Cintas Trades at 36 Times Earnings on 8% Growth as It Bets $5.5 Billion on UniFirst

---

Begin with the quality, because Cintas's excellence is not in dispute and is the foundation of the bull case. In the third quarter of fiscal 2026, reported on March 25, 2026, the company posted revenue of $2.84 billion, up 8.9% from $2.61 billion a year earlier, with organic growth — stripping out acquisitions and currency — of 8.2%. Gross margin reached $1.45 billion, or 51.0% of revenue, which management flagged as an all-time high. Diluted earnings per share rose 9.7% to $1.24. Net income was $502.5 million, up 8.4%. The company raised full-year guidance to revenue of $11.21 billion to $11.24 billion and adjusted EPS of $4.86 to $4.90. It returned roughly $1.45 billion to shareholders through buybacks and dividends in the quarter, and it has now raised its dividend for forty-five consecutive years. These are the numbers of a thoroughbred, and they deserve to be stated plainly before any skepticism begins.

The question is not whether Cintas is a great company — it manifestly is — but whether the price the market is paying for it leaves any room for the ordinary disappointments that befall ordinary businesses. At roughly 36 times trailing earnings, Cintas carries a multiple typically reserved for asset-light, secular compounders whose growth is structurally insulated from the economy. Cintas's growth is not so insulated: it is, at its core, a bet on how many Americans are working, how many businesses are opening, and how many of them choose to outsource their shirts, mats, and restroom supplies. This essay examines the gap between that software-style price and the employment-linked, decelerating, partly cyclical reality of the business — the bought growth now layered on top of the organic, the quality-of-earnings details inside the adjusted numbers, the denominator effects flattering the optics, and the asymmetry of paying for perfection.

A multiple priced for software on a business priced by headcount

Start with what 36 times earnings actually demands. A multiple like that is normally the market's way of saying it expects durable, high-teens-or-better growth for years, with margins that expand on near-zero incremental cost and a business model insulated from the economic cycle. That description fits a dominant software platform. It fits Cintas only in the loosest analogical sense. Cintas rents uniforms, floor mats, restroom and hygiene supplies, first-aid cabinets, and fire-protection services to roughly a million business customers. Its revenue is overwhelmingly a function of how many "wearers" its customers employ and how many locations they operate. When a customer adds workers, Cintas bills for more uniforms; when a customer cuts headcount or closes a site, the billing shrinks. The topline is, in the most literal sense, indexed to employment and business formation.

That is a fundamentally different animal from software, and it deserves a fundamentally different valuation framework. Software margins compound because code costs nothing to copy; Cintas's 51% gross margin, impressive as it is, is earned by laundering, delivering, and maintaining physical goods through a continent-spanning network of plants and routes, against real costs in labor, energy, fuel, and cotton. The growth that justifies a 36-times multiple has to come from somewhere, and for Cintas it comes from a slow, route-by-route grind of selling more services into existing accounts and winning new ones — admirable, durable work, but not the frictionless, exponential kind that the multiple implies. Pricing a headcount-linked services firm as though it were a secular software platform is the central tension here: the market has decided Cintas's quality is so high that the ordinary rules of its category no longer apply, and history is unkind to that decision when it is made at a peak valuation.

The growth is decelerating beneath the record headline

Look closely at the trajectory of the organic number, because the headline of "record revenue" obscures the trend underneath it. Organic growth in the third quarter of fiscal 2026 was 8.2%. That is a strong figure in absolute terms — most businesses would celebrate it — but it sits below the double-digit organic rates Cintas posted in the post-pandemic surge, when reopening, re-hiring, and pricing power all worked in concert. The company's own full-year guidance, at $11.21 billion to $11.24 billion on a base near $10.3 billion the prior year, implies high-single-digit total growth, of which an increasing share will come from acquisition rather than from the underlying engine. The compounding machine is still compounding, but it is doing so at a visibly slower rate than the period that earned it the multiple it now carries.

This matters enormously at 36 times earnings, because a premium multiple is, in effect, a bet on the durability and acceleration of growth, not merely its presence. When growth decelerates from the low double digits toward the high single digits, the math of a premium multiple turns hostile: the same multiple applied to a slower grower is a richer multiple, and the market's willingness to keep paying it depends on a story of re-acceleration that the recent numbers do not yet support. Cintas is not stumbling — 8.2% organic is a fine result — but it is decelerating, and a stock priced for permanent best-in-class growth has little tolerance for the ordinary maturation that every great business eventually experiences. The skeptic's point is not that the business is bad; it is that the price assumes the growth rate of a younger, faster Cintas than the one the most recent quarter actually describes.

Bought growth is now propping up the organic story

Here is the development that should sharpen every investor's attention: Cintas agreed to acquire UniFirst, its largest publicly traded rival in uniform rental, in a transaction announced in March 2026 and valued at roughly $5.5 billion. The strategic logic is sound — consolidating a fragmented industry, absorbing routes, gaining density — and Cintas has a long record of digesting tuck-in acquisitions well. But the timing and the framing are telling. A company that genuinely believed its organic engine could carry a 36-times multiple on its own would not need to bolt on its largest competitor to keep the growth narrative intact. The UniFirst deal — pursued by Cintas across multiple attempts before it finally closed — is, in part, an admission that buying growth is becoming a necessary supplement to making it.

This is the bought-growth-masks-organic-stall frame, and it deserves to be applied carefully and fairly. Cintas's organic growth is still positive and healthy at 8.2%, so this is not a case of acquisitions papering over an organic collapse. But the direction of travel is unmistakable: as organic growth decelerates, the acquired growth becomes a larger fraction of the total, and the reported "8.9% revenue growth" headline increasingly reflects what Cintas bought rather than what it built. Acquired growth is lower-quality growth for a premium-multiple stock, because it is purchased with shareholder capital, it carries integration risk, it consumes the balance sheet, and it does not demonstrate the organic vitality the multiple is supposed to reward. When a 36-times compounder starts leaning on a roughly $5.5 billion deal to keep its growth line elevated, the question is not whether the deal makes sense — it may well — but whether the organic business underneath still justifies the price on its own. A roughly $5.5 billion deal makes the company bigger; it does not make the per-share math at 36 times any easier.

Quality of earnings: where "adjusted" does its quiet work

Now turn to the texture of the profit, because the cleanest-looking income statements reward the closest reading. Cintas guided to adjusted diluted EPS of $4.86 to $4.90 for fiscal 2026, and management was explicit that this adjusted figure excludes non-recurring transaction expenses tied to the UniFirst acquisition. That is a perfectly standard and defensible exclusion — deal costs genuinely are one-time. But it is also the beginning of a pattern worth watching: as a serial acquirer integrates a multibillion-dollar target, the gap between adjusted and GAAP earnings tends to widen, populated by transaction costs, integration charges, amortization of acquired intangibles, and restructuring. Each item is individually justifiable; collectively they can flatter the number that analysts anchor their multiple to.

There is a second, subtler quality-of-earnings wrinkle in the operating line. Cintas reported operating income at 23.2% of revenue in the quarter, versus 23.4% a year earlier — a slight decline. Management's framing was that, excluding a one-time gain in the prior-year period, operating margin would actually have risen 40 basis points. That is very likely true and fair. But notice the mechanics: the comparison is being normalized by reference to a prior-period item, which is exactly the kind of adjustment that, repeated across quarters, lets a company present a smoother and more favorable margin story than the unadjusted GAAP figures alone would tell. None of this is aggressive accounting, and Cintas's reputation for clean reporting is well earned. The point is narrower and forensic: at 36 times earnings, the investor is paying a premium for the adjusted number, and the adjusted number is increasingly the product of judgment calls about what counts as recurring. The wider that adjusted-to-GAAP gap grows during the UniFirst integration, the more the premium multiple rests on a figure that management, rather than the unadorned accounts, defines.

The denominator illusion in the buyback

Watch the share count, because part of the per-share growth story is arithmetic rather than economic. Cintas returned roughly $1.45 billion to shareholders through buybacks and dividends in the third quarter alone, and it has a fresh $1 billion repurchase authorization from late 2025. Buybacks are a legitimate and often excellent use of cash, and Cintas's long history of shrinking its share count has genuinely amplified per-share returns. But they also create a denominator illusion that flatters the very EPS figure the market capitalizes at 36 times.

When earnings per share grow 9.7% while net income grows 8.4%, the difference — more than a full percentage point — is the buyback doing its quiet work, lifting per-share earnings faster than the underlying profit pool is actually expanding. This is real value for continuing holders, and no one should pretend otherwise. But it is worth naming precisely, because a premium multiple applied to buyback-inflated EPS double-counts the benefit: the investor pays a rich multiple for growth, and part of that growth is itself a function of returning capital rather than generating it. The faster the share count falls, the more EPS growth outruns income growth, and the more the headline number flatters a business whose actual profit engine is expanding at the more modest 8% pace. At a cheaper multiple this would be a footnote; at 36 times earnings it is part of how the price is sustained.

The cyclical truth the secular multiple ignores

Confront the cyclicality directly, because it is the fact the valuation most wants to forget. Cintas's revenue is tied to employment, business formation, and the operating health of its million customers. When the economy expands and companies hire, Cintas's wearer counts rise and new accounts open; when the economy contracts, headcount falls, locations close, and customers trade down on service levels. This is not a hypothetical: it is the structural reality of any business that bills by the worker and by the location. Cintas is more defensive than a pure-cyclical industrial — it has recurring rental contracts, a diversified customer base, and consumable products that customers need regardless of the cycle — but "more defensive than a steelmaker" is not the same as "secular grower immune to the economy."

A 36-times multiple, however, prices Cintas as though its growth were independent of the employment cycle — as though a recession that cut U.S. payrolls, slowed business formation, and pushed customers to defer or cancel services would leave Cintas's compounding untouched. That assumption has not been tested in this cycle, and it runs against the grain of how the business actually earns its revenue. The danger in a priced-for-perfection cyclical is not that the cycle is certain to turn tomorrow; it is that the multiple leaves no room for the cycle to turn at all. If U.S. employment growth stalls, Cintas's organic rate — already decelerating to 8.2% — would feel it directly, and a stock priced for permanent acceleration would have to reprice toward the reality of a business that grows with the labor market rather than above it. The market is paying a secular price for an employment-linked truth, and the bill comes due whenever the labor market does what labor markets eventually do.

The price-taker beneath the pricing power

Consider the cost side, because the margin story has an exposure the multiple discounts. Cintas's record 51% gross margin is real and hard-won, built on route density, plant efficiency, and disciplined pricing. But the inputs beneath that margin — cotton and synthetic fabric for the garments, energy and water for the laundering, diesel for the delivery fleet, and above all labor for the plants and routes — are commodities and wages over which Cintas has limited control. In a benign cost environment, as recently prevailed, those inputs are a tailwind and the margin prints records. In an adverse one — a spike in energy, a surge in wage inflation, a jump in fuel — they compress the very margin the bull case extrapolates to infinity.

Cintas has pricing power and passes costs through over time, and that is a genuine strength. But pricing power operates with a lag, and a 51% gross margin presented as an "all-time high" is, by definition, a peak — a level that is more likely to mean-revert than to keep ratcheting upward indefinitely. A premium multiple built on the assumption that record margins are a permanent new baseline, rather than a favorable point in a cost cycle, is making an extrapolation that the history of input costs does not support. The skeptic does not claim margins must fall; only that paying 36 times earnings for a business at peak gross margin assumes the peak holds, and peaks, by their nature, do not always.

What the bulls genuinely get right

In fairness, the bull case is formidable, and Cintas's quality genuinely supports a premium — the only honest debate is how large a premium and whether the cycle is being priced at all. Cintas is the dominant player in a fragmented, structurally attractive industry, with a route-density advantage that compounds: the more stops per route, the lower the marginal cost of each, a real moat that competitors cannot easily replicate. Its recurring rental model produces sticky, contracted, high-retention revenue that is far more durable than a transactional business. Its 45-year dividend-growth streak and consistent buybacks reflect a capital-allocation discipline that has rewarded shareholders for decades. The 51% gross margin and the steady operating leverage are evidence of genuine execution, not financial engineering. The UniFirst acquisition, whatever the bought-growth concern, removes a meaningful competitor and adds density that should compound for years. And critically, Cintas has earned its premium multiple repeatedly across cycles — the stock has looked expensive for most of its history and has rewarded holders who paid up anyway, because the compounding kept arriving. The "expensive" call on Cintas has been wrong far more often than it has been right.

The honest synthesis is that Cintas is one of the finest compounders in the American market, executing superbly, with a real moat and a long runway in a consolidating industry — and that it is also an employment-linked, partly cyclical, decelerating services business now leaning on a roughly $5.5 billion acquisition, priced at 36 times earnings as though none of those qualifiers existed. The bull is right that the quality, the moat, the capital discipline, and the track record are all genuine and have humbled skeptics before. The skeptic notes that the growth is decelerating, that bought growth is supplementing organic, that the EPS is buyback-flattered and adjusted-number-dependent, that the gross margin is at a cyclical peak, and that the multiple prices a perfection the labor market does not guarantee.

The asymmetry of paying for perfection

Pull it to the level of risk and reward, because that is where the discipline lives. Buying a wonderful business is not the same as making a wonderful investment; the price paid determines which one you get. At a moderate multiple, Cintas's quality, moat, and durable growth would be an easy and comfortable hold, with the cycle as a manageable risk and the compounding as the reward. At 36 times trailing earnings on 8% underlying growth, the calculus inverts. The premium multiple has already priced the quality, the moat, and years of continued compounding; what it has not priced is any meaningful disappointment — a recession that hits wearer counts, a deceleration that continues, an integration of UniFirst that proves harder than the smooth ones before it, or a gross margin that proves to be a peak rather than a plateau.

That is the priced-for-perfection asymmetry. If everything goes right — organic growth holds, UniFirst integrates cleanly, margins stay at records, employment stays strong — the stock can grind higher and grow into its multiple, as it has before. But the reward for that best case is modest, because so much of it is already in the price. The penalty for the ordinary case — a slowdown, a margin give-back, a recession that touches the customer base — is a multiple that has a long way to compress, from 36 times toward something an employment-linked services business more naturally deserves. The business will very likely keep performing; Cintas is too good and too well-run for that to be in serious doubt. Whether the stock rewards anyone who pays 36 times for it from here depends not on Cintas's execution, which is assured, but on the absence of the ordinary disappointments that a perfection price leaves no room to absorb.

The kicker

Cintas is a genuinely great company, and nothing in this piece disputes its excellence — the dominant position, the route-density moat, the record 51% gross margin, the 45-year dividend streak, and the disciplined capital returns are all real and all earned. But the market has confused the quality of the business with the immunity of the business, pricing an employment-linked, decelerating, partly cyclical services firm at 36 times earnings as though its growth floated free of the labor market that feeds it, while that growth slowed to 8.2% organic and the company reached for a roughly $5.5 billion acquisition to keep the top line elevated. That is a software multiple on a headcount business, a peak margin extrapolated as a baseline, and a buyback-flattered EPS dressed as organic vitality. The shirts will keep getting laundered and the routes will keep running with their usual precision. Whether the stock rewards anyone who pays for permanent perfection at the top of a cost cycle depends on something Cintas cannot control and the price has already spent in advance.

The finest uniform-rental compounder in America raised its dividend for the forty-fifth straight year and printed a record gross margin while the market, admiring the streak, paid thirty-six times earnings for a business whose every shirt and floor mat is billed against how many people its customers happen to employ — and decided, on the strength of a long and genuinely brilliant record, that this employment-linked, decelerating, acquisition-leaning services firm should be valued as though it stood outside the labor cycle that has fed its growth from the very beginning, which is the one thing about renting work clothes that no amount of operational excellence has ever managed to repeal.