Chapter 5

Priced at $165.92

Between April and the end of July 2026, Accenture's shares lost and then partly regained roughly a third of their value while the earnings the company is expected to produce barely moved. That gap — a large price change sitting on top of a small numbers change — is the whole of this closing act. The reader arrives holding the pieces: a payroll-and-client-list business whose unit price the company says it cannot measure, a margin promise kept on the adjusted line, and an FY2026 plan that spends about 1.8 times free cash flow across acquisitions and buybacks. What follows is the price the market has put on all of it, stated as arithmetic, and the dated occasions on which that arithmetic gets tested.

The reaction itself came on the results. On 18 June 2026 the stock fell 17.97% in one session, from $156.01 to $127.98, on 41.7 million shares — about eight times the April daily average. The print underneath that reaction beat consensus adjusted EPS ($3.80 against roughly $3.71) and, in the same table, raised GAAP and adjusted EPS guidance, held free-cash-flow guidance at $10.8 billion to $11.5 billion, and lifted the capital-return floor to at least $9.5 billion from at least $9.3 billion [1]. The single line the company cut was the top of the local-currency revenue range, from 3%–5% to 3%–4% [1]. Alongside it, new bookings of $19.32 billion were down 3% in local currency at a book-to-bill near 1.0 [2]. The market priced the ceiling on the top line and the softness in the order book, not the earnings or the cash.

A re-rating, not a downgrade

The de-rating is far larger than the estimate revision behind it. Over the six months to 31 July 2026, consensus FY2027 normalized EPS moved from $14.89 to $14.68 — a 1.4% cut — and FY2028 from $16.16 to $15.79, a 2.3% cut; the FY2027 and FY2028 revenue lines came down 1.7% and 2.4% on the same marks, so the trim is a top-line trim, not a margin call.

Source: consensus normalized-EPS and revenue marks (180-day vs current), and the 52-week price range, as reported. Estimate momentum from consensus data; 52-week high $291.09.

Against a consensus that moved by one to two points, the stock at $165.92 is 43.0% below its 52-week high of $291.09 [3]. Expressed as a multiple on unchanged FY2027 consensus EPS, the shares moved from about 19.8 times at that high to 11.3 times at the July close. The numbers the company is expected to earn are close to where they were; almost the entire change is in what a dollar of those earnings costs.

The recovery is the same mechanism in reverse. The closing low was $124.44 on 30 June 2026; by 31 July the stock had risen 33.3% to $165.92, its five best sessions of the four-month price file all falling in July. Over that same stretch the estimate revisions were inert — every 30-day change in FY2027 and FY2028 revenue and EPS sits inside 0.1% — and the corpus holds no dated news item after 27 June. A third of the current quote was added in a month with no change in the consensus underneath it, which makes the entry point unusually sensitive to sentiment rather than to numbers.

Source: daily closing prices, as reported. The 18 June session marks the −17.97% drop; the closing low of $124.44 was set 30 June. The price file covers only 1 April to 31 July 2026.

The drawdown that isn't

The single most quotable relative-valuation figure in this run reverses direction once the windows are matched. The company's own price history in the corpus covers just 84 sessions, from 1 April to 31 July 2026, and is flagged as partial. The "-17.6% drawdown from a three-year high" that the facts table reports is therefore measured against a four-month high of $201.33, not a genuine three-year peak. Every peer file, by contrast, holds five years of sessions, so the peer drawdowns — IBM −32.1%, Cognizant −39.0%, TCS −48.1%, Infosys −43.5%, HCLTech −32.5%, a −39.0% median — are true multi-year figures. Measured on the nearest comparable basis, Accenture's own 52-week high of $291.09, the decline to $165.92 is 43.0%, in line with the peer median rather than less than half of it.

Source: peer drawdowns over five-year windows, as reported; Accenture's four-month-window figure from the run's valuation series; Accenture's 52-week figure derived from the $291.09 52-week high and the $165.92 close.

The distinction matters because the four-month figure is the one that invites the conclusion that Accenture fell less than its peers. On matched windows it fell about the same. Whatever the market repriced, it repriced Accenture roughly as hard as the rest of the group.

What the price embeds

At the 31 July close the equity is capitalized at $104.93 billion. That price carries 12.0 times FY2026 and 11.3 times FY2027 consensus normalized EPS ($13.87 and $14.68), a 10.4% trailing and 10.7% forward-consensus free-cash-flow yield, and a guided capital return of at least $9.5 billion — equal to 9.1% of market capitalization — for the year in progress.

Source: multiples derived from the $165.92 close and FY2026/FY2027 consensus normalized EPS [3]; free-cash-flow yields per the run's valuation series; capital return of at least $9.5 billion from Q3 FY2026 guidance [1].

The return at this price is dominated by the cash yield and the share-count arithmetic, not by growth — and the growth the price does embed is mostly bought. Consensus FY2027 revenue of $76.61 billion is 4.1% above FY2026, but management said on the Q3 call that it expects to enter FY2027 with "slightly below 2%" of inorganic growth [4], funded by the roughly $9 billion of FY2026 acquisition spend it now plans [6]. Broker models put the organic line near 1.85%. A buyer at $165.92 is underwriting roughly 2% organic growth, plus purchased revenue, plus a slow grind of operating margin — consensus moves it about 45 basis points over three years — not the mid-single-digit organic figure the headline growth rate suggests.

Sources: consensus FY2027 revenue growth as reported; the acquired share from management's Q3 FY2026 call [4]; the organic residual from broker models. The components draw on different sources and do not sum precisely.

The other half of the return is cash coming back. In Q3 FY2026 alone the company returned $2.2 billion — $1.2 billion repurchasing or redeeming 6.0 million shares and $1.0 billion in dividends at $1.63 per share, a 10% increase over the FY2025 rate — leaving roughly $3.2 billion of repurchase authority at 31 May 2026 [5]. At a 10.7% forward free-cash-flow yield the guided payout alone is close to a tenth of market value in a single year, and that is the arithmetic that dominates the outcome at this price. The caveat carried forward from the cash-quality act (Ten to Thirty Basis Points) applies directly here: FY2025's 1.42x cash conversion — the anchor of the trailing yield — was lifted about 65% by a $1,519 million accrued-payroll swing that management itself guides back toward 1.2x in FY2026. The yield is real; whether it repeats at FY2025's level is the open question the buyer inherits.

The disagreement, priced

The sell-side posture is neutral and its dispersion is wide. Across 25 targets the mean is $178.89 and the median $179.00 — about 8% above the close — inside a range that runs from a $130 low to a $275 high, more than double the low [3]. The rating split is 11 strong-buy, 3 buy and 13 hold, with no sell and no underperform; five price-target cuts landed in the trailing 90 days, and three of the most recent dated actions — DBS to $133, Susquehanna at $140, Deutsche Bank to $136 — sit below the current price [3]. The July rally carried the quote above several of the most recently published targets, so the marginal buyer is ahead of a coverage base that has stopped cutting but has not re-engaged.

The wide target range hides a narrow disagreement about the actual income statement. Thirteen brokers cluster FY2028 operating margin between 15.74% and 16.20%; the modeled near-term profit line is close to settled. The dispersion lives in the per-share and narrative inputs — the pace of the buyback, the free-cash-flow level, the delivery headcount, and how much AI-related bookings translate into revenue — rather than in the next two years of earnings. That is consistent with a market that repriced the durability of the model rather than its forecast: the contest over who sets the price of an hour of labour (Who Sets the Unit Price) is precisely the kind of question a multiple, not an estimate, absorbs.

Sources: cash conversion and buyback economics from the cash-quality and capital-allocation acts [5]; forward-growth split from the Q3 FY2026 call [4]; AI-deflation and peer-outgrowth evidence developed earlier in the report.

What is still open, and when it gets tested

Five questions decide which of those reads is right, and none can be closed from the current corpus. Whether FY2025's cash conversion repeats once the $1,519 million accrued-payroll swing reverses. Whether the roughly $9 billion of FY2026 software and platform acquisitions changes the labour model or merely buys revenue at prices the company does not disclose — the allocation act (Spending 1.8x Free Cash Flow) framed this deal shift and left it deliberately open. Whether a buyback that has run at an average $303.83 against a $165.92 close is a return of capital or a deferred bet on the price recovering. Whether consensus organic growth near 1.85% is too low or too high. And whether the AI price deflation peers are already quantifying reaches this book of work through renewals of its fixed-price contracts.

Two of these are self-reversing items already sized and sitting in the FY2026 base. Management attributes about a point of the local-currency revenue drag to the U.S. federal business and expects to anniversary that headwind and return the unit to growth in the fourth quarter [6]; separately it quantified roughly $100 million of Q3 revenue impact from the Middle East conflict, all in consulting-type work [7]. Both are dated, both are sized, and both are embedded in the base off which FY2027 consensus is built.

The calendar puts firm dates on the resolution. Initial FY2027 guidance — the first real test of the just-under-2%-inorganic and roughly-2%-organic framing — lands on the fiscal fourth-quarter call and the investor day that follows it.

Source: Accenture investor-relations events calendar, as published [3].

What the market repriced this summer was the durability of an hour of Accenture's labour, not the next two years of its earnings, which barely moved. On the evidence, $165.92 pays for a business whose return is carried by a roughly 10% cash yield and a shrinking share count rather than by growth, and whose growth is about half acquired at prices never disclosed. The strongest fact against reading that as conservative is that the same cash yield rests on a conversion rate management expects to fall, and the buyback that would compound it was executed well above today's price. Whether the July recovery was the market looking through a demand air-pocket or front-running a re-rating it has not yet earned is the disagreement that the 1 October call and the 14 October investor day will begin to settle — the first moment the company puts a number on FY2027 against a price that has already moved twice this summer without one.