Person:

Choudhury, Prithwiraj

The use of machine learning (ML) for productivity in the knowledge economy requires considerations of important biases that may arise from ML predictions. We define a new source of bias related to incompleteness in real time inputs, which may result from strategic behavior by agents. We theorize that domain expertise of users can complement ML by mitigating this bias. Our observational and experimental analyses in the patent examination context support this conjecture. In the face of “input incompleteness,” we find ML is biased towards finding prior art textually similar to focal claims, and domain expertise is needed to find the most relevant prior art. We also document the importance of vintage-specific skills and discuss the implications for artificial intelligence and strategic management of human capital.

We study migration in the right tail of the talent distribution using a novel dataset of Indian high school students taking the Joint Entrance Exam (JEE), a college entrance exam used for admission to the prestigious Indian Institutes of Technology (IIT). We find a high incidence of migration after students complete college: among the top 1,000 scorers on the exam, 36% have migrated abroad, rising to 62% for the top 100 scorers. We next document that students who attended the original “Top 5” Indian Institutes of Technology (IIT) were 5 percentage points more likely to migrate for graduate school compared to equally talented students who studied in other institutions. We explore two mechanisms for these patterns: signaling, for which we study migration after one university suddenly gained the IIT designation; and alumni networks, using information on the location of IIT alumni in U.S. computer science departments.

We exploit plausibly exogenous variation in the staggered entry of new managers into India’s 42 public R&D labs between 1994 and 2006 to study how alignment between the CEO and middle-level managers affect research productivity. We show that the introduction of new lab managers aligned with the national R&D reforms raised patenting and multinational licensing revenues by 58% and 75%, respectively, and scientist research productivity, including a 16%, 10%, 11%, and 22% increase in h-indices, number of coauthors, publications, and citations per scientist, respectively. Using natural language processing (NLP) techniques on the set of research abstracts produced among these scientists, we also find that overall mood and sentiment increased by 8.5% following the first managerial change.

Past research offers mixed perspectives on whether domain experience helps or hurts algorithm-augmented worker performance. Reconciling these perspectives, we theorize that intermediate levels of domain experience are optimal for algorithm-augmented performance, due to the interplay between two countervailing forces—ability and aversion. Although domain experience can increase performance via increased ability to complement algorithmic advice (e.g., identifying inaccurate predictions), it can also decrease performance via increased aversion to accurate algorithmic advice. Because ability developed through learning by doing increases at a decreasing rate, and algorithmic aversion is more prevalent among experts, we theorize that algorithm-augmented performance will first rise with increasing domain experience, then fall. We test this by exploiting a within-subjects experiment in which corporate information technology support workers were assigned to resolve problems both manually and using an algorithmic tool. We confirm that the difference between performance with the algorithmic tool versus without the tool was characterized by an inverted U-shape over the range of domain experience. Only workers with moderate domain experience did significantly better using the algorithm than resolving tickets manually. These findings highlight that, even if greater domain experience increases workers’ ability to complement algorithms, domain experience can also trigger other mechanisms that overcome the positive ability effect and inhibit performance. Additional analyses and participant interviews suggest that, even though the highest experience workers had the greatest ability to complement the algorithmic tool, they rejected its advice because they felt greater accountability for possible unintended consequences of accepting algorithmic advice.

We study whether, when, and how better connectivity through nonstop flights leads to positive innovation outcomes for firms in the global context. Using unique data of all flights emanating from 5,015 airports around the globe from 2005 to 2015 and exploiting a regression discontinuity framework, we report that a 10% increase in nonstop flights between two locations leads to a 3.4% increase in citations and a 1.4% increase in the production of collaborative patents between those locations. This effect is driven primarily by firms as opposed to academic institutions. We further study the characteristics of firms and firm locations that are salient to the relation between nonstop flights and innovation outcomes across countries. Using a gravity model, we posit and find that the positive effect of nonstop flights on innovation is stronger for firms and subsidiaries with greater innovation mass (e.g., stocks of inventors and R&D spending), located in innovation hubs or countries that are deemed technology leaders, and that are separated by large cultural or temporal distance. This paper was accepted by Alfonso Gambardella, business strategy. Supplemental Material: The online appendix and data are available at https://doi.org/10.1287/mnsc.2023.4682 .

The number of Global Mobile Inventors (GMIs), inventors moving across borders during their career, has increased more than tenfold over the past two decades, and the corridors of mobility have shifted towards a growing presence of emerging markets. We document that GMIs that have patented in a given technology before moving are 70% more likely to be among the pioneering inventors in that technology once they arrive at destination, which we interpret as evidence of knowledge diffusion across borders. Returnees, which are typically inventors from emerging markets that go back after having spent some time in the US and other advanced economies, are twice as likely to file pioneering patents once returned than migrants when arriving abroad. Finally, we find that the more central the GMIs in the network of inventors during the early stages of the technology life-cycle at destination, the faster the technology-specific knowledge is absorbed by local inventors.

This paper argues that intra-firm geographic mobility is an understudied mechanism that can help mitigate coordination failures in a geographically distributed organization. The paper presents an organizing framework on how intra-firm geographic mobility creates value for firms and discusses how intra-firm geographic mobility can create value for individual workers. The paper concludes by presenting a future research agenda for intra-firm geographic mobility in light of emerging phenomena such as global collaborative patenting by multinationals, temporary colocation of knowledge workers, and nonstandard work.

We investigate the relationship between the presence of migrant inventors and the dynamics of innovation in the migrants’ receiving countries. We find that countries are 25 to 60 percent more likely to gain advantage in patenting in certain technologies given a twofold increase in the number of foreign inventors from other nations that specialize in those same technologies. For the average country in our sample, this number corresponds to only 25 inventors and a standard deviation of 135. We deal with endogeneity concerns by using historical migration networks to instrument for stocks of migrant inventors. Our results generalize the evidence of previous studies that show how migrant inventors "import" knowledge from their home countries, which translates into higher patenting in the receiving countries. We interpret these results as tangible evidence of migrants facilitating the technology-specific diffusion of knowledge across nations.

Firm-induced migration typically entails firms relocating workers to fill value-creating positions at destination locations. But such relocated workers are often exposed to external employment opportunities at their destinations, possibly triggering turnover. We conceptualize the firm-induced migration path, consisting of the relocated workers’ place of origin and destination, as relevant in determining worker performance and turnover postrelocation. Using a unique data set from a large Indian technology firm that hires talent from both large cities and smaller towns, we document robust econometric patterns by exploiting the firm’s randomized assignment of workers to production centers across the country. These production centers are located in the largest technology cluster in India (Bangalore), smaller technology clusters, and noncluster locations. We find that the firm-induced migration path shapes both worker performance and turnover. Compared with workers from large cities, workers from smaller towns achieve higher performance when relocated to Bangalore than to other production centers, but are also more likely to join competing firms. Fine-grained data on employment and human-capital-augmentation opportunities at workers’ destination locations, and on socioeconomic conditions in workers’ places of origin, help us rule in an abductive explanation: across firm-induced migration paths, differences in external labor-market opportunities between workers’ places of origin and their destinations, as well as intrafirm skill-development opportunities at the destination, are related to heterogeneous human-capital outcomes. Supplemental Material: The e-companion and data files are available at https://doi.org/10.1287/mnsc.2022.4361.

Supervised machine learning (ML) methods are a powerful toolkit for discovering robust patterns in quantitative data. The patterns identified by ML could be used for exploratory inductive or abductive research, or for post-hoc analysis of regression results to detect patterns that may have gone unnoticed. However, ML models should not be treated as the result of a deductive causal test. To demonstrate the application of ML for pattern discovery, we implement ML algorithms to study employee turnover at a large technology company. We interpret the relationships between variables using partial dependence plots, which uncover surprising nonlinear and interdependent patterns between variables that may have gone unnoticed using traditional methods. To guide readers evaluating ML for pattern discovery, we provide guidance for evaluating model performance, highlight human decisions in the process, and warn of common misinterpretation pitfalls. An online appendix provides code and data to implement the algorithms demonstrated in the paper.