How Blockchain Technology Could Prevent Bank Runs

Banking - Shutterstock photo
Credit: Shutterstock

Silicon Valley Bank (SVB) suffered the largest and fastest bank run in history: Customers withdrew $42 billion in a single day from SVB, leaving the bank with $1 billion in negative cash balance. That's $4.2 billion an hour, or more than $1 million per second for ten hours straight. To understand the extent of this event, the previous largest bank run in modern U.S. history took place at Washington Mutual in 2008, which totaled $16.7 billion over the course of 10 days.

The withdrawals at SVB unfolded at a speed enabled by digital banking and were likely fueled in part by viral panic spreading on social media platforms and in private chat groups – making the run likely the first fueled by social media.

This predicament brings to mind the Federal Reserve Bank of Richmond's 2018 paper, which stated “banking activity is viewed as inherently fragile — that is, prone to runs. A run is when many depositors withdraw their funds to avoid losing those funds if the bank becomes insolvent. Notably, a run can occur whether the bank is insolvent or not — that is, the fear of a run may be enough to produce one.” In other words, the rumor becomes a self-fulfilling prophecy.

A bank run is a consequence of a liquidity issue, as described in a 2015 Chicago Booth Review paper – it’s a mismatch between assets (loans) and liabilities (deposits) – more specifically, a mismatch of short-term assets versus short-term liabilities. If the bank only holds a fraction of the deposits in cash (i.e., reserve ratio, a Fed requirement) – as all banks do – it will be almost impossible to accommodate a withdrawal of all (or most) deposits at the same time.

Bank runs are not something new, and spreading rumors on the stability of a financial institution or any organization is not new either. But it is the magnitude of the run and the speed that it can spiral that has changed due to digital banking and social media. The advances in technology might have worked to the detriment of SVB, but technology could also be utilized to provide real-time and transparent information to combat social media rumors from disseminating and potentially mitigate such an event.

How could we adjust to the advances in technology and cultural changes to prevent or at least mitigate the next bank run from happening? Here, we look at two different potential solutions, one proposed by the New York Fed, and the other that relies on blockchain technology.

Solution One: Mitigating the Risk of Runs on Uninsured Deposits: The Minimum Balance at Risk

The goal of the Minimum Balance at Risk (MBR) is to create a tradeoff between the liquidity of investments and their risk: investors who withdraw to preserve liquidity do so at the cost of greater risk to their principal; this tradeoff mitigates the incentives to run.

The basic mechanics of the MBR are as follows: a small fraction of each depositor’s uninsured deposit (the amount that exceeds the amount insured by the FDIC per depositor), which is called the minimum balance at risk, would be available for withdrawal only with a delay.

For example, the MBR could be 10 percent of a depositor’s maximum uninsured deposit balance over the previous thirty days. Let's say you had $1 million of uninsured deposits during the past thirty days, you would be able to withdraw all but $100,000 (10 percent) of your uninsured deposits immediately. The remaining $100,000 would only be available to you with a delay, for example, forty-five days.

To further reduce the incentive to withdraw, a portion of your MBR would be subordinated to the deposits of those who did not withdraw. If you requested to withdraw all your uninsured deposits, but another customer left all her uninsured deposits in the bank, you would receive your $900,000 while the 10 percent ($100,000) stayed in the bank for those forty-five days; but in that time period, your $100,000 would be subordinated to the other customer. In other words, if the bank fails during that period, you would absorb any loss up to your $100,000 before the other customer lost a penny.

This subordinated deposit structure significantly reduces the incentive for any bank customer to act as the first mover, because they will always be at a loss. If no one wants to be a first mover, then we may end up with no sudden withdrawals and hence preventing a bank run. MBR might be a plausible solution, but it is missing transparency concerning the bank’s liquidity risk condition, especially in real time. The bank's customers decision are still based on rumors and outdated information.

Blockchain technology might provide transparency that is critically needed to better assess the organization’s assets and liabilities, whether it’s a financial organization or any organization. What if transactions could be recorded on a transparent immutable ledger, providing real time accounting information of the organization’s assets and liabilities balances? TapestryX is developing such a solution utilizing blockchain technology, but you can take this a step further with tokenization.

Solution Two: Tokenization for Liquidity Risk Management

Liquidity risk management within financial services organizations can benefit from tokenization. The SVB case clarifies this benefit. Its failure had several underlying issues, one of them relating to the “liquidity gap” – the bank's customers had no real time information on the liquidity state of the bank. If there were checks and balances in place that were transparent for customers to see, mitigating actions could have been taken in time.

In the case of SVB, the available bank information on its assets and liabilities balances was outdated, as it reflected the 2022 fourth quarter. If everyone could have viewed in real-time the balances of the banks, the viral rumors on social media might have had no impact on the situation at SVB.

Tokenizing both assets and liabilities would have shown the bank reserves and capital at hand and assured customers in real time that the bank was not insolvent and has not been susceptive to unreasonable leverage, which might have prevented the social media panic.

After the FTX collapse, there has been a rush of efforts to provide Proof of Reserves from several centralized crypto exchanges. Proof of Reserves only shows that a firm has some assets to service its debts; it does not provide any information on the organization’s liabilities. If a company can transparently demonstrate that it has $1 billion in reserves/assets, but if its liabilities are not clear for everyone to see (for example, if it had $10 billion in liabilities), its solvency would come under question.

Therefore, it would be more useful to show “Proof of A Going Concern.” If both assets and liabilities of an organization can be tokenized, on-chain analytics can be used to understand if the organization has enough assets – short term and long term – to service its liabilities. In other words, it answers and verifies whether an organization is a “going concern” and if it will meet its financial obligations when they are due.

Furthermore, it would provide real-time information on the risk and leverage that the organization has been exposed to. In the case of SVB, it would have clearly informed customers that the bank had made no bad loans or had any excess leverage exposure, hence, the rumors that the bank is about to fail were most likely unfounded.

People might still spread rumors on the solvency of a financial organization. But we could mitigate the risk of bank run and the probability of future bank failure with the adoption of a blockchain technology solution.

The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.

Merav Ozair, PhD

Dr. Merav Ozair is a global leading expert on Web3 technologies, with a background of a data scientist and a quant strategist. She has in-depth knowledge and experience in global financial markets and their market microstructure.

Read Merav's Bio