Working Papers
When Competition Compels Change: Trade, Management, and Productivity </span>
- ▷ Abstract
Competition reallocates resources toward more productive firms, raising aggregate productivity. In this paper, I show that competition also operates within firms, inducing organizational change that raises managerial efficiency. I exploit a product-specific import competition shock in India, assembling novel data on family-managed firms—the predominant form of corporate governance worldwide—with tenure records and family ties for over 6 million company executives and directors. Using an event-study design, I show that exposed firms respond to import competition by replacing family managers with non-family, professional executives. Firms that professionalize experience productivity gains of over 30 percent. To quantify the aggregate implications of these managerial adjustments, I develop a quantitative model embedding a management choice within a model of heterogeneous firms with monopolistic competition. Firms trade off the non-pecuniary private benefits associated with family management against the profit gains from professionalization. Using my model, I estimate that import liberalization increased aggregate productivity in India by 12 percent, with within-firm improvements in managerial allocation explaining nearly 30 percent of this increase.
We study the micro sources and macro consequences of worker-job matching across countries with large income differences. Using internationally comparable data on over 120,000 individuals in 30 countries, we document that workers' skills align more closely with their jobs' skill requirements in higher-income countries, indicative of more meritocratic labor market matching. We interpret this fact through an equilibrium matching model with cross-country differences in three fundamentals: (i) endowments of worker skills and job requirements determining match feasibility; (ii) technology determining the returns to matching; and (iii) idiosyncratic frictions capturing how nonproductive traits affect matching. A development accounting exercise based on the model, estimated separately for each country, shows that variation in matching frictions explains only a small share of cross-country output gaps. However, improved worker-job matching substantially amplifies the gains from adopting frontier endowments and technology.
Work in Progress
Management in India
- ▷ Abstract
We study how family control and professionalization shape firm performance and organization using nationwide administrative records on over 350,000 registered Indian firms spanning the 1980s to the present, including director genealogies and tenure histories. We construct firm-year measures of family presence on boards and deploy event-study designs around leadership transitions and shifts toward professional boards. Supermajority family boards are systematically smaller and less productive: revenues, profits, and exporting are lower when more than three quarters of directors are family, relative to otherwise similar firms. Dynamic estimates around patriarch exit show post-transition improvements in operating outcomes and profitability ratios. Together, the evidence maps who professionalizes, when, and with what consequences for scale, productivity, and managerial practices in Indian firms.
Improving Tax Administration via AI: Randomized Evaluation of a Decision Support Tool
- ▷ Abstract
Legal decision makers in tax administration face heavy caseloads and large, heterogeneous case records. We design and deploy an AI decision support system that complements human judgment by summarizing case histories, surfacing the most relevant precedents with citations, and estimating the probability that a further appeal would succeed. The tool will be embedded in the existing workflow of quasi-judicial tax officers in India and rolled out using a randomized evaluation. We will measure impacts on decision quality, consistency, and efficiency, and examine how access to probability estimates shifts appeals toward cases with stronger legal merit. Primary outcomes include higher-court affirmation rates, alignment with precedent, processing time, backlog, and appeal selection.
AI and Judicial State Capacity
- ▷ Abstract
Access to timely justice is a cornerstone of state capacity, yet judicial systems across the Global South face severe administrative bottlenecks and chronic backlogs- in India alone, it would take nearly 300 years to clear the backlog of over 50 million pending cases. This study evaluates whether artificial intelligence can meaningfully strengthen the state's judicial capacity through one of the largest randomized controlled trials ever conducted in courts. We partner with over 1,500 courtrooms across Andhra Pradesh and Odisha in India to introduce an AI platform that provides real-time transcription, translation, and automated case-flow management for judges and court staff. The experiment measures how workflow automation affects judicial productivity, hearing duration, and case resolution rates, using daily administrative and survey data on court proceedings and judgments. The study will provide the first causal evidence on how frontier AI technologies can improve bureaucratic efficiency in core state institutions, enhance decision quality, and inform the design of scalable digital infrastructure for justice delivery in low- and middle-income countries.
How Much Do Firms Save? Financial Frictions and the Microeconomic Implications of the Euler Equation
- ▷ Abstract
Neoclassical growth models with standard parameter values provide powerful self-financing incentives to financially constrained entrepreneurs. This fundamental prediction has important implications for capital misallocation. If entrepreneurs can indeed save themselves out of financial constraints, capital misallocation should disappear on its own and impose a small, transient cost to aggregate productivity. This is contrary to a large body of empirical work that has documented high dispersion in marginal products of resources, particularly capital. Using micro firm-level data and the staggered implementation of a financial liberalization policy in India, I provide causal evidence on the relationship between financial constraints and self-financing by individual firms. I find that the behavior of treated firms that see an easing in their financial constraints does not conform with standard predictions of the neoclassical growth model.
Labor Regulations, Firm Dynamics and Sectoral Reallocation: Evidence from India
- ▷ Abstract