I think readers should know that you chose one of the worst, if not
thee worst, ten-year periods for the S&P 500.
Your statement misses the point. Although the majority of people may
not retire into one of the worst financial periods in history, you can
never know that beforehand. It's no different than life insurance.
Chances are good that most 30 year olds will never make use of the
term insurance they own, but you never know when you'll be one of the
unlucky few that do. The chance of loss is great enough that
individuals aren't willing to play the odds. If they were, the life
insurance AND annuity industries would be practically non-existent.
There's "cherrypicking" that's done to mislead and there's
"cherrypicking" that's done to precisely further a point. Sometimes
examining the worst case scenario is exactly appropriate.
By the by... historically, the stock market is down roughly 1 out of
every 3 years. So while the last decade may have been particularly
brutal, roughly 1/3rd of Americans will likely retire into some sort
of down market and would have done so with nearly any time period
chosen.
And just for fun, I did sit down last night and take the time to look
back over a longer period of time (I expected someone to nit-pick the
data). I started with the 70's and looked at every rolling 10 year
period up until 2009. In other words, 1970 - 1979... then 1971 -
1980.. then 1972 - 1982, etc.... The annuity came out significantly
better in almost every single rolling period. The reason is that
investors fail to realize that the stock market rarely performs close
to its historical average. If it did, the annuity wouldn't be such an
enticing offer. But instead, the market is more commonly up 30% one
year and down 20% the next. Due to those wide variations, the
"guaranteed 5% years" give the annuity a boost that the general market
can't keep up with. Errantly, investors too often apply the logic that
"the market averages X% annually, so I'm probably not going to use
that 5% guarantee too often. And I'm not paying 2% for something I
don't use". The truth is that since 1970, an annuity investor would
have employed the 5% guaranteed growth a total 15 years out of the 40
(over 35% of the time).
It's also worth noting that I used the published returns for the S&P
500 and the Barclays' Agg bond index. In reality, those are indexes
and cannot be invested in directly. Rather, one would have had to buy
VFINX, AGG, or some other proxy. Although the expense ratios and
trading costs are (presumably) small, they would have further eroded
the market returns making the annuity shine that much brighter.
Monte Carlo simulation offers a more appropriate comparison in my
opinion.
I encourage you to run one. I like them too! But I don't have the time
or expertise to run one given the unique "either/or" earnings
structure of the annuity. In order to do a comparison, you would have
to run a basic monte carlo of the market and then re-evaluate that
exact same "run" using the annuity's "market minus 2% or 5%" scheme.
You can't merely compare two separate monte carlo runs. The randomly
applied variable would need to be held constant over both runs. Quite
a difficult task.
Backtesting isn't perfect by any means (past performance is no
guarantee of future results), but it is at least indicitive of
potential reality and allows side by side comparisons.
