Fellow Monkeys - just soliciting some technical expertise on behalf of a friend. Here is his message to me:
"yo! i wanted your help with something. have you read anything Aswath Damodaran has written? if not then you might have to. i have to find out the risk premium for the BSE 200 or any other indian indices, through the CAPM method. to do so you need a risk free bond rate, the equity return and the beta for stocks. the problem that i came across is that the indian stock markets have been around for 25-30 years now, however the power sector stocks (which i am basing my analysis on) have been around for only 4-5 years. the question then arises if i am calculating the equity return on the indices for a 25-30 year period, then for how long should i calculate the beta for the power sector stocks to figure out the risk component. as you see there will be a mismatch in time periods, between market returns and its risk profile. this aswath damodaran guy is supposed to have written papers on the same problem. that's why i asked.
let me know if you can give me an answer and the hypothesis you worked on to get it. if its a publicly available document that would be great."
It's an interesting question. My first instinct (I have not read any Damodaran) is to say that it would be some form of weighted average...but I have a suspicion that the solution is not a simple linear one. Anyone have any ideas?
no. equity risk premium is independent of sector and therefore your beta calculation, so it doesnt matter. normally you would have a global ERP, now its often estimated at 5-8%, and then add a political risk premium to cost of equity for your country. 5 years is more than enough for a beta calculation for the sector in your specific market.
.....
Isn't the market beta always 1? (assuming we are looking at the BSE in isolation?)
The market risk premium would just be the average market return minus the average risk free rate for whatever timescale you were looking at.
Or have I misunderstood the question?
Yeah, time series for ERP and beta should be different. Most commonly, beta is estimated using a regression of monthly data over a 5 year period (60 observations). This is what Ibbotson uses, and Damodaran recommends in his textbook. ERP is typically analyzed over the entire available history. For example, in the U.S. the ERP is usually estimated as the difference between S&P500 returns and U.S. govt bond returns from 1926 to present. As Damodaran warns, you get into differences depending on whether you use an arithmetic or geometric average to calculate the ERP. Also, some people use 3 month t-bills as the risk-free rate, and some use long term govt bonds (10 year). This will produce large differences in ERP estimate (it will swing from 4.5% to 8.5% depending on assumptions). But yeah, for international CAPM build-up usually use a global index and add country risk premium for political risk, etc.
Check out this paper:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1274967
Sorry, I know kinda a lot about this because it's always a hotly contested issue in litigations.
And Curlyarmstrong, in theory the market beta is one, but I think he's asking specifically for power stocks in India, which are just a small sub-segment of the global market, and will probably have a beta different than one.
Thanks, homies. Input much appreciated.
Sequi sed commodi veniam doloremque tempora. Id tenetur corrupti eos eveniet. Perspiciatis unde vel consequatur qui molestias rerum.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...