Lecture 12

Question 1

Why is, using equal weights, the average beta these days higher than 1?

Answer 1

Nowadays, there are more small firms who haver higher beta than larger firms who have lower beta and since the weights are equal, true market beta tends to be higher than 1.

Question 2

What risk indicator does CAPM use?

Answer 2

Beta, not volatility (sigma)

Question 3

What is the slope of a SML?

Answer 3

Sharpe Ratio

Question 4

How do we test CAPM, according to Fama and French? Also include the 3 empirical steps.

Answer 4

A positive risk-return relation (SML) lies at the heart of finance theory and is a testable hypothesis. CAPM predicts that expected returns increase with beta. We measure it in 3 empirical steps;
1. Estimate betas (time-series regressions)
2. Approximate expected returns by averaging realized returns
3. Estimate CAPM Risk Premium (cross-sectional regression)

–> We need the expected returns and estimated betas and we regress them cross-sectional.

Question 5

What does the SML show?

Answer 5

The linear relationship between expected returns and systematic risks

Question 6

What outcomes are possible from this test?

Answer 6

1. Disaster –> negative relationship between returns and beta
2. No evidence to support –> no relationship between returns and beta
3. Good outcome –> positive relationship between returns and beta (as expected according to CAPM)

Question 7

What is a time series regression in the Fama-Macbeth test?

Answer 7

One firm regressed over multiple periods

Question 8

What is a cross-sectional regression in the Fama-Macbeth test?

Answer 8

Multiple firms in one point in time

Question 9

How do Fama Macbeth estimate the betas without recursively method?

Answer 9

Regress stock returns on market returns

Question 10

What is recursive estimation in estimating the betas and the cross-sectional returns?

Answer 10

We estimate betas using a rolling window.

Question 11

According to CAPM, what should be the two coefficients?

Answer 11

The first coefficient should be zero because all returns should be explained by the Beta and not by the Alpha.
The second coefficient is the slope and according to CAPM, should be the Market Return Portfolio (MRP).

Question 12

How to obtain the Fama Macbeth estimates?

Answer 12

We estimate the betas using a rolling window period (recursive). We then regress a beta of one rolling window on returns of the next period and we get coefficients for every rolling window. We then average all these coefficients and obtain the Fama Macbeth estimates.

Question 13

In the graph, what turns out from the Fama Macbeth tests?

Answer 13

The estimated SML is flatter than the predicted CAPM.

Question 14

Since the estimated SML is flatter than the predicted CAPM, what will investors do?

Answer 14

Since the estimated SML tells us that low beta stocks (with lower risk) actually get higher returns than predicted by CAPM, investors will buy low beta stock and sell or short high beta stocks because you actually get less return than predicted by CAPM for higher risk.

Question 15

With an estimated flatter SML, and investors buying low beta stocks and selling high beta stocks, what happens over time?

Answer 15

If everyone buys low beta/high alpha stocks, the alpha decreases and vice versa.

Question 16

In CAPM, only beta matters so not any other factors like size or book-to-market ratio.

Question 17

What is alpha?

Answer 17

Excess return that is not explained by beta. In other words, the risk-adjusted excess return of the stock.

Question 18

What is an alternative test to test CAPM?

Answer 18

GRS test. All alphas should be zero.

Question 19

What are 3 critics to the CAPM test?

Answer 19

1. CAPM predicts expected returns and in the test, they use realized returns.
2. CAPM is not testable unless we know the exact composition of the true market portfolio = joint hypothesis problem
3. Benchmark error due to proxy for market, which means not the entire market portfolio is used. Because only a part of the market is used in the model, the estimates tend towards zero, which is attenuation bias.

Question 20

How to find other risks next to beta?

Answer 20

1. Data-driven. Risk = comovement among stocks so we can use data to spot variation in returns. A small number of factors account for the most variation in returns.
2. Economic thinking.
3. Fama-French model. Beta, size and book-to-market ratio

Question 21

What method do we use for the data-driven approach and what is an advantage of this method?

Answer 21

Principal component analysis. Five factors are generally enough to match a lot of variation in portfolio returns.

The advantage of this method is that it is data-driven, so we do not need to guess factors.

Question 22

What could be risk factors that investors want to hedge against other than beta?

Answer 22

Labor risk
Unexpected inflation
Consumption risk
Changes in investment opportunity set

Question 23

What is the essence of a hedge?

Answer 23

You want to hedge against the risk of an asset by taking a position in another asset that makes the correlation of the two close to zero or even negative.

Question 24

How do we measure these additional sources of risk?

Answer 24

1. We use betas from Fama Macbeth estimates.
2. We need a ‘return’ of the risk. This will be a mimicking portfolio that tracks the changes in the risk factor.

Question 25

How to obtain a factor mimicking portfolio?

Answer 25

Go long on shares high in the certain risk
Go short on shares low in the certain risk

Return of the certain risk = Long position - short position

Question 26

Does a factor mimicking portfolio has a high risk beta?

Answer 26

Yes because it is long in high risk stocks and short in low risk stocks, therefore the beta is relatively high

Question 27

What is the relationship of a long-short portfolio (factor mimicking) to the risk-return trade-off?

Answer 27

High returns on the portfolio when returns of risk are high. Low returns when risk returns are low.

Question 28

How to hedge these kinds of market risks such as labor risk?

Answer 28

Go short in high labor risk stocks and long in low labor risk stocks.

Question 29

If rising temperatures cause economic activity to slow down, then stocks that do well when temperatures rise are good for the investor. This makes
the climate risk premium negative. Why?

Answer 29

If rising temperatures cause economic activity to slow down, then stocks that do well when temperatures rise would be seen as hedges against climate risk. This would imply that investors would accept lower returns on these stocks, as they offer protection in a scenario where temperatures are rising and the economy is slowing.

In this case, the risk premium for climate risk, would be negative. This is because investors would demand lower compensation (a negative premium) for holding stocks that perform well in adverse climate conditions (i.e., rising temperatures).