Net margin below operating margin does not clear a company of an investment mark

Claim

The 2026-08-04 Broadridge analysis cleared the name of earnings-quality distortion with a single test: net margin 15.0% sits below operating margin 17.4%, therefore no equity-mark distortion, therefore "reported earnings are operating earnings."

That conclusion was wrong, and the test could not have caught it. $2.00 of FY26's $9.60 GAAP EPS — 20.8% — was a non-cash digital-asset mark. Broadridge's own 8-K GAAP-to-adjusted bridge lays it out: 9.60 + 1.74 amort + 0.15 acq + 0.11 restr − 1.94 digital-asset gains − 0.06 investment gain. The company carries digital assets marked to market, so GAAP EPS now carries crypto beta — and it had already turned, with an $11M loss in Q4 alone.

A portfolio-specific passage was removed from the public build.

Why the tripwire cannot work

The inequality only inverts when the non-operating gain exceeds the combined drag of interest and tax. A $227M gain against $7.48B of revenue moves net margin by roughly 3 points — enough to matter enormously to EPS, nowhere near enough to push net margin above operating margin. The test has a floor below which it is silent, and that floor is far above the level at which a mark starts distorting the multiple.

The failure is worse than a miss, because the test was reported as conclusive. A silent check invites a second look; a check that returns "clear" ends the inquiry.

The conclusive test instead

1. Compare operating income to pretax income. The gap is where non-operating items live. This catches a mark of any size, not just an inverting one.
2. Read the cash-flow statement's non-cash reconciling lines. A mark must be reversed out of operating cash flow, so it is named there explicitly.
3. Read the company's own GAAP-to-adjusted bridge in the 8-K or earnings release when one exists. It is the issuer telling you which lines it does not consider operating.

What it cost

Every downstream multiple on the name was flattered: clean GAAP EPS is $7.60 not $9.60, clean trailing P/E 21.4x not 16.95x, clean EV/EBITDA 13.3x not 12.04x, and clean net income 3yr CAGR 12.2% not 21.3%. FCF, OCF, operating income, gross margin, revenue and adjusted EPS were all untouched — the mark is non-cash — so the cash case survived intact while the GAAP headline was overstated by a fifth for eight weeks.

A forward trap follows from the same fact: FY27's GAAP compare will look bad against a mark-inflated FY26 base for non-operating reasons. Use adjusted EPS and FCF on this name.

Related

• [[pitfall-net-margin-above-operating-margin-is-a-tripwire]] — the sibling note, describing the case where the inequality does invert. This note is the complement: the tripwire's silence is not evidence of absence.
• [[pattern-ai-build-inflates-earnings-while-destroying-fcf]] — the same shape in a different guise: non-operating marks inflating reported earnings while cash tells another story.

History

• 2026-09-28 — written after the BR re-analysis retracted the 2026-08-04 clearance. Cause classified as wrong test, not wrong data: every input the baseline used was correct, and the diagnostic applied to them could not detect what was there.