December 8, 2008

A simple idea

I have written about the overall market a few times in the past (see here). The problem with market analysis is that it is fuzzy and not actionable. For example, if someone were to say that the market is undervalued, what does that mean and what should one do about it?

Most of the times, market analysis is just noise. Good to hear and entertain yourself, but not really actionable. That said, there are a few times when the overvall market levels can be analysed and some buy or sell decisions can be taken.

I have generally invested in the overall market via an ETF (exchange traded funds). An ETF can be bought or sold like a stock and esentially represents an underlying index (see here for more details).

I bought heavily (by my standards) in 2003-2004 and held the ETFs for 2-3 years. I started moving out by 2006 and was completely out by mid 2007. As I have written in the past, I have followed a very simplistic approach on investing in ETF’s : Buy when the PE is below 12 and start selling once it crosses 17-18.

There are several valid drawbacks of using this approach, such as
- PE data is based on historical data. However the market is not stationary (which means that the index of year 2000 is not same as index of 2008), and hence the data is not comparable across time horizons
- PE data is backward looking, whereas the returns will depend on what happens going forward.
- There is no hard and fast rule that the market is undervalued below a PE of 12 and overvalued above 18. If the economy is going into a recession then a PE of 12 is not low as the earnings are about to fall off.

In spite of all the above drawbacks, I have found that buying below a PE of 11-12 and selling above 17-18 works well over a 2-3 year time frame. The returns are not fantastic, maybe a 40-50% returns over a 2-3 year time frame. However as it involves a minimal effort, I think it is worth it. The above approach may not give big returns, but it definitely gives better returns than a fixed deposit.

I have also uploaded an analysis of the market data – PE, P/B etc for the last 9 years (see file history data.xls in google groups). The data is downloaded from the nse website. As you can see from the data, the market has been below the current PE (11.9) only 2% of the days. I personally think that the odds are decent at the current market levels.

12 comments:

VISHNU said...

Hi Rohit,

This is the time to pick stocks.(Senior Bonds as some say , But I am not sure how to play here).

There are plenty of stocks around where I getting my hands full.

I feel you are looking at historical data merely based on the price alone.I think you need to calculate intrinsic value , competitive position of each companies and buy. Though not a bad idea , I still dont do it because I dont understand some of the companies + I can get better price somewhere else.

Thanks
Vishnu

Anonymous said...

hello rohit,

I am looking for companies with long history of consistently paying dividends and solid growth in Indian mkts. Where can I find such information ?( with data atleast 20/30 yrs )

Regards,

Vikas Rana said...

Thanks Rohit for another good post. It think it is good to have some exposure to Index unless one is WB, depends on how much success one has with value picking over longer periods of time.

You're right in terms of PE estimates. Following is what some experts said about these same few months back:
"
http://www.nseindia.com/marketinfo/indices/indexwatch_nifty.jsp
I would put figures like this for India:
Acceptable : 10-12 PE, P/B < 2 and Dividend yield around 2.5%
Cheaper : 10 P/E, P/B <2, DY ~ 3%

If you see I have put 0.5% more on DY because due to higher inflation and higher cost of money, companies may tend to retain more of their earnings."

Vikas

Rohit Chauhan said...

hi vishnu
i am talking of buying the index based on a quick valuation approach. the gains are not very high, but decent enough considering the low effort involved.

invidual companies definitely have more value, but need more time and work

Rohit Chauhan said...

hi anonymous
i havent found any specific source for last 20 yrs of data. at best 10 yrs of data if the company provides it in the annual report

vic - A PE of 12 or less is decent. below 10, it would be very attractive.

regarding crisil - yes it is a part of my core portfolio

regards
rohit

Anonymous said...

Simple idea and I guess it will work. I too have been thinking about the same for the last 2 months :-)

Please let me know the ETFs (both Sensex and Nifty-fifty based) you find suitable.

Please also let me know if there are any broader BSE-500 type ETFs in the market.

I hope my ICICIDirect account is all I need for buying such a fund.

A random sweet guy said...

IMO, one should always be invested in stocks, fixed income and other asset classes (including gold and real estate). The asset allocation will depend on risk profile and current levels of market.

As market (and the PE) falls one can start shifting more money into stocks, and as it rises one can slowly move out of stocks. This can be a gradual (done once every 2-3 weeks) process.

If during the past 2-3 weeks, market has fallen by more than 5%, place a buy order (on any index stock you like, or a mutual fund). The amount can be fixed, or can vary depending on how much the market has fallen.

A random sweet guy said...

Another VERY IMPORTANT factor to be feared of, when investing in stocks is the 'Crash' of the market. It can wipe out the gains made over several months (or even years).

Personally, I've observed 2 minor crashes in the Indian market, and one major (the current subprime), and I can probably say that whenever the market falls by over 10% in a single day, take a hint and SELL completely. Return back after 2-3 months.

Rohit Chauhan said...

hi lucky
you can buy nifty BEES ..i think they are good enough. not sure if there is an ETF for BSE 500

Rohit Chauhan said...

hi amey
i agree with you portfolio allocation approach of changing allocation % based on market condition and valuation.

however i do not agree with your timing approach. selling when the market 'crashes' and coming back after 2-3 months is not a dependable strategy. it would have worked based on what we saw recently. clearly the future may not resemble the past.

also how do you define the crash - 10% drop, 3 % drop ? how do you arrive at the time to re-enter ..1 month , 2 months ? a lot of time this approach would mean sitting out the sudden turns in the market. i prefer to stay in the market if i believe in the stocks i am holding

regards
rohit

Anonymous said...

Dear Rohit,

Any specific reason for picking CRISIL over ICRA. Just looking at the numbers ICRA seems to be cheaper??

Respectfully,

Rohit Chauhan said...

Hi anonymous
no specific reason. i have followed crisil for a long time and found it be a good company and is now available at good valuation.
will look at ICRA too
regards
rohit