Deep Value Screen
Stocks trading at low earnings and book value multiples — classic Graham-style deep value candidates.
What Is a Deep Value Screen?
Deep value investing is the oldest systematic approach to equity selection. Benjamin Graham's foundational insight — that the market periodically prices securities below their intrinsic value, and that a disciplined investor can exploit this gap — forms the intellectual basis for the strategy. The simplest implementation focuses on two metrics: price-to-earnings (P/E) and price-to-book (P/B). Companies trading at low multiples on both dimensions are, in theory, available at a discount to both their earnings power and their asset value.
Historically, the evidence for deep value investing is strong. The Fama-French "value factor" (long low P/B, short high P/B) is one of the most robustly documented equity return premiums in academic finance, observed across nearly a century of US data and replicated in international markets. However, the value premium has been notably weak since roughly 2007, and especially from 2017 to 2022 — a period of sustained growth-stock outperformance driven by low interest rates, platform-economy dynamics, and the rising economic value of intangibles that do not appear on balance sheets under US GAAP.
The debate about whether the value premium is dead, in a prolonged drawdown, or structurally shifting is genuinely unresolved. Reasonable people disagree, and the empirical evidence through 2026 is still being interpreted. What is fair to say: deep value has worked over very long time horizons, tends to be highly cyclical, and requires patience and psychological fortitude that most investors find difficult to maintain through multi-year underperformance.
A practical deep value screen excludes financial companies (where book value has a different meaning) and requires positive earnings — buying loss-making companies just because they are "cheap" on book value is a different strategy (distressed investing) with different risk characteristics.
The criteria in plain English:
- P/E below 10 — you are paying less than 10 years of current earnings for the business
- P/B below 1.5 — the stock is available at or near asset value
- The company is profitable — negative P/E ratios indicate losses, not cheapness
- Reasonable financial health — extreme leverage can make a "cheap" stock appropriately cheap
How to Run This Screen in GeminIQ
Step 1. Open the GeminIQ Screener
Step 2. Add the following filters:
| Filter | Operator | Value |
|---|---|---|
| P/E TTM | ≤ | 10 |
| P/E TTM | ≥ | 1 |
| PB | ≤ | 1.5 |
| Debt To Equity | ≤ | 1.5 |
The P/E floor of 1 removes companies with negative earnings (where P/E is reported as negative by convention) from the results. The debt-to-equity ceiling ensures you are not surfacing companies that appear cheap primarily because leverage has impaired the equity value.
Step 3. Set Sort By to P/E TTM, direction Ascending — cheapest on earnings first.
Step 4. Review the sector composition of results. Deep value screens naturally surface cyclical industries — energy, basic materials, regional financials, traditional retail — at trough valuations. Determine whether the low multiple reflects a cyclical trough (potentially attractive) or structural deterioration (value trap).
GeminIQ Tip: For each candidate, open the Visualizations tab and chart revenue and net income over 10+ quarters. A company with low P/E because earnings are at a cyclical peak is very different from one with low P/E because earnings have been consistently modest. The trailing multiple does not distinguish between these — the trend does.
What Aggregator Data Misses for This Screen
Book value and intangibles. Under US GAAP, internally developed intangible assets — brand value, customer relationships built organically, proprietary technology created in-house — are generally expensed rather than capitalized. This means a consumer brand company that spent decades building a recognizable brand has a balance sheet that understates economic book value, while an acquisition-heavy company that purchased the same brand carries it as goodwill. Two companies with identical economic assets can show wildly different P/B ratios depending on whether their asset base was built internally or acquired.
Aggregators further complicate this by sometimes stripping goodwill and intangibles from book value to calculate "tangible book value," which changes P/B dramatically for acquisition-heavy companies. GeminIQ reports P/B using total equity as-filed, with goodwill and intangibles visible as separate balance sheet lines — allowing you to calculate both the reported and tangible variants.
Earnings normalization. Deep value screens based on low P/E are highly sensitive to the earnings figure used. Aggregators frequently substitute "adjusted earnings" for GAAP net income when computing P/E ratios, stripping out restructuring charges, impairment charges, and litigation settlements. A company with 10× adjusted earnings may trade at 6× GAAP earnings if it is taking recurring "one-time" charges. GeminIQ's P/E is calculated from as-filed GAAP net income, which is the legally reported figure and the basis for most academic value research.
Pension and lease obligations inflating apparent book value. A company with a large unfunded pension deficit and significant off-balance-sheet operating leases can show an artificially high book value if those obligations are not fully reflected. Since ASC 842 and SFAS 158, more of these obligations appear on-balance-sheet, but the treatment varies enough that directly examining the source filing for each candidate is warranted.
GeminIQ builds its financial statement database from raw SEC filings, not from third-party financial data APIs.
This screen is educational and does not constitute investment advice. Past performance of any strategy does not guarantee future results.