Timing clarification: Capital One is scheduled to release second-quarter results today, July 21, 2026, at approximately 4:05 p.m. ET, followed by its earnings call at 5:00 p.m. ET. The event date provided—July 21—is today rather than tomorrow.
This quarter is less about headline EPS—which will remain distorted by Discover integration costs, purchase accounting, and the addition of Brex—and more about whether Capital One can demonstrate that its enlarged franchise is beginning to generate the promised operating leverage.
The most important questions are:
The setup is mixed. Credit indicators through May were encouraging, loan balances grew from March, and Q2 should benefit from seasonal balance growth and debit-network revenue synergies. Against that, expenses should step higher, the acquired Discover portfolio is still in a temporary growth “brownout,” and management’s limited quantitative guidance has made it difficult for investors to underwrite normalized returns.
| Metric | 1Q26 |
|---|---|
| Reported diluted EPS | $3.34 |
| Adjusted diluted EPS | $4.42 |
| Net revenue | $15.2 billion |
| Pre-provision earnings | $6.8 billion |
| Provision for credit losses | $4.1 billion |
| Net charge-offs | $3.8 billion |
| Allowance build | $230 million |
| Net interest margin | 7.87% |
| Adjusted efficiency ratio | 49.71% |
| CET1 ratio | 14.4% |
| Period-end loans | $447.8 billion |
| Period-end deposits | $489.1 billion |
First-quarter reported earnings included $477 million of Discover amortization expense and $415 million of Discover integration expense, reducing diluted EPS by a combined $1.08. Similar noise is likely in Q2, now with Brex purchase accounting added to the mix.
Investors should therefore prioritize adjusted pre-provision earnings, adjusted efficiency, underlying credit costs, and capital generation over reported EPS alone.
Capital One entered Q2 with a relatively constructive consumer-credit backdrop. In Q1:
Monthly data subsequently showed further improvement in card credit:
| Metric | April 2026 | May 2026 |
|---|---|---|
| Domestic Card net charge-off rate | 4.94% | 4.82% |
| Domestic Card 30+ day delinquency rate | 3.44% | 3.33% |
| Auto net charge-off rate | 1.20% | 1.45% |
| Auto 30+ day delinquency rate | 4.02% | 4.24% |
The card data suggest Q2 losses should compare favorably with Q1, although June can influence the quarterly result and monthly charge-offs are sometimes affected by recovery and debt-sale timing.
Auto deserves closer scrutiny. The May metrics weakened from April, and Capital One has been growing originations while accepting a somewhat higher subprime mix. Management previously attributed the Q1 auto reserve build to strong growth, mix, and a modestly weaker vehicle-value outlook. Investors should listen for any change in used-car prices, payment performance, or the quality of newer vintages.
Q1 NIM fell 39 basis points sequentially to 7.87%, largely because of:
Management indicated that Q2 receives approximately nine basis points from one additional day. It also expected the $76 billion quarter-end cash position to decline due to tax payments, debt maturities, and normalization after seasonal deposit inflows.
Domestic Card balances already increased from $254.0 billion at March 31 to $258.0 billion at May 31, while auto loans rose from approximately $85.7 billion to $88.0 billion. That should improve earning-asset mix and support sequential net interest income.
The key question is whether excess liquidity actually came down. Continued strong deposit gathering is strategically favorable, but if cash remains elevated, it could again dilute NIM.
Base case: a sequential NIM recovery driven by calendar days, card balance growth, and lower cash, though not necessarily a full return to late-2025 levels.
Q1’s adjusted efficiency ratio of 49.71% benefited from seasonally low marketing and the timing of planned investments. Management explicitly said marketing would be heavier in Q2 and later quarters.
Q2 will also include:
That makes a sequential increase in both marketing and operating expenses likely. An adjusted efficiency ratio above Q1’s 49.7% would not necessarily be a thesis breaker, but management needs to demonstrate that the increase is planned and attached to measurable growth opportunities.
The market will be less forgiving if higher expenses are accompanied by weak revenue, delayed synergies, or another refusal to provide any bridge toward normalized returns.
Investors should separate:
Expense synergies are back-loaded because they depend on technology conversions and application decommissioning. Management has said they should be fully realized only after integration is completed in the first half of 2027.
Capital One continues to target $2.5 billion of total Discover synergies by mid-2027.
The revenue side is further advanced:
The expense side remains more dependent on future systems conversions.
The temporary weak point is Discover loan growth. Legacy Discover card balances were down 1.2% year over year in Q1 following earlier underwriting cutbacks and additional Capital One tightening in higher-balance revolvers. Management expects that “brownout” to intensify somewhat before the technology integration is completed.
Key milestones to monitor:
For Q2, investors should not demand strong legacy Discover balance growth. They should demand on-time migration milestones, stable customer experience, realization of debit synergies, and continued confidence in the $2.5 billion target.
Capital One completed the Brex acquisition on April 7 for approximately $4.5 billion. Management previously estimated that the transaction would reduce Q2 CET1 by a little over 40 basis points.
Q2 will provide the first detailed look at:
Management is intentionally pursuing an “enablement” strategy rather than an immediate, deep integration. That could preserve Brex’s growth culture, but it also means investors should expect investment spending to precede much of the earnings benefit.
A constructive update would quantify at least some combination of Brex’s revenue run rate, growth, funding benefits, investment needs, or path to profitability.
Capital One ended Q1 with a 14.4% CET1 ratio after repurchasing $2.5 billion of stock. The estimated Brex impact would mechanically bring that closer to 14.0% before considering Q2 earnings, balance-sheet changes, and additional repurchases.
That still appears well above the 12.5% capital level management uses when discussing the post-Discover ROTCE framework.
Investors should watch:
Another $2.5 billion buyback would signal confidence in earnings and capital durability. A meaningful slowdown may indicate either greater macro caution or a preference to preserve capacity for growth and integration.
COF closed July 20 at $206.81, down approximately 16.6% from January 2, compared with declines of roughly 5.6% for American Express and 13.3% for Synchrony. COF is only about 2.1% above its April 21 close following the Q1 report.
The underperformance suggests the market is already discounting some combination of:
That lowers the bar for the headline quarter somewhat, but it raises the importance of forward-looking clarity.
The likely operating story is better seasonal revenue and NIM, continued healthy card credit, and materially higher expenses. That could still produce a solid quarter, but the stock’s reaction will depend less on reported EPS than on whether management improves visibility into the post-integration earnings model.
The cleanest positive signal would be a combination of stable credit, sequential NIM expansion, controlled adjusted expenses, on-schedule Discover conversions, and a clearer bridge to normalized ROTCE. The principal risk is that Capital One continues to add attractive long-term investments faster than investors can see the resulting earnings power.