Equity Premium Puzzle

Question 1

What is the equity premium puzzle (Mehra & Prescott, 1985)

Answer 1

Found that equity risk premium between 1889 and 1978 imply a relative risk aversion that is significantly higher than 250 vs expected value ~1

Question 2

Explanation 1 - Selection Bias

Answer 2

Study focused on US

• -> most success stock market in 20th Century
• -> other countries displayed LR returns but still had + RP
• -> using US same as choosing outlier from sample, creates upward bias

Question 3

Explanation 2 - Survivor Bias

Answer 3

Exchanges can go bust/be disrupted by factors such as war, hyperinflation, political revolution

• -> e.g. shangai SX shit down during 1949 during communist takeover
• -> risk of disruption/shutdow needs to be captured
• -> using only EX that survive overstates returns
• -> surviving EX also have longest time series/more data points can’t observe counterfactual for closed EX

Question 4

Explanation 3 - Cannot Observe Expected Returns

Answer 4

Only actual returns are observable

• ->Fama and French (2002) suffest 2nd 1/2 of 20th Century saw constant actual returns > exp returns
• -> Goetzmann and Ibbotson (2005): equity risk premium between 1792 and 1936 was only 3.6% in US

Question 5

Explanation 4 - Poor Proxy for Consumption

Answer 5

GDP figures extrapolated from surveys and include a lot of smoothing

• ->Savov (2011) uses garbage as a measure of consumption using quarterly estimates published by US Environmental Protection Agency
• -> more volatile than GDP consumption
• ->uses methid to CRRA from 81 to17 (still too high)

Jagannathan & Wang (2007): create consumptio tracking portfolio using Q4 on Q4 consumption
–>can be tracked more frequently than aggreegate consumption

Question 6

Explanation 5 - Limited Data Set

Answer 6

1900-2005 only provides 105 independent years
–> too small for statistical analysis

Not sufficient to explain away equity premium puzzle though

Consensus is that findings supporting EPP have statistical power

Question 7

Explanation 6 - Irrationality

Answer 7

ERP puzzle due to excessively strong assumptions in models of investor behaviour
–> real life investors inconsistent with model assumptions

Question 8

Explanation 7 - Loss Aversion

Answer 8

Bernatzi and Thaler (1995): Investors extremely risk averse suffer from “myopic loss aversion”

• -> unwilling to invest in high risk assests w/o high risk premium
• -> keen to maintain current level of consumption

Question 9

Explanation 8 - Narrow Framing

Answer 9

Investors see each individual investment for its inherent risk–> Inv decisions based on idiosynchratic risk
–> doesnt apply well to fund managers–> dont realise losses fro stock price falls