Where does a banking shock actually reach you?

Your operating cash is counterparty concentration.

Through the two accounts you never thought of as risk: your company’s operating cash and your personal deposits, both likely parked above the insured limit at one or two institutions. That is counterparty concentration — the same single-point-of-failure pattern as your equity, in a different layer of the stack. This is a modelled scenario, not a forecast, and it sits alongside the other scenarios we stress-test.

The 2023 episode made the channel concrete. Silicon Valley, Signature, and First Republic each relied heavily on uninsured deposits, and a founder’s runway can sit frozen for the days a resolution takes — exactly when you need to make payroll. The contagion mechanics echo the 2008 global financial crisis: one counterparty’s failure repriced confidence in every balance sheet that looked like it.

What does the exposure cost when it breaks?

Liquidity, at the worst possible moment.

It costs liquidity at the worst possible moment. Stanford economists estimate the market value of U.S. bank assets sits roughly $2.2 trillion below stated value, leaving a large set of institutions vulnerable to an uninsured-depositor run (Stanford SIEPR, 2026). The pressure is structural: a ~$2 trillion commercial-real-estate maturity wall refinances into high rates through 2027, and the first FDIC-insured failure of 2026 — Illinois’s Metropolitan Capital Bank & Trust, $261M in assets — was tied to CRE (Bisnow, 2026).

A frozen operating account during a lock-up you already cannot exit is two illiquidity events stacked in one session — and both correlate hard with today’s risk environment, where a U.S. debt and rate shock is the most plausible trigger for the next deposit run.

What does engineered resilience look like here?

A mapped counterparty surface.

It looks like a mapped counterparty surface. You stop assuming the bank is risk-free infrastructure and start reading it as a position: which institution holds what share of your cash, how much sits uninsured, and what a 48-hour resolution does to your runway. Exposure named, not discovered mid-run — then hedged with the same tail-risk instruments you would use on any other concentrated position.

How it works

The Sandbox Engine.

The scenario runs against your book the way you run a chaos test against production.

model: diffusion sampling · data: deposit + CRE-stress regimes · coverage: company + personal cash · stated failure rate: 4.2% false-comfort

It compiles extreme banking-stress paths — deposit flight, correlated regional failure, resolution delay — against your cash and equity layout, and shows where liquidity breaks first. The engine uses diffusion over fat-tailed distributions, not Gaussian VaR, which underweights the tail a run lives in. The published 4.2% false-comfort rate stays in view; a model that hides its error rate is the failure mode.

“My deposits are FDIC-insured — why model this?”

Because the insured limit is not where your money is.

For a founder holding company runway and personal liquidity, the balance above the cap is uninsured by definition, and resolution timing — not the eventual guarantee — is what freezes payroll. The diagnostic prices that gap so you can spread counterparties before a headline forces it.

3 fields · 48-hour document · no call, no sequence.