💹 Financial Contagion Network

Bank network: asset-liability links transmit shocks. When a bank's equity falls below zero it defaults, sending losses to creditors (Eisenberg-Noe clearing). See systemic risk and too-big-to-fail.

FinanceInteractive
Click bank to select · Shock Selected Bank · Propagate to spread losses

How it Works

This simulation models a network of banks connected by interbank loans. Each bank has external assets A, interbank receivables, external liabilities, and interbank liabilities. Equity E = A + receivables − liabilities. When equity drops below zero, the bank defaults.

The Eisenberg-Noe clearing mechanism computes how losses propagate. When bank i defaults, its creditors receive a proportional share of its remaining assets. If this causes a creditor's equity to turn negative, that creditor also defaults, triggering further rounds.

Equity: E_i = A_i + Σ_j π_ji · p_j − L_i Default: E_i < 0 Clearing vector p* = fixed point of: p_i = min(L_i, A_i + Σ_j π_ji · p_j) where π_ji = interbank liability share Cascade: iterate until no new defaults

Key insight: the topology matters. A complete network diversifies small shocks but creates systemic fragility under large shocks. A ring network localizes losses but can create chain-reaction cascades.

Frequently Asked Questions

What is financial contagion?

Financial contagion is the spread of financial distress across institutions through direct linkages (interbank loans, derivatives) or indirect channels (fire sales, confidence effects). A default by one bank imposes losses on its creditors, potentially triggering further defaults.

What is the Eisenberg-Noe clearing model?

The Eisenberg-Noe model (2001) computes the unique clearing payment vector in an interbank network. Each bank pays its obligations in proportion to its liabilities, subject to limited liability. The algorithm iterates until a fixed point is reached.

What is systemic risk?

Systemic risk is the risk that failure of one institution triggers a cascade of failures throughout the financial system. It arises from interconnectedness, common exposures, and liquidity spirals that can cause solvent institutions to become insolvent.

What does too-big-to-fail mean?

Too-big-to-fail refers to banks so large and interconnected that their failure would cause systemic collapse. Governments therefore bail them out, creating moral hazard — banks take excessive risks knowing they will be rescued.

How does network structure affect contagion?

Complete networks (every bank lends to every other) are more resilient to small shocks but amplify large ones. Ring networks concentrate losses. Core-periphery structures, common in real interbank markets, create a vulnerable core.

What is a default cascade?

A default cascade occurs when an initial bank failure causes losses to creditor banks, pushing them into default, which in turn causes further losses downstream. The cascade can amplify the original shock many times.

What triggers a financial shock in the model?

External shocks include asset value drops (real-estate crash, sovereign debt write-downs), withdrawal of wholesale funding, or sudden loss of market confidence. In this simulation you can select a bank and apply an asset shock.

What is the recovery rate in bank defaults?

The recovery rate is the fraction of a defaulted bank's liabilities that creditors actually receive. In fire-sale scenarios, asset liquidation below book value reduces recovery rates, amplifying contagion losses.

How did the 2008 crisis illustrate financial contagion?

The 2008 crisis began with US subprime mortgage losses but spread globally through complex structured products, counterparty risk, and liquidity freezes. Lehman Brothers' failure triggered a global credit crunch demonstrating real-world default cascades.

What policies reduce financial contagion?

Policies include capital requirements (Basel III), central clearing for derivatives, bail-in mechanisms that impose losses on creditors rather than taxpayers, and macroprudential oversight to monitor network exposures.

About this simulation

This simulation places a handful of banks in a ring, complete, random, or core-periphery interbank network, each with its own assets, equity, and interbank obligations, then lets you click a bank, hit it with an asset shock, and step through the Eisenberg-Noe clearing rounds to watch losses ripple through creditors. A bank turns red the instant its equity crosses zero, and each Propagate click resolves one more wave of the resulting default cascade.

🔬 What it shows

A circular graph of banks sized by assets, with green equity bars beneath each node, red highlighting for defaulted banks, orange for shocked-but-solvent ones, and edge colour showing which interbank links are actively transmitting losses.

🎮 How to use

Set bank count, interbank density, shock size, and capital ratio with the sliders, choose a topology from the dropdown, click New Network to rebuild it, click any bank node to select it, then Shock Selected Bank and Propagate repeatedly to watch the cascade unfold round by round.

💡 Did you know?

A fully connected (complete) interbank network can absorb small shocks better than a sparse ring network because losses are diversified across every creditor — but past a critical shock size, that same dense connectivity turns into a superhighway for contagion, spreading defaults to nearly everyone at once.

Frequently asked questions

Why doesn't a shock immediately default every connected bank?

Shocking a bank only changes its own assets and equity directly; neighbouring banks only take losses once you click Propagate, which runs one round of the Eisenberg-Noe clearing calculation and pushes a proportional share of the defaulted bank's shortfall onto its interbank creditors.

What determines how much a creditor bank loses?

Each defaulted bank's loss is split among creditors in proportion to the size of their interbank claim (l.amount / ibLiab), so a bank with a larger exposure to a failing counterparty absorbs a proportionally larger hit to its own equity.

Why does core-periphery topology behave differently from random?

In the core-periphery setting, the top third of banks are densely interlinked while peripheral banks connect to the core only about half the time — concentrating risk so that a shock to a core bank can cascade through many more creditors than the same shock hitting a peripheral one.

What does the capital ratio slider actually protect against?

Capital ratio sets each bank's equity buffer as a fraction of its assets; a higher ratio means a bank can absorb a bigger asset shock before its equity turns negative, so raising this slider makes the whole simulated system visibly more resistant to the same shock size.

Does the simulation ever run out of contagion on its own?

Yes — the cascade is a fixed-point process: each Propagate click checks whether any solvent bank's equity has just turned negative, and once a round produces no new defaults, further clicks change nothing, mirroring how Eisenberg-Noe clearing converges to a unique stable payment vector.