Date of Submission


Date of Award


Institute Name (Publisher)

Indian Statistical Institute

Document Type

Doctoral Thesis

Degree Name

Doctor of Philosophy

Subject Name

Computer Science


Economics and Planning Unit (EPU-Delhi)


Afridi, Farzana (EPU-Delhi; ISI)

Abstract (Summary of the Work)

The role of social contacts in finding jobs, career mobility and other labor market outcomes is well acknowledged both theoretically and empirically in labor economics. The existence of individuals’ social networks at their workplace is a pervasive phenomenon, leading to the widespread use of social ties for information, influence and referrals on both sides of the labor market. The central theme of this thesis is to examine the interplay of ‘off-workplace socio-economic interdependence’ and outcomes at the workplace. Understanding this interdependence has the potential to devise policies which can impact labor productivity, especially in developing countries. This thesis divided into five chapters, explores the association and mechanisms through which social networks may manifest themselves at workplace and affect workers’ behavior. These chapters rely on primary data from the garment manufacturing sector (India), in which the production process takes place in large assembly lines involving strong complementarities in labor input. The first chapter gives the introduction and brief description of the thesis. The second chapter uses high-frequency worker-level panel data from garment factories to find a positive impact of workers’ network size on their own and thereby line performance. Our theoretical model and empirical analysis show that monitoring (mentoring) by higher ability types from own social network makes the low-ability type worker put in higher effort, leading to an increase in line output, even in the absence of explicit, individual performance linked incentives. This monitoring (mentoring) takes place through the increased threat of social sanctions arising from the reputation of being a defaulter as own network size in the production line increases. Chapter 3 builds on this context and explores what happens when team performance determines the worker’s financial payoff. We use a minimum effort coordination game framework to show that socially connected teams have higher output and better coordination due to a greater degree of pro-social motivations. We test this model’s predictions through a unique lab-in-the-field experiment design that recruited garment factory workers for a real-time effort-based task and shut down other alternative channels. We find that while social incentives augment team productivity, financial incentives may not always give desired results. These two chapters use the familiarity-of-characteristics based network (caste and residential clusters) as a proxy for socio-economic interdependence. Chapter 4, on the other hand, explores actual connections and interaction patterns of the garment factory workers (horizontal and vertical ties), focusing on women, a group that has been historically under-represented at managerial positions. We find that women’s personal ties exhibit patterns inimical to career advancement, given the management’s dependence on in-house referrals (who are mostly males) for recruitment and promotion. The fifth chapter concludes and summarizes the policy implications. It also discusses future research on social networks and on-the-job outcomes for historically disadvantaged groups (such as women). The micro-econometric data used in this thesis is unique and innovative in itself (whether from factories or experiment design). Nevertheless, the key findings apply to situations where production occurs in large teams with limited observability of peers’ effort. This thesis contributes to the literature on worker incentives, management practices, and firm behavior when workers are complements and informal channels are prevalent for accessing information and influence in the labor market.


ProQuest Collection ID:

Control Number


Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.


Included in

Mathematics Commons