Evercore Inc. EVR

Key data

The business

Evercore advises on the biggest corporate decisions. Two engines. Investment Banking, the advisory desk that runs mergers, sales, restructurings, and capital raises, plus a small equities research and sales operation, is the overwhelming majority of revenue. Investment Management, wealth management for rich families and institutions, is the modest remainder. The product is a phone call from a senior banker a board already trusts, made in the most consequential moment a company faces. The money is a fee, often a percentage of the deal's size, collected when the deal closes. The moat is the roster: the named senior bankers clients actually follow, and Evercore has built the deepest independent bench outside the giant banks. A board advised on a ten-billion-dollar merger pays Evercore tens of millions, and the cost of delivering that advice is mostly the banker's own pay.

The numbers

The story is the swing, not the level. Read the quarters in order.

Q1 2026 was enormous partly for a mechanical reason: the first quarter always carries a tax benefit as equity awards vest, and Q1's tax rate ran under 3%. Q2 handed much of it back. Revenue still rose 19% from a year earlier, but net income slipped 2% and diluted EPS fell to $2.32 as the compensation ratio and a normal 28% tax rate returned. The top line is still growing; the profit line stopped. Q2 2026 is the inflection.

That is the deal cycle drawn in full: a 2021 boom, a trough in 2023 as boardrooms froze, and a climb back to a record $3.88B in 2025. Over 2022 to 2025 revenue compounded 11.8% and free cash flow per share 29.1%, but diluted EPS compounded only 6.6%, because the diluted share count rose 8.1% even as the firm spent about $2.1B buying stock back. That gap is the whole tell. Stock-based pay, near 20% of revenue, issues shares almost as fast as repurchases retire them, so per-share earnings lag the business. What this memo believes the market underweights is that the reported multiple flatters the price, because these earnings sit near a cyclical peak and the count keeps creeping up. The print that settles it is the Q3 tax-normalized net income against the +56% first-half comparison.

Operating leverage is the one thing that makes the peak so seductive:

When deals flow, incremental margins are high, which is exactly why the stock overshoots at the top and undershoots at the bottom.

Management

John S. Weinberg runs it, and the capital record is clean: about $3.4B returned through buybacks and dividends over five years, a $3.46 dividend at a 1.2% yield, a payout near 20% of earnings. But the buyback runs uphill. Shares outstanding rose while the firm spent, and 2025's $661.8M was repurchased at an average $276.84, the highest price it has ever paid. Insiders sold $5.05M across four sellers over the past year with no buys, a rounding error against an $11.5B firm at 0.04% of value, and the filings here do not split planned sales from discretionary ones, so read it as routine rather than a message. The real management lever is the compensation ratio, the share of every dollar paid to bankers, and it is what moved Q2's profit. CEO pay is not disclosed in this bundle.

How it fails or surprises you

The deal cycle rolls over. Advisory revenue is almost pure deal flow, and when boardrooms freeze it can halve fast, as diluted EPS did from $17.08 in 2021 to $6.37 in 2023. The first sign is two straight quarters of a shrinking announced-deal backlog with the compensation ratio climbing to hold talent through the drought. Avoiding this is the whole game.

The compensation treadmill (proves the read wrong). Diluted shares rose 8.1% in three years despite heavy buybacks. If stock-based pay keeps issuing faster than repurchases retire, per-share earnings stall even in a strong cycle and the reported P/E overstates value. This is the fact the memo's own numbers explain least well, and the thing that would break the quality-compounder story.

Private-capital and restructuring surprise up (right tail). The newer engines, private capital advisory, fund placement, and restructuring, grow through both halves of the cycle and lean counter-cyclical when M&A stalls. The market prices Evercore as one deal-cycle bet. If those lines keep compounding, the next trough is shallower than 2023's. The print that reveals it: advisory down in a soft quarter while total revenue holds.

Closing thoughts

This is largely a distribution the market already prices. The stock is down about 23% from its 52-week high and sits below both its 50- and 200-day averages, so the market has started discounting the deceleration a careful reader would have watched for. The tell named last time, a decelerating Q3 against the +56% first-half comparison, has not printed yet, but the price moved ahead of it, and Q2's flat-year profit was the first crack. The fatter tail remains the downside: a cycle roll-over costs more than the private-capital upside adds, because the earnings are cyclically full and the share count keeps rising.

The bet is still that companies keep hiring Evercore's bankers through the cycle and that the current deal wave runs into 2027. What breaks it is a freeze in announced deals with the compensation ratio climbing to hold talent, and the pair that tells you first is quarterly advisory revenue against the prior year alongside that compensation ratio. If Q3 net income, normalized for tax, comes in below a year ago while the diluted count keeps rising, the record was the peak, not the baseline.