Skip to content
Company Deep DiveGS

[GS] Goldman Sachs: Can High Returns Preserve Capital?

Editorial illustration for [GS] Goldman Sachs: Can High Returns Preserve Capital?
Published 11 min read

Summary

Goldman Sachs posted $20.34 billion of Q2 revenue and 23.5% ROE; Q3 will test whether banking and financing growth can persist without further CET1 erosion.

Goldman Sachs provides investment banking, markets, asset management and wealth management services to companies, institutions, governments and individuals. Its Goldman Sachs Q3 2026 earnings call is scheduled for 2026-10-13 and covers FY2026 Q3 ended 2026-09-30.[1] In Q2, net revenue was $20.34 billion, up 39%, net income was $6.63 billion and annualized ROE was 23.5%; Goldman gave no quarterly revenue or profit guidance and only expected a roughly 20% full-year 2026 effective tax rate.[2][3]

Three issues matter most in the coming results. First, investment-banking fees must continue converting after Q2's $3.395 billion and a higher backlog because transactions produce fees only when completed or priced. Second, Equities and FICC financing must preserve high returns as the balance sheet expands because financing balances also consume capital. Third, fee growth must move Asset & Wealth Management's first-half pre-tax margin above 24% toward the roughly 30% medium-term target because that determines whether the revenue mix becomes less trading-dependent.[2][3]

Company Background and Business Structure

Goldman Sachs is a global financial institution founded in 1869 and headquartered in New York, with capital markets and institutional clients still at its center. It reports Global Banking & Markets, Asset & Wealth Management, and Platform Solutions; their 2025 net revenues were $41.453 billion, $16.679 billion and $151 million, respectively, for a total of $58.283 billion.[4]

The three segments have distinct revenue engines. Global Banking & Markets includes advisory, equity and debt underwriting, FICC and Equities intermediation and financing, and transaction banking since Q4 2025. Asset & Wealth Management includes management fees, incentive fees, private banking and lending, and investment results. Platform Solutions mainly contains Apple Card and exited-business remnants, and Goldman has agreed to transfer the program to another issuer.[4]

Financial History and Current Position

Revenue and profit rose in 2025, but expenses grew faster. Full-year net revenue was $58.283 billion, up 9%, net income was $17.18 billion, diluted EPS was $51.32 and ROE was 15.0%; operating expenses were $37.54 billion and the efficiency ratio rose from 63.1% to 64.4%. Goldman returned $16.78 billion to common shareholders, including $12.36 billion of buybacks and $4.42 billion of dividends, and ended the year with a 14.3% Standardized CET1 ratio.[4]

Profit accelerated in the first half of 2026 while capital use increased. Q2 net revenue was $20.34 billion, net income was $6.63 billion, diluted EPS was $20.98, and annualized ROE and ROTE were 23.5% and 25.5%. First-half net revenue was $37.565 billion, net income was $12.26 billion, the efficiency ratio was 58.8%, credit-loss provision was $417 million, and capital returns were $11.74 billion.[2][3]

Balance-sheet changes limit a direct extrapolation of those high returns. Total assets rose from $1.809 trillion at year-end 2025 to $2.128 trillion in Q2 2026, deposits were $558 billion, the Standardized CET1 ratio fell to 12.9%, and the supplementary leverage ratio fell to 4.3%. Q1's 13.2% effective tax rate also benefited from $895 million of employee share-award tax benefits, which added $2.91 to quarterly EPS.[2][5]

Operating Model

Goldman's net revenue is driven by completed transactions, client activity, financing balances and fee-bearing assets. Advisory and underwriting fees are recognized after completion or pricing, intermediation follows client trading activity, and financing revenue comes from client balances. Management fees approximate average fee-bearing AUS multiplied by the blended fee rate, while private banking and lending also depend on deposit spreads.[3][4]

Pre-tax profit is net revenue less credit-loss provision and operating expense. Compensation and transaction-based expenses increase with revenue, but year-to-date compensation accruals can adjust more slowly when revenue falls. Net income then depends on the effective tax rate and discrete items such as employee share-award tax benefits.[3][5]

A dealer bank's net income cannot be treated directly as distributable cash. Repos, reverse repos, securities lending and client receivables and payables create large operating-cash-flow swings, while distributions ultimately depend on CET1 and leverage headroom. Goldman returned $11.74 billion in the first half of 2026 as total assets expanded by about $319 billion from year-end 2025.[2][3]

Industry and Competitive Position

Goldman competes with large universal banks and independent investment banks, with its strength concentrated in advisory, underwriting and institutional markets. It said it ranked first in announced and completed M&A, equity and equity-related offerings, and leveraged lending in Q2 2026, second in high-yield debt, while Equities set a record.[2] Those self-reported rankings show breadth but do not replace comparable cross-bank revenue shares; Goldman's heavier markets and investment-banking mix can also create greater quarterly volatility than a universal bank.

Core Debates

What sustains Goldman Sachs investment-banking fees after a record advisory and underwriting quarter?

Investment-banking durability depends on backlog converting into completed transactions. Q2 fees were $3.395 billion, up 55%, including $1.378 billion of advisory, $985 million of equity underwriting and $1.032 billion of debt underwriting; backlog at June was higher than at December 2025.[3] Incremental fees reach segment pre-tax profit only after compensation and transaction expense, so all three fee lines, backlog direction and Global Banking & Markets pre-tax profit must be read together. If backlog remains higher while fees decline for two consecutive quarters, the current understanding would weaken.

Is Goldman Sachs financing revenue a repeatable balance-driven stream or a cyclical one that fades with market activity?

Financing quality must be judged alongside capital returns. Q2 Equities financing revenue was $3.259 billion and FICC financing was $1.216 billion; Global Banking & Markets generated $15.520 billion of net revenue, $7.499 billion of pre-tax profit and 27.5% ROE, versus 14.4% a year earlier.[3] Client financing balances lift revenue but also consume risk-weighted assets and leverage exposure. Financing growth with lower segment ROE, or financing and intermediation falling together after volatility declines, would weaken the case for stability.

Can fee growth and rising assets under supervision lift Goldman Sachs Asset and Wealth Management toward its disclosed roughly 30% pre-tax margin?

Durable Asset & Wealth Management revenue depends on net inflows, fee levels and cost together. Q2 management and other fees were $3.355 billion, up 20%, ending AUS was $4.041 trillion and long-term net inflows were $91 billion; first-half segment pre-tax margin was 24%, against Goldman's roughly 30% medium-term target.[2][3][4] Net inflows, rather than market appreciation alone, can keep expanding the fee base. AUS growth with single-digit fee growth, or a margin stuck near 24%, would weaken the assessment.

After simultaneous balance-sheet growth and large buybacks, how much capital headroom supports Goldman Sachs returns above 20%?

The sustainability of high returns depends on both earnings quality and capital headroom. Q2's 12.9% Standardized CET1 ratio was below 14.3% at year-end 2025, while Goldman disclosed a 10.9% total requirement at December 2025; total assets reached $2.128 trillion, quarterly capital returns were $5.36 billion and ROTE was 25.5%.[2][4] Financing growth and distributions both pressure capital ratios, while share-award tax benefits added 1.7 percentage points to first-half annualized ROE. ROE holding up while CET1 continues to fall would show that some returns are being purchased with headroom.[3]

Risks and Falsifiers

Concentration in capital-market activity is the most direct cyclical risk to earnings. If M&A and issuance, volatility and financing demand fall together while compensation and the expanded balance sheet do not contract as quickly, revenue and pre-tax profit would decline and the first-half 58.8% efficiency ratio would rise. An efficiency ratio no higher than 62% and ROTE above 20% during quieter markets would weaken that risk.[3][4]

The consumer exit can still leave markdowns and contract costs. The Apple Card transfer remains incomplete, and related markdowns reduced 2025 Platform Solutions net revenue by $2.26 billion, leaving only $151 million for the year. Two consecutive quarters without another related markdown and with a narrower pre-tax loss would falsify the residual-drag concern.[3][4]

Transaction delays can break the transmission from backlog to revenue. Closed issuance windows or delayed large deals would reduce investment-banking fees and segment pre-tax profit while compensation accruals adjust with a lag. Two quarters of simultaneous year-over-year growth in advisory and underwriting with improving segment pre-tax profit would weaken the concern.[3][4]

Financing expansion can consume capital headroom. Client balances may be hard to shrink quickly in a downturn, while higher risk-weighted assets and leverage exposure lower CET1 and SLR and constrain buybacks. Two quarters of financing growth with a stable Standardized CET1 ratio and segment returns above 20% would falsify the risk.[2][3]

AUS growth may not produce equivalent fee growth. If new assets concentrate in lower-fee liquidity products or advisory accounts, management-fee growth would slow and the margin could remain near 24%. Two quarters of double-digit fee growth and a margin above 24% would strengthen the durable-revenue assessment.[3][4]

Regulatory changes may narrow the room for simultaneous buybacks and balance-sheet growth. A higher stress capital buffer when it is reset in 2027, or continued risk-weighted-asset expansion, could pressure buybacks, book-value growth and ROE. Two quarters with stable CET1 while buybacks and ROTE both hold would weaken this risk.[2][4]

What to Watch Next

  • For fee conversion, compare advisory, equity underwriting and debt underwriting with $1.378 billion, $985 million and $1.032 billion, and test backlog direction against segment pre-tax profit. Broad growth confirms conversion; two fee declines despite a higher backlog weaken it.
  • For financing quality, compare Equities and FICC financing with $3.259 billion and $1.216 billion and segment ROE with 27.5%. Revenue and returns holding together confirms quality; revenue up with ROE down falsifies it.
  • For durable fees, compare $4.041 trillion of AUS, $91 billion of long-term inflows, $3.355 billion of fees and the 24% margin. Double-digit fees and margin above 24% confirm progress; rising AUS with single-digit fees weakens it.
  • For capital headroom, compare the 12.9% CET1 ratio with the 10.9% requirement, $2.128 trillion of assets, $4.00 billion of buybacks and 25.5% ROTE. Stable CET1 and returns confirm room; stable returns with a falling ratio weaken it.

Conclusion

Goldman's earnings are driven by transaction completion, intermediation and financing activity, and fee-bearing assets, while capital and expenses determine how much revenue becomes return. Q2's $20.34 billion of net revenue and 23.5% annualized ROE show strong performance, but backlog is not revenue and financing growth accompanied asset expansion and a CET1 decline from 14.3% at year-end 2025 to 12.9%. The central question is whether growth can persist without continuing to consume capital headroom.[2][3][4]

Quartz attributed the Q2 beat mainly to Equities, citing $7.42 billion of revenue, up 72%, and reporting Bloomberg's view that the business surpassed every previous bank record for a third straight quarter.[6] That outside interpretation agrees with Goldman's disclosed strength in Equities financing and intermediation, but it also sharpens the concentration question. It was the only qualifying external view, so it cannot be treated as a market consensus.

The assessment would strengthen if advisory and underwriting rise together, financing revenue and segment ROE hold together, management fees remain in double-digit growth with margin above 24%, and CET1 stops falling. It would weaken if backlog rises while fees decline, financing grows as returns fall, AUS expands without fee growth, or high ROE accompanies another capital-ratio decline.

Sources

[1] GS earnings calendar updated 2026-09-12 · 2026-09-12 · earnings_calendar

[2] GS Q2 2026 earnings release 2026-07-14 · 2026-07-14 · 8-K · https://www.sec.gov/Archives/edgar/data/886982/000088698226000294/gs-20260714.htm

[3] GS 10-Q filed 2026-08-03 · 2026-08-03 · 10-Q · https://www.sec.gov/Archives/edgar/data/886982/000088698226000297/gs-20260630.htm

[4] GS 10-K filed 2026-02-25 · 2026-02-25 · 10-K · https://www.sec.gov/Archives/edgar/data/886982/000088698226000091/gs-20251231.htm

[5] GS 10-Q filed 2026-05-01 · 2026-05-01 · 10-Q · https://www.sec.gov/Archives/edgar/data/886982/000088698226000118/gs-20260331.htm

[6] Quartz GS Q2 2026 analysis 2026-07-14 · 2026-07-14 · Quartz · https://finance.yahoo.com/markets/stocks/articles/goldman-sachs-q2-2026-earnings-120234207.html

Related:GS

Want deeper analysis?

Ask drillr anything about GS — powered by SEC filings, earnings calls, and real-time data.

Try drillr.ai for free

drillr can make mistakes. Information only — not investment advice. Learn more