Regarding really cheap stocks.... (I love big questions.)
First of all, a cheap no-growth company means P/Book below
1.0, P/E in single digits, 5% dividend or better, etc. There have
been a few moments when GM would have filled that requirement
recently...
But cheap growth companies are harder to define. You have
to make a guess of the prospective (i.e. future) growth rate.
First of all, what is the recent growth rate? In my opinion, growth
rates should never be calculated in terms of changes in earnings.
The reason is that earnings are too easily manipulated. Earnings
are bounded by sales and sales are a lot more stable, so use
them. More particularly, use sales per share. If you use straight
sales to calculate growth rates, you will artificially inflate the
growth rate when the company makes an acquisition that
dilutes the stock. This is frequently the case with non-dividend
paying high-tech stocks. The Value Line charts give historical
Sales/share figures. Tech stock investors are currently making
this error all over the place. For an example, compare the growth
rates in sales per share of GM and NSM over the last 17 years
(available in Value Line).
Second, determine what the limits to future growth are. One
limit is the niche that the sales are going to is being filled, or
worse, being eliminated. For instance, buggy whip makers
probably were showing reasonable growth rates in 1900, but
their niche had its days numbered. Here is a niche calculation
I posted for Altera/Xilinx:
Message 2549390
Somewhere on SI I posted a calculation that allowed you
to estimate total niche size by the rate of slowdown in sales
rowth (i.e. sales deceleration), but I can't find it.
Third, check to see if that growth rate is horribly risky. This is
the case with "fad" businesses. A recent example were the
micro breweries. Or CLCDF. There has to be some sort of
barrier to entry to keep the competition out. Like patents.
Fourth, take a good look at the earnings as a percent of sales.
(By the way, I only deal with companies that have positive
earnings, and in addition, every few years I get reminded to
leave the financial stocks alone. If earnings have been
anomalously low or high, ignore some part of that, and look
at historical margins. Use the historical margins to compute
what current earnings should be from current sales (per share).
Finally, you have a (sustainable) growth rate, and the amount
of earnings per share in the trailing 12 month (ttm) period. A
cheap stock is one that is cheaper than the growth rate in
percent, multiplied by the ttm earnings per share (with
historical margins). From there, you can either buy and hold
(and be at the mercy of the market, or you can try and steal
from the tiger by catching a bounce. I don't like buy and
hold, cause it's boring, so I like to try and steal some cash
from the market.
When stealing cash from the market, you want an obscure
stock (most investors don't know what it does) that has just
crashed by 50% in one day. That will bring it up on the radar
screens of all the other dead cat resuscitators. Analyze such
stocks for safe and cheap, and if it passes, buy immediately
with both hands. Don't worry about being too early, with a
safe cheap stock it doesn't matter, it won't go down much
more. (Or it wasn't safe. :) Instead, worry about being too
late.
In order to find such cheap stocks, look for the stocks that
manage to tank badly enough to make a new 52-week low
at the same time as they make the "dog" hit parade for single
day drops. You have to explore for these daily. The more
obscure the better. The reason is that big price moves attract
attention, and since 98% of all investment is long rather than
short, you can faithfully rely on the fact that an increase in
awareness is bullish for a stock, even if it's bad news.
Some of my recent favorites:
AAM a bank stock, (Don't do this at home). Note the big gap
down in late April:
tscn.com
CLFY, a software company. Again note the big gap down in
mid April (it was a good month for bottom feeders).
tscn.com
ASPX did the same thing last April:
tscn.com
CUBE caught it in May:
tscn.com
Rules:
(1) Don't chase stocks up, instead wait for fresh road kill, there
will be more within a year. I promise.
(2) Don't chase stocks down, it could be a trap.
(3) Don't keep a stock more than two months. This is for two
reasons. First, you need your capital free for the next morsel.
Second, your analysis could change when new information
hits the market. The whole idea here is to extract money from
a market swing, not bet on the long-term future of a company
or bet on the nature of the next piece of news. (Sort of like
dating. Look at all your dates as future spouses, have your
fun with the good ones, but don't marry them, they can/will
change :)
I don't have a good rule for selling, it always seems like I leave
the party too early...
Look for a stock that has dropped, say, 75% or more from its
52-week highs. If it shows as a good cheap growth stock, as
outlined above, it is reasonably safe, (in my opinion) if you buy
it within a few weeks of its big crash, (and the bigger the better)
I should mention that investing this way requires a willingness
to read the financial reports filed with the SEC and is a lot of
work. In addition, this all sounds somewhat better than it works.
If it worked really well, I'd be retired already...
Comments?
-- Carl |
