# Use of Relative Valuation and Selection of Peer Group

### The Use of Relative Valuation

Relative valuation can be used to value both private and listed companies.

Case 1: A private company may not have a share price in the market. In this case, you can select comparable listed companies to value it using their multiples as follows.

Case 2: Upon the relative valuation methodology derived from DCF, you can estimate the value of a target company using the fundamental factors. That will be explained later.

Case 3: You can judge whether the stock prices of listed companies are undervalued or overvalued by comparing the multiples formed in the stock market to the intrinsic ones. You can also use the multiples obtained from regression analysis to compare among listed companies. That will also be explained later.

### Four Steps to Verify Appropriateness of Relative Valuation Application

The application of relative valuation is simple. However, you can make errors easily in the calculation process. Selecting comparable companies and then finding meaningful multiples are not easy. You can use four methods to use relative valuation appropriately.

1. #### Definition check for consistency

The first step is to ensure that you consistently defined the multiples of the selected companies by checking their numerators and denominators.

##### Step 1: Consistency check

The numerator and denominator of relative valuation comprise as follows.

As shown in the table above, if the numerator is the market value of equity, the denominator must also come with a financial figure related to equity. If the numerator is an enterprise value, the denominator must also accompany a figure associated with the firm value. For example, when dividing the equity value by EBITDA, it may provide an awkward result. That is because EBITDA includes interest that is the cash flow attributable to debtholders, whereas the denominator only considers shareholders’ value.

##### Step 2: Uniformity check

Next, we need to take a closer approach and see if the determinants used in the numerator and denominator are applied uniformly to all comparable companies.

For example, the PER formula is ‘PER = Share price / EPS’. Here, you need to check whether the share price is of the latest one day for all of them or the average of the past six months or one year.

The same is in the EPS. That should be from the previous fiscal year(Current PE) for all of them, or the last fourth quarter(Trailing PE), or the next year’s forecast(Forward PE). It is also necessary to check whether all the companies include non-recurring items equally in the EPS or exclude. One more thing. You should see whether the share number for calculating EPS is the number of current floating shares or the diluted number of shares that are assumed to have all options exercised.

If the above criteria are not applied uniformly, errors may occur among the multiples. For example, suppose that Company A uses forward PE, and Company B uses trailing PE. If both are a growth company, A’s PE can be relatively low, while B’s can be high. That is because the EPSs in the numerator increase year by year as they grow so that A’s forward PE may seem lower than B’s trailing PE.

1. #### Sense of the distribution of multiples across industries

It is good to have a sense of how the average multiple of the industry you want to analyse is and which industries have high and low multiples as well. Then, you can see which stocks or sectors are undervalued, overvalued or appropriate. You can also recognise outliers and see how they affect the multiple you are analysing.

##### Step 1: Distribution test

By arranging the multiples of companies in the industry or sector you are interested in and looking at the distribution, you can get the average multiple. You could also recognise which companies are undervalued or overvalued. If you look at the distribution of not only the target industry but also the related ones or all ones over time, you could see whether the target industry is undervalued or overvalued.

##### Step 2: Average and outlier check

When you find the average multiple of the peer group, if the multiple of a particular company is too high, you may need to remove it from the group. For such an outlier, the stock price might be temporarily too high, or the financial figure could fall significantly, increasing the multiple.

##### Step 3: Biases in estimating multiples

As for PER, you may exclude companies with losses in calculating the average PER of the sector. In this case, the average PER will get biasedly high because the number includes only the companies in the black, not in the red. The solution to this problem is to use the total amounts. You may put the total market cap of all companies in the numerator and the sum of all profits and losses in the denominator. That will take into account all surplus and deficit companies that make up the industry.

1. #### Fundamental Analysis

It is conducive to analyse what fundamental factors determine each multiple and see how the multiple changes as each factor changes.

##### Step 1: Identify determinants

All multiples are affected by three variables: risk, growth rate, and cash flow. Higher growth rates, lower risk, and higher cash flows form higher multiples.

#### Dividend Discount Model: Share price = P0 = DPS1 / (Ke-gn)

Theoretically, the fundamental factors that determine each multiple can be derived from the dividend discount model.

#### PSR = P0/S0= (1+gn) / (Ke-gn)

These intrinsic multiples can have meaningful results when compared to the multiples from the stock market. For example, if, in the same industry, A has a PER of 7, and B has 12, we can determine that B is overvalued. However, it could be wrong because the market may acknowledge that B’s growth rate is higher than A’s. For PBR, suppose A has 0.8, and B has 1.4. We cannot judge that B is unconditionally overvalued because B’s ROE may be higher than A’s.

##### Step 2: Analysis of the relationship between determinants and multiples

It is useful to understand how much a multiple changes as a fundamental factor moves. For example, it is to see how much PER increases as the growth rate increases by 1%, which is a sensitivity analysis. That is possible when you calculate an intrinsic multiple with the fundamental factors.

1. #### application test

You select comparable companies to value a target company. The easiest way to do it is to choose companies in the same industry. However, this method is not always correct. That is because, no matter how carefully selected, there are differences among them. In that case, the following tests will help.

##### Step1: Definition of a comparable company

A comparable company is the one that has similar risk, growth rate, and cash flow properties to the company that you evaluate. If such companies exist, it is ideal to value the target company with those. Many look for comparable companies in the industry or sector in which a target company belongs. That assumes that the company in the same field would have similar risks, growth rates, and cash flows as the target company. However, even if they belong to the same group, it is tough to satisfy everything. Therefore, you need to control the differences among the target company and the comparable companies to get a more reliable multiple.

In particular, this controlling can be very useful when you need to find them globally. Suppose you find multiples in the semiconductor industry. Then, you would look for comparable companies in Asia, the United States, and Europe. In this case, if you adjust the differences in risk, growth rate and cash flow attributes of firms across countries, you will get better comparable indicators.

##### Step 2: Controlling for differences across firms

There are three ways to control differences among comparable companies after selecting them; Subjective controlling, use of PEG, and regression analysis.

First, Subjective controlling

In the process of relative valuation, you choose two. One is comparable companies to compare a target company to, and the other is the multiples of them. Here, you can eliminate specific companies from the list by analysing each risk, growth, and cash flow attribute. You can also increase or decrease the final multiple by your discretionary calculation. The upside is that you know the companies well through analysis so that you could get appropriate multiples. However, the downside is that, paradoxically, the view is subjective.

Second, PEG

Another way is to use a modified multiple, PEG, which is the PER divided by the company’s growth rate. When you judge whether a stock is undervalued or overvalued only with PER, you could misjudge. That is because each company’s potential growth rate is different. Thus, if you calculate the PEGs with the growth rate of each company and then compare them to the industry average PEG, you will more accurately judge the appropriateness of their stock price levels.

Third, Sector regression analysis

If you find at least one more variable affecting the PER of your target industry, you can value the target company through regression analysis.

Suppose there are n companies. You determined that growth rate and risk would have a significant impact on the PER. Then, you can derive the following equation through regression analysis using those variables.

#### Theoretical PER = a – b*σ + c*g

a, b, c = coefficient, σ = risk, g = growth rate

Now, you can obtain the intrinsic PER of a target company with that equation. After then, you can compare it to the PER from the stock market and then judge whether it is overpriced, underpriced, or appropriate.