Search runs in your browser, across every published page on this site.

Jamie Dimon's fortress balance sheet model

Model Dimon's fortress balance sheet: hold a high equity ratio on a bank's assets, compare the return on equity against a thinner balance sheet, and see what the buffer costs and buys.

Jamie Dimon of JPMorganChase, photographed in 2025 Jamie Dimon Reviewed Aug 6, 2026 Rule located in JPMorganChase, 2026

Portrait: Lauren Hurley / No 10 Downing Street, Open Government Licence v3.0, via Wikimedia Commons. Self-hosted by JMM.

Interactive model

Put your numbers through the rule

The source establishes the rule. The values below belong to you, and the output is JMM's deterministic calculation.

Make the assumptions yours

Every field recalculates immediately. Changed values can be copied into a shareable URL.

Fortress capital buffer$130,000,000
Fortress ROE
21.7%
ROE at 8% equity
33.8%
Extra buffer dollars
$50,000,000
Fortress leverage
7.7×

ROE is modeled earnings divided by equity at your entered asset return and funding cost. It is not a forecast of any bank’s actual return.

Inspect the calculationThe same result in a readable record view

Fortress balance sheet

Equity ratio
13.0%
ROE
21.7%
Earnings
$28,250,000

Comparison balance sheet

Equity ratio
8.0%
ROE
33.8%
Earnings
$27,000,000
Source and translation

What the source says, and what the calculator adds

Source-supported ruleDimon’s 2025 letter treats the buffer as a permanent commitment rather than a cycle call — the firm raised its dividend twice "while continuing to reinforce our fortress balance sheet" — and its own exhibit shows the cost and the result: CET1 up from 7.0% before the crisis to 14.6% in 2025, with ROTCE still at 20%.
JMM calculationJMM quantifies the tradeoff: earnings at each entered equity ratio, the ROE each ratio produces, the extra buffer dollars, and the leverage implied by the fortress ratio.
Formula and methodEarnings = assets × asset return − (assets − equity) × funding cost. ROE = earnings ÷ equity. Buffer = assets × fortress ratio − assets × comparison ratio.
Decision notes

What changes the answer

The buffer is priced in ROE terms

A thinner balance sheet produces a higher modeled ROE at the same asset return. The calculator shows the exact cost of the fortress, which is the trade Dimon accepts.

The fortress pays off in the downturn

The value of surplus capital appears when losses arrive, which this steady-state model cannot show; that is why the limitation is stated beside the ROE.

Carrying surplus capital costs return every year and pays for itself once a decade. That trade has been settled by events.

This calculator shows the cost side plainly: at the same asset return, a thinner balance sheet produces a higher return on equity, so the fortress is expensive in every ordinary year. The reason to accept that bill is not modellable here, because it arrives as an option — the ability to buy a failing competitor at the bottom rather than become one. JPMorgan has exercised that option repeatedly, most recently in the 2023 regional bank failures, and the banks that optimised the other way are the ones it bought. JMM’s read: the argument is finished, and the honest caveat is that it is much easier to hold surplus capital when you are the largest bank in the country and cheap deposits arrive during every crisis.

Not the calculator’s limits. The rule’s.

  1. Capital ratios describe quantity, not quality

    A high ratio against assets that reprice badly is not safety. 2023 made the point precisely: the failures were institutions whose reported capital looked adequate right up until the securities behind it were marked.

  2. Fortress capital is partly a competitive weapon

    A buffer that lets the largest bank absorb rivals in a crisis produces private returns and concentration at the same time. The shareholder case and the systemic case are not the same case.

Limits

What this model does not know

  • The equity ratio here is equity over total assets. CET1 — the 14.6% in the cited exhibit — is measured against risk-weighted assets and is a different, larger-looking number. Do not read one as the other.
  • The model uses flat entered returns and funding costs with no credit losses, rate paths, or regulatory detail.
  • ROE here is modeled leverage arithmetic, not a forecast of any bank’s actual return.
  • Capital quality, resolution, and stress-testing regimes are outside the model.
Questions people ask

Before you use the result

Does JPMorgan run at exactly 13% equity over assets?

No. The 13% and 8% here are editable illustration inputs. The letter reports a 14.6% CET1 ratio for 2025, which is measured against risk-weighted assets, not total assets.

Is higher ROE always better?

Not under the fortress philosophy. The model shows the arithmetic; the source argues the buffer is worth the cost, and reports 20% ROTCE in 2025 while carrying it.

Take the answer further

The next question this page cannot answer

Keep researching Jamie Dimon