To average or not to average?
July 5, 2011 by Jack Marrion
I have often heard agents suggest that averaging should be used to measure index movement if one thinks the market will be volatile. Such an approach would be used instead of measuring the index gain or loss from one start point to one end point—the annual point-to-point method.
The belief is that averaging fares better when the market is moving around a
lot. The reality is, no, averaging does not usually result in a higher index
gain, except in the early stages of a bear market.
In both the 2000 and 2008 bear markets, averaging the daily or monthly values
extended the positive gains for a couple of months past the period when the APP
method turned negative. However, when the bull market returns, averaging delays
a return to positive territory.
A look back
If you look at 12-month periods over the past
decade and use an annual reset approach, the mean annual S&P 500 index
return was 8.14 percent. By contrast, a monthly averaging approach produced a
mean return of 4.56 percent, or 56 percent of what the APP calculation
generated. Does that mean averaging methods produced half the returns of APP
methods?
No, and for a couple of reasons. The first is that most approaches,
especially APP methods, use caps that limit the gain, but averaging methods
usually have higher caps. If you apply a 10 percent APP cap and a 12 percent
averaging cap, averaging produces 76 percent of what the APP method does.
The second reason is more annuities
offering averaging methods also offer uncapped methods where participation is a
straight percentage of the index gain, or the uncapped gain less a yield spread,
which tends to raise the index gain even more.
If you look at other historic periods when the market was volatile you get
similar results. Averaging is not a better choice for choppy markets. Yet this
does not mean averaging is a bad choice. It simply means that any method that
needs index gains to produce interest will suffer in down markets.
Could look good
From spring 2010 to spring 2011, the
unaveraged index showed double digit returns, but averaging the values limited
annual gains to around 1 percent. Why the huge disparity? The reason is the
markets began to dip in April 2010 and didn’t recover until October. Indeed, the
stock market gain from fall 2010 to spring 2011 was just about the same as the
gain from spring to spring.
However, this negative effect ended. In fact, if the indices do dip this
summer, yearly averaging could look very good. As an example, if the indexes
dropped 5 percent each month from spring until autumn, an annual point-to-point
look would show a loss in October, but averaging results in strong
returns.