Issues in Business

Valuation

Cost of capital

FAS internal training

7 September, 2005
What we seek to answer through this presentation:
Q - What is Cost of capital?
Q - What are the various components?
Q - How do we value/calculate each component?
What is the time interval used
Which index used
Time period
Q – Practical implications for calculating cost of capital
Q – Alternative methods of calculating cost of capital
Q – Some empirical evidences to support
Q – Usage of cost of capital for the purpose of Valuation
Q – Extract of some research papers
Q - Some follow up readings
Q – Case study/Assignment
 What is cost of capital
• Cost of capital is the compensation expected by various
providers of capital for the opportunity cost of investing
their funds in one particular business instead of other(s)
with equal risk.
• Cost of capital must:
– Comprise a weighted average of the cost of all sources of capital
– Post tax costs
– Nominal rates of return
– Adjusted for systematic risk borne by each provider
– Market value weights
– Subject to change with change in inflation, systematic risk and
capital structure
• Weighted average cost of capital (WACC) is the discount
rate to convert expected FCF to all the capital providers
into present value.
 WACC
• WACC=Kb(1-t)(B/V) + Kp(P/V) + Ks(S/V)
Kb : cost of debt
B: Market value of debt
V: Market value of enterprise
Kp: After tax cost of preferred stock
Ks: opportunity cost of equity capital
P: Market value of preferred stock
S: Market value of equity
• Other possible entries include:
– Leases (operating and capital)
– Subsidized debt
– Convertible or callable debt/preferred stock
– Minority interest
– Warrants and executive stock options
– Income bonds, commodity index bonds, extendable, puttable or
retractable bonds
 Some simplifying assumptions
• No distinction of callable and non callable debt
• Non interest bearing liabilities are not included
such as accounts payable, though there is
implied financing cost which if separated from
operating cost will complicate the process
• It is theoretically correct to use different WACC
for each projection year, however we usually use
one WACC for the entire forecast period as at
one point of time a company’s capital structure
will not reflect capital structure of the company
for the rest of its life.
 Calculation of WACC
Developing market Opportunity cost Opportunity
value weights of non equity cost of equity
capital capital

♫Circularity issue – As we need to know market value weights

to determine WACC - which needs cost of equity. And cost
of equity cannot be calculated without knowing WACC, as it is
calculated by discounting FCF with WACC.
♫Existing or target capital structure
♫Capital structure of comparable companies in the industry
Estimating the existing capital structure
• Market value of listed capital structure
• Problem arises when some of them are not listed
– Calculate the present value of stream of financing
payments using YTM of equivalent issue as discount
rate
• Complex calculations in other/hybrid securities:
– Debt type financing – fixed or variable/leases
– Equity linked
– Minority interests
 Some other securities
• Option features: Option pricing
• Swaps: if associated with a specific outstanding
instrument, we estimate the value of the ‘synthetic
security’
• Leases: No differentiation of capital and operating
leases, however operating leases may be kept out of
valuation analysis - if insignificant
• Warrants and ESOP: Using option pricing
• Convertible securities: Using option pricing
• Minority interest: is the claim of outside shareholders in
subsidiary companies – Using DCF or market multiples
 Calculation of WACC
Developing market Opportunity cost Opportunity
value weights of non equity cost of equity
capital capital

Straight Investment grade debt – Fixed and variable rate

Below investment grade debt – Junk bonds
Subsidized debt – Tax free debt
Foreign currency denominated debt
Leases
Straight preferred stock
 Calculation of WACC
Developing market Opportunity cost Opportunity
value weights of non equity cost of equity
capital capital

CAPM

Cost of equity = Rf + beta [E(Rm) – Rf]

Average beta for entire market portfolio is 1
Unusual to find beta higher than 2 and that lower than 0.3
Cost of equity increases linearly as a function of the measured
undiversifiable risk beta
Expected return SML
Market portfolio rate

Rm

Rf

1 Beta
 Risk free rate of return
Q -Why risk free rate of return?
A -In finance, expected return on risky investment is always measured
relative to the risk free rate – with the risk creating an expected risk
premium that is added on to the return on risk free asset.
Q – What is risk free asset?
 No variance in return -Actual return should always be equal to
expected return
 No reinvestment risk – Duration matching strategy
 No default risk – Government or proxy if Govt does not borrow OR
through using interest rate parity on forward currency contracts
Probability = 1

Expected return Returns

Forward rateHC,FC = Spot rate (1+interest rateHC)/(1+ interest rateFC)

 Risk free rate of return
• Rf is return on security/portfolio
which has no default risk and Alternatives for Rf:
completely uncorrelated with T –bills (short term GoI
returns on anything else in the securities)
economy. 10 Year T – bonds (10
• Theoretically Rf will be return on a Year GoI bonds)
zero beta, minimum variance 30 Year T- bonds (Long
equity portfolio. However due to term GoI bonds)
non availability of data for same it
cannot be constructed. Most widely used
10 Year GoI securities rate is generally recommended as its
duration is closer to the cash flows of the company being valued
More appropriate compared to long term rate (30 years) as its price
is less sensitive to unexpected changes in inflation and lower liquidity
premium built into it.
 Issue for discussion
• Risk free rate should be used of the currency in which
cash flows of the firm are estimated.
Change in interest rate Low

If we assume
reflect
Purchasing power parity
High
Level of inflation
If the differences in interest rates
With equal impact on
across the two currencies does not
adequately reflect the difference in Cash flows Discount rate
inflation, the values obtained using
different currencies will be
difference OTHERWISE NOT
Over valuation
 Market risk premium
• A forward looking rate based on past data
• Market risk premium measures what investor on an average demand
as extra return for investing in this portfolio relative to the risk free
asset
• Risk for the purpose of risk premium should be measured from the
perspective of the marginal investor given that the marginal investor is
well diversified.
• Therefore the risk that an investment adds to a diversified portfolio
should be measured and compensated
• Only the undiversifiable – market component of risk should be
rewarded.
• Company specific risk aspect is handled separately by beta.
• Practical implication is that the market risk premium data is available
in the market without reference to which rf used. Hence for the
purpose of the analysis, consistency should be maintained for the two
rates.
• Historical arithmetic average to be downward adjusted by 1.50-2.00%
towards survivorship bias
 Market risk premium
• Different methodologies used for estimating
market risk premium:
 Time period used √ Since inception
√ 50/20/10 years
Longer vs. latest
√ Standard error
 Choice of risk free security
Time period SE
 Arithmetic vs geometric averages use
– Moving average
5 years 8.94%
• For emerging markets with limited history, we
10 years 6.32%
should not use market risk premiums (local),
and should rather go through country risk 25 years 4.00%
premium route using matured market risk 50 years 2.83%
premium and adding the risk of the country
under study – This topic has already been
discussed by Punita and Hormazd

No statistically significant changes in the risk premium between 1926 to 1995.
Issues in market risk premium
Issues
Arithmetic average or Geometric average

•Arithmetic average is the best estimate of future

expected returns as all possible paths are given equal
weightage
•Arithmetic average considers all the data points
independent
•Arithmetic average is always higher than geometric
average

Greater the interval for taking average – Lower the

arithmetic average. Hence concluded that true market
premium lies between arithmetic and geometric
averages
Issues in market risk premium
Issues Issues
Survivorship bias Either historical data or Ex
ante estimates

•Survival imparts a bias to ex post •Ex ante estimates are based on

returns current value of share market
•Empirical evidence – US market relative to projections of earnings or
annual return exceeded median cash flows.
return on a set of 11 countries with •E(Rm)=D/S+g approach may be
continuous histories dating to 1920s used for Ex ante estimates. Further
by 1.9% in real terms or 1.4% in future cash flows may be estimated
nominal terms. for the purpose.
•However it is not necessary that the •Many investment banks have
out-performance will be continued started publishing estimates of the
for next decade. Hence a downward market risk premium using ex ante
adjustment. approach.
•These ex ante approach based
premium is generally lower than
historical based.
 Beta
• Betas measure the risk added on to a diversified
portfolio rather than total risk – therefore an
investment may be high risk individually but may be
low risk in terms of market risk
• Betas measure the relative risk, and thus are
standardised around 1
• Beta measures the risk added on by the investment
being analysed to a portfolio
• However in practice, beta is calculated relative to
stock market index rather than portfolio
ISSUES
-Just equity portfolio or to include other asset classes
-Diversified domestically or internationally
 Beta
• Choice of market indices: Index used beta
Dow 30 0.99
– Equity/+ debt
S&P 500 1.13
– BSE/NSE
NYSE composite 1.14
• Time period: No standards – Wilshire 5000 1.05
Theoretically 2 to 10 years – MS Capital Index 1.06
Practically 3 to 5 years
Longer periods should be used for
firms with stable business mix and Time period used Beta
estimated
leverage ratios. Shorter periods for
3 years 1.04
ones who have recently
restructured, been acquired, 5 years 1.13
divested business or changed their 7 years 1.09
financial leverage etc. 10 years 1.18
 Beta
• Choice of return Return interval
used
Beta
estimated

interval: Daily 1.33

– Daily Weekly 1.38

– Weekly Monthly 1.13

– Monthly Quarterly 0.44

– Quarterly Annual 0.77

– Annually
Daily and weekly interval are likely to have
significant bias due to non trading problem
and illiquid stocks and speculation

Most preferred
 Beta
• Betas should be extracted from more than one
source
• Also should be compared with industry beta
• If beta from two sources vary by more than 0.2
or it is more than 0.3 from the industry average,
consider using industry average
• When using industry average, unlever the beta
and relever it using the company’s capital
structure
BL = Bu ((1+(1-t)(D/E))
 Beta – in practise
• Practitioners further adjust beta towards
one. Rationale is that over time there is a
tendency on the part of betas of all
companies to move towards one on
account of increase in size over time, their
becoming more diversified and have more
assets in place producing cash flows.
 Alternatives to regression betas
• Modified regression betas:
– Based on fundamental factors – Income statement and balance
sheet
– Eg High payout – low beta
– High variability of earnings and covariability – High beta
Beta*=0.7997 + 2.28 sd + 0.21 D/E -0.000005 m cap
* Rosenberg and Marathe
 Alternatives to regression betas
• Relative risk measures
– Relative volatility = (Std dev./Average std dev across all assets)
– Accounting betas - use accounting earnings than traded prices
• Biased up for safer firms and biased down for risky firms
• Influenced by non operating factors
• At max, quarterly figures are available hence less data points
 Alternatives to regression betas
• Bottom up betas
– Determined by three variables
• Type of business
• Degree of operating leverage
• Degree of financial leverage
• Bottom up betas – steps
– Identify the business/es that make up the firm/asset/project
– Estimate the unlevered beta for the units – adjusted for change
in operating leverage
– Take a weighted average of the adjusted unlevered betas
– Calculate the leverage for the firm
– Estimate the levered beta

BL = Bu ((1+(1-t)(D/E))
 Limitations of CAPM
The model makes unrealistic assumptions
The parameters of the model cannot be estimated precisely
• - Definition of a market index
• - Firm may have changed during the 'estimation' period'
The model does not work well
• - If the model is right, there should be
–a linear relationship between returns and betas
– the only variable that should explain returns is betas
• - The reality is that
– the relationship between betas and returns is weak
– Other variables (size, price/book value) seem to
explain differences in returns better.
 Others measures for CoC
CAPM -No transaction costs Betas measured
-The diversified portfolio includes all traded against market
investments, held in proportion to their market portfolio
value
APM Investments with the same exposure to market Betas measured
risk have to trade at the same price – No against multiple
arbitrage (unspecified) market
risk factors
Multi factor No arbitrage assumption Betas measured
model against multiple
specified macro
economic factors
Proxy Over very long periods, high returns on Proxies for market
model investments must be compensation for higher risk, include market
market risk capitalisation and
P/BV ratios
 Extract of research papers
Research paper by Roelof Salomons and Henk Grootveld on
“The Equity Risk Premium – Emerging versus Developed Markets”

Developed countries Emerging countries

Canada China
France India
Germany Indonesia United States
Italy Korea 1889-1978 – Equity Risk Premium = 6.00% pa
Japan Malaysia
UK Pakistan 1802-1990 - Equity risk premium = 5.3% pa
US Philippines 1988 – 2001 – Equity risk premium = 3.7% pa
Taiwan MSCI
Thailand
Argentina
1988 – 2001 – Equity risk premium = 1.8% pa
Chile G7 countries
Columbia 1988 – 2001 – Equity risk premium = 3.6% pa
Peru IFC index of emerging markets
Venezuela
Mexico 1985 – 2001 – Equity risk premium = 12.7% pa
Brazil

*All the above averages are based on equal weightage to countries

 Extract of research papers
James Dow paper on Cost of Capital

Section on beta: “If it has Greek letters, don’t do it.”

TODAY WE WILL DISREGARD THIS ADVICE AND DO ß

Country Risk premium

France 4.5%
Germany 4.8%
International comparison
Japan 4.3%
UK 6.1%
US 4.5%
 Extract of research papers
Javier Estrada paper on “Systematic risk in Emerging Markets – The D CAPM

This paper is a critic to CAPM especially in the Emerging markets. Since

CAPM measures risk by beta – which follows from an equilibrium in which
investors display mean-variance behavior. Risk here is measured by the
variance of returns.
The semivariance of returns is considered more plausible measure of risk
and can be used to generate an alternative behavior (mean-semivariance
behavior) – called DOWNSIDE beta and an alternative pricing model (D-
CAPM)
 Some other papers – follow up reading

“Testing for time variation for beta in India” by Syeed Abusar Moonis
and Ajay Shah

“Do stock markets penalise environment unfriendly behavior?

Evidence from India” by Shreekant Gupta and Bishwanath Goldar

The presentation on “Cost of Capital” dated 15 March, 2004 by

Nuclear Energy economics and policy analysis
 Assignment
A copy of the small assignment is
distributed to you as a homework

Thank you