From the November 2010 issue of Research Magazine • Subscribe!

How to Get an A on the Exam

Our columnist has been "grading" a variety of different variable annuity products in the past year; in this article, Prof. Milevsky instructs readers in how to get an "A" on the exam.

In the last few issues of Research magazine, I have been “grading” different variable annuity products and focusing on the value of promises associated with guaranteed living income benefits (GLIBs). According to industry estimates, GLIBs now account for almost 80 percent of new sales of VAs, which were approximately $31 billion in the first quarter of 2010, according to Morningstar. The volume of this business is remarkable when you consider that GLIBs are an insurance rider that didn’t really exist a mere 15 years ago. If you missed some of these product evaluations, you can find links to previous Annuity Analytics articles and related in-depth information at my newly re-designed website www.MosheMilevsky.com.

As these columns have hit the newsstands, so to speak, many readers (and some manufacturers) have contacted me to ask exactly how I evaluate these products and specifically what it takes to get an A on this so-called final exam. Since I’m in the education business and am used to these questions from generations of anxious college students, in this month’s column I will discuss the factors that are likely to earn a product an A+ (a.k.a. the coveted 10/10.) Consistent with the exam analogy, however, I can’t quite solve all the questions for students in advance, but I’ll offer some guidelines. Allow me to focus on five general factors.

1. The Buyer

If I can play the role of financial psychoanalyst for a moment, I believe there are three distinct personality types for whom the variable annuity makes sense. Imagine the financial advisor as a parent with three types of children. The first child is concerned with income-tax risk, and specifically the fear that income-tax rates will be increasing (or tax cuts will be phased out) very soon. The second child is concerned with a lack of corporate pensions, guaranteed retirement income and the impact of longevity risk. The third child may or may not be worried about what scares the first two children, but is fearful about financial markets in general and the volatility of stock prices in particular. The third child is concerned about investments losing value and wants protection before wading back into the water. For the time being, the third child’s money is under the mattress. Occasionally these personalities are merged into one composite (and complicated) individual, but at a minimum, there is a distinct risk that worries them.

So, the ideal product makes it clear, in both design and description which personality (child) it is targeting. Products that perform well in targeting the proper fear score highly in my ranking. I am not a fan of products that take the “kitchen sink” approach to riders and confuse the phobias. Case in point is a so-called nursing home rider on a GLIB. At first glance it appeals to those fearful of dying penniless in a decrepit old-age home. But upon further inspection these only allow for an increase in your allowable withdrawal rate. Is this truly nursing home coverage? I don’t think so. The prescribed therapy does not match the client’s phobia.

2. The Rates

A 6 percent rate is always larger that a 5 percent rate. A compound return is better than a simple return, and stacking (i.e., giving you the best of both) is better than either a roll-up or a step-up, by definition. This is simple math and relatively easy to compare across products. Guaranteed payout rates that increase with age are better than flat constant ones, etc. Increases that end after 10 years are less preferable than those with 15 years to run, etc. Numbers lend themselves to comparison, but too often end up the only thing being considered. Numbers crowd out subtle but important differences that spreadsheets can’t evaluate. I would hate to pick insurance company A over insurance company B simply because they were offering a bonus of 5 percent versus 5.5 percent. There are myriads of other factors where they may differ. I doubt that a 50 basis point difference in a guaranteed payout, which is worth a lot less than you think on a present value basis, can reverse the tide. In fact, even 1 percent spreads may not be worth the hassle. Sure, a 30-year U.S. Treasury bond that pays a 4 percent coupon is undeniably preferred to a 30-year bond paying 3.5 percent. There are no ifs, ands or buts. But the same can’t and should not be said about VAs. The lesson here is that a product that screams “we pay more than the competition” — even if it is true — will not necessarily garner a top score. Sizzle might sell the steak, but it doesn’t satisfy the health examiners.

3. The Investments

While the consensus from wholesalers and other product experts seems to be that the promises and guarantees are what differentiate variable annuity products and companies from each other, I lament the ongoing marginalization of the investment and subaccount chassis.

I don’t remember who lent me this analogy (and I apologize to the owner) but great lifeboats on a cruise ship shouldn’t be the reason you take the cruise. (And neither should the food, for that matter.) Ultimately, consumers are buying investments, real companies and taking risks. They want appreciation over time, so the step-ups and ratchets can actually work. Let me make this perfectly clear. If all they want is lifetime income, then buy a single premium income annuity, which might pay 8 percent, 10 percent or even 12 percent for life. They want upside and so there must be a reasonable expectation that money managers can deliver it. So, while I certainly don’t want to start ranking and comparing 10,000 different subaccounts, and the relative merits of growth fund A versus value fund B, I appreciate products that offer a wide variety of investment choices — popular, cost effective and uncorrelated — within their lineup. Moreover, if they allow for exchange-traded funds or at least some low-cost index funds, they score extra points. At the very least, allow me to protect something that needs protecting. Condominiums do not require lifeboats.

4. The Gotchas

I am borrowing a phrase and concern from fellow annuity connoisseur Mark Cortazzo, but too many policies have “gotchas” buried deep within the prospectus or product description. These are insurance company escape valves that you stumble across by chance, or someone who has actually bothered to read all the documentation and fine print points them out to you. And although they are disclosed in the legal sense of the word, they certainly aren’t highlighted or emphasized. Obviously, the fewer “gotchas,” the better. For example, products that allow too much company wiggle room on investment fund choices allowed, or the exact amount of income to be received in the event the account is annuitized, or excessive discretion with costs, are all “gotchas.” I’ll confess here that when I discover these items myself, after an hour of mindless reading, they drive me batty. The response of “we don’t expect that to happen” does not cut it in a post-Lehman world. My home insurance hedges my risk 24/7. So should my living benefit.

5. The Fees

No review of variable annuities is complete without a discussion of fees. I am talking about the base mortality and expense fee, the investment management fee and the additional fees for the riders, which are computed based on the protected value and are therefore equivalent to even higher fees on an asset basis. It goes without saying that all else being equal, products with lower fees score higher. This is a rather obvious tautology. But on a more subtle level, products that allow the client/advisor to pick and choose the insurance and investment features they like and then base the all-in cost on the choices made, score even better. For example, if an actively managed growth fund is more costly to insure and guarantee — from the company’s perspective — compared to a balanced passive portfolio, then why not have the insurance fees reflect this? Likewise, if a guarantee made to a 50-year-old female (who is expected to live longer) is more expensive to provide compared to a similar guarantee made to a male, then the fees should reflect that as well. Should your car insurance premiums be based on the average driving record of someone in your zip code, or based on your personal driving record? I think I know the answer. So here is a thought: Start underwriting variable annuities.

***

Variable annuity policies aren’t commodities and can’t easily be compared in the same way a generic equity fund can be contrasted across fund complexes. At last count of mine, they can differ across 20 different dimensions and any attempt to rank them is fraught with difficulties. That said, there are aspects that are more valuable relative to others, and that’s what grading is all about.

Moshe A. Milevsky, Ph.D. is a professor at York University in Toronto, Executive Director of the non-profit IFID Centre at the Fields Institute and CEO of the QWeMA Group (www.qwema.ca). His recently co-authored book Pensionize™ Your Nest Egg: How to Use Product Allocation to Generate a Secure Retirement, was published (by Wiley, in Canada) in September 2010.

Reprints Discuss this story
This is where the comments go.