Week 6 - Equity Premium and Asset Allocation Flashcards
(22 cards)
What is equity premium?
The equity premium is the difference between the average returns on stock market index and the risk free rate
- estimated to be about 6% during the 100 years
- Stocks have a higher return on average than bonds
- however in a given year, stocks may have much lower returns than bonds
What is risk aversion?
- risk aversion: demanding compensation for bearing risk
- The term risk-averse describes the investor who chooses the preservation of capital over the potential for a higher-than-average return
- the return pattern for stocks and bonds is consistent with risk aversion
- -> stocks have higher average returns and higher standard deviations historically…explanation: investors demand higher returns on average for stocks to offset the higher variation in the returns
What is the equity premium puzzle?
- many economists think the 6% equity premium is too big under standard rational model of investor preference and market risk
- yes, economists agree that stocks are riskier
- but not risky enough to justify 6% per year for the level of risk aversion observed in the experiments
Explanations of Equity Premium
- survivorship bias:
- -> we only looked at the US market which has been the best performing market over the past 100 years
- -> if you look at the whole world, than the premium is not that big
- maybe we are measuring risk incorrectly or incompletely
- it is hard to estimate future expected returns with historical realized returns
- investor sentiment and psychology
Rethinking the equity risk premium
- long-term government bonds have gained 11.5% a year on average over the past three decades, beating the 10.8% increase in the S&P500
- on the other hand, many people claim “stocks are dead”
- retail investors seem to agree (they have been pulling money out of stock mutual funds every year since 2008)
- general consensus of academics and CFO’s is that the ERP is perhaps closer to 3-4%
Are stocks less risky in the long-run?
- annualized return volatility decreases with investment horizon. BUT, what matters is your wealth at the end of your investment horizon, not the annualized return
- short fall risk decreases with investment horizon
- but this ignores the size of potential losses, which for some of the possible outcomes amount to complete ruin
- if you end up with a loss at the end of the investment horizon, the magnitude of the loss tends to be bigger, when the horizon is longer
- a better way to quantify the risk of a long-term investment is the market price of insuring it against short fall
- -> the cost of such long-term insurance premium is very high and increases with horizon
- -> it costs 20% of the initial value of portfolio to insure against shortfall risk over 10 year
- -> a 25 year policy would costs 30% of the initial portfolio value
Diversify across time…
adding risks
Diversify across assets…
dividing risks
Expected portfolio return
E(rp) = wE(r1)+(1-w)E(r2)
Variance of a two-risky asset portfolio
variance = w^2variance(r1)+(1-w)^2variance(r2)+2w(1-w)Covariance(r1,r2)
Correlation between r1,r2
=cov(r1,r2)/SD(r1)SD(r2)
Covariance r1, r2
=correlation(r1,r2)SD(r1)SD(r2)
Diversification between two risky assets
- the portfolio has the same expected return as the individual stocks but it also has a smaller variance
- the portfolio variance will always be lower as long as the stocks are not perfectly correlated
- if correlation is sufficiently low we can find a portfolio with lower variance than either of the assets
How is the mean-variance boundary formed?
- we picked the expected return we want, and then choose the weights of the portfolio so that the variance of the portfolio is minimized
- -> min portfolio variance; subject to E(rp) = K
What does mean-variance portfolio theory tell us?
- the MV portfolio theory says that any investor will choose the optimal portfolio from the set of portfolios that:
- -> maximize expected return for a given level of risk
- -> minimize risk for a given level of expected return
How do you find the minimum variance portfolio?
Suppose you want to minimize the variance of a portfolio with many securities already in it…
- Find two securities already in the portfolio with different covariances with the portfolio
- add a little weight to the security with a lower cov(ri,rp)…subtract a little from the security with the higher covariance
- the portfolio variance is a little lower. Repeat
The variance of the portfolio will be minimized when all the securities have the same covariance with the portfolio
cov(r1,rp) = cov(r2,rp) = … = cov (rn, rp)
- an asset’s influence on a portfolio’s variance primarily depends on how it covaries with the other assets in the portfolio
How do you find the tangency portfolio?
- form a portfolio using all the risky securities
- find two securities already in the portfolio with different risk premium to covariance ratios:
E(ri)-rf/cov(ri,rp) - Add a little weight to the security with a higher ratio, and subtract a little from the security with the lower ratio
- Keep repeating steps 1-3
Finding the tangency portfolio
E(r1)- rf/cov(r1,rT) = E(r2) - rf/cov(r2,rT)…
- the ratio of risk premium to covariance is the same for all assets
Implications
Suppose the following is true:
- everyone is risk averse
- we all agree on the same values for the expected returns, variances, and covariances of the securities
Result: everyone will hold some combination of the tangency portfolio and diskless asset - the more risk averse you are, the higher your percentage of the diskless asset
- if people have different inputs, they will perceive different tangency portfolios and thus different mean-variance frontiers
How does investor risk aversion factor into portfolio selection?
- a portfolio manager will offer the same risky portfolio to ALL clients regardless of risk aversion
- the clients risk aversion comes into play in capital allocation (risky vs. risk-free), not in determining mix off risky assets
How could different efficient frontiers be obtained by portfolio managers?
- different constraints
- different inputs used
CAL vs. CML
The CML is a special case of the CAL, where the CML is tangent to the efficient frontier. Because all investors will hold the same risky portfolio, this tangent portfolio is the market portfolio.