Value Investing Calculator
The Graham Formula
Benjamin Graham's intrinsic value formula · The Graham Number · Free calculator
V = EPS × (8.5 + 2g) × 4.4 / Y
Complete guide to Graham's 1962 and 1974 formulas, the Graham Number defensive-investor ceiling, worked examples, and a working calculator. When they work, when they break.
1962
Original formula
1974
Revised with AAA yield
8.5
Base P/E (no-growth)
22.5
Graham Number constant
What Is the Graham Formula?
The Graham Formula is a shortcut Benjamin Graham published for estimating the intrinsic value of a common stock. He introduced the original version in the 1962 edition of Security Analysis and revised it in 1974 to account for prevailing interest rates. It is not the centerpiece of Graham's work — that honor belongs to the concept of margin of safety. But it is probably the most-cited formula in value investing because it is simple enough to compute on a napkin.
The revised 1974 formula is:
V = EPS × (8.5 + 2g) × 4.4 / Y
EPS = trailing earnings per share · g = expected 7-10 year growth rate · Y = current AAA bond yield
There is also a second Graham creation — the Graham Number — which is not the same as the Graham Formula but is often confused with it. The Graham Number is a mechanical ceiling derived from Graham's defensive-investor rules in The Intelligent Investor: P/E below 15, P/B below 1.5. Combine those and you get:
Graham Number = √(22.5 × EPS × BVPS)
Both formulas are built into the calculator below. Graham intended them as starting points for analysis, not as final verdicts. He would be horrified at the idea of buying a stock purely because the Graham Number said it was cheap. His full analytical framework — laid out in exhaustive detail in Security Analysis — involves dozens of additional checks on financial strength, earnings stability, dividend record, and management quality.
Graham Formula Calculator
Runs entirely in your browser. No data is collected or transmitted. Defaults are for a hypothetical mid-cap consumer staples company; replace them with the numbers for the stock you are analyzing.
Graham Formula Calculator
Enter the numbers below. All calculations run in your browser — nothing is sent anywhere.
Diluted EPS from the most recent 10-K or 10-Q.
Analyst consensus or your own estimate. Be conservative.
Moody's Seasoned AAA yield — look up on FRED.
Shareholders' equity ÷ diluted shares outstanding.
Optional — used to compute your margin of safety.
$98.00
Margin of safety vs. price: 38.8%
$122.50
V = EPS × (8.5 + 2g)
$53.03
√(22.5 × EPS × BVPS) · -13.1% MoS
How to read this: The revised formula is Graham's most-cited intrinsic-value estimate, but it is a rough approximation, not a precise number. Use it as a sanity check, not a price target. A margin of safety below 25% means you are paying close to fair value — Graham wanted 33-50% or more before he would buy.
The Revised 1974 Formula Explained
Every term in the formula is doing a specific job. Understand them and you will understand why the formula works (and when it does not).
EPS · the earnings engine
Trailing 12-month diluted earnings per share. Graham specifically warned against using peak-cycle or one-time inflated EPS. Normalize across a full business cycle if you can. Exclude non-recurring gains.
8.5 · the no-growth baseline
The P/E multiple Graham believed a stable but non-growing company deserved. Equivalent to an 11.8% earnings yield, which roughly matches the long-run real return on US stocks. Non-negotiable — do not change this.
2g · the growth premium
Graham adds 2 P/E points for every percentage point of expected 7-10 year growth. A company growing 8% gets 8.5 + 16 = 24.5 times earnings. Graham capped this implicitly by warning against growth rates above 20%.
4.4 / Y · the interest-rate adjustment
4.4% was the AAA corporate bond yield when Graham published the revised formula in 1962. Dividing by the current yield Y scales the valuation up when rates are low and down when rates are high. It is a crude but effective discount-rate adjustment.
Put together, the formula is saying: “Start with a base P/E of 8.5 for a no-growth company, add 2 points for every point of expected growth, then adjust up or down based on whether interest rates are lower or higher than the 1962 baseline.” That is a remarkably compact way to communicate the three big drivers of equity valuation: earnings, growth, and discount rate.
The Graham Number — Defensive Investor Ceiling
The Graham Number comes from two of Graham's mechanical rules in The Intelligent Investor, Chapter 14:
- Rule 1: Price-to-earnings ratio should be below 15.
- Rule 2: Price-to-book ratio should be below 1.5.
- Combined: P/E × P/B should not exceed 22.5 (which is 15 × 1.5).
Solving that constraint for price gives you the formula: maximum price = √(22.5 × EPS × BVPS). This is the Graham Number. If a stock trades below it, it satisfies Graham's defensive-investor price ceiling. If it trades above, it does not.
The Graham Number is extremely strict. In today's market, very few large-cap stocks clear it. That is not a bug — Graham wrote the rule during a period where many stocks did qualify, and modern markets have permanently higher valuations. Use the Graham Number as one screen among many, not as a standalone buy signal.
Critically, the Graham Number is hostile to asset-light businesses. A software company with $2 EPS and $3 BVPS has a Graham Number of roughly $11.62, which looks absurd compared to the stock's actual value. Graham designed this formula for industrial companies with real book value — factories, inventory, machinery. For modern tech companies, the Graham Number is not the right tool.
1962 vs. 1974 — What Changed and Why
Original (1962)
V = EPS × (8.5 + 2g)
Graham's first published intrinsic-value shortcut. A clean equation with two inputs: current earnings and expected growth. Interest rates are nowhere in the formula — which was a reasonable omission in an era of stable rates around 4%.
Revised (1974)
V = EPS × (8.5 + 2g) × 4.4 / Y
After watching 1970s inflation spike corporate bond yields toward 9%, Graham added the 4.4/Y term. When rates are higher than 4.4%, the multiplier is below 1 and the valuation falls. When rates are lower, the multiplier is above 1 and the valuation rises. This is a primitive but effective discount-rate correction.
Practical note: Always use the 1974 revised version unless you are specifically trying to replicate a 1962 analysis. The difference matters enormously. At a 5.5% AAA yield (early 2026), the 4.4/Y multiplier is 0.80 — the revised formula produces valuations 20% lower than the original. At 2% yields (2020), the multiplier was 2.20 and the revised formula produced valuations more than double the original.
Worked Examples
Three hypothetical companies showing where the formula produces reliable signal and where it breaks down.
Hypothetical Consumer Staples Co.
Modest undervaluation — textbook Graham candidateEPS
$5.00
Growth
6%
AAA Yield
5.5%
BVPS
$28.00
Price
$62.00
Revised Value
$82.00
Graham Number
$56.12
Margin of Safety
24.4%
A slow-growing, profitable, asset-heavy company is exactly the type of business Graham designed his formula for. The 6% growth assumption is defensible, book value supports the Graham Number, and the 5.5% AAA yield is close to Graham's 1974 baseline of 4.4%. The formula gives a reliable estimate here.
Hypothetical Industrial Cyclical
Use with caution — cyclical earnings distort the outputEPS
$8.00
Growth
3%
AAA Yield
5.5%
BVPS
$55.00
Price
$75.00
Revised Value
$92.80
Graham Number
$99.50
Margin of Safety
19.2%
Cyclical industrials have earnings that swing 3x from peak to trough. Using trailing peak EPS will make the stock look cheap; using trough EPS will make it look expensive. Normalize across a full cycle before plugging anything into Graham's formula.
Hypothetical Software Growth Story
Formula breaks — do not use this outputEPS
$1.50
Growth
25%
AAA Yield
5.5%
BVPS
$6.00
Price
$180.00
Revised Value
$70.20
Graham Number
$14.23
Margin of Safety
-156.4%
A 25% growth rate plugged into (8.5 + 2g) produces a P/E of 58.5, which Graham himself would have rejected. Software companies with minimal book value are also hostile to the Graham Number. This is an example of where the formula cannot help you — you need a different valuation framework.
I Document Every Trade — Even the Losses
Options record: 1W-8L. Net worth: 100% GSE preferred. Get the unfiltered updates.
Unsubscribe anytime. I respect your inbox more than Congress respects property rights.
When the Graham Formula Works
The Graham Formula is a sharp tool when applied to the companies Graham actually had in mind. It works best for:
- ✓Mature, profitable businesses with at least 10 consecutive years of positive earnings. Think consumer staples, utilities, industrial manufacturers.
- ✓Modest growth expectations in the 3-12% range. Anything higher and Graham's 2g adder overshoots what the formula was designed to handle.
- ✓Asset-heavy companies where book value reflects real economic value — factories, equipment, inventory. The Graham Number is particularly useful here.
- ✓Stable interest rate environments where the AAA yield is reasonably close to the 4.4% baseline. The 4.4/Y correction is crude and gets wild at extremes.
- ✓Screening and sanity checks — as one input among several. Never as the sole decision criterion for a large position.
When the Graham Formula Breaks
The formula is not an oracle. It has specific failure modes. Recognize them before you trust the output:
- ✗Negative or near-zero EPS. The formula multiplies by EPS, so unprofitable companies produce meaningless or negative values. Use a different framework for turnarounds.
- ✗High-growth technology companies. A company growing 30% gets a P/E multiple of 68.5 in the formula — higher than any stock Graham would have touched. The formula breaks because Graham's 2g linear adder was calibrated for a 5-15% growth world.
- ✗Asset-light software businesses. The Graham Number is useless when book value is trivial. Microsoft's Graham Number would be far below its actual value — because Microsoft's real assets (code, brand, network effects) do not show up on the balance sheet.
- ✗Extreme interest rate environments. At 1% AAA yields (2020), the 4.4/Y multiplier hits 4.4x, inflating valuations to absurd levels. At 15%+ yields (1981), the multiplier crushes valuations below what even panicked markets agreed on.
- ✗Cyclical businesses at cycle peaks. Using peak EPS for a commodity producer or homebuilder makes the stock look cheap right before earnings collapse. Normalize across a full cycle.
- ✗Earnings manipulation. If a company is juicing EPS through aggressive capitalization, reserve reversals, or one-time gains, the formula rewards the manipulation. Read the 10-K before you trust any EPS input.
In all of these cases, you need to either fix the inputs (normalize EPS, cap growth, sanity-check the yield) or use a different valuation framework entirely. The formula will happily spit out a number for any inputs you give it. Your job is to know when that number means something and when it is statistical noise.
How I Use Graham's Formula in 2026
I do not use the Graham Formula as a primary valuation tool. My main framework is a DCF with explicit free-cash-flow projections, a weighted average cost of capital, and a normalized terminal value. But the Graham Formula sits on my desk as a sanity check — a second opinion I consult whenever my DCF is producing a surprising number.
Specifically: if my DCF says a stock is worth $80 and the Graham Formula says it is worth $25, I go back and audit my DCF assumptions. Usually I am over-modeling growth or under-modeling the discount rate. The Graham Formula is blunt but honest — it cannot be fooled by fancy spreadsheet gymnastics.
I also use the Graham Number as a first-pass screen when hunting for deep-value names. If I am looking at small-cap industrials or regional banks, a stock trading below its Graham Number goes on the watch list. Most will not pan out — the low valuation is usually earned — but the screen catches a few legitimate bargains every year.
For the full Graham framework — including all the qualitative and quantitative filters beyond these two formulas — see my breakdowns of Security Analysis, the free Security Analysis PDF, and The Intelligent Investor. The formulas are a five-minute shortcut; the full framework is a five-year education.
The Source Material
Graham's formulas appear in both of his major books. To understand the formulas the way Graham understood them, you need the books — not just the equations.
Frequently Asked Questions
What is the Graham Formula?
The Graham Formula is Benjamin Graham's shortcut for estimating the intrinsic value of a stock. The revised 1974 version is V = EPS × (8.5 + 2g) × 4.4 / Y, where EPS is trailing earnings per share, g is the expected 7-10 year annual growth rate (as a number, e.g. 8 means 8%), and Y is the current AAA corporate bond yield. The 8.5 multiplier represents the base P/E Graham assigned to a no-growth company, the 2g term rewards growth, and the 4.4/Y adjustment scales the answer to today's interest-rate environment.
What is the Graham Number?
The Graham Number is a separate formula Graham introduced for defensive investors: V = sqrt(22.5 × EPS × BVPS), where BVPS is book value per share. It comes from combining two of Graham's rules of thumb — a P/E below 15 and a price-to-book below 1.5 — into a single ceiling. If a stock's price is below the Graham Number, it clears Graham's mechanical defensive-investor filter. If it is above, it does not. Use it as a screen, not a precise valuation.
Is the Graham Formula still accurate in 2026?
It is directionally useful but should never be the only input. The formula was designed in a 1960s-70s world dominated by industrial companies with tangible book value, stable earnings, and modest growth expectations. It does not handle asset-light software businesses, negative-growth cyclicals, or companies that reinvest heavily at the expense of reported earnings. Use it as one sanity check alongside DCF, comparables, and qualitative analysis — not as a standalone verdict.
What is the difference between the 1962 and 1974 versions?
The original 1962 version was V = EPS × (8.5 + 2g). In 1974, Graham added the 4.4/Y adjustment to account for the fact that when interest rates rise, the present value of future earnings falls, so stocks are worth less. Practically, the revised formula produces lower valuations in high-rate environments (like 2026) and higher valuations when rates are low. Always use the revised formula unless you are specifically trying to replicate a 1962 result.
Why did Graham pick 8.5 as the base P/E?
Graham believed a stable business with zero growth deserved a P/E of roughly 8.5 — equivalent to an 11-12% earnings yield. That roughly matches the long-run average earnings yield of the stock market. The 2g adder then rewards growth: a company growing 10% per year would be worth (8.5 + 20) = 28.5 times earnings in the original formula. Graham himself acknowledged this was a simplification. He warned that his formula did not belong in a textbook as gospel — it was a shortcut for a working investor.
What growth rate should I use in the Graham Formula?
Use your best estimate of 7-10 year annual earnings growth. Analyst consensus is a starting point but tends to be too optimistic. Historical 5-10 year earnings growth is a useful sanity check. Graham himself warned that growth rates above 20% are almost never sustainable — most companies growing that fast mean-revert to single digits within a decade. If you cannot justify a number below 15%, your margin of safety should be large enough that even a 5% growth rate still leaves the stock attractive.
Where do I find the AAA bond yield?
Use Moody's Seasoned AAA Corporate Bond Yield, published daily on the Federal Reserve Economic Database (FRED) under series AAA. In early 2026, that yield is roughly 5.5%. The 10-year Treasury yield is a reasonable proxy if you cannot find the Moody's number, but Graham specifically chose AAA corporate because it reflects the opportunity cost of equity capital more accurately than sovereign debt.
What margin of safety should I require?
Graham argued that the price should be at least one-third below your intrinsic value estimate — a 33% margin of safety. For higher-quality businesses with predictable earnings, some investors accept 25%. For cyclical or speculative companies, demand 50% or more. Remember that the Graham Formula itself has large error bars. A 10% margin of safety on a rough estimate is not a margin of safety at all — it is a rounding error.
When does the Graham Formula break?
It breaks when any input is unreliable. Negative EPS makes the formula nonsensical. Growth rates above 20% produce absurd multiples. AAA yields below 1% (as seen in 2020-2021) inflate the formula to unreasonable levels. Asset-light software businesses with minimal book value can produce misleading Graham Numbers. Companies that heavily buy back stock distort EPS growth. Use the formula when the inputs are stable and defensible, not as an oracle.
How does the Graham Formula compare to a DCF?
A discounted cash flow (DCF) model is more flexible and more honest about its assumptions — you explicitly project free cash flow, choose a discount rate, and estimate a terminal value. The Graham Formula bakes all those decisions into three constants (8.5, 2, and 4.4). That is its weakness (less precision) and its strength (faster, fewer inputs to manipulate). Use both. If your DCF and the Graham Formula both say a stock is undervalued, you have real signal. If they disagree sharply, you have found something worth investigating.
Recommended Resources
Tools & books I actually use and recommend
TradingView
Best charting platform out there. Real-time data, screeners, and a community of millions of traders.
Try TradingViewThe Psychology of Money
Morgan Housel on why managing money is about behavior, not intelligence. Short, brilliant chapters you'll re-read.
View on AmazonThe Intelligent Investor
Ben Graham's timeless guide to value investing. The book Warren Buffett calls "the best investing book ever written."
View on AmazonSome links above are affiliate links. I only recommend products I personally use. See my full disclosures.
Keep Exploring
Security Analysis by Benjamin Graham
The graduate-level companion. Free 6th edition PDF + deep breakdown of Graham's full framework.
Read moreCore ConceptMargin of Safety Explained
Graham's central concept. Why a 33-50% margin is the foundation of every value trade.
Read moreChecklistValue Investing Checklist
The full filter set Graham used, distilled into a workable modern checklist.
Read moreDownloadSecurity Analysis (Free PDF)
Download the free 6th edition and read Graham's original derivations.
Read moreGuideThe Intelligent Investor Guide
Chapter-by-chapter breakdown of Graham's other masterpiece. Home of the Graham Number.
Read morePhilosophyValue Investing
The philosophy Graham invented, applied to modern markets.
Read more