Blessing in Disguise? Environmental Shocks and Performance Enhancement
71 Pages Posted: 24 Jul 2018 Last revised: 7 Oct 2019
Date Written: October 6, 9
This study examines the consumer satisfaction and performance of firms in the service sector in response to reputation shocks caused by random and exogenous air pollution events. The study incorporates machine learning techniques, such as social media sentiment analysis and topic modelling, to analyze review texts and show that pollution shocks during haze episodes lead to significant decreases in consumer satisfaction, which can be explained by the changes in consumers’ mood rather than service quality. Moreover, while the level of consumer satisfaction immediately reverts to and then exceeds previous levels after the haze dissipates, such improvement in service quality is not persistent in the long run. The underlying mechanism is that firms with managers closely monitoring customer reviews and actively responding to negative feedback show signiﬁcant improvements following a temporary reputation crisis due to a pollution shock. Our findings provide novel empirical evidence on how drivers of organizational reputation change in times of crisis and highlight the importance of realizing deﬁciencies in operation even in the absence of negative shocks.
Keywords: Air Pollution, Service Quality, Customer Review, Natural Language Processing, Machine Learning
JEL Classification: Q51, Q53, Z30, D83, D22
Suggested Citation: Suggested Citation