This page is quite old hence its rather spartan appearance.
Why not check out our Latest Stories page for our newest articles or search our site for anything.
VALUE INVESTING
|
|
By
I shifted my attention, only temporarily mind, over to the Qualiport discussion board in recent days, where a debate has been raging for some time now over the methods used to select shares for their approach. If I recall correctly, the opening shots of the war were not fired by me, but by one of my colleagues who challenged the concepts – primarily that of trying to forecast earnings per share (EPS) growth many years into the future, in order to discount this to a net present value, or NPV. The purpose of this technique, which some of them call "value" (something for which I will never forgive them) is to see whether the share's NPV is higher than the current price. If it is, then the share may appear undervalued on this basis. Their idea of "value" though, I must stress, bears no relationship at all to what I mean by the term, which is the kind of value shares we discuss on the value board. Just to make the point, value to me means cheap on current, actual fundamentals. To Qualiport followers, it means cheap on NPV grounds. The difference is enormous. My view is based on actual figures, theirs on long term forecasts, with all the risks involved in taking such a view. To be fair, they do not use this idea alone, but in conjunction with several other features. Stuff like the quality of management, sustainable economic advantage, brands and so on. The criteria of the Qualiport are published in full here. Generally speaking, it is the Warren Buffett approach. Or at least his approach of recent years. I understand that in his early days he was a value player, in the proper sense of the word. Anyway, the main idea is to find long term buy and hold shares that will outperform the market. Long-term buy and hold, or "LTB&H", is an admirable strategy, and one which I fully support for a great deal of investors. But what I and my colleague were criticising were the methods QP'ers use to locate likely shares for the strategy. I opined that those methods, far from creating market beating performance, were likely to produce underperformance, and that such an investor would be better off in a much less carefully chosen selection of large capitalisation companies. Perhaps top FTSE 100 shares or sector cap leaders, some simple approach to blue chip investing like that. I won't repeat all my arguments here through lack of space. I pulled out in the end because these sorts of arguments rarely change peoples' minds, so ultimately they are often fruitless and eventually become repetitive and boring. At least I don't think I changed anybody's mind there. Can't know for sure. But what happens is that the entrenched views are aired and everyone goes away at the end of it convinced that they were right in the first place, just as entrenched as they were before – including me, of course – and nothing changes. But it got me thinking about whether the value style (and I use the word from now on only in my sense) is actually something that could be used in a LTB&H portfolio. Such investing requires that you choose your shares at the start, then more or less ignore them for a very long time, reinvesting dividends if you don't need the income. I'm not sure it would work, for a variety of reasons. Assuming only (say) FTSE 350 companies are desired, one problem is that my shares are rarely that large. So the criteria would have to be weakened considerably to admit these. Secondly I very rarely find suitable selections anyway, whatever the capitalisation, so again the filters would need to be widened in order to choose a portfolio of, say, five to ten shares at one go. All right, I can live with that because the target time has changed. We are no longer looking at my usual two year or so view, but ten years perhaps, maybe more. Which of the pyad filters would I have to dilute or scrap? Well, price to book value (P/BV) would have to be watered down very substantially or maybe dropped altogether in order to admit such shares. Debt, the lack of it and the presence of net cash, would have to severely weakened too, I suspect. Might be able to still get away with price to earnings (P/E) ratio maximum of one third below the market, and with a yield 50% above it, though. So to summarise, I would trawl on the P/E and yield criteria, looking for qualifying shares in the FTSE 350 with minimum debt/maximum cash and lowest P/BV. One can do this with CD-REFS, for example, playing around with the filters until enough shares qualify for the desired portfolio size. I would have a problem, though, with earnings per share (EPS). In the normal shortish-term approach, I require a decent rise over the next year in EPS as forecast by analysts. I know the risks of this, but it is a necessary evil to achieve success. But if we are now looking at a long-term value portfolio of ten years or more, one year's EPS rise means little. So what do I do? Take the Qualiport view and try and forecast EPS for the next 20 years? No way. So here's what I would do. I would utterly ignore EPS forecasts. Does this sound odd? Here's why. Since I know that neither I nor anyone else can make a long-term forecast that has the slightest degree of any semblance of accuracy at all, it appears to me that one is better off ignoring it than being lured falsely into a share because a long-term forecast appears attractive. As an odds player, I would put it like this. One has a better chance of selecting a better performing share at random than of selecting one by trying to use long-term EPS forecasts. As a rough test then, I took May REFS and set the following parameters: P/E max 15 The diluted pyad long term value criteria. These are the trawl results, in descending size order: Anglo American (LSE: AAL) Six shares qualify. I had to play around with the filters for a bit in order to set them such that I had a portfolio of between five and ten shares. Too tight and not enough were thrown up. Too loose and I had too many. It was the P/BV and gearing levels which required repeated tweaking to produce a portfolio of this size. And I stress that no regard whatsoever has been paid to future EPS. Also, and very importantly, I have not applied any sense of smell to these shares. Thus the choice is more or less mechanical really, although it would be dynamically mechanical because the filters would have to varied every you time you ran it, given that P/E and yield are a function of the market averages. A few comments on some of these shares may be of interest. Barratt, a housebuilder, was a great value play for me a while back at the beginning of the current house price boom when it was a fully qualifying pyad share. Unigate has been discussed quite a bit on the value board, I know that one or two of our regulars fancy it. Anglo is a major mining house, and I've always had a bit of soft spot for serious mining companies like this and Rio Tinto. Halma is a favourite share of a colleague at TMF, who never, ever, agrees with anything I say, so Jim, there you go, a little bit of common ground for a change. Smith and B. Energy appear to be poor quality shares, but that's the point, it is a value portfolio, so these two perhaps add a bit of bottom-fishing. Note that I am not suggesting that anyone invest in these shares. It was an experiment. I personally would not buy them on the criteria I have used because I am not a long-term player. But this is the kind of approach I might suggest as one way of looking for a long-term value-based portfolio, as opposed to my other LTB&H ideas which I find attractive as well, as mentioned above on FTSE 100 shares. Equally, you could easily do something similar with value criteria on the Footsie itself, if you wanted to restrict the selections to that field rather than the 350. They would have to be loosened further, though, to produce enough shares. Would this portfolio beat the Qualiport long term? Well, it would take the long term to find out of course, so there's no quick answer. My money would be on this value portfolio, or a Footsie-derived portfolio as I describe above, without any doubt in my mind. The value underanalyser portfolio v. the QP overanalyser one.
Yield min 4.0%
P/BV max 3
Gearing max 15%
Member of FTSE 350
British Energy (LSE: BGY)
W H Smith (LSE: SMWH)
Unigate (LSE: UNIG)
Barratt Developments (LSE: BDEV)
Halma (LSE: HLMA)Related Links