not require repeated observations of individual network members. We apply our estimator to data from Tennessee's Student/Teacher Achievement Ratio (STAR) Project. Without observing the latent network in each classroom, we identify and estimate peer and contextual effects on students' performance in mathematics. We found that peer effects tend to be larger in bigger classes, and that increasing peer effects would significantly improve students' average test scores.
    Keywords: Social networks, Peer effects, Misclassified links, Missing links, Mismeasured network, Unobserved network, Classroom performance
    Date: 2019–07–30
  2. By: Victor Aguirregabiria; Jiaying Gu; Yao Luo; Pedro Mira
    Abstract: This paper presents a dynamic structural model to evaluate economic and public health effects of the diffusion of COVID-19, as well as the impact of factual and counterfactual public policies. Our framework combines a SIR epidemiological model of virus diffusion with a structural game of network production and social interactions. The economy comprises three types of geographic locations: homes, workplaces, and consumption places. Each individual has her own set of locations where she develops her life. The combination of these sets for all the individuals determines the economy's production and social network. Every day, individuals choose to work and consume either outside (with physical interaction with other people) or remotely (from home, without physical interactions). Working (and consuming) outside is more productive and generates stronger complementarities (positive externality). However, in the presence of a virus, working outside facilitates infection and the diffusion of the virus (negative externality). Individuals are forward-looking. We characterize an equilibrium of the dynamic network game and present an algorithm for its computation. We describe the estimation of the parameters of the model combining several sources of data on COVID-19 in Ontario, Canada: daily epidemiological data; hourly electricity consumption data; and daily cell phone data on individuals' mobility. We use the model to evaluate the health and economic impact of several counterfactual public policies: subsidies for working at home; testing policies; herd immunity; and changes in the network structure. These policies generate substantial differences in the propagation of the virus and its economic impact.
    Keywords: COVID-19; Virus diffusion; Dynamics; Production and social networks; Production externalities; Public health
    JEL: C57 C73 L14 L23 I18
    Date: 2020–05–11
  3. By: Mohamed Belhaj (Aix-Marseille Univ, CNRS, EHESS, Ecole Centrale, AMSE, Marseille, France.); Frédéric Deroïan (Aix-Marseille Univ, CNRS, EHESS, Ecole Centrale, AMSE, Marseille, France.); Shahir Safi (Aix-Marseille Univ, CNRS, EHESS, Ecole Centrale, AMSE, Marseille, France.)
    Abstract: We consider agents organized in an undirected network of local complementarities. A principal with a limited budget offers costly bilateral contracts in order to increase the sum of agents' effort. We study excess-effort linear payment schemes, i.e. contracts rewarding effort in excess to the effort made in absence of principal. The analysis provides the following main insights. First, for all contracting costs, the optimal unit returns offered to every targeted agent are positive and generically heterogeneous. This heterogeneity is due to the presence of outsiders, who create asymmetric interaction between contracting agents. Second, when contracting costs are low, it is optimal to contract with everyone and optimal unit returns are identical for all agents. Third, when contracting costs are sufficiently high, it becomes optimal to target a subset of agents, and optimal targeting can lead to NP-hard problems. In particular, when the intensity of complementarities is sufficiently low, a correspondence is established between optimal targeting and the densest k subgraph problem. Overall, the optimal targeting problem involves a trade-off between centrality and budget spending-central agents are influential, but are also more budget-consuming. These considerations can lead the principal to not target central agents.
    Keywords: networked synergies, aggregate effort, optimal group targeting, linear contract
    JEL: C72 D85
    Date: 2020–04
  4. By: Dieter Wang (Vrije Universiteit Amsterdam); Julia Schaumburg (Vrije Universiteit Amsterdam)
    Abstract: We propose a dynamic network model for the study of high-dimensional panel data. Crosssectional dependencies between units are captured via one or multiple observed networks and a low-dimensional vector of latent stochastic network intensity parameters. The parameterdriven, nonlinear structure of the model requires simulation-based filtering and estimation, for which we suggest to use the smooth marginalized particle filter (SMPF). In a Monte Carlo simulation study, we demonstrate the SMPF’s good performance relative to benchmarks, particularly when the cross-section dimension is large and the network is dense. An empirical application on the propagation of COVID-19 through international travel networks illustrates the usefulness of our method.
    Keywords: Dynamic network effects, Multiple networks, Nonlinear state-space model, Smooth marginalized particle filter, COVID-19
    JEL: C63 C32 C33
    Date: 2020–05–10
  5. By: King, Maia
    Abstract: Network transmission of infection or information can have serious social, economic and political effects. Heuristics are often used to address the computationally hard optimal seeding problem, and to approximate SIR models of epidemics. This paper develops a new heuristic for the probabilities of node-to-node diffusion in networks. The simple formula uses De Morgan’s laws to eliminate the double counting of signals found in diffusion centrality. It provides a new measure of centrality — word-of-mouth centrality — which gives the average probability that a signal emitted by a node will be received by other nodes in the network by diffusion. The paper also gives two further centrality measures for the cases when some nodes obstruct or conceal signals, called obstructed centrality and visibility centrality.
    Date: 2020–05–05
  6. By: Michel Grabisch (CES - Centre d'économie de la Sorbonne - CNRS - Centre National de la Recherche Scientifique - UP1 - Université Panthéon-Sorbonne, PSE - Paris School of Economics); Fen Li (Department of Entomology - Michigan State University [East Lansing] - Michigan State University System)
    Abstract: We provide a detailed study of the threshold model, where both conformist and anti-conformist agents coexist. Our study bears essentially on the convergence of the opinion dynamics in the society of agents, i.e., finding absorbing classes, cycles, etc. Also, we are interested in the existence of cascade effects, as this may constitute a undesirable phenomenon in collective behavior. We divide our study into two parts. In the first one, we basically study the threshold model supposing a fixed complete network, where every one is connected to every one, like in the seminal work of Granovetter. We study the case of a uniform distribution of the threshold, of a Gaussian distribution, and finally give a result for arbitrary distributions, supposing there is one type of anti-conformist. In a second part, we suppose that the neighborhood of an agent is random, drawn at each time step from a distribution. We distinguish the case where the degree (number of links) of an agent is fixed, and where there is an arbitrary degree distribution. We show the existence of cascades and that for most societies, the opinion converges to a chaotic situation.
    Keywords: threshold model,anti-conformism,absorbing class,opinion dynamics
    Date: 2019
  7. By: Shujie Ma; Liangjun Su; Yichong Zhang
    Abstract: This paper proposes a logistic undirected network formation model which allows for assortative matching on observed individual characteristics and the presence of edge-wise fixed effects. We model the coefficients of observed characteristics to have a latent community structure and the edge-wise fixed effects to be of low rank. We propose a multi-step estimation procedure involving nuclear norm regularization, sample splitting, iterative logistic regression and spectral clustering to detect the latent communities. We show that the latent communities can be exactly recovered when the expected degree of the network is of order log n or higher, where n is the number of nodes in the network. The finite sample performance of the new estimation and inference methods is illustrated through both simulated and real datasets.
    Date: 2020–05
  8. By: Vinay Reddy Venumuddala
    Abstract: In this study, we make use of empirically observed occupational stratification patterns, in order to identify the relationship between education and social mobility of individuals - the latter is approximated by the social distance of an individual's occupation from his/her household's traditional niche occupation. Our study draws upon a novel occupational network construction proposed in Lambert (2018), with slight adjustments, to empirically identify social stratification patterns using cross sectional household surveys available in the Indian context. We use IHDS-2 data-set for the purpose of our study.
    Date: 2020–05
  9. By: Eugen Dimant
    Abstract: This paper uses a novel experimental design to study the contagion of pro- and anti-social behavior and the role of social proximity among peers. Across systematic variations thereof, we find that anti-social behavior is generally more contagious than pro-social behavior. Surprisingly, we also find that social proximity amplifies the contagion of anti-social behavior more strongly than the contagion of pro-social behavior. Anti-social individuals are also most susceptible to the behavioral contagion of other anti-social peers. These findings paired with the methodological contribution inform the design of effective norm-based policy interventions directed at facilitating pro-social behavior and reducing anti-social behavior in social and economic environments.
    Keywords: behavioral contagion, peer effects, anti-social & pro-social behavior
    JEL: C91 D64 D90
    Date: 2020
  10. By: Agostino Capponi; Felix C. Corell; Joseph E. Stiglitz
    Abstract: Banks usually hold large amounts of domestic public debt which makes them vulnerable to their own sovereign’s default risk. At the same time, governments often resort to costly public bailouts when their domestic banking sector is in trouble. We investigate how the interbank network structure and the distribution of sovereign debt holdings jointly affect the optimal bailout policy in the presence of this "doom loop". Rescuing banks with high domestic sovereign exposure is optimal if these banks are sufficiently central in the network, even though that requires larger bailout expenditures than rescuing low-exposure banks. Our findings imply that highly central banks can use exposure to their own government as a strategic tool to increase the likelihood of being bailed out. Our model thus illustrates how the "doom loop" exacerbates the "too interconnected to fail" problem in banking.
    JEL: G01 G21 G28 H63 H81
    Date: 2020–05
  11. By: Michael Bailey; Abhinav Gupta; Sebastian Hillenbrand; Theresa Kuchler; Robert Richmond; Johannes Stroebel
    Abstract: We use anonymized data from Facebook to construct a new measure of the pairwise social connectedness between 180 countries and 332 European regions. We find that two countries trade more with each other when they are more socially connected and when they share social connections with a similar set of other countries. The social connections that determine trade in each product are those between the regions where the product is produced in the exporting country and those where it is used in the importing country. Once we control for social connectedness, the estimated effect of geographic distance on trade declines substantially, and the effect of country borders disappears. Our findings suggest that social connectedness increases trade by reducing information asymmetries and by providing a substitute for both trust and formal mechanisms of contract enforcement. We also present evidence against omitted variables and reverse causality as alternative explanations for the observed relationships between social connectedness and trade flows.
    Keywords: international trade, social connectedness, contract enforcement, information frictions
    JEL: F10 F50 F60
    Date: 2020
  12. By: Taisuke Ehara (Aoyama Gakuin University); Kaoru Kuramoto (Aoyama Gakuin University); Yosuke Kurihara (Aoyama Gakuin University); Toshiyuki Matsumoto (Aoyama Gakuin University); Satoshi Kumagai (Aoyama Gakuin University)
    Abstract: Owing to its high degree of credibility, influencer marketing is incorporated into the advertising activities of many companies. However, it is often unclear for the company which market initiator (defined as the consumer to whom a company should first provide information) should be given information for marketing purposes. A company?s choice of market initiators in social networking services influences advertising speed as well as marketing cost.In this study, market initiator candidates were identified to determine the most suitable market initiator in the information diffusion simulation. The cost effectiveness of each candidate was evaluated in terms of influencer-market elasticity. First, we optimized the information diffusion model in influencer marketing and considered whether this or mass marketing is best for companies. Influencer-market elasticity was then determined based on the information obtained from the simulation. Using such elasticity, we clarified whether companies should request advertising from influential users, with an emphasis on speed of advertising, or from insensitive users, with emphasis on advertising costs. The proposed methods were applied to actual companies, and the most suitable market initiator was identified for the maximum and minimum values of influencer-market elasticity in a certain period.
    Keywords: information diffusion modelsocial networking servicecentrality index
    JEL: M10 M31 L19
    Date: 2020–02

This nep-net issue is ©2020 by Alfonso Rosa García. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at For comments please write to the director of NEP, Marco Novarese at <>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.