To Tweet or not to Tweet? The Determinants of Tweeting Activity in Initial Coin Offerings

Abstract

Our research explores the causes of Twitter activity in highly technological start-ups that finance their activities via initial coin offerings (ICOs). By relying on weekly data of 297ICOs for the period 2015-2020 (35,459 observations), we examine how major exogenous events affect the number of tweets issued by the start-up. Then, we explore how the community of followers reacts to the tweets. We discover that events external to firms reduce ICOs' tweeting activity. Moreover, our evidence indicates that the followers' reaction is positively related to the tweets issued by the firm and negatively related to major events unrelated to the firm. Interestingly, followers' reaction has an inverted U-shaped relation with the firm's Twitter volume, suggesting that excessive Twitter activity can harm the further dissemination of tweets. Our results, robust to alternative estimation techniques, emphasize the important role of Twitter as an information disseminator, legitimizer, and endorser for highly opaque firms.

Anies Rasyid Baswedan’s Political Communication in Facing the 2024 Election

“…The following figure presents the categorization of tweets, retweets generated due to an account's tweet, and retweet activities. The @aniesbaswedan account had 1588 tweets and 1145 retweets, which can be seen by the total number of tweets and retweets resulting from those tweets, according to the interpretation of Moro et al (2023). The tweet content generated by the account can reflect and describe the tasks and responsibilities played by the account.…”

Big Data, Proxies, Algorithmic Decision‐Making and the Future of Management Theory

“…For illustration, we turn to Google Trends (GT). Management research (Liu et al, 2022;Moro et al, 2023;Reinwald et al, 2023) and economics (Castelnuovo and Tran, 2017;Lai et al, 2022) increasingly 'borrow' data obtained from GT. However, this enthusiasm is not always matched with sufficient scrutiny for the quality of the data (Cebrián and Domenech, 2023;Fradkin, 2014;Nuti et al, 2014), and the theoretical ramifications of their use.…”