Estimating Required Return using different Methods

In this lecture I will talk about different methods to estimate the cost of equity.

1. CAPM
a. Beta estimate of a public company
b. Adjustment of raw beta
c. Beta estimate of private company
2. Arbitrage Pricing Theory
3. Fama French Model or Three Factor Model
4. Extension of Fama French Model (Carhart Model)
5. Extension of Fama French Model (Pastor and Stambough, 2003)
6. Bond Yield Plus Risk Premium (BYPRP
7. Constant Dividend Growth Model

1. CAPM
Capital Asset pricing model is the market equilibrium model to estimate the required return of an asset as a linear
function of its systematic risk measured by stock’s beta. The key insight of the CAPM is that investors assess the
risk of an asset in terms of its contribution to systematic risk of their portfolio.
It is assumed that all the investors are rational and all rational investors hold the market portfolio. So when
investors consider an asset for investment, he considers only how much additional risk will be added to his existing
market portfolio, and investor demands risk premium for that additional risk or systematic risk only.

We know that risk is the variability of return as measured by standard deviation or variance. Suppose, market
portfolio is the DGEN.The return series and risk as measured by standard deviation are depicted in the following
figure.

40.0%
30.0% Standard devaition of return of DGEN =9.12%
20.0%
10.0%
Return

0.0%
DGEN

October

October
October

October

October
June

June

June

June

June
February

February

February

February

February
-10.0%
-20.0%
-30.0%
-40.0%
Monthly return series from 2007-2011

Now consider a rational investor who have already holding the market portfolio as per CAPM, is considering
investing in AB bank share. Return series of AB bank and standard deviation are depicted in the following figure.
 Standard deviation of retrun of AB bank= 20.38%
Covariance with DGEN retturn= 1.03%
Contribution of risk if added to DGEN or beta = 1.24
100.0% Correlation of DGEN with AB bank return= 0.56

50.0%

Return
0.0% ABBANK

August
August

August

August

August
November

November

November

November

November
May

May

May

May

May
February

February

February

February

February
-50.0%
Monthly return of AB Bank from 2007-2011

Notice the line graph of the DGEN and AB bank, returns series of these two are not moving in a perfectly same
way, ( correlation of .56 less than 1). Similarly , there are some times , when DGEN moves up , AB banks moves
down and vice versa which will contribute the dispersion of return series to be lower. That is there is some risk
reduction (lowering of return series dispersion). If we combine these results, we can conclude that adding AB bank
to DGEN will not exactly add risk of 20.38% which is the standard deviation of AB Bank return. Actual risk
addition will be lower than this 20.38%. This additional risk is known as systematic risk or market risk.

According to CAPM, the relation between an asset’s risk and return is:

𝑹𝒊 = 𝑹𝒇+ 𝜷𝒊 ∗ (𝑹𝒎 − 𝑹𝒇 )
Where
𝑅𝑖 is expected rate of return of asset 𝑖

𝑅𝑓 is the risk free rate of return

𝑅𝑚 is the expected market rate of return

𝝈𝒊,𝒎
𝛽𝑖 is the sensitivity of asset’s 𝑖 return to market return and computed as 𝜷𝒊 = 𝝈𝟐𝒎

1.1 Beta Estimate for a public limited company

The simplest estimate of beta is found by ordinary least square method or linear regression of return on stock as
dependent variable and the return of market as the independent variable. The resulting regression line slope or beta
estimate is known as unadjusted beta or raw beta.

Most commonly you collect the recent month end close price data of market and individual stock for the last 61
months and from these price date ,you estimate the return series of market and individual assets for the 60 months.
Then you regress the return of individual stocks on the returns on market (think of return of individual assets as
dependent variable and return of market as independent variable)

Or alternatively you can collect the weekly return series of market and the individual stocks for the latest 2 years
(52*2 = 104 weeks) and regress the return of individual stocks on the returns on market

Or simply you can calculate the beta by dividing the covariance between the return of market and individual assets
by the variances of the market return
1.2 Adjustment of Raw Beta

The beta of a stock has a tendency in the future period to revert to its mean value of 1, the beta of an average
systematic stock beta. Because valuation is forward looking, it is more logical to adjust the beta so that it can more
closely reflect the future beta.

The most commonly used example is the Blume Adjusted Beta using the following equation

Adjusted beta = 2/3(unadjusted beta) + 1/3(1)

For example AB stock has a raw beta of 1.3, using the Blume adjustment, we find adjusted beta as

= 2/3(1.3) +1/3(1)

= 1.20 which can be used in CAPM equation instead of raw beta of 1.2 to more closely
predict the future.

1.3 Beta Estimate for Non-public or Thinly Traded Company

Although market price data is available for publicly traded company from which analyst can calculate the beta, but
for non-public or thinly traded company, market price is not available. So analyst faces a problem here in
calculating beta. Analysts address this problem by indirectly estimating the beta of non-public company based the
beta of peer public company in the same industry.

The procedures must take into account the effects of beta of the differences of financial leverages between non-
public company and the peer benchmark public company. In other words, if we assume that a nonpublic
company and its most closely matched peer public company has same business risk, so the only difference of
risk between them is financial risk ( Total risk = Business Risk + Financial Risk) .

First the equity beta of the benchmark public company is unlevered to estimate the benchmark public company’ the
asset beta – reflecting the systematic arising from just the business risk or economics of industry. Next, the
benchmark asset beta is re-levered to reflect the financial leverage of the non-public company.

For example, Metro spinning is a publicly listed company having beta of 1.3, and D/E ratio of 40%, and marginal
tax rate of 30%. Now, we have Malek Texlile which is non-public company and we want to calculate the beta of
Malek Taxtile. Assume that Both Metro Spininng and MakekTexlinehas same business risks.

Here at first we unlevered the beta of Metro Spinning to estimate its asset beta using the following equation
1
Asset beta = Equity beta * ( 𝐷 )
1+(1−𝑡𝑎𝑥)( )
𝐸

1
= 1.3 * ( )
1+(1−0.3)(0.4)
 = 1.016

Now we will re-lever the asset beta of Metro Spinning by the financial leverage of the Makek Textile. Assume the
financial leverage of Malek Textile is 50% and tax rate is 30%. So equity beta of the Malek texline can be
estimated by the following equation

𝐷
Equity Beta/Beta = Asset Beta * 1 + (1 − 𝑡𝑎𝑥)( )
𝐸

= 1.016 * 1+ (1-0.3)*0.5

= 1.37

Notice that beta estimate of Malek textile is higher than that of Metro Spinning (1.37>1.3) . As both has same
business risk but Malek Textile has more financial leverage than that of Metro Spinning ( 50%>40%) , so it is
logical that Malek Textile will have higher beta.

2. Arbitrage Pricing Theory

Though CAPM claims that required return of an asset is a linear function of its systematic risk and the only risk
factor in required return estimation is market factor but many of the empirical studies point out some serious
deficiencies in the model as an explanation of the link between risk and return. There was mixed support for a
positive linear relationship between rates of return and systematic risk for portfolios of stock, with some recent
evidence indicating the need to consider additional risk variables or a need for different risk proxies.
One especially compelling challenge to the efficacy of the CAPM was the set of results suggesting that it is
possible to use knowledge of certain firm or security characteristics to develop profitable trading strategies, even
after adjusting for investment risk as measured by beta. Typical of this work were the findings of Banz, who
showed that portfolios of stocks with low market capitalizations (i.e., “small” stocks) outperformed “large” stock
portfolios on a risk-adjusted basis, and Basu, who documented that stocks with low price-earnings (P-E) ratios
similarly outperformed high P-E stocks. More recent work by Fama and French also demonstrates that “value”
stocks (i.e., those with high book value-to-market price ratios) tend to produce larger risk-adjusted returns than
“growth” stocks (i.e., those with low book-to-market ratios) and small cap company tends to produce higher return
than big cap company on risk adjusted basis.

Given the limitations of CAPM, the investment and academic community searched for an alternative asset pricing
theory to the CAPM that was reasonably intuitive, required only limited assumptions, and allowed for multiple
dimensions of investment risk. The result was the arbitrage pricing theory (APT), which was developed by Ross in
1976 has three major assumptions:

1. Capital markets are perfectly competitive.

2. Investors always prefer more wealth to less wealth with certainty.

3. The stochastic process generating asset returns can be expressed as a linear function of a

set of K risk factors (or indexes).

APT says that there are many risk factors that drive the return of stocks in contrast to single market factor in
CAPM. The theory assumes that the stochastic process generating asset returns can be represented as a K factor
model of the form:
 E (𝑹𝒊 ) = 𝛌𝟎 + 𝜷𝒊𝟏 𝛌𝟏 + 𝜷𝒊𝟐 𝛌𝟐 + 𝜷𝒊𝒌 𝛌𝒌
Where:

λ0 = the expected return on an asset with zero systematic risk expected return on asset i when all the risk factor
premium is zero

λk= the risk premium/ factor premium related to the kth common risk factor

𝛽ik=factor sensibility ; that is, how responsive asset i is to the kth common factor. (These are called factor betas or
factor loadings.)

Two terms require elaboration: λk and 𝛽ik . As indicated, λk terms are the multiple risk factors expected to have an
impact on the returns of all assets. Examples of these factors might include inflation, growth in gross domestic
product (GDP), major political upheavals, or changes in interest rates. The APT contends that there are many such
factors that affect returns, in contrast to the CAPM, where the only relevant risk to measure is the covariance of the
asset with the market portfolio (i.e., the asset’s beta).Given these common factors, the 𝛽ik terms determine how
each asset reacts to the jth particular common factor. To extend the earlier intuition, although all assets may be
affected by growth in GDP, the impact (i.e., reaction) to a factor will differ. For example, stocks of cyclical firms
will have larger 𝛽ik terms for the “growth in GDP” factor than will noncyclical firms, such as grocery store chains.
Likewise, you will hear discussions about interest-sensitive stocks. All stocks are affected by changes in interest
rates; however, some experience larger impacts. For example, an interest-sensitive stock would have a 𝛽k interest
of 2.0 or more, whereas a stock that is relatively insensitive to interest rates would have a 𝛽k of 0.5. Other examples
of common factors include changes in unemployment rates, exchange rates, and yield curve shifts. It is important
to note, however, that when we apply the theory, the factors are not identified.

Illustration: assume that there are two common risk factors: one related to unexpected changes in the level of
inflation and another related to unanticipated changes in the real level of GDP. If we further assume that the risk
premium related to GDP sensitivity is 0.03 and stock that is sensitive to GDP has a bj (where j represents the GDP
factor) of 1.5, this means that this factor would cause the stock’s expected return to increase by 4.5 percent (= 1.5 ×
0.03).To develop this notion further, consider the following example of two stocks and a two factor model. First,
consider these risk factor definitions and sensitivities:

λ1= unanticipated changes in the rate of inflation. The risk premium related to this factor is 2 percent for every 1
percent change in the rate (k1 = 0.02)

λ2 = unexpected changes in the growth rate of real GDP. The average risk premium related to this factor is 3
percent for every 1 percent change in the rate of growth (k2 = 0.03)

λ0 = the rate of return on a zero-systematic risk asset (i.e., zero beta) is 4 percent (k0 = 0.04)Assume also that there
are two assets (x and y) that have the following response coefficients to these common risk factors:

bx1 = the response of asset x to changes in the inflation factor is 0.50 (bx1 = 0.50)

bx2 = the response of asset x to changes in the GDP factor is 1.50 (bx2 = 1.50)

by1 = the response of asset y to changes in the inflation factor is 2.00 (by1 = 2.00)

by2 = the response of asset y to changes in the GDP factor is 1.75 (by2 = 1.75)
These factor sensitivities can be interpreted in much the same way as beta in the CAPM; that is, the higher the level
of bij, the greater the sensitivity of asset i to changes in the jth risk factor. Thus, the response coefficients listed
indicate that if these are the major factors influencing asset returns, asset y is a higher risk asset than asset x, and,
therefore, its expected return should be greater. The overall expected return equation will be:

E (Ri) = λ0 + λ1bi1 + λ2bi2

= 0.04 + (0.02) bi1+ (0.03) bi2

Therefore, for assets x and y:

E (Rx) = 0.04 + (0.02)*(0.50) + (0.03)*(1.50)

= 0.0950

= 9.50%

And

E (Ry) = 0.04 + (0.02)(2.00) + (0.03)(1.75)

= 0.1325

= 13.25%

3. Fama French Model or Three Factor Model

Though CAPM contends that only driver or only factor of equity return is the market factor but after 1980s,
substantial empirical evidence showed that market factor alone failed to explain the return. In the US and other
equity market, evidence suggests that Small cap stocks or value stocks generate higher return over the long run
than the CAPM predicts.

In 1993 researcher Fama and French addressed the perceived weakness of the CAPM in a model with Three
Factors, which is known as Fama French Model (FFM) or Three Factor Model (TFM).

The 3 factors of return in the FFM is namely

1. RMRF or market factor similar to CAPM, which is measured as return on market value weighted equity
index in excess of risk free rate
2. SMB ( Small Minus Big) or size factor which is the return to a portfolio of small capitalization stocks less
the return to a portfolio of large capitalization stocks

3. HML (i.e., high minus low) or value factor which is the return to a portfolio of stocks with high ratios of
book-to market values less the return to a portfolio of low book-to-market value stocks

The FFM estimate of required return on equity can be expressed as

Ri = Rf + β1 mkt factor RMRF + β2 size factor SMB + β3 value factor HML

The FFM equation says that required return on equity is not just only function of market factor as in CAPM, two
additional factors of return - the size factor and the value factor play role in the required return on equity.
Like market risk factor premium, average historical estimate of size factor and value factor premium is positive.
This means that investor will require higher required for holding small size company compared to average size
company though both small size and average size company have the same market beta. The logic and intuition is
that small company is relatively more risky than an average size company and CAPM market beta cannot capture
this risk arising from the size factor. Similar to size, required return on high value company (company having high
book value to market price ratio) is higher compared to average value company though both high value and
average value company have the same market beta

Like CAPM, neutral or average value of beta (B1) for market factor in the FFM is 1 , but neutral value or average
value of beta for size and value factor is zero . For a security, the beta value of zero for size and value factor means
that the security has no size or value bias. A beta value of positive for size factor means that security is a smaller
company relative to average size company, similarly a negative beta value for size factor means that security has
relatively higher size than a average size company in the market.
Illustration of Estimation of Historical Market , Size and Value Factor Premium

Market Return Return

Risk Market Size Return on Return on Value
Return on Small on Big
Free Premiu Factor company company factor
(DSEX cap cap
Year Rate m Premium with High with low Premium
) company company
1 BV to MP BV to MP
6
ratio ratio
2 10= (9—
3 4= 3-2 5 7=(6-5) 8 9
8)
1 15.00% 10% 5% 18.00% 14.00% 4.00% 16.50% 14.50% 2.00%
2 22.00% 10% 12% 25.00% 21.00% 4.00% 23.50% 21.50% 2.00%
3 13.00% 10% 3% 15.50% 12.00% 3.50% 18.50% 12.50% 6.00%
4 24.00% 10% 14% 26.50% 19.00% 7.50% 29.50% 23.50% 6.00%
5 15.00% 10% 5% 17.50% 14.00% 3.50% 20.50% 14.50% 6.00%
6 26.00% 10% 16% 28.50% 25.00% 3.50% 27.50% 25.50% 2.00%
-
7 10% -22% -9.50% -13.00% 3.50% -10.50% -12.50% 2.00%
12.00%
8 15.00% 10% 5% 17.50% 14.00% 3.50% 16.50% 14.50% 2.00%
9 30.00% 10% 20% 32.50% 29.00% 3.50% 32.50% 29.50% 3.00%
10 27.00% 10% 17% 29.50% 26.00% 3.50% 29.50% 26.50% 3.00%
11 22.00% 10% 12% 26.50% 21.00% 5.50% 23.50% 21.50% 2.00%
12 20.00% 10% 10% 22.50% 16.00% 6.50% 21.50% 19.50% 2.00%
Averag
18.08% 10% 8% 20.83% 16.50% 4.33% 20.75% 17.58% 3.17%
e
The table above shows market risk premium RMRF is 8%, Size factor premium is 4.33% and the value factor
premium is 3.17%, so the FFM equation stands at

Ri = Rf + β1*8% + β2 4.33% + β3 3.17%

Suppose a share of company X has sensibility to market factor or B1 is 1.1 and sensibility to size factor or B2
is 1.5 and sensibility to value factor is B3 -0.5, so the required return on share X is

Also assume that risk free rate is 6%,

Rx = 6% + 8%*1.1 + 1.5* 4.33% + 3.17% * - (0.5)

Rx= 19.71%

FFM estimates required return to be 19.71% whereas if CAPM is used the required return would have been

Rx = 6% + 8%*1.1

= 14.8%

CAPM required return is lower than FFM return because CAPM assumes that investors do not require size factor
premium and value factor premium and this factor premium are already in market factor premium or this size
factor and value factor premium are due to market inefficiency. But considerable studies and empirical work have
shown that small cap company and high value company outperform big cap and low value company over a long
period of time horizon both in US and other developed market.

FFM says in the above example that the required return of 19.71 % comes from three risk sources, market risk ,
size factor and value factor risk. Moreover the FFM also says that company X is a very small size company as
depicted by the size factor beta of 1.5 and size factor contributes to its return of 4.33% *1.5 = 6.5%. Beta to value
factor of -0.5 means that X is a low value company (or growth company) and this value factor contributes
negatively to its required return.

With FFM, it is also possible to construct portfolio to the investment objective of portfolio manager.

For example, suppose 3 companies A, B, C are to be combined in a portfolio and investment manager want that
portfolio return to go up by 1% when market return goes up by 1% and portfolio return to go up by 3% when
market return of small size companies go up by 1% and portfolio return to go up by 2% when portfolio of value
company goes down by 1%. Following information is available

Factor Risk Premium Estimate

Market factor premium 7%
Size factor premium 4%
Value factor premium 3%
Company Beta or sensibility Beta or sensibility Beta or sensibility Expected Return
to market factor to size factor to value factor
A 15%
1.20 0.50 (0.50)
B 20%
1.00 2.00 (1.00)
C 18%
1.50 (1.00) 2.00

Now if we want to build such portfolio of stocks A, B, C then according to requirement, Portfolio should be
following characteristics

Portfolio market beta = 1

Portfolio size beta = 3 and

Portfolio value factor beta = -2

So weight of each of stock in the portfolio should be such that

WA*1.2 +WB *1 + WC *1.5 = 1

WA*.5 +WB *2+ WC *(-1) = 3

WA*(-.5) +WB *(-1) + WC *2 = -2

Now if we solve this simultaneous equation , we can find the weight of A , B and C to construct a portfolio having
exposure or beta to market factor , size factor and value factor of 1 ,3 and -2.

4. Extension of Fama French Model (Carhart Model)

There have been other interesting characteristic-based approaches to estimating a multifactor model of risk and
return. Three of those approaches are described here. First, Carhart directly extends the Fama-French three-factor
model by including a fourth common risk factor that accounts for the tendency for firms with positive (negative)
past returns to produce positive (negative) future returns. He calls this additional risk dimension a momentum
factor and estimates it by taking the average return to a set of stocks with the best performance over the prior year
minus the average return to stocks with the worst returns. In this fashion, Carhart defines the momentum factor—
which he labels PR1YR—in a fashion similar to SMB and HML. Formally, the model he proposes is:

Ri = Rf + β1 mkt factor RMRF + β2 size factor SMB + β3 value factor HML + β4 momentum factor PR1YR

He demonstrates that the typical factor sensitivity (i.e., factor beta) for the momentum variable

is positive and its inclusion into the Fama-French model increases explanatory power by as much

as 15 percent.
 5. Extension of Fama French Model (Pastor and Stambough, 2003)
Another extension of Fama French work is the Pastor and Stambough model or PSM . PSM claims that investors
require additional risk premium for holding illiquid share and adds fourth factor to FFM . Liquidity premium
estimate is found by taking the difference between the return on illiquid shares and return on the liquid shares and
takes the following form.

Ri = Rf + β1 mkt factor RMRF + β2 size factor SMB + β3 value factor HML + β4 liquidity factor LIQ

Build Up Model Estimate of Estimating Required Rate of Return

Another widely used tool of estimating the required rate of return especially for private and closely held companies
is the Buildup method. The method sums up risk free rate plus one or more risk premium to estimate the return

Ri = Risk free rate + one or more risk premium

Suppose we want to cost of equity or required return on equity using build up method.

Risk free rate is 6%. The market risk premium is 7%. The average systematic risk or beta is 1 for a publicly traded
company. Now suppose we want to estimate the required return on equity of a private company which is very small
in size ,

As company is not publicly traded, so its illiquid share, so investor will demand liquidity premium. Similarly due
to small size, investor will require size premium. So we can add this risk premium to the required return of a
average systematic risky publicly traded company to find the the required return of that private , small size
company as follows

RI = 6% + 7%*1 + liquidly premium + Size premium

If Liquidity premium is = 2% and size premium is 3%, so the required return is

Ri = 6% +7%*1 + 2% + 3%

= 18%.

6. Bond Yield Plus Risk Premium (BYPRP)

For companies with publicly traded debt, BYPRP approach provides a quick estimate of the cost of equity.

The estimate is

BYPRP = YTM on company’s long term debt + Risk Premium

YTM on company’s long term debt covers both

1. Real interest rate and premium for expected inflation

2. A default risk premium for bond obligation

The risk premium in BYPRP estimates covers the additional default risk arising from equity issue.
Suppose ACI has corporate bond listed with DSE having a yield of 12.5%. If Risk premium for equity issue is 4%
then

BYPRP estimate of cost of equity = 12.5% + 4%

= 16.5%

7. Dividend Growth Model

Cost of equity of a stock can also be calculated using the dividend growth model. According the to the
constant growth dividend model, we know that price of a company (assuming price is equal to intrinsic value
of the company and market is efficient) is

𝑫𝟏
𝑷=
𝑲𝒆 − 𝒈
Where P= Price = Value of the company
D1 = Expected Dividend at the end of the year
Ke = Cost of Equity
g = growth rate in dividend

If we can do some algebra, the above equation can be written as follows

 𝑫𝟏
𝑲𝒆 = + 𝒈
𝑷𝟎

8. Calculating Weighted Average cost of equity

When calculating WACC, we need to have weight of different sources of capital such as debt, preference share and
common stock. This weight may be based on market value weight, book value weight, and the target weight of
each sources of capital in the capital structure.

Book value weight does do reflect the most recent market condition as most of assets are reported on Balance sheet
using historical cost basis. So Book value weight is least preferred. Market value weight has, though reflects the
most recent market condition, but is affected by the short term market price variation. Plus the current capital
structure may not reflect what a company target. For example company may have debt to equity ratio 30% but
company may find that if it can have capital structure at 50% debt and 50% equity , its WACC will be lowest,
which is its target capital structure . So when estimating WACC, weight preference should be target weight first,
then market value weight, then if none of two is found, Book value weight .