More from the Author

• Article | Strategic Management Journal | February 2019

  The Ethnic Migrant Inventor Effect: Codification and Recombination of Knowledge Across Borders

  Prithwiraj Choudhury and Do Yoon Kim

  Ethnic migrant inventors may differ from locals in terms of the knowledge they bring to host firms. We study the role of first-generation ethnic migrant inventors in cross-border transfer of knowledge previously locked within the cultural context of their home regions. Using a unique dataset of Chinese and Indian herbal patents filed in the United States, we find that an increase in the supply of first-generation ethnic migrant inventors increases the rate of codification of herbal knowledge at U.S. assignees by 4.5%. Our identification comes from an exogenous shock to the quota of H1B visas and from a list of entities exempted from the shock. We also find that ethnic migrant inventors are more likely to engage in reuse of their prior knowledge, whereas knowledge recombination is more likely to be pursued by teams comprising inventors from other ethnic backgrounds.

  Keywords: skilled migration; ethnic migration; first-generation migrant; cultural context; knowledge flows; knowledge reuse; knowledge recombination; recombinant creation; H1B visas; Knowledge Sharing; Knowledge Use and Leverage; Knowledge Dissemination; Immigration; Ethnicity; Cross-Cultural and Cross-Border Issues;

  Citation:

  Choudhury, Prithwiraj, and Do Yoon Kim. "The Ethnic Migrant Inventor Effect: Codification and Recombination of Knowledge Across Borders." Strategic Management Journal 40, no. 2 (February 2019): 203–229.  View Details

  CiteView DetailsFind at Harvard Read Now

  Related

• Working Paper | HBS Working Paper Series | 2018

  Work from Anywhere or Co-locate? Autonomy versus Learning Effects at the United States Patent Office

  Prithwiraj Choudhury, Cirrus Foroughi and Barbara Larson

  While employees might prefer work arrangements that offer greater autonomy, such as work from anywhere (WFA) policies, there are possible negative productivity effects of WFA, due to a lack of learning from co-located peers and increased coordination costs. We study the effects of WFA on productivity at the United States Patent and Trademark Office (USPTO) and exploit two features of the setting that lend themselves to econometric analysis: First, our setting allows us to tease out how the lack of opportunities to learn from co-located peers affects productivity. Second, we exploit a natural experiment in which the implementation of WFA was driven by negotiations between managers and the union of patent examiners, leading to exogeneity in the timing of individual examiners’ transition to WFA. We observe mixed results of WFA across experienced and new hires: for experienced hires, WFA results in a 3.9% increase in output and a 24% reduction in turnover, without affecting the proportion of rework; however, for new hires, an increase in output due to WFA is followed by an additional increase in rework, indicating a negative learning effect from the lack of co-location with experienced peers. We employ micro-data on cost of living, degree of autonomy, and proxies for examiner effort to shed light on mechanisms. Back-of-the-envelope welfare estimates indicate that the 3.9% increase in patent grants at the USPTO could potentially create $1.16 billion in value for the U.S. economy.

  Keywords: Working Conditions; Performance Productivity; Learning; Cost vs Benefits;

  Citation:

  Choudhury, Prithwiraj, Cirrus Foroughi, and Barbara Larson. "Work from Anywhere or Co-locate? Autonomy versus Learning Effects at the United States Patent Office." Harvard Business School Working Paper, No. 19-054, November 2018.  View Details

  CiteView Details Read Now

  Related

• Working Paper | HBS Working Paper Series | 2018

  Developing Theory Using Machine Learning Methods

  Prithwiraj Choudhury, Ryan Allen and Michael G. Endres

  We describe how to employ machine learning (ML) methods in theory development. Compared to traditional causal inference methods, ML methods make far fewer a priori assumptions about the functional form of the underlying model that best represents the data. Thus researchers could use such methods to explore novel and robust patterns in data, which could in turn lead to inductive theory building. ML’s strengths include replicable identification of novel patterns in data. ML methods also address several issues raised by scholars (such as “p-hacking” and confounding local effects with global effects) pertinent to the norms of empirical research in the fields of strategy and management. We provide a step-by-step roadmap that illustrates how to use four ML methods (decision trees, random forests, K-nearest neighbors, and neural networks) to reveal patterns in data that could be used for theory building. We also illustrate how ML methods can illuminate interactions and non-linear effects better than traditional methods. In summary, ML methods could serve as a complement to both existing inductive theory-creating methods, such as multiple-case inductive studies and traditional methods of causal inference.

  Keywords: Machine learning; theory building; induction; Decision Trees; random forests; k-nearest neighbors; neural network; p-hacking;

  Citation:

  Choudhury, Prithwiraj, Ryan Allen, and Michael G. Endres. "Developing Theory Using Machine Learning Methods." Harvard Business School Working Paper, No. 19-032, September 2018. (Revised October 2018.)  View Details

  CiteView Details Read Now

  Related