I cordially invite you to attend my dissertation defense scheduled for Thursday, April 18, from 8:30 am to 10:00 am EST in Room 312, Scheller College of Business.
You are also welcome to join remotely via Zoom: https://gatech.zoom.us/j/98528244134.
Li Ding
Ph.D. Candidate in Operations Management
Scheller College of Business | Georgia Institute of Technology
Area: Operations Management
Committee Members: Dr. Basak Kalkanci (Chair), Dr. Seyed Emadi (UNC-Chapel Hill), Dr. Manpreet Hora, Dr. Vinod Singhal, Dr. D.J. Wu
Title: Towards a Better Future of Work: Enhancing Efficiency, Promoting Fairness, and Upholding Labor Standards in the Contemporary Workplace
Essay 1: Trips for Tips? Implications of Tips on Drivers’ Search Behavior
Motivated by the increasing adoption of tipping in ride-sharing services, we examine how tips influence drivers’ relocation decisions and subsequent outcomes such as driver earnings and system efficiency in an empirical setting. We first show that drivers’ relocation decisions are not only driven by their inherent heterogeneity, but are also affected by their learning from past experiences. We propose a discrete-choice structural model that captures drivers’ learning from their past experiences in a Bayesian fashion and estimate the model using a large-scale and granular taxi trip dataset. We find that drivers’ sensitivity to tip amount is greater than that of the fare amount, suggesting that the source of income is an important leading factor that shapes service providers’ relocation decisions. We perform a series of counterfactual analyses to find the impact of policies such as hindering learning by delaying the tip information, and a service charge that replaces the tip with a fixed percentage of the fare on driver earnings and system performance. We find that while tipping is increasingly adopted to supplement driver income, it can create an unintended consequence of lower system efficiency. Although hindering drivers from learning about tips reduces their hourly income by 12%, it improves the system efficiency, indicated by a 1% decrease in wait time for drivers and 10% decrease in the probability of unmet demand for customers. Replacing tips with a service charge not only increases drivers’ hourly income by 4%, but also leads to a small but positive improvement in system efficiency.
Essay 2: Reducing Gender Disparities in Online Labor Platforms: The Role of Performance Feedback and Task Exposure
Given the persistence of gender disparities in task selection and earnings in online labor platforms, we investigate the underlying behavioral mechanisms that drive these disparities and the interventions to mitigate them. We conduct a two-phase experiment on Prolific, where online workers are recruited to grade elementary school-level math problems of two difficulty and payment levels. We find that women earn significantly less than men because they are less likely to select the more difficult task compared to men of comparable capabilities, despite a higher payment for the same expected task completion time. This difference in task selection is partially driven by women’s negatively biased self-evaluation of their performance, particularly for more difficult tasks. In terms of mitigation strategies, we show that performance feedback alone may not be sufficient to motivate women to pursue the more difficult tasks later on, because women often choose to work on the easier task in the first phase, for which they have relatively unbiased self-evaluation. Supplementing performance feedback with early exposure to challenging tasks can effectively reduce the gender disparities in task selections and significantly boost women’s hourly earnings without adversely affecting men’s earnings. Our study provides actionable insights for online labor platforms to tackle gender disparities and build a more inclusive and equitable online work environment.
Essay 3: Beyond Health Risks: The Impact of COVID-19 on Labor Violations in Global Supply Chains
Many global organizations such as the International Labor Organization have called for an in-depth understanding of the impacts of COVID-19 on labor practices globally beyond the immediate health risks. Responding to these calls, we examine the impacts of the pandemic on the prevalence of worst forms of labor violations in global supply chains and whether the impacts (if any) persist over time. While business shutdowns, production slowdowns, and reduced demand during the pandemic can lead to fewer opportunities of labor violations, the supply chain disruptions induced by unexpected changes in orders as well as diverted compliance resources and attention might aggravate the risks of such violations. Motivated by the empirical ambiguity, we perform Year-to-Year difference-in-difference analyses to examine the impact of the pandemic outbreak on labor violation incidents in global supply chains. Our analysis covers 2385 globally public-listed firms and their supply chains in two different industries (apparel and electronics) and across 8 years (from 2015 to 2022). We find that the pandemic increases labor violations in global supply chains, and the impact persists over a 2- to 3-year post period. Moreover, heterogeneity analyses show that the impact is exacerbated for buying firms with more suppliers, more diversified geographic exposure of suppliers, but is mitigated by longer-term relationship with suppliers. This study provides practical insights for managers, policymakers, and industry regulators on strategies to assess and mitigate the risk of labor violations induced by the pandemic.