Is the advice influencers share about cryptocurrency of real investment value? A new paper suggests that it isn’t.
As cryptocurrency continues to grow in popularity, so have so-called crypto-influencers who offer advice about how to invest in crypto through their growing social media channels.
“The primary results of the paper indicate that crypto influencers’ tweets are initially associated with positive returns, consistent with them getting investors’ attention,” says Ken Merkley, a professor of accounting at the Indiana University Kelley School of Business.
“However, these tweets are followed by negative longer-horizon returns, consistent with pump and dump schemes but at a minimum suggesting influencers’ advice has little long-term investment value.
“What is also troubling is that these effects are most pronounced for tweets issued by crypto-influencers proclaiming to be crypto experts and for self-described experts with many Twitter followers.”
These findings are especially relevant when considering the billions of dollars that investors have lost in crypto scams. In September, the FBI reported that Americans lost $5.6 billion to crypto scams in 2023, a 45% jump from the year before. Several recent high-profile cases suggest that figure is rising exponentially.
The researchers examined the returns associated with about 36,000 tweets issued by 180 of the most prominent crypto social media influencers, covering more than 1,600 crypto assets over a two-year period ending in December 2022. Their efforts differ from past research involving celebrity endorsements, as it only focused on information provided by people directly and prominently involved in the crypto market.
They obtained lists of available crypto currencies and daily price data from the data aggregator CoinGecko. Through internet searches they obtained lists of the most prominent crypto-influencers accounts and merged them with data on tweets posted on Twitter. Their unit of observation was an individual mention of a crypto currency in an influencer’s tweet on a certain day. They focused on short term results from the first to second day and then to up to 30 days later.
The mean one-day return for crypto-influencer tweets was 1.83% (and 1.57% for two days). But analysis of subsequent returns shows they begin to decline substantially as early as five days after the tweets. The mean return from day two to day five is -1.02%, suggesting that over half of the initial positive returns are eliminated within five days.
The returns continue to become increasingly negative, as average cumulative returns ending 10 and 30 days after the tweet are -2.24% and -6.53% respectively.
“Back-of-the-envelope estimates suggest an individual buying small, lesser-known crypto tokens on the tweet date and holding the investment for 30 days would incur losses of 7.9%, generating an annualized loss of 62.8%,” says coauthor Mark Piorkowski, an assistant professor of accounting.
“Post-event returns are more negative when the influencer self-describes as an expert and even more negative for experts who have a higher following.
“One can only profit from cryptoinfluencers’ advice by exiting the position shortly after the initial tweet, a strategy that is often not viable,” Piorkowski adds.
The researchers also used machine-learning methods to classify tweets. They found that this pattern of results strengthens when the tweets have a more positive sentiment or relate to buy recommendations, suggesting that crypto-influencers are usually sending positive signals in their tweets.
“Overall, the evidence in our study suggests that investors should be cautious in following crypto-influencers’ investment advice, as most gains dissipate soon after the tweets,” says coauthor Brian Williams, a professor of accounting.
“Some crypto-influencers may simply be chasing trends or promoting crypto-assets that will gain them the most visibility and followers, without a regard to the actual future performance of the asset.”
The new paper appears in the .
Joseph Pacelli, an associate professor of business administration at Harvard University is a coauthor of the paper.
Source:
