dcf-valuation

DCF thinking valuation method (a way of thinking based on discounting future cash flows, including rough estimation + sealing the position for ten years)

R — Original text (Reading)

Buying a stock is buying a company, and buying a company is buying the discounted value of the company's future cash flows, period! Discounting future cash flows is only a way of thinking; only for companies within one's circle of competence can an investor understand them through rough estimation. No one really uses formulas, at least the survivors all don't use formulas. To emphasize again: discounting future cash flows refers to a way of thinking; those who want to use formulas should go use them at school.

— Duan Yongping, Chapter 1 The Faith of Investing / Chapter 5 Valuation Logic

I — Methodological skeleton (Interpretation)

DCF is a way of thinking, not a formula. The essence of valuation is answering "how much money can this company earn over its entire future life cycle," not opening Excel to build a model and adjust parameters. Munger said he had never seen Buffett calculate a DCF, and Buffett himself also said there is no formula for calculating intrinsic value.
Rough estimation—"a 300-jin fat person doesn't need a scale." If you need a calculator to figure out that it is cheap, then it is not cheap enough. Valuation only requires "one glance showing it is obviously cheap"—when Yao Ming walks in, you know he is very tall without a ruler. Rough estimation of course must consider growth; discounting itself is the result after considering growth.
Sealing the position for ten years—using the time dimension to test the depth of understanding. Before buying, ask yourself: if the stock market closed for 10 years, would you still be willing to hold? "Why hold a stock for 10 days if you don't plan to hold it for 10 years" does not mean you really have to hold it for 10 years, but rather tests whether you truly understand this company's future.
PE is a rearview mirror; E/E is more honest. PE's denominator is historical profit and does not reflect debt. General Motors had a PE of only 5 times yet went bankrupt. E/E (Enterprise Value / Earnings = market cap + debt - cash / profit) lets investors see the money truly required to buy the entire company.
Valuation is "kung fu"—it requires years of accumulation. Buffett decided in 5 minutes to invest in Goldman Sachs, backed by 50 years of accumulated understanding. Valuation ability equals the depth of one's understanding of the business; there is no shortcut.

A1 — Applications in the book (Past Application)

Case 1: Yahoo DCF teaching case (sum-of-the-parts valuation method)

Problem: Yahoo's stock price was sluggish, and the market ignored the large amount of assets it held (Yahoo Japan, Alibaba Group shares)
Use of the methodology: Duan Yongping used "rough estimation" sum-of-the-parts: cash 3 + listed assets 6.7 + Yahoo's own business 9.6 (giving it a 12x PE) = about $19.3 per share. Adding the "free" Taobao + Alipay (estimated at 50 billion), Yahoo was obviously cheap. In reality, only 7 billion was spent to buy the entire Yahoo (after deducting listed assets)
Conclusion: "No matter how you look at it, it's worth 30 dollars," while the price at the time was far below that
Result: Duan Yongping bought a large amount of Yahoo, and Alibaba's subsequent listing verified the judgment

Case 2: Apple valuation (guessing profit of 50 billion 5 years later)

Problem: In 2011, Apple's market value was about 300 billion, and many people felt it had "risen too much" and didn't dare buy
Use of the methodology: Don't look at past stock prices or PE; only ask "how much money can Apple make in the future." Judge that Apple may be the first company on Earth to exceed 50 billion in annual profit, so a 500 billion market value is very reasonable. Don't look at Apple's past 5 billion market value; only look at "current earnings, cash on the books, and possible future earnings"
Conclusion: "I think the price I paid is still very cheap"
Result: In 2015, Apple's net profit reached 53.4 billion, validating the rough estimate. Duan Yongping continued to hold and predicted profit would exceed 100 billion within 10 years

Case 3: NetEase (didn't carefully value it but it was obviously cheap at a glance)

Problem: NetEase's stock price fell to less than $1 in 2001, at a trough
Use of the methodology: "When a 300-jin fat person walks in, you know he's fat without a scale." Based on his experience in the gaming industry, Duan Yongping judged that NetEase had a market value of less than $20 million yet had a huge user base and cash flow potential, and he saw at a glance that it was cheap. "When I bought NetEase, I really didn't carefully value it"
Conclusion: No need for precise calculation; if it's obviously cheap, buy
Result: Held for 8-9 years, average purchase price about 0.25 (after stock split), most sold at 30-35, return over 100x

A2 — Trigger Scenarios (Future Trigger) ★

In what situations will users need this skill?

Currently using indicators such as PE/PB to judge whether a stock is expensive, but the more you calculate, the more confused you become——for example, "Is a PE of only 8 very cheap" or "Is PE=50 too expensive"
Trying to build a DCF model to precisely calculate enterprise value, agonizing over the parameter choices for discount rate and growth rate——opened Excel but doesn't know what numbers to fill in
Facing a company whose business you understand, hesitating over whether the current price is worth buying at——"I can understand this company, but I don't know whether this price can be bought now"

Language signals (activate when these appear in the user's words)

"What's this PE multiple, can I buy it"
"I calculated a DCF model/discounted cash flow, and the result is..."
"Is this company cheap/How much is it worth/How to value it"

Distinction from adjacent skills

With good-business(good business identification method): Good business identification judges "is this a good business" (qualitative), while DCF thinking valuation judges "is this price cheap" (pricing). First use good-business to confirm a good company, then use DCF thinking to judge the price.
With opportunity-cost(opportunity cost decision framework): Opportunity cost solves the comparison problem of "buy A or buy B," while DCF thinking solves the absolute judgment problem of "is A itself worth this price." DCF is an input to opportunity cost—you must first know how much A is worth before you can compare it with alternatives.
With capability-circle(circle of competence decision framework): Circle of competence judges "can you understand this company," while DCF thinking requires you to already be within your circle of competence in order to "roughly estimate." Companies you cannot understand cannot be valued, so skip them directly.

E — Executable Steps (Execution)

After the skill is activated, the agent should execute the following steps:

Confirm whether the user truly understands this company's business
- Q: Can you explain in one sentence what this company makes money from? If the stock market closed for 5 years tomorrow, would you still be willing to hold it?
- Completion standard: The user can clearly describe the company's business model and competitive advantage, and can give a "lock it up for ten years" judgment. If not—stop the valuation, go back to the step of understanding the company, or give up on the target.
Use "rough estimation" to make a rough value judgment
- Guide the user to think: roughly how much money can this company make in the next 5-10 years? (Not a precise number, but a range judgment of "worth at least X hundred million")
- Use the Enterprise Value perspective: market cap + debt - cash = the actual money needed to buy the entire company; divide by annual profit—how many years to recoup?
- Test: "If this were not a listed company, and you bought the whole company at this price, would you be willing?"
- Completion standard: Reach a qualitative conclusion of "clearly cheap / about right / clearly expensive," rather than a precise number.
- Stop condition: If the user says "I cannot judge future profits," jump back to Step 1 (indicating insufficient understanding), or suggest giving up on the target.
Compare with opportunity cost to make the final decision
- Compare the judgment result with the best alternative investment opportunity the user has (at least compare with long-term government bond returns)
- Confirm the margin of safety: rough estimated value vs. current price—is there a clear gap? If you need a calculator to figure out that it is cheap, it is not cheap enough.
- Completion standard: The user can answer "At this price, I can buy it and not worry for 5-10 years" or "Clearly not cheap, do not touch it."

B — Boundary ★

Do not use this skill in the following situations

Scenarios requiring precise valuation modeling(such as M&A pricing, private equity transactions)—Duan Yongping's method is a qualitative judgment framework and does not produce precise numerical values. "Anyone who tells you they can calculate a company's value is fooling you."
Short-term trading timing—The time dimension of DCF thinking is 5-10 years or more, and short-term price fluctuations are not within consideration.
Assets that cannot generate cash flow(such as Bitcoin, gold, collectibles)—Duan Yongping explicitly said, "I do not know how to view things without cash flow." Such assets require a completely different analytical framework.
Judging the top of cyclical stocks—When PE is lowest is often when cyclical stock profits are highest. PE may look cheap but is actually extremely expensive. The denominator needs to use average profit across a full cycle rather than current profit.

Failure modes the author warns about in the book

Misled by PE: General Motors' PE was around 5 times for a long time, but its debt was extremely high and it eventually went bankrupt. PE only reflects the past, not the debt structure. PE=8 may mean earnings are about to decline sharply (at the top of a cyclical stock); it looks cheap but is actually extremely expensive.
Precise error: Spending a lot of time adjusting DCF model parameters (discount rate, growth rate); the resulting number looks precise but is actually meaningless. "People who use formulas often get trapped in details and ignore the whole."
Long-term holding of the wrong stock: Locking it up for ten years is not mindless holding—the premise is buying the right company. "Some people originally planned to hold a stock for only 10 days, but because they lost money decided to hold it for 10 years. This kind of decision often makes people lose even more."

The author's blind spots / limitations of the era

The cases focus on consumer (Moutai) and technology (Apple/NetEase), and do not cover cyclical stocks, financials, utilities, and other categories; the valuation of these industries requires supplementing with different analytical dimensions.
As a large investor, Duan Yongping's operating logic (concentration, long-term, no market timing) may not be optimal for small-capital investors, who have time costs.
The assumption that "a good company's long-term stock price will definitely reflect its value" may in reality face the problem that market irrationality can last far longer than expected.

Easily confused adjacent methodologies

Graham-style "cigar butt" investing: Both belong to value investing but have different focuses—Graham focuses on the margin of safety in net asset value, while Duan Yongping focuses on the sustainability of future cash flows. "Old Buffett discovered that the discounted future cash flows of bargain issues were even smaller than those of growth companies"—this was an evolution driven by Munger.
Wall Street analysts' DCF models: Similar in form but fundamentally different—analysts pursue precise calculations and parameter sensitivity analysis, while Duan Yongping pursues the business intuition of "being able to see that it's cheap without a calculator."

Related skills (to be filled in Phase 3)

depends-on: [ownership-mindset, good-business]
contrasts-with: []
composes-with: [opportunity-cost]

Audit information

Verification passed: V1 ✓ / V2 ✓ / V3 ✓
Test pass rate: Pending Phase 4
Distillation time: 2026-04-16