Even if you buy long term, you need to care. Valuation is very important. Go google finance and look at the 10-year chart of Microsoft, Wal Mart, Pfizer and Johnson and Johnson. All this are super blue chips in US. During the early 2000-2002, there is a big cap bubble in US. Those folks that buy big caps during that period and hold it long term, practically have zero capital gain. The earnings do grow as forecast, but, the multiple contracts, leaving shareholders with almost zero capital gain. Their dividend yield aren't that great either.

Great stuff there kinwing as well as other forumers who contributed. Yes, process is extremely important. Most people thought when the price of the share they long go up, then, their process is correct, this is a wrong way to view investments. Sometimes, a good process may still lead to bad result. But, if your process is sound, you should do well over the long run.

Some of us here may not be comfortable in revealing our position because whenever you make your position public, you feel a certain obligation to stick to that position and you can't think independently.

For those who like to read message board for tips, read this piece of research paper : Confirmation Bias and Investment Performance : Evidence from Stock Message Board

As for criticism on DCF model, actually, all the PE ratios and stuff can link back to DCF model. DCF is good because it can allow you to look at what drives the company as well as stimulate various conditions. As long as your cash flow forecast is consistent and your inputs are logical and consistent, then, the valuation derived should be a good guide. Do a sensitivity analysis on your inputs. With excel, it is rather easy to do all this stuff. Stress test the company and try to kill it, if the company still survive, then, most likely you got a bargain in your hands. As for some common errors on DCF, I have share this piece in the Value Investing thread before but I am going to share it again on this thread. Download the attachment below.

On air asia, please take a look at SIA performance since its stock listing. SIA is the most profitable airline in the world. If Air Asia turn out well, it would probably be like SIA and that's a big IF. But, if you look at SIA financials, you will know why Air Asia is not a good idea especially at such a price. It may go up, but, it doesn't means it is a good company.


Attached File(s)
 CommonErrors.pdf ( 106.4k ) Number of downloads: 23

I think you may have misunderstood the difference between Discounted Cash Flow and Discounted Residual Income. I think the DCF model you talk about is actually a discounted residual income model. Yes, I agree with you on the discounted residual income model. Discounted residual income model is a bunch of BS created by accountants who thought that accounting numbers is good representation of actual economic condition.

Discouted Cash Flow model is basically a model which discount the expected CASH FLOW, that's hard solid cash. It takes into account of capital expenditure and working capital requirement. Underlying the model is a simple assumption that a business is worth only as much as the cash that it can generate, which is very sensible.

As we are buying equities without control, as Kinwing have mentioned previously, we can adjust that using a discounted dividend model (a subset of discounted cash flow) instead. But, that would probably undervalue the company. But, you can compensate that by doing a discounted cash flow to equity model. The fair value, should be within the range of two values.

There are complications involve when using DCF to value young and high growth firms because cash flow projection is difficult. But, even if you use multiples such as PE and you whack an industry PE into the high growth firm, you are implicitly assuming that the company, will become like the mature players in the industry in the long run and earn that sort of cash flow that a mature player earns. So, you are basically doing some forward looking assumption without realising it. If you use historical PE, you may be assuming that the past performance can repeat itself. It is just that when you use multiples, these assumption are implicit. In DCF, the assumption is explicit.

There are various shortcomings in DCF model assumption. The problem is not the model itself but the problem lies in the inconsistencies in applying the assumption. For example, the capital expenditure assumption must tie back with the growth assumption. You can't assume that the growth is 30% a year with minimal CAPEX. Another possible error is the terminal value growth rate, some analyst whack a very high growth rate into their terminal value. What's their justification? Cause the industry grow at 5% for 30 years, so, I think a 4% growth rate is possible for terminal value. But, terminal value is a valuation to an infinite years. What is 30 years when compared to infinite amount of years. By whacking a 4% growth, the company theoritically at one point of time will be bigger than the entire output of the world because the world do not grow so fast.

There are flaws in every valuation model, be it multiples or dcf, I think the best way is to use a combination of both. Check with various technique.

Plus, for those who like to read analyst report for TP, always find out how they come up with it. Even some big foreign banks TP have some real inconsistency problem which result in an above average target price.

Just my two cents.

This post has been edited by the snowball: Nov 7 2010, 12:07 PM

I think Dreamer's strategy is actually the same as your strategy of buying below intrinsic value, cause by definition, if we use the dividend yield as the measures of market return, a 2X-3X spread over FD means the market risk premium is high. This tend to happen during a crisis and prices tend to fall below intrinsic value. It is just a layman and easily understandable ways to say the same thing.

I use to think like you too. But, now, I prefer a certain amount of dividends. It is hard to find management that can deploy capital into something that have incremental return on capital. Even you take the most fantastic company the world have ever seen- Microsoft, they have some fantastic return on equity. Although they pay dividends and buy back a lot of shares, the cash flow generated from its Office, Windows and Server division is still amazingly large. So, the dividends is less than its cash flow. Thus, Microsoft keep some of the cash flow it generated. Since Microsoft have above average ROE, you may think that it is good for Microsoft to retain the earnings to do R&D. But, most of Microsoft R&D is on things that they lose a lot of money, like Bing and they are throwing a lot of money into Windows Phone 7. So, even though MSFT itself have fantastic return on equity, the retain portion is wasted on those that do not generate that high of return of equity. You may say that this is a good deployment of capital in tech industry, but, if XBOX is anything to go by, Microsoft need to throw 8-10 years of cashflow into something before it can generate some decent result. So, not sure whether this is good deployment of capital. The point is, even companies with high Return on Equity, the portion retain may not be reinvested into division that have that high ROE, so, the cash, in a sense is better return to shareholders.

As you mention that you manage to do 20+% compounded growth for 5 years now, the benchmark should be a company that can return 20+% ROE in its reinvested earnings just to match your performance. Very few companies manage to pull that off, yes, even Microsoft can't do that because the money is thrown into division with lousy returns. The historical ROE of US is about 12%, I guess, for Malaysia, it should be around there or lesser as our leverage should be a bit lower due to some Chinese businesses tend to have a more conservative capital structure. So, in your case, since your return is 20+%, I think, most if not all of the time, you should hope companies to pay you dividend because you can actually do better with the money.

So, if you have some good stock picking skill like foofoosasa, one should hope for dividend.

The retention ratio is not that important in this discussion as I am comparing between your own incremental ROE vs the companies incremental ROE. In addition, it is not the average ROE that matters, it is the incremental ROE that matters, that is what ROE the incremental retained earnings manage to generate.

I believe the company you talk about that churn out 30-80% ROE are tech, pharma and recently IPO-ed companies. I am not sure how you define ROE, but, if you define it as accounting ROE i.e. Net Income/ Accounting Net Book Value rather than finance ROE i.e. Net Income/ Market Cap then, the company are most likely in the industry I mention above. This is due to the fact that there are certain accounting rules that require them to expense R and D rather than keep it at the books, so, the ROE is actually not a good reflection of true economics of the business because a lot of asset are actually off the balance sheet.

Another possible explaination for such a high ROE is that it is a human resource based business. For example, it is a pure play investment bank (i.e. no trading) or consultancy firm. Again, such companies have a lot of off balance sheet assets that is not reflected in the balance sheet.

In any way, businesses above should not be judged on ROE because their profitability and growth is not a function of their capital allocation decision. It is more of a function of how they increase their brand awareness, customer loyalty and retaining their best talent. All this are off balance sheet assets. So, accounting ROE become unimportant in such businesses because their key success factors is not capital allocation. For example, take coca cola, a management should be judged on how much its retain its brand awareness and image or even increase it as such action would result in an increase in bottom line. He should not be judged on how good he is spending money to build the factory. The damage done on the coca cola brand image on its bottom line should be much more severe than a wrong capital allocation decision.

Another type of companies with such an impressive ROE is those with a recent IPO, but, those advantage tend to fade within 3 years.

I ran a stock screen using Capital IQ with a very basic requirement of non-tech and non-pharma, sustainable >30% ROE of 10 years, debt to equity less than 1.25, out of the universe of 500k companies worldwide, only 200 fits the bill.

In any case, I would like to hear the list of companies you have mention. If it is cheap, I may even buy it. Not necessary need to be in Malaysia. I have access to most markets as long as it is not companies that is listed in Timbaktu.

This post has been edited by the snowball: Nov 7 2010, 09:55 PM

It is ok to not disclose. Unless you finish buying the stock already, it is actually better to keep your stock pick secret.

The company is impressive in terms of financials. Just browse through, if add back off balance sheet liabilities, the leverage will be more significant but it is how retails works, which explain the ROE. But, be careful of the licensing deal they have. Keep a close eye on that. I know of a HK company got wiped because of that. Thanks for the sharing.

I use Capital IQ. But, mere mortals like us would not be able to afford it unless you have USD13k/year to throw around. It is actually for hedge funds and asset management guys. I got it because it is in my university database. Even in my school, I can't access it as it is for MBA student not undergrad. I somehow manage to convince the librarian to allow me to use it.

Capital IQ is fantastic. It is the best software out there. The figure is accurate and you can check back to the annual report because it provides an instant link. If you know who is Li Lu, the man in running to replace Buffett before he voluntarily pull out, he actually have some stake in Capital IQ. It is way way way better than Bloomberg because it is designed for FA rather than TA. If you like trading, perhaps Bloomberg is a better platform.