Old Investing Advice Gems from Wallstraits days

[freedom]

I understand that DCF can't precisely calculate the present value of the company, but it does give some indication with certain assumption as long as the earning/cashflow is not too volatile.

[D123]

Hence, I don't use DCF at all during my analysis. I use a more holistic approach where I look at quantitative and qualitative factors, and "simpler" approaches such as P/B and PER as well as NAV (in certain cases) works well enough.

In a sense, applying a multiple like P/E ratio to earnings to get a valuation is akin to a shortcut DCF calculation.

For example, say you make a few estimates of this company based on your study of its fundamentals:
1) owner's earnings of $X amount is easily achievable
2) $X can grow at 6% per year for 10 years
3) low interest rates would persist during this period so you use a low discount rate of 5%
These are estimates that you would usually need for a DCF calculation.

Using the annuity formula, you can calculate that you just need to slap a multiple of about 10 to $X to get a DCF valuation of those 10 years of cash flows.

So with a set of assumptions, you can derive a P/E multiple, which you can then use as a reality check. Of course, you can also arrive at the same multiple of 10 using another set of assumptions, but the point is to get your understanding of the company as on the mark as possible, and derive a set of assumptions that is most rational. Once you have that, a DCF calculation is as easy as referring to an annuity table.

Just my two cents.

[freedom]

I always thought PE is a derivation from ROE, PB, plus DCF etc...

[Satchmo]

Thanks D.O.G for ur comprehensive replies yet again. Now I am clearer on how to build up a proper portfolio instead of investing haphazardly. Probably will have lots more questions as I go along

DCF depends on the discount factor used, doesn't it. How does one arrive at what value to use?

But logically, like what MW said, DCF looks pretty difficult to use, especially in our constantly and rapidly changing world nowadays.

[wsreader]

Investment gurus can appraise a company and get roughly right about its valuation. Novice like me only know how to do the DCF and end up precisely wrong. But we will still do DCF to make us feel like we are carrying out due diligence

[d.o.g.]

hi d.o.g. From the last paragraph, it seems that you calculate the margin of safety using some other valuation methods. Can I know what they are? Or are you now more actively managing your portfolios to reduce the conservative requirement of margin of safety?

To repeat myself:

"I only believe that, based on their track record, their current balance sheet and the price I am paying, my future investment results will at least be satisfactory."

This seems pretty self-explanatory and conservative to me. I'm not asking for the moon - as long as we do not encounter depression-type conditions, the companies I choose will do OK, and assuming I don't pay too much to buy in, I will do OK too.

I would like to know how do you determine at what price to buy if you do not do intrinsic value calculation through DCF?

Also, I believe DCF can be done on companies with predictable cash flow for the past 5-10 years. With such predictability, we can project the cash flow into the future to get the intrinsic value. Since we might not have perfect projections due to random errors, we use a margin of safety (I prefer MOS of at least 25% for stable companies). What's your take on this? Thanks.

I did not say I do not use DCF. It is merely one of the many tools available. A good carpenter carries a toolbox with many different tools to cope with different situations. A good investor should do no less. So he should be familiar with DCF, P/E, P/B, Dividend Yield etc.

In my own experience I have seldom found companies whose operations were stable enough to be DCF-able. I already mentioned the shipping trusts and KGT as examples of DCF-able securities. Most companies' cash flows are too unpredictable for DCF to be of any real use. The old Wallstraits members who subscribed to the Intellivest service saw in real time how most of the Intellivest portfolio companies did NOT live up to the expectations of the DCF analysis.

Use the right tool for the right job. As Charlie Munger has said, to a man with a hammer, everything looks like a nail. Sometimes the object in question IS a nail and the hammer works great. But many other times it isn't and the results are unlikely to be good. So don't be a man with a hammer (an investor who can only DCF) - carry a toolbox (learn to use many valuation methods).

I understand that DCF can't precisely calculate the present value of the company, but it does give some indication with certain assumption as long as the earning/cashflow is not too volatile.

Herein lies the HUGE caveat: "as long as the earning/cashflow is not too volatile". A random survey of say 30 companies over the last 5 years will make it abundantly clear that earnings and cashflow ARE very volatile. In other words the usability of DCF is more the exception than the rule.

DCF depends on the discount factor used, doesn't it. How does one arrive at what value to use?

Aha. You have found the devil in the details of DCF. In theory the discount factor is taken as a premium to the long-term government bond rate, on the basis that the long-term goverment bond represents a risk-free return and is therefore the floor in terms of required return. In practice the premium to be used varies wildly as different people ascribe different premia to country risk, business risk, forex risk, governance risk, commodity price risk, interest rate risk etc.

As a result you end up with "garbage in garbage out". With DCF and an arbitrary discount rate you can get any NPV you want. This leads to DCF being popular for (ab)use by consultants in M&A deals - if the buyer WANTS to buy, the consultant will obligingly project future cashflows optimistically and use a low discount rate so that NPV is very high and the purchase price looks like a discount.

[thinknotleft]

As I recall I had a debate with Sage over his reasons for selling one stock to buy another. His rationale was that stock A was better than stock B so he sold B to buy A. My quibble was that he didn't show that stock B was the worst holding in the portfolio, that it was not only worse than A which wasn't in the portfolio, but that it was also worse than C, D, E and F which were already in the portfolio. He refused to discuss the matter further so that was it.

I have some rough recollection that I used to agree fully with d.o.g. and brk on this. Now, I will agree to some extent.

A stock need not be the worst holding, in order for one to sell it to buy other stock. It could be that the stock has occupied too large a proportion of the portfolio. Hence, the stock is first sold partially for diversification's sake.

Or it can be a change in philosophy. The stock is first bought on the high risk, higher gains philosophy. The investor becomes more cautious. So he sell the stock bought on a high risk, higher gains basis, while keeping the lower risk (and maybe lower reward-to-risk) stocks.

Or, it can be sold for non-investing reasons i.e. the key investor threatens to withdraw its funds if the stock is not sold.

I think in the end fundamentally the big divergence was whether DCF should be the only valuation tool used. Sage felt so, while BRK and I did not agree.

Today I seldom use DCF because I rarely find businesses that lend themselves well to a DCF analysis. Exceptions include the shipping trusts and KGT, where revenues are fixed for a long period of time and expenses are modest in relation to revenues.

There may be many ways to adjust DCF for riskier or less stable businesses.

One is to increase the discount rate. Two is to use multiple DCF which are then multiplied by your subjective probabilities (the probabilities should sum to 1). Three, multiply the DCF by a risk discount factor (say 0.8 x DCF value). etc.

Personally, I perferred discounted earnings (DE) approach over discounted cashflow, since there is an academic paper that notes that DE has similar performance as DCF and DE is easier to compute.

[freedom]

I think in the end fundamentally the big divergence was whether DCF should be the only valuation tool used. Sage felt so, while BRK and I did not agree.

Today I seldom use DCF because I rarely find businesses that lend themselves well to a DCF analysis. Exceptions include the shipping trusts and KGT, where revenues are fixed for a long period of time and expenses are modest in relation to revenues.

There may be many ways to adjust DCF for riskier or less stable businesses.

One is to increase the discount rate. Two is to use multiple DCF which are then multiplied by your subjective probabilities (the probabilities should sum to 1). Three, multiply the DCF by a risk discount factor (say 0.8 x DCF value). etc.

Personally, I perferred discounted earnings (DE) approach over discounted cashflow, since there is an academic paper that notes that DE has similar performance as DCF and DE is easier to compute.

I agree that to find a perfect DCF-able company is very difficult. but to certain extent, as long as the earning volatility is within a range, adjusted DCF/DE (plus increase the discount factor, projected growth/discount rate) is not that bad to get a sense of margin of safety. Of course, if it is to be used to calculate the intrinsic value, it could be very off.

[wee]

Sage's business model with Intellivest is flawed in my view. A subscriber expects to learn from his every decision and thought process. However, many times i felt he was holding back in his sharing, and this leads to frustration. Perhaps he know something that is not suitable to be shared publicly, or because he had made some bad choices or errors which he was too embarrassed to admit. Afterall, he is supposed to be Sage.

His DCF is also too simplistic IMO. 10% or 15% p.a. growth in earnings for 10 straight years is very aggressive assumption. I don't see the point of making aggressive assumptions and then try to apply a 50% MOS. Valuation should be reasonable, or MOS will not be meaningful.

Lastly, his screens looks good in theory, but it failed to screen out the bunch of suspected fraud companies.

[yeokiwi]

I had taken a look at my cdp record from 2001 to 2011. The changes in the portfolio were very significant. Even from year 2007 to 2011, the changes were rather drastic too.

So, in these years, it is either I was very fickle minded or the stocks' valuation had changed such that I was divesting them away and purchased other stocks that met the selection requirements.

It is not easy to hold a stock for ten years. The fundamental of the stock may change or a better stock may appear that cause me to switch from one to the other.

Although at the last count, there are more than 5 stocks that are in my portfolio for more than 4 years.

Not suprisingly, the longest lasting counter is STI-ETF that I bought using CPF. more than 8 years.