One of the bulls’ major reasons for being optimistic on
the stock market is their view that stocks are reasonably valued at 14 to 15 times
earnings, well within past norms. They consistently
state this view on financial TV and in print without ever being challenged by
their interviewers. The far smaller number
of bears, on the other hand, contends that the market is substantially overvalued. We believe that the bulls are using a flawed
model that would not have had predictive value in the past, and that the bears
will prove to be correct.
Simply put, the bulls use the current price of the
S&P 500 and divide it by estimated
forward-looking operating earnings to arrive at the current price-to-earnings
(P/E) multiple. Therefore, based on
today’s S&P close of 1698 and consensus estimated 2014 operating earnings
of $122, they come up with a reasonable P/E of 13.9 times. The bears use the same price, but divide it
by trailing reported (GAAP) earnings of $90.96, resulting in a P/E of 18.7, at
the high end of the range of historical valuations. In addition, when reported earnings are cyclically
smoothed to dampen distortions, the P/E multiple is 19.7.
We have three major problems with the way that the
majority determines the value of the market.
First, operating earnings differ significantly from earnings calculated
in accordance with “generally accepted accounting principles”, commonly
referred to as “GAAP” or reported earnings. Operating earnings start with
reported earnings, and then add back a number of expenses considered
non-recurring, such as severance pay, plant closings, inventory write-downs,
opening and closing of facilities and any other number of expenses that
corporate managements may choose to write down.
In the past 15 years or so, companies have gotten a lot more creative
about what items they can write off, and now a large number of expenses that used
to be considered a normal cost of doing business are called “unusual”, even
when these write-offs are taken year after year. In other words, in too many cases what is
called operating earnings is pure fiction, and not calculated in accordance
with generally accepted accounting principles.
Second, the long-term average P/E ratio of 15 is based on
trailing reported earnings, not on operating earnings. Prior to the last 14 years of sequential
bubbles, the 71 year average P/E on this basis was 14.5 (rounded to 15). Operating earnings, as they exist today, did
not even exist until after the mid-1980s, when they came into vogue partly as a
means of making earnings look better than they would have been under the accepted rules. Since operating earnings always exceed
reported earnings, often by significant amounts, the P/E on operating earnings
has averaged three multiples below the P/E on reported earnings. Therefore, it is likely that if operating
earnings had a long history, the average P/E would have been only 12, rather
than the 15 on reported earnings. In
that case, even on 2014 operating earnings of $122, the market would still be
Third, but not least, estimates of year-ahead operating
earnings are notoriously unreliable. In
the last 28 years, estimates were too high 76% of the time, often by amounts
exceeding 20%. In May 2008 the estimate
for the year was $89, and eventually came in at $50. At the same time, the 2009 estimate was as
high as $110. The final number was $57.
In sum, the use of forward operating earnings to
determine the value of the market can be extremely hazardous. In our view, the market is selling at 19.7
times cyclically-smoothed reported earnings, about 31% higher than the
historical average of 15, let alone the average multiple of 7-to-10 times seen
at the bottom of past bear markets. At
present levels the market is already discounting a highly optimistic outlook
that leaves it increasingly vulnerable to the serious U.S. and global economic
and political risks that can come to the fore at any time.