Index Annuities: Are the Critics Wrong?

Posted at on November 30, 2009

“Equity-indexed annuities? Don’t bother,” said Consumer Reports Money Adviser in September 2005. That sums up the attitude of many fee-only advisors, and mine as well. I understand the theoretical case for principal-protected products, but that doesn’t make me a fan of the actual products in the marketplace. High commissions, high surrender charges, high taxes and a lack of transparency — is that the best that we can do?

Jack Marrion, Geoffrey VanderPal and David F. Babbel have produced a serious defense of index annuities: “Real World Index Annuity Returns,” Wharton Financial Institutions Center, October 5, 2009. David F. Babbel’s presentation, “Un-Supermodels and the FIA,” at the November 2008 Ibbotson Associates/IFID Centre Conference on Guaranteed Living Income Benefit Insurance Products contains more technical details.

They argue that index annuity critics have made flawed assumptions about contract design, nonguaranteed elements (such as participation rates), consumer behavior and stock market returns. As a result, the critics’ conclusions do not fairly describe the actual or even hypothetical performance of index annuities.

What is the money’s worth of index annuities; that is, the expected present value of benefits divided by the expected present value of premiums? Critics have said that it is less than 85%; the defenders argue that it could be more than 95%, which would put index annuities in the same league as immediate annuities for efficiency.

This new research is long overdue, but it falls well short of what is needed to make a solid case for index annuities. There is a real danger that each new set of methodological refinements will lead to oscillating conclusions about the merits of the products. This is not a good situation if you are trying to incorporate the best academic research into the guidance that you give to clients.

Here are the questions that I have:

1. What are the results if you take account of the variable costs of hedging and the insurer’s development, distribution, issuing, administrative and capital costs?

The critics and defenders do stochastic modeling of stock returns (with different assumptions), but they don’t do stochastic modeling of the cost of the hedging instruments (often over-the-counter call options) that provide the upside potential.

And let’s get down to the nitty-gritty of the manufacturing costs of index annuities. What are the commissions and other distribution costs? What does it cost to develop the product, file the necessary forms with each state (with maddeningly different regulations), deliver the contracts, process each transaction and provide the annual statements? What is the cost of the capital that must support the annuities?

When you buy an index annuity, you are paying an insurance company to construct and manage a portfolio of derivatives and fixed-income investments that provides some upside potential with limited downside risk.

How much are you paying for this service?

2. What are the results if you take account of taxes?

The tax treatment of nonqualified annuities is a mixed bag. Gains are tax deferred, but distributions are interest first, principal second, and they are taxed as ordinary income.

With capital assets, such as mutual funds, you have a lower tax rate on long-term capital gains, a step-up in basis at death, and the ability to harvest losses to offset gains.

When you defer income tax, you incur the risk of being subject to a higher tax rate in the future.

On the other hand, starting in 2010 you will be able to use gains in nonqualified annuities to pay for long-term care benefits.

David F. Babbel and Ravi Reddy have analyzed the taxation of annuities (“Measuring the Tax Benefit of a Tax-Deferred Annuity,” Journal of Financial Planning, October 2009), but they did not consider all of the aspects of it.

3. What other factors need to be considered?

Marrion, VanderPal and Babbel mention the creditor protection benefit of annuities, and annuities also avoid probate. Some contracts provide a guaranteed minimum income benefit that would need to be valued.

By design, index annuities have only a small number of indexing choices, such as the S&P 500. What is the opportunity cost of this limitation?

What about the lack of liquidity, due to surrender charges and tax deferral? Three researchers quantified the cost of illiquidity for non-index fixed annuities (S. Browne, M.A. Milevsky, T.S. Salisbury, “Asset Allocation and the Liquidity Premium for Illiquid Annuities,” Journal of Risk and Insurance, September 2003). Another researcher recently studied the broad impact of liquidity on asset pricing and portfolio construction (Francis A. Longstaff, “Portfolio Claustrophobia: Asset Pricing in Markets with Illiquid Assets,” American Economic Review, September 2009).

How much additional yield should you demand as compensation for the illiquidity of index annuities?

4. What are the pros and cons of different product designs?

There are many interest-crediting formulas, and they produce different results. Based on my experience with older point-to-point designs, which can give you nothing to look forward to after a big loss in the index account, it seems that a reset feature is important.

What do normative theories of rational choice and behavioral finance have to say about index annuity product design?

Do different product designs have significantly different hedging costs?

5. Why should I tolerate a black box?

What are index annuities indexed to? There’s a reference value, such as the S&P 500, but almost all index annuities give the insurer the right to adjust the interest-crediting factors (such as participation rate and cap) after issue, subject to minimums and maximums. This makes sense, because the insurer’s hedging costs will fluctuate as interest rates and stock market volatility change. However, insurers do not explicitly link their interest-crediting factors to their hedging costs, and they don’t offer much explanation about the determination process.

For example, this is from American Equity Investment Life Insurance Company’s Actuarial Opinion of Nonguaranteed Elements in their 2008 Annual Statement: “Interest crediting rates, PRs [participation rates], AFRs [asset fee rates], and caps are established by management after considering earned rates and the spread assumed in pricing, and weighing marketing and competitive considerations.”

For Marrion, VanderPal and Babbel, managerial discretion creates more doubts about the critics’ analyses: “This human element introduces a random variable that cannot be quantified, thereby making any attempt to project any returns ultimately subjective.”

For me, managerial discretion is an additional risk factor that should be considered in comparing index annuities with other financial products. Have insurers consistently met their profit objectives when setting nonguaranteed elements, or have they been willing to give up profits to buy market share? How can I use past performance to judge future performance without knowing this?

6. Why shouldn’t I wait for better products to come along?

Index annuities typically have significant surrender charges for a long time, so you are locked in after purchase. Does the interim benefit of principal protection adequately compensate buyers for making an immediate purchase, instead of waiting to see if there are better ways to get principal protection?

Why can’t I separate the investment and risk management functions and buy a principal protection wrapper for an investment portfolio?

So I’ll ask my first question again: I mean, really, are today’s index annuities the best that we can do? Really? That’s it, America?