Comments on: WOW! Prices Roll Back

By: Dean Morel

Dean Morel — Tue, 09 Mar 2010 10:56:39 +0000

Thanks for the link Sean, interesting stuff. I also penned you a long reply to your question on valuation, but it seems to have disappeared. Bother that! In summary valuation is more art than science. The reason changing the terminal rate changes the value so much in your example, is that your terminal rate starts so soon. If it started at year 10 then it would have a much smaller effect.
I’d argue mean reversion works as well for an established company as it does for a country. I know there was more, but brain dead now after a late lecture.
cheers – Dean

By: Sean

Sean — Thu, 04 Mar 2010 09:53:05 +0000

Hi Buffetfan, thanks for your thoughts!

I’m not sure if Buffet would look at Telstra. Unfortuantely I bought some and I’m increasingly getting worried it’s a value trap. Still, he does have a large stake in Burlington and Conoco Phillips which to me don’t particularly strike me as sticking to his principles. I suppose the devil is in the detail and application. I think I’ll apply a 15% stoploss to my Telstra purchase.

I found this site interesting.
http://www.empireinvesting.com.au/

They attempt to apply classic value investing principles. I also like that they share their info and that it is about the local market.

By: BuffettFan

BuffettFan — Mon, 01 Mar 2010 09:44:47 +0000

Sean
While I don’t use terminal growth rate in the sense that you do, I think I understand your position. Yes, I agree that the calculation results are very sensitive to some variables. I have attempted to address the problem these ways:
1. Develop a short list of investments. Over the years, I have reduced the list to the 300 largest Aust equities because I have been burnt by exchange rates, consider that other investments don’t have the long term returns, small stocks don’t have enough liquidity or competent management etc
2. Only invest in industries that I understand (usually because I have worked in them.) Hopefully that helps narrow the range of forecasts.
3. Only invest in companies with strong fundamentals eg good financials, quality management and all the other things Buffett et al talk about
4. Research as widely as possible, listen to analysts’ opinions, but be prepared to avoid going with the crowd.
5. Prepare a best guess, optimistic figure and a pessimistic figure to calculate a likely range of values
6. Then compare that range to the other stocks that look likely at the moment. Often, the entire range of one stock will be different to the range of another. I don’t try to do Monte Carlo or anything too “scientific” as all I am trying to do is decide if stock ABC is a better long term bet than stock XYZ.
7. Then compare all the other fundamentals to get an overall comparison between the stocks.
8. Then go with what I think is best

Yes it’s messy, but after studying stacks of books and doing lots of simulations, so far it’s the closest I can get to emulating Buffett.
Hope this helps…….

By: Sean

Sean — Sun, 28 Feb 2010 03:28:33 +0000

Hi Buffetfan and Dean, one problem I have with fundamental analysis of individual companies is that the valuation/NPV and IRR are affected greatly by the terminal growth rate, which is the hardest variable to know with any accuracy. In fact I would say it’s always a massive guess.

I would argue that it is easier to estimate the fair value of an index for a country than any company because you have a reasonable estimate of GDP growth and profit share% and can invest based on mean reversion.

For example with Telstra currently, almost every analyst report I read has a terminal growth rate of 2%+. The market from what I can see values it at about 1% at current prices. Half the time the market turns out to be right. There is too much firm specific risk to be able to rely on mean reversion for individual stocks for it to be more than a guess.

Here are some scenarios, all with moderate 9% discount rate :
1. Telstra revenues down 10% in next 3 years then zero terminal grwoth rate NPV: 2.2
2. Revenues down 10% next 3 years, 1% terminal growth rate NPV 2.5
3. Revenues down 10% next 3 years, 2% terminal growth rate NPV 3.0
4. Revenues down 10% next 3 years, 3% terminal grwoth rate NPV 3.6
5. Flat revenues for next 3 years, 0% terminal growth rate NPV 2.9

From 0-3% terminal growth rate, the valuation of the company changes by 60% (ie $1.4). Obviously the market and actual outcome depend on management, the sector, government, economic growth, but how can you accurately predict the terminal growth rate within a 0-3% range ? Of course the range could be larger than this say -1 to 5%

By: BuffettFan

BuffettFan — Wed, 24 Feb 2010 10:58:37 +0000

Valuation model: a spreadsheet based on Buffettology, StockVal, and StockValue (by John Malloy) plus a lot of fine tuning by reading Buffett, Graham, Fisher et al. One objective is to avoid a lot of hassles and mathematical issues by estimating the (after tax)IRR of investing in each stock for 5 years. Then rank the stocks by IRR. Then start at the top and eliminate those with poor fundamentals, low liquidity etc. Then invest in about the top 20. Continually monitor and recalculate. Rebalance whenever major changes occur, probably half a dozen transactions a year.
Lots of assumptions – require a fair bit of industry, accounting knowledge etc.
The principal is to buy a stock only after comparing it to all the other candidates on the short list. After all, we don’t buy a stock in isolation: there is always a hidden opportunity cost, even if it is only a bank deposit rate.

When I retired, I spent almost an entire year (full time) researching all the Technical/trading techniques, bought some trading software and did countless simulations and paper trades. At the end, I came to the conclusion that short term trading was a lot of work and the after tax results couldn’t beat the All Ords over a reasonable time frame. That’s when I became a Buffett fan, hopefully for obvious reasons.

By: Dean Morel

Dean Morel — Wed, 24 Feb 2010 09:41:59 +0000

Hi BuffettFan, Thanks for your comment. I’d love to discuss your valuation model someday.
On StockVal. I just checked and funnily WOW is ranked 23rd out of the stocks they cover! Fancy that. It was fifth and now third on their “very low required return” list.

Congrats on buying WOW and having the temerity to hold for so long. More investors could profit from your time horizons.

I hope you share more about your investing success as this site is all about helping individual investors.

By: BuffettFan

BuffettFan — Mon, 22 Feb 2010 22:57:19 +0000

I’ve studied the books by Brian McNiven (the developer of StockVal) and I vaguely remember disagreeing with some of his assumptions in the model. Consequently, I think that Woolworths is a reasonable buy at the moment, but not a screaming bargain. My model forecasts that if you buy it and keep it for 5 years, then you’ll probably earn 14% pa in dividends and capital gains (lots of assumptions, of course). The most likely range of returns is between 5% and 21%. This ranks WOW at 23 out of 50 on my list of desirable investments.

Which reminds me, I bought WOW in about 1993. I still have them, but they will probably go once I have done a little more research after the recent half year results.

Which reminds me, I created my own indexed fund in 1993 when I was too busy at work to devote the time to investing. I just bought the top 10 stocks and made their weighting in my portfolio proportional to their market capitalisation. I hardly ever sold anything, just rebalanced by buying when I added more money. It worked a treat, followed the All Ords closely enough and I could control the capital tax effect of sales. After taxes and fees, it beat most of the fund managers year in year out. I think it has to be the best return per hour of your time spent of all the investing techniques.

Has anyone else tried this?