Intrinsic value is what a business is worth based on the cash it will generate, independent of what its stock traded for today.
The standard way to estimate it is a discounted cash flow model. The arithmetic is something a spreadsheet does in a second. The judgment is hard, and the judgment is the whole thing.
The formula
Written out, you project the business's free cash flow for five to ten years, estimate a terminal value capturing everything after that, discount all of it back to today at a rate reflecting the risk, and add it up.
The four steps
One: estimate future free cash flows. Project five to ten years, based on historical performance, management guidance, industry growth rates, and your own read on what the business can deliver. The longer the projection, the more uncertain it gets.
Two: calculate the terminal value. Everything beyond the explicit forecast has to be captured somehow. The standard approach is the perpetuity growth model.
Where g is the assumed long-term growth rate, usually 2% to 3% and tied to long-run GDP growth, and r is the discount rate. Some analysts use an exit multiple instead. Both methods work. Both can be manipulated to produce whatever number you wanted in the first place, so be honest about which assumptions you chose and why.
Three: discount everything back to present value. Use a rate reflecting the risk of those cash flows. For most businesses that's the weighted average cost of capital, blending the cost of equity and the cost of debt by the company's capital structure. Riskier businesses get higher rates.
Four: sum it and convert to a per-share number. Add the discounted cash flows to the discounted terminal value and you have the intrinsic value of the business. Divide by shares outstanding for intrinsic value per share. Compare to the price.
A complete worked example
Here's a full DCF on a company I'll call ABC, start to finish, so you can follow every number.
The inputs: projected free cash flows of $50M, $60M, $70M, $80M and $90M over five years. Terminal growth 2%. Risk-free rate 3%. Beta 1.2. Expected market return 10%. Cost of debt 5%. Tax rate 25%. Market value of equity $1 billion, market value of debt $500 million.
Step A, cost of equity using CAPM. 3% + 1.2 × (10% − 3%) = 3% + 8.4% = 11.4%
Step B, the weighted average cost of capital. (1,000/1,500 × 11.4%) + (500/1,500 × 5% × 0.75) = 7.6% + 1.25% = 8.8%
Step C, discount the projected cash flows at 8.8%.
| Year | Free cash flow | Discount factor | Present value |
|---|---|---|---|
| 1 | $50M | 0.9191 | $45.96M |
| 2 | $60M | 0.8445 | $50.67M |
| 3 | $70M | 0.7757 | $54.30M |
| 4 | $80M | 0.7124 | $56.99M |
| 5 | $90M | 0.6543 | $58.89M |
Step D, the terminal value. ($90M × 1.02) ÷ (8.8% − 2%) = $91.8M ÷ 6.8% = $1,350M. Discounted to today at 0.6543, that's roughly $883M.
Step E, sum it. $45.96M + $50.67M + $54.30M + $56.99M + $58.89M + $883M, or about $1,150M of intrinsic value. Small differences from rounding are normal.
Step F, compare to the market. Current market value is equity plus debt, $1,000M + $500M = $1,500M. Intrinsic value of $1,150M sits below that, which says ABC looks overvalued on these assumptions.
Note where the value actually came from. The terminal value is $883M of a $1,150M answer. More than three quarters of this valuation rests on a single assumption about what happens after year five.
Why a single DCF number is a lie
The output is extraordinarily sensitive to the inputs, and most people never see how sensitive. Watch the same company revalued with modest changes.
| Scenario | WACC | Terminal growth | Intrinsic value |
|---|---|---|---|
| Base case | 8.8% | 2% | ~$1,150M |
| Higher WACC | 10% | 2% | ~$987M |
| Lower WACC | 7% | 2% | ~$1,593M |
| Higher growth | 8.8% | 3% | ~$1,470M |
| Lower growth | 8.8% | 1% | ~$1,024M |
Move the discount rate from 7% to 10% and the valuation swings from $1,593M to $987M. That's a 38% difference from an assumption neither of us can know. The terminal growth rate has similar leverage.
A single DCF number creates false precision. A range across reasonable assumptions tells you the truth: the valuation depends heavily on what you assume about the future, and the future is uncertain.
So always run the sensitivity. Build the model, then rerun it with the discount rate 50 and 100 basis points higher and lower, and with terminal growth a point above and below your base case. If the valuation holds up across that range, your conclusion is robust. If small changes produce wildly different answers, the result is fragile and shouldn't carry a large position.
What the DCF is actually for
This is the part that gets missed. A DCF's greatest strength isn't the number at the bottom. It's that it forces you to state what has to happen for the price to make sense.
If you believe a company is worth $100 a share, the model tells you why. Maybe revenue has to grow 15% for five years. Maybe operating margin has to climb from 18% to 24%. Maybe capital spending has to stay modest and the cost of capital has to stay low.
Now those assumptions are visible, and anyone including you can challenge them. That's the value. It's also exactly why a DCF is dangerous in the wrong hands: a spreadsheet will produce almost any valuation you like if you feed it the inputs that get you there.
Where it stops working
DCF needs reasonably predictable cash flows. A mature business with a stable competitive position is a good candidate. A pre-revenue company, a deep cyclical at a cycle turn, or a business facing a technology shift that could reset its economics entirely are all cases where the projection is a guess wearing a decimal point.
Banks and insurers need different treatment too, since their balance sheets work differently and standard free cash flow doesn't mean the same thing.
For those, relative valuation against peers or an asset-based approach will serve you better than forcing a DCF and believing the output.
What to do with this
Take a company you own with a reasonably steady business. Pull five years of free cash flow from its filings. Grow it at whatever rate you honestly believe, apply a discount rate somewhere between 8% and 10%, add a terminal value at 2% growth, and total it up.
Then do the part that matters: run it again at a discount rate two points higher.
If the two answers bracket the current price, the market is roughly where you are and there's no edge in the trade. If both answers sit far above or far below the price, you've found something worth investigating. That gap, not the point estimate, is the whole output of this exercise.
Valuation is the heart of Book One.
Book One covers the three financial statements, competitive position, management, industry, the full range of valuation methods, and benchmarking against peers, with addenda devoted to free cash flow, discounted cash flow in depth, and calculating a target price.