Silicon Valley Bank: Social media ‘amplified’ bank run, researcher says

Christoph Schiller, assistant professor of finance at the Arizona State University W. P. Carey School of Business, discusses how social media may have accelerated the collapse of Silicon Valley Bank.

Video transcript

[AUDIO LOGO]

JULIE HYMAN: On March 9 of this year, depositors initiated $42 billion in withdrawals from Silicon Valley Bank. That caused the first Twitter-fueled bank run, that's what Congressman Patrick McHenry, chair of the House Financial Services Committee, that's what he called it shortly thereafter. Five finance professors now from the US and Europe examined this theory.

Christoph Schiller, Finance Professor at Arizona State Business School, is one of them, and he's with us right now. Chris, thanks for being here. So this was something that was talked about at the time, right, whether, in fact, the discussions on Twitter contributed to this. How do you, first of all, quantify that, whether it did?

CHRISTOPH SCHILLER: Well, our approach to quantifying that was to just actually observe what people were saying on Twitter. So we used statistical programs and scraped and pulled essentially every single tweet that was sent throughout the year about banks and bank holding companies and constructed measures of the number of tweets that people have sent about these banks, the content of these tweets, who sent them, and that allowed us to try to get closer to that answer.

DIANE KING HALL: But, Chris-- Diane here-- correlation doesn't exactly always equal causation. So what would you say is the-- you know, could that really-- can social media really contribute to a run on banks?

CHRISTOPH SCHILLER: So we structured our analysis in a way that sort of covers this from a couple of different angles. One thing we did was measure the amount of talk about these banks in a period long before the run. So we sort of captured the amount of tweets about these different bank holding companies in January and February, long before the run occurred, and then tried to understand how that sort of amount of talking in January and February is related to the decline in returns and the run on these banks during the run period.

So we have sort of a separation between when we observed the exposure to being some bank that gets talked a lot about on Twitter versus when the actual run happens, which helps us speak to that somewhat. And then during the actual run period, we also tried to observe when the Twitter conversations happened before we observed when the bank's returns change or when they go down so that we can try to separate is this Twitter conversation happening because returns are crashing, or are returns crashing because the Twitter conversations are happening?

JULIE HYMAN: And, likewise, is there a way of teasing out whether this would have happened anyway but this hastened the deposit flight from the bank?

CHRISTOPH SCHILLER: That is a difficult question to answer. So one thing that we do is our analysis is not sort of solely focused on SVB or on those two or three banks. We take a broader approach and look at the entire banking sector, essentially. And if we exclude firms like First Republic and Silicon Valley from our analysis, we still find very similar results indicating that sort of Twitter contributed to declining returns in firms that had a lot of this Twitter conversation.

Now, of course, we know that other media, other sort of communication channels like WhatsApp and Slack also played an important role. So maybe this would have happened anyway. But our analysis suggests that Twitter at least amplified and sort of poured additional fuel on the pace and on the intensity of this run.

DIANE KING HALL: So I know, Chris, you can't get the data on the deposit outflows with regard to that bank. But how-- would you say that we are not considering social media as impact enough on something like this with regard to a regional banking crisis?

CHRISTOPH SCHILLER: Well, we hope to, in the future, be able to maybe add some data on the deposit flows. That's just something that we cannot really do at this stage because we're not regulators. But in the future, this data might be available, and we might be able to say a little bit more about this.

I think one thing that the study highlights is that the very, very important sort of feature in any kind of bank run that is of coordination among depositors is at a completely sort of different stage now that depositors, investors, and so forth can communicate in sort of real time on a platform that's very visible and observable to anyone. And so that's, I think, a new quality or a new type of risk that we document in this paper.

What are the implications sort of going forward? Well, maybe we'll understand this risk a little bit better, and we'll sort of all pay a little bit more attention to this or banks will have to highlight how much they're exposed to this or they'll have to report on this. But it is something that we think is here to stay as these platforms are very visible and are very present in our everyday life.

JULIE HYMAN: Well, I guess then the final question is, what to do about it, right? We did get the Federal Reserve's report today on the collapse of SVB and what contributed to it. They-- the main culprit was the management of the bank, they said, but regulators and supervisors, bank supervisors, also played a role. Is there anything you can do about Twitter? I mean, I don't know what you would do. But is there anything that could be done?

CHRISTOPH SCHILLER: I think you're making an important point, which is that Twitter alone does not bring down these banks, right? What we find in the paper, too, is that Twitter amplified already pre-existing risks, right? These-- the run was much more intense when Twitter conversation was stronger but only for those banks that had sort of weaker balance sheets, had higher number of uninsured deposits, and so on. So I think what can be done in the first stage is have more stable sort of balance sheets that don't allow for this type of run.

On the other hand, maybe, you know, going forward there will be-- investors will take this type of risk into consideration more, will, in ex-ante, look at which banks get talked a lot about, which banks have more Twitter conversation without any kind of run so that we can think of this as an additional risk factor that needs to be taken into consideration. I don't think there will be any way of shutting down the platforms or putting any kind of limitations on the amount of tweets you can send. I think that's not very realistic.

DIANE KING HALL: Yeah. All right, social media versus SVB. Our thanks to you, Christoph Schiller, Arizona State University finance professor. We appreciate you coming on.