Price is what you pay; value is what you get. Estimate value from cash flows, or compare with similar companies.
Price comes from the stock market and is known to all. Value is the valuer's own estimate at a point in time. No formula gives a precise number, because the inputs are uncertain. At best, value is an educated estimate. That is why valuation is called both an art and a science: it needs knowledge, experience and professional judgement.
Buying a business as an investment
Selling a business as an investment
Mergers and acquisitions
Owners' general sense of what it is worth
Fair treatment of stakeholders in an equity swap
Accounting, tax, regulatory and legal needs
Whatever the reason, the purpose is to relate price to value: is it fairly priced, over-priced or under-priced? Because of the uncertainty, valuers usually present several scenarios showing how value changes with the main inputs.
Buffett, quoted in the book: "There are only two sources of value in a business: Earnings and Assets." Every asset produces periodic earnings plus a final inflow when it is sold.
| Asset | Periodic earnings | One-time inflow |
|---|---|---|
| Bond | Coupons | Redemption or sale |
| Equity share | Dividends | Sale of the share |
| Real estate | Rent | Appreciated value on sale |
| Business | Earnings (cash flows) | Sale of tangible and intangible assets, if earnings fall short |
Value = cost to create the asset. Suits only a buyer choosing between buying and making. Most stock market investors cannot build the company themselves, so it is generally not suitable for financial investors. Strategic, long-term investors may use it. Often needs technical assessment by engineers.
Value = what an investor would pay for the asset's cash flows, discounted at the return they expect.
Value from the prices of similar assets, using ratios such as P/E, P/B, EV/EBITDA.
Start with a bond. It pays 9% a year and repays its face value of Rs 1,00,000 after 10 years. Investors also expect 9% for this maturity and credit quality. Its value today is the present value of all its cash flows at 9%, which comes to exactly the face value. If investors expect more than 9%, the value is below face value; if less, above.
Every asset is valued the same way. For equity, dividends replace coupons and the sale price replaces redemption. The difference: a bond's cash flows and timing are known; equity's are uncertain. So DCF valuations of businesses carry significant judgement error and can be very sensitive to some inputs.
The stream of free cash flows (inflows over outflows)
When each cash flow arrives
The return investors expect
What an investor would pay today
Analysts' DCF values differ mainly because they estimate the cash flows and the discount rate differently.
| Model | What is discounted | Discount rate | Best for |
|---|---|---|---|
| Dividend discount model (DDM) | Expected future dividends | Cost of equity | Companies paying regular, substantial dividends: mature companies in defensive industries |
| Free cash flow to equity (FCFE) | Cash flow available to equity shareholders, instead of actual dividends | Cost of equity | Companies that pay little or no dividend, and companies in high growth. Book example: Alphabet (Google's parent) has never paid a dividend |
| Free cash flow to firm (FCFF) | Cash flow before any payments to any source of capital | WACC | When the company has no objective debt policy, so future borrowing cannot be estimated for FCFE |
Also called the perpetual growth model. It values a company whose dividend is expected to grow forever at a constant rate.
D1 = dividend expected at the end of the year. k = cost of equity. g = constant growth rate, which must be lower than k.
Operating cash flow
− Capital expenditure
− Interest payments
± Net borrowings (repayments)
= Free cash flow to equity
Operating cash flow
− Capital expenditure
− Tax benefit on interest payments
= Free cash flow to the firm
EBIT × (1 − tax rate)
+ Depreciation and non-cash charges
− Increase (+ decrease) in non-cash working capital
− Capital expenditure (+ sale of assets)
Non-cash charges added back include amortisation and loss on sale of assets. A gain on sale of assets is deducted. (In the book the direct FCFF list ends with the label "Free cash flow to equity"; the section is about FCFF.)
A fast-growing company's growth may be unsustainable and even above its cost of capital, so a single constant growth rate does not fit. Value it in two stages:
Estimate how long high growth lasts. Forecast each year's free cash flow and discount it
Gordon model on the cash flow after high growth, or an exit value: EBITDA (or EBIT) at the end × EV/EBITDA (or EV/EBIT) of comparable firms
It is a future amount, so bring it back to present value too
Value of equity (FCFE) or of the firm (FCFF)
To get enterprise value, the book deducts surplus cash, cash equivalents and short-term investments not needed for current operations from the firm value.
It should reflect the risk in the cash flows. FCFF is discounted at WACC (debt and equity). FCFE is discounted at the cost of equity. Cost of debt is usually the prevailing interest rate for borrowers of comparable credit quality. Cost of equity is the return the company's shareholders require.
Ke = Rf + β × (Rm − Rf)
WACC = Ke × We + Kd × (1 − Tax) × Wd
Rates
Capital mix and dividend
DCF is complicated and rests on many assumptions. Relative valuation compares what we pay (price) with what we get (earnings or assets). It does not give an absolute value, but shows whether something looks cheap or expensive, which supports a buy, sell or hold call.
The book's example: a company has paid Rs 5 or more in dividend every year for five years.
| Price (Rs) | 50 | 100 | 150 | 200 |
|---|---|---|---|---|
| Dividend (Rs) | 5 | 5 | 5 | 5 |
| Dividend yield | 10.00% | 5.00% | 3.33% | 2.50% |
| Price / dividend | 10 | 20 | 30 | 40 |
Price to dividend = what the market pays for one rupee of dividend.
Coined by Peter Lynch. A high P/E can be justified by high growth, but high growth may not last. Lynch: PEG below 1 may be treated as undervalued. Use rules of thumb with care; comparing PEGs across competitors is more useful.
| A Ltd | B Ltd | |
|---|---|---|
| Price | Rs 120 | Rs 140 |
| P/E | 12x | 14x |
| Expected growth | 10% a year | 15% a year |
| PEG | 12 ÷ 10 = 1.2x | 14 ÷ 15 = 0.93x |
| Verdict | Cheaper on P/E | Better value once growth is counted |
EPS, and so P/E and PEG, depend on the company's capital structure (equity has the residual claim). A retail investor cannot change that; a controlling shareholder can. So from an acquirer's view, or for a takeover target, use ratios neutral to capital structure: EV/EBIT or EV/EBITDA (see 3.1.7 for EV).
Preferred for capital-intensive industries, where historical asset costs and depreciation choices distort EBIT.
Preferred for other industries.
Lower ratios look more attractive, all else equal; higher growth or lower risk earns a premium.
P/E, EV/EBITDA and EV/EBIT cannot be used when profit is negative, and are too high to mean much when a company has only just broken even. Sales are never negative, so EV/Sales works. But only if the company is likely to turn and stay profitable. If no turnaround is likely, its "going concern" status is doubtful and it should be valued at liquidation value.
Here assets replace earnings. Two ratios from earlier chapters: ROE = net profit ÷ net worth (return on the book value of equity); ROCE = EBIT ÷ capital employed (debt + net worth).
How much an investor pays for ownership of one unit of the company's net assets.
Preferred here. Their assets are mostly monetary, so book value is close to fair value.
Historical cost accounting means book values often do not reflect fair value.
Their key assets, people and self-generated intellectual property, are not on the balance sheet.
Because accounting is conservative, book value is often used for a conservative value of equity. Lower P/B looks more attractive, all else equal, but a higher ROE deserves a premium.
| AFB Finance | LKH Finance | |
|---|---|---|
| Share capital | 500 | 500 |
| Share premium | 3,200 | 2,100 |
| Reserves and surplus | 6,200 | 5,400 |
| Total equity | 9,900 | 8,000 |
| Book value per share | 9,900 ÷ 50 = Rs 198 | 8,000 ÷ 50 = Rs 160 |
| Market price | Rs 200 | Rs 175 |
| P/B | 1.01 | 1.09 |
On P/B alone, AFB Finance is less expensive, even though its share price is higher.
Used together with ROCE, it shows the return an investor actually earns on the price paid.
Net worth 1,00,000 + debt 1,00,000 = 2,00,000
Market cap 5,00,000 + debt 1,00,000 − cash 0 = 6,00,000
ROCE of 45% on 2,00,000 is only 15% on 6,00,000
Pay at most 45 ÷ 20 = 2.25x capital employed, that is EV of Rs 4,50,000
NAV = market value of assets minus liabilities. Unlike book value, it uses market values. It can be stated in total or per share. Used for very asset-heavy businesses such as real estate, shipping and aviation.
Embedded value: present value of expected net future cash flows (probability-adjusted) from policies now in force.
Adjusted book value: fair value of assets minus fair value of liabilities. Includes off-balance sheet items.
Operating metrics instead of financial ones: a closed steel plant by its production capacity; an e-commerce start-up by users, transactions or transaction value.
So use the maximum, minimum and average of the multiples, look at each company's own history over several years, and compare across peers and the industry.
From prices in the stock market.
From similar deals completed recently. More relevant and more authentic: someone actually agreed to buy at that value.
Conglomerates such as ITC and L&T run many businesses under one umbrella. Value each business separately, as an independent business, using the methods above, then add them up. The book calls this sum-of-the-parts (SOTP or SOP) valuation.
Valuations of companies like WhatsApp, Zomato, LinkedIn and Facebook are hard to explain with numbers. Such deals use new measures: eyeballs, page views, footfall, ARPU, number of users. As Buffett would say, these must eventually become profits for owners. Without that, valuations last only as long as the story and the next buyer, and can fall like a pack of cards, as in the dot-com boom of 2000 to 2001.
It may look objective, but it is very subjective: the inputs have no generally accepted standards. It is also not timeless; it can change dramatically when the business changes. Complicated models do not make a value precise. They only give a false impression of preciseness.
Tap the card to flip. On a keyboard: space flips, right arrow = got it, left arrow = again.
Scoring works like the real exam: +1 for a right answer, minus 0.25 for a wrong one, 0 for a skip. Skipping is allowed.