Warren Buffett claims P/E Ratio has
Nothing to do with Valuation
We all know that the P/E Ratio is one of
the most useful ways to find undervalued stocks. Right? Almost every book on
Value Investing seems to recommend using the P/E ratio. I have happily been
using the P/E ratio for many years.
But I must admit that I sometimes
struggled to understand exactly why some stocks with a P/E of 30 are bargains
while other stocks with a P/E of 10 might still be over-priced. I learned how to
make sure the "E" part of the ratio was based on adjusted or
normalized earnings. This removes distortions caused when earnings are
abnormally high or low due to unusual or one-time items. I did a lot of
calculations, thinking and reading which helped me to refine my methods of using
the P/E ratio.
I even published two articles on the
subject to my Web Site.
the P/E Ratio
that P/E ratio of 20 indicate a buy or is it a sell?
After all of this analysis and effort I
was satisfied that I had honed to a sharp edge my understanding of exactly what
level of P/E was attractive in different circumstances. Having read and studied
intensively many books on Value investing including three books about Warren
Buffett's methods I found that I was very comfortable with Buffett's methods and
I imagined that I was becoming quite a skilled disciple of his.
So...imagine my confusion at reading on
March 13, 2001 the following sentence which Warren Buffett wrote in his year
2000 annual report for his holding company Berkshire Hathaway. Warren Buffett
wrote "Common yardsticks such as dividend yield, the ratio of price to
earnings or to book value, and even growth rates have nothing to do with valuation except to the extent they provide clues
to the amount and timing of cash flows into and from the business."
My reaction was... Warren, why has thou
forsaken me? You say the price to earnings ratio and the price to book value
ratio have nothing to do with valuation? Are these not the very foundation of
value investing? Have you given up on value investing? in short, have you
gone senile, taken leave of your senses?
Having invested a lot of personal time
and energy into valuing stocks using the P/E ratio and to a lesser extent the
price to book value ratio, I did the sensible thing and put this offending
sentence out of mind.
But a few things were still bothering
me. When I analyse a stock, I estimate the intrinsic value of the stock and
compare it to the price. I also consider the P/E ratio but I was starting to
think that maybe the P/E was not needed since the intrinsic value was my main
gauge of value. It also bothered me that the P/E level at which a stock becomes
a bargain can vary over a huge range depending on circumstances.
I did more calculations to determine
exactly what level of P/E was justifiable under different assumptions about
earnings growth, the dividend pay-out ratio and the required rate of return. I
was very pleased to create a table to help investors judge when the P/E
represented a bargain in different circumstances. This table is available at my
Web Site How to Pick
Stocks Using the P/E ratio.
But I was still bothered. My table of
maximum justifiable P/E ratios was created by calculating the present (or
intrinsic) value of cashflows (per dollar of initial earnings) that could be
expected from stocks with different growth, dividend pay-out ratio and required
rate of return assumptions.
Finally it dawned on me, all I really
needed to look at was simply the ratio of the stock's price to its calculated
intrinsic value. You simply can't properly determine the maximum appropriate P/E
ratio without knowing the growth, dividend pay-out ratio and required rate of
return. There are no short cuts you have to know or estimate all of those. But
if you know all of those you can simply calculate the intrinsic value of the
share and compare it to the price. Let's call this the ("P/V") or
Price to Value ratio.
Warren Buffett was right after all
(surprise!) the P/E is of no real use without additional information. He has
recognized that the P/E ratio and book value are simply too crude to use
directly as value indicators, particularly when he is able to calculate an
actual intrinsic value for a share.
Using the P/E ratio is like trying to
estimate the weight of a person by looking at their shadow. It only works if you
start factoring in the angle of the sun and you need to estimate how fat the
person is, which may not be apparent from the shadow. It's really not too
accurate. In order to judge the P/E you need to know all these other factors. I
think that in effect Warren Buffet would simply want to look at more indicators,
if available, rather than trying to judge someone's weight by their shadow.
At this point I realized that my
calculation of the intrinsic value was all I needed to actually calculate my
estimate of the P/V (Price to Value) ratio of the stock. There was no need or
value for me to look at the P/E ratio as such. (Why look at shadows to judge
weight if there is a scale available?)
Itís useful to look at the various
value ratios as a progression as summarized below.
Uncorrected P/E ratio - Calculated by dividing the price by the actual
trailing net income per share. The advantage of this ratio is that it is easily
and mechanically calculated. This ratio is provided in portfolio tracking
programs such as on YAHOO. The severe
disadvantage of this method is that the earnings are not normalized or adjusted
for unusual items. There can be no assurance that the earnings are in any way
representative of future sustainable earnings. There can be no rule of thumb for
this P/E. Using this uncorrected P/E to judge which stocks are bargains is
likely to lead to some severe errors.
P/E Ratio based on normalized earnings - Most often calculated by using
future earnings estimates provided by analysts. Sometimes calculated by
adjusting actual net income for unusual items or after ascertaining that the
actual earnings are sustainable. More useful than the uncorrected version.
Useful for gross indications, for example anything over 30 is more likely to be
over-priced. Ratios under 10 are possible bargains. But there is quite a spread
and its hard to know if a P/E of 20 is a bargain or not without also looking at
the growth and the dividend pay-out ratio.
PEG Ratio. - Divides the P/E based on normalized earnings by the
estimated growth rate. This is a big improvement as it attempts to normalize for
the growth rate. Some rough rules of thumb for appropriate PEG ratios are
possible. The disadvantage is that it is not a mechanical calculation, it
ignores the impact of the dividend pay-out ratio and it does not deal with the
very common situation where the growth rate is expected to change, such as a
forecast of a few years at a very high growth followed by a lower more
sustainable growth rate in the long term.
P / V Price to Value Ratio - This gets much more directly to the heart of
the matter. It directly indicates if the price is higher than or lower than the
calculated value. Recognizes that if one knows the normalized earnings level,
the growth rate and the required rate of return then the intrinsic value can be
directly calculated and compared directly to the price, rather than trying to
work through the shadows with the P/E ratio.
Strangely, one does not often see
references to calculating the P/V ratio. Many books do talk about comparing the
intrinsic value to the price but it seems more clear to call this the P/V ratio.
In summary if you can determine the
initial adjusted (and sustainable) earnings per share, the growth rate (which
may be in several stages) and the dividend pay-out ratio. Or if you can
otherwise forecast the cashflows, then you can directly calculate the intrinsic
value and P/V ratio using available tools. This is simply much more direct
(although more work) than looking at the P/E ratio and then trying to figure out
what P/E level represents a bargain. It seems that Warren Buffett already knew
this and perhaps the rest of us need to catch up and catch on.
September 14, 2001