Version 1.1 · September 2026
How to write an investment thesis for a stock: layout, sections, concepts, checklists and valuation methods.
Sources integrated:
- Peter Lynch, One Up on Wall Street — 103 highlights extracted from the book. Marked in the rulebook with the
[LYNCH]tag. - Warren Buffett’s investment principles — widely known rules drawn from the Berkshire Hathaway shareholder letters and his talks. Marked with the
[BUFFETT]tag. - Calculation methods are marked with the
[FORMULA]tag. Untagged items are general thesis-writing practice.
This rulebook was written to be kept open beside you while writing a thesis. At the start of each thesis section you will find what that section is for, what goes into it and which checklist applies. Every technical concept has its own concept card: what the concept is, what it shows, how it is calculated, how it is interpreted, an example and a pitfall.
Table of Contents
- What a thesis is and how to use this rulebook
- The layout of the thesis: what to write section by section
- Concept cards: financial terms and what they mean
- Rules integrated from Lynch
- Buffett’s rules of thumb
- Methods for calculating the future share price
- Templates
- Appendix: Formulas and glossary
1. What a thesis is and how to use this rulebook
An investment thesis is the document that puts in writing, before you buy, why you are buying a stock, what would prove you right and what would prove you wrong. The aim is not to convince yourself but to test yourself. A good thesis makes the sell decision easier far more than the buy decision: when the price falls, the answer to “has the story broken, or has only the price fallen?” is already written in the thesis.
1.1 Five principles
- It must be explainable to a child. Lynch’s yardstick: if you can explain a stock’s story to your family, your friend or your dog in a way a child could understand, you have grasped the situation. A thesis that cannot be told in two minutes is not a thesis.
- It must be falsifiable. Every thesis contains the sentence “if this happens, I was wrong.” Without that sentence a thesis is a wish list.
- You invest in the company, not in the stock market. The thesis is about the company’s earnings, assets and story, not about what the market will do next year. Lynch: “Predicting the economy is futile. Predicting the short-term direction of the stock market is futile.”
- The sell conditions are written before the purchase. Deciding while you hold the stock is hard; the “when do I sell” section of the thesis is tied to predetermined signals, not to price movements.
- The thesis is a living document. Every few months the story is reviewed again; if the company has changed category (from fast grower to stalwart, from turnaround to cyclical), the thesis is reclassified.
“Before buying a stock, I like to be able to give a two-minute monologue that covers the reasons I’m interested in it, what has to happen for the company to succeed, and the pitfalls that stand in its path.” — Peter Lynch, One Up on Wall Street, “The Two-Minute Drill”
1.2 Workflow
- When an idea is born, first read Section 4.1 (personal preparation) and 4.2 (discovery is not a buy signal).
- The thesis is written in the order given in Section 2, keeping the section headings exactly as they are. The “What to write” list under each section is filled in; every question in the “Questions” list is answered in one sentence.
- For every term you do not understand, look at the concept card in Section 3; the card tells you where each number is taken from and how to interpret it.
- The company is placed in one of the six categories in Section 4.3; that category’s checklist and sell signals are copied into the relevant sections of the thesis.
- Section 4.4 (the 13 attributes) and 4.5 (what to avoid) are scored; the Buffett rules in Section 5 are interrogated under the headings “Business, Management, Financials, Price, Behavior”.
- The target price is calculated with at least two of the methods in Section 6, using bear/base/bull scenarios.
- The one-page summary in Section 7 is filled in. If the summary cannot be filled in, the thesis is not yet finished.
- After the purchase, the review template is applied every three months; every decision is recorded in the journal.
1.3 The difference between the two investors and how they are used together in the thesis
The rulebook draws on two different schools; knowing what each one is resolves the places that look like contradictions.
Peter Lynch (managed the Fidelity Magellan Fund from 1977 to 1990) is a stock picker. His method: find the companies you observe in everyday life (the store, the product, the workplace) before Wall Street notices them, put the company into one of six categories and form expectations according to the category. He holds hundreds of stocks, rotates frequently, and also goes into cyclicals and turnarounds. His key concepts: category, story, edge, the P/E–growth relationship, the tenbagger.
Warren Buffett (Berkshire Hathaway) is a business buyer. He sees a stock not as a piece of paper but as a piece of a business; he buys a small number of outstanding businesses at a reasonable price and holds them for a very long time. He generally stays away from turnarounds and cyclicals; he looks for a durable competitive advantage (a moat), high returns on capital and honest management. His key concepts: circle of competence, moat, owner earnings, intrinsic value, margin of safety, Mr. Market.
How they combine in the thesis: Lynch says where the idea will come from and what type of company it is (Sections 2–4, 6). Buffett makes you ask whether this company is a business worth owning and how much to pay for it (Sections 5–9, 12). A stock may fall into Lynch’s “fast grower” category and still fail Buffett’s moat test; in that case the thesis says so explicitly and the position is sized down accordingly.
2. The layout of the thesis: what to write, section by section
A thesis consists of the 15 sections below (0 through 14). The section headings are the same in every thesis, so that theses can be compared with one another and it is clear where to look during the quarterly review. Target for total length: 6 to 12 pages. The cover summary must fit on a single page.
| # | Section | The question it answers in one sentence | Length |
|---|---|---|---|
| 0 | Cover summary | What is this stock, why now, what is the target, what would make me sell? | 1 page |
| 1 | The two-minute story | How would I tell the story to a child? | ½ page |
| 2 | Company and business model | What does the company do, where does the money come from? | ½–1 page |
| 3 | Category | Which of the six categories is it, and what is expected from that category? | ½ page |
| 4 | My edge | What do I know about this company that the market doesn’t? | ¼ page |
| 5 | Moat and niche | Why can’t competitors take this profit? | ½ page |
| 6 | Industry and competition | Is the industry growing, are there new entrants, is the industry “hot”? | ½ page |
| 7 | Management and capital allocation | What are they doing with the money they earn, are they putting in their own money? | ½ page |
| 8 | Financials | Are earnings growing, will the balance sheet hold, is the cash real? | 1–2 pages |
| 9 | Valuation | What am I paying for these earnings, and what will be paid in the future? | 1–2 pages |
| 10 | Risks and falsification conditions | What would mean I was wrong? | ½–1 page |
| 11 | Catalysts and timing | What has to happen for earnings to rise, and when? | ¼–½ page |
| 12 | Position and trading plan | How much, at what price, when do I add, when do I sell? | ½ page |
| 13 | Monitoring plan | What will I look at every three months? | ¼ page |
| 14 | Decision journal | When, at what price, why, and what did I do? | table |
Section 0 · Cover summary
Purpose. Even if the entire thesis is forgotten, this page must be enough to make a decision. One year after the purchase, only this page will be read.
What to write.
- Company name, ticker, exchange, date, price at the time the thesis was written, market capitalization, number of shares outstanding.
[LYNCH]Category (slow grower / stalwart / fast grower / cyclical / turnaround / asset play) and a one-sentence rationale.- Thesis sentence: “I am buying company X because A, B, C; within N years I expect earnings to be … and the stock to reach the … range.“
- Target price range: bear / base / bull and their probabilities; expected annual return (Section 6).
- Maximum purchase price (I do not buy above this price) and position size (percentage of portfolio).
- Three sell triggers (copied from Section 12).
- Date of the next review.
Section 1 · The two-minute story
Purpose. [LYNCH] To tell the story in plain sentences: why I am interested, what has to happen for it to succeed, what the pitfalls along the way are. The sample monologues Lynch gives for each category are the template for this section:
| Category | The axis of the monologue |
|---|---|
| Slow grower | The dividend: how many years it has been raised, whether it has ever been cut, what happened in recessions, whether there is a new business that will accelerate growth. |
| Cyclical | Business conditions, inventories, prices: which phase the cycle is in, plant capacity, cost cutting, which way earnings will turn. |
| Asset play | What the assets are, what they are worth, what is left per share after deducting debt, how cheaply I am buying the remaining business, whether insiders are buying. |
| Turnaround | What the company has done to repair its fortunes, whether the plan is working, whether the unprofitable divisions have been sold, whether there is a share buyback. |
| Stalwart | Where it sits in its P/E range, whether it has already surged in recent months, whether there is something that will accelerate growth (new market, spin-off, divestiture). |
| Fast grower | Where and how it can keep growing fast; whether the formula has been copied in other regions; whether debt is excessive; how much runway is left before the market is saturated. |
Questions. Can I tell this story to a child? Is there anything concrete beyond “the guy on the bus says it’s going to be taken over”? Does the story have substance, or is it only “sizzle”?
Section 2 · Company and business model
Purpose. To write, in my own words and citing sources, how the company makes money.
What to write.
- Products and services; who the customer is, why they buy from this company, how often they buy.
- Segment breakdown: each product group’s contribution to sales and to profit (percentage).
[LYNCH]The percentage-of-sales rule: what percentage of the company’s sales does the product I care about represent? L’eggs was everything for the small Hanes; Lexan was part of a 6.8 percent slice of the giant GE. Even if the product is “the next Pampers,” if it means nothing to the shareholder, the thesis ends there.[LYNCH]Repeat purchase or one-time sale? Do people have to keep buying the product (drugs, soft drinks, razor blades), or do they buy it once and that’s it (toys, Sensormatic’s surveillance systems)? With one-time sales, a slowdown in growth does not slow earnings, it sends them crashing.[LYNCH]“Could any idiot run it?” A simple, boring, understandable business is a plus; sooner or later an idiot really will be running it. A motel chain instead of fiber optics, pantyhose instead of communications satellites.[LYNCH]Technology producer or technology user? Instead of a computer company in a price war, the company that benefits from cheaper computers (ADP); instead of the scanner maker, the supermarket that installs the scanners.[BUFFETT]Circle of competence: do I understand this business well enough to judge what it will look like ten years from now? If I don’t, no matter how attractive the business is, it is outside the thesis (concept card: 3.1).- Key customer and supplier dependencies (carried over to Section 10).
Questions. Can I describe the company in one sentence? What percentage of profit comes from a single product? Is the product a habit or a fad? Will the business most likely be the same ten years from now?
Section 3 · Category
Purpose. [LYNCH] To place the stock in one of the six categories, because the category determines “what I should expect from this stock.” You don’t expect a fourfold gain from Coca-Cola in two years; you don’t expect a dividend from a turnaround. The category also selects the valuation method (Section 6) and the sell signals (Section 12).
What to write.
- Category and rationale: growth rate (earnings), company size, position in the industry.
- Whether the company was previously in another category (Dow Chemical went from fast grower to a cyclical-tinged plodder; Chrysler from cyclical to turnaround, then back to cyclical).
- The return range expected from this category: 30–50 percent in a stalwart, then rotate; 10- to 40-bagger potential in a fast grower, but bankruptcy risk; timing in a cyclical; patience and a raider in an asset play.
- The category’s checklist (copied from Sections 4.3 and 4.12).
Questions. Does size limit growth? (It was mathematically impossible for GE to triple in the foreseeable future: it represented one percent of U.S. GNP.) What happens if I pick the wrong category: mistaking Ford for a stalwart like Bristol-Myers means being caught unprepared for an 80 percent decline.
Section 4 · My edge
Purpose. [LYNCH] To write honestly what I know about this stock better than the ordinary investor or Wall Street. The person with an edge can always beat the person without one; if there is no edge, the thesis is nothing but a gamble.
What to write.
- Type of edge: professional (I work in the industry, I am a supplier, I am a customer), local (I observe the store, the product, the crowds), analytical (a report everyone can see but nobody reads), temporal (the Wall Street lag: institutions and analysts have not discovered it yet).
- Evidence of the edge: what I saw, what I counted, whom I talked to.
[LYNCH]Separate the source of the idea from the tipster: “Uncle Harry is buying, and he’s rich” is not an edge. If an expert is speaking about his own field, that is valuable; if he is speaking about another field, it is noise.[LYNCH]Discovery is not a buy signal: the fact that Dunkin’ Donuts is always crowded does not mean I should buy the stock; it is only a clue leading to a story that must be developed.
Section 5 · Economic moat and niche
Purpose. To show what protects the company’s profit from competitors. [BUFFETT] Buffett’s “moat” and [LYNCH] Lynch’s “niche” are the same question: why can’t competitors take this profit? (Concept card: 3.2.)
What to write.
- Type of moat and evidence: brand and pricing power, cost advantage, switching costs, network effects, license/scale/local monopoly (the Boston Globe’s 90 percent of Boston print advertising revenue; Philip Morris’s brands).
[LYNCH]The price-increase test: “If you find a business that can raise its prices year after year without losing customers, you have a terrific investment.” Price increases and volume over the past five years are examined together.[BUFFETT]Pricing power is also Buffett’s single test of a business: if you have to hold a prayer session before raising the price by 10 percent, it is a bad business.[LYNCH]The source of the niche can be dullness: the funeral business, bottle caps, coupon clearing, waste oil collection. Wharton graduates and investment bankers don’t want to enter these businesses; the competition never comes.- Direction of the moat: is it widening or narrowing? The trend of the profit margin relative to competitors (Section 8) is the evidence for this.
Section 6 · Industry and competition
Purpose. To assess the industry’s growth rate, new entrants, and whether the industry is “in fashion.” [LYNCH] Lynch’s order of preference is surprising: an industry with no growth at all, then a low-growth industry, and a high-growth industry last.
What to write.
- The industry growth rate and the company’s growth relative to the industry (is it taking market share: beer grows slowly, Anheuser-Busch grew fast; hotels grow 2 percent a year, Marriott 20 percent).
[LYNCH]Hot-industry warning: for every hot product there are a thousand MIT graduates trying to figure out how to make it cheaper in Taiwan. Carpets in 1950, electronics in 1960, computers in 1980: the industry’s growth did not guarantee the companies’ prosperity; two hundred new competitors arrived and nobody made another dime.[LYNCH]A negative-growth industry does not attract a herd of competitors (over fifteen years Philip Morris went from 14 to 90 while Xerox fell from 160 to 60).- New entrants: how many companies have entered the market in the past three years, is capacity increasing? In cyclicals, a new entrant has to cut prices; everyone’s profits fall.
[LYNCH]The “next X” warning: if the company is being touted as “the next IBM/McDonald’s/Intel,” it is a bad sign for both the imitator and the original.- Percentage of institutional ownership and the number of analysts covering the company: the lower both are, the better (it means the Wall Street lag is still in effect).
Section 7 · Management and capital allocation
Purpose. To judge management not by what it says but by what it does with the money. [LYNCH] “Base your purchases on the company’s prospects, not on the president’s résumé or his speaking ability.”
What does capital allocation mean? Every year the company earns a profit; management puts that money into one of five places: (1) reinvestment in the business (new plants, stores, R&D), (2) acquiring other companies, (3) paying down debt, (4) dividends, (5) share buybacks. Which one it chooses determines what the shareholder earns over the long run more than the profit itself does. This section examines which of these five doors management walks through, and whether the door it chooses is the right one for the shareholder.
What to write.
[LYNCH]Insider buying: are the executives buying shares with their own money? Seven vice presidents from the lower ranks buying a thousand shares each is more meaningful than the president buying five thousand. Insider selling means nothing on its own (tuition, diversification); but if nine executives are selling most of their holdings while the stock has gone from 3 to 12, pay attention. A company whose insiders are buying will not go bankrupt within six months.[LYNCH]Buyback or dilution? A buyback is the simplest and best way to reward the shareholder (“the purest of synergies”). The opposite is dilution: Navistar printed millions of shares and recovered, but the stock did not. The five-year trend of shares outstanding is entered in the table (concept card: 3.13).[LYNCH]Acquisitions and diworseification: while Gillette scattered its razor-blade profits across cosmetics, lighters, and digital watches, the stock stayed at 35 instead of 100. Companies flush with cash overpay for acquisitions, expect too much, and manage them badly. A related business (synergy) is preferred; but synergy sometimes materializes and sometimes doesn’t.[BUFFETT]The one-dollar test: does every $1 retained create at least $1 of market value over time? If not, the money should have gone back to the shareholder as a dividend or a buyback (concept card: 3.12).[BUFFETT]Candor and openness: does management write the bad news plainly in the annual report, or does it only talk about the good quarters? Does it resist the “institutional imperative” (doing something because a competitor does it)?- Ownership: is founder/family/employee ownership high (usually a good sign, according to Lynch)? Is executive compensation tied to earnings, to the share price, or to size?
Section 8 · Financials
Purpose. To show with numbers that earnings are growing, the balance sheet holds, and the cash is real. Lynch’s rule: the cheaper the paper, the more valuable the information; skip the glossy pages and go to the balance sheet.
Three statements, three questions. An annual report contains three main financial statements, and each answers a separate question:
| Statement | Question | Items used in the thesis |
|---|---|---|
| Income statement (profit and loss) | How much did it sell this year, how much profit did it make? | Sales, gross profit, operating profit, pretax profit, net profit, earnings per share (EPS) |
| Balance sheet | On a given day, what does it own, whom does it owe? | Cash and marketable securities, inventories, receivables, property, plant and equipment, short-/long-term debt, shareholders’ equity |
| Cash flow statement | What actually came into and went out of the till? | Cash from operations, capital expenditure (capex), depreciation, dividend and buyback payments, borrowing |
The income statement says “profit,” but profit is an accounting opinion; the cash flow statement shows the real money in the till. A prolonged divergence between the two (profit but no cash) is one of the most important warning signs in the thesis.
What to write (each with a table covering at least five years; the concept card for each term is in Section 3).
[LYNCH]Earnings growth. “The only growth rate that really counts is earnings.” Don’t confuse growth with expansion. Are earnings erratic or steady (Shoney’s: 116 consecutive quarters)? The earnings line and the price line are plotted together: when the price runs far ahead of earnings (Avon 1972, The Limited 1983 and 1987), the result is disaster. (Card 3.5)[LYNCH]Pretax profit margin. Compared with competitors in the same industry; the highest margin = the lowest-cost operator = the one that survives on a bad day. (Card 3.6)[LYNCH]Cash position and debt. Net cash = cash + marketable securities − long-term debt; net cash per share is deducted from the price. A company with no debt cannot go bankrupt. (Cards 3.7, 3.8)[LYNCH]Free cash flow. What is left after normal capital expenditure; a ratio to price of 10 percent is standard, 20 percent is terrific. (Card 3.9)[BUFFETT]Owner earnings. Buffett’s version of free cash flow; the input to the DCF. (Card 3.10)[BUFFETT]Return on equity (ROE) and return on invested capital (ROIC). How efficiently the company uses the money in its hands. (Card 3.11)[LYNCH]Book value and hidden assets. Stated book value is often unrelated to real value, in both directions. (Card 3.14)[LYNCH]Inventories. A red flag if they are growing faster than sales. (Card 3.15)[LYNCH]Dividend and payout ratio. Track record and cushion in slow growers. (Card 3.16)[LYNCH]Dependence on a single customer. Dangerous if 25–50 percent of sales go to a single customer.
Questions. How many times have earnings fallen in the past ten years, and why? Where is the margin relative to competitors, and which way is it heading? How many lira/dollars of net cash per share? Does free cash flow line up with accounting profit? Is the share count falling?
Section 9 · Valuation
Purpose. To answer the question “What is this company worth, and is today’s price cheap relative to that?” with at least two methods. The details of the methods are in Section 6; here, what to write is listed.
What to write.
[LYNCH]P/E context. The company’s own historical P/E range, the P/E of industry peers, the overall market P/E. Compare the P/E with the earnings growth rate. (Card 3.3)[LYNCH]Fair P/E ≈ growth rate; the formula with dividends: (growth + yield) / P/E. (Card 3.4)[LYNCH]Cash and subsidiary adjustment. Net cash per share and separately valuable subsidiaries are deducted from the price; the P/E of the remaining “core business” is calculated (the Ford example, Section 6.7).[LYNCH]Avoid an excessively high P/E. It is like extra weight in the saddle (Avon, Polaroid, EDS: a P/E of 50).[BUFFETT]Intrinsic value and margin of safety. Intrinsic value = the cash the business will generate over its lifetime, discounted to the present; buy at a clear discount to it. (Cards 3.17, 3.18)[FORMULA]Scenario table: EPS, P/E, target price, and probability for bear / base / bull; probability-weighted expected price and annual return; loss/gain asymmetry.- Maximum purchase price: the highest price that delivers the targeted annual return (e.g. 15 percent) in the base scenario.
Section 10 · Risks and the thesis’s falsification conditions
Purpose. The most honest section of the thesis: what I don’t know, and what would mean I was wrong. Lynch: things are never clear on Wall Street; when they are clear, it’s too late to profit from them. Decisions are made with incomplete information, but what is missing gets written down.
What to write.
- Falsification conditions: “If one of these three things happens, the thesis is wrong” — measurable, dated, observable. Example: “If same-store sales fall two quarters in a row,” “If net debt/EBITDA exceeds 3,” “If the key customer contract is not renewed.”
- Business risks: single customer/supplier, regulation, technological change, the product being a fad, currency, raw materials.
- Financial risks: debt maturities, likelihood of dilution, sustainability of the dividend, hidden liabilities (Bhopal, Johns-Manville-type unquantifiable damages: stay away).
- Category risks: fast grower → insufficient financing and the Gulliver dilemma; cyclical → buying in the wrong phase (50%+ loss, years of waiting); turnaround → the list of failed turnarounds gets erased from memory; asset play → management eroding the assets with debt; stalwart → diworseification; slow grower → dividend cut.
- Valuation risk: P/E compression: if growth falls from 25 percent to 15 percent, the multiple falls too, a “double whammy” for the loyal shareholder.
- My own risk: do I understand this stock, or have I been taken in by the sizzle? Would a loss affect my daily life?
Section 11 · Catalysts and timing
Purpose. For earnings (and therefore the price) to advance, “something dynamic has to happen.” To write what that something is and roughly when.
What to write.
[LYNCH]Five ways to increase earnings: reduce costs; raise prices; expand into new markets; sell more in old markets; revitalize, close, or sell a losing operation. Which of these is the company doing, and when?- Concrete catalysts: number of new plants/stores, product launch, spin-off, division sale, share buyback program, capacity leaving the industry, regulatory change.
- In an asset play, the catalyst comes from outside: a corporate raider, a takeover, a bidding war. Waiting may be required; patience is part of the thesis.
[LYNCH]Timing: don’t try to time the market; but two periods carry a high probability of bargains: year-end tax selling and market crashes (“corrections push outstanding companies down to bargain prices”).- Big results take years, not months; the thesis’s horizon is written down (2–5 years).
Section 12 · Position and trading plan
Purpose. To answer the questions of how much money, at what price, when do I add, and when do I sell, before the price moves.
What to write.
- Position size.
[LYNCH]Number of stocks to own = the number of opportunities where I have an edge and that have passed all the tests of the research; “dumb diversification is the curse of the small investor.”[BUFFETT]Concentrate on the best ideas; the twenty-hole punch card. Limits by category: in fast growers and turnarounds, a single position is smaller (bankruptcy risk); in stalwarts it can be held larger. - Buying plan. Maximum purchase price; staged buying (one-third + adding on declines); can I really carry out the sentence “when I’m down 25 percent, I’m a buyer, not a seller”?
- Condition for adding. If the stock has fallen but the fundamentals are positive, holding is good and buying more is better (The Limited 1984). Adding is done only if the story is intact; “it fell, it got cheaper” is not a reason on its own.
- Sell triggers. The category’s Lynch sell signals (Section 4.3) and the thesis’s falsification conditions (Section 10) are copied here verbatim. Reaching the price target is not a reason to sell on its own; if the story is still good, the target is updated. Rotating out of a stalwart after a 30–50 percent gain is the exception.
[BUFFETT]Three situations in which to sell: (1) the thesis turned out to be wrong, (2) the business has deteriorated permanently, (3) there is a clearly better opportunity. “The price went up” is not one of them.[LYNCH]Don’t sell because insiders have started selling; don’t sell because an outsider (Petrie) has stopped buying either.
Section 13 · Monitoring plan
Purpose. To list in advance what to look at after the purchase. Lynch: “Keeping up with a company whose stock you own is like playing an endless hand of stud poker”: every quarter a new card is turned over.
- Every three months: the latest quarterly report; are earnings as expected; earnings line/price line; inventories; debt; insider buying/selling; institutional ownership and analyst count; store/product observation (“is there an air of prosperity?”).
[LYNCH]Growth phase: start-up (risky), rapid expansion (the safest and most profitable: the formula is being copied), maturity/saturation (troubled: no room left to expand). Is the company moving from one phase to the next? If Holiday Inn has reached the foot of Gibraltar, “where else can it expand, Mars?”- Retell the story: is the two-minute monologue still the same? If it has changed, the thesis is updated and the category is questioned again.
- At least one hour of investment research per week; collecting dividends and tallying profit and loss don’t count.
Section 14 · Decision journal
Date · Price · Action (buy/add/trim/sell/hold) · Why (one sentence) · Which section of the thesis changed · Next check date. The journal is kept as a table at the very end of the thesis; no row is ever deleted. The goal is to answer the question “why did I do that?” a year later from the record, not from memory. Lynch’s warning: “the stock went up, so I was right / it went down, so I was wrong” cannot be written in the journal as a reason.
3. Concept cards
This section explains every technical term used in the thesis in the same layout: What it is · What it shows · How to calculate (and where to find the number) · How to read it · Example · Trap · Where it is used in the thesis. If you run into a term you do not understand while writing a section, come back here.
Source of the numbers: the company’s annual report (the 10-K in the US; in Turkey, the financial statements and annual report filed on KAP). The income statement, balance sheet and cash flow statement are all inside this report. Finance sites (Yahoo Finance, Macrotrends, Koyfin, KAP, Fintables) serve the same line items ready-made, but their definitions may differ; state in the thesis which definition you used.
3.1 Circle of competence [BUFFETT]
What it is. The boundary of the businesses you genuinely understand. According to Buffett, what matters is not the size of the circle but knowing where its edge lies.
What it shows. Whether you can forecast what a company will look like ten years from now with reasonable confidence. If you cannot forecast it, you cannot value it no matter how cheap it is; and what you cannot value, you cannot buy.
How to test it. Can you answer these questions without writing them down: Where does the company make its money? Why don’t its customers go to a competitor? Ten years from now, who will buy this product, and why? What were the last two major changes in the industry, and how was the company affected? If the answers are hesitant, you are outside the circle.
Example. For many years Buffett stayed out of technology stocks; he said he did not understand them. He bought Apple only after he came to understand it as a consumer-products company. Lynch’s version: the intuition he gained by eating donuts or buying tires was something he could never gain in laser beams.
Trap. A stock outside the circle looks as if it is inside because “everyone is buying it.” The only way to widen the circle is reading and time, not price action.
In the thesis. Section 2 (business model). The thesis should open with the sentence “this business is inside my circle of competence because …”; if it is not, write explicitly “it is outside, but I am entering with a limited position for the following reason.”
3.2 Economic moat [BUFFETT] · Niche [LYNCH]
What it is. Like the water-filled ditch around a medieval castle, a structural advantage that protects the company’s profits from competitors’ attacks. Without a moat, high profits attract new rivals and the profit drifts down to the average over time.
What it shows. Whether the company can keep today’s high margin and return on capital in the future. The moat is the basis for the “growth will continue for years” assumption used in valuation; without a moat, long-term DCF assumptions are invalid.
Types and evidence.
| Type of moat | What it means | Where to look for evidence |
|---|---|---|
| Brand / pricing power | Customers pay this company more for the same product | Price increases can be made without losing volume; gross margin above competitors |
| Cost advantage | Produces the same product cheaper than everyone else | Pretax margin is the highest in the industry; scale, location, process |
| Switching costs | Moving to another supplier is expensive or troublesome for the customer | Customer churn is low, contract terms are long, integration is deep |
| Network effect | Every new user makes the product more valuable for the others | Margin rises as market share grows; the second player is far behind |
| License / regulation / local monopoly | Law or geography blocks competitor entry | Number of permits is limited; the only regional newspaper, the only gravel pit |
How to read it. Buffett’s test: “If they gave me billions of dollars, could I destroy this company?” If the answer is “yes, easily,” there is no moat. Lynch’s test: “Can it raise prices year after year without losing customers?” The direction of the moat also matters: is it widening (margin rising, share growing) or narrowing (new technology, new entrants)?
Example. The Boston Globe: 90 percent of print advertising revenue in Boston (local monopoly). Philip Morris’s brands (pricing power; from 14 to 90 in fifteen years). See’s Candies: Buffett raised prices every year and customers did not leave.
Trap. “Excellent management” is not a moat; management changes. “A great product” is not a moat either; products get copied. A moat is what protects the profit independently of the product and the management.
In the thesis. Section 5. Name the type of moat, put the evidence in the table, and state its direction.
3.3 Price/earnings ratio (P/E) [LYNCH]
What it is. The share price divided by annual earnings per share. “How many times this company’s one-year earnings am I paying?”
What it shows. How much the market is willing to pay for the company’s future earnings growth. Lynch’s reading: the P/E is the number of years it would take to earn back the initial investment if the company’s earnings stayed flat. P/E 10 = ten years; P/E 40 = forty years (“by then Cher may be a great-grandmother”).
How to calculate.
P/E = Share price / Earnings per share (EPS) EPS = Net income / Shares outstanding
Earnings for which period? There are three usages: the last 12 months (trailing), the estimate for the current fiscal year (forward), and the previous fiscal year. State in the thesis which one you used; when comparing, use the same definition throughout. One-off items (the gain on selling a division, a lawsuit settlement) are stripped out of earnings; otherwise the P/E looks abnormally low or high.
How to read it. A P/E on its own is meaningless; it is compared against three things:
- The company’s own history: where is it within the P/E band of the last 5–10 years?
- Peers in the same industry: what does “at a discount to the sector” mean, how many points?
- The earnings growth rate: fair P/E ≈ growth rate (card 3.4).
Lynch’s 1988 ranges (the logic still holds today, calibrate the numbers): utilities 7–9, stalwarts 10–14, fast growers 14–20. Lowest for slow growers, highest for fast growers; cyclicals swing back and forth in between. Comparing P/Es across categories is apples and oranges: a P/E that is a bargain for Dow Chemical is not a bargain for Wal-Mart.
The market’s P/E. The aggregate P/E of the index shows whether the market as a whole is expensive or cheap. Lynch’s examples: 20 in 1971 (madness, followed by the 1973–74 crash), 8 in 1982, 16 in 1987 (twice the price of five years earlier for the same earnings; it should have been a warning). When interest rates are low, P/Es rise because bonds become less attractive; but optimism carries P/Es to ridiculous levels independently of interest rates too.
Example. K mart: price 35, EPS 3.50 → P/E 10. An investor buying 100 shares for $3,500 owns $350 of earnings a year; if earnings stay flat, the investment is paid back in ten years.
Trap. (1) In cyclicals the P/E looks low when earnings are at their peak, and the stock is mistaken for cheap; in fact it is time to sell. (2) If earnings are close to zero the P/E goes to infinity and becomes meaningless. (3) A high P/E does not mean “bad,” it means “high expectations”; if there is growth to justify the expectations there is no problem, if not it is “extra weight in the saddle.”
In the thesis. Section 9. The three comparisons (history, sector, growth) go into the table.
3.4 PEG and the Lynch score [LYNCH]
What it is. Two simple ratios that relate the P/E to the earnings growth rate. Lynch’s rule: in a fairly priced company, the P/E equals the earnings growth rate (in percent).
What it shows. Whether the multiple you pay is covered by growth. Paying a P/E of 6 for a company growing 12 percent is attractive; paying a P/E of 12 for one growing 6 percent is not.
How to calculate.
PEG = P/E / earnings growth rate (in percent, e.g. 15) Lynch score = (earnings growth rate % + dividend yield %) / P/E
For the growth rate, use the average of the realized earnings growth of the last 3–5 years and a reasonable estimate for the next 3–5 years. Dividend yield = annual dividend per share / price.
How to read it.
| PEG | Lynch score | Meaning |
|---|---|---|
| 0.5 | 2 and above | Very positive; the level you are looking for |
| 1.0 | ~1.5 | Reasonable / acceptable |
| 2.0 | Below 1 | Very negative; growth does not cover the price |
Example. A company growing 15 percent, paying a 3 percent dividend, P/E 6: score (15+3)/6 = 3, terrific. One growing 12 percent, paying 3 percent, P/E 10: (12+3)/10 = 1.5, acceptable.
Trap. If growth is above 30 percent or below 5 percent the formula breaks down. In cyclicals and turnarounds the growth rate is meaningless; the formula is not used. When interest rates are high, the same PEG should count as more expensive.
In the thesis. Section 9, the quick screen, and the anchor for the exit multiple in Section 6.2.
3.5 Earnings per share (EPS) and earnings growth [LYNCH]
What it is. Net income divided by the number of shares outstanding; the money one share “earned” that year. Earnings growth is the year-over-year change in this figure.
What it shows. According to Lynch, the one thing that determines the share price in the long run: “The only growth rate that really counts: earnings.” The price can decouple from earnings in the short run, but sooner or later it returns to the earnings line.
How to calculate. Given ready-made at the bottom of the income statement (prefer “diluted EPS”: options and convertible bonds are taken into account). Growth:
Annual growth = EPS_this year / EPS_last year − 1 Compound growth (CAGR, n years) = (EPS_last / EPS_first)^(1/n) − 1
How to read it. Look at two things: speed and consistency. Speed determines the category (2–4 percent slow grower, 10–12 stalwart, 20–25 fast grower). Consistency determines reliability: Shoney’s raised earnings 116 quarters in a row; erratic earnings are the mark of a cyclical or a troubled company. Lynch is cautious about growth above 25 percent: “the 50-percenters are usually found in hot industries.”
Earnings line / price line. On the same chart, multiply EPS by a multiple (e.g. 15) and plot it alongside the price. When the price drops clearly below the earnings line it is a buying zone; when it rises clearly above, a selling zone (Avon 1972: the price decoupled from earnings, then collapsed; The Limited in 1983 and 1987, the same).
Growth ≠ expansion. Opening new stores or increasing sales is not growth; if earnings are not rising, expansion gives the shareholder nothing. The five ways to increase earnings (Section 4.7) tell you the source of the growth; the ability to raise prices is the most valuable source.
Trap. A share buyback raises EPS even without earnings growing (the share count shrinks); this is the “magic” effect, but check the trajectory of total earnings separately as well. Conversely, dilution lowers EPS while total earnings rise (Navistar).
In the thesis. Section 8 (the five-year table), Section 6.2 (the starting point of the projection).
3.6 Pretax profit margin (Pretax margin) [LYNCH]
What it is. How many cents of every dollar of sales are left after all costs (raw materials, wages, depreciation, interest) are deducted and before taxes are paid.
What it shows. How good the company’s cost structure is relative to its competitors. The company with the highest margin in an industry is by definition the lowest-cost operator; when business conditions deteriorate, it has the best chance of survival.
How to calculate.
Pretax margin = Pretax profit / Sales
Both are on the income statement. Operating margin (before interest) and net margin (after tax) give similar information; Lynch prefers pretax because it includes the interest burden but is not affected by differences in tax rates.
How to read it. Comparison across industries is meaningless: supermarkets (Albertson’s) 3.6 percent, pharmaceuticals (Merck) 25 percent and above; the average company around 5 percent (students guess 20–40 percent). Compare with competitors in the same industry and look at the five-year direction. Two uses: in a stock to be held for the long term, a relatively high margin (evidence of the moat); in a turnaround story, a relatively low margin (large room for improvement; if the margin goes from 2 to 5 percent, earnings rise two and a half times).
Example. Ford 1987: 71.6 billion in sales, 7.38 billion pretax profit → 10.3 percent.
Trap. A jump in the margin in a single year usually comes from a one-off item (an asset sale); use the five-year average and trend. In cyclicals the margin swings with the cycle; do not mistake the peak margin for permanent.
In the thesis. Section 8, the comparison table with competitors; in Section 5 as evidence of the moat.
3.7 Net cash and cash position [LYNCH]
What it is. The cash in the company’s till plus marketable securities that can easily be turned into cash, after long-term debt is deducted.
What it shows. Two things. (1) Safety: if cash exceeds debt the company cannot go bankrupt; if cash is growing relative to debt the balance sheet is improving, and the reverse means it is deteriorating. (2) Hidden value: part of the share price is already the money in the till; once that money is deducted, you see what you are actually paying for the business itself.
How to calculate. On the balance sheet, cash and cash equivalents plus marketable securities under “current assets”; long-term debt under “long-term liabilities.”
Net cash = Cash + Marketable securities − Long-term debt Net cash per share = Net cash / Shares outstanding Price of the core business = Share price − Net cash per share Core P/E = Price of the core business / Core EPS
Lynch does not count short-term debt; he assumes the other current assets (inventory, receivables) cover it. If you want to be more conservative, deduct total debt (see 3.20, enterprise value).
How to read it. Net cash per share is a floor for the share price: Ford is unlikely to fall much below $16, because there is $16 of cash per share. If net cash is near zero or negative (debt exceeds cash), it is the debt card (3.8), not this one, that takes the lead.
Example (Ford 1987). Cash 5.672 + marketable securities 4.424 = 10.1 billion; long-term debt 1.75 billion → net cash 8.35 billion → $16.30 per share. The stock is at $38; the finance subsidiary is worth $16.60 per share (1.66 EPS × P/E of 10). The automobile business is being bought for 38 − 16.30 − 16.60 = $5.10 and is expected to earn $7: core P/E of 0.7.
Trap. Cash does not always make a difference: at Bristol-Myers, $5 of net cash per share is insignificant on a $40 stock; if the stock falls to 15 it becomes a big deal. Cash held abroad may be subject to tax; restricted cash (collateral) is not free.
In the thesis. Section 8 (table), Section 9 (price adjustment), Section 10 (the floor of the bear case).
3.8 Debt, debt/equity and types of debt [LYNCH] [BUFFETT]
What it is. The company’s interest-bearing obligations to others; equity is the part that belongs to shareholders (assets − liabilities). The debt/equity ratio shows whose money the company is operating with.
What it shows. Whether the company can survive a bad year. Lynch: “A company with no debt can’t go bankrupt.” Buffett: leverage magnifies returns in good times and finishes off the company in bad times; “if smart people go broke, the reason is usually leverage.”
How to calculate.
Debt/equity = Total interest-bearing debt / Shareholders' equity (balance sheet) Net debt = Total debt − Cash Net debt / EBITDA = how many years of operating cash it takes to pay off the debt (see 3.20) Interest coverage = Operating profit / Interest expense (income statement)
How to read it. It depends on the industry (utilities and real estate are naturally indebted); but the general rule: debt/equity below 0.5 is comfortable, above 1 calls for attention, net debt/EBITDA above 3 is heavy. If interest coverage drops below 3, earnings are going to interest. Lynch’s distinction: bank debt is the most dangerous, because it can be called on demand; funded debt (bonds) is safe until maturity, and as long as the interest is paid the creditor waits. Among troubled companies, choose the one with the superior financial position; avoid those loaded with bank debt.
Survival test in turnarounds. Cash / annual cash burn = how many years it can last. During its crisis period, Apple, with $200 million in cash and zero debt, made it clear it was not going bankrupt.
Trap. Lease obligations and pension deficits may not appear as debt on the balance sheet; look at the footnotes. Debt may look low while a large loan is due within a year; write down the maturity schedule.
In the thesis. Sections 8 and 10.
3.9 Free cash flow (FCF) [LYNCH]
What it is. The money left over from the cash a company generates from operations after the investment spending (capital expenditure, capex) it makes to sustain and grow the business. “The cash you take in and don’t have to spend.”
What it shows. Whether the earnings are real. Accounting profit rests on estimates (depreciation schedules, provisions for receivables); cash is in the till. If there is profit for a long time but no cash, either the profit is inflated or the business keeps swallowing investment (Lynch’s Pig Iron Inc. example: it makes a profit, but the money goes into new furnaces; Philip Morris’s cash, by contrast, is free).
How to calculate. From the cash flow statement:
FCF = Cash from operations − Capital expenditure (capex) FCF yield = FCF / Market capitalization (or FCF per share / price)
How to read it. Lynch’s thresholds: an FCF yield of 10 percent is standard (it matches the minimum return expected from long-term stock ownership); 20 percent is terrific; a sustainable 50 percent means “mortgage the house and buy all the shares you can find.” If the ratio of FCF to net income (cash conversion) is close to 1, earnings are of high quality; if it is persistently below 0.5, the earnings are on paper.
Example. A $20 stock with $2 of FCF per share → 10 percent; $4 → 20 percent.
Trap. There are two kinds of capex: maintenance (keeping the existing plant running) and growth (a new plant). The two are not separated on the cash flow statement. The FCF of a company making growth investments looks temporarily low; this is not bad, but the distinction must be written down in the thesis (card 3.10 resolves this). A single year’s FCF is volatile; use a 3–5 year average.
In the thesis. Section 8; the input to the DCF in Section 6.4.
3.10 Owner earnings [BUFFETT]
What it is. Defined by Buffett in his 1986 shareholder letter, the answer to the question “if I owned this business, how much money would really go into my pocket each year?” It is a more careful version of free cash flow: it deducts only the investment required to keep the business going as it is today, not growth investment.
Why it exists, what it shows. Accounting profit misleads for two reasons. (1) Depreciation is a non-cash expense: the cost of a machine bought ten years ago is deducted from profit piece by piece every year, but that money was paid long ago. So profit shows real cash as lower than it is. (2) But machines really do wear out and need replacing; this maintenance capex is often larger than depreciation (inflation, technology). So depreciation has to be added back and replaced with the real replacement cost. Owner earnings do exactly that: they add the non-cash expense back on top of profit and take out the money that really has to be spent to keep the business standing. What remains is the money the owner can use freely: pay dividends, buy back shares, invest in growth.
How to calculate.
Owner earnings = Net income
+ Depreciation and amortization (cash flow statement, "cash from operations" section)
+ Other non-cash charges (e.g. part of stock-based compensation, debatable)
− Maintenance capex (investment required to keep the business at today's volume)
± Change in working capital (in a growing business inventory and receivables grow too; this swallows cash)
Maintenance capex is not reported separately; ways to estimate it: (a) if the company discloses a “maintenance/sustaining investment” figure, use it; (b) if not, take the lower of the last 5 years’ average depreciation and total capex; (c) in a company that is not growing, assume total capex ≈ maintenance capex.
How to read it.
- Owner earnings / Net income. If close to 1, earnings are turning into cash. If persistently above 1 (depreciation large, investment small), the business is “light”: the kind Buffett likes. If persistently well below 1, the business is swallowing capital: reported earnings are not going into the shareholder’s pocket.
- Owner earnings yield = Owner earnings / Market capitalization. Compare with the bond yield; it should be clearly above the bond, because the stock is risky and owner earnings can grow.
- Growth test. If the company says “earnings are growing” but owner earnings are not growing, the growth keeps demanding new capital; this is the kind of business Buffett avoids.
Example. A company with net income 100, depreciation 40, total capex 70 (of which 45 is maintenance and 25 is for a new market), and a working capital increase of 10. Owner earnings = 100 + 40 − 45 − 10 = 85. Free cash flow would have been 100 + 40 − 70 − 10 = 60. The difference (25) is the money the company chose to spend on growth; owner earnings show “what would have gone into the pocket even without growing.” Buffett’s example: Berkshire’s businesses such as insurance and See’s require less investment than depreciation, so owner earnings are higher than reported earnings; in heavy-industry businesses it is the reverse.
Trap. Maintenance capex is an estimate; an optimistic estimate inflates owner earnings. Always calculate two values (with the assumption depreciation = maintenance, and with the total capex assumption) and show the band between them in the thesis. Adding depreciation back and putting nothing in its place (using EBITDA) is the mistake Buffett criticizes most harshly: “The tooth fairy doesn’t pay for capex.”
In the thesis. Section 8 (cash quality), Section 6.4 (DCF input). If owner earnings are used instead of FCF in the DCF, a more conservative intrinsic value results.
3.11 Return on equity (ROE) and return on invested capital (ROIC) [BUFFETT]
What it is. ROE: how many cents of profit each dollar the shareholders have left in the company produced that year. ROIC: how many cents of operating profit each dollar of both shareholder and creditor money (total invested capital) produced.
What it shows. How efficiently the company uses money; in other words, the quality of growth. Buffett’s preference: not growth in earnings per share, but steady, high ROE achieved with little or no debt. Because earnings grow simply because retained earnings accumulate; the skill lies in making the accumulated money work at a high return too. A business’s long-term return converges toward the return on the capital it reinvests: a business with an 8 percent ROIC will earn you around 8 percent over the long run no matter how cheaply you buy it; a business with a 25 percent ROIC makes up for it over time even if bought a little dear (this is the math behind the “wonderful company at a fair price” rule).
How to calculate.
ROE = Net income / Average shareholders' equity (income statement / balance sheet) ROIC = After-tax operating profit / (Debt + Equity − Excess cash) After-tax operating profit = Operating profit × (1 − tax rate)
How to read it. ROE above 15 percent is good, above 20 percent exceptional; but it can be inflated with debt (as equity shrinks, ROE grows). So read ROE together with debt/equity, or use ROIC; ROIC is unaffected by debt. Compare ROIC with the company’s cost of capital (roughly 8–10 percent): if it is above, growth creates value; if below, growth destroys value (a company that grows yet gets poorer). Ten years of consistency matter more than a single-year peak.
Example. Two companies, both growing 10 percent. A’s ROIC is 30 percent: it reinvests a third of its earnings for growth and can distribute the remaining two-thirds. B’s ROIC is 10 percent: it must reinvest all its earnings, nothing is left for the shareholder. Same earnings growth, entirely different value.
Trap. In companies doing large buybacks, equity shrinks a great deal and ROE looks meaninglessly high or negative; use ROIC. Goodwill (from acquisitions) inflates invested capital; calculate ROIC both with and without goodwill: the first shows management’s acquisition skill, the second the business itself.
In the thesis. Section 8 (five-year ROE/ROIC table), Section 7 (quality of capital allocation).
3.12 The one-dollar test [BUFFETT]
What it is. From Buffett’s 1983 letter: every dollar of earnings the company retains rather than distributes should, over time, create at least one dollar of market value.
What it shows. Whether management’s capital allocation (card 3.11 and Section 7) is delivering results; that is, whether retained earnings are creating value for the shareholder or being burned. If they are not creating value, the money should have been distributed as dividends or buybacks.
How to calculate.
Retained earnings (n years) = Σ (Net income − Dividends) (over n years) Increase in market value (n years) = Market value_end − Market value_start Ratio = Increase in market value / Retained earnings → should be above 1
n should be at least 5, preferably 10 years; in a single year, market swings distort the result. Pick the start and end prices from similar points in the market cycle, or normalize with an average P/E.
How to read it. If the ratio is clearly above 1 (e.g. 2: every retained dollar created two dollars of value), management is using the money well and is right to keep earnings inside. If it is below 1, the company is growing but the shareholder is getting poorer; the activists pushing for dividends/buybacks are right.
Example. Total earnings of 500 million over ten years, 100 million in dividends → 400 million retained. Market value rose from 1 billion to 2.2 billion → increase of 1.2 billion; ratio 3. Successful. The reverse: 400 million retained, market value from 1 billion to 1.1 billion → ratio 0.25; the money was burned (the typical result of diworseification).
Trap. The increase in market value may have come purely from P/E expansion (a P/E of 20 instead of 2); that is the market’s doing, not management’s. Check: did EPS and owner earnings grow at a similar rate?
In the thesis. Section 7.
3.13 Share count, buybacks and dilution [LYNCH]
What it is. The number of shares outstanding is “how many slices the company is cut into.” A buyback is the company purchasing its own shares from the market and retiring them; dilution is issuing and selling new shares, or giving them to employees.
What it shows. The number of slices of the same pie. If the share count falls, each share’s claim on earnings grows (the “magical effect on earnings per share”: if the company buys back half its shares, EPS doubles even though total earnings stay the same). If it rises, the opposite: Navistar recovered, but because of dilution the shareholders did not.
How to calculate. “Weighted average diluted shares outstanding” on the income statement; a five-year series.
Annual change in share count = Share count_this year / Share count_last year − 1 Buyback yield = Cash spent on buybacks / Market cap
How to read it. A 2–4 percent annual decline is a healthy buyback; a continual increase (options, capital raises) works against the shareholder. But the price of the buyback matters: if the company buys its own stock when it is expensive it destroys value, when it is cheap it creates value (Buffett: a buyback makes sense only when the stock is below intrinsic value). Lynch: on October 20, 1987, companies’ buyback announcements stopped the panic.
Trap. If buybacks are happening but the share count is not falling, the repurchased shares are going to employee options; the net effect is zero. Always look at the net share count, not the buyback announcement.
In the thesis. Section 7, Section 8 table.
3.14 Book value and hidden assets [LYNCH]
What it is. Book value (shareholders’ equity): the assets on the balance sheet minus liabilities; book value per share is this divided by the share count. The price-to-book (P/B) ratio is how many times that the market is paying.
What it shows. In theory, what would be left for shareholders if the company were liquidated. In practice it often bears no relation to real value, because assets are recorded at historical cost: land bought thirty years ago sits at its purchase price, and the tunnels of a bankrupt railroad stand as “assets.”
How to calculate.
Book value per share = Shareholders' equity / Share count (balance sheet) P/B = Price / Book value per share
How to read it. Lynch’s two warnings:
- It overstates. Penn Central had $60 of book value per share while going bankrupt (tunnels bored through mountains, useless railcars). Danger when the left side (assets) is bloated while the right side (liabilities) is real: 400 million in assets, 300 million in liabilities = 100 million book value; but if the assets fetch 200 million in liquidation, the real value is minus 100 million. “The company is worth less than nothing.”
- It understates. Natural resources, real estate, brands, patents, subsidiaries, tax-loss carryforwards look very low on the books: Handy and Harman’s book value of $7.83 exceeded $19 when its metal inventory was counted at current prices. These are hidden assets, and they are the source of asset plays.
Rule: if you are buying a stock for its book value, understand line by line what those values really are, and revalue them with your own estimate (current price, saleability, taxes).
Trap. “P/B below 1, so it’s cheap” is meaningless on its own; it is meaningful for banks and insurers, meaningless for brand and software companies (the brand is not on the books). If goodwill (the leftover difference from acquisitions) makes up most of book value, real assets are fewer.
In the thesis. Section 8; an input to Section 6.7 for asset plays.
3.15 Inventories [LYNCH]
What it is. The raw materials, work in progress and finished goods the company has not yet been able to sell; under current assets on the balance sheet.
What it shows. The balance between demand and production. If inventory is growing faster than sales, the product is not selling; discounts, write-downs and lower earnings are coming soon.
How to calculate.
Inventory growth vs. sales growth (annual; write the two side by side) Inventory turnover = Cost of goods sold / Average inventory Days of inventory = 365 / turnover
How to read it. If inventory growth clearly exceeds sales growth, it is a red flag (Lynch watches this at both manufacturers and retailers; in turnarounds, “inventories are growing twice as fast as sales” is a sell signal). In cyclicals: “When the parking lot fills up with ingots, it’s definitely time to sell the cyclical; in fact, you may be a little late.” The reverse is also a good sign: if inventory is melting away and days of inventory are falling, demand is strong.
Trap. A deliberate inventory build (a new product launch, buying before raw material prices rise) is not bad; look at management’s explanation. Seasonality is large in retailers; compare the same quarters.
In the thesis. Sections 8 and 13 (quarterly follow-up).
3.16 Dividends, dividend yield and payout ratio [LYNCH]
What it is. Dividend: the portion of the company’s earnings distributed to shareholders in cash. Yield: the annual dividend as a proportion of the share price. Payout ratio: what percentage of earnings is distributed.
What it shows. Two different things. In slow growers the dividend is the return itself; you would not buy them for any other reason. In other companies, dividend policy is management’s honest confession about growth opportunities: “Companies pay generous dividends when they can’t find new ways to use the money to expand the business.”
How to calculate.
Dividend yield = Annual dividend per share / Price Payout ratio = Total dividends / Net income (or dividend per share / EPS)
How to read it. Checklist for a slow grower: was the dividend paid every year, was it raised regularly, what happened in the last three recessions? If the payout ratio is low (30–50 percent), the company has a cushion for a bad year: it can earn less and still protect the dividend. Above 70 percent the dividend is fragile; it gets cut when earnings fall. A dividend record also limits how far the share price falls in a recession (shareholders keep collecting the check).
Trap. A very high yield (8 percent+) usually shows the market expects a dividend cut; sustainability matters, not the yield. Dividends are generally at a tax disadvantage relative to share buybacks; look at total shareholder return (dividends + buybacks).
In the thesis. Section 8; an input to Section 6.6 for slow growers.
3.17 Intrinsic value [BUFFETT]
What it is. The sum of the cash (owner earnings) a business will leave its owner over its remaining life, discounted to the present at an appropriate rate. According to Buffett, the only logical definition of “value”; ratios like P/E and P/B are shortcuts to it.
What it shows. What the company is worth, independent of price. Price changes every day; intrinsic value changes only when the company’s capacity to generate cash changes. “Price is what you pay; value is what you get.”
How to calculate. The DCF in Section 6.4. Its essence:
Intrinsic value = Σ [ Owner earnings_t / (1 + r)^t ] + Terminal value / (1 + r)^N
What “discounting” means: 110 dollars a year from now is equivalent to 100 dollars today at a 10 percent return; every future cash flow is shrunk according to how many years away it is and the required return (r), and brought to the present. Terminal value: the value of all the years after the forecast period, in a single formula.
How to read it. Buffett’s warning: intrinsic value is not a precise number but a range; two people will get different results from the same data. Therefore (1) work with rough assumptions, don’t fiddle with decimals; (2) build a sensitivity table (how does value change when r and growth change); (3) apply a margin of safety (card 3.18). If most of the value comes from terminal value (70 percent+), the thesis rests on “forever” and is fragile.
Example. The example in Section 6.4: intrinsic value per share ≈ 36; buy limit with a 30 percent margin ≈ 25.
Trap. DCF is highly sensitive to inputs; a small change in growth or the discount rate moves the value by 30 percent. That does not make the method worthless; it makes visible the question “which assumption is carrying the thesis.” Bad use: deciding on a target price first, then picking assumptions to make the DCF fit it.
In the thesis. Section 9.
3.18 Margin of safety [BUFFETT]
What it is. The gap between the intrinsic value estimate and the price paid. Benjamin Graham’s concept; according to Buffett, the three most important words in investing.
What it shows. Room for error. Since the intrinsic value estimate is rough, a buffer is needed for bad luck and unknowns. The margin ensures you don’t lose money even if the estimate turns out 30 percent wrong. If the bridge holds 10 tons, you drive across it with a 5-ton truck.
How to calculate.
Margin of safety = 1 − Price / Intrinsic value Buy limit = Intrinsic value × (1 − required margin)
How to read it. The required margin varies with the predictability of the business: in a steady stalwart 25 percent may suffice; in a cyclical, indebted or single-customer business, 50 percent. The margin also determines the return: a company bought at half its value returns 100 percent as price approaches value. If there is no margin (price ≈ value), the expected return is only as much as the growth in owner earnings.
Trap. A margin does not make a bad business good; in a business that keeps losing value, a “cheap” price becomes even cheaper tomorrow (a value trap). The margin comes first from business quality, then from price: “a wonderful company at a fair price.”
In the thesis. Sections 9 and 12 (buy limit).
3.19 Mr. Market [BUFFETT]
What it is. Graham’s parable: a partner with a manic-depressive temperament who comes to your door every day and names a price for the shares he holds. Some days he is euphoric (high price), some days despairing (low price). He does not force you; you can accept his offer or ignore it.
What it shows. That price is mood, not information. Mr. Market is there to serve you, not to guide you. You sell in his euphoria and buy in his despair; if you have no valuation of your own, you get caught up in his mood.
How to use it. The intrinsic value/buy limit in the thesis is compared with Mr. Market’s offer: if the offer is below the limit, buy; if above, wait; if the story has not changed, a price drop is an opportunity, not a reason. Lynch’s version of the same idea: “Market declines are great opportunities to buy stocks in companies you like.” “Be fearful when others are greedy, and greedy when others are fearful” is the same rule.
Trap. Mr. Market is sometimes right: if the price has fallen, first check the story (Section 10 decay conditions), then say “Mr. Market is wrong.” Taking every decline for an opportunity is as mistaken as taking every rise for confirmation.
In the thesis. Section 12 (adding condition), Section 14 (decision rationale).
3.20 Market cap, enterprise value and EBITDA [FORMULA]
What it is. Market capitalization (market cap): share price × share count; the price of the shareholders’ stake. Enterprise value (EV): the cost of buying the whole company together with its debt. EBITDA: earnings before interest, taxes, depreciation and amortization; roughly the raw cash the operations generate.
What it shows. Market cap ignores debt; two companies can have the same market cap while one is debt-free and the other heavily indebted. EV corrects for this. EBITDA is used to compare companies with different debt and depreciation structures; EV/EBITDA is the cousin of P/E that takes debt into account.
How to calculate.
Market cap = Price × Share count Enterprise value (EV) = Market cap + Total debt − Cash EBITDA = Operating profit + Depreciation and amortization EV / EBITDA (by sector, 6–8 reasonable, 12+ expensive; roughly) Net debt / EBITDA (debt load; above 3 is heavy)
Trap. EBITDA ignores depreciation; but machines really do wear out (card 3.10). In capital-intensive businesses EBITDA is very misleading; it is Buffett’s least favorite number. In the thesis, EBITDA is used only for debt load and sector comparison, not for valuation.
In the thesis. Section 0 (market cap), Section 8 (debt load).
3.21 Compound annual growth rate (CAGR) [FORMULA]
What it is. The constant annual rate at which a quantity (earnings, price, a portfolio) is deemed to have grown from start to finish.
What it shows. It makes returns over different periods comparable. Which is better, “100 percent in five years” or “60 percent in three years”? CAGR: 14.9 percent and 17.0 percent.
How to calculate.
CAGR = (Ending value / Starting value)^(1/n) − 1 n: number of years Rule of 72: the time for a quantity to double ≈ 72 / annual growth percentage
Example. From 60 to 120.7 in five years: (120.7/60)^(1/5) − 1 = 15 percent. Earnings growing at 15 percent double in roughly 72/15 ≈ 4.8 years.
In the thesis. Section 6 (target return), Section 8 (earnings growth).
4. Rules integrated from Lynch
Every item in this section is based on the highlights drawn from One Up on Wall Street. The numerical thresholds (P/E ranges, margins) date from 1988; carry the logic, not the absolute values, into the present and calibrate them against current sector data.
The essence of Lynch’s method, in three sentences. (1) The amateur’s edge is being able to see, in everyday life, the companies the professional cannot see or is not allowed to buy. (2) Every stock falls into one of six categories, and the category determines what to expect from that stock, what to look at, and when to sell. (3) In the long run, price follows earnings; so follow the story and the earnings, not the market.
4.1 Before you start: personal preparation
“There’s no point in studying the financial pages until you’ve looked in the mirror.” Three questions before buying any stock:
- Do I own a house? Invest in a house before you invest in a stock. (People spend months choosing a house and minutes choosing a stock.)
- Do I need the money? Invest only what you could lose without the loss affecting your daily life in any way in the foreseeable future.
- Do I have the personal qualities it takes to succeed? Patience, self-reliance, common sense, a tolerance for pain, open-mindedness, detachment, persistence, humility, flexibility, a willingness to do independent research, a willingness to admit mistakes, and the ability to ignore general panic. Being able to make decisions with incomplete or imperfect information. Being able to resist your gut instincts: the moment when people are most convinced that stocks will go up coincides with the moment when the opposite happens.
Investing in stocks is an art, not a science; history and philosophy are better preparation than statistics. The math you need is learned in the fourth grade. Don’t be like the Greeks who debated how many teeth a horse had instead of looking at the horse: study the company. The rooster’s crowing does not cause the sunrise: beware of cause-and-effect confusion in market commentary.
4.2 The edge and sources of ideas
- Stop listening to the professionals. The smart money is wrong 40 percent of the time; you can’t tell when it changes its mind and sells; your sources are better because you can keep track of them.
- The Wall Street lag. A stock isn’t considered “attractive” until several large institutions and respected analysts have endorsed it. The Limited: IPO in 1969, first institution in 1975, two institutions holding 0.6 percent in 1979, more than thirty analysts in 1985 and the stock far beyond its fundamentals. The amateur’s edge is this gap.
- “Checked by number four.” The fund manager looks for reasons not to buy the exciting stock: “too small,” “no track record,” “unionized,” “the competition will kill it.” Nobody ever lost his job losing money on IBM; lose it on La Quinta and it’s “what’s the matter with you?” You don’t have to invest like an institution; if you’re a surfer, a trucker, or a retiree, you have an edge.
- Institutional constraints create opportunity. Because of the 5/10 percent rules and market-cap thresholds, the big funds are limited to 90–100 of 10,000 companies; the small fast growers are screened out.
- Where ideas come from: near home, the shopping mall, the workplace. Pep Boys’ clerk, supplier, sign painter, and floor washer all saw the success before Wall Street did. The ordinary person comes across two or three candidates a year.
- Ask about the competitors. Ask a company executive about his most successful competitor; there is no more positive sign than grudging admiration (United Inns → La Quinta).
- Discovery ≠ purchase. The initial information is treated like an anonymous tip stuffed into your mailbox; you develop the story, then you decide.
- Separate the tip from the tipster. Even if the tipster is smart and rich and his last tip worked out. People in the paper industry hand out drug tips; health-care people talk about paper-company takeovers.
4.3 The six categories
Why categories? Lynch’s observation: most of investors’ mistakes come from expecting a stock to behave like a category it doesn’t belong to. Whoever takes Ford for a “blue chip” like Bristol-Myers is caught unprepared for an 80 percent drop; whoever expects Coca-Cola to quadruple in two years is disappointed. Once the size has been determined, the company is placed in one of six categories. “Big companies have small moves, small companies have big moves.” Companies don’t stay in the same category forever; the category is questioned anew at every review.
| Category | Definition | Expectation | What to check |
|---|---|---|---|
| Slow grower | Large, mature company growing at the rate of GNP (2–4 percent); the chart is as flat as a map of Delaware. A generous dividend is the typical sign. | Dividends. Limited capital gains. Lynch doesn’t much care for them; sells what he buys at a 30–50 percent gain or when the fundamentals deteriorate. | Has the dividend always been paid, has it been raised regularly; the payout ratio (a cushion if low); is there a new line of growth. |
| Stalwart | Giants growing 10–12 percent a year: Coca-Cola, Bristol-Myers, Kellogg, P&G. They don’t go bankrupt; good friends in a crisis. | A 30–50 percent gain, then rotate; no tenbagger expected. Protection in a recession. | Where the P/E stands relative to its own history and the sector; has it already shot up in recent months; diworseification; is the long-term growth rate holding; performance in previous recessions. |
| Fast grower | Small, aggressive new enterprise growing 20–25 percent a year; can be in a slow-growth industry (Anheuser-Busch, Marriott). | 10- to 40-baggers, even 200; one or two of them in a small portfolio make a career. Risk: insufficient financing, P/E contraction once it runs out of steam. | Is the product a significant part of the business; is growth in the 20–25 percent range (be wary of the 50 percenters); has the formula been duplicated in more than one city; is there still room to grow; P/E ≈ growth rate; is expansion accelerating; is institutional ownership low; is the balance sheet sound and are the earnings substantial. |
| Cyclical | Sales and profits rise and fall in a regular fashion: automobiles, airlines, tires, steel, chemicals, defense. The chart is a map of the Alps. | Rises faster than the stalwarts in a recovery; 50 percent+ loss and years of waiting in the wrong phase. The most misunderstood category: mistaken for blue chips. | The phase of the cycle; inventories and supply-demand; new entrants (danger); capacity additions; commodity spot/futures prices; union contracts; end demand; expect the P/E to shrink as earnings approach the peak. If you work in the industry, this is where your edge is most valuable. |
| Turnaround | Battered, non-growing company on the brink of bankruptcy: Chrysler, Penn Central, Lockheed, GPU. Types: “bail-us-out-or-else,” “who-would-have-thunk-it,” “little-problem-we-didn’t-anticipate,” “perfectly-good-company-inside-a-bankrupt-company.” | Lost ground is made up very quickly; the category least related to the general market. The failures drop off the list; the success rate is low, the gains are large. | Can it survive a raid by its creditors: cash, debt, debt structure, how many years it can lose money; what is left for the shareholder in bankruptcy; have unprofitable divisions been sold off; is business coming back; are costs being cut; is there dilution (Navistar). Stay away from unquantifiable liabilities (Bhopal). |
| Asset play | A company sitting on something valuable that Wall Street has overlooked: cash, real estate, timber, oil, metal reserves, licenses, tax losses, subsidiaries. Pebble Beach: the gravel pit alone was worth more than the whole company. | Where the local edge is most useful; the catalyst is a raider. Patience required; nowadays it gets noticed faster. | The real value of the assets; hidden assets; debt to be subtracted (creditors come first); is management eroding the assets with debt; are insiders buying; is a raider waiting in the wings. |
Sell signals by category
| Category | When to sell and signals |
|---|---|
| Slow grower | At a 30–50 percent gain or when the fundamentals deteriorate. Two years of market-share loss and a new advertising agency; R&D cut back, resting on past successes; two unrelated acquisitions and a search for “the forefront of technology”; the balance sheet went from no debt/millions in cash to no cash/millions in debt; even at a low price the dividend yield isn’t attracting interest. |
| Stalwart | The price has risen above the earnings line, or the P/E has strayed far outside its normal range → sell, buy it back on a dip or buy another stalwart. New products had mixed results; P/E of 15 while peers are at 11–12; no insider buying in the past year; the division contributing 25 percent of earnings is vulnerable to a slowdown; growth is slowing and the cost-cutting opportunities are exhausted. |
| Fast grower | The end of the second phase of rapid expansion: no new stores opening, the old ones falling apart, the kids don’t like the product. Forty analysts with “highest recommendation,” institutional ownership at 60 percent, three magazines praising the CEO. Same-store sales down 3 percent; new stores disappointing; two top executives left for a competitor; just back from a “roadshow”; P/E of 30 while the most optimistic growth estimate is 15–20 percent. The P/E has reached absurd levels (Avon, Polaroid at 50). Don’t sell early and lose a potential tenbagger. |
| Cyclical | Toward the end of the cycle; when things actually start to go wrong: costs rising, plants at full capacity, capacity investment under way; inventories building up; commodity prices falling, futures below spot; new competitors; two union contracts expiring; end demand slowing; a lavish new plant in the budget; still unable to compete with foreign producers. The leading edge sells a year in advance. |
| Turnaround | After the turnaround: the troubles are over, everybody knows it, the stock should be reclassified (Chrysler at $48 is now a cyclical). Debt that fell for five quarters rose in the latest one; inventories growing twice as fast as sales; P/E inflated relative to expectations; the strongest division sells half its output to a single slowing customer. |
| Asset play | Wait for the raider; a takeover, a bidding war, or a leveraged buyout doubles or triples the price. Sell: a 10 percent new share issue for diversification despite the discount; the division expected to sell for $20 million brought $12 million; a tax cut reduces the value of the loss carryforward; institutional ownership went from 25 percent to 60 percent. |
4.4 The 13 attributes of the perfect stock
The perfect company can’t be found; but if it can be imagined, its favorable attributes can be recognized. The common logic: each of these attributes keeps institutional investors and analysts away from the stock, so you can buy it cheap before Wall Street notices. In the thesis, every item is scored as “yes / no / partly.”
4.5 Stocks to avoid
- The hottest stock in the hottest industry. It has nothing behind it but hope and thin air; it doesn’t go down slowly, and it doesn’t stop where you got on. If you’re not skilled at selling (the fact that you bought it is a clue), the profit turns into a loss.
- “The next something.” On Broadway as on Wall Street, the next one almost never is; it’s also a bad sign for the original it’s being compared with.
- The company that diworseifies. Synergy sometimes happens (Gillette + Foamy), and usually doesn’t (Marriott + auto parts). If the core business is terrible, acquisitions are a good strategy (Berkshire, Loew’s); otherwise, be suspicious.
- The whisper stock. A hypnotic story, emotional appeal, all sizzle and no steak. Longshots almost never pay off. It’s better to miss the first move and see whether the company’s plan is working: wait for the earnings; tenbaggers can be found in proven companies too. “When in doubt, tune in later.”
- The company dependent on a middleman. 25–50 percent of sales to a single customer (SCI Systems → IBM; Tandon).
- The company with the exciting name. The more attractive the name, the bigger the crowd; the opposite of the dull name.
4.6 The numbers: “the famous numbers”
Each number’s detailed card is in Section 3; here is Lynch’s rule in a single line.
| Number | Rule | Card |
|---|---|---|
| Percent of sales | What percent of sales does the product you’re interested in account for? If it’s small, even “the next Pampers” is meaningless to the shareholder. | — |
| P/E ratio | In a fairly priced company, P/E ≈ the earnings growth rate. Look at the company’s own P/E history and the sector. Stay away from excessively high P/Es. | 3.3, 3.4 |
| Market P/E | If a few stocks are inflated, most of them are. 1971: 20 (madness), 1982: 8, 1987: 16. | 3.3 |
| Cash position | Cash + securities − long-term debt. Net cash per share is subtracted from the price; it’s the floor. | 3.7 |
| Debt | A company with no debt can’t go bankrupt. Bank debt is more dangerous than bond debt. | 3.8 |
| Dividends | The record over the years; what happened in recessions; the payout ratio. | 3.16 |
| Book value | Look at the real value, not the stated value; buying on book value alone is dangerous. | 3.14 |
| Hidden assets | Natural resources, real estate, brands, patents, tax losses, subsidiaries; carried on the books at historical cost. | 3.14 |
| Cash flow | Free cash flow / price: 10 percent is standard, 20 percent is terrific. | 3.9 |
| Inventories | If they’re growing faster than sales, it’s a red flag. | 3.15 |
| Growth rate | Only earnings growth counts; the ability to raise prices is the most valuable source of growth. | 3.5 |
| Pretax margin | Compare with others in the same industry; the highest margin is the lowest-cost operator. | 3.6 |
4.7 Five ways to increase earnings (the catalyst list)
- Reduce costs
- Raise prices
- Expand into new markets
- Sell more of its product in the old markets
- Revitalize, close, or otherwise dispose of a losing operation
These are the factors to research while developing the story; this is where your edge is most useful. Every thesis records “yes / no / planned” for each of the five items.
4.8 Growth phases and rechecking the story
| Phase | What happens | For the investor |
|---|---|---|
| Start-up | The kinks in the basic business are being worked out; success unproven. | The riskiest. |
| Rapid expansion | The successful formula is being copied into new markets. | The safest phase, and where the most money is made. |
| Maturity / saturation | No room left to expand; other ways of increasing earnings are needed. | The most problematic; the P/E contracts. |
Every few months: the latest report, are earnings as expected, are the stores still attractive, has a new card turned up. With fast growers, the question “what’s going to keep them growing?” is asked anew every time.
4.9 Portfolio: how many stocks, rotation, cash
- The number of stocks to own is the number of opportunities where you have an edge and that pass all the tests: it could be one, it could be a dozen. Don’t spread out into unknown companies just for the sake of diversity.
- Don’t go to cash; stay in the market forever and rotate stocks according to the fundamentals. Deciding that a certain amount will always stay in the stock market saves you from badly timed moves.
- Get in and out according to what the price has done relative to the story: if a stalwart has delivered the expected 40 percent and there’s nothing new, sell it and buy another stalwart that hasn’t gone up yet; if you don’t want to sell all of it, sell part.
- If you can’t say “when it drops 25 percent, I’m a buyer,” you’ll never make a decent profit in stocks.
- Don’t water the weeds and pull out the flowers: don’t hold a winner just because it went up, or a loser just because it went down. When favorable cards turn up, increase your bet; in the opposite case, do the opposite.
- Serious money is made by compounding 20–30 percent gains in stalwarts.
4.10 Timing and the market
- Within the next month, year, or three years, the market will drop sharply; declines push the companies you like to bargain prices. Predicting the direction one or two years out is impossible. The bells never ring; things never become clear until it’s too late.
- The recession-every-five-years theory isn’t written into the Constitution; your chance of predicting a recession is zero.
- Newsletter writers and advisors turn bullish/bearish at exactly the wrong moments (late 1972: 15 percent bearish, the 1974 bottom: 65 percent bearish, the 1987 peak: 80 percent bullish). Crowd sentiment is a contrary indicator.
- The best time to buy a stock is the day you’re convinced you’ve found solid merchandise at a good price. Two periods with a high probability of bargains: year-end (tax selling) and market collapses.
- You don’t have to be right all the time, or even most of the time, to come out ahead. The biggest winners are surprises; big results take years. The stock you expected at the outset isn’t the one that goes up.
4.11 The twelve silliest (and most dangerous) things people say about stock prices
If any of these sentences appears as a justification in the thesis or the decision journal, the decision is void.
- If it’s gone down this much already, it can’t go much lower.
- You can always tell when a stock’s hit bottom.
- If it’s gone this high already, how can it possibly go higher?
- It’s only $3 a share: what can I lose?
- Eventually they always come back.
- It’s always darkest before the dawn.
- When it rebounds to $10, I’ll sell.
- What me worry? Conservative stocks don’t fluctuate much.
- It’s taking too long for anything to ever happen.
- Look at all the money I’ve lost: I didn’t buy it!
- I missed that one, I’ll catch the next one.
- The stock’s gone up, so I must be right, or… the stock’s gone down so I must be wrong.
The accompanying truths: If a stock goes to zero, you lose your entire investment whether you bought at $50 or at $2. A company being in bad shape doesn’t mean it can’t get worse. A rising price doesn’t mean you’re right, and a falling one doesn’t mean you’re wrong. You lose nothing by not owning a successful stock; you don’t have to “kiss all the girls.” The stock doesn’t know you own it.
4.12 Final checklist (Lynch)
Stocks in general
Category-specific
Lynch’s one-line reminders
- Don’t overestimate the skill and wisdom of professionals.
- Take advantage of what you already know.
- Look for companies outside the radar.
- Invest in companies, not in the stock market.
- Ignore short-term fluctuations.
- Over the long term, stock returns are both predictable and superior to bonds.
- Common stocks aren’t for everyone, nor even for every phase of life.
- In the stock market, a bird in the hand is worth ten in the bush.
- Invest in dull, mundane, out-of-favor, simple companies.
- Companies growing 20–25 percent in no-growth industries are ideal.
- Be suspicious of companies growing 50–100 percent a year.
- Avoid the hot stocks in hot industries.
- Among troubled companies, pick the one with superior financial condition; avoid those loaded with bank debt.
- Management skill matters but is hard to assess; look at the prospects.
- Weigh the P/E carefully; at an exorbitant price you won’t make money even if everything goes right.
- Find a story line to follow.
- There’s always something to worry about; be open to new ideas.
- Devote at least as much time to picking a new stock as you would to picking a new refrigerator.
5. Buffett’s rules of thumb
The rules below summarize the widely known principles drawn from Warren Buffett’s Berkshire Hathaway shareholder letters, annual meeting remarks, and interviews. Items in quotation marks are his well-known sayings; the rest are paraphrases. Every rule follows the same layout: Rule · What it means · Why · How it is applied in the thesis. The calculation details for the terms are in the concept cards in Section 3.
The essence of Buffett’s approach. A share is a piece of a business; when you buy it, think as if you were buying the whole company. Good businesses are few: understandable, protected from competitors, generating a lot of cash on little capital, honestly run. Buy such a business at a price clearly below its value and, whatever the market does, hold it as long as the business stays good. The method rests on four pillars: business, management, financials, price; the fifth pillar is temperament, which makes it possible to apply the other four.
5.1 On the business
Circle of competence
- Rule. Invest only in businesses you understand. “Risk comes from not knowing what you’re doing.”
- What it means. Understanding = being able to estimate, with reasonable confidence, what the company will look like ten years from now. What matters is not the size of the circle but knowing where its edge is.
- Why. Valuation is forecasting future cash flows; a business you cannot forecast you cannot value, and a business you cannot value you cannot call cheap or expensive.
- In the thesis. The first sentence of Section 2. Card 3.1.
A simple, understandable business
- Rule. Prefer businesses where change is slow and the outcome is still predictable ten years out.
- What it means. Chewing gum, soft drinks, insurance, railroads: businesses that will make money the same way ten years from now. In a fast-changing industry, today’s leader can be tomorrow’s loser.
- Why. Buffett’s observation: change is the investor’s enemy, not friend; in a race where you do not know who will win, you are merely placing a bet. This meets Lynch’s “any idiot can run it” and “boring business” preferences at exactly the same spot.
- In the thesis. Section 2: “Will this business most likely be the same ten years from now?”
A durable moat
- Rule. Look for a business with a structural advantage that will withstand competitors’ attacks for years.
- What it means. Brand, cost advantage, switching costs, network effects, licenses (card 3.2). The test: “If someone gave me billions of dollars, could I take this company down?”
- Why. Without a moat, high profits attract competitors and profits revert to the mean; every growth assumption in a long-term valuation rests on the moat.
- In the thesis. Section 5. The type of moat is named, its evidence is written down, and its direction (widening/narrowing) is stated.
Pricing power
- Rule. The single most important thing when evaluating a business: can it raise prices without losing market share?
- What it means. If raising prices by 10 percent requires a “prayer session,” the business is weak. In inflationary periods, this is the only thing that protects the shareholder.
- Why. A price increase is costless profit growth; among Lynch’s five ways to increase earnings, it is the most valuable.
- In the thesis. Section 5, the price-increase test: the last five years’ price increases and volumes side by side.
A consistent operating history
- Rule. Prefer businesses with a ten-year record of steady earnings; “turnarounds seldom turn.”
- What it means. Buffett does not try to fix troubled companies; he buys the ones that are already good. When a bad business meets good management, it is usually the reputation of the business that remains intact.
- Why. Past earnings stability is the best indicator of future predictability.
- Tension with Lynch. Lynch selectively buys turnarounds and earns large gains from them; Buffett stays away. The thesis states explicitly which approach you adopt; if a turnaround thesis is being written, the position is kept small, per Buffett’s warning.
- In the thesis. Section 8 (ten-year earnings table), Section 3 (category).
Low capital requirements
- Rule. Avoid businesses that constantly need new capital to grow; look for businesses that generate a lot of cash on little capital.
- What it means. See’s Candies: growth does not require new factories, and the profit stays with the owner. An airline: every step of growth demands new planes and new debt; the profit never reaches the shareholder.
- Why. This is where the gap between owner earnings (card 3.10) and reported earnings opens up. In a capital-intensive business, reported earnings are high while the money that actually reaches your pocket is low.
- In the thesis. Section 8: owner earnings / net income ratio; capex / sales ratio.
Avoid commodity businesses
- Rule. A company whose product is indistinguishable from its competitors’ is a price taker; take an interest only if it is the lowest-cost producer.
- What it means. Steel, paper, airline seats, crude oil: the customer buys the cheapest, and margins depend on the cycle.
- Why. There is no pricing power; profit is the result of the industry’s supply-demand balance, not of the company.
- Link to Lynch. This is precisely Lynch’s cyclical category; Lynch plays these with timing, Buffett generally does not.
- In the thesis. Section 6 (industry), Section 3 (category).
5.2 On management
Able and honest managers
- Rule. Look for three qualities: intelligence, energy, integrity; without the last one, the first two will sink you.
- What it means. Judge management by its record, not its talk: did it keep past promises, did it admit mistakes, how does it speak to shareholders?
- Why. The shareholder cannot see the day-to-day running of the business; management’s honesty is the precondition for the reliability of every number in the thesis.
- In the thesis. Section 7.
Rational capital allocation
- Rule. A manager’s real job is deciding where to put the money the business earns.
- What it means. Five options: reinvestment in the business, acquisitions, debt repayment, dividends, buybacks. The right choice depends on the return of each option: if the return on reinvesting in the business is high, reinvest; if not, distribute. A buyback makes sense only when the stock trades below intrinsic value.
- Why. Over the long run, shareholder returns are determined less by the size of earnings than by where the earnings go. Most CEOs come up through operations and never learned capital allocation.
- In the thesis. Section 7: for each of the five options, what the company did over the last five years and what return it earned.
The one-dollar test
- Rule. Every $1 of retained earnings must, over time, create at least $1 of market value.
- What it means. The concrete exam of management’s capital allocation (card 3.12). If it does not create that value, the money should have been paid out.
- Why. Retaining earnings is making an investment decision on the shareholder’s behalf; the result must be measured.
- In the thesis. Section 7: ten-year ratio of retained earnings to the increase in market value.
Candor
- Rule. Trust management that writes openly about bad news and mistakes in the shareholder letter; do not trust management that only talks about the good quarters.
- What it means. “Adjusted” earnings, perpetual “one-time” charges, stock-based compensation not counted as an expense: these are the opposite of candor.
- Why. Management that deceives itself in public deceives itself in private too; in the end, the company pays the price.
- In the thesis. Section 7: the “bad news” sections of the last five annual reports are compared.
Resistance to the institutional imperative
- Rule. Look for management that stays away from acquisitions, expansions, and fashionable projects done “because the competitor is doing it.”
- What it means. Buffett’s observation of the “institutional imperative”: institutions act not because something makes sense but because others are doing it; for every strategic move, subordinates promptly produce a report justifying it.
- Why. This is the source of Lynch’s diworsification.
- In the thesis. Section 7: the rationale and outcome of recent acquisitions.
Management that thinks like an owner
- Rule. Prefer managers who have a significant part of their own wealth in the company and who talk about the business, not the stock price.
- What it means. Look at the structure of executive compensation: is it tied to return on capital, rather than to the share price or to size? This meets Lynch’s insider-buying rule at exactly the same spot.
- In the thesis. Section 7: ownership percentage, compensation structure, insider purchases.
5.3 On financials
Return on equity, not earnings per share
- Rule. Measure the quality of growth with ROE/ROIC; look for steady high returns achieved with no or low debt.
- What it means. Earnings grow simply because retained earnings accumulate; the skill lies in making the accumulated money work at high returns too (card 3.11).
- Why. Long-term returns converge toward the return on reinvested capital; a business with an ROIC of 8 percent, however cheaply bought, will earn you around 8 percent over the long run.
- In the thesis. Section 8: ten-year ROE and ROIC series, together with debt/equity.
Owner earnings
- Rule. Measure the value of a business not by reported earnings but by the money that actually reaches the owner’s pocket.
- What it means. Net income + depreciation − the investment required to sustain the business ± working capital (card 3.10). Depreciation is a non-cash expense and is added back; but machines really do wear out, and the cost of replacing them is subtracted.
- Why. Two companies can report the same net income; one can distribute all of it, while the other must spend all of it keeping the factory standing. The second one’s “profit” does not exist for the shareholder.
- In the thesis. Section 8 (cash quality), Section 6.4 (DCF input).
High and stable margins
- Rule. Look for businesses whose margins are higher than their competitors’ and stable over the years; those that have made cost-cutting a habit, not those that announce it as a “program.”
- What it means. High margins are the numerical evidence of a moat; stability is the evidence of pricing power (card 3.6).
- In the thesis. Section 8, Section 5.
Little debt
- Rule. A good business delivers good results without debt too; debt magnifies returns in good times and threatens survival in bad times.
- What it means. “When smart people go broke, the reason is usually leverage.” Debt takes away your right to make mistakes; a debt-free company can wait out a bad year, an indebted company cannot.
- Why. Directly tied to Buffett’s Rule No. 1 (never lose money): the most frequent cause of permanent loss of capital is debt.
- In the thesis. Sections 8 and 10 (card 3.8).
Skepticism toward accounting embellishments
- Rule. Do not trust EBITDA, “adjusted” earnings, or stock-based compensation that is not counted as an expense.
- What it means. “The tooth fairy doesn’t pay for capex”: EBITDA, which ignores depreciation, makes a capital-intensive business look more profitable than it is. Stock-based compensation is a real expense; if it is not an expense, what is it?
- In the thesis. Section 8: the gap between reported earnings and owner earnings is explained.
5.4 On value and price
Intrinsic value
- Rule. The value of a business is the sum of the cash it will leave to its owner over its remaining life, discounted to the present at an appropriate rate.
- What it means. Card 3.17 and Section 6.4. Not a precise number but a range; two people with the same data will arrive at different results, and both can be reasonable.
- Why. Ratios such as P/E and P/B are shortcuts to this; a shortcut used without knowing the real road misleads.
- In the thesis. Section 9.
Margin of safety
- Rule. Buy clearly below your estimate of intrinsic value; leave room for estimation error, bad luck, and the unknown.
- What it means. Card 3.18. 25 percent in a stable business, 50 percent in an uncertain one. Graham’s “three most important words.”
- Why. There is no certainty in investing; the margin ensures you do not lose money when you are wrong, and at the same time it is the source of the return.
- In the thesis. Sections 9 and 12 (buy ceiling).
“Price is what you pay; value is what you get.”
- Rule. Price movement is not information; the value of the business rests on information.
- What it means. The company did not get worse because the stock fell, nor better because it rose. Lynch’s twelfth silly thing (“it went up, so I must be right”) is the same mistake.
- In the thesis. Section 14: price movement cannot be the rationale for a decision.
“It’s far better to buy a wonderful company at a fair price than a fair company at a wonderful price.”
- Rule. Business quality first, price second.
- What it means. The point where Buffett parted from Graham (under Charlie Munger’s influence). Time is the friend of the good business and the enemy of the mediocre one: a wonderful business accumulates value every year and makes up for being bought a bit expensively; a mediocre business keeps losing value even if bought cheaply (the “value trap”).
- Why. The ROIC math (card 3.11): over the long run, returns converge toward the business’s return on capital.
- In the thesis. Sections 5 and 8 are written first, Section 9 afterward; the order is deliberate.
Mr. Market
- Rule. The manic-depressive partner who quotes you a price every day is there to serve you, not to guide you.
- What it means. Card 3.19. If the quote is absurd, ignore it; sell into his euphoria, buy into his despair.
- In the thesis. Section 12: the condition for adding is tied to the story, not the price.
Rule No. 1: Never lose money. Rule No. 2: Never forget rule No. 1.
- Rule. Avoid permanent loss of capital.
- What it means. A temporary price decline (story intact, price low) is different from a permanent loss (the business broke down, debt sank it, the stock went to zero). The first is an opportunity, the second a disaster. Recovering from a 50 percent loss requires a 100 percent gain.
- Why. The enemy of compounding is the large loss; the difference is made not in average years but in disaster years.
- In the thesis. Section 10: the probability and size of permanent loss in the bear scenario; if it exceeds 50 percent, the position is reduced.
Be fearful when others are greedy and greedy when others are fearful
- Rule. The crowd’s emotion is a contrary indicator.
- What it means. The same as Lynch’s observations that “corrections create bargains” and “advisors turn bullish/bearish at exactly the wrong moment.”
- In the thesis. Section 11 (timing), Section 12 (adding).
When to sell
- Rule. In three situations: (1) the thesis turned out to be wrong; (2) the economics of the business have permanently deteriorated; (3) there is a clearly better opportunity. “The price went up” is not one of them.
- What it means. “Our favorite holding period is forever”: a good business is not sold because its price rose; it keeps accumulating value over time.
- Tension with Lynch. Lynch rotates out of stalwarts after a 30–50 percent gain; Buffett holds. In the thesis, the choice is made by category: Lynch’s P/E band rule for stalwarts and fast growers, Buffett’s “hold” rule for companies with an extraordinary moat.
- In the thesis. Section 12.
5.5 On behavior
The twenty-punch card
- Rule. Imagine you can make only twenty investments in your lifetime; make every decision accordingly.
- What it means. A limited allowance makes every decision more careful and weeds out the “let me try a little” purchases.
- In the thesis. The effort of writing a thesis is the application of this rule; a stock not worth writing a thesis for is not bought.
Concentration
- Rule. Diversification is protection for those who do not know what they are doing; for those who do, concentrating on the best ideas makes more sense.
- What it means. Overlaps with Lynch: as many stocks as you have an edge in. Putting money into the thirteenth or fourteenth best idea means taking it away from the first.
- In the thesis. Section 12: position size is tied to thesis quality.
The ten-year test
- Rule. If you aren’t willing to own a stock for ten years, don’t even think about owning it for ten minutes.
- What it means. The question “Would I be happy owning this company even if the stock market closed for ten years?” forces you to see the stock as a business, not a piece of paper.
- In the thesis. Section 1: the sentence “ten years from now, this company …” is added to the two-minute story.
You don’t have to swing
- Rule. There are no called strikes in investing; swing only at the pitch right down the middle of the strike zone.
- What it means. If there is no opportunity, sitting in cash is acceptable. (Tension with Lynch’s “don’t go to cash” rule: Lynch, as a fund manager, always stays invested; for the individual investor, Buffett’s flexibility applies. In the thesis: waiting while the price is above the buy ceiling is a decision, and it is recorded in the journal.)
- In the thesis. Section 12.
Ignore macro forecasts
- Rule. Forecasts of interest rates, elections, and the economy do not enter the decision process; what the company will look like in ten years does.
- What it means. Lynch’s “forecasting the economy is futile.” Macro can appear as a scenario in the risks section, not as a reason to buy.
- In the thesis. A scenario in Section 10; never in Section 0.
Temperament beats intellect
- Rule. Investing requires not a superior IQ but the ability to think independently of the crowd and to be boring.
- What it means. The same as Lynch’s list of qualities (Section 4.1): patience, detachment, ignoring panic.
- In the thesis. Section 10, “my own risk.”
Read the annual report
- Rule. Read every day; investment knowledge accumulates like compound interest.
- What it means. Lynch’s “at least one hour of research a week” rule and his observation that “the cheaper the paper, the more valuable the information.”
- In the thesis. Section 13.
5.6 Where the rules fit in the thesis
| Thesis section | Buffett rules to apply |
|---|---|
| 1 · Two-minute story | Ten-year test |
| 2 · Company and business model | Circle of competence; simple business; low capital requirements; avoid commodity businesses |
| 5 · Moat and niche | Durable moat; pricing power; “could I take it down with a billion dollars?” |
| 7 · Management | Honest and able management; rational capital allocation; one-dollar test; candor; institutional imperative; thinking like an owner |
| 8 · Financials | ROE/ROIC; owner earnings; margins; little debt; accounting skepticism |
| 9 · Valuation | Intrinsic value (DCF); margin of safety; price ≠ value; wonderful company at a fair price |
| 10 · Risks | Rule No. 1: permanent loss of capital; leverage; macro only as a scenario |
| 12 · Position and plan | Concentration; twenty-punch card; three reasons to sell; Mr. Market; you don’t have to swing |
| 13 · Monitoring | Reading the annual report; watching the business, not the price movement |
6. Methods for estimating the future share price
No single method is sufficient on its own; every thesis uses at least two methods and three scenarios. The shared purpose of the methods is not a precise price but an answer to the question “if I buy at this price, what is my expected annual return, and how much room do I have to be wrong?” Every method follows the same layout: what it does, which category it suits, the steps, an example, the trap.
6.1 The basic logic
Over the long run the share price follows earnings (Lynch’s earnings-line / price-line charts). Price breaks down into two components:
Price = Earnings per share (EPS) × Price/earnings multiple (P/E) Future price (n years out) = EPS_n × P/E_n EPS_n = EPS_0 × (1 + g)^n g: annual earnings growth rate Total return = (Price_n + dividends received over n years) / Price_0 Annual return (CAGR) = (Total return)^(1/n) − 1
So the return comes from three sources: earnings growth, change in the multiple (P/E expansion or contraction), and dividends. The thesis writes these three out separately, because a thesis that rests on multiple expansion (“the market will re-rate this company”) is far more fragile than one that rests on earnings growth (“the company will earn more”). The first depends on the market changing its mind; the second depends on the company doing its job.
6.2 Method A · Earnings × multiple projection (Lynch style)
What it does. Carries today’s earnings forward at a growth rate, applies a reasonable P/E to that year’s earnings, and computes the annual return from the resulting price relative to today’s price. The simplest, most transparent method.
Who it suits. The main method for the fast grower, stalwart, and slow grower categories.
Steps.
- Take today’s EPS (strip out one-time items; trailing twelve months if the business is not cyclical).
- Write three assumptions for the earnings growth rate g: bear / base / bull. The base assumption is derived from the company’s earnings growth over the last 5–10 years and the catalysts in Section 11; treat growth above 25 percent with Lynch’s caution.
- Horizon n = 3–5 years. Compute EPS_n.
- Assumption for the exit P/E (P/E_n): between the midpoint of the company’s own historical range and the normal range for its category; a lower multiple if growth is going to slow (a fast grower turning into a stalwart lowers the multiple). Never carry today’s inflated multiple into the future.
- Target price = EPS_n × P/E_n; add dividends; compute the CAGR.
- Upper limit for the purchase price: the price that delivers the target CAGR (e.g. 15 percent) in the base scenario = (Target price + dividends) / (1 + target CAGR)^n.
Example. EPS_0 = 4.00; price = 60 (P/E 15); base growth 15 percent; n = 5; exit P/E 15; no dividend.
- EPS_5 = 4.00 × 1.15^5 = 8.05 → Target = 8.05 × 15 = 120.7 → CAGR = (120.7/60)^(1/5) − 1 = 15 percent.
- Bear: growth 8 percent, P/E 11 → EPS_5 5.88, target 64.6, CAGR 1.5 percent.
- Bull: 20 percent, P/E 18 → 9.95 × 18 = 179, CAGR 24 percent.
- Purchase ceiling for a 15 percent target CAGR (base) = 120.7 / 1.15^5 = 60. Today’s price sits exactly at the limit; there is no margin of safety, so buy in stages or wait.
Trap. The entire result hangs on two assumptions (g and the exit P/E); choosing both optimistically easily doubles the target. That is why the bear scenario is mandatory, and if the exit P/E is to be set higher than today’s, the justification is written down.
6.3 Method B · Fair P/E and the Lynch formula
What it does. Compares the P/E with growth to give a quick “cheap / reasonable / expensive” verdict, and provides an anchor for the exit multiple in Method A (card 3.4).
Fair P/E ≈ long-term earnings growth rate (in percent) [P/E = growth → PEG = 1]
PEG = P/E / growth rate 0.5 very favorable · 1 reasonable · 2 very unfavorable
Lynch score = (growth rate + dividend yield) / P/E
below 1 poor · 1.5 okay · 2 and above what you are looking for
Fair price (rough) = EPS_0 × growth rate
Use. Quick screening. Limits. The formula breaks down when growth is above 30 percent or below 5; it is meaningless for cyclicals and turnarounds. When interest rates are high the fair P/E falls (the P/E competes with the bond yield).
6.4 Method C · Discounted cash flow (DCF): Buffett’s intrinsic value
What it does. Estimates, year by year, the cash the company will generate in the future, shrinks (discounts) each year’s cash by the logic of “how much would this be worth if I had it today,” and adds it up. The resulting number is the present value of the business; it is compared with the price (card 3.17).
Why “discounting” is necessary. 110 lira a year from now is equivalent to 100 lira today for someone expecting a 10 percent return; 100 lira ten years from now is worth only 100/1.10^10 ≈ 38.6 lira today. Cash in distant years is worth less because waiting is costly and uncertainty grows. The discount rate r represents both of these together: the return you want and the risk you take on.
Who it suits. Stalwarts, slow growers, and fast growers with a strong moat; businesses with predictable cash flow. Not used for cyclicals and turnarounds, because the inputs cannot be forecast.
Intrinsic value = Σ [ FCF_t / (1 + r)^t ] (t = 1..N) + Terminal value / (1 + r)^N Terminal value (Gordon) = FCF_N × (1 + g_t) / (r − g_t) r : discount rate (required return; e.g. 9–12 percent, higher for a risky business) g_t : perpetual growth (should not exceed 2–3 percent; no company grows faster than the economy forever) N : explicit forecast period (5–10 years) Intrinsic value per share = (Intrinsic value − net debt + excess cash) / share count Purchase price ceiling = Intrinsic value per share × (1 − margin of safety) [margin 25–50 percent]
What the terminal value is. The company keeps generating cash after year N; instead of writing out every year individually, the value of all the years after N is computed with a single formula: the value of a cash flow growing at a constant rate (g_t) forever. This formula is extremely sensitive to the difference r − g_t; pushing g_t above 3 percent inflates the value rapidly.
Steps.
- As the input, take the average free cash flow or owner earnings of the last 3–5 years (smooth out the cycle effect; cards 3.9, 3.10).
- Growth for the first 5 years (base and bear), lower growth for the following 5 years.
- Choose r: 10 percent is a reasonable starting point; 12–15 for an indebted, cyclical, or single-customer business.
- Look at the terminal value’s share of total value: if it exceeds 70 percent, the value rests on “forever,” which is fragile.
- Write a sensitivity table for r and g (r 9/10/12, g_t 2/3).
- Apply the margin of safety.
Example. FCF_0 = 100 million; 10 percent for the first 5 years, 5 percent for the next 5; r = 10 percent; g_t = 2.5 percent.
| Year | FCF | Discount factor 1/(1.10)^t | Present value |
|---|---|---|---|
| 1 | 110.0 | 0.909 | 100.0 |
| 2 | 121.0 | 0.826 | 100.0 |
| 3 | 133.1 | 0.751 | 100.0 |
| 4 | 146.4 | 0.683 | 100.0 |
| 5 | 161.1 | 0.621 | 100.0 |
| 6 | 169.1 | 0.564 | 95.5 |
| 7 | 177.6 | 0.513 | 91.1 |
| 8 | 186.5 | 0.467 | 87.0 |
| 9 | 195.8 | 0.424 | 83.0 |
| 10 | 205.6 | 0.386 | 79.3 |
| Total 1–10 | 936 |
Terminal value = 205.6 × 1.025 / (0.10 − 0.025) ≈ 2,810; its present value = 2,810 × 0.386 ≈ 1,083. Intrinsic value ≈ 936 + 1,083 ≈ 2,020 million. Net debt 200 million, share count 50 million → per share ≈ 36.4. With a 30 percent margin, the purchase ceiling ≈ 25.5. Terminal share 1,083/2,020 = 54 percent: acceptable.
Sensitivity (value per share).
| g_t = 2% | g_t = 2.5% | g_t = 3% | |
|---|---|---|---|
| r = 9% | 39.9 | 41.6 | 43.6 |
| r = 10% | 35.3 | 36.4 | 37.7 |
| r = 12% | 28.8 | 29.4 | 30.1 |
The table shows that the right way to put it is not “the value is 36” but “the value is somewhere between 29 and 44, depending on the assumptions.”
Trap. DCF is highly sensitive to its inputs; that does not make the method worthless, it makes the question “which assumption is carrying the thesis” visible. Bad practice: deciding on the target price first and then choosing the assumptions that make the DCF fit it.
6.5 Method D · Reverse DCF: what is the market pricing in?
What it does. Takes today’s price as the intrinsic value and solves the DCF backwards: how many years of growth, at what percentage, are needed to justify this price? Then it asks: is that growth consistent with the company’s history, the size of the industry, and Lynch’s “Gulliver dilemma” (room to grow)?
How it is done. In the DCF table, vary the growth rate to find the g that sets the value per share equal to today’s price (“goal seek” in a spreadsheet, or trial and error).
Interpretation. If the growth the market is pricing in is above your base scenario, the stock is expensive; if it is below, either the market knows something (investigate in Section 10) or there is an opportunity. Lynch’s Avon example is the essence of reverse DCF: a P/E of 50 required Avon to sell a billion bottles of perfume; with every other housewife in America already an Avon lady, that was impossible.
6.6 Method E · Dividend discount model (slow growers, utilities)
What it does. Values the company solely on the dividends it pays; the dividend is treated as a cash flow growing at a constant rate (Gordon growth model).
Fair price = D_1 / (r − g) D_1: next year's dividend, g: sustainable dividend growth Expected return ≈ dividend yield + dividend growth
Example. A dividend of 2 lira per share next year, growing 4 percent a year, required return 9 percent: fair price = 2 / (0.09 − 0.04) = 40. If the stock is at 32 it trades at a 20 percent discount; expected return = 2/32 + 4 percent ≈ 10.3 percent.
Check. If the payout ratio exceeds 70 percent, do not be optimistic about g; the dividend record must have been tested in recessions (Lynch’s slow grower checklist, card 3.16). Trap. As r − g shrinks the formula becomes hypersensitive; never let g approach r (with r at 9 percent and g at 8 percent → the value inflates tenfold).
6.7 Method F · Asset-based valuation (asset plays, net cash adjustment)
What it does. Values the company not on earnings but on the present value of the assets it owns. It has two forms: net asset value (liquidation logic) and sum of the parts (valuing each division separately and adding them up).
Net asset value (NAV) = Σ (current market value of each asset) − all debt − hidden liabilities NAV per share = NAV / share count Discount = 1 − Price / NAV per share Sum of the parts (SOTP) = Σ (each division's earnings × that division's industry P/E) + net cash − debt P/E of the core business = (Price − net cash per share − value of subsidiaries per share) / core EPS
Example (Lynch’s Ford calculation, SOTP in its simplest form). The stock is at 38 dollars. Net cash per share is 16.30. The finance subsidiary (Ford Credit etc.) earns 1.66 dollars per share; at the customary P/E of 10 for finance companies, that is worth 16.60 dollars. The auto business is being bought for 38 − 16.30 − 16.60 = 5.10 dollars; analysts expect 7 dollars of earnings from the auto business. Core P/E 0.7: “a real bargain even though the stock had gone up tenfold since 1982.”
Check. Value the asset with your own valuation (current price, saleability, taxes), not the stated book value (card 3.14: Pebble Beach, the Handy and Harman hidden asset; the Penn Central book-value trap). Debt is deducted first; creditors stand ahead in line. Trap. Closing the discount requires a catalyst (a raider, a spin-off, a sale); without a catalyst the discount can persist for years, and management can meanwhile erode the asset with debt.
6.8 Method G · Cyclicals: normalized earnings × mid-cycle multiple
What it does. A single year’s earnings for a cyclical company is misleading (very high at the peak, zero at the trough); instead, it computes the average earnings over a full cycle (“normalized earnings”) and applies a mid-cycle multiple to that.
Normalized EPS = average EPS over a full cycle (7–10 years) (or average margin × today's sales) Target price = Normalized EPS × mid-cycle P/E Peak warning: when earnings are at the peak the P/E looks low (trap); at the trough, with earnings zero/negative, the P/E looks infinite (may be an opportunity).
Example. An automaker’s EPS over the last ten years: 1, 3, 6, 8, 9, 4, −2, 0, 3, 7 → average 3.9. Mid-cycle P/E 10 → normalized value 39. The stock is currently at 70 (P/E 10 on EPS of 7, “looks cheap”) → in fact 18 times normalized earnings, expensive. When the same stock drops to 25 at the trough with EPS of −2, the P/E is meaningless, but it trades at a 36 percent discount to normalized value.
Interpretation. For cyclicals a low P/E is not a buy signal; most of the time it is a sell signal (Lynch: as earnings approach the peak the P/E contracts, and the leading crowd sells a year ahead). Ratios less affected by the cycle, such as price/sales or price/book, and leading indicators such as inventories, capacity, and commodity prices are added to the table.
6.9 Turnarounds
What it does. Estimates the “normalized” earnings after the trouble is over (after the unprofitable division is sold and costs are cut: when Lockheed dropped the L-1011, EPS went from 1.50 to 10.78), then applies the multiple of the category the company will belong to once it has recovered.
Order matters. First the survival test (card 3.8): cash / annual cash burn = how many years it can last; debt maturities; the likelihood of dilution (at Navistar the recovery never reached the shareholders). If it cannot survive, valuation is meaningless.
Target price = Normalized EPS × (post-turnaround category) P/E Bear scenario = what is left to shareholders in bankruptcy (usually zero) × its probability
Example. The company is losing money today, the stock is at 8. Once the unprofitable division is closed and costs are cut, EPS of 2 dollars is expected; after the turnaround it will be a stalwart, P/E 12 → target 24 (a threefold gain). Probability of bankruptcy 30 percent, in which case shareholders get zero. Expected value = 0.7 × 24 + 0.3 × 0 = 16.8; still attractive relative to today’s 8, but the position is kept small in light of the 30 percent chance of a total wipeout.
6.10 Scenario table and asymmetry
This table is filled in for every thesis:
| Scenario | Assumptions (growth, multiple, catalyst) | EPS_n | P/E_n | Target | Probability | CAGR |
|---|---|---|---|---|---|---|
| Bear | The story breaks; growth slows; the multiple contracts | 25% | ||||
| Base | The thesis plays out as planned | 50% | ||||
| Bull | Catalysts arrive early and strong | 25% | ||||
| Expected price | = Σ (probability × target) |
Upside/downside asymmetry = (Bull target − Price) / (Price − Bear target)
Asymmetry rule: the ratio should be at least 2–3. If the permanent loss of capital in the bear scenario exceeds 50 percent (Buffett’s rule number 1), the position is kept small or the stock is dropped from the thesis. Lynch’s floor: net cash per share is written down as the floor of the bear scenario.
Example (the company from 6.2). Price 60; bear 64.6; base 120.7; bull 179. Expected = 0.25×64.6 + 0.5×120.7 + 0.25×179 = 121. Asymmetry = (179 − 60)/(60 − 64.6): even the bear scenario is above the price, no loss → the asymmetry is very favorable. But be careful: 8 percent growth in the bear scenario may still be optimistic; if a real bear case is written (growth 0, P/E 9 → target 36), the asymmetry is (179−60)/(60−36) = 5, still favorable.
6.11 Which method for which category
| Category | Main method | Cross-check method | Trap |
|---|---|---|---|
| Slow grower | E · Dividend model | A (low growth, low multiple) | Assuming the dividend is sustainable |
| Stalwart | A · Earnings × multiple; C · DCF | B · Lynch score; historical P/E band | Carrying today’s inflated multiple into the future |
| Fast grower | A · Earnings × multiple (the multiple must contract) | B · PEG; D · Reverse DCF | Assuming growth will last forever; P/E contraction as a “double whammy” |
| Cyclical | G · Normalized earnings | P/S, P/B; leading indicators | Mistaking a low P/E on peak earnings for cheap |
| Turnaround | 6.9 · Normalized earnings × category P/E | Survival test; F · Liquidation value | Dilution; liabilities that cannot be measured |
| Asset play | F · NAV / SOTP | Catalyst analysis | Trusting book value; forgetting the debt; waiting without a catalyst |
7. Templates
The templates in this section are copied into a new thesis file and filled in.
7.1 One-page thesis summary
| Field | Content |
|---|---|
| Company / Ticker / Exchange | |
| Date / Price / Market cap / Share count | |
| Category (Lynch) and rationale | |
| Thesis sentence | I am buying … because (1) … (2) … (3) …; in N years I expect earnings to be … and the stock to reach the … range. |
| My edge | |
| Moat / niche | |
| Key figures | Earnings growth 5y: … · Pre-tax margin: … (competitor: …) · Net cash/share: … · Debt/equity: … · FCF yield: … · ROIC: … · P/E: … (historical: …, sector: …) · PEG: … · Lynch score: … |
| Target price (bear / base / bull, probability) | |
| Expected CAGR (base) / Buy ceiling | |
| Catalysts (five ways) | |
| Thesis-breaking conditions (3) | 1. … 2. … 3. … |
| Sell triggers (category) | |
| Position size / buying plan | |
| Next review |
7.2 Full thesis skeleton
# [Company] — Investment Thesis (v1, [date])
## 0. Cover summary
(Table 7.1)
## 1. The two-minute story
Why I'm interested · What needs to happen · Pitfalls · Ten years from now this company …
## 2. Company and business model
Products · Customer · Segment percentages · Repeat purchase or not · Boringness · Technology user or not · Circle of competence
## 3. Category
Which one · Why · Expectation · Category checklist (4.12)
## 4. My edge
Type · Evidence · Tip vs. tipster distinction
## 5. Moat and niche
Type · Evidence · Price-raising test · Direction
## 6. Industry and competition
Growth · New entrants · Is it hot · "The next X" · Institutional ownership / analyst count
## 7. Management and capital allocation
Insider buying · Buybacks/dilution · Acquisitions · One-dollar test · Candor · Ownership and compensation
## 8. Financials (5–10-year tables)
Earnings growth · Margin (vs. competitors) · Net cash · Debt · FCF and owner earnings · ROE/ROIC · Book value/hidden assets · Inventories · Dividend · Share count · Customer concentration
## 9. Valuation
P/E context · PEG/Lynch score · Cash adjustment · At least two methods · Sensitivity · Scenario table (6.10) · Buy ceiling
## 10. Risks and thesis-breaking conditions
Falsification sentences · Business · Financial · Category · Valuation · My own risk
## 11. Catalysts and timing
Five ways · Concrete events · Horizon
## 12. Position and plan
Size · Buying · Adding · Sell triggers (7.4)
## 13. Monitoring plan
Quarterly checklist (7.3) · Growth phase · Is the story the same
## 14. Decision log
(Table 7.5)
7.3 Quarterly review
7.4 Sell trigger table
| # | Trigger (measurable) | Source | Check frequency | Action |
|---|---|---|---|---|
| 1 | Thesis-breaking condition | Quarterly | Sell | |
| 2 | Lynch category signal | Quarterly | Trim / sell | |
| 3 | Valuation (P/E band) | Monthly | Trim, rotate | |
| 4 | Better opportunity | Ongoing | Rotate |
7.5 Decision log
| Date | Price | Action | Reason (one sentence) | Thesis section changed | Next check |
|---|---|---|---|---|---|
8. Appendix: Formulas and glossary
8.1 Formulas
P/E = Price / EPS PEG = P/E / earnings growth rate (%) Lynch score = (growth % + dividend yield %) / P/E Earnings yield = EPS / Price = 1 / P/E (compare with the bond yield) Net cash = Cash + marketable securities − long-term debt Net cash per share = Net cash / share count Core P/E = (Price − net cash per share − value of affiliates) / core EPS Pre-tax margin = Pre-tax profit / Sales Free cash flow (FCF) = Cash from operations − capital expenditure Owner earnings (Buffett) = Net income + depreciation/amortization − maintenance capex ± working capital FCF yield = FCF / Market cap ROE = Net income / Equity ROIC = After-tax operating profit / (Debt + Equity − excess cash) Debt/equity = Total debt / Equity Net debt / EBITDA = (Debt − Cash) / (Operating profit + depreciation) Interest coverage = Operating profit / Interest expense Payout ratio = Dividends / Net income One-dollar test = Increase in market cap (n years) / Retained earnings (n years) → above 1 Market cap = Price × Share count Enterprise value (EV) = Market cap + Debt − Cash EPS_n = EPS_0 × (1+g)^n · Target price = EPS_n × P/E_n CAGR = (Ending / Beginning)^(1/n) − 1 Rule of 72: doubling time ≈ 72 / annual growth % Buy ceiling = (Target price + dividends) / (1 + target CAGR)^n DCF: Value = Σ FCF_t/(1+r)^t + [FCF_N(1+g_t)/(r−g_t)]/(1+r)^N Gordon: Price = D_1 / (r − g) NAV = Σ market value of assets − liabilities − hidden liabilities Margin of safety = 1 − Price / Intrinsic value Expected price = Σ (probability_i × target_i) Upside/downside asymmetry = (Bull − Price) / (Price − Bear)
8.2 Glossary
| Term | Meaning | Card |
|---|---|---|
| Depreciation | Spreading the cost of a long-lived asset over the years as an expense; there is no cash outflow | 3.10 |
| Raider | An activist/acquirer who moves into a company to unlock its hidden assets | 4.3 |
| Mr. Market | The moody imaginary partner who quotes a price every day (Graham) | 3.19 |
| One-dollar test | Every 1 dollar of retained earnings must create at least 1 dollar of market value | 3.12 |
| Capex (capital expenditure) | Spending on long-lived assets such as factories, machinery and stores; split into maintenance and growth | 3.9, 3.10 |
| Thesis-breaking condition | A measurable event written down in advance which, if it occurs, shows the thesis is wrong | Sections 2/10 |
| DCF | Discounted cash flow; the present value of future cash | 6.4 |
| Book value | Assets − liabilities on the balance sheet; recorded at historical cost | 3.14 |
| Cyclical | A company whose earnings rise and fall regularly with the economic cycle | 4.3 |
| EPS | Earnings per share | 3.5 |
| EV (enterprise value) | Market cap + debt − cash; the cost of buying the whole company | 3.20 |
| P/E | Price / earnings per share | 3.3 |
| EBITDA | Earnings before interest, depreciation and taxes | 3.20 |
| Whisper stock | A long shot with no substance but an attractive story | 4.5 |
| Margin of safety | The gap between the intrinsic value estimate and the price paid | 3.18 |
| Moat | A structural advantage that protects a company’s profits from competitors | 3.2 |
| Fast grower | A small, aggressive company growing 20–25 percent a year | 4.3 |
| Intrinsic value | The present value of the cash a business will generate over its lifetime | 3.17 |
| Working capital | Inventory + receivables − payables; swallows cash in a growing business | 3.10 |
| Diworseification | Unrelated acquisitions that weaken the core business | 4.5 |
| Institutional imperative | Institutions acting because “others are doing it” rather than on logic | 5.2 |
| Net cash | Cash + marketable securities − long-term debt | 3.7 |
| Tenbagger | A stock that rises to ten times its purchase price; Lynch’s baseball term | 4.3 |
| PEG | P/E divided by the growth rate; around 1 is reasonable | 3.4 |
| ROE / ROIC | Return on equity / return on invested capital | 3.11 |
| Stalwart | A large company growing 10–12 percent a year that holds up in a crisis | 4.3 |
| Owner earnings | The cash left to the owner after deducting the investment needed to sustain the business | 3.10 |
| Free cash flow | Cash from operations − capital expenditure | 3.9 |
| Dilution | The shrinking of existing shareholders’ stake through the issuance of new shares | 3.13 |
| Terminal value | In a DCF, the value of all cash flows after the explicit forecast period | 6.4 |
| Turnaround | The comeback of a battered company on the brink of bankruptcy | 4.3 |
| Asset play | A company sitting on a valuable asset the market has overlooked | 4.3 |
| Wall Street lag | The years that pass before a good company is noticed by institutions and analysts | 4.2 |
| Slow grower | A mature company growing at the pace of GNP, held for its dividend | 4.3 |
| Circle of competence | The boundary of the businesses you truly understand | 3.1 |
Version note. This rulebook is a living document. When a new book is read, the rules drawn from it are added as a separate section and linked by tag to the relevant thesis sections in Section 2; new concepts are added to Section 3 as cards. The valuation methods section is expanded with every new method used.


























