What About Market Timing to Reduce Investment Risk?
Sometimes it's just not productive to be in the market, so a common
request is to find a way to time fund purchases as a way to either
improve your overall returns, or at least to reduce the risk and
volatility of your returns. The problem doing this with mutual
funds is twofold:
1) It's very hard to actually time the market well,
2) The fund companies have made it more difficult. They have
introduced Early Redemption Fees (ERF's) that make it difficult to
get in and out of the market very often without either destroying
your returns with their extra fees every time you switch, or they
may just ban you for overtrading in their funds.
With that in mind we developed a simple calendar based system that
improves the odds of getting better returns, and historically
reduces volatility.
Seasonal Stock Market Trends
The most popular of the calendar based systems is explained by the
adage "Sell in May and go away." Simply put, you stay out of the
market for the 6 months beginning with May, and buy again in the
beginning of October. And actually those 6 months do underperform.
But exploring that a little deeper, you find that the 2 worst
months of the year since 1950 have been August and September. Armed
with that we looked at the performance of the Fidelity Equity
system from January 1995 to January 2006. Specifically we looked at
the performance during each of our quarterly holding periods.
The 3rd quarter (July thru September) underperformed on our system
as well. The compounded return in that quarter was -1.1%, by far
the worst. In addition, out of 8 negative quarterly returns, 4 of
them were during the 3rd quarter, including the 3 worst quarters we
had in the backtest (1998, 2001, 2002).
By going to cash in the 3rd quarter holding period, the overall
return of the Fidelity Equity system improved slightly (just over
2%). However, the maximum drawdown improved from about 27% to 13%.
While these are backtested results, it shows a significant
improvement in the volatility of the system returns.
There are other forms of seasonality or calendar based trading as
well. Over a longer cycle, there is the Presidential cycle. This
tracks which year of the Presidential election cycle we are in,
with the proposition that the year before a Presidential election
and the election year itself outperform the other 2 years.
Over the short term, there is the month end and holiday seasonal
bias. This was first popularized by Norm Fosback in his book Stock
Market Logic in the 1970's. He noted that the last 3-4 days of
each month along with the first 1-2 days outperform then rest of
the month for most months. Hulbert's Financial digest still tracks
this, and it continues to perform well after all these years. This
bias is why we execute the trades for both the Fidelity and ETF
systems after the 1st week of the month.
Next we will cover Market Timing options with ETF's.