| Author: | Xu, Mingze |
| Title: | Sustainability events and operations performance : event study, event mining, and data-driven event analytics |
| Advisors: | Wong, W. Y. Christina (SFT) |
| Degree: | Ph.D. |
| Year: | 2023 |
| Subject: | Clothing trade -- Environmental aspects Clothing trade -- Economic aspects Sustainable development Data mining Industrial management -- Technological innovations Hong Kong Polytechnic University -- Dissertations |
| Department: | School of Fashion and Textiles |
| Pages: | xiv, 227 pages : color illustrations |
| Language: | English |
| Abstract: | The fashion industry is facing sustainability challenges due to the pollution it creates and the nature of its energy- and resource- intensive operations. Researchers and managers are increasingly being urged to develop proactive sustainability strategies to meet the expectations of stakeholders and to cope with the challenges sustainability poses in the ever-changing business environment. Stakeholders are expecting firms to be more transparent in their sustainability activities, by adopting practices such as sustainability reporting. However, due to a lack of understanding of the value of sustainability reporting, many companies remain in a dilemma about whether and when to produce sustainability reports, which involve voluntary reporting that is beyond mandatory requirements. Apart from the pressure to meet stakeholder expectations, firms face many challenges from events taking place in the business environment that can influence their operations, supply chain activities, and competitiveness. These can not only challenge individual firms but also the wider industry and market development. In view of these expectations and challenges, this thesis offers innovative solutions to address this rapidly changing environment with the aim to help managers make decisions. Insights are gained from conducting three studies, with the results reported herein. The first study examines whether and when sustainability reporting can improve the performance of manufacturing firms based on signaling theory and stakeholder theory. The extant literature shows that sustainability reporting (SR) can improve a firm's market and financial performance through signaling its superiority to external stakeholders (investors and customers); this is known as the costly signaling effect. However, less is known about how internal stakeholders, like operations employees and senior management, can use SR to improve operational efficiency. We hence conducted five event studies to estimate the abnormal performance of US reporting and non-reporting manufacturers using Global Reporting Initiative (GRI) standard data from 1999-2020, which included 1254 firm-year observations. The findings suggest there are time-lagged positive effects of GRI reporting on the abnormal return on assets (ROA), labor productivity, COGS/Sales, Tobin's q (short term), and market value (marginal) due to the costly signaling (i.e., GRI reporting). Furthermore, by using media exposure, first-time reporting, and reporting frequency as proxies for signal observability, our regression results show that they can improve profitability, in terms of the abnormal ROA, and operational efficiency, in terms of labor productivity, through a "reverse" signaling effect. However, these proxies fail to improve market value and COGS/Sales, suggesting some weaknesses in the signaling effects. These results suggest that executives should pay more attention to internal stakeholders (employees) and sustainable operations when investing in GRI reporting. This study fills the research gap about the role of SR in driving financial performance and productivity, grounded in the integrated framework of stakeholder theory and signaling theory. As observed through the process of conducting the first study, there are limitations in conducting event studies. We observe that, in order to gain an understanding of an event and how it may affect firms, event identification is the beginning and indeed the most important step. Currently, this is still a manual process that relies heavily on the researcher's efforts, leading to several limitations in terms of efficiency, capacity, and comprehensiveness with respect to gaining insights on events that occur at high frequencies in the market. In particular, the existing event studies approach is to focus on a single event, which is an unrealistic scenario and fails to take into account the actual complexities of the business environment. To address these limitations, in Study 2, we designed and developed an approach called EventMining to identify multiple event clusters from textual company data available online. This approach adopts Natural Language Processing (NLP), which is a mainstream element of Artificial Intelligence (AI) technologies. The EventMining approach can help researchers and managers automatically collect, pre-process, analyze, and identify multiple event cases. Using company news collected from Thomson Reuters, we demonstrate that multiple event cases can be identified by EventMining and with less researcher intervention needed. By focusing on multiple event cases and replacing previous manual work, the proposed EventMining approach advances event study methods in terms of event identification. Based on four designed modules, EventMining collects, preprocesses, and analyzes textual data and eventually identifies event cases. Our application of the proposed approach to company news demonstrates its utility and robustness. To the best of our knowledge, EventMining is among the first efforts in machine-based event identification and event case generation. The proposed approach contributes to gaining an understanding of the event patterns contained in complex text data. In a volatile, uncertain, complex, and ambiguous market, managers have to adapt and respond to events as they arise in the business environment. While sustainability events are an important stream of events that now concern managers and stakeholders, less attention has been paid so far to Emerging Sustainability Events (ESEs), which are sustainability events that are still in the formation process. Although the use of EventMining in the second study provides a tool for managers to identify multiple event cases in their business environment, the question remains as to what are the important events that might trigger market reactions. In Study 3, we employ the EventMining approach and extend Study 2 with the design of a data-driven event analytics system. Based on a large-scale event study including 120 ESEs, our empirical findings show that a series of ESEs can trigger market reactions. Unlike the event study literature hypothesized general effects of events, we find that ESEs may have localized effects that can affect specific groups of firms. Abnormal returns can indicate investors' concerns about ESEs. The results also highlight the most important ESEs concerning investors. To the best of our knowledge, this study is among the first to highlight the importance and impact of ESEs for gaining managerial insights. This study sheds light on the role of emerging events in prompting managers to develop proactive strategies and operations. |
| Rights: | All rights reserved |
| Access: | open access |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/13730