Valuing with multiples: comparables and forecasted fundamentals

What is relative valuation?

In relative valuation , an analyst compares a market measure (the share price, market capitalization or enterprise value) with a fundamental from the income statement, balance sheet or cash flow statement, and judges the share against similar companies. Absolute approaches, such as discounted cash flow (DCF) models, estimate the value directly instead.

The logic is the principle of no arbitrage : similar assets should trade at similar prices. No two firms are identical, but if investors see two firms as close substitutes, their multiples should be close too.

What people do with the verdict matters for the exam. A share whose multiple is well above its peers’ may be overvalued, so investors cut their exposure or sell short. One well below may be undervalued, so investors buy and wait for the gap to close. Issuers act too: they sell shares they see as overvalued and buy back those they see as undervalued. A gap can also have good reasons, such as faster growth or lower risk, which is why the analysis does not stop at the comparison.

Multiples are simple and need fewer assumptions than a DCF model, but every multiple hides assumptions about growth, risk and payout. Learning Module 5 introduced multiples and relative valuation.

What are the five steps of the method of comparables?

Memorize

The method of comparables compares a firm’s multiple with those of companies that have similar characteristics, to judge its value against its market price. It is a five-step relative valuation process :

1. Choose a group of companies with similar characteristics: the peer group , ideally with similar cash flow, risk and growth.
2. Select an appropriate multiple.
3. Calculate, compare and interpret the multiples of the company and its peers.
4. Decide whether the company is overvalued, undervalued or fairly valued against its peers, using past, current or expected future figures.
5. Use one or more multiples to support a DCF valuation, or to set a value range against the share’s actual trading range.

Relative valuation by comparablessimilar assets should trade at similar prices

• 1. Choose peers Memorize similar cash flow, risk and growth
• 2. Select a multiple Memorize numerator and denominator for the same claimants
• 3. Calculate and compare Memorize median or mean; outliers; not meaningful values
• 4. Judge the value Memorize over-, under- or fairly valued, on past, current or future figures
• 5. Apply Memorize support a discounted cash flow value or set a value range

Exam tip Peers, multiple, compare, verdict, apply.

The rest of this module follows the steps. Lesson 4 is about step 1, Lessons 2 and 3 are about step 2, and Lesson 5 is about the time basis in step 4.

How do you choose a multiple? Match the numerator to the denominator

A multiple divides a value to claimants (the numerator) by a fundamental that belongs to the same claimants (the denominator). The rule is consistency of numerator and denominator : a price or market capitalization numerator needs a denominator that belongs to shareholders, such as net income, dividends, free cash flow to equity (FCFE) or book value of equity. Enterprise value needs one that belongs to lenders and shareholders together, such as revenue, earnings before interest and taxes (EBIT), earnings before interest, taxes, depreciation, and amortization (EBITDA), free cash flow to the firm (FCFF) or book value of capital.

In the table, P/E is price-to-earnings, P/S price-to-sales, P/B price-to-book and EV enterprise value. Lessons 2 and 3 define each multiple.

Statement	Shareholders (price or market capitalization)	All capital providers (enterprise value)
Income statement	P/E, P/S	EV/EBITDA, EV/EBIT, EV/sales
Balance sheet	P/B	firm value or EV to book value of capital
Cash flow statement	dividend yield, market capitalization to FCFE	EV to FCFF

Define and calculate the multiple the same way for the company and every peer. Watch for different accounting methods, reporting periods, one-time items and timing. A dividend multiple suits firms with stable cash dividends and a steady payout ratio; comparing payers with non-payers on it is poor practice.

Timing deserves care. Suppose three building-products makers end their latest quarters in March, February and January. If each P/B uses the price on its own quarter-end date, market moves between those dates distort the comparison, so use prices from the same date. The book values still come from different dates (one firm may have bought back shares since), and that cannot always be fixed.

How do you compare a company with its peers? Mean, median and outliers

Find where the company’s multiple sits in the peer distribution. The central tendency of peer multiples , the peer mean or median, is taken as the fair multiple for the group. Peers far from the middle are more likely to be mispriced or wrongly included, so they are the outliers to examine.

Look at the spread and the shape, not only the middle. A skewed distribution of multiples has a few very high values that pull the mean well above the median (positive skew). That is the heart of mean versus median : with positive skew the median is the better measure of the center. Decide on a consistent rule for outliers, and leave out peers whose multiple is one of the not meaningful multiples , such as a negative P/E after a net loss.

The value from a peer multiple is the multiple times the company’s own fundamental. Applied to EPS it gives a value per share. Applied to total net income it gives a market capitalization (net income of EUR 52 million times 14.0 is EUR 728 million).

$$V = \text{peer multiple} \times \text{company fundamental}$$

Mean, median and skew are covered in Quantitative Methods: see mean, median, mode and outliers.

What is the method of forecasted fundamentals?

A justified multiple is an estimate of the fair value of a multiple, to compare with the actual multiple. Under the method of comparables it is the peer average or median. Under the method of forecasted fundamentals it comes from the drivers of value: profitability, growth and the reinvestment of earnings.

A trailing multiple uses the prior 12 months of the denominator. A forward multiple uses the next 12 months. Lesson 5 goes into the time bases.

Start from the constant growth model of Learning Module 6: $P_t = \dfrac{Div_{t+1}}{r_e - g}$, where $r_e$ is the cost of equity and $g$ the long-run growth rate. Divide both sides by book value of equity (BVE) per share, write next year’s dividend as forward EPS times the payout ratio, and use return on equity (ROE), $ROE = EPS_{t+1} \div BVE_t$. The result is the justified price-to-book ratio .

$$\frac{P_t}{BVE_t} = \frac{ROE \times \text{Payout ratio}}{r_e - g} = \frac{ROE - g}{r_e - g}$$

The second form uses the sustainable growth rate , the growth a firm can fund by retaining earnings at its return on equity.

$$g = (1 - \text{Payout ratio}) \times ROE$$

P/B depends on the spread between ROE and the cost of equity. An ROE above $r_e$ gives a P/B above 1. For two firms with the same spread, the faster grower deserves the higher multiple.

The same idea works with more complex models than constant growth. The key step is always to name the value drivers of the multiple you use. Lesson 2 does it for P/E, P/S and the dividend yield, and Lesson 3 for EV/EBITDA.

How do you combine the two methods? Adjustment and regression

Compare multiples and also correct for the differences in fundamentals that make them differ. There are two ways:

1. The adjustment approach modifies a multiple by a fundamental variable so it compares more directly.
2. The regression approach estimates the multiple statistically after controlling for one or more company characteristics, for example with a multiple regression across the peer group.

Justified multiplean estimate of the fair value of a multiple

• Method of comparables Memorize peer mean or median
• Method of forecasted fundamentals Memorize built from profitability, growth and reinvestment
• Both combined correct the peers for differences in fundamentals
  • Adjustment approach Use it for example the price/earnings-to-growth ratio
  • Regression approach Know the idea predicted multiple from the peers' fundamentals

Expected growth is a main reason P/Es differ. The price/earnings-to-growth (PEG) ratio divides the P/E by expected EPS growth. Enter growth as a percentage number, so 20% growth is 20. As with P/E, a higher PEG points to a share more likely overvalued. It assumes P/E rises in a straight line with growth.

$$\text{PEG} = \frac{\text{P/E}}{\text{Expected EPS growth rate}}$$

With more than one driver, regress the peers’ multiples on their fundamentals, for example P/B on ROE and growth. The intercept $a$ and the slopes $b_1$ and $b_2$ are estimated once for the whole peer group. Only $ROE_i$, $g_i$ and the error $e_i$ vary from firm to firm. The error is the gap between firm $i$‘s actual multiple and its predicted one.

$$\text{P/B}_i = a + b_1 \times ROE_i + b_2 \times g_i + e_i$$

Predicted above actual: the share is more likely undervalued. Predicted below actual: more likely overvalued. A large gap: more likely mispriced.

The regression uses the same drivers as the forecasted-fundamentals formula but takes its numbers from the peers, so it builds in the method of comparables without relying on the constant growth model. Like the PEG ratio it assumes straight-line relationships, and real ones may not be. A small sample makes any conclusion fragile. Regression itself is taught in Quantitative Methods: see the least squares regression line.

What is the last step? Applying the multiple

Apply the multiple directly, multiplying the justified multiple by the company’s fundamental to get a value today. Or apply it indirectly, using multiples in a terminal value to set the value at the end of a discounted cash flow forecast.

In the indirect case, the intrinsic value today is the present value of the forecast cash flows plus the present value of a terminal value, where the terminal value is the final year’s fundamental times an appropriate multiple. Lesson 5 works an example.

The right peers for a terminal value are the firms the company will resemble then, often larger and slower-growing, not today’s peers.

Go deeper Why 'no two firms are identical' does not end the argument

No two firms are the same, true. But the differences that matter (growth, risk, payout) can be measured. The adjustment and regression approaches correct for them, so comparisons stay useful. What breaks the method is a peer group whose fundamentals the analyst cannot explain.