Comparative analysis between Zombies and Non-zombies

Economic downturns and financial crises lead to a rise in the number of zombie firms, yet these firms do not necessarily decline during periods of recovery (Banerjee and Hofmann, 2018). In an era of growing uncertainty and resource constraints, stakeholders may face increasing difficulty in sustaining zombie firms. This highlights the need for further research to better understand corporate zombiism and to develop more effective strategies for handling these unsustainable firms.

This thesis seeks answers to three research questions: (i) Is there a misclassification of zombie companies? (ii) Do zombies exhibit a different level of narrative tone compared to nonzombies? (iii) How does the capital market respond to the narrative tone disclosed by zombies in comparison to non-zombies? The sample consists of companies listed on the FTSE All-Share Index from 2013 to 2022. After excluding financial and utility companies, the final sample consists of 457 firms. Due to different requirements of the research questions and the unavailability of certain data, each question is addressed with a different sample set. Specifically, 3,379 fiscal years are used to analyse zombie misclassification, 1,376 annual reports are used to identify differences in narrative tone between zombies and non-zombies, and market reaction is analysed using 1,295 event days.

Four zombie identification methods are used to address the first research question on zombie misclassification: CHK08 (Caballero, Hoshi and Kashyap, 2008), AAM18 (Adalet McGowan, Andrews and Millot, 2018), UGLR18 (Urionabarrenetxea et al., 2018), and BH20 (Banerjee and Hofmann, 2020). The results show that 574 out of 3,379 observations have been identified as zombies. However, only 8 observations have been classified as zombies by all four methods unanimously. Consequently, given the high level of misclassification, this study proposes a new approach for identifying zombies. The advantage of the new method is a reduction of Type II error, where fewer zombies are misclassified as financially healthy companies.

This study uses Henry’s (2008) narrative wordlists to address the second research question on differences in narrative tone between zombies and non-zombies. The results show that zombies show a higher level of negative tone, a lower level of positive tone, and a lower level of net tone compared to non-zombies. These findings suggest distinct communication strategies employed by the two groups of companies and indicate that zombies may not manipulate narrative tone to the same extent as non-zombies.

The event study methodology is applied across three cumulative abnormal return periods, and five regression models are used to address the third research question on capital market reaction to narrative tone. The findings reveal a positive association between zombies’ negative tone and abnormal stock returns, as well as a negative association between zombies’ net tone and abnormal stock returns. In contrast, the market reaction to non-zombies’ narrative tone appears to be muted. These results suggest that investors are sensitive to the narratives disclosed by zombies. Despite the increased degree of pessimistic language, investors seem to hold optimistic views, perhaps anticipating zombie reorganisations or turnarounds.

This thesis contributes three new analyses to the literature. Firstly, as a high degree of zombie misclassification was found, this study proposes a new multi-method identification approach that reduces Type II error, thus, it may appeal to risk-averse stakeholders. Secondly, this study opens a new research avenue by analysing the narrative tone of zombies and non-zombies. The findings reveal distinctive narrative features between these two groups, enabling a better understanding of their communication strategies. Thirdly, this thesis pioneers an investigation into the relationship between capital market reaction and the narrative tone disclosed by zombies and non-zombies. The results highlight the importance of narrative tone as a factor in communication strategies used by zombies to shape investor confidence.

The findings are likely to interest policymakers, analysts, investors, and managers. Policymakers, by better identifying zombies, would introduce new anti-zombie policies that would more efficiently distribute national resources. Analysts and investors would gain a better understanding of the information encoded in narratives, thereby enabling them to make more accurate assessments of zombie companies. Finally, since the capital market is sensitive to zombies’ qualitative information, managers may consider improving the quality of narratives as part of their communication strategies.