Former Goldman Sachs Vice President Matt Levine: How can stablecoins solve the bank run problem?

Original Title: “Stablecoins Can Have Bank Runs”

Author: Matt Levine, Bloomberg Opinion columnist, former Goldman Sachs vice president

Translated by: Chris, Techub News

Cryptocurrencies are rapidly retracing the path of traditional finance.

Compared to the complex and chaotic financial crises of the traditional financial world, the financial crises in cryptocurrencies are often more intuitive and straightforward. You can understand a credit crisis by analyzing the 2008 global financial crisis, but that is quite complex; the 2022 cryptocurrency crisis is more like a simplified version of it, where everything happens faster and more transparently, with some participants even sharing their experiences in real-time on Twitter and YouTube. Although these crises in cryptocurrencies are simple, they allow us to understand similar financial phenomena more clearly, almost like a vivid textbook.

I once wrote a line: “Cryptocurrency is the result of smart, ambitious interns in traditional financial firms controlling a simulated market.” This line illustrates why the world of cryptocurrencies is so educational. It places those young, smart minds in an environment with few constraints, allowing them to explore, make mistakes, and learn from them. This process is not only fun but also teaches us a lot about finance.

Stablecoins are somewhat like an abstract form of banks in the cryptocurrency world. Here’s how it works: you give 1 dollar to the stablecoin issuer, and they give you a “receipt,” which is the stablecoin, representing “1 dollar.” Then, you can use this stablecoin as if it were dollars in a cryptocurrency environment like a blockchain or cryptocurrency exchange. It can be used for trading; you can use a 1 dollar stablecoin to buy 1 dollar worth of Bitcoin, and the person selling you the Bitcoin will then have your stablecoin.

Meanwhile, the stablecoin issuer will hold onto your 1 dollar and invest it, trying to earn profits from these investments to cover operational costs and management salaries. You (or anyone holding the stablecoin) can usually return the stablecoin to the issuer at any time to exchange it back for 1 dollar. At that point, the stablecoin issuer must come up with that 1 dollar to return to you.

This process illustrates the operational logic of stablecoins: they are both like currency and an abstract form of a bank, with their core relying on the promise to peg 1 dollar to the stablecoin to gain trust, and maintaining and operating the entire system through investments of those dollars.

The operation of stablecoins is somewhat similar to bank deposits, but there are some key differences between them. If you understand how banks operate, you will know several ways these systems can go wrong. Here are two famous examples:

• Bank Risk Investments:

The stablecoin issuer (like a bank) holds your funds and has the right to invest those funds and earn profits from them. The more profits the issuer earns, the more they can retain. However, if the investments fail and lead to losses, most of the losses will ultimately be borne by the depositors (i.e., the stablecoin holders). Since most of the funds actually belong to the depositors, if the issuer loses all that money, they will not have enough funds to compensate the depositors. This creates a risk-taking incentive for the issuer: if the risky investment succeeds, they can make a lot of money; if it fails, the losses primarily affect other people's money.

• Potential for “Runs”

Even if the stablecoin issuer invests all funds in very safe assets, some of these assets may be long-term investments. If everyone demands to redeem their money today, the issuer may not be able to immediately liquidate these long-term investments and may have to sell them at a loss, resulting in insufficient funds to meet everyone’s demands. This dynamic is well-known and thus has a self-reinforcing characteristic: if you think a run might happen, you should withdraw your money as soon as possible (before the issuer runs out of funds), and if everyone thinks and acts this way, it will indeed lead to a run.

These two issues (bank risk investments and runs) are often interconnected, with a common cause of runs being that banks made some poor investment decisions with depositors' money.

Sometimes, banks may lose a significant amount of money in investments and go bankrupt before anyone realizes it. In such cases, even without a run, a bank may fail due to investment losses.

On the other hand, runs can also occur without investment losses, purely due to liquidity mismatch issues. That is, a bank may have enough assets to repay all depositors, but these assets are long-term or not easily liquidated, while the funds demanded by depositors are “immediate liabilities” that can be withdrawn at any time. When too many people demand withdrawals at the same time, the bank may not be able to quickly liquidate enough assets to meet these demands, leading to a run, even if those assets themselves have not lost value.

To address the investment risk issues faced by banks, there are two main countermeasures:

1. Prudential Supervision: Regulatory agencies closely monitor banks' investment behaviors to ensure they do not engage in high-risk or inappropriate investments. This regulation helps banks avoid serious investment mistakes.

2. Capital Regulation: Banks are required to maintain a certain ratio of extra funds (i.e., capital buffers) to ensure that even if investments incur losses, the bank still has non-depositor funds to absorb those losses. This regulation reduces the risk of investment failures directly impacting depositors.

For the issue of runs, there are also several common solutions:

1. Liquidity Regulation: Banks are required to keep enough cash or other highly liquid assets to ensure they can promptly pay when depositors demand withdrawals. This regulation ensures that banks can meet customers' withdrawal needs in the short term.

2. Lender of Last Resort: If a bank holds good but illiquid assets, and at a certain moment all depositors want to withdraw funds, central banks like the Federal Reserve will provide loans to the bank to ensure it can temporarily weather the storm while waiting for assets to be liquidated. This mechanism prevents banks from failing due to temporary liquidity issues.

3. Deposit Insurance: The government promises that if a bank does fail, they will compensate depositors for their losses (usually within certain limits). This assurance gives depositors peace of mind, preventing large-scale runs due to fears of bank failures.

The stablecoin space largely lacks those protective mechanisms found in traditional banking. While various proposals attempt to introduce some of these elements, overall, stablecoin issuers typically do not have the regulatory frameworks or safety measures that banks possess.

Take Tether, for example, which is currently one of the largest stablecoin issuers. Tether ran into trouble in 2019 for making extremely risky investments with customer funds. At that time, its public financial reports even showed that its capital adequacy ratio was only 0.2% (which has since improved), meaning it had almost no extra funds to absorb potential losses.

Another example is TerraUSD, whose investment strategy can be described as highly concentrated risk. This strategy led to TerraUSD losing its peg during a run in 2022.

Regulatory solutions for stablecoins might look like this:

The most direct solution is to impose bank-like regulatory requirements on stablecoin issuers. In other words, “stablecoin issuers should invest funds in relatively safe assets that should have high liquidity, and they should hold a certain proportion of their own capital to ensure that even if these assets incur losses, there are still enough funds to repay users holding stablecoins.” This is a relatively simple answer, but implementing it effectively requires addressing many detailed issues.

Another bolder idea is to allow stablecoin issuers to access support mechanisms similar to those of banks.

In simple terms, a good regulatory solution should include the following aspects:

1. Safe Investment Policies: Ensure that stablecoin issuers invest customer funds in low-risk, highly liquid assets.

2. Capital Requirements: Require issuers to hold sufficient own capital as a buffer to address potential losses.

3. Liquidity Management: Ensure that issuers have enough liquidity to meet users' redemption demands, preventing situations similar to bank runs.

4. Systemic Support: Consider providing stablecoins with support mechanisms similar to traditional banks, such as deposit insurance or emergency liquidity support.

A paper written by Gordon Liao, Dan Fishman, and Jeremy Fox-Geen, who are employees of stablecoin issuer Circle Internet Financial, is very interesting. This paper explores “Risk-Based Capital Requirements for Stable Value Tokens”, and it is clear that Circle has some interest in lenient stablecoin regulation, but the paper indeed raises some important points about the relationship between stablecoins and banks.

One point made is that stablecoins are, in many ways, more transparent than banks. For cryptocurrency enthusiasts who are skeptical of the opacity of the traditional financial system, this transparency may be seen as an advantage. However, the opacity in the traditional financial system has its reasons.

Liao, Fishman, and Fox-Geen point out in their paper that while this transparency may seem attractive to supporters of the cryptocurrency world, in reality, certain opacities in the traditional banking system are designed to protect the stability and efficiency of the system. The complexity and certain degree of opacity of banks help maintain confidence and stability during market fluctuations, while excessive transparency may exacerbate market panic during crises.

In the traditional banking system, the main reason for bank runs is usually that people believe others will also rush to withdraw their money. How do you know if everyone will run? The roots of this idea are often due to rumors, poor financial reports, panicked television interviews, and other signals. A drop in a bank's stock price may indicate problems with deposits.

However, in the world of stablecoins, the situation is more direct and transparent. Because stablecoins are traded on public markets, their prices directly reflect market confidence in them. If a stablecoin's trading price is $1.0002, that may indicate there is currently no risk of a run; but if its price drops to $0.85, it almost certainly indicates that a run is occurring in the market.

This transparent price signal makes it easier for people to gauge market sentiment, but it also means that price fluctuations of stablecoins will quickly convey market panic, potentially accelerating the occurrence of runs. This public market reaction not only reflects the fundamentals of stablecoins but can also become a self-fulfilling prophecy, as when people see prices drop, they are more inclined to further sell or redeem, causing prices to continue to fall, thereby exacerbating market instability. While this transparency aids in the dissemination of market information, it can also become a trigger for chain reactions during crises.

The main way to address this issue is for stablecoin issuers to maintain high liquidity for most of their funds.

Stablecoins and traditional banks need to adopt different strategies in managing financial risks, especially when facing higher risks of coordinated runs.

Specifically, fiat-backed stablecoins typically hold highly liquid assets, avoiding excessive maturity mismatches (i.e., inconsistencies in the maturity dates of assets and liabilities) and relatively low credit risks. This is because the risk of runs for stablecoins is much higher than for traditional banks, so they need to maintain high asset liquidity to quickly respond to redemption demands.

Since the asset pools held by stablecoins are usually more resilient and specifically separated for the interests of token holders, the capital buffers required for stablecoins (to absorb financial losses) are typically less than those for banks. In other words, because the asset pool is robust enough and the risk of runs is lower, stablecoins do not need as much capital as banks to address potential losses.

However, for tokenized deposits supported by traditional loans and fractional reserves, due to the increased risk of runs from tokenization, they may require more capital than traditional deposits, even if both have the same asset backing. The reason is that tokenized deposits inherit the inherent asset-liability mismatch issues in bank balance sheets, so similar capital and solvency regulatory mechanisms may need to be applied to manage these risks.

The phrase “blockchain blockchain blockchain” may emphasize the core role of blockchain technology in stablecoins and its impact on the unique nature of stablecoin systems.

The use of tokenization and distributed ledgers not only introduces financial risks but also brings additional risks related to technology, infrastructure, and operations. These non-financial risks have been highlighted in public consultations and proposals from regulatory agencies.

Specifically, the application of cryptographic technology, permanent record-keeping, and traceable transactions can somewhat reduce certain security and compliance risks. However, these technologies also present new challenges, particularly in assessing the capital needed for these risks. Such risks are typically referred to as operational risks in traditional banking.

The difficulties in assessing these operational risks stem from several factors:

1. Lack of Sufficient Historical Data: Due to the relative novelty of blockchain technology, there is limited historical data on operational losses, making risk assessment more complex.

2. Dependence on Technology Choices: The technology choices made by issuers can significantly impact the capital needed to absorb losses. This dependence becomes more pronounced with the rapid development and upgrading of infrastructure.

3. Rapid Changes in Infrastructure: In a constantly evolving technological environment, assessing and managing these operational risks becomes more challenging. The choices and changes in technology can significantly affect issuers' ability to respond to potential risks.

On one hand, it can be imagined that stablecoin issuers, due to their blockchain-based characteristics of transparency, traceability, and digital native nature, are less likely to lose your funds compared to traditional banks. The public ledger and cryptographic protection of blockchain make it theoretically lower the risk of fund loss, as every transaction can be recorded and verified.

But on the other hand, it can also be imagined that precisely because of these same technological characteristics, stablecoin issuers may be more likely to lose funds. Possible reasons include:

• Technical Complexity: The complexity and novelty of blockchain technology may lead to operational errors or system vulnerabilities, especially in a rapidly evolving environment.

• Dependence on Technological Infrastructure: Stablecoins rely entirely on digital infrastructure, and if the system encounters issues or is attacked, it may lead to failures in fund management.

• Lack of Mature Operational Processes: Compared to traditional banks, stablecoin issuers may lack sufficiently mature operational processes and contingency mechanisms to address potential technical failures or errors.

Last year, the U.S. experienced a small banking crisis, which forced me to take time to rethink the banking industry. I once wrote a line:

The essence of banks is a way for people to collectively bear long-term and risky investments without having to pay particular attention to those risks. Banks make everyone safer and more profitable by spreading risks among numerous depositors.

When you and I deposit money in a bank, we think this money is very safe, it is money in the bank, and we can withdraw it at any time to pay rent or buy a sandwich. But in reality, banks take these deposits to issue long-term, fixed-rate 30-year mortgages. Homeowners cannot borrow 30-year money directly from me because I might need that money tomorrow to buy a sandwich. But they can borrow collectively from us because banks reduce this risk by spreading liquidity risk among many depositors.

Similarly, banks also lend to small businesses that might go bankrupt. These businesses cannot borrow money directly from me because I am unwilling to bear the risk of potential losses, but they can borrow collectively from us because banks reduce the credit risk by spreading it among many depositors and borrowers.

The opacity of the traditional banking system gives banks more power to invest customer funds in risky ventures. This opacity once helped banks operate in a relatively stable environment, as the complexity behind it made it difficult for customers to fully understand what banks were doing. However, last year's regional bank crisis partially illustrates that this opacity is no longer as effective as it once was.

Today, due to the widespread dissemination and digitization of information, the public can more easily access relevant information about banks. Rumors and panic can spread rapidly across the globe via the internet, and people's expectations of banks increasingly lean towards market valuation, focusing more on real-time market performance rather than long-term stability.

As a regulator from the Federal Deposit Insurance Corporation (FDIC) said last year, the game is still the same, just more intense. This statement highlights the new challenges facing modern banking: although the basic operational logic of banks has not changed, the transparency and speed of information dissemination have made market reactions more rapid and intense. The risk management advantages that traditional banks gained from opacity have become more vulnerable in the new information environment, and banks thus need to manage risks more cautiously to cope with more complex and rapidly changing market sentiments.

The “magic” of traditional banks lies in their ability to aggregate a series of high-risk investments and then issue senior claims on those investments, which manifest as dollars: 1 dollar in a bank account is 1 dollar, even if it is backed by a pile of risky assets. This arrangement allows customers to believe that their deposits are safe and can be used without risk when needed.

However, stablecoins abandon this “magic.” Although a 1 dollar stablecoin is almost equivalent to 1 dollar in most cryptocurrency scenarios, its market price can fluctuate. When market conditions are good, its trading price may be $1.0002 or $0.9998, but in adverse conditions, it may drop to $0.85. Stablecoins represent a form of banking that does not have the traditional bank guarantee of “1 dollar equals 1 dollar,” but instead reflects its proximity to 1 dollar through a 24/7 real-time market.

This situation raises new regulatory issues. Because stablecoins do not have the implicit guarantees and opacity of traditional banks, market prices directly reflect the risk status of the underlying assets and market confidence. This real-time market feedback not only changes the way stablecoins operate but may also signal the future direction of traditional banking.

In the future, as financial markets become further digitized and transparent, traditional banks may also face similar challenges. The asset risks and market confidence of banks may be more directly reflected in market prices than they are now, rather than relying on internal safeguards of the banking system. This shift could reshape our understanding of banks and financial stability and force regulators and market participants to adjust their risk management strategies.