Chapter 21 International Asset Pricing

1
Chapter 21International Asset Pricing
Learning objectives ? Asset pricing
models The CAPM and the IAPM Arbitrage
pricing theory (factor models) ? Factor model
applications Macroeconomic factors of
return Global, country, and industry
factors The value premium and the size
effect Momentum effects The currency risk
factor
Butler, Multinational Finance, 4e
2
What makes a good investment?

• In which firm would you rather invest?
• A firm with high earnings growth or a firm that
  is in financial distress?
• A firm whose stock has recently risen in price or
  a firm whose stock has recently fallen in price?
• Studies of asset prices in international markets
  can shed light on these questions.

Asset pricing models
3
On asset pricingand informational efficiency...

• It is the theory that decides
• what can be observed.
• Albert Einstein

Asset pricing models
4
The traditional capital asset pricing model (CAPM)

• Perfect financial markets
• Frictionless markets
• Rational investors with equal access to costless
  information and market prices
• Homogeneous expectations
• Everyone can borrow and lend at the riskless rate
  of interest rF

Asset pricing models CAPM
5
The CAPM capital market line
Capital market line
Expected return Erj

Efficient frontier
M
Investment opportunity set
ErM
rF
sM
Standard deviation of return
Asset pricing models CAPM
6
The CAPM security market line
rj rF ßj (ErM - rF) ßj ?j,M (sj /sM )
Asset pricing models CAPM
7
The one-factor market model
rj aj ßj rM ej
bj ?j,M (sj /sM )
Asset pricing models CAPM
8
An international version of the CAPM

• In addition to the CAPM assumptions, suppose
• Purchasing power parity holds
• Investors in each country have the same
  consumption basket
• This leads to an international version of the
  CAPM called the International Asset Pricing Model
  (IAPM)

Asset pricing models IAPM
9
The International Asset Pricing Model

• Like the CAPM, the market portfolio in the IAPM
  includes all assets in the world weighted
  according to their market values
• Investors also hold a hedge portfolio of domestic
  and foreign bonds
• as a store of value (like rF)
• as a hedge of the market portfolios currency risk

Asset pricing models IAPM
10
The IAPM capital market line
Capital market line
Expected return Erj

Efficient frontier
W
Investment opportunity set
ErW
rHedge portfolio
sW
Standard deviation of return
Asset pricing models IAPM
11
Mean returns and betas

• Unfortunately, market model betas estimated with
  unconditional models have no power to explain
  security returns.
• Fama and French (1992) conclude
• ...the relation between market beta and average
  return is flat, even when beta is the only
  explanatory variable.
• Fama and French, The Cross-Section of Expected
  Stock Returns, Journal of Finance, June 1992.

Asset pricing models IAPM
12
Arbitrage pricing theory (APT)

• rj aj ß1jF1 ... ßKjFK ej (21.5)
• where rj random rate of return on asset j
• aj expected return on asset j
• ßkj sensitivity of asset j to factor k where
  k1,...,K
• Fk systematic risk factor k
• ej a random error term

Asset pricing models APT
13
The one-factor market model

• Regressing rj on rM yields
• rj aj bjrM ej (21.3)
• Subtracting asset js mean return
• mj aj bj mM
• from both sides of (21.3) and rearranging yields
  a one-factor market model in excess return form
• rj mj bjFM ej (21.6)
• where FM (rM - mM)

Asset pricing models APT
14
Beta as a regression coefficient

• rj aj bj rM ej ? rj - mj bj FM ej
• where FM rM-mM

Asset pricing models APT
15
APT factors

• Roll Ross (1995) identified 4 APT factors
• rj mj b1jF1 b2jF2 b3jF3 b4jF4
  ej (21.7)
• F1 industrial production
• F2 risk premia (corporate - government bond
  yield)
• F3 term premia (long-term T-bond minus T-bill
  yield)
• F4 inflation
• When the market return was included as a fifth
  factor, its coefficient was not significant.
• Roll Ross, The Arbitrage Pricing Theory
  Approach to Strategic Portfolio Planning,
  Financial Analysts Journal, Jan/Feb 1995.

Macroeconomic factors
16
The relative importance of global, country, and
industry factors

• There is a long-term trend toward increasing
  capital market integration
• Integration tends to increase the correlations
  between national indices, which reduces the
  benefits of international diversification
• Cross-country correlations typically are lower
  than cross-industry correlations
• Consequently, diversification across countries
  usually brings greater benefits than
  diversification across industries

Macroeconomic factors
17
Fama Frenchs value premium

• Fama French fit a 3-factor model
• rj mj bj (rM - mM) bZj FSize bDj
  FDistress ej
• FSize the difference in mean return between
  the smallest 10 and the largest 10 of firms
• FDistress (relative financial distress) the
  difference in mean return between value and
  growth stocks
• Value stocks high equity book-to-market ratios
• Growth stocks low equity book-to-market ratios
• Fama and French, The Cross-Section of Expected
  Stock Returns, Journal of Finance, 1992

Value and size
18
The value premium and the size effect

• Firm size Small firms outperformed large firms
  by an average of 7 percent per year
• Relative financial distress Value stocks
  outperformed growth stocks by an average of 12
  percent per year
• After controlling for size and relative financial
  distress, the market factor contributed nothing
  to the explanatory power of the regression

Value and size
19
The value premium in U.S. stocks
Annual returns to portfolios ranked on equity
book-to-market
Value stocks
Growth stocks
Value and size
20
The international value premium

• Fama and French extended their study to
  international stocks
• Value stocks have higher mean returns than growth
  stocks in 12 of 13 international markets
• The difference in mean return is 7.60 per year
• Fama French, Value versus Growth The
  International Evidence, Journal of Finance,
  December 1998.

Value and size
21
A cross-country comparisonof the value premium
Annual return
Value and size
22
What makes a good investment?

• In which firm would you rather invest?
• A firm with high earnings growth or a firm that
  is in financial distress?
• Firms in financial distress have higher
  expected returns than growth firms
• But this difference in mean return could merely
  represent a systematic risk factor, so that you
  get what you pay for

Asset pricing models
23
Momentum in U.S. stocks

• Jegadeesh and Titman categorized stocks into 10
  equal-sized portfolios according to return over
  the preceding 6 months
• Winners - stocks with the highest return over the
  preceding 6 months
• Losers - stocks with the lowest return over the
  preceding 6 months
• Jegadeesh Titman, Returns to Buying Winners
  and Selling Losers Implications for Stock Market
  Efficiency, Journal of Finance, March 1993.

Momentum
24
Momentum in European stocks

• Rouwenhorst examined momentum in 12 European
  stock markets
• Past winners outperformed losers by more than one
  percent per month after correcting for risk
• Return continuation lasts for about one year, and
  then is partially reversed
• K. Geert Rouwenhorst, International Momentum
  Strategies, Journal of Finance, February 1998.

Momentum
25
Momentum in the U.S. Europe
Mean monthly returns (winners - losers)
Months relative to portfolio formation 12 24 36
Momentum
26
Cumulative momentum returns
Cumulative returns (winners - losers)
Months relative to portfolio formation
Momentum
27
Cross-country momentum returns
Mean monthly returns (winners - losers)
Momentum
28
What makes a good investment?

• In which firm would you rather invest?
• A firm whose stock has recently risen in price or
  a firm whose stock has recently fallen in price?
• These results are difficult to reconcile with
  the efficient markets hypothesis because of the
  curious reversal in return after about one year

Asset pricing models
29
Exposure to currency riskas a regression
coefficient
rd md bf sd/f ed
Currency risk
30
The diversifiability ofcurrency risk exposure
Currency risk
31
Is currency risk priced in the US?

• Jorion added a currency risk factor
• rjd mjd b1j(rMd-mMd) bSjfsd/f
  ejd (21.13)
• rjd mjd b1jF1d ... b4jF4d bSjfsd/f
  ejd (21.14)
• In actively traded US markets, the currency risk
  factor is subsumed into the other factors
• There remains considerable cross-sectional
  variation among US-based MNCs
• Jorion, The Pricing of Exchange Rate Risk in
  the Stock Market,
• Journal of Financial and Quantitative Analysis,
  Sep. 1991.

Currency risk
32
Is currency risk priced outside the US?

• De Santis and Gérard fit a conditional version
  of a 2-factor model
• Conditional models allow risks to vary over time
• Different national markets have different
  currency risk exposures
• Currency risk was a small fraction of total risk
  in the United States
• Currency risk was a significant proportion of
  total risk in Germany, Japan, and the United
  Kingdom
• De Santis Gérard, How Big is the Premium for
  Currency Risk, Journal of Financial Economics,
  September 1998.

Currency risk