Goldman Sachs (NYSE: GS) made the analyst consensus look like it was set in a different market. Net revenue of $18.68 billion arrived $2.0 billion above the $16.83 billion estimate, and diluted EPS of $17.01 crushed the $12.86 consensus by 32%. This beat wasn't a one-line-item story. It was broad, deep, and concentrated in the businesses where Goldman has spent years building scale.
Revenue rose 23% year over year, accelerating from the 15% growth seen in the first half. The engine was Global Banking & Markets, with net revenue hitting $13.46 billion, up 32% from a year ago. Equities stood out. Revenue from the Equities franchise more than doubled in some sub-lines: Equities financing hit $3.37 billion, nearly doubling from $1.72 billion a year ago, and Equities intermediation rose 52% to $3.08 billion. The combined $6.44 billion in Equities revenue represents 34% of total firm revenue, up from 25% a year ago. That is not a cyclical uptick. It reflects structural market share gains in prime brokerage and derivatives intermediation that compound across rising volumes.
Investment banking fees of $3.22 billion rose 21% year over year. Advisory set a record at $1.65 billion. Equity underwriting jumped 43%, and Debt underwriting rose 15% on investment-grade strength. The backlog declined sequentially, as is typical after a strong quarter, but the decline was modest. The advisory record may not stand for long.
Asset & Wealth Management contributed $5.05 billion in revenue, a 14% year-over-year increase. Management and other fees hit a record $3.59 billion, up 22%, driven by higher average assets under supervision. Total AUS reached a record $4.10 trillion, marking the 35th consecutive quarter of long-term fee-based net inflows. This fee stream is becoming a more predictable counterweight to trading and banking volatility. Incentive fees surged to $587 million from $76 million a year ago, driven by fund distributions, but that line is inherently lumpy.
The only blemish was Platform Solutions. Revenue there collapsed 72% to $170 million, reflecting net markdowns on the Apple Card loan portfolio transferred to held for sale in Q4 2025. That segment is in wind-down mode. The losses are contained and declining.
Operating expenses rose 20% year over year to $11.31 billion, pushed by higher compensation and transaction-based costs. The compensation ratio was 27.4% of revenue, up from 26.8% a year ago, but still well below the 30%+ levels that once defined Goldman. The first nine months' efficiency ratio improved to 59.4% from 62.1% a year ago. The firm is getting more operating leverage, even as it invests in headcount (up 3% sequentially on campus hires) and technology.
The capital return story is where the quarter gets interesting. Goldman returned $4.49 billion to common shareholders, including $3.0 billion in share repurchases (2.9 million shares at an average cost of $1,048.73). The buyback pace looks aggressive relative to the earnings trajectory. The standardized CET1 ratio fell to 12.4% from 12.9% last quarter, driven by higher risk-weighted assets. The advanced CET1 ratio held at 13.6%. The supplementary leverage ratio remained at 4.3%. The buyback is consuming capital faster than organic generation, narrowing the CET1 buffer above regulatory minimums. That is a deliberate choice, but it leaves less room for error if markets turn.
The ROE of 18.7% for the quarter and 20.7% year-to-date are well above Goldman's medium-term target of 15-17%. A $450 million charitable contribution to Goldman Sachs Gives reduced EPS by roughly $1.00 and ROE by 1.1 percentage points. Strip that out, and the underlying ROE was closer to 20% for the quarter. That is the kind of number that justifies the premium valuation, but it also raises the bar for the next quarter.
What to watch next: the sustainability of the Equities financing surge. Prime brokerage revenue is sensitive to both client leverage and market levels. If volatility compresses or client risk appetite fades, that line can normalize quickly. The advisory backlog, while down sequentially, remains elevated by historical standards. The M&A cycle has further to run. The real question is whether Goldman can maintain this earnings power through a full cycle, or whether Q3 2026 will look like a peak in hindsight. For now, the firm is firing on most cylinders, and the buyback suggests management shares that confidence.