Quant View: A Look at a Winning Formula
By Ted Murphy
Special to TheStreet.com
11/28/97 12:57 PM ET
What if I told you there is a strategy that always works,
returning relative outperformance of +45.3% a year during a
10-year backtested period? Well, there is.
Not to keep you in suspense, it's the strategy of buying
stocks with low price-to-earnings ratios based on next year's
actual reported earnings. I know. That's a tough trick to pull
off.
In a study I did 10 years ago, using historical data from 1976
to 1987, I found that buying the lowest P/E decile of stocks
using next year's actual reported earnings returned an
average of 44.3% a year, versus an average of -1.0% a year
for purchasing stocks in the highest, most expensive P/E
decile.
For me, this backtest explains the rules of the game played
by many institutional investors and analysts. If you are
successful in projecting the next one or two years in actual
earnings, you can consistently and significantly outperform
the market and your peers by buying stocks that are cheap
on your projected earnings.
Surprisingly Cheap Stocks Were the Best
The study also showed that the real payoff came when the
company surprised investors, posting stronger than
expected earnings.
Buying low P/E stocks using the current consensus wasn't
nearly as successful as perfect foresight. The spread
between purchasing the top versus bottom P/E deciles
based on the consensus estimate of the next year's
earnings was only 5.3% in this study, versus 45.3% for the
spread based on actual earnings.
Low P/E Using Perfect Foresight vs.
Consensus Estimates
Annualized returns, quarterly observations from 1976 to
1987
P/E Using:
Low P/E
Decile
High P/E
Decile
Spread
Actual EPS (+1Yr)
44.3
-1.0
+45.3
Estimated EPS (+1Yr)
24.1
18.8
+5.3
No Need for Home Runs
The study continued by looking closely at the group of
unexpectedly cheap stocks, the companies that ended up
cheap because earnings grew rapidly and above consensus.
This unexpectedly cheap group made up 39% of the
companies in the lowest price-to-earnings decile.
The return of these surprisingly cheap stocks was 62.7%,
even better than the 44.3% return for the entire decile of
stocks that were cheap on reported earnings.
So how do we find these unexpectedly cheap stocks?
It is important to note that only 20% of this key
outperforming group were home-run stocks that started from
an extremely low earnings base and then exploded upwards.
The vast majority of this special group of stocks -- 80% --
were starting from P/E multiples already below the market
median, in P/E deciles 2 through 5.
My conclusion at that time, with which I still feel
comfortable, was that the most attractive group of stocks
were those selling at a below-market multiple where
earnings came through above expectations, turning a
moderately priced stock into a winner.
Ted Murphy (ted@pdgm.com) operates the MarketPlayer
Web site. Prior to MarketPlayer, he was a partner at
Equinox Capital Management.
c 1997 TheStreet.com, All Rights Reserved. |
