Product manufacturers in the fixed index annuity (FIA) world seem to be broken down into two groups.
Those who believe, build, and use exotic indices and those who don't.
The latter warn about the dangers of back-tested results. They'll say things like, "have you ever seen a bad back-test?" Surely, you haven't.
(Related: Know Your Clients. Well.)
And I believe most advisors are aware of the risks associated with the newest flavor of the month index birthed by some of the brightest numerically oriented minds within the insurance industry towers. However, more intriguing is the total disregard to a much bigger fault or risk associated with back-tested results, which is present in traditional and exotic indices.
Here's the two common components of back-testing in FIA's:
- How the index would've performed during previous time periods prior to the index's existence; and,
- The amount of an index's growth a FIA would've participated in prior to the FIA's existence.
In regard to the first component, this only applies to indices too newly formed to have historical data to reference. I believe most advisors understand the risks associated with these back-tests, so I will move on without discussing this further. The risks of the second component are much less understood and seemingly never discussed.
The second component applies to freshman indexes as well as indexes of tenure, like the Dow Jones Industrial Average or the S&P 500. What's the danger to back-testing an FIA's would have been growth, for a period when it didn't exist, using actual historical performance of either of these mammoth of banality indices? Simple, crediting rates are susceptible to changes within an insurer's overall book yield.
Maybe this isn't a glaring error. A few years ago, at a home office event, a seasoned producer proclaiming, "an FIA will do better than the market." To prove his point, he showed that company's marketing piece, which illustrated an FIA outperforming the S&P 500 over a specific 10-year period. Obviously, this wasn't intended to implicate this product would outperform the S&P 500 every 10-year period, or even most, but that's the way he took it. That's also the way he presented it to his clients.
As a side note, I've always wondered if they rescinded his contract due to a clear lack of understanding of their products.
Is this mistake too egregious to contemplate any advisor north of the industry median committing? Yes, but many advisors are making a similar mistake.
Recently I saw a FIA illustrated with a high enough annual participation rate (it also had a spread) to give an annualized rate of return of around 90% of the S&P 500's gains over the last 40 years, although its only be in existence for the last few.
The advisor was very confident in this illustration. It showed nearly 40 years, both good and bad. Thankfully, in this case, it was an easy conversation with the now current client. I simply asked, "do you really believe you can get 90% of the good with none of the bad?"
He didn't and neither do I.