טכניון מכון טכנולוגי לישראל
הטכניון מכון טכנולוגי לישראל - בית הספר ללימודי מוסמכים  
M.Sc Thesis
M.Sc StudentPotashinski Irina
SubjectThe Equity Premium Puzzle and Indexed Bonds
DepartmentDepartment of Industrial Engineering and Management
Supervisor Professor Avner Bar Ilan


Abstract

  The equity premium puzzle was first presented by Mehra and Prescott (1985). The data examined by Mehra and Prescott (1985) showed a significant 6.18% annual equity premium. At the same time, according to their theoretical analysis and based on earlier estimates of typical risk aversion parameter of individuals, this gap in average returns should be markedly smaller. Since most individuals are risk averse, the riskier the asset, the larger should be the risk premium. However, this premium is too high for the additional risk of stocks relative to bonds and conventional risk aversion parameters. So there is no puzzle in the existence of the risk premium, but its size.  We can therefore say that the puzzle is quantitative.

There are two categories of explanations an empirical and a theoretical. We have focused on the empirical side of the problem. The aim of this research is to examine the existence of the equity premium puzzle in Israel and other countries with different types of relatively riskless securities; in particular bonds indexed to the consumer price index. It is common to test the existence of the equity premium puzzle when the riskless securities are nominal bonds. In addition to nominal bonds we have used CPI indexed bonds and bonds indexed to a basket of foreign currency. The idea was that indexed bonds are a better measure of riskless security since their real, not nominal yield, are relatively constant.

      We have also tried to examine if including indexed bonds can partly solve the equity premium puzzle and if the result is unique for Israel or is common among  other countries, where indexed securities exist.

    For Israel we find that for the arithmetic sample in most of the examined periods the equity premium puzzle exists. In the geometric sample we got smaller values, but still in most of the examined periods we found that the equity premium puzzle existed for unindexed bonds and for foreign currency linked bonds. For consumer price index linked bonds, in the geometric sample, in most of the examined periods we found that the equity premium puzzle did not exist.

     The result that we received for Israel is that including linked bonds can partly solve the equity premium puzzle. The result is unique to Israel, and does not extend to other countries.