Bank Run¶

A self-reinforcing mass withdrawal or refusal to renew short-term claims on a financial institution, where a finite liquid pool rewards early exit and each withdrawal worsens later claimants' prospects, so even a solvent institution can be drained by the race.

Version: v3 · 2026-09-07 · History
Domain-specific #: 80
Origin domain: economics finance
Subdomain: banking and financial stability
Aliases: Deposit Run, Financial Run

Core Idea¶

A bank run is a self-reinforcing mass withdrawal, redemption, or refusal to renew short-term claims on a financial institution or runnable funding vehicle. Claimants can exit faster than the institution's assets can be converted into par liquidity. Because the immediately available pool is finite, early exit improves an individual claimant's chance of full repayment while each withdrawal worsens the position of those who remain^[1]. That first-mover advantage can shift the institution from continued funding into a run equilibrium.

The identity does not require initial insolvency. A solvent institution may own assets whose long-run value exceeds its liabilities while still being unable to meet simultaneous demands without selling at discounts^[1]. The run can then produce the very losses and liquidity shortage that claimants feared. Retail deposit queues, wholesale creditors refusing overnight rollover, money-market fund redemptions, and digital withdrawals are different operational forms of the same financial event.

Structural Signature¶

The runnable institution or vehicle — an intermediary financed by short, redeemable, or renewable claims.
The illiquid asset side — commitments cannot be converted to par liquidity as fast as claims can leave.
The finite liquid pool or backstop — cash and immediately available support cannot satisfy universal simultaneous exit.
The first-mover advantage — earlier claimants expect better repayment than later claimants.
The withdrawal or non-rollover action — claims are redeemed, transferred, or allowed to mature without renewal.
The endogenous deterioration — each exit reduces what remains or forces costly liquidation.
The equilibrium transition — expectations and action move the institution from continued funding to a run.

What It Is Not¶

A bank run is not identical to insolvency. Insolvency concerns the value of assets relative to liabilities; a run concerns the timing and convertibility of claims. It is not Funding Fragility, which is the structural exposure present before anyone runs. It is not a Fire Sale, although forced liquidation often transmits the run into market prices and balance-sheet losses.

Nor is every simultaneous financial exit a bank run. A broad credit crunch can involve lenders refusing new credit without racing to redeem claims on one runnable intermediary. Generic users leaving a platform lack the par-claim and maturity-transformation machinery.

Scope of Application¶

Retail banking: depositors transfer or withdraw demand deposits.
Wholesale banking: professional creditors refuse to roll repo, commercial paper, or other short-tenor funding.
Shadow banking: runnable vehicles promise liquid or stable-value claims backed by less-liquid assets.
Money-market funds: investors redeem shares whose stable value depends on liquid reserves or sponsor support.
Digital banking: mobile transfers compress the event from visible queues into rapid remote outflows.
Currency boards or reserve arrangements: claimants race against a finite pool of liquid backing, subject to the exact institutional claim structure.

Clarity¶

The abstraction separates condition, event, and consequence. Funding Fragility is the maturity- and confidence-sensitive balance sheet. Bank Run is the coordinated withdrawal event. Fire Sale is one possible liquidation consequence. Insolvency may be an initial fundamental, a result of the run, or absent. This decomposition prevents every bank failure from being explained as the same thing.

Manages Complexity¶

The event can be represented by six roles: runnable claim, asset-conversion delay, liquid pool, first-mover advantage, withdrawal feedback, and equilibrium transition. Interventions map onto roles. Deposit insurance weakens the incentive to exit; lender-of-last-resort lending enlarges credible liquidity; liquidity requirements change the pool; longer funding maturities reduce runnability; gates slow the timing but can also intensify pre-gate exit^[2].

Abstract Reasoning¶

The run is an equilibrium-selection problem with financial differentia. Continued funding may be sustainable if claimants expect one another to remain. If they expect exit, leaving early becomes individually rational because later claimants face a depleted pool and more forced liquidation. Their action then validates the expectation. Fundamentals can select or strengthen the bad equilibrium, but pure coordination can also matter^[3].

The abstraction also separates stocks from flows. The institution may have enough total asset value but insufficient liquid flow capacity at the speed claims are leaving. Converting assets faster can reduce their realized value, making liquidity stress feed solvency stress^[4].

Knowledge Transfer¶

Within finance the mechanism transfers literally across claimant classes and technologies. A wholesale run has no retail queue, yet short claims, finite liquidity, rollover refusal, and first-mover advantage remain. Digital access changes speed but not identity. Transfer outside finance should be conservative: an exit cascade is only analogous unless it preserves redeemable claims, par-liquidity expectations, and maturity transformation.

Examples¶

Canonical¶

A deposit-taking bank funds long-term loans with deposits withdrawable on demand. A rumor prompts some depositors to transfer funds. The bank uses cash, then sells loans or securities at discounts. Those sales reduce the resources available to remaining depositors, making their early withdrawal increasingly rational. A liquidity event can thereby create balance-sheet losses.

Wholesale¶

A dealer finances longer-lived assets through overnight repo. A few professional counterparties observe the same adverse information and refuse to renew large tranches. No one stands in line, but the loss of rollover drains funding within days^[5]. This is the Wholesale Funding Run species.

Structural Tensions¶

Liquidity versus solvency: each can cause or amplify the other, but they remain analytically distinct.
Fundamentals versus coordination: poor assets can trigger a run, while expectations can drain a solvent institution.
Transparency: information can reassure claimants or synchronize exit.
Backstop credibility: an announced facility works only if claimants believe its size, speed, and legal reach.
Exit restrictions: gates preserve liquidity after activation but can create a race to leave before activation.

Structural–Framed Character¶

Bank Run is domain-specific. The equilibrium and feedback skeletons are portable, but redeemable financial claims, maturity transformation, par liquidity, reserves, rollover, and lender backstops are constitutive.

Structural Core vs. Domain Accent¶

Remove deposits, creditors, par redemption, and illiquid financial assets. What remains is a first-mover coordination problem in which exit depletes a shared pool and makes further exit rational. That structural residue is carried by Coordination Problem and Equilibrium Selection; the financial claim machinery defines this domain node.

Funding Fragility is the strict enabling condition.
Coordination Problem and Equilibrium Selection supplies the multiple-equilibria skeleton through Funding Fragility.
Information Cascade is only a conditional mechanism when sequential observation suppresses private signals.
Self-Fulfilling Prophecy applies to some confidence-driven runs but not universally to runs revealing genuine insolvency.
Fire Sale is a frequent downstream consequence rather than the event's genus.

Relationships to Other Abstractions¶

Current abstraction Bank Run Domain-specific

Parents (2) — more general patterns this builds on

Bank Run presupposes Funding Fragility Domain-specific

A bank run presupposes runnable short-funded claims against assets that cannot meet simultaneous withdrawal at par.
Bank Run is a decomposition of, typical Expectancy-Mediated Outcome Prime

Bank runs typically instantiate expectancy-mediated outcomes because anticipated withdrawal by others makes early withdrawal individually attractive and thereby helps produce the run that was anticipated.

Children (1) — more specific cases that build on this

Wholesale-Funding Run Domain-specific is a kind of Bank Run

Wholesale-Funding Run is Bank Run specialized to concentrated professional creditors refusing to roll short-tenor funding under shared information.

Hierarchy paths (11) — routes to 9 parentless roots

Bank Run → Funding Fragility → Coordination Problem and Equilibrium Selection → Coordination → Concurrency

Show alternative paths (10)

Neighborhood in Abstraction Space¶

Bank Run sits in a sparse region of the domain-specific corpus (97^th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Unclustered & Miscellaneous (1565 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-09-08

Not to Be Confused With¶

Funding Fragility is the pre-event runnable balance-sheet structure.
Insolvency is a deficit of asset value relative to liabilities.
Fire Sale is forced liquidation into a price-depressed market.
Wholesale Funding Run specializes claimant type, funding tenor, and rollover mechanism.
Liquidity Freeze or Credit Crunch is a broader market-wide withdrawal of lending.

References¶

[1] Diamond, Douglas W. and Dybvig, Philip H. “Bank Runs, Deposit Insurance, and Liquidity”. Journal of Political Economy, 1983. The sequential-service payout at the centre of the Diamond-Dybvig model, under which claimants are paid in arrival order out of a finite pool, so that withdrawing early raises one's own expected recovery and lowers everyone else's. The Diamond-Dybvig result that the run equilibrium arises for a bank whose assets held to maturity would cover its deposits, so that premature liquidation, not prior insolvency, creates the shortfall. registry ↩a ↩b

[2] Li, et al. “Runs and Interventions in the Time of COVID-19: Evidence from Money Funds”. SSRN Electronic Journal, 2020. Evidence for two of the five interventions named here: redemption gates and fees accelerated rather than slowed exit from prime money funds in March 2020, and the Federal Reserve's MMLF liquidity backstop stopped that run. The deposit-insurance, liquidity-requirement and funding-maturity clauses rest on other work. registry ↩

[3] Goldstein, Itay and Pauzner, Ady. “Demand-Deposit Contracts and the Probability of Bank Runs”. The Journal of Finance, 2005. The global-games treatment in which fundamentals uniquely select the equilibrium and thereby fix a computable probability of a run, while the runs so selected remain panic-driven rather than fundamentals-forced. registry ↩

[4] Shleifer, Andrei and Vishny, Robert. “Fire Sales in Finance and Macroeconomics”. Journal of Economic Perspectives, 2011. The fire-sale channel: a forced sale clears at a dislocated price because the natural buyers are constrained sellers themselves, and the resulting mark-downs deplete the selling institution's balance sheet. registry ↩

[5] Gorton, Gary and Metrick, Andrew. “Securitized banking and the run on repo”. Journal of Financial Economics, 2012. The wholesale form of the run - repo counterparties raising haircuts and declining to renew, with no depositor queue - documented across the 2007-2008 panic; the paper measures the run through spreads and haircuts rather than timing how fast a single dealer's funding drains. registry ↩