Wall Street Banks Head Into Record Q3 Earnings as Yields Peak

The setup: a record year, and one more test of it

The big U.S. banks open third-quarter earnings season on October 13, and they arrive with more wind at their back than at any point since the post-pandemic trading boom. New York State Comptroller Thomas DiNapoli's office reported this week that Wall Street firms booked $45.9 billion in profits in the first half of 2026 — 51.3% more than the same period last year, and already more than the $45.3 billion forecast for the entire year2. If the first-half pace holds, annual profits could exceed $90 billion, comfortably surpassing 2025's record of $65.1 billion and rivaling the 2009 peak even after adjusting for inflation238.

That backdrop frames this week's reports as less a question of whether the banks made money and more a question of how durable the boom is. JPMorgan Chase, Goldman Sachs, Wells Fargo and Citigroup report Tuesday, October 13; Bank of America and Morgan Stanley follow on October 14; U.S. Bancorp and Charles Schwab close out the mega-cap slate on October 157.

What Wall Street is actually earning

The profit surge is not one engine but several firing at once. Underwriting fees — the money banks earn helping companies sell stocks and bonds — jumped 68% in the first half of 2026 versus a year earlier, tracking a 76.5% surge in global equity issuance, a large chunk of it from the SpaceX initial public offering, which was the biggest IPO on record25. Debt issuance rose 11.3%, with the AI hyperscalers among the biggest borrowers25. Global M&A set a record for any half-year, and equity issuance hit its highest level since 2021, according to the Comptroller's report5.

The trading desks have kept up their end. Goldman Sachs has posted three consecutive all-time quarterly records for stock trading at any bank, and Jefferies — the firm that reports ahead of everyone else and is watched as a bellwether — delivered record third-quarter investment banking net revenues of $1.33 billion, up 17% year over year, with advisory revenues up 25% to $818 million and equity underwriting up 69% to $306 million46. Global dealmaking has already crossed $4 trillion this year, and Jefferies flagged a healthy backlog heading into 20276.

The yield problem hiding in plain sight

What makes this earnings season genuinely unusual is the interest-rate environment underneath it. The 10-year Treasury traded above 5.3% during the quarter, at levels not seen in over two decades, while the 30-year bond yield climbed above 5.7% to touch 24-year highs1. Elevated long yields normally compress bank profitability over time — they raise funding costs, pressure asset valuations on the government bonds banks hold on their books, and compete directly with equities for investor capital17.

So far, the banks have been riding the transition rather than suffering it. Steep yields create profitable carry opportunities, and the volatility that comes with a bond market that doesn't believe the inflation fight is over has generated substantial fee income on the trading desks1. But the strain is starting to show at the margins. JPMorgan's net interest income — the spread between what it earns on loans and what it pays depositors — has been the anchor of its earnings power, and the firm has guided to roughly $105.5 billion of it for full-year 2026, raised from $103 billion after a blowout second quarter131419. That is a big number, but the growth rate on the core banking engine is no longer uniformly accelerating: in the second quarter, NII excluding Markets was up only 4% year over year19.

The catch, as one earnings-season primer put it plainly, is that savers aren't sitting still — with Treasury bills paying around 4%, money migrates out of zero-interest checking, and banks have to pay up to keep deposits7. Watch whether interest income keeps growing as fast as rates. Wells Fargo has guided to about $50 billion of net interest income for the year7.

What analysts expect on October 13

Consensus looks for JPMorgan to earn roughly $5.82 to $5.84 per share on about $50.6 billion in revenue, with the bank guiding to investment banking and markets revenue up mid-to-high teens from a year ago714. Goldman's consensus is unusually dispersed — aggregators cluster between roughly $13.30 and the mid-teens, on net revenue near $17 billion to $18 billion — reflecting genuine uncertainty about summer trading volumes after a second quarter in which diluted EPS surged 92% year over year to $20.981517. Wells Fargo is expected near $1.85; Bank of America near $1.17, though BofA has guided to investment banking fees of $1.6 billion to $1.8 billion, down 10% to 20% year over year, a divergence Jefferies' record quarter makes look like timing rather than demand67.

The consensus bets, in short, are on a seasonal step-down from an extraordinary first half, but still-solid year-over-year growth. The one number that matters more than any headline: provisions for credit losses. That is where a weakening consumer shows up first, and there are warning signs that it is weakening — American households now owe more than $1 trillion on credit cards, with card delinquencies at levels not seen since the aftermath of the Great Recession, and total household debt at a record $18.8 trillion as of the first quarter5. JPMorgan's card charge-off outlook for 2026 was lowered to about 3.2%, suggesting contained risk for now1419.

The regulatory tailwind nobody is talking about enough

The other big story under the surface is that Washington is loosening the leash. The FDIC's third-quarter 2026 Call Report instructions implement a final rule that lowers the community bank leverage ratio from 9% to 8% and extends the grace period from two quarters to four2123. More consequentially for the firms reporting this week, the OCC, Fed and FDIC released a comprehensive set of proposals in March to recalibrate capital requirements — replacing the current dual standardized/advanced framework with a single consolidated framework for the largest banks, and revising the GSIB surcharge methodology to be more risk-sensitive and eliminate cliff effects26. The agencies estimate that, combined with the stress-testing changes, the proposals would produce a net 5% decline in CET1 requirements for Category I and II holding companies — money that flows straight toward lending capacity and buybacks26.

The Fed is also preparing to reindex the asset thresholds — $100 billion, $250 billion, $700 billion — at which banks face stepped-up stress tests and reporting, adjusting for inflation and growth since 2019. Under a nominal-GDP reindexing, the top threshold could move to roughly $960 billion and the lower bar to about $150 billion, freeing banks like U.S. Bancorp, Capital One, PNC and Truist to grow without triggering the most intrusive oversight, and potentially unleashing a wave of mid-size consolidation29. The Fed has separately finalized stress-test scenarios and stress capital buffer mechanics, moving the buffer notification date to September 30 and its effective date to January 1, alongside a final rule adding a global market shock component to the annual test25.

For the banks reporting this week, capital flexibility translates directly into shareholder returns: JPMorgan announced a $50 billion buyback program and intends to raise its quarterly dividend to $1.65, supported by a 14.1% standardized CET1 ratio19.

The reading that matters

Pull it together and one interpretation holds up better than the others. Wall Street's record year is real, but it is a narrow, market-driven boom — trading volatility, AI-driven underwriting and debt issuance, reopened M&A — layered on top of a rate environment that is quietly turning against traditional spread lending. The Comptroller himself flagged AI, inflation and global conflicts as growing concerns posing increasing risk to public finances should the industry stumble810. The divergence within the group is telling: JPMorgan expects mid-to-high-teens growth in investment banking fees while Bank of America expects a double-digit decline67. When the whole sector's results are this dependent on one bank's trading franchise and one technology sector's borrowing binge, the record is more fragile than the $90 billion headline suggests.

The bullish case — that deregulation plus a Fed that hiked in September but is now unlikely to move again soon, with October hike odds collapsing after weak jobs data, gives banks a prolonged window of elevated margins and freed-up capital129 — is genuinely strong. But the vigilance points are equally clear: whether net interest income keeps growing as deposits reprice upward, whether provisions keep ticking higher at JPMorgan, and whether the bond portfolios on bank balance sheets keep bleeding paper value as the 30-year sits at 24-year highs719. Bank earnings matter beyond bank investors because banks see the real economy first — who is paying cards, who is borrowing, who is doing deals7. This quarter, what they report will tell us whether the AI-and-volatility gold rush is broad enough to carry a consumer that is increasingly paying for it in interest.