A.4.3 Screening Metrics

Basic Prerequisites: Core Financial Ratios (margins, leverage, yields) and 02 · Graham Criteria.

---

1. Intuition & Practical Objective

Graham’s screen asked “is it cheap and safe?” The working practitioner’s screen asks a complementary question: “is it a good business that is also reasonably priced?” This page is the metric layer - the handful of numbers a value/quality screen actually computes, and the two ways to combine them: a hard filter (all-or-nothing gates) and a composite ranking (z-scores summed into a score).

Three principles:

1. A metric is only useful with a threshold and a comparison set. “Operating margin of 12%” means nothing until you say ”≥ 8%, and above the sector median.” Every metric below is quoted with the threshold a screen would apply.
2. Growth without profitability is a money-loser; profitability without cash is an illusion. Hence the screen pairs a growth metric (revenue CAGR), a profitability metric (operating margin), a solvency metric (Net Debt/EBITDA), and a cash metric (FCF yield). Each catches a failure the others miss.
3. Filtering and ranking are different jobs. A hard filter shrinks the universe to names that clear every gate; a composite rank then orders the survivors so you read the best candidates first. Use both - the filter for hygiene, the rank for attention.

---

2. Mathematical Ground Truth & Derivations

10-year revenue CAGR - the compound annual growth of the top line:

$$\text{RevCAGR}={\left(\frac{S_t}{S_{t-10}}\right)}^{1/10}-1.$$

Using the compound rate (not the simple average of annual growth) is essential: it is the single rate that takes $S_{t-10}$ to $S_t$ in ten years, and it is the only correct summary when year-to-year growth varies. A screen requires it above a floor (e.g. $\ge 8\%$) to exclude melting ice cubes and prove durable demand.

Operating margin - profit from the core business per dollar of sales:

$$\text{OpMargin}=\frac{\text{EBIT}}{S}.$$

It is the cleanest single measure of pricing power and cost discipline (it excludes interest and tax, so it is comparable across capital structures). Sector-relative by construction.

Net Debt / EBITDA - years of operating cash flow needed to clear net debt:

$$\text{ND/EBITDA}=\frac{\text{TotalDebt}-\text{Cash}}{\text{EBITDA}}.$$

The lower, the safer; a screen typically caps it at $2.0$ (above that, a downturn threatens solvency). It is undefined when EBITDA $\le 0$ - a negative-EBITDA firm is automatically excluded, which is correct.

FCF yield - cash actually thrown off per dollar of market value:

$$\text{FCFy}=\frac{FCF}{\text{MC}},FCF=\text{CFO}-\text{Capex}.$$

The cash counterpart to the earnings yield; a screen floor of ~$3\%$ demands the business converts accounting profit into real cash.

Composite score (z-score sum). Because metrics have different units, standardise each before combining. For a metric $x$ across the universe with mean $\mu$ and standard deviation $\sigma$:

$$z_i=\frac{x_i-\mu }{\sigma },{\text{score}}_i=z(\text{CAGR})+z(\text{OpMgn})+z(-\text{ND/EBITDA})+z(\text{FCFy}).$$

Sign-flipping leverage (so lower debt scores higher) aligns every term so that higher composite = better. (This z-score-sum construction is exactly the composite that Piotroski, Altman, and every quantamental screen formalize - see 06 · Advanced Extensions.)

---

3. Computational Implementation - filter, then rank

Runs on the standard library only. It first applies the owner’s four hard gates (Rev CAGR ≥ 8%, Op margin ≥ 8%, Net Debt/EBITDA ≤ 2.0, FCF yield ≥ 3%), then ranks the survivors by the composite z-score.

The filter rejected Omega Builders (7% margin, 3.29× leverage - under-earning and over-borrowed) and Kappa Pharma (6.1% growth, below the floor - a stalling business, even though its balance sheet is pristine). The ranking then separates the survivors: Sigma Tools leads on growth and near-zero leverage; Delta Chemical - despite crossing every gate - ranks last because its leverage is the highest of the three and its FCF conversion the thinnest. That is the point of combining a filter with a composite: the filter answers “is it admissible?”, the score answers “which one do I read first?“

---

4. Failure Modes & First-Principles Breakdowns

1. Thresholds are not universal. An 8% margin is stellar in grocery and terrible in software; a 2.0× leverage cap is generous for utilities and reckless for cyclicals. Every metric must be screen-relative to its sector - an absolute threshold applied across industries produces a portfolio of accidental industry bets (the Penman comparative principle, Core Financial Ratios).
2. CAGR hides the path and the endpoint. A 12% ten-year CAGR can be a smooth compounder or a company that tripled once in year 3 and shrank since. Always pair CAGR with the annual series and the most recent two years - a falling headline that still clears the ten-year gate is a warning, not a pass.
3. EBITDA-based leverage flatters capex-heavy firms. Net Debt/EBITDA ignores the capital required to keep the business running; a capital-intensive firm with 2.0× EBITDA leverage may be far more stretched than a software firm with the same ratio. Cross-check with Net Debt/FCF and interest coverage.
4. The composite is only as good as its inputs, and can be gamed by a single outlier. A firm with one enormous FCF year can dominate the z-scores. Winsorize extreme values and check that the ranking is stable across reasonable thresholds before trusting it.
5. Screen-optimism: the data hygiene trap. Running these gates on today’s survivors with restated numbers bakes in survivorship and look-ahead - the screen “works” because you already know who lived. Use point-in-time, as-reported data (Data Sources & Corporate Data).

---

5. References

• Graham, Benjamin: The Intelligent Investor (2003 annotated ed.)
• Palepu, Krishna & Healy, Paul: Business Analysis and Valuation: Using Financial Statements (Cengage)
• Fridson, Martin & Alvarez, Fernando: Financial Statement Analysis: A Practitioner’s Guide (Wiley)
• O’Shaughnessy, James P.: What Works on Wall Street (McGraw-Hill, 4th ed.)
• Gray, Wesley R. & Carlisle, Tobias E.: Quantitative Value (Wiley, 2013)
• Novy-Marx, Robert: “The Other Side of Value: The Gross Profitability Premium” (JFE, 2013)

---

6. Connected Graph Bridges

• Back: 02 · Graham Criteria · Core Financial Ratios
• Continue: 04 · The Research Workflow (the screen feeds the reading) · 06 · Advanced Extensions (factors & the quantamental composite)
• Data: Data Sources & Corporate Data (point-in-time inputs) · Accounting Quality & Red Flags (why the earnings behind the margin may be fake)
• Quant translation: Quantitative Fundamental Investing · Fundamental Multi-Factor Models