Selecting Good Companies (ROE, PB and PER)

[arthur]

P/B, by itself, is quite meaningless. You have to use P/B with other information like ROE etc. Even in the situation where P/B is more applicable, P/RNAV (or P/Expected book value) is more useful.

You are right to point out that P/RNAV is more useful. However the word "expected book value" is already self explanatory.
Not many ppl would be able to calculate to a good estimation. In fact, given a recessionary biz climate (which P/B would approach the lowest), the probability of P/RNAV getting lower is higher than avg (ie. sell assets for cost reduction,etc)

Finally, I wish to reiterate one of my earlier points. It is very seductive to conclude from d.o.g's analogy that one should always buy good business (or high ROE business) at reasonable prices. If you take a step back, the assumption behind d.o.g's analogy is that good business now implies good business in the future. And, this assumption is very tenuous, given the mean-reverting tendency of business performance.

Well.. for d.o.g analogy that good biz now implies good biz in future, that would depends also on the company biz itself.

Let's have a look at our forum top left header.

It shows Ben Graham with a caption "Investing is most prudent when it is most biz-like" simply implies by understanding the biz of the counter which one is buying, there exist a higher probability knowing whether this biz would survive thru the recession and emerge stronger.

Example: PG.

Proctor and Gamble-> a mega cap counter with hundreds of products selling around the world. Its products? Simply soap powder, toothpaste, dish washing detergent, etc. Yet, we can believe strongly no matter how bad the recession, PG will not collapse short of an accounting fraud like Enron.

PG P/B ratio approaches 2 when it was at March 09. Low P/B if we look at the value historically.


Still I respect thinknotleft's viewpoint on this and the reason why I'm bringing this up together with a case study is for discussion purpose and to let the newbies differentiate employing the P/B value freely vs. using it together with the understanding of knowing the biz organisation well.

No ill feelings alright?

Cheers!

[mrEngineer]

P/B, by itself, is quite meaningless. You have to use P/B with other information like ROE etc. Even in the situation where P/B is more applicable, P/RNAV (or P/Expected book value) is more useful.

Finally, I wish to reiterate one of my earlier points. It is very seductive to conclude from d.o.g's analogy that one should always buy good business (or high ROE business) at reasonable prices. If you take a step back, the assumption behind d.o.g's analogy is that good business now implies good business in the future. And, this assumption is very tenuous, given the mean-reverting tendency of business performance.

I did not know RNAV means expected book value. Learnt something new today. I was actually calculating future NTA values based on expected ROE and I term it as BV. So perhaps I was indirectly calculating RNAV too.

Yes I agree that fantastic business performance does not continue forever. Thats precisely the reason why I am asking when can we find P/RNAV becoming overvalued.

For e.g. if a company is now trading at P/B of 6 and assuming it will have consistent ROE of 35% performance for the next few years. The 70 rule of thumb will say that the company will double its NTA in 2 years and quadruple in 4 years. But now the P/B is already 6 which means for the NTA to reach the current price, it will have to take between 4-5years. Is it considered overvalued?

[d.o.g.]

Finally, I wish to reiterate one of my earlier points. It is very seductive to conclude from d.o.g's analogy that one should always buy good business (or high ROE business) at reasonable prices. If you take a step back, the assumption behind d.o.g's analogy is that good business now implies good business in the future. And, this assumption is very tenuous, given the mean-reverting tendency of business performance.

I fully agree that assuming good past performance means good future performance is a very dangerous assumption. However I think the odds improve a lot when the business is of a stable nature e.g. consumer products or recurring services.

Indeed, the companies I have found that have had good 10 year records are mostly in this category. I am quite confident they will continue to do at least as well in the future, provided there are no significant changes to management or business model. If I can buy them at a good price I would be quite happy to leave them alone to grow for 10 years. I would be very surprised if I came back 10 years later and my investment had not grown significantly in terms of book value, cash dividend payout, and market value.

For companies that are benefitting from temporary supply/demand imbalances, reversion to the mean is quite likely as the high returns encourage supply to increase and eventually drive down returns.

If one is not sure that the business is of a type that can continue to generate high ROEs in future, then it is probably better to stick to a Graham-type approach i.e. buy ordinary businesses at cheap prices.

Buying a 10% ROE business at 0.5x book value will also yield highly satisfactory results after 10 years, since $1 of book value compounded at 10% will become $2.59. Bought at $0.50, the investor will have quintupled his investment if it now sells at book value. At a more conservative 0.75x book value, he has still almost quadrupled his money. Graham's own experience suggested that the P/B multiple improved quickly enough that followers of this approach need not wait 10 years. Indeed, Graham suggested selling after making 50% or waiting 2 years, whichever came first.

Perhaps the common idea is to pay less in P/B than the ROE would suggest. So a 10% ROE business should be bought below book value, a 20% ROE business below 2x book value etc. And of course you need time to help out. Without the effect of time, compounding cannot do its magic.

For e.g. if a company is now trading at P/B of 6 and assuming it will have consistent ROE of 35% performance for the next few years. The 70 rule of thumb will say that the company will double its NTA in 2 years and quadruple in 4 years. But now the P/B is already 6 which means for the NTA to reach the current price, it will have to take between 4-5years. Is it considered overvalued?

There are very few companies that can average 35% ROE for any decent period of time. The ones I know of do so by paying out a very high proportion of earnings, so their book value doesn't really grow. In the current case, you pay $6 for $1 of book value which generates $0.35 in earnings. Assuming a 100% payout ratio you get ~5% yield. OK yield, but if the business doesn't continue to generate the same ROEs the investor may suffer a cut in dividend, plus losses on his investment capital.

If a company can indeed compound internally at 35% for any decent period of time it should rapidly become a giant in its industry, since (by the rule of 72) it will double every 2 years and (by the rule of 115) it will triple every 3.3 years. So in 10 years it will be over 27 times(!) its present size.

If you think about it, this would be an incredible business. Nobody in their right mind would list such a business - equity would be too precious. Just borrow some money from the bank, since you'll be able to pay it off in double-quick time.

[mrEngineer]

Digging up a old post to discuss something that came across my mind. As we frequently advocate stable consistent and high ROE as an indicator of company performance and can be used as a method to shortlist good companies., how about we use FCFE / Equity instead? The reason is because Income figures are often manipulated and sometimes can create extreme values of ROE. If we use FCFE instead, we can say that operating cashflow can be less manipulated and it can be used to assess extremely stable cashflow companies.

Any comments?

[D123]

Digging up a old post to discuss something that came across my mind. As we frequently advocate stable consistent and high ROE as an indicator of company performance and can be used as a method to shortlist good companies., how about we use FCFE / Equity instead? The reason is because Income figures are often manipulated and sometimes can create extreme values of ROE. If we use FCFE instead, we can say that operating cashflow can be less manipulated and it can be used to assess extremely stable cashflow companies.

Any comments?

It might be helpful to define FCFE before we start discussing using different interpretations of FCFE. How would you calculate it?

____________________________________________________________


ROE is return on equity, and doesn't explicitly advocate using reported earnings. So far, I don't think you can go too wrong using owner's earnings to calculate ROE.

For a definition of owner's earnings, I will defer to Buffett:

"If we think through these questions, we can gain some insights about what may be called 'owner earnings.' These represent (a) reported earnings plus (b) depreciation, depletion, amortization, and certain other non-cash charges such as Company N's items (1) and (4) less ( c) the average annual amount of capitalized expenditures for plant and equipment, etc. that the business requires to fully maintain its long-term competitive position and its unit volume. (If the business requires additional working capital to maintain its competitive position and unit volume, the increment also should be included in ( c) . However, businesses following the LIFO inventory method usually do not require additional working capital if unit volume does not change.)"

This is taken from his 1986 letter to Berkshire Hathaway shareholders. It is worth reading the whole section in the Appendix to get the context.
The full letter can be found here:
http://www.berkshirehathaway.com/letters/1986.html

[mrEngineer]

I use this definition. FCFE = operating cash flow – capital expenditures – Interest Expense and Debt Repayments

[Drizzt]

I use this definition. FCFE = operating cash flow – capital expenditures – Interest Expense and Debt Repayments

this is usuallly negative? hard to be positive unless it generates alot of astounding cashflow

[Satchmo]

Using FCFE is a good idea, but of course have to watch out for and discount extraordinary cash inflow/outflow.

Usually negative? Have to depend on the industry, biz nature of the companies and their accounting policies i guess. I am sure there are plenty of companies that generate regular positive FCFE.

[freedom]

I think it is negative probably because in the formula "FCFE = operating cash flow – capital expenditures – Interest Expense and Debt Repayments"

since debt repayments are deducted, how about borrowings or equity raising? added in?

for those large cap companies, maybe they are in net cash position, they normally maintain some amount of debt. and the debt probably roll over year after year. if debt repayment is deducted, it is quite easy to have this "FCFE" negative

[yeokiwi]

There are companies that do not have any debt or very little debt and at the same time, their capital expenditure is low.
Vicom, Kingsmen are in this category.Even road builders like OKP is also having positive FCFE.

I think it is also highly related to the kind of industry that the company is in. For property companies, it is hard to have positive FCFE annually due to debt repayment but once in a few years, they may exhibit tremendous cashflow..