Valuation Principles — what a business is worth

Price is not value

Two quotations open this chapter, and both are worth memorising.

The distinction is the foundation of the whole chapter. Price is available from the stock market and known to all. Value is based on the evaluation and analysis of the valuer at a point in time.

And then the sentence candidates skip and later regret: there is no formula or method to throw a precise number on the valuation of an asset. There are uncertainties in every input. The final output is at best an educated estimate, and only then if adequate due diligence has been done. That is why valuation is often considered an art as well as a science — it needs knowledge, experience and professional judgment.

Why anyone values anything

The workbook lists six reasons:

• Buying a business as part of an investment exercise
• Selling a business as part of an investment exercise
• Mergers and acquisitions
• A general sense of value of the business to its owners
• Fair treatment to different sets of stakeholders in an equity swap
• Accounting, taxation and other regulatory and legal requirements

Whatever the reason, the purpose of valuation is to relate price to value and estimate whether something is fairly priced, over-priced or under-priced. Because the process has limits, valuers typically present multiple scenarios showing what happens to the output when the primary variables change.

Two sources of value

Every asset produces two streams: periodic earnings and a final inflow on sale.

Businesses exist for the same reason: to generate cash flows as earnings, with the potential to realise cash from the sale of tangible and intangible assets if earnings are not sufficient, or die down, and owners collect their money by liquidating.

Lenders think exactly the same way. A lender first checks whether the business can generate the cash flows to meet the obligation. Collateral is not the primary consideration — it is only a fall-back if the cash flow estimates prove wrong. Lenders want to be paid from the cash inflow stream.

And one warning to carry through the chapter: the capability of business assets to pay all liabilities and settle equity holders is always doubtful. Assets may be worth a lot less than they appear on the balance sheet — but the outstanding liabilities must be settled in full.

Three approaches to valuation

(i) Cost based valuation. Value the asset at the cost of creating it. Suitable only for a buyer choosing between buying and making. Most stock market investors have no option to build and run a company themselves, so this approach is generally not suitable for financial investors. Strategic investors intending to carry on the business long term may use it. It often involves technical assessment by engineers and is seldom employed by financial investors.

(ii) Cash flow based valuation (intrinsic valuation). Value the asset by what an investor would be willing to pay for the cash flow it generates — by discounting those cash flows at a rate reflecting the return the investor expects. It splits two ways:

• Risk neutral valuation — adjust the cash flows by the probability of realising them, then discount at the risk free rate. Insurance companies are typically valued this way (embedded value and appraisal value).
• Real world valuation — estimate the most likely cash flow and discount at risk-free rate plus a suitable risk premium for the uncertainty.

(iii) Selling price-based approach (relative valuation). Value the asset from the price of other similar assets, using ratios such as P/E, P/B and EV/EBITDA.

Discounted cash flow, starting from a bond

Take a bond paying 9% interest per annum, redeemed at the end of the 10th year at its face value of ₹1,00,000. Prevailing interest rates for that maturity and credit quality are also 9%.

Its value is the present value of all future cash flows discounted at 9%. Because the coupon and the discount rate are the same, the value is exactly ₹1,00,000. Push the expected return above 9% and the bond is worth less than ₹1,00,000; push it below and the bond is worth more. That is simply present-value arithmetic.

Every asset is priced the same way. Replace the bond with equity and coupons become dividends, redemption value becomes expected sale proceeds. But there is one enormous difference: for a bond, both the quantum of cash flows and their timings are known with certainty; for equity, both are unknown and uncertain.

That is why the workbook says DCF is conceptually the most appropriate approach when three things are known with certainty:

1. The stream of future cash flows
2. The timings of those cash flows
3. The expected rate of return of investors — the discount rate

Once you have those three, it is simple mathematics. In practice, DCF valuations come with significant error of judgment because quantum and timing cannot be measured with certainty, and the models can be very sensitive to some input factors.

The method itself: find the inflows over outflows at different points in time — the Free Cash Flows (FCF) — and bring them to today at an appropriate discount rate. The two principal factors that drive a DCF are estimating the expected cash flows and determining the discount rate, and the value can vary across analysts because of differences in exactly those two.

The three DCF models

Dividend Discount Model (DDM). Discount the company's expected future dividends at the cost of equity. Suitable for companies that pay regular and substantial dividends — so more suitable to matured companies in defensive industries.

Two complications: equities have perpetual life, theoretically, and dividend payments are not contractual. So DDM requires estimates and assumptions.

The Gordon growth model — also called the perpetual growth model — values a dividend-paying company whose dividend grows perpetually at a constant rate:

P = D₁ ÷ (k − g)

where D₁ is the dividend expected at the end of the year, k is the cost of equity, and g the constant growth rate. The model assumes g is lower than k, and that cost of equity is the required rate of return of equity shareholders, under the stringent assumptions of no information asymmetry and perfect capital markets.

Free Cash Flow to Equity (FCFE). DDM has an obvious hole: it cannot be used for companies that do not pay dividends. Even well-performing companies may retain everything for reinvestment — the workbook's example is Alphabet Inc., the parent of Google, which has never paid a dividend.

FCFE values equity by discounting the free cash flow available to equity shareholders instead of the dividends actually paid:

FCFE models are most useful for companies in a high growth phase. But there is a catch: in a high growth phase it is inappropriate to assume a constant growth rate, because the growth is unsustainable in the long run and may even be higher than the cost of capital. So value in two stages:

Value of equity = Present value of FCFE during the high growth phase + Value of the perpetual stream of FCFE after the high growth phase (the terminal value)

You assess how long high growth lasts, estimate FCFE for each of those years, discount them, then use the Gordon growth model with FCFE in place of dividends for the perpetual stream. The terminal value has to be discounted further to its present value — that last step is the one candidates forget.

Free Cash Flow to Firm (FCFF). The problem with FCFE: unless a company has an objective debt policy, net borrowings and repayments cannot be objectively estimated. Guessing them can lead to significant bias in valuation.

FCFF is the free cash flow before taking into consideration any cash flows pertaining to any source of capital:

Analysts often compute it by the indirect method instead:

FCFF = EBIT × (1 − Tax rate) + Depreciation and non-cash charges − Increase (add decrease) in non-cash working capital − Capital expenditure incurred (or add sale of assets)

Besides depreciation, the non-cash charges added back include amortisation of capital expenses and loss on sale of assets. Gains on the sale of assets are deducted from both FCFF and FCFE.

Because FCFF is available to all sources of capital, it is discounted at the Weighted Average Cost of Capital (WACC). That gives the value of the firm. Then:

Value of equity = Firm value − minority interest − preferred share capital − interest-bearing debt

Like FCFE, FCFF for a high-growth company splits into cash flows during the high growth phase plus a terminal value.

Terminal value

Forecasting is required, and a rudimentary forecast can be derived by extrapolating historical growth rates. Analysts refine that with judgement, taking account of the internal determinants of growth — the proportion of earnings ploughed back into the business and the return on equity expected on it (in FCFE models).

The terminal value is usually computed with the perpetual growth model. Note the ceiling: the perpetual growth rate is often capped at the long-term nominal GDP growth rate of the markets in which the company operates, because it is not possible for a business to grow faster than the economy till perpetuity. An analyst may use a lower rate if that seems appropriate.

The alternative: take the expected sale value of the business at the end of the high growth phase, by multiplying EBITDA (or EBIT) at the end of that period by an appropriate EV/EBITDA (or EV/EBIT) multiple drawn from comparable firms. Either way, the terminal value is added to the growth-period cash flows and then discounted to present value.

The discount rate

The discount rate must reflect the risks in the cash flows and therefore the expectations of investors.

• FCFF is discounted at WACC — it considers both debt and equity
• FCFE is discounted at the cost of equity

Cost of debt in most valuation exercises is the prevailing interest rate in the economy for borrowers of comparable credit quality. Cost of equity is the rate of return required by the company's common shareholders.

Cost of equity comes from the Capital Asset Pricing Model (CAPM), which establishes the relationship between risk and expected return. It has three components: the risk-free rate (Rf), the return expected on a stock market portfolio (Rm), and a proxy for the systematic risk of the firm reflecting both its business and financial risks — the equity beta, being the sensitivity of the stock's return to the market's return.

Ke = Rf + β × (Rm − Rf)

(Rm − Rf) is the market risk premium. β × (Rm − Rf) is the equity risk premium to be compensated to the investor over and above the risk-free rate.

WACC = [Ke × Equity ÷ (Equity + Debt)] + [Kd × (1 − Tax) × Debt ÷ (Equity + Debt)] = (Ke × We) + (Kd × (1 − Tx) × Wd)

Note where the tax shield sits: only on the cost of debt, because interest is tax deductible and dividends are not.

The workbook's closing caution on DCF: it may result in erroneous output if sufficient rigour does not go into estimating the cash flows and the discount rate.

Relative valuation

DCF is complicated, given the assumptions that go into estimating future cash flows. Relative valuation takes the other route: instead of an absolute number, compare what we get with what we pay. What we pay is the price. What we get is the earnings and the assets. Compare price with earnings and with assets, and you get not the absolute valuation, but whether something is cheap or expensive.

Dividend yield and price to dividend

Dividend Yield = Dividend per share ÷ Current price of stock

Take a company that has paid a dividend of ₹5 or more for five years:

At ₹50 the business yields 10% on a price-to-dividend of 10 times. Compare with a bond: a 10% coupon at a 30% tax rate gives a post-tax return of about 7%, while the equity is offering 10% — plus potential upside if earnings go up, which debt does not offer because debt is usually redeemed at face value or the original principal.

At ₹200 the yield is 2.5% on a price-to-dividend of 40 times, clearly inferior to the bond. If equity yields are generally higher than bond yields, equity is available cheap — typically true when markets are down. During bull markets equity yields are quite lower than bond yields.

And then the warning. A stock yielding more than its peers and more than the market may be seen as a value pick — it may not necessarily be so. A high dividend payout may indicate limited avenues for expansion and investment, which in turn may limit capital appreciation.

That leads to a term worth learning: companion variables are the relevant and related company fundamental financial metrics that must be checked alongside a market-determined multiple. P/E is related to the growth rate and the ROE of the firm; EV/EBITDA is related to the ROI. They tell you whether a lower valuation multiple is truly an under-priced good pick or a reflection of fundamental weakness.

Earnings yield and P/E

Earning Yield = Earnings per share ÷ Current price of stock

Price to Earnings Ratio = Current price of stock ÷ Earnings per share — the reciprocal of earnings yield.

The P/E ratio indicates the amount of money an investor needs to invest to receive 1 unit of profit. It can be trailing (current market price with historical EPS) or forward (with forecast EPS). Investors pay a higher price for earnings if they expect above average growth, or a turnaround.

All else held constant, a stock with a higher P/E than its peer group and the market P/E is considered expensive, and a relatively low P/E is considered undervalued.

But "all else held constant" is doing heavy work. Companies with higher growth potential or lesser risk should trade at a premium to their peer group; companies with lower growth potential or higher risk should trade at a discount.

One technical point: earnings are for a period, price is at a point. So the period of reference for the EPS must be appropriate. Investors usually assess companies on future potential rather than past performance, so estimated future earnings matter more than past reported EPS — but the reference period for a forecast is somewhat subjective. If the current year's EPS is expected to be unduly low or high, investors may compare on next year's expected earnings. Analysts must exercise judgement on which period to use.

PEG — the growth adjusted P/E

PEG Ratio = (Current price ÷ EPS) ÷ Growth rate = P/E ÷ Growth rate

The term was coined by Peter Lynch. He believed a high P/E can be justified by high growth potential — while warning that a high growth regime may not continue for very long, and investors should be cautious of that. His rule of thumb: if the PEG ratio is less than 1, the business may be treated as undervalued.

The workbook's own illustration:

On P/E alone, A looks the better investment. Factor in growth and B is the better investment. That reversal is exactly what the exam tests.

The workbook adds the sensible caveat: a benchmark based on a rule of thumb may not be appropriate in all circumstances, but investors can compare PEG across companies to find those attractively priced against their competitors.

EV/EBIT and EV/EBITDA

A business can be funded by common equity, preferred share capital and debt. Because common equity has the residual interest, the return on equity is affected by the current capital structure. So EPS — and therefore P/E and PEG — are impacted by capital structure.

A retail investor cannot change the capital structure. A controlling shareholder can. So from an acquirer's perspective, a ratio neutral to capital structure is more suitable, and when a company is valued as a potential acquisition target, EV/EBIT or EV/EBITDA is more appropriate. Both ratios are neutral to capital structure.

Which of the two? In capital-intensive industries, differences in the historical cost of assets and the choice of depreciation method cause major discrepancies — so prefer EV/EBITDA. For other industries, EV/EBIT is preferable.

As with P/E, lower is more attractively valued, all else held constant, and higher growth potential or lesser risk attracts a premium valuation.

EV/Sales

P/E, EV/EBITDA and EV/EBIT cannot be applied if the underlying profit metric is negative. And for companies that have just broken even, profit is far below long-term potential, so those multiples come out too high to be meaningful.

EV/Sales is the answer, because sales can never be negative.

But only conditionally. EV/Sales is suitable only for companies likely to turn profitable and sustain that profitability. If a company is loss-making and a turnaround is not likely in the foreseeable future, the going concern nature of the company becomes doubtful — and such companies have to be valued at liquidation value.

Asset based multiples

Recall the two return ratios: ROE = Net profits ÷ Equity capital or net-worth, and ROCE = EBIT ÷ Total capital employed (Debt + Net-worth). They show returns on the book values of equity, and of equity plus debt.

Price to Book Value

Profit-based ratios measure how much you invest to earn a unit of profit. Price to book measures how much you invest to gain ownership of the assets.

Price/Book = Market capitalisation ÷ Balance sheet value of equity, or equivalently Price per share ÷ Book value per share

P/B is preferred for valuing companies in the financial sector rather than sectors with more tangible assets. The reason is the reliability of the book value numbers: most assets of financial companies are monetary assets, so book value closely reflects fair value. In capital-intensive firms, historical cost accounting means balance sheet values often do not reflect fair value. And in service sectors and technology firms, the most important assets — human capital and self-generated intellectual property — find no place in the balance sheet at all.

Because accounting follows the principle of conservatism, book value of equity is often used to arrive at a conservative value of equity, even where the ratio is not ideal for non-financial companies.

Lower P/B is more attractive, all else held constant — but companies with higher ROE should command a premium, as the equity is more efficiently employed.

The workbook's example:

AFB: BVPS = 9,900 ÷ 50 = ₹198; P/B = 200 ÷ 198 = 1.01 LKH: BVPS = 8,000 ÷ 50 = ₹160; P/B = 175 ÷ 160 = 1.09

So AFB Finance is less expensive on P/B, even though its share price is higher.

One practical caution: compare a company's multiple with the average for the entire industry rather than against only one peer — and be careful with averages. If the industry is fragmented with outlier players who pull the mean up while the majority hover around the median, verify both mean and median multiples before taking a call.

EV to Capital Employed

EV = Value of Equity + Value of Debt − cash and cash equivalents

EV to Capital Employed = Enterprise Value ÷ Capital Employed (Total Equity + Total Debt)

Used together with ROCE, this gives a sense of the return on the capital you actually invest. The workbook's example:

• Net-worth ₹1,00,000, debt ₹1,00,000, market capitalisation ₹5,00,000, cash nil, ROCE 45% per annum
• Capital employed = 1,00,000 + 1,00,000 = ₹2,00,000
• EV = 5,00,000 + 1,00,000 = ₹6,00,000
• EV to Capital Employed = 6,00,000 ÷ 2,00,000 = 3

The business earns 45% on the capital employed, but you are paying 3 times that capital. So your money generates only one third of the ROCE — 15% (45% on ₹2,00,000 is 15% on ₹6,00,000).

Turn it round: if the investor wants a minimum 20% return on capital, he would not pay more than 45 ÷ 20 = 2.25 times capital employed — that is, ₹4,50,000.

Net Asset Value

NAV of equity is the market value of an entity's assets minus the value of its liabilities. It differs from book value or net-worth because it uses the market value of assets, not the book value. It can be expressed for total equity or divided by shares outstanding for NAV per share. Used in extremely asset-oriented businesses such as real estate, shipping and aviation.

Other metrics for other industries

Trading multiples and transaction multiples

Relative valuation is basically intuitive — we do this all the time in our personal lives. Buying an apartment, you find the price of comparable apartments in that locality and use it as your indicative value for negotiation. It is highly useful, quick, and needs limited computations and assumptions.

Its weakness: it reflects the current market mood, which may be quite optimistic or pessimistic. So always use parameters like maximum, minimum and average when working with relative valuations.

Comparables come from two places:

• Trading multiples — from the stock market
• Transaction multiples — from other similar transactions consummated in the recent past

Transaction multiples are more relevant, because they represent the willingness of some entity to actually acquire the asset at that value, and are therefore relatively more authentic than the prices at which shares are traded.

Sum-of-the-parts

Several businesses operate as a cluster or bundle of businesses rather than one — the workbook names ITC and L&T. The best way to value them is to value each business separately and then add the valuations up. That is Sum-Of-The-Parts (SOTP). Each vertical is treated as an independent business, valued on earnings and assets as above, and then simple summation gives the total.

New-age businesses

People wonder about the valuations of e-commerce and technology companies — the workbook names WhatsApp, Zomato, LinkedIn and Facebook. It is honest: it is difficult to put the numbers together to arrive at the valuations at which these transactions are happening, and calls it "our own limitation to understand the value proposition".

In the new-age economy people use parameters such as eyeballs, page reviews, footfall, ARPU and number of users to justify exorbitant valuations. As Buffett would put it, all of these should ultimately translate into profits for owners at some point in time. Where there is no visibility of that, valuations sustain only while there is a storyline, people believe the story, and the next buyer is available — and fall like a pack of cards in the absence of those. The workbook points to the dot-com boom of 2000–2001 as the demonstration.

Is valuation objective?

It may appear so, but it is a very subjective exercise, because the inputs required in every method are quite subjective without any generally accepted standards. Further, valuation is not timeless — it can change dramatically if the circumstances of a business change.

The conclusion is one of the most quotable lines in the workbook: there is no precise estimate of value, and complicated quantitative models need not mean the valuation is precise; it only means a false impression of preciseness.

Nine considerations to carry into practice

1. If the earning power of a business is high, book value of shares could be less important. But if earning power is low, book value becomes very important.
2. As a share reflects part ownership in a business, to value a share you must value the entire business.
3. EV, and not market capitalisation, is the true value of the firm for a private owner.
4. P/E for a leveraged firm may be deceptive — look at debt levels in the business.
5. Look at the consolidated numbers, not just the standalone numbers.
6. Focus on ROE and not EPS — EPS does not account for retained earnings.
7. Leverage improves ROE, but excessive leverage is risky.
8. Differentiate between ROCE and ROE. ROCE reflects the true return on capital; ROE could be manipulated by high leverage.
9. ROCE and ROE should be closely knit. Any wide variation should trigger investigations.