Basic Prerequisites: 02 · Profitability through 04 · Liquidity & Leverage.
1. Intuition & Practical Objective
Every ratio in this folder is a flow-over-stock question, and every ratio inherits the two weaknesses of that form: the numerator and denominator are accounting constructs that can be gamed, and the ratio’s level is only meaningful relative to a peer group and a time series. This page is the defense layer. It collects the ways ratios fail - not as a list of “gotchas” but as consequences of first principles - and gives the working rules that keep a screen honest. If the hub table is the engine, this page is the quality control.
The deepest principle: a ratio is only as trustworthy as the line items that feed it, and line items are produced under accrual accounting. Because earnings are accrued (recognized when earned, not when cashed), management has discretion over when revenue and expense hit - and that discretion is exactly what Sloan’s accruals anomaly, Beneish’s M-score, and Penman’s reformulation all police. Every red flag on this page traces back to that single fact.
2. Mathematical Ground Truth & Derivations
The accrual identity - where manipulation lives. Net income and operating cash flow differ by accruals:
Sloan (1996) shows the accrual component of earnings is less persistent than the cash component, so high-accrual (low-cash) earnings are systematically overvalued - the accruals anomaly. Beneish’s M-score operationalizes this as a multivariate screen for likely manipulators. The ratio-level red flag is immediate: any profitability ratio (ROE, ROA, margins) is inflated when earnings are accrual-heavy - so pair every earnings ratio with the cash-flow version of the same question (e.g. FCF yield vs. earnings yield; CFO vs. net income).
The negative-denominator singularity. For a loss-making or balance-sheet-damaged firm:
These are mathematical singularities in the ratio, not economic facts about the firm - the output stops carrying meaning even though the company still exists. The practice rule is to change the lens (EV/Sales, book-based anchors, the Z-score) rather than to read the broken ratio literally.
Cross-sectional vs. absolute. The comparison theorem of this folder: a ratio’s information content is its deviation from a relevant peer/benchmark, not its raw size. Fama–French 1992 established P/B and E/P as cross-sectional return drivers; the same discipline applies to margins, turnover, and leverage - all of which vary hugely and legitimately across industries (high-margin/low-turnover luxury vs. low-margin/high-turnover grocery, the Penman Ch 11 figure).
3. Computational Implementation - a red-flag scanner for the sample firm
Stdlib only. Computes the standard earnings-quality and distress red flags from the statements: accruals ratio, the FCF-vs-earnings gap, and the negative-denominator stress test that shows why broken ratios must be skipped, not read.
4. Failure Modes & First-Principles Breakdowns
The catalog, each tied to its first principle.
- Accrual-gamed earnings (first principle: accrual discretion). ROE/margins/P/E all inflate when earnings carry heavy positive accruals. Red flag: earnings yield ≫ FCF yield, or CFO persistently < net income. Check: the Sloan accrual ratio (computed above); the Beneish M-score for the full screen. → Accounting Quality & Red Flags.
- Negative/tiny denominators (first principle: the ratio’s singular point). Losses kill P/E; near-zero equity kills ROE and D/E; negative EBITDA kills NetDebt/EBITDA. Rule: identify the singularity, skip the broken ratio, switch to EV/Sales, book anchors, or the Altman Z. Never quote a negative P/E as “cheap.”
- Absolute-threshold reading (first principle: ratios are comparative). A “current ratio above 2” or “P/E below 15” as a universal rule ignores that the right level is sector- and growth-dependent. Rule: compare to sector medians and the firm’s own history; treat absolute cutoffs as starting filters, not verdicts.
- Convention mismatch (first principle: units must match). Averages vs. ending balances, before- vs. after-tax operating income, EBITDA with/without leases - any of these changes the number more than the signal. Rule: fix one convention and keep it constant across the comparison set.
- Financials/cyclicals distort the standard set (first principle: flow-stock alignment). Banks’ “equity” and “leverage” are regulatory constructs; cyclicals’ earnings collapse at the trough, inflating P/E exactly when the firm is cheapest. Rule: for financials use book/regulatory ratios; for cyclicals use mid-cycle normalized earnings (ties to Equity Valuation).
- Survivorship/look-ahead in screens (first principle: data hygiene). Testing a ratio screen on today’s live firms bakes in survivorship; using restated data leaks the future. Rule: point-in-time, as-reported data - Data Sources & Corporate Data.
5. References
- Sloan, Richard: “Do Stock Prices Fully Reflect Information in Accruals and Cash Flows About Future Earnings?” (TAR, 1996)
- Beneish, Messod: “The Detection of Earnings Manipulation” (FAJ, 1999)
- Dechow, Ge & Schrand: “Understanding Earnings Quality…” (JAE, 2010)
- Penman, Financial Statement Analysis and Security Valuation
- Fama & French, “The Cross-Section of Expected Stock Returns” (JF, 1992)
- Schilit, Perler & Engelhart, Financial Shenanigans
6. Connected Graph Bridges
- Back: 02 · Profitability · 04 · Liquidity & Leverage · Index Hub
- Forward: 06 · Advanced Extensions
- Defense layer: Accounting Quality & Red Flags · Quantitative Fundamental Investing