Fair value is the most misunderstood number in investing. This lesson explains exactly how ZEME builds one, and how much weight it can carry.
See it live: Valuation ยท Fundamentals
It is not one model. It is six, blended by company type.
ZEME follows the standard institutional framework, cost of equity from CAPM plus a country risk premium, then a full WACC across the capital structure. What changes is which models are blended and how:
| Company type | Blend |
|---|---|
| Banks / financials | Justified P/B (40%) + Dividend Discount (30%) + Sector P/E (30%) |
| Dividend payers | DCF (30%) + Sector P/E (25%) + DDM (25%) + Graham Number (20%) |
| Growth / other | DCF (40%) + Sector P/E (30%) + Graham (20%) + EV/EBITDA (10%) |
Banks get equity-cost models only. A discounted cash-flow model on a bank is close to meaningless, because for a bank, debt is the raw material rather than a financing choice. Any platform showing you one DCF number for every company on the exchange is not doing this step.
Why a blend rather than the best model
Because each model breaks in a different, known way:
- DCF is exquisitely sensitive to the discount rate and the terminal growth assumption. Small changes in either move the answer enormously.
- Sector P/E inherits whatever mispricing the sector already has. If the whole sector is expensive, it says the stock is fine.
- Graham Number is deliberately conservative and will call most growth companies expensive.
- DDM requires a stable dividend, which most PSX companies do not have.
Blending does not remove these errors. It stops any single one from dominating.
The part that matters most
A fair value is a statement about the company. It is not a statement about the next 30 sessions.
Nothing in this course connects fair value to short-term movement, and that is not an accident. Our entire touch-rate body of work, 73,159 observations, found the drivers of a 20% move in a month were volatility, extension and price level, not valuation. A stock can be 40% undervalued and sit still for two years.
Use fundamentals to decide what you are willing to own. Use the measured behavioural work to decide when. Collapsing those two questions into one number is the most common analytical mistake we see.
Honest limits
- Fair values are only as good as the reported financials behind them, and reporting quality varies widely across this exchange.
- The country risk premium is a judgement input, not a measurement.
- We have not published a study showing ZEME's fair-value gap predicts returns on PSX, because we have not run one to a standard we would defend. Until we do, treat it as a screening and context tool.
Do this now, practical activity
- Open Valuation and pick a stock trading well below its fair value.
- Open the same stock in SuperCharts. How has it behaved over the last year?
- Now check it against the Reading a PSX Chart scoreboard, is it in motion, or quiet?
- Write one sentence answering: would I own this company, and a separate sentence answering, is this the month to buy it?
Two questions, two sentences, deliberately. Most losing trades in this market come from answering the first question and acting as though you had answered the second.
Measured: ZEME blends up to 6 valuation models with weights that change by company type. Against that, our 73,159-observation study found the measured drivers of a +20% move within 30 sessions were volatility, extension and price level, valuation was not among them.