Download a PDF of my research statement: Click Here
Download a PDF of my teaching statement: Click Here
"Study the optimal targeting of information by a Designer who wants to persuade a set of agents connected in a social network."
I study information diffusion in a social network where a third party can control the precision of information as well as who initially receives information. Applications include spreading of news by digital media outlets, lobbyists persuading senators to contribute to a project and others. A designer engages in a bayesian persuasion game with multiple agents, but is constrained to send information privately to a subset of agents. The agents can communicate this information to each other through links in a social network. The designer relies on this word-of-mouth com- munication channel to diffuse information. The optimal precision of information sent by the designer reflects a fundamental tradeoff: precise information increases diffusion, but reduces the designer’s ability to manipulate the agents’ beliefs to- wards his objective. The optimal seeding strategy involves choosing an agent with the highest “influence”- related to (and a generalization of) Banerjee et al. (2013)’s diffusion centrality. As preferences become more diverse, I show that there exists equilibrium with sub- optimal spreading where the designer caters only to agents relatively more aligned towards his objective.
"Study an augmented-SIR model with fargsighted strategic agents and government."
The canonical epidemiology model of disease dynamics is considered where agents can be in one of five states: susceptible, infected, hospitalized, recovered or dead. This is enriched by allowing agents to social distance from others, and the government to impose lockdowns and trace-test the population with imperfect efficacy. Each of these forces interact with each other to produce a rich set of dynamics that are not captured by the canonical model. The pairwise substitutability between lockdown, social distancing and tracing-testing is explored. Agents’ choice to social distance and the governments’ choice to lockdown are characterized by a novel set of conditions that offer some insights into the trade-offs faced by them. The predictions are then fit to the US data on the evolution of Covid-19 to calibrate the key parameters of prevalence, tracing-testing, and behavior. The forward-backward sweep method from optimal control finds the optimal policy which is then numerically implemented using the particle swarm algorithm. Finally, inspired from the varied experiences of countries around the world, four policy experiments are conducted: increased testing efficiency, no lockdown at all, increased government control, and making agents myopic.
Work In Progress
"An adversary aims to send an object to a target through a network, and nodes along the path can conduct costly checks to intercept the item."
We present a model of network security where an adversary aims to send a (potentially destructive) object to a target via one of multiple paths. Along each path, there are intermediaries who can periodically involve in costly checks to intercept and destroy the item. The adversary wins if the object reaches the target, and the target (along with the intermediaries) wins if the object is intercepted before reaching the target. We characterize the equilibrium of the game and discuss the welfare properties in terms of the network structure. We provide a mechanism design approach to design the least-cost security system to minimize the loss to the economy. (To be updated)
"Consumers in a social network suffer from data externality, and a data-buyer aims to purchase this data while maintaining privacy concerns."
(To be updated)
"Determine the key factors behind India's growth story in the 1990s using the novel KLEMS dataset."
The dynamics of Economic Growth in India continues to engage economists and still remains an enigma. The trends and patterns of growth observed in India have seen acceleration in growth in Indian economy in the period following macroeconomic reforms and policy changes in investment and trade regimes. However, when and how did India transformation from itself Hindu rate of growth to the present growth regime continues to be debated. The present study using INDIA KLEMS dataset provides a highly novel, very valuable, and quite distinctive perspective on India’s economic growth. A unique dataset comprising 27 sectors of Indian economy at a disaggregate industry level for a period of 30 years attempts to understand the dynamics of India’s growth from the contribution of industries that comprise the Indian economy. This DATA SET offers a new way of analyzing the dynamics of growth including the sources of growth. Our growth empirics allow evaluation of the relative significance of multi factor productivity growth Vis-a-Vis input accumulation in accounting for output growth. In addition, we are able to document the industry contributions to aggregate growth. In this way, we are able to analyze how important are different sectors of the economy- agriculture, manufacturing, construction, market as well as non-market services in accounting for the observed growth in India.
"Study the impact of a rehabiltation project on costs and profits for artisan communities in Kolkata.."
This paper aims at adding to the burgeoning literature on the informal sector in de- veloping countries. Kumartuli is the traditional home of the artisans who work round the year to supply idols for the various Hindu pujas. Plagued with general ignorance and a lack of development, this art form is slowly dying out. In 2007, the government took a baby step towards the development of Kumartuli by sanctioning a rehabilitation project that would see each artisan obtain a permanent concrete studio to work in. Till today, the only steps that have been taken is the make-shift relocation of some artisans so that land for the rehabiliation project could be acquired. However, we find that the artists are stuck in a historically conditioned equilibrium whereby the artists do not want to leave their old studios and shift to the relocated region. This prevents rehabilitation to start in the first place. In this paper we try to obtain the welfare implication of the rehabilitation project in an ex ante sense. We use data from a census survey on the 110 artisans present in Kumartuli to obtain details of the components of revenue and costs for these artisans. We show that while the rehabilitation would benefit the worst off artists,there are limits to this benefit. We further try to ascertain the long term prospects of the profession with the help of an accompanying attitudinal survey conducted by us.