|
on Network Economics |
| By: | Syngjoo Choi; Sanjeev Goyal; Frédéric Moisan (Division of Social Science) |
| Abstract: | We consider a setting in which individuals can purchase information at a cost and form costly links to access information purchased by others. The theory predicts that in every equilibrium of this game the network is a `star'. For small groups, there exists a unique purchase configuration -a pure influencer outcome, in which the hub node purchases information while all others free ride. For large groups, there exists, in addition, a pure connector outcome in which the hub purchases no information and the peripheral players purchase information. We test these predictions on a new experimental platform with asynchronous activity in continuous time. We start with a baseline setting where subjects only see their own payoffs. We find that subjects create a star network. In small groups, the hub purchases equilibrium level information, but in large groups the hub purchases excessive information and as a result earns low payoffs. To study the reasons for this excessive investment we propose a treatment in which subjects see everyone's payoffs. We find that in small groups the pure influencer out- come obtains but that in large groups the pure-connector outcome now becomes common, suggesting that information and group size interact in powerful ways to shape networks and payoffs. |
| Date: | 2022–04 |
| URL: | http://d.repec.org/n?u=RePEc:nad:wpaper:20220077&r= |
| By: | Edoardo Gallo; Yohanes E. Riyanto; Nilanjan Roy; Tat-How Teh |
| Abstract: | This paper examines experimentally how reputational uncertainty and the rate of change of the social environment determine cooperation. Reputational uncertainty significantly decreases cooperation, while a fast-changing social environment only causes a second-order qualitative increase in cooperation. At the individual level, reputational uncertainty induces more leniency and forgiveness in imposing network punishment through the link proposal and removal processes, inhibiting the formation of cooperative clusters. However, this effect is significant only in the fast-changing environment and not in the slow-changing environment. A substitution pattern between network punishment and action punishment (retaliatory defection) explains this discrepancy across the two social environments. |
| Date: | 2022–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2203.04001&r= |
| By: | Michael Kastoryano; Nicola Pancotti |
| Abstract: | Risk assessment and in particular derivatives pricing is one of the core areas in computational finance and accounts for a sizeable fraction of the global computing resources of the financial industry. We outline a quantum-inspired algorithm for multi-asset options pricing. The algorithm is based on tensor networks, which have allowed for major conceptual and numerical breakthroughs in quantum many body physics and quantum computation. In the proof-of-concept example explored, the tensor network approach yields several orders of magnitude speedup over vanilla Monte Carlo simulations. We take this as good evidence that the use of tensor network methods holds great promise for alleviating the computation burden of risk evaluation in the financial and other industries, thus potentially lowering the carbon footprint these simulations incur today. |
| Date: | 2022–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2203.02804&r= |
| By: | Tamas David-Barrett |
| Abstract: | Reputation-based cooperation on social networks offers a causal mechanism between graph properties and social trust. Recent papers on the `structural microfoundations` of the society used this insight to show how demographic processes, such as falling fertility, urbanisation, and migration, can alter the logic of human societies. This paper demonstrates the underlying mechanism in a way that is accessible to scientists not specialising in networks. Additionally, the paper shows that, when the size and degree of the network is fixed (i.e., all graphs have the same number of agents, who all have the same number of connections), it is the clustering coefficient that drives differences in how cooperative social networks are. |
| Date: | 2022–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2203.00372&r= |
| By: | Ozan Candogan; Nicole Immorlica; Bar Light; Jerry Anunrojwong |
| Abstract: | Individuals increasingly rely on social networking platforms to form opinions. However, these platforms typically aim to maximize engagement, which may not align with social good. In this paper, we introduce an opinion dynamics model where agents are connected in a social network, and update their opinions based on their neighbors' opinions and on the content shown to them by the platform. We focus on a stochastic block model with two blocks, where the initial opinions of the individuals in different blocks are different. We prove that for large and dense enough networks the trajectory of opinion dynamics in such networks can be approximated well by a simple two-agent system. The latter admits tractable analytical analysis, which we leverage to provide interesting insights into the platform's impact on the social learning outcome in our original two-block model. Specifically, by using our approximation result, we show that agents' opinions approximately converge to some limiting opinion, which is either: consensus, where all agents agree, or persistent disagreement, where agents' opinions differ. We find that when the platform is weak and there is a high number of connections between agents with different initial opinions, a consensus equilibrium is likely. In this case, even if a persistent disagreement equilibrium arises, the polarization in this equilibrium, i.e., the degree of disagreement, is low. When the platform is strong, a persistent disagreement equilibrium is likely and the equilibrium polarization is high. A moderate platform typically leads to a persistent disagreement equilibrium with moderate polarization. Lastly, more balanced and less polarized initial opinions are more likely to lead to persistent disagreement than to consensus. |
| Date: | 2022–02 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2202.12453&r= |
| By: | Benson Tsz Kin Leung |
| Abstract: | This paper analyzes a model of innovation diffusion with case-based individuals a la Gilboa and Schmeidler (1995, 1996, 1997), who decide whether to consume an incumbent's or an entrant's product based on their and their social neighbors' previous consumption experience. I analyze how diffusion pattern changes with individuals characteristics, innovation characteristics and social network. In particular, radical innovation leads to higher initial speed but lower acceleration compared to increment innovation. Social network with lower degree of homophily or higher exposure of reviews from early adopters speed up diffusion of innovation. |
| Date: | 2022–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2203.05785&r= |
| By: | Jase Clarkson; Mihai Cucuringu; Andrew Elliott; Gesine Reinert |
| Abstract: | In this work, we introduce DAMNETS, a deep generative model for Markovian network time series. Time series of networks are found in many fields such as trade or payment networks in economics, contact networks in epidemiology or social media posts over time. Generative models of such data are useful for Monte-Carlo estimation and data set expansion, which is of interest for both data privacy and model fitting. Using recent ideas from the Graph Neural Network (GNN) literature, we introduce a novel GNN encoder-decoder structure in which an encoder GNN learns a latent representation of the input graph, and a decoder GNN uses this representation to simulate the network dynamics. We show using synthetic data sets that DAMNETS can replicate features of network topology across time observed in the real world, such as changing community structure and preferential attachment. DAMNETS outperforms competing methods on all of our measures of sample quality over several real and synthetic data sets. |
| Date: | 2022–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2203.15009&r= |
| By: | Federico Forte; Pedro Elosegui; Gabriel Montes-Rojas |
| Abstract: | This paper studies the network structure and fragmentation of the Argentinean interbank market. Both the unsecured (CALL) and the secured (REPO) markets are examined, applying complex network analysis. Results indicate that, although the secured market has less participants, its nodes are more densely connected than in the unsecured market. The interrelationships in the unsecured market are less stable, making its structure more volatile and vulnerable to negative shocks. The analysis identifies two 'hidden' underlying sub-networks within the REPO market: one based on the transactions collateralized by Treasury bonds (REPO-T) and other based on the operations collateralized by Central Bank (CB) securities (REPO-CB). The changes in monetary policy stance and monetary conditions seem to have a substantially smaller impact in the former than in the latter 'sub-market'. The connectivity levels within the REPO-T market and its structure remain relatively unaffected by the (in some period pronounced) swings in the other segment of the market. Hence, the REPO market shows signs of fragmentation in its inner structure, according to the type of collateral asset involved in the transactions, so the average REPO interest rate reflects the interplay between these two partially fragmented sub-markets. This mixed structure of the REPO market entails one of the main sources of differentiation with respect to the CALL market. |
| Date: | 2022–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2203.14488&r= |
| By: | Jan Schulz; Daniel M. Mayerhoffer; Anna Gebhard |
| Abstract: | Across income groups and countries, individual citizens perceive economic inequality spectacularly wrong. These misperceptions have far-reaching consequences, as it is perceived inequality, not actualinequality informing redistributive preferences. The prevalence of this phenomenon is independent of social class and welfare regime, which suggests the existence of a common mechanism behind public perceptions. The literature has identified several stylised facts on how individual perceptions respond to actual inequality and how these biases vary systematically along the income distribution. We propose a network-based explanation of perceived inequality building on recent advances in random geometric graph theory. The generating mechanism can replicate all of forementioned stylised facts simultaneously. It also produces social networks that exhibit salient features of real-world networks; namely, they cannot be statistically distinguished from small-world networks, testifying to the robustness of our approach. Our results, therefore, suggest that homophilic segregation is a promising candidate to explain inequality perceptions with strong implications for theories of consumption and voting behaviour. |
| Date: | 2022–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2203.14254&r= |
| By: | David William Hughes |
| Abstract: | This paper considers estimation of a directed network model in which outcomes are driven by dyad-specific variables (such as measures of homophily) as well as unobserved agent-specific parameters that capture degree heterogeneity. I develop a jackknife bias correction to deal with the incidental parameters problem that arises from fixed effect estimation of the model. In contrast to previous proposals, the jackknife approach is easily adaptable to different models and allows for non-binary outcome variables. Additionally, since the jackknife estimates all parameters in the model, including fixed effects, it allows researchers to construct estimates of average effects and counterfactual outcomes. I also show how the jackknife can be used to bias-correct fixed effect averages over functions that depend on multiple nodes, e.g. triads or tetrads in the network. As an example, I implement specification tests for dependence across dyads, such as reciprocity or transitivity. Finally, I demonstrate the usefulness of the estimator in an application to a gravity model for import/export relationships across countries. |
| Date: | 2022–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2203.15603&r= |
| By: | Alem, Yonas; Dugoua, Eugenie |
| Abstract: | Interactions among peers of the same social network play significant roles in facilitating the adoption and diffusion of modern technologies in poor communities. We conduct a large-scale randomized controlled trial in rural India to identify the impact of information from friends on willingness to pay (WTP) for high-quality and multi-purpose solar lanterns. We offered solar lanterns to seed households from 200 non-electrified villages and randomly assigned three of their friends to two communication treatments (unincentivized and incentivized) that led to different exposure to their seed friend. We also introduce a second treatment to investigate whether the seed’s gender impacts the magnitude of peer effects. We show that unincentivized communication increases WTP for solar lanterns by 90% and incentivized communication by 145%, but gender doesn’t seem to matter. We also show that learning from others is the mechanism that drives the increase in WTP. Our findings have significant implications for policies that aim at promoting the diffusion of new technologies in developing countries. JEL: O33, D83, Q21, Q42 Keywords: Technology Adoption; Peer Effects; Information Flow; Solar Lante |
| Keywords: | technology adoption; peer effects; information flow; solar lantern |
| JEL: | O33 D83 Q21 Q42 |
| Date: | 2021–04–02 |
| URL: | http://d.repec.org/n?u=RePEc:ehl:lserod:110858&r= |
| By: | Sudo Yi; Deok-Sun Lee |
| Abstract: | We study the structure of the international trade hypergraph consisting of triangular hyperedges representing the exporter-importer-product relationship. Measuring the mean hyperdegree of the adjacent vertices, we first find its behaviors different from those in the pairwise networks and explain the origin by tracing the relation between the hyperdegree and the pairwise degree. To interpret the observed hyperdegree correlation properties in the context of trade strategies, we decompose the correlation into two components by identifying one with the background correlation remnant even in the exponential random hypergraphs preserving the given empirical hyperdegree sequence. The other component characterizes the net correlation and reveals the bias of the exporters of low hyperdegree towards the importers of high hyperdegree and the products of low hyperdegree, which information is not readily accessible in the pairwise networks. Our study demonstrates the power of the hypergraph approach in the study of real-world complex systems and offers a theoretical framework. |
| Date: | 2022–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2203.05762&r= |
| By: | Nathan Canen; Ko Sugiura |
| Abstract: | When using dyadic data (i.e., data indexed by pairs of units, such as trade flow data between two countries), researchers typically assume a linear model, estimate it using Ordinary Least Squares and conduct inference using "dyadic-robust" variance estimators. The latter assumes that dyads are uncorrelated if they do not share a common unit (e.g., if one country does not appear in both pairs of trade flow data). We show that this assumption does not hold in many empirical applications because indirect links may exist due to network connections, e.g., different country-pairs may have correlated trade outcomes due to sharing common trading partner links. Hence, as we prove, then show in Monte Carlo simulations, "dyadic-robust" estimators can be severely biased. We develop a consistent variance estimator appropriate for such contexts by leveraging results in network econometrics. Our estimator has good finite sample properties in numerical simulations. We then illustrate our message with an application to voting behavior by seating neighbors in the European Parliament. |
| Date: | 2022–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2203.03497&r= |
| By: | Thomas J. Sargent; John Stachurski |
| Abstract: | This textbook is an introduction to economic networks, intended for students and researchers in the fields of economics and applied mathematics. The textbook emphasizes quantitative modeling, with the main underlying tools being graph theory, linear algebra, fixed point theory and programming. The text is suitable for a one-semester course, taught either to advanced undergraduate students who are comfortable with linear algebra or to beginning graduate students. |
| Date: | 2022–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2203.11972&r= |
| By: | Jan Schulz; Daniel M. Mayerhoffer |
| Abstract: | The nexus between debt and inequality has attracted considerable scholarly attention in the wake of the global financial crisis. One prominent candidate to explain the striking co-evolution of income inequality and private debt in this period has been the theory of upward-looking consumption externalities leading to expenditure cascades. We propose a parsimonious model of upward-looking consumption at the micro level mediated by perception networks with empirically plausible topologies. This allows us to make sense of the ambiguous empirical literature on the relevance of this channel. Up to our knowledge, our approach is the first to make the reference group to which conspicuous consumption relates explicit. Our model, based purely on current income, replicates the major stylised facts regarding micro consumption behaviour and is thus observationally equivalent to the workhorse permanent income hypothesis, without facing its dual problem of `excess smoothness' and `excess sensitivity'. We also demonstrate that the network topology and segregation has a significant effect on consumption patterns which has so far been neglected. |
| Date: | 2022–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2203.14259&r= |
| By: | Harshil Sahai; Mike Bailey |
| Abstract: | This paper studies the role of social networks in spatial mobility across India. Using aggregated and de-identified data from the world's largest online social network, we (i) document new descriptive findings on the structure of social networks and spatial mobility in India; (ii) quantify the effects of social networks on annual migration choice; and (iii) embed these estimates in a spatial equilibrium model to study the wage implications of increasing social connectedness. Across millions of individuals, we find that multiple measures of social capital are concentrated among the rich and educated and among migrants. Across destinations, both mobility patterns and social networks are concentrated toward richer areas. A model of migration suggests individuals are indifferent between a 10% increase in destination wages and a 12-16% increase in destination social networks. Accounting for networks reduces the migration-distance relationship by 19%. In equilibrium, equalizing social networks across locations improves average wages by 3% (24% for the bottom wage-quartile), a larger impact than removing the marginal cost of distance. We find evidence of an economic support mechanism, with destination economic improvements reducing the migration-network elasticity. We also find suggestive evidence for an emotional support mechanism from qualitative surveys among Facebook users. Difference-in-difference estimates suggest college attendance delivers a 20% increase in network size and diversity. Taken together, our data suggest that - by reducing effective moving costs - increasing social connectedness across space may have considerable economic gains. |
| Date: | 2022–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2203.05595&r= |
| By: | R. Vilela Mendes; Tanya Araújo |
| Abstract: | Networks with long-range connections, obeying a distance-dependent power law of sufficiently small exponent, display superdiffusion, L´evy flights and robustness properties very different from the scale-free networks. It has been proposed that these networks, found both in society and in biology, be classified as a new structure, the fractional networks. Particular important examples are the social networks and the modular hierarchical brain networks where both short- and long-range connections are present. The anomalous superdiffusive and the mixed diffusion behavior of these networks is studied here as well as its relation to the nature and density of the long-range connections. |
| Date: | 2022–03 |
| URL: | http://d.repec.org/n?u=RePEc:ise:remwps:wp02232022&r= |