The Goldman Sachs Group
The Wall Street firm that advises on the biggest mergers, trades and lends to hedge funds and other investors, and manages money for institutions and the wealthy.
The scores
Growth3/4 ↑
Revenue rose 9% in 2025 and 25% so far in 2026, but it tracks the market cycle and fell 22% from 2021 to 2023.
Business quality3/4 ↑
An outside count puts it first in M&A and second in fees, and financing demand exceeds capacity; the new mix is untested by a downturn.
Financial position3/4 →
Capital 100 basis points above its requirement and large liquid assets, but a dealer depends on funding markets and the buffer narrowed.
Execution record3/4 ↑
Missed its returns target with the consumer bank in 2023, exited it and has beaten the target since; the Apple Card exit still costs money.
Per share capture4/4 →
The share count fell 4.4% in a year, buybacks far exceed share pay, and acquisitions are paid in cash.
Potential2/4 →
More upside needs the market cycle to stay strong and hedge funds to keep borrowing, both outside Goldman's control.
Risk profile3/4 →
One dependency dominates: the markets and the regulators who set capital. Diversified clients and no binary event inside the horizon.
Valuation2/4 ↑
The price pays for returns above management's own through-cycle target, so it assumes the plan works and 2026's market mostly lasts.
Evidence quality3/4 →
Everything central is filed and the league tables are independent, but durability and financing capacity rest on management's word.
The Verdict
We rate Goldman Sachs Neutral. The business is solid: it led global mergers advice in 2025, its lending to hedge funds is up 82% this year, and it earned about 22% on its net worth after goodwill, well above its own 15% to 17% target. The price is full: at about 2.6 times that net worth, buyers are paying for returns near 18% to 20% for years to come. The view hinges on whether this year is a new normal or a peak, and the chief executive himself says the market has cooled while the pipeline of deals shrank.
The long caseHedge funds keep growing and borrowing, and Goldman lends them the most; that financing must keep growing when markets turn.
The evidence under this - 9 claims
- Return on tangible equity is management's non-GAAP figure; the filed return on equity was 18.7% for the quarter and 20.7% for nine months, a point and a half lower.
- About $975M of tax benefits on share awards added 1.1 points to return on equity in the nine months, a benefit that depends on the share price at vesting.
- The targets were set at the 2020 investor day and have not been raised, though 2026 runs well above them; a target the firm is beating by five points says the year is unusual, or the target is cautious.
- Asset & Wealth Management's own target of a 30% margin compares with 25% in 2025, so the half of the firm meant to steady returns is below its target.
- The cost of equity is assumed: at 9% and 4% growth the implied return is 17%, inside the top of the target, so the conclusion is sensitive by about a point per point of discount rate.
- Goldman earned 22.4% on tangible equity in 2026 to date, so the implied return is below what the business is earning now; the question is only whether 2026 is a normal year.
The chief executive has reason to talk down expectations after a record year; the third quarter still grew revenue 23% and earned 18.7% on equity, which is not a cooling business.
The company
When a company wants to buy another, sell shares or borrow in the bond market, it hires a bank to advise it and find buyers. When a hedge fund wants to trade, it needs a firm to trade with it and lend it money against its holdings. Goldman does both, for the largest clients in the world.
Its rivals are a handful of full-service banks, mainly JPMorgan and Morgan Stanley, plus independent advisers, private lenders and electronic traders who each take one piece.
The evidence under this - 6 claims
- The chief executive on the same call called prime brokerage cyclical and said leverage would come down in a weaker market; 'capacity' is a claim about rivals' balance sheets that nothing filed measures.
- The IMF's October 2026 note on hedge funds says their leverage, much of it borrowed from prime brokers, can amplify stress, which is the same balance sheet seen from the side of the risk.
How it makes money
Most of Goldman's money comes from Global Banking & Markets: fees for advising and raising capital, spreads earned trading shares, bonds and currencies, and interest on loans to hedge funds. In the first nine months of 2026 that unit earned $41.7B of $56.2B. The rest is Asset & Wealth Management, which charges a fee on the money it looks after, $10.0B of fees in nine months, growing 19%.
The evidence under this - 5 claims
Where the money comes from, stream by stream
| Stream | Share of sales (9M 2026) | How it earns it |
|---|---|---|
| Equities: trading and financing | 34% | Trading shares and derivatives for clients, and lending to hedge funds against their positions. |
| Fixed income, currency and commodities | 21% | Trading bonds, currencies and commodities for clients, and lending against them. |
| Investment banking fees | 17% | M&A advice and selling companies' shares and bonds. |
| Asset and wealth management fees | 18% | Management and incentive fees on client assets. |
| Private banking, lending and investments | 6% | Lending to wealthy clients and gains on the firm's own investments. |
| Platform Solutions and other | 4% | The Apple Card programme being wound down, and other markets revenue. |
Who runs it
David Solomon has run Goldman since 2018 and chaired it since 2019; John Waldron is president and Denis Coleman finance chief. Directors and officers own about 0.5% of the company, and insiders sold shares this year without buying any.
Most of Solomon's $47M of pay for 2025 is in shares that pay out in full only if the return on equity averages 16% or more from 2026 to 2028.
The evidence under this - 4 claims
Sales cluster in the January, February, April and May trading windows after results, the usual pattern at a firm whose staff are paid largely in stock, and the group still owns about ten times what it sold.
Why it wins, and who could beat it
Why customers come back
The biggest deals go to the bank that sees the most buyers and can arrange the most financing, and that is the bank already doing the most deals. LSEG, which counts deals for the industry, put Goldman first in mergers advice in 2025 with a third of the value, and second in all banking fees.
Each part feeds the next: the deals bring underwriting and trading, and the trading desk's balance sheet funds the loans hedge funds buy, up 82% this year.
The evidence under this - 4 claims
- JPMorgan earns more total investment banking fees with a lending balance sheet Goldman cannot match, so scale alone does not decide who leads.
- Evercore, an adviser with no balance sheet, wins M&A mandates against Goldman, so the advice does not need the trading arm to win.
The threat
Each rival takes one piece. Evercore advises without lending, so it has no conflict to sell. Apollo and other private-credit funds lend to companies directly, skipping the bank's bond desk. JPMorgan and Morgan Stanley offer the whole bundle, JPMorgan with a bigger balance sheet. Electronic market makers win the simplest trades. And JPMorgan is taking over the Apple Card.
The evidence under this - 5 claims
The rivals: EVR, APO, JPM, MS, and 1 more that is private, in-house or listed abroad.
Who else does this, rival by rival
| Ticker | Rival | Where it competes | How far along | Priced against it |
|---|---|---|---|---|
| EVR | Evercore and other independent advisersPriced against it: Different product | Global Banking & Markets | On sale | Different product |
| APO | Apollo and the private-credit fundsPriced against it: Not checked | Global Banking & Markets | On sale | Not checked |
| JPM · MS | JPMorgan and Morgan StanleyPriced against it: Yes | Global Banking & Markets | On sale | Yes |
| Private | Citadel Securities and Jane StreetPriced against it: Private | Global Banking & Markets | On sale | Private |
| JPM | JPMorgan Chase, the Apple Card's new issuerPriced against it: Yes | Platform Solutions | On sale | Yes |
| MS | Morgan Stanley's wealth business and the large asset managersPriced against it: Yes | Asset & Wealth Management | On sale | Yes |
How big it could get
How big this could get
LSEG put all the fees banks earned worldwide for advice and raising capital at $117.4B in 2024, up 14%, of which roughly $33B was for mergers advice. Goldman took about a tenth of that mergers pool. This measures only one slice of Goldman: trading, lending and money management, most of its revenue, have no outside measure of their size.
The evidence under this - 3 claims
The results
The numbers
Sales grew 18% over the twelve months to Jun 2026, faster than the 8.9% in the year to Dec 2025.
It keeps 32% of sales as profit, up from 18% in the year to Dec 2023.
Yearly sales growth has ranged from −20% to +33% over the last 5 years, with sales falling in 2 of them.
Shareholders' money earned 17% a year.
The share count fell 14% in 4.5 years, and pay in shares costs 5.2% of sales.
All the numbers
Filed figures from the company's 10-K and 10-Q, read 2026-10-26. Fiscal years end in December.
| Dec 2021 | Dec 2022 | Dec 2023 | Dec 2024 | Dec 2025 | 12 mo to Jun 2026 | |
|---|---|---|---|---|---|---|
| Size and growth | ||||||
| Sales | $59.3B | $47.4B | $46.3B | $53.5B | $58.3B | $66.2B |
| Sales growth | +33% | −20% | −2.3% | +16% | +8.9% | +18% |
| Profit | ||||||
| Net margin | 36% | 24% | 18% | 27% | 29% | 32% |
| What growth costs, per sales dollar | ||||||
| Pay in shares | 4.0% | 8.6% | 4.5% | 5.0% | 5.9% | 5.2% |
| Capital spending | 7.9% | 7.9% | 5.0% | 3.9% | 3.5% | 3.3% |
| Cash | ||||||
| Cash from operations | $6.30B | $8.71B | −$12.6B | −$13.2B | −$45.2B | −$39.4B |
| Free cash flow | $1.63B | $4.96B | −$14.9B | −$15.3B | −$47.2B | −$41.5B |
| Balance sheet | ||||||
| Cash and short-term investments | $261B | $242B | $242B | $182B | $164B | $187B |
| Debt | - | $61.0B | $75.9B | $69.7B | - | - |
| Net cash | - | $181B | $166B | $112B | - | - |
| Returns and shares | ||||||
| Shares (diluted) | 356M | 358M | 346M | 334M | 318M | 305M |
What growth costs
For Goldman the main cost is pay, which rises with revenue. Even so, costs took 59.4 cents of each revenue dollar in the first nine months of 2026, against 62.1 cents a year earlier, and staff numbers fell while revenue grew 25%.
The evidence under this - 3 claims
The last report (13 Oct)
Quarter ended June 2026, filed figures.
In the third quarter of 2026 revenue was $18.68B, 23% more than a year earlier, though 8% below the second quarter. Share trading and lending to hedge funds drove it: equities revenue rose 72%. Mergers advice had its best quarter ever. Bond trading was flat, and the Apple Card exit cost money again.
Two things cooled. The backlog of deals fell, and the capital ratio slipped to 12.4% from 12.9% as the bank lent more.
What it did to the thesisIt confirmed the growth in lending to hedge funds and in fees the case rests on, but showed the cycle turning: revenue fell from the quarter before, the deal backlog shrank and the capital cushion narrowed. It left open whether returns hold in a weaker year.
The quarter in detail
Sales growth from the previous quarter, filed figures.
Earnings per share $17.01 against $12.86 expected.
The evidence under this - 5 claims
What management did with the money
Goldman returns most of what it earns: about $17B in 2025, and $4.49B in the third quarter of 2026, mostly by buying back shares, which cut the share count 4.4% in a year. It bought two fund managers for cash and is handing the Apple Card to JPMorgan.
Insiders in 2026Q3: 4 purchases and 7 sales, from their Form 4 filings.
The evidence under this - 4 claims
The price, and where we disagree
In short. Full. The price pays for returns above Goldman's own target lasting for years, so it needs this year to be close to normal rather than a peak.
The shares fell from $1,152 in July to $876 by 23 October, while revenue kept growing. Most of the fall came in September, when the Federal Reserve raised rates for the first time in three years.
The fourth-quarter results on 19 January 2027 show whether fees and lending hold as the deal backlog falls, and new capital rules expected by the end of 2026 decide how much money Goldman can return.
The evidence under this - 6 claims
The chief executive has reason to talk down expectations after a record year; the third quarter still grew revenue 23% and earned 18.7% on equity, which is not a cooling business.
The surcharge and Basel rules have been promised and delayed before, and the new stress test adds a second market shock for large trading books, which could raise Goldman's requirement rather than lower it.
How we read the multiple
For a bank the useful measure is the price against its net worth after goodwill, the value of past acquisitions.
- Goldman's net worth after goodwill was $336.61 a share in June 2026, and buyers pay about 2.6 times that.
- Assume shareholders want 10% a year and the bank grows 4% to 5% a year for ever.
- Then 2.6 times net worth pays for a return near 18% to 20% a year.
- Goldman earned 16% in 2025 and targets 15% to 17% through a full cycle.
Against its own history. Buyers pay far more for each dollar of Goldman's net worth than in its own past, while paying about the usual price for its earnings: they trust the earnings more.
The evidence under this - 5 claims
- The cost of equity is assumed: at 9% and 4% growth the implied return is 17%, inside the top of the target, so the conclusion is sensitive by about a point per point of discount rate.
- Goldman earned 22.4% on tangible equity in 2026 to date, so the implied return is below what the business is earning now; the question is only whether 2026 is a normal year.
- The targets were set at the 2020 investor day and have not been raised, though 2026 runs well above them; a target the firm is beating by five points says the year is unusual, or the target is cautious.
- Asset & Wealth Management's own target of a 30% margin compares with 25% in 2025, so the half of the firm meant to steady returns is below its target.
Priced against its rivals
The competitors its own annual report names, each confirmed against the rival's filing. Measured on price for each $1 of yearly profit, cheapest first:
| Ticker | Company | price for each $1 of yearly profit |
|---|---|---|
| GS | The Goldman Sachs Group | 12.7× |
| JPM | JPMorgan Chase & Co. | 14.0× |
| MS | Morgan Stanley | 14.6× |
The middle of the other 2 is 14.3. The Goldman Sachs Group, Inc. is cheapest of 3.
Rivals' figures come from a data provider, so read the spread of the group rather than any single pair.
On earnings Goldman is the cheapest of its two closest rivals, a small gap that fits a firm whose profit swings more with markets than theirs.
Private or foreign rivals that cannot be priced: Citadel Securities, Jane Street.
Where we disagree with the market
This year is a peak, and rising interest rates will end it.
The steadier mix is real, but the price already pays for most of it.
The market has sold Goldman twice this year on fears of a peak. The first time, in the spring, it was wrong within a quarter. We think the business really is steadier than in past cycles, but that the price is high because of that, not low despite it.
We would be too cautious if returns stay near 18% through a year when deal fees fall.
What analysts say, and the evidence
The evidence under this - 6 claims
- The cost of equity is assumed: at 9% and 4% growth the implied return is 17%, inside the top of the target, so the conclusion is sensitive by about a point per point of discount rate.
- Goldman earned 22.4% on tangible equity in 2026 to date, so the implied return is below what the business is earning now; the question is only whether 2026 is a normal year.
- 'Durable' is management's definition and no filed table reconciles it; equities financing, the largest part of it, grew 96% in a year, which is not what a stable annuity does.
- Firm revenue fell from $59,339M in 2021 to $46,254M in 2023 on the filed statements, so the business has shown how far it can fall within two years in a downturn.
- Return on tangible equity is management's non-GAAP figure; the filed return on equity was 18.7% for the quarter and 20.7% for nine months, a point and a half lower.
- About $975M of tax benefits on share awards added 1.1 points to return on equity in the nine months, a benefit that depends on the share price at vesting.
What could go wrong, and how we would know
The risks
- Markets turn. Revenue fell 22% from 2021 to 2023, and the steadier mix management describes has not yet been through a downturn.
- Hedge funds borrow less. Lending to hedge funds is the fastest-growing line, and the IMF warns that their borrowing can turn quickly when markets fall.
- Capital rules tighten. New stress tests add a second market shock for big trading banks, which could raise what Goldman must hold.
- Buybacks at a high price. Buying shares at about three times net worth leaves net worth per share barely changed if returns fall.
- Insiders sell, none buy. Insiders sold shares in the past year and bought none, even after a quarter's fall.
The evidence under this - 7 claims
- The chief executive on the same call called prime brokerage cyclical and said leverage would come down in a weaker market; 'capacity' is a claim about rivals' balance sheets that nothing filed measures.
- The IMF's October 2026 note on hedge funds says their leverage, much of it borrowed from prime brokers, can amplify stress, which is the same balance sheet seen from the side of the risk.
Book value per share, which keeps the acquired goodwill, rose 4.2% in the nine months, and the share count fell 4.4%, so the shareholder's claim on earnings rose even where the tangible book did not.
Sales cluster in the January, February, April and May trading windows after results, the usual pattern at a firm whose staff are paid largely in stock, and the group still owns about ten times what it sold.
What we still do not know
- How much hedge funds are borrowing from banks now, after September's rate rise.
- How large Goldman's deal backlog is in dollars; it only says which way it moved.
- How much the coming capital rules change what Goldman must hold.
The evidence under this - 3 claims
The surcharge and Basel rules have been promised and delayed before, and the new stress test adds a second market shock for large trading books, which could raise Goldman's requirement rather than lower it.
What would change our view
- Neutral→AttractiveThe price pays only for the target
It needs:
- ○The price falls to about 2 times net worth after goodwill, from 2.6, so it pays only for the top of Goldman's own target.about $687.14 today, 22% below the last close
- Neutral→AttractiveThe higher return survives a weaker year
It needs all of:
- ○Goldman earns 18% or more on net worth after goodwill for a full year in which deal fees fall.checked at the next results, due by 19 Jan
- ○Asset & Wealth Management fees pass $15B over four quarters, from about $13.1B now.checked at the next results, due by 19 Jan
- Neutral→UnattractiveThe returns or the capital give way
It needs any of:
- ○The return on net worth after goodwill falls below 12% in a quarter, from 20.5% in the third quarter of 2026.checked at the next results, due by 19 Jan
- ○The capital ratio falls below 11.5%, from 12.4%, against an 11.4% requirement.checked at the next results, due by 19 Jan
We would also be wrong if
- Lending to hedge funds falls 30% below its 2026 average within a year.
- Buybacks are suspended outside a market crisis.
What is coming
- December 2026 Basel III endgame and surcharge rules finalised: How much capital Goldman must hold, and so how much it can return or lend.
Every dated event ahead
- December 2026 Basel III endgame and surcharge rules finalised: How much capital Goldman must hold, and so how much it can return or lend.
- 19 January 2027 Q4 2026 results: Whether fees and financing hold as the backlog falls, the capital ratio, and the full-year return on tangible equity.
- December 2027 Apple Card transfer completed: The end of Platform Solutions' losses.
Since we wrote this
- 26 OctThesis writtenNeutral, medium convictionWritten two weeks after a strong third quarter, with the shares a quarter below July's high on rate rises and a falling backlog.
Sources 17
Sources - 17
Every claim above cites one of these, or states the reasoning that derived it. The quote is frozen at the date it was read: a link can rot, and an agent fetch returns a reading of a page rather than the page. Backs counts the whole claim ledger, so it can name a claim this page does not surface.
- S01primarycompany filing2026-02-25The business and segment descriptions, the competition section, the Apple Card transition, the return targets, the return history and the executive officers.backsC01C07C08C18C33C35C43C52
- S02primarycompany filing2026-01-152025 net revenues by segment and line, net earnings, earnings per share, return on equity, book value per share and the fourth-quarter columns.backsC01C04C14C24C43
- S03primarycompany filing2026-10-13Third-quarter and nine-month revenues by segment and line, earnings, returns, capital ratios, the balance sheet, assets under supervision, the backlog statement, capital returned, geography, headcount and the share-award tax footnote.backsC02C03C04C11C14C15C16C18C19C20C21C22C24C39C40C41C42C44
- S04primarycompany filing2026-08-03Tangible book value per share and tangible common equity, goodwill and intangibles, the Innovator and Industry Ventures notes, the backlog direction and unsecured borrowings.backsC11C22C24C39
- S05primarytranscript2026-10-13Goldman Sachs third-quarter 2026 earnings call transcriptReturn on tangible equity, durable revenues, fundraising, wealth inflows, the M&A ranking claim, the capital buffer, the chief executive's view of the cycle, and twelve analysts' questions, among them on buybacks at three times tangible book and on rising bond yields.
Demand for Equities financing continues to exceed available industry capacity.
backsC04C05C06C09C10C17C19C28C39C40C51C49 - S06primarycompany filing2026-03-20Beneficial ownership of directors and officers, the chief executive's 2025 pay and its mix, the performance share unit metric and scale, 2025 shareholder return and capital returned.backsC36C37C50
- S07primarycompany filing2026-10-26Every open-market transaction by an insider in the year, with the Rule 10b5-1 checkbox read from each filing.backsC38
- S08primaryvendor data2026-10-26Research pack for GS, frozen at content/GS/research-pack.jsonThe price and its history, market value, the analyst split and its dated series, price-target drift, the vendor's valuation history, FINRA short interest and the filed-competitor cohort.backsC23C24C26C27C28C46
- S09primaryindustry data2025-01-06The global investment banking fee pool, each bank's estimated share of it, and the M&A fee ranking.
Investment banking fees reached US$117.4 billion during full year 2024, posting a 14% increase compared to full year 2023
backsC13 - S10secondarynews2026-01-02LSEG's 2025 M&A ranking and fees by bank, and total investment banking fees for JPMorgan and Goldman.
It was paid US$4.6 billion in M&A fees, followed by JPMorgan at US$3.1 billion and Morgan Stanley at US$3 billion, Citi at US$2 billion and Evercore at US$1.7 billion
backsC12 - S11primarygovernment data2026-10-06IMF blog: Hedge funds improve market functioning but can also amplify stress International Monetary FundThe growth of hedge fund assets and their gross exposures, and the IMF's reading of the leverage prime brokers supply.
Since 2013, gross assets under management have tripled, to $13 trillion.
backsC29 - S12primarycompany filing2026-02-24Morgan Stanley's own description of Institutional Securities, used for the competitor gate.backsC34
- S13primarycompany filing2026-02-13JPMorgan's own description of its corporate and investment bank, used for the competitor gate.backsC34
- S14primarycompany filing2026-02-26The independent adviser's own description of its business and its named competitors.backsC31
- S15primarycompany filing2026-02-24The private-credit substitute's own description of its direct origination platform.backsC32
- S16primaryregulator2026-09-30Federal Reserve Board press release: stress test transparency and stress capital buffer averaging final rules Federal ReserveThe content and timing of the final stress-test rules: averaging of the stress capital buffer from 2028 and two market shocks for large trading books.
These banks will now be tested against two global market shock components each year, and the Board will use the shock that produces the largest losses for each firm to calculate the firm's stress test results.
backsC45 - S17secondarynews2026-01-07The name of the new issuer, the transition period and the size of the balances moving.
Apple announced Wednesday that JPMorgan Chase is the new issuer of the Apple Card, replacing Goldman Sachs.
backsC52