<THE FOOLISH FOUR>
Foolish Four Report
by Robert Sheard
LEXINGTON, KY. (Mar. 12, 1998) -- With all of the recent discussions about which Dow Dividend Approach is the best, our Foolish colleagues an ocean away (The Motley Fool in the UK) are also taking an interest in the topic, as they're developing similar approaches for the Footsie (FTSE).
So today I want to move over for a different voice to be heard in the discussion. David Berger spearheads the efforts to develop a Beating the Footsie area, and the following is a column David wrote in the not-too-distant past. -- Robert Sheard
I was emailed a couple of weeks ago by an expatriate Briton living in the USA. Chris Larmour is his moniker. In his electronic missive he mysteriously alluded to a refinement he had developed of the Beating the Dow investment strategy.
"Enlighten this old Fool, old man!" came my response.
And he did. And it was so impressive that I have summarily named his refinement, "The Larmour Value Variant," or LVV for short. (With his agreement, of course.)
The LVV involves an attempt to quantify the degree of cheapness of a high yield share. You'll remember that the strategy hinges upon buying out of favour large cap shares, which have been identified by their high dividend yield. In other words, they are cheap in relation to their dividend.
'Twas Michael O'Higgins who pointed out that the cheapest of the ten highest yielders in the Dow tended to underperform, while the second cheapest -- his Penultimate Profit Prospect, or PPP -- was a stupendous, if volatile, performer over the years. The Unemotional Value variants (the official Foolish Four now uses the UV4 variant) capitalise on this by eliminating the cheapest share if it is also the highest yielder, on the basis that if it is so suspiciously cheap it is likely to be a share in genuine trouble. This does seem to work in terms of boosting returns, but it is an empirical technique. To quote Robert Sheard, who wrote the blurb for the Unemotional Value section:
"While not perfect, this signal [i.e. the cheapest share if it's also the highest yielder] has been consistent enough that using it to decide when to exclude the #1 stock and when to keep it has generated much better returns than either of the other two methods out of which it grew."
What Chris has done by means of his now eponymously named variant is to refine the UV variant by quantifying when to ditch a share, using the Price to Yield ratio as a guide.
What the Price to Yield ratio tells us about a share is how much each percentage point of yield is costing us. If each percentage point of yield is too cheap, then chances are there might be something truly amiss with the company. There are, after all, no truly free lunches. See below:
Price Yield P/Y 50 5% 10 40 10% 4 200 10% 20
The second company down really does seem to be coming cheaply, too cheaply, probably. On the Dow, Chris has worked out the following rough guide to the cheapness or otherwise of the stocks:
Under 10 -- "No-no"
10 - 20 -- Nice range
25+ -- Too richly valued or stable income producing to bother with.
Obviously, as the valuation of the stocks in the universe from which they are picked changes, it is possible that these levels will change. Anyway, Chris is doing some backtesting in which he rejects the cheapest highest yielder(s) if they have a PY ratio below 10. So far, the most impressive result is that in 1987 the Unemotional Value 4 variant when combined with a PY ratio screen would have returned more than 30%, more than any other multi-stock mechanical method for that year.
Of course, the crux for the LVV will come with backtested data and ultimately with prospectively tested data if the historical returns do indeed look good enough to follow the strategy in the future. Chris is currently backtesting his variant on the Dow and when we get a database of historical FTSE share price and yield data it will be fascinating to see how and if it works here.
It will be vital to compute average levels for the PY ratio in the FT30 or FTSE100 as a whole. If these do not tend to change over time to any great extent, then it may be reasonable to set some kind of "absolute" levels for the PY ratio cut-offs. Otherwise it may be more appropriate to set the cut-offs as a percentage of the index's average PY ratio at that time. The same will be true for the Dow.
There is no mystique to the LVV. It is simply a way of quantifying the relative cheapness of a share and a refinement of the method of throwing out the cheapest share if it is also the highest yielder as likely to be an underperformer. My suspicion is that the PY ratio may just provide the solution to the fact that the FT30 can contain some extremely poorly performing companies for alarmingly long periods of time (BICC is a case in point), before they are thrown out of the index. With the PY ratio as a screen, these mangy dogs would simply not make it into the final cut, thus hopefully boosting returns.
Philosophically yours,
David Berger (TMF FoolUK)
[Want to be the first Fool on your block to get a copy of Robert Sheard's forthcoming book? Click here to pre-order your copy of The Unemotional Investor.]
TODAY'S
NUMBERS
Stock Change Last -------------------- UK - 13/16 46.94 IP - 3/8 51.50 MO -1 1/16 43.50 EK - 3/16 62.81 |
Day Month Year
FOOL-4 -1.25% 1.89% 7.13%
DJIA -0.19% 1.33% 9.50%
S&P 500 +0.14% 1.96% 10.26%
NASDAQ +0.41% -0.36% 12.33%
Rec'd # Security In At Now Change
12/31/97 289 Int'l Pape 43.13 51.50 19.42%
12/31/97 291 Union Carb 42.94 46.94 9.32%
12/31/97 206 Eastman Ko 60.56 62.81 3.72%
12/31/97 276 Philip Mor 45.25 43.50 -3.87%
Rec'd # Security In At Value Change
12/31/97 289 Int'l Pape 12463.13 14883.50 $2420.38
12/31/97 291 Union Carb 12494.81 13658.81 $1164.00
12/31/97 206 Eastman Ko 12475.88 12939.38 $463.50
12/31/97 276 Philip Mor 12489.00 12006.00 -$483.00
CASH $77.19
TOTAL $53564.88
|