I'm an Assistant Professor of Management at the Wharton School at the University of Pennsylvania. My research focuses on entrepreneurship and human capital, with a particular focus on emerging economies. My work draws on a variety of methodologies, with a focus on new computational methods and text as data. I received my Ph.D. at Columbia Business School.
The emergence of large language models (LLMs) has opened new avenues for integrating artificial intelligence into research, particularly for data annotation and text classification. However, the benefits and risks of using LLMs in research remain poorly understood, such that researchers lack guidance on how best to implement this tool. We address this gap by developing a foundational framework for implementing LLMs for annotation in management research, providing structured guidance on key implementation decisions and best practices. We illustrate the implementation of this framework through an empirical application: classifying sustainability claims in crowdfunding projects to assess the performance relationships of these claims. We demonstrate that while LLMs can match or exceed traditional methods' performance at lower cost, variations in prompt design can significantly affect results and downstream analyses. We thus develop procedures for sensitivity analysis and provide documentation to help researchers implement these robustness checks while maintaining methodological integrity.
Small unregistered firms contribute to a substantial proportion of global economic activity, particularly in developing regions. In explaining variation in productivity in these types of informal firms, research has focused primarily on the adoption of effective business practices and access to capital, with little focus on fundamental positioning. This article explores the nature of differentiation in microenterprises, introducing a text‐based measure of differentiation using state‐of‐the‐art sentence embeddings. Using a combined sample of nearly 10,000 microenterprises across eight developing countries, I examine whether (and which) microenterprises differentiate, whether differentiation is related to performance (and for whom), and whether any existing policy interventions affect differentiation.
We demonstrate how a novel synthesis of three methods—(a) unsupervised topic modeling of text data to generate new measures of textual variance, (b) sentiment analysis of text data, and (c) supervised ML coding of facial images with a cutting-edge convolutional neural network algorithm—can shed light on questions related to CEO oral communication. With videos and corresponding transcripts of interviews with emerging market CEOs, we use this synthesis of methods to discover five distinct communication styles that incorporate both verbal and nonverbal aspects of communication. Our data comprises interviews that represent unedited expressions and content, making them especially suitable as data sources for the measurement of an individual's communication style. We then perform a proof-of-concept analysis, correlating CEO communication styles to M&A outcomes, highlighting the value of combining text and videographic data to define styles. We also discuss the benefits of using our methods versus current research methods.
Support organizations in under-resourced settings often prescribe “best practices,” yet firms facing severe resource constraints frequently struggle to adopt them -- or to benefit when they do. We argue that practices differ in the reliability of their returns under constraint: some are reliable, generating consistent value even in low-resource settings, whereas others are contingent, yielding gains only when firms can meet underlying resource or coordination requirements. We examine how external support shapes both practice adoption and productivity in such contexts. Using data from 1,480 smallholder coconut farmers in the Philippines, we show that several "Good Agricultural Practices'' (GAPs) deliver negligible or even negative returns for the most constrained farms. Access to Philippine Coconut Authority extension services helps farmers navigate this challenge by steering adoption toward reliable GAPs and away from contingent ones. Benefits may therefore arise from judicious subtraction as well as addition. The study reframes capability development in poverty settings as a portfolio-optimization problem, and positions support organizations as critical brokers of that optimization.
We study a randomized controlled trial of an entrepreneurial training program in Zimbabwe that successfully converted many individuals from itinerant self-employment and casual labor into full-time business owners, generating significant business growth and higher incomes for treated individuals. We find that the treatment additionally increased participants' rates of civic engagement, with treated individuals being substantially more likely to contact political leaders across all levels of society with grievances (including traditional leaders, government agencies, local councilors, and MPs). However, none of our hypothesized mechanisms, such as networks, income, or empowerment, adequately explain this result. Instead, exploratory analyses reveal that the effect is broadly mediated by entrepreneurship itself -- specifically, treated individuals' attempts to grow their business, including hiring employees, taking out loans, and opening business accounts. These steps toward business growth appear to have exposed entrepreneurs to a broader array of institutional failures, leading them to voice more complaints about challenges such as corruption, cash shortages, and unfair loan terms. We conclude that policies designed to promote entrepreneurship may generate the unintended consequence of fostering greater civic engagement, as entrepreneurs encounter and respond to institutional constraints that impede their business activities. Our work suggests that even the smallest businesses engage in forms of grassroots lobbying.
This paper examines how skilled immigrants hired at headquarters (HQ), an increasingly important source of talent for multinational firms, shape two key aspects of global product strategy: where to offer products and the degree of standardization versus local adaptation of those products. We assemble a unique dataset by linking 71,527 consumer packaged goods launched in 84 countries by 337 U.S.-based multinationals from Mintel GNDP to USCIS microdata on all H-1B workers hired by those firms. We leverage these data to develop an instrument that isolates random variation in firm's ability to obtain workers of specific nationalities and use NLP techniques to create a novel adaptation-standardization index at the product level. We find that hiring workers from a certain country at HQ causally increases product launches in that country. Whether those products become more standardized or localized depends on the existence of a foreign subsidiary. Immigrants shift products toward greater standardization when a firm has a subsidiary in the focal country. When a subsidiary is not present, immigrants cause products to be more locally adapted. Our study highlights the critical role of HQ talent in shaping global strategy.