Long response to Long post
> Finally, let me leave you with something Bill Miller said in one of
> those "esoteric" interviews published on the WWW:
> Q:
> So you wouldn't put much store in these historical studies, like
> O'Shaughnessy's or Dreman's, that show the success of low-P/E
> strategies in the past?
> A:
> "No, I do. Since the future hasn't happened yet, the past is the only
> thing you have to work with. But I think that when you look at those
> things you really have to be careful and put them in context. Let's
> see the context that gave rise to them. Let's see why it might have
> worked. Let's see if it makes sense with what we know about
> current finance theory. Let's get into it a little bit as opposed to
> saying "Gee, I have these correlations and they sure do look
> robust. Let's go invest this." It's like Buffett said, if all there was to
> investing was looking up the data and figuring out the data, the
> richest people in the world would be librarians."
Twister, I would suggest you read Haugen's "New Finance" for
explanations of why these low PB (PE, PS, etc.) stocks outperform,
as well as Dreman's new version of Contrarian Investor. I am
sure I'll be forgetting something, but here are two reaons that I
remember:
1. Most individual investors prefer not to own "dogs," hence
these stocks are undervalued for psychological reasons.
Despite my value leanings, I have experienced this personally.
When I was gifted some KO, I got their annual report and felt
a true sense of pride of being on the winning team. That didn't
stop me from recently selling the KO a few months ago, but it
did actually make me think twice. Moreover, what makes
better party conversation, I bought some AMAZON and it
is going through the roof or I bought a stock at half its
book value and in 2 years I hope to make some money
on it.
Mutual fund managers (i) do not want to have to keep justify
owning these dogs to their shareholders and (ii) are worried
about underperforming their benchmarks from quarter-to-quarter,
while this style of value investing requires a long time horizon
and can underperform for quite some time.
2. The performance of most companies tends to revert to
the mean within a few years, so if you buy cheap companies,
within in a few years you'll have better then even odds of
making decent money as valuations reflect improved company
performance.
As for a Buffet-like "value" approach of buying companies based
on their intrinsic value going out to eternity, I for one have no
ability to do such a forecast and I don't think most analysts can
either (In Contrarian Investing, Dreman has some stats on how
analysts are inaccurate in predicting earnings, and I can't imagine
most analysts forecasting with any reliability the earning for a
company 5 or 10 years out). If you have any pointers, this
would be a good way to diversify my strategy.
|
