On the close of Friday 4 September 2026, Paychex traded at $121.71. Sperio values a business from its filed accounts and a discount rate, giving no credit of any kind to future growth, and arrives at $185.14.
That is a 52% gap. It is the fifty-second widest gap in the index this week, and it is the one the Sunday ranking put first. Explaining why is most of the article.
Why this one, and not the widest discount on the list
The widest discount in the index this week belonged to a fertiliser company whose net income had fallen 83%. The one after it was an office landlord. Ranking by discount answers "what looks cheap", and it answers it reliably with whichever company is deteriorating fastest, because a collapsing denominator makes any price look like a bargain.
So the ranking works the way a fund's does: two hard gates first, then an ordering.
The gates are these. Does the business durably earn more on its capital than the capital costs — return on invested capital clearing ten percent in four fifths of the readable fiscal years and in each of the last three. And are both lines going the right way — revenue and operating income each compounding upward over the full readable history. Of the 457 companies the board had valued, 109 cleared both.
What orders the survivors is an expected rate of return: the upside if the company is worth what its franchise record supports, the downside if it is worth only its trough case, weighted by how strongly the board argued for it. That weighting is worth naming rather than burying — the conviction score is a model output, it has moved more than twenty points on a single company in a day, and that is exactly why it weights the ordering instead of deciding it.
Paychex came out first, and it came out first at a two-year, a three-year and a five-year horizon. Nothing about its position depends on the window chosen.
What the forty percent is
Return on invested capital, across the readable record: about 40%. The supporting lines are consistent with it rather than in tension with it — a 27.0% net margin, a 35.7% free-cash-flow margin, $2.3bn of free cash flow, and a 47.1% return on equity.
The part that matters more than any of those:
- earnings per share, over the four fiscal years the window covers: +14%
- net income: +13%
- share count: −1%
Almost none of the per-share growth is arithmetic. This is the opposite of the pattern where a company retires a fifth of itself and reports the result as growth: here the treasury did essentially nothing, and what reached the shareholder is what the business earned.
The number that does not fit
Revenue has compounded at 8.0% over the full readable history, and at 9.2% over the last three years. Operating income has compounded at 8.0% across the same long record — the two lines that the second gate tests, both going up, at the same rate, for as long as the accounts can be read.
Earnings have not kept up. Over the recent window, +14% in earnings per share is about 4.3% a year — the board's own summary uses that figure and calls it anemic. Three different windows give three different numbers (4.3% annualised over the recent record, 6.5% over the last year, 8.0% over the long history) and all of the short ones sit below the revenue line.
That gap is the whole question. Revenue growing at eight to nine while earnings grow at four means margin is being given away somewhere, and the accounts alone do not say whether that is a cost of acquiring growth or the price of keeping it. Last week's subject had the same shape at a much larger scale and on a much worse business. Here it sits on a company earning forty percent on its capital, which makes it a question rather than a verdict.
The downside, priced
A single valuation cannot be argued with. Three can.
| the trough case | $115.29 |
| no growth credited | $185.14 |
| growth credited on the franchise record | $296.35 |
The first line is the one that decides how much of a position a professional would take, and here it does the thing a trough case is not supposed to do: it sits below the price. At $115.29 against the $121.71 the market was paying, a buyer who is wrong about growth and right about nothing else is down a few percent, not protected. That is disclosed on the row rather than filtered out of it, because a fragility that is named can be sized and a fragility that disqualifies a name is a fragility nobody ever reads.
The rate sweep says the same thing from a different direction. At today's cost of capital — 9.77%, built on a 4.77% ten-year Treasury as FRED reported it on 4 September — 109 companies clear both gates and four of them are worth more than their price even in the trough case. Raise the cost of capital by a hundred basis points and 94 still clear the gates but one keeps that cushion. Paychex is not among the four today, and it is not the one that survives.
There is a second disagreement worth putting in writing. On a full-cycle operating multiple this company trades at 34.5× EV/EBIT — ninety-ninth of the 246 names on the cheap-and-growing screen, which is the only screen in this project that has ever been measured against realised returns. The expected-return ranking puts Paychex first. The backtested screen puts it in the middle of the pack. Those two instruments do not agree about this company, and the honest thing is to say which of them has evidence behind it: the one that ranks it ninety-ninth.
The screen also picked its competitor
Sixth on the same list, with the same two gates cleared, is ADP: a 48% return on invested capital, $277.62 against a no-growth valuation of $391.75. Two of the top six names in a ranking of 457 companies are the two large American payroll processors.
That is not confirmation. A screen ranks companies one at a time and has no idea that two of its answers are the same bet — the same customers, the same sensitivity to how many people are on American payrolls, the same exposure to what happens to the interest earned on money held between an employer and its employees. Anyone taking both is taking one position twice.
The case against, in the board's own words
Sperio runs a devil's advocate whose only job is to attack the thesis. It is reproduced without softening, because a recommendation that publishes only its bull case is one that cannot be retracted honestly later:
The core SMB client base is highly cyclical and vulnerable to sluggish GDP growth. Furthermore, anemic 4.3% annualized EPS growth and a high PEG ratio suggest agile competitors may be capping its pricing power.
Both halves land. Paychex sells to small and medium businesses, which are the first to stop hiring and the first to fail, so its unit of revenue is a headcount it does not control. And the pricing-power charge is not rhetorical — it is the same 4.3% that the section above could not reconcile with an 8% revenue line. The bear case and the arithmetic are pointing at the same place from opposite ends.
The debt line is flagged too, at 1.2 times equity. It does not block the name and it is not being hidden.
Why now: nothing has happened
There is no catalyst. The verdict has not moved, nothing in the accounts changed this week, and the gap has simply been open for a while. Saying so is more honest than manufacturing a reason.
One date is worth watching. Third-quarter results are scheduled for 29 September 2026, with consensus at $1.32 a share. For context and not as a prediction: the last four quarters came in at $1.32 against $1.31, $1.71 against $1.67, $1.26 against $1.23, and $1.22 against $1.20 — four beats, none of them larger than 2.4%. A company that clears its guidance by a rounding error every quarter is a company guiding carefully, not one surprising anybody.
The number to read on that day is not the beat. It is whether the distance between the revenue line and the earnings line closed.
What this is, and is not
Every figure here comes from accounts filed with the SEC, from a board valuation dated 3 September 2026, and from a single ranking run dated 6 September 2026. The price used is the close of Friday 4 September 2026. The cost of capital is built from FRED observations as of 4 September 2026. The valuation gives no credit to growth beyond what the franchise record earns, and the ranking that produced this name orders candidates by an expected rate of return — a way of comparing them to each other, not a forecast of what any of them will do.
The board that produced those figures will run on any company you name. What it will not do is decide for you — nothing here is investment advice.