Reading: The Challenges of Relative Valuation

The second key challenge is which multiple(s) to pick. As we have seen in the introduction to this module, financial analysts generally use multiples to either estimate the value of the firm's equity of the enterprise value. The most popular multiples for equity valuation is clearly the Price/Earnings (P/E)-Ratio. To estimate enterprise value, many analysts rely on the EV/EBITDA-ratio. The list below summarizes the valuation multiples that are most commonly used:

 

Popular multiples to estimate Equity value

• Price/Earnings (P/E-Ratio) and its variations (PEG and Relative PE)
• Price/Book equity
• Price/Sales

 

Popular multiples to estimate Enterprise value (EV)

• EV/EBITDA
• EV/EBIT
• EV/Sales
• EV/Book assets
• EV/Cash flow

 

When picking multiples, it makes sense to stick to the following guidelines and recommendations:

• Understand the definition of the multiple
• Understand its economic relation to value
• Check the empirical distribution
• Dealing with negative value indicators

 

Definition of the multiple

It sounds almost too simple, but the first step should always be to make sure that we understand what the multiple actually does and how the various elements are defined.

• Remember that multiples with EV in the numerator will produce an estimate of enterprise value whereas multiples with P in the numerator will produce an estimate of equity value (or stock price).
• In the case of enterprise valuation, remember that EV excludes excess cash.
• Check whether the value driver that is used to normalize the valuation (e.g., EBITDA) reflects the most recent accounting data (so-called trailing multiples) or future expected accounting data (so-called forward multiples). If we work with trailing multiples, we should apply them to our most recent financials. In turn, if we work with forward multiples, we should use our future expected financials to conduct relative valuation. More on this in a later section of this module.
• In most instances, we should also make sure that the numerator and the denominator of the multiple reflect the claims of the same group of stakeholders. What does this mean? 
• In the case of enterprise valuation, we are interested in the value of all financial claims against the firm (debt plus equity). Therefore, the earnings figure we use should also be one that reflects the firm's ability to generate money for all providers of capital (e.g., EBITDA, EBIT). 
• In contrast, for equity valuation, the relevant earnings figure is one that reflects the firm's ability to generate money for its shareholders (net income).

 

Economic relation to value

Second, it is important to understand how and why a particular multiple should be able to explain the market valuation of assets. This is a crucial point. 

As we will show in a later section of the module, many of the popular valuation multiples can be derived from a simple valuation model that is based on the present value of expected future cash flows. For these multiples, it is fairly easy to understand their economic relation to value. For example, the P/E-Ratio can be expressed as a function of the firm's payout policy, future growth, and risk. Firms that are identical with respect to these three dimensions should have the same P/E ratio according to economic theory.

For other multiples, it is less obvious that they really reflect the way markets value assets. For example, in the late 1990s, internet firms were sometimes assessed based on the number of clicks on their website or, even better, their cash burn rate (e.g., how much cash firms spend on their projects).  Arguably, a high number of clicks implies great visibility and an attractive offering, whereas a high cash burn rate suggests that the firms in question have a lot of attractive projects (hence the high investments). However, it is not difficult to come up with alternative interpretations of these multiples. For example, a high cash burn rate could also be typical of firms with inefficient cost management and poorly defined projects---both not very attractive characteristics to investors.

In general, we should avoid multiples for which we cannot easily understand how and why they should correlate with value.

 

Empirical distribution of the multiple

Once we have computed the valuation multiples for our peer group, it makes sense to take a closer look at the individual multiples. The reason is a very simple one: If the comparable firms are really comparable and if the valuation multiple really correlates with value, then the comparable firms should have similar valuation multiples!

As an illustration, consider the two valuation situations that are summarised in the table below:

 

 

In both situations, there are 5 comparable firms and the median valuation multiple of the five comparable firms is 10 (e.g., an EV/EBIT-Ratio of 10). However, a closer look at the data reveals that the two situations might be very different. 

• In the first case, there is no clearly visible pattern. The individual multiples are scattered between 1 and 100. A statement such as "the comparable firms are typically valued at a multiple of approximately 10" does not seem to do justice to this distribution. In all likelihood, Multiple 1 will produce a rather unreliable valuation.
• In the second case, the picture is much clearer. While the multiples also exhibit some variation, it does indeed look as if a multiple of 10 could approximately describe the typical valuation of the comparable firms. Consequently, the resulting valuation should also be more reliable.

 

Dealing with negative value indicators

Another important factor that relates to the empirical distribution of the multiple concerns the comparable firms with negative value indicators. Remember from the valuation example of Ford Motors in the previous section that we had to exclude two of the twelve comparable firms because their value indicator (net income) was negative. Consequently, the resulting median multiple of 9.0 was based on ten rather than twelve comparable companies. 

The problem is that we did not just randomly exclude two companies from the peer group. To the contrary, we excluded the two comparable firms with the poorest performance based on the chosen value indicator. The implication is that the resulting valuation multiple is systematically biased towards better performing firms! Put differently, the multiple of 9.0 is not a representative valuation for all firms in the car manufacturing industry. It only applies to comparatively well-performing firms, i.e., firms that perform well enough to generate positive net income. 

This bias can be a serious valuation challenge in situations where many of the peers have negative value indicators. This is often the case in young and rapidly growing industries or market segments. 

Luckily, the recommended workaround is fairly simple: Work your way up the income statement! 

• Because net income is after all operating expenses, depreciation, interest expenses, taxes, etc. it is the one earnings figure that is most likely to be negative. If many of the firms in question have negative net income, it is therefore not advisable to work with the P/E ratio. 
• EBIT, the firm's operating profit, is before interest expenses and taxes. Consequently, it is less likely to be negative than net income. However, because EBIT reflects a profit measures for all providers of capital, not only the shareholders, it is advisable to use EBIT to normalize enterprise value rather than equity value.
• Firms that operate in an investment-intense environment often face significant depreciation charges, which lower EBIT. If that is the case, and if we believe that different comparable firms have different investment policies, it makes sense to work with EBITDA rather than EBIT and use the EV/EBITDA ratio rather than the EV/EBIT ratio.
• Finally, if many firms have negative EBITDA, we can go all the way to the top line of the income statement and use an EV/Sales multiple. However, in such situations, the result of the valuation will most likely be rather coarse.

 

Example

Let's look at these considerations with a comprehensive example. Suppose we want to value a firm with the following characteristics (values in thousands):

• Net income: 1'000
• EBIT: 5'000
• EBITDA: 8'000
• Net debt: 20'000

We want to estimate both the equity value and the enterprise value. To do so, we have compiled the following information about 5 comparable firms (Firms A to E).

 

 

Based on this information, what's the theoretical value of our company?

We can use the information about the comparables to compute valuation multiples. To do so, remember that the enterprise value (EV) corresponds to the value of equity plus net debt. In the case of Firm A, for example, the enterprise value is 90'000 (= 50'000 + 40'000).

The following table summarizes the various valuation multiples of the comparables. In computing these multiples, we stick to the rule that the numerator and the denominator should reflect the same financial claim. Consequently, we use enterprise value when working with EBITDA and EBIT and equity value when working with net income:

 

 

For example, in the case of Firm A, the multiples are the result of the following calculations:

• EV/EBIT = 90'000/7'500 = 12
• EV/EBITDA = 90'000/11'250 = 8
• P/E = 50'000/500 = 100.

 

We can now use these multiples to estimate the value of our firm. To do so, keep in mind that we have collected the following financial information about our firm:

• Net income: 1'000
• EBIT: 5'000
• EBITDA: 8'000
• Net debt: 20'000

 

This allows us to estimate the following enterprise values:

• EV/EBIT multiple: EV = EBIT × Median EV/EBIT multiple = 5'000 × 10 = 50'000
• EV/EBITDA multiple: EV = EBITDA × Median EV/EBITDA multiple = 8'000 × 6 = 48'000.

 

In words, our valuation implies an enterprise value of roughly 50'000 for our firms. Given a net debt of 20'000, the theoretical equity value is, therefore, approximately 30'000.

 

Using the P/E ratio is more problematic in the specific valuation situation. 

• The reason is that two out of five peers have negative net income.
•  Moreover, a look at the last column in preceding table shows that the remaining valuation multiples exhibit significant variation. They range from 14.4 to 100.

 

The mathematical median value of the comparables' P/E ratio is 56. Given a net income of 1'000, the theoretical equity value of our firm would be 56'000; and given a net debt of 20'000, the theoretical enterprise value would be 76'000. However, given the concerns discussed above, this valuation should be taken with a grain of salt.