Research
For my research statement, please click here.
WORKING PAPERS
This paper studies learning through social networks, where agents update their beliefs by weighting those of their peers. We introduce dynamically updated weights, allowing agents to pay little attention to contacts with poor information at first, but more later on, if the latter have in the meantime acquired better information from more knowledgeable agents. We derive explicitly how social influence depends on agents’ popularity (eigenvector centrality) and expertise (information precision). We further show that even completely uninformed agents can contribute to social learning, and that providing better information to certain agents may lead society to worse assessments
Conformism under incomplete information
|
We develop a simple Bayesian network game in which players, embedded in a network of
social interactions, bear a cost from deviating from the social norm of their peers. All agents
face uncertainty about the private benefits and the private and social costs of their actions. We
prove the existence and uniqueness of a Bayesian Nash equilibrium and characterize players’
optimal actions. We then show that denser networks do not necessary increase agents’ actions
and welfare. We also find that, in some cases, it is optimal for the planner to affect the payoffs
of selected individuals rather than all agents in the network. We finally show that having
more information is not always beneficial to agents and can, in fact, reduce their welfare. We
illustrate all our results in the context of criminal networks in which offenders do not know
with certitude the probability of being caught and do not want to be different from their
peers in terms of criminal activities.
social interactions, bear a cost from deviating from the social norm of their peers. All agents
face uncertainty about the private benefits and the private and social costs of their actions. We
prove the existence and uniqueness of a Bayesian Nash equilibrium and characterize players’
optimal actions. We then show that denser networks do not necessary increase agents’ actions
and welfare. We also find that, in some cases, it is optimal for the planner to affect the payoffs
of selected individuals rather than all agents in the network. We finally show that having
more information is not always beneficial to agents and can, in fact, reduce their welfare. We
illustrate all our results in the context of criminal networks in which offenders do not know
with certitude the probability of being caught and do not want to be different from their
peers in terms of criminal activities.
Nudges, networks and social preferences
with Ranjula Bali Swain (Stockholm Sch of Econ), Erik Gråd (Södertörn U), and Shyam Ranganathan (Virginia Tech)
Nudges are interventions that steer individuals’ decisions without reducing their choice set or imposing additional costs on them. This paper proposes a general model of behavioural change in situations that can be classified as public goods games. We develop a model that takes social preferences into account, and examines how individual choices are affected by various types of nudges (informative, emotional, and social), and combinations thereof. We find conditions under which society converges to a uniform behaviour, and introduce a new centrality measure to identify the individuals that are more influential in inducing pro-social behaviour to their peers. Based on the proposed framework, experiments aimed at providing a deeper understanding of what drives individuals’ decisions related to the provision of public goods can be designed.
Recent research suggests that interpersonal and social networks play an important role in shaping political opinions. This paper studies the evolution of political beliefs using a network model of social learning. Agents communicate their information, discuss their opinions with their peers, and update their beliefs accordingly. Although information originating from better-informed agents receives ceteris paribus a larger weight, individuals filter incoming information based on their political ideology. The paper also studies networks with individuals or groups of individuals who are not interested in learning or exchanging of information, but rather in promoting their own views to other agents. The features that make such groups influential are identified and discussed.
WORK IN PROGRESS
Debt relief and moral hazard: estimating the effect of bankruptcy protection on loan repayment
with Theresa Kuchler (NYU), Sharada Shridhar (NYU), and Constantine Yannelis (U Chicago)
Personal bankruptcy provides households with insurance against adverse financial shocks, but at the same time induces moral hazard by alleviating the consequences of non-repayment. Taking advantage of the introduction of a comprehensive bankruptcy protection framework in Greece, we use a large data set of individual mortgages to estimate the effect of the new legislation on the loan repayment rate, and study its effect on repayment costs.
Political competition and parties' ability to influence their candidates' platforms
In primary elections, candidates running for their party's nomination face a dilemma: notwithstanding their personal views, appearing more partisan would foster their chances of getting nominated but could at the same time alienate moderate or independent voters in the forthcoming general election. In this paper it is shown that even within a probabilistic voting setting, which allows candidates some room for manoeuvring, primaries remain a quite effective mechanism for parties to discipline their prospective nominees on both sides of the party's ideological spectrum. This, however, is not always true for candidates with extremely radical views, who may hold on to their position, at the price of a reduced chance of winning the nomination.