Jim and I regularly cruise for decent net net opportunities
and I was somewhat surprised to see such an extensive list
of not only net nets but profitable ones. Recognizing many
in the list immediately, I concluded that there are two
major problems here:
1) the data for the screen is faulty
2) the list confuses the difference between a net net and a net net
buy, i.e. there has to be a margin of safety in the discount to
net net value ( Graham wanted a 33% discount at least).
I went through the first bunch of companies you listed - by
going through the SEC filings- and came up with the following
ratios (market cap/net net).
ACMM 1.0
ASFN 0.97
VOX 0.90
CSTL 1.66
CHRB 1.70
CRAN 0.92
EDIN 0.94
ESP 1.10
HBW 0.95
IDG 2.07
I stopped at this point, though I may go through the rest
of the list when I have more time. It is obvious that these
are not at all "profitable net net buys" a la HYDEA, JBM,
or PSO. These three had ratios < 0.70 at the time they
were discussed, thereby providing a decent expected return
of near 50%. The best ratio in the bunch - 0.9 - only provides an
expected return of 11% - not a significant margin of safety IMO.
Moreover, VOX and ESP among several others seem to be
actively collapsing, with sales shortfalls greater than
30% and recent ventures into negative income.
I have commented in the past that I can take a list of
15 net nets down to 1 or none in about 30 minutes with
SEC Edgar. This is not because I am seeking the best ones,
but because I am simply seeking true net nets at 2/3 of
net net value, as Graham advocated. To do this, I don't
even look at the income statement or the business they
are in. If I find one, then I will look at the other aspects,
but I rarely take this next step because 2/3 net nets a la Graham
just are not out there in any great quantity.
Screens have inherent limitations - even the best ones.
Historical testing of the net net hypothesis should recognize
the problems in the data and the potential pitfalls in not
recognizing the 2/3 criteria for a buy.
Good Investing,
Mike |
