Social Media Sentiment and Classification Shifting

Abstract

This study investigates the role of social media sentiment in classification shifting. Prior literature finds that social media can either reduce accruals management by reducing information asymmetry or increase it by raising market expectations. Using a sample of 21,635 firm-year observations based on data from Chinese A-share listed firms from 2009 to 2020, we examine how social media sentiment from the stock forum EastMoney affects classification shifting in China. We find that managers increase classification shifting in response to negative social media sentiment, and our channel tests support the market pressure hypothesis. We confirm that negative social media sentiment increases classification shifting by exerting greater market pressure on managers. We also find that, if accruals or real earnings management is constrained, firms resort to classification shifting to address negative social media sentiment. Finally, we show that negative social media sentiment triggers more stock volatility and price drops. Our study contributes to the earnings management literature by providing novel evidence that sheds light on the market pressure exerted by social media in shaping earnings management strategies.