Corporate Finance CAPM Again

1
Corporate FinanceCAPM Again
2
Issues in applying the CAPM

• What riskfree rate should we use? Time horizon
  (maturity) of the asset?
• How should we measure beta? Relative to which
  index? How many months of data?
• How should we measure the market premium?

3
Beta

• Often, monthly return data is used. Five years
  of data historical data or 12 x 5 60
  observations.
• In Japan, the TOPIX index and Nikkei 225 Index
  are often used as proxies for the market
  portfolio.

4
Year Price Return 98 50
99 100
100 00 60 -40
Arithmetic average return is 30 0.5 (100(-40))
Geometric average return is 9.5 (1.20.5 1)
Example, from Damodaran
5
Market Premium

• Often the historical average is used. Monthly
  returns data? Annual returns data?
• Why do we use the historical average?
• How far back should we go? How many observations?

6
Nikkei 225 Index from 1976.01 to 1998.12
7
Cash flows in future
Historical data
Today
CAPM one period model
8
Sources of Error market premium

• Ferson and Locke (1998) publish a set of
  interesting findings.
• We find that the great majority of the error in
  estimating the cost of equity capital is found in
  the risk premium estimate, and relatively small
  errors are due to the risk measure, or beta.
  This suggests that analysts should improve
  estimation procedures for market risk premiums,
  which are commonly based on historical averages.
  (p. 485)

9
Cont.

• Risk premium is estimated using three methods
  historical, regression, GMM.
• Betas estimated using three methods regression,
  ones, GMM.
• Greatest error when using the historical risk
  premium.

10
Appendix Betas and Emerging Markets

• Adjusting with a country beta
• Using a modified model

11
Adjusting with Country Beta

• For example, a Japanese corporate in the auto
  industry is expanding into Peru. We need data to
  estimate the beta for an automobile company in
  Peru.
• We could use a proxy. Take the beta for the
  Japanese corporate and use it as a proxy. This
  proxy must be adjusted.
• One adjustment is to multiply the proxy beta by a
  country beta (Japan, Peru).

12
(No Transcript)
13
Cont.

• If the country beta is greater than one, this
  implies that Perus market is riskier (more
  volatile) than the Japanese market. Thus, the
  proxy beta must be adjusted upward to take into
  account for the overall greater risk in the
  Peruvian market.

14
Country Credit Rating

• When working with emerging markets, the use of
  betas could be problematic.
• Beta is estimated by running a regression of the
  country stock index return against a world stock
  index such as that published by Morgan Stanley.
• What if the stocks traded in the country in
  question have very little weight in the world
  index?
• What if the emerging market is not liquid? What
  if the stock market does not exist?

15
Basic Model
16

• Notice, that gamma (one) should be negative.
  Higher credit rating then the expected return
  should be lower.
• CCR is the index published by Institutional
  Investors.

17
(No Transcript)