Stress Tests, CCAR, and the Stress Capital Buffer

Introduction

The Federal Reserve stress tests are the annual exercise that determines how much capital each large US bank can return to shareholders, making them the single most consequential regulatory event on the FIG calendar. Every June, the Fed publishes stress test results for banks with over $100 billion in assets. Every August, it announces stress capital buffer (SCB) requirements derived from those results, effective October 1. And every June through August, banks announce dividend increases and buyback authorizations calibrated to their resulting capital headroom. For FIG analysts, stress test outcomes are not abstract regulatory outputs; they are the direct input to DDM payout assumptions, excess capital deployment analysis, and M&A feasibility screening.

The Dodd-Frank Act Stress Tests (DFAST) subject large bank holding companies to a hypothetical severely adverse economic scenario projected over nine quarters. The Fed designs the scenario, runs its own models against each bank's balance sheet and income statement, and publishes the results.

The 2025 severely adverse scenario assumed unemployment spiking to 10% (a 5.9 percentage point increase from Q4 2024), real GDP declining 7.8% through Q1 2026, and house prices falling 33% through Q3 2026. Under these assumptions, the Fed projects each bank's pre-provision net revenue (PPNR), loan losses, trading losses, and capital ratios quarter by quarter. The critical output is the peak-to-trough CET1 decline: the maximum drop from the starting CET1 ratio to the lowest point during the stress period.

In 2025, the aggregate CET1 decline across all tested banks was 2.7 percentage points (versus 2.8 percentage points in 2024), with individual bank results varying significantly based on loan portfolio composition, trading book risk, and fee revenue resilience.

Stress Capital Buffer (SCB)

The SCB is each bank's individualized capital buffer, derived directly from stress test results. It equals the bank's peak-to-trough CET1 decline in the severely adverse scenario plus four quarters of planned common stock dividends, floored at 2.5% of risk-weighted assets. The SCB replaced the old CCAR qualitative assessment (which included subjective "pass/fail" judgments on capital planning processes) with a purely quantitative framework. A bank's total CET1 requirement is the 4.5% minimum plus the capital conservation buffer (replaced by the SCB, minimum 2.5%) plus the G-SIB surcharge if applicable. The SCB is announced in August and takes effect October 1, remaining in force until the following year's results. Banks that breach their SCB face automatic restrictions on dividends, buybacks, and discretionary bonus payments through the Maximum Distributable Amount (MDA) framework.

The 2025 stress test produced favorable results for most large US banks, reflecting both improved loss resilience and a less severe unemployment shock than the 2024 scenario (5.9 percentage points versus 6.3).

The most dramatic shift was Goldman Sachs, whose SCB fell from 6.2% to 3.4%, reducing its total CET1 requirement from 13.7% to 10.9%. Goldman's 2024 result had been driven by outsized trading losses in the stress scenario; the 2025 scenario produced more moderate trading stress, and Goldman successfully requested reconsideration of specific model assumptions. The reduction freed approximately $35 billion in excess capital relative to requirements, enabling Goldman to boost its quarterly dividend by 33.3% and expand its buyback authorization.

The SCB framework creates a hard mathematical link between stress test outcomes and capital distribution capacity. A bank's distributable capital is the amount above its total CET1 requirement (minimum + SCB + G-SIB surcharge) plus any management buffer (typically 50-150 basis points above the regulatory requirement). Everything above that threshold can theoretically be returned through dividends and buybacks.

JPMorgan's 2025 position illustrates the mechanics. With a CET1 ratio of approximately 14.5% and a total requirement of approximately 11.5%, it holds approximately 300 basis points of excess. On approximately $2 trillion in RWA, that translates to roughly $60 billion in excess capital, which JPMorgan is deploying through its $50 billion buyback authorization and a 7% dividend increase. Bank of America, with only approximately 140 basis points of excess on its requirement, has significantly less distributable capacity, constraining its buyback programs relative to peers.

For M&A, the constraint operates through pro forma capital analysis. An acquisition that creates goodwill (the premium above tangible book value) directly reduces CET1 because goodwill is deducted from regulatory capital. If the pro forma CET1 ratio would fall near or below the SCB-inclusive requirement, the deal cannot proceed without a compensating capital raise. FIG bankers model the interaction between acquisition pricing (which determines goodwill), pro forma capital ratios, and the acquirer's SCB to determine the maximum affordable premium.

The stress test framework also intersects with Basel III Endgame in important ways. If the Endgame reproposal changes how RWA is calculated (through new standardized approaches or the output floor), the stress test denominator changes, potentially altering SCB outcomes. FIG analysts monitoring the Endgame must also model the second-order effects on stress test capital requirements, creating a complex interaction between two regulatory frameworks that are evolving simultaneously.

Stress tests are the bridge between the static capital requirements of Basel III and the dynamic reality of bank capital management. They translate theoretical loss scenarios into binding capital constraints that directly determine dividends, buybacks, M&A capacity, and ultimately the valuation of every large US bank.