Bank Run

What Is a Bank Run?

A bank run occurs when a large number of depositors simultaneously attempt to withdraw their funds from a financial institution, driven by fears that the institution may not be able to meet all withdrawal demands. The phenomenon is rooted in a fundamental feature of modern banking: banks do not hold all deposited funds in reserve. Instead, they operate under a fractional reserve system, lending out the majority of deposits to borrowers while retaining only a fraction as liquid cash.

Under normal conditions, this works well because depositors rarely all need their money at the same time. But when confidence erodes, a self-reinforcing cycle begins: withdrawals reduce the bank's available reserves, which increases the risk that remaining depositors will lose access to their funds, which motivates even more withdrawals. Economists Douglas Diamond and Philip Dybvig formalized this in their 1983 model, demonstrating that a bank run can be the individually rational choice for each depositor even when the bank is fundamentally solvent.

The result is that a perfectly healthy bank can fail not because its loans are bad, but because it cannot convert long-term assets (mortgages, bonds, business loans) into cash fast enough to satisfy a sudden wave of withdrawals. This mismatch between short-term liabilities (deposits) and long-term assets (loans) is the structural vulnerability at the heart of every bank run.

How It Works

The mechanics of a bank run follow a predictable pattern driven by the asset-liability mismatch inherent in fractional reserve banking:

1. A trigger event erodes depositor confidence: this could be negative news about the bank's financial health, a broader economic downturn, or rumors of insolvency
2. Early depositors withdraw their funds while the bank still has sufficient liquid reserves to honor requests
3. Withdrawals accelerate as remaining depositors observe the growing lines (physical or digital) and fear being left with nothing
4. The bank exhausts its liquid reserves: it cannot sell long-term assets (loans, bonds) quickly enough or without significant losses
5. The bank either receives emergency liquidity from a central bank or regulator, or it fails and is placed into receivership

Reserve Ratio and the Tipping Point

A bank's vulnerability to a run depends heavily on its reserve ratio: the proportion of deposits held as liquid reserves versus the amount lent out. A simplified view of how reserves deplete during a run:

Bank Balance Sheet (Simplified)
─────────────────────────────────────────
Total Deposits:     $10,000,000,000
Liquid Reserves:     $1,500,000,000  (15%)
Loans & Securities:  $8,500,000,000  (85%)

Day 1: 5% withdrawal  → Reserves: $1,000,000,000
Day 2: 8% withdrawal  → Reserves: $200,000,000
Day 3: 3% withdrawal  → Reserves: INSUFFICIENT
        ↳ Bank must sell assets at a loss or seek
          emergency lending from the central bank

The key insight: only a fraction of depositors need to withdraw before the bank hits a liquidity crisis. In the example above, just 16% of total deposits exhausts all liquid reserves. This is why bank runs are self-fulfilling: the fear of insolvency causes the very insolvency that depositors feared.

Digital Bank Runs

Modern bank runs unfold at internet speed. Unlike historical runs where depositors physically queued at branch offices, today's depositors can initiate transfers via mobile apps, online banking portals, and wire services. This compresses the timeline from weeks or days to hours.

Silicon Valley Bank's collapse in March 2023 demonstrated this vividly: depositors attempted to withdraw approximately $42 billion in a single day, roughly 25% of the bank's total deposits. The bank failed within 48 hours of the first signs of distress, making it the fastest major bank failure in modern history.

Historical Examples

Northern Rock (2007)

Northern Rock was the first significant bank run in the United Kingdom in over 150 years. On September 14, 2007, after the BBC reported that the bank had sought emergency funding from the Bank of England, long queues formed outside branches. The bank's website crashed and phone lines were overwhelmed. Approximately £2 billion was withdrawn, about 8% of total deposits. The UK government eventually nationalized the bank in February 2008.

The episode exposed gaps in British deposit insurance: at the time, only the first £2,000 of deposits were fully covered, with 90% coverage on the next £33,000. This partial guarantee failed to prevent panic. The UK subsequently raised its deposit protection limits.

Silicon Valley Bank (2023)

SVB was the 16th largest bank in the United States, with $209 billion in total assets. Its collapse was triggered by a combination of factors: rising interest rates caused massive unrealized losses on the bank's $21 billion bond portfolio (which yielded an average of 1.79% while benchmark rates exceeded 3.9%), and a slowdown in venture capital funding forced startup clients to draw down their deposits.

On March 8, 2023, SVB announced it had sold its bond portfolio at a $1.8 billion loss. The next day, its stock price fell 60% as depositors rushed to withdraw funds. By March 10, regulators seized the bank. It was the largest U.S. bank failure since Washington Mutual in 2008. The FDIC invoked a systemic risk exception to guarantee all deposits, including the roughly 94% of SVB deposits that exceeded the standard $250,000 insurance limit.

Prevention Mechanisms

Deposit Insurance

Deposit insurance is the primary tool for preventing bank runs. The U.S. Federal Deposit Insurance Corporation (FDIC) guarantees deposits up to $250,000 per depositor, per insured bank, per ownership category. When depositors know their funds are protected regardless of the bank's fate, the incentive to withdraw early disappears.

However, deposit insurance has limits. Large institutional depositors, corporate treasury accounts, and high-net-worth individuals often hold balances far exceeding the insured limit. These uninsured deposits remain vulnerable to runs, as SVB demonstrated: most of its depositors were tech companies with multi-million dollar accounts.

Central Bank as Lender of Last Resort

Central banks can provide emergency liquidity to solvent banks experiencing temporary withdrawal pressure. By lending against a bank's illiquid but sound assets (such as mortgages and government bonds), the central bank bridges the gap between short-term withdrawal demands and long-term asset maturity. This prevents a liquidity crisis from becoming a solvency crisis.

Full Reserve Banking

Full reserve banking eliminates the structural vulnerability entirely by requiring banks to hold 100% of deposits in reserve. While this prevents runs by design, it also eliminates the bank's ability to earn interest by lending deposits, fundamentally changing the banking business model. No major economy currently uses a full reserve system.

Bank Runs in Crypto

Cryptocurrency markets experience dynamics that closely mirror traditional bank runs. When confidence in a stablecoin issuer or lending platform wavers, users rush to redeem or sell their tokens before others do, creating the same self-reinforcing withdrawal spiral.

Stablecoin Depegs as Bank Runs

The connection between bank runs and crypto was starkly illustrated during the SVB crisis. Circle, the issuer of USDC, disclosed that approximately $3.3 billion of its roughly $40 billion in reserves was held at Silicon Valley Bank. USDC immediately lost its dollar peg, trading as low as $0.87 on exchanges like Coinbase and Curve. This was a depeg event driven by the same fear that drives traditional bank runs: uncertainty about whether the backing assets could cover all redemptions. The peg was restored within 72 hours once the FDIC guaranteed SVB deposits. For a deeper analysis of how depegs cascade through DeFi, see the research on stablecoin depeg contagion.

The broader category of crypto bank runs includes the collapse of lending platforms like Celsius and Voyager in 2022, where users attempted to withdraw funds faster than the platforms could liquidate their illiquid positions.

Transparency as a Defense

Unlike traditional banks, crypto projects can offer real-time transparency into their reserves. Proof of reserves attestations and on-chain verification allow users to independently confirm that a stablecoin's reserves match or exceed its circulating supply. This transparency can reduce the information asymmetry that fuels panic withdrawals.

For example, USDC publishes monthly third-party attestations, and its reserves are primarily held in U.S. Treasury Bills and cash at regulated financial institutions. However, attestations have limitations: they are point-in-time snapshots that do not cover future solvency, custodian-bank operational risk, or smart-contract risk. The evolution of these mechanisms is covered in depth in the research on stablecoin proof of reserves.

How Spark Addresses Run Risk

Bitcoin-native Layer 2 solutions like Spark take a fundamentally different approach to the trust problem. Rather than relying on fractional reserves or custodial promises, Spark uses a self-custodial model where users retain control of their own funds. There is no pooled reserve that could be subject to a run, because each user holds their own virtual UTXOs with FROST threshold signatures ensuring that no single party can freeze or seize funds.

Risks and Considerations

• Contagion risk: a bank run at one institution can spread to others as depositors question the health of similar banks. The SVB failure triggered runs at Signature Bank and First Republic Bank within days.
• Moral hazard: government guarantees (like the FDIC's systemic risk exception for SVB) prevent runs but may encourage riskier behavior by banks that expect to be bailed out.
• Speed of digital runs: mobile and online banking has made runs faster and harder to contain, reducing the window for regulators to intervene.
• Stablecoin exposure: stablecoins backed by fiat reserves held in traditional banks inherit the run risk of those banks, as the USDC-SVB episode demonstrated. See the stablecoin depeg risk glossary entry for more on this dynamic.
• Incomplete insurance: the FDIC's $250,000 limit leaves large depositors exposed, and crypto deposits are generally not covered by any government insurance program.

This glossary entry is for informational purposes only and does not constitute financial or investment advice. Always do your own research before using any protocol or technology.