Goldman Sachs spent most of the last decade trading below the value of its own assets. After the 2008 crisis, when the firm famously converted from an investment bank to a bank holding company to access Federal Reserve funding, the market spent years treating it as a relic of a wilder era, worth less than the sum of its parts. The consumer banking detour, the Marcus experiment that burned through billions and then quietly retreated, did not help. For a long stretch, owning Goldman stock meant believing the firm would eventually be allowed to be Goldman again.
That moment appears to have arrived. The numbers filed this summer are, by any honest reading, extraordinary. The tension is that the market has noticed, and then some.
The quick read
The 10-Q filed August 3, 2026 shows Goldman earning at a pace it has not sustained since its 2021 peak, with a return on equity that most banks would frame and hang on the wall. The stock, at roughly $1,022, sits at a price-to-earnings multiple in the 88th percentile of its own decade-long history, our data show. The math says the business is performing; the mood, meaning what the market is currently willing to pay per dollar of profit, says the performance will not only continue but expand. Those two readings can coexist, but only if the deal-making boom that drove this quarter is not a one-time event.
A Quarter That Demands an Explanation
The simplest way to describe the second quarter ended June 30, 2026: Goldman nearly doubled its profit versus the same quarter a year earlier. Net revenues reached $20.34 billion, up from $14.58 billion in Q2 2025, and net income followed at a comparable pace. Think of it this way: for every dollar the firm earned in that quarter last year, it earned nearly two this year. That is not a rounding error or an accounting quirk. Something structural shifted.
The engine was Global Banking and Markets, the division that houses trading desks and deal advisory. It generated more than half of total firm revenue for the quarter, a dramatic jump year-over-year, driven by what the Q2 2026 earnings call described as record revenues in both equities and fixed-income trading, alongside investment banking fees at their highest level since 2021. Large-cap corporate mergers and acquisitions surged by volume in the first half of 2026, with Goldman leading more than a trillion dollars in announced deals. That last figure is worth sitting with: more deal value than most countries produce in a year, flowing through one firm’s advisory business in six months.
For anyone who watched Goldman limp through 2022 and 2023, when investment banking fees collapsed and the firm was still digesting its consumer lending losses, this quarter reads like a different company. It is, in some ways, the same firm that posted a full-year 2025 EPS of $51.32, itself a near-record, now accelerating further.
What the Filed Numbers Actually Show
Return on equity, the ratio of profit to the shareholders’ money funding the business, is the metric Goldman’s own management uses to judge itself. An ROE of 10% means the firm earns ten cents for every dollar of equity. Goldman’s annualized Q2 2026 ROE of 23.5% is nearly double its pace from the same quarter a year earlier. For context, most large banks consider anything above 15% a strong result. Goldman is running well above that bar.
Book value per common share, which is the accounting value of what each share owns in the business after all debts, stood at $367.67 as of June 30, 2026. The stock trades near $1,022. That means the market is paying roughly three times what the accountants say each share is worth in hard assets. On tangible book value, which strips out intangible items and gives a more conservative floor, the ratio sits at about 2.97. Historically, Goldman has traded much closer to book during periods of uncertainty. The current premium reflects confidence that the firm will keep earning well above its cost of capital for years, not quarters.
The capital return picture adds another layer. Goldman declared a quarterly dividend of $5.00 per share in July 2026, up from the prior $4.50, while also conducting substantial share buybacks during the quarter. Together, that is more capital returned to shareholders in three months than many financial firms earn in a year. The firm’s CET1 ratio, a regulatory measure of financial cushion where higher means safer, held comfortably above regulatory minimums, suggesting the firm is not stretching to fund these returns.
Where the Crowd Has Gathered
The stock is up nearly 40% over the past twelve months, our data show, and sits about 11% below its 52-week high. The trailing price-to-earnings ratio of 17.8 sits at the 88th percentile of Goldman’s own decade-long range, against a decade median of 11.8. That gap is the math-versus-mood tension in plain arithmetic: the business is priced as if the current earnings pace is not a peak but a new floor.
Analyst opinion is broadly supportive but not unanimous. Wells Fargo analyst Mike Mayo raised his valuation estimate to $1,325 on July 15, 2026, maintaining an Overweight rating, and Barclays analyst Jason Goldberg moved his estimate meaningfully higher on the same date. UBS analyst Erika Najarian, by contrast, maintained a neutral stance on August 3, 2026, raising her valuation estimate modestly to $1,150 while describing a “modestly higher valuation outlook” without conviction. HSBC’s Saul Martinez upgraded from Reduce to a neutral stance on July 21, 2026, a move that signals less pessimism rather than enthusiasm. The range of valuation estimates across more than twenty analysts per Zacks data as of August 19, 2026 reflects genuine disagreement about whether this quarter’s results are repeatable.
The honest uncertainty is this: Goldman’s trading and deal-making revenues are, by nature, volatile. The 2021 earnings peak was followed by a collapse and then a trough. The current boom is real and filed. Whether it persists depends on M&A volumes, market volatility, and the AI-driven investment cycle that Goldman executives cited on the earnings call as a structural tailwind. None of those factors are in the 10-Q. They are bets on the future, and the current price embeds a fairly optimistic version of that future.
Reading the numbers
- Net revenue, Q2 2026: $20.34 billion. What it is: total income before expenses, from the 10-Q for the quarter ended June 30, 2026. What it means here: this is the highest single-quarter revenue Goldman has reported in recent history, nearly 40% above the same quarter last year. Everyday version: if a neighborhood bakery normally sold $14,000 worth of bread in a quarter and suddenly sold $20,000, the owner would want to know whether the block got more foot traffic permanently or just had a festival.
- Annualized ROE, Q2 2026: 23.5%. What it is: return on equity, the profit the firm generates for every dollar shareholders have put in, annualized from one quarter. What it means here: Goldman is earning nearly a quarter of its equity base in a single year, roughly double its Q2 2025 pace. Everyday version: a household with $100,000 in savings earning 23.5% would collect $23,500 in a year; at last year’s 12.8% pace, they’d have collected $12,800.
- P/E ratio: 17.8, versus decade median of 11.8. What it is: price-to-earnings, the price tag per dollar of annual profit, our data computed from filed EPS and current market price. What it means here: the market is paying 51% more per dollar of Goldman’s earnings than it has on a typical year over the past decade. Everyday version: a coffee shop that normally sells for 12 times its annual profit is now being offered at 18 times, because buyers expect profits to keep rising fast.
- Book value per share: $367.67; stock price: ~$1,022. What it is: book value is the accounting net worth per share from the June 30, 2026 10-Q; the stock price is the market’s current bid. What it means here: the market pays nearly three times the accountants’ valuation of each share. Everyday version: a house appraised at $368,000 selling for $1,022,000 means the buyer is paying for the neighborhood’s future, not just the building today.
- FY2025 EPS: $51.32, from the 10-K for fiscal year ended December 31, 2025. What it is: diluted earnings per share for the full year. What it means here: the 2025 full year was already near the 2021 record of $59.45 EPS, and Q2 2026 alone ran at an annualized pace above that. Everyday version: a worker who had their best salary year ever in 2025 is now on track for an even bigger paycheck in 2026.
If earnings continue near the filed 2025-to-2026 trend, our data show the decade-median multiple of 11.8 would imply a price well below today’s level, while a multiple in the upper quartile of the historical range would sit closer to current prices. The condition that matters is whether deal volumes and trading revenues hold. That question is open, and the filed numbers cannot answer it.
Sources
- Goldman Sachs 10-Q, quarter ended June 30, 2026 (SEC EDGAR)
- Goldman Sachs 10-K, fiscal year ended December 31, 2025 (SEC EDGAR)
- Goldman Sachs Q2 2026 Earnings Press Release
- Goldman Sachs Q2 2025 Earnings Press Release
- Goldman Sachs Q2 2026 Earnings Call Transcript (The Motley Fool)
- Goldman Sachs Price-to-Tangible-Book (GuruFocus)
- Goldman Sachs Analyst Valuation Estimates (Zacks, August 19, 2026)
- Goldman Sachs Analyst Ratings Roundup (Barchart)
- Goldman Sachs Dividend History (StockEvents)
- Goldman Sachs 2008 Bank Holding Company Conversion (Goldman Sachs)
