The Second Coming of Social Media: Predicting Bank Runs
- By Lachlan Colquhoun
- March 13, 2023
Silicon Valley Bank's (SVB) failure last week has sent tremors through the U.S. and global financial sector and equity markets. But according to one social media analytics company, the signs of a run on the bank were clear a week ago.
In the wake of the SVB collapse, the biggest bank failure in the U.S. since the global financial crisis of 2008, social media analytics platform LunarCrush claimed it accurately forecasted the bank run.
According to LunarCrush analytics, the bank's social engagements skyrocketed by an exponential 78,102.50% over the last week. Meanwhile, social contributors and engagement increased by 50,950% and 252,194.8%, respectively.
These metrics clearly indicated a significant increase in customer activity on social media platforms, indicating that SVB’s customers were becoming increasingly concerned about the bank's financial health.
LunarCrush’s prediction was proven accurate when SVB experienced a sudden surge in customer activity, with many customers withdrawing their deposits from the bank. This situation proved disastrous for the bank, leading to widespread panic and a potential financial collapse which saw U.S. regulators step in.
By analyzing critical metrics on social media platforms, LunarCrush enables companies to make informed decisions and stay ahead of the competition.
The company’s usual business is in cryptocurrency, where it uses social media to seek to identify market trends and help investors make buy and sell decisions. It collects social media activity on more than 4,000 cryptocurrencies and 300-plus NFT projects to identify deviations in trends and new influencer behavior, helping to spot changes.
Users tailor the LunarCrush platform according to their favorite investments, and through social media, they get another source of intelligence around essential market sentiments.
Meme stocks
Predicting a financial failure might be one use for a social media predictor, but it is by far not the only one in this rapidly developing area.
In the investment space, U.S. investment company VanEck has launched a Social Sentiment Exchange Traded Fund,which selects 75 stocks based on comments made on social media.
The authors used backdated information to validate the investment case in a whitepaper that accompanied the fund launch.
“It’s a little like searching for a needle in a haystack, but the stakes are high, so it’s worth trying some different approaches”
The launch generated much publicity and interest when it launched in March 2021 but has since been something of an underperformer, losing an annualized 34% of its value or 41% in total. It currently has a modest USD54 million in assets.
Talking about something, it would appear, does not translate directly into share market performance.
But does this undermine the idea that social media can be analyzed for accurate predictions?
Not just finance
The finance sector is not the only one experimenting with social media as a predictor.
One U.S. study, for example, uses AI analysis of social media language crossed with historical mortality data to predict future changes in opioid-related deaths.
Still, in health, there is potential for using AI and social media to predict future pandemics such as COVID-19.
Researchers at the University of California have received a USD1 million grant to develop an AI-based early warning system to examine social media posts to help predict future pandemics.
The idea is that infectious diseases “are sociobiological phenomena and leave both social and microbiological footprints.” AI and public data, such as tweets, may help “monitor human society for signs of unusual activities that reflect the emergence of novel pathogens with pandemic potential.”
The project builds on earlier work by UCI and UCLA researchers, including a searchable database of 2.3 billion U.S. Twitter posts collected since 2015 to monitor public health trends.
The researchers’ plan is to analyze tweets and other social media in the months leading up to the COVID-19 outbreak to determine if any patterns or trends could have provided an early warning on the outbreak.
“It’s a little like searching for a needle in a haystack,” said Andrew Noymer, an associate professor of population health and disease prevention at UCI, in the 2022 announcement.
“But the stakes are high, so it’s worth trying some different approaches.”
Lachlan Colquhoun is the Australia and New Zealand correspondent for CDOTrends and the NextGenConnectivity editor. He remains fascinated with how businesses reinvent themselves through digital technology to solve existing issues and change their entire business models. You can reach him at [email protected].
Image credit: iStockphoto/kimberrywood