The Limits of DCF Models
What discounted cash flow models can't do, the ways they mislead even careful users, and how to use them well: as a tool for thinking, not a machine for answers.
The short version
A DCF is only as good as its assumptions, and the most important ones, long-run growth and the discount rate, are the hardest to know. Small differences produce very different values, so a model can be tuned to justify almost any price. DCFs also struggle with young companies, businesses that are hard to predict, and banks and insurers, whose cash flows work differently. Used well, a DCF is a way to make assumptions explicit, find out what a price implies, and focus research on what matters most. Combine it with multiples and a margin of safety rather than trusting it alone.
By the end of this lesson, you'll be able to
- Explain the main limitations of DCF models
- Recognize how a model can be tuned to justify a chosen price
- Use a DCF as a thinking tool alongside other methods
Why this matters
A finished DCF looks authoritative: rows of numbers ending in a value per share. That appearance is part of the danger. This module has shown how much a valuation depends on its assumptions. This lesson pulls those lessons together into the main limits of DCF models, and how to use them well despite them.
The main limits
- Garbage in, garbage out. The model only rearranges your assumptions. If the growth forecast is wrong, so is the value, however careful the math.
- The distant future dominates. Most of the value usually sits in the terminal value, the part that's hardest to know.
- Small changes, big swings. A point on the discount rate or long-run growth can move the answer 10% to 20%, as Terminal Value and Sensitivity showed.
- Some businesses don't fit. Young companies with no positive cash flow, businesses that change constantly, and banks and insurers, whose cash flows work differently, all need other approaches.
Warren Buffett has written about the second-to-last point directly. In his 1992 letter, he explained that a bond's coupon and maturity define its future cash flows, while for a stock the analyst must estimate the future "coupons" themselves. He added that Berkshire sticks to businesses it believes it understands, which means relatively simple and stable ones, because for complex or constantly changing businesses it doesn't consider itself able to predict future cash flows.1
A model can justify almost any price
Two analysts value STU Corp, each with assumptions that sound reasonable. Both are hypothetical.
| Analyst A | Analyst B | |
|---|---|---|
| Growth, years 1 to 5 | 12% | 8% |
| Discount rate | 9.0% | 10.5% |
| Long-run growth | 3.5% | 2.5% |
| Value per share | $59.36 | $35.42 |
Neither set of assumptions is outlandish, yet one value is 68% higher than the other. Against a $50 price, Analyst A sees a bargain and Analyst B sees a stock to avoid. If you already want to buy, it's easy to drift toward Analyst A's assumptions without noticing. That's confirmation bias, from Fact, Interpretation, and Confirmation Bias, wearing a spreadsheet.
The rest of this lesson is for members
Still to come in this lesson: How to use a DCF well, the key takeaways, and 2 quiz questions with explanations.
Membership opens every lesson in the curriculum, the capstone projects, the interactive tools, and the members-only channels in the Hyperion Entities Discord. The first lesson of every module stays free.
Go deeper
- Reverse DCFValuation and Modeling (depth track)
- Valuation Under UncertaintyValuation and Modeling (depth track)
- High-Growth and Negative-Earnings CompaniesValuation and Modeling (depth track)
Notes and sources
- Berkshire Hathaway Inc.. Warren E. Buffett, 1992 Letter to Shareholders. Checked 2026-10-01. Back to text