Mutual Funds Versus VAs

 

Saturday, Mar 07,2009, 2:48:54 PM   Click:

Copyright 2009 SourceMedia, Inc.All Rights Reserved Bank Investment Consultant

February 2009

RETIREMENT EDGE; Pg. 32 Vol. 17 No. 2

1590 words


Mutual Funds Versus VAs; Do the higher costs of variable annuities and variable life insurance products cause them to underperform mutual funds? Not always.

Craig L. Israelsen



It's no secret that variable products cost more than mutual funds. The average annual expense ratio for variable annuities (VA) is about 2.4% (or 240 basis points), compared with 1.5% for variable universal life (VUL) and variable life (VL) policies.

Compare those costs with the average annual expense ratio for stock mutual funds of 1.4% and 1.1% for bond funds, producing an overall average of 1.3% for all 26,529 stock and bond mutual funds in the Morningstar Principia database.

Variable products are contracts of insurance as well as investment products that use mutual funds as the underlying investment vehicle. Because they're two products in one, variable annuities and variable life insurance have higher expense ratios than mutual funds. The question is: Do the higher costs of variable products cause them to systematically underperform mutual funds?

Understandably, variable products aren't parallel because mutual funds don't provide life insurance, so it's not exactly a fair comparison. Moreover, when implemented correctly, variable products may qualify for more favorable tax treatment than mutual funds. Nevertheless, baseline performance comparisons, and comparisons of the underlying costs can help you understand the pros and cons of variable products and when they make the most sense for your clients.

We used large-cap U.S. equity blend (the middle ground between growth and value) as the underlying investment in the variable products. As of Nov. 30, 2008, there were 2,025 U.S. equity mutual funds in the Morningstar Principia database classified as "large-cap blend." They had an average expense ratio of 1.24% and their average three-year annualized return as of Nov. 30 was -9.58%. (See "Equity Portfolios Comparison: Equity Portfolios" on page 33.)

By comparison, there were 2,771 large-cap blend VUL/VL sub-accounts with an average annual expense ratio of 1.33% and an average three-year performance of -10.11%. (VL and VUL are combined and reported together.) Finally, the average expense ratio among 6,976 large-cap blend VAs was 2.29% and the average three-year annualized return was -11.15%.

In the large-cap blend category, the higher costs associated with variable products hurt their performance relative to the performance of mutual funds, at least in absolute terms. However, the real question is whether or not variable products performed better than expected given the difference in their expense ratio?

For example, as a category, the expense ratio of VUL/VL products that invest in underlying U.S. large-cap blend funds was nine basis points higher than the average large-cap blend mutual fund (1.24% vs. 1.33%). In light of that, we would expect large-cap blend VUL/VL products to underperform large-cap blend mutual funds by nine basis points. However, as shown in the chart "Performance Premium: Equity Portfolios," VUL/VL products underperformed mutual funds by about 40 basis points. Thus, as a category, large-cap, blend-based VUL/VL products showed a negative performance premium over the past three years.

VA products with an underlying investment in large-cap blend U.S. equities had an average expense ratio of 2.29%, or 105 basis points more than the average expense ratio for large-cap blend mutual funds. The average three-year performance for the VA products was -11.15% compared with -9.6% for mutual funds, or a difference of 157 basis points. While the "expected" underperformance of VA products was 105 basis points, the actual underperformance was 157 basis points, producing a negative performance premium of 52 basis points.

Interestingly, when the variable products (VUL/VL and VA) invested in underlying non-U.S. equity large-cap blend mutual funds, they showed, on average, a positive performance premium relative to comparable (non-U.S. equity large-cap blend) mutual funds. The positive premium was about 50 basis points for VUL/VL products and just under 40 basis points for VA products. Variable products using world stock funds as the underlying investment did not fare as well relative to world stock mutual funds. The negative three-year performance premium was 120 basis points for VUL/VL products using world stock sub-accounts and 50 basis points for VAs.

The cost and performance differentials for conservative and moderate allocation funds as well as intermediate-bond funds are shown in the "Cost and Pperformance Comparisons: Balanced & Bond Portfolios" on page 34.

In the conservative category, portfolios are typically 30% to 40% invested in equities and 60% to 70% invested in fixed income and cash. VUL/VL products in this group had a small relative underperformance (compared with the expected underperformance of 13 basis points). Conservative VA products underperformed conservative mutual funds by 65 basis points more than expected. However, in the moderate and intermediate bond allocations, VUL/VL products performed considerably better than expected relative to mutual funds, while VA products generated small positive premiums.

BEHIND THE NUMBERS

Of the six fund categories examined, the average performance of VUL/VL products outperformed comparable mutual funds (in absolute terms) in several categories: non-U.S. large-cap blend, moderate allocation and intermediate bond. Clearly, if they outperformed in absolute terms they also outperformed in relative terms; that is, the performance was better than expected based on the differences in expense ratios. VA products were less likely to generate a performance premium, but did so on a relative basis in the same three categories. How is this possible?

Part of the answer comes from the fact that variable products don't use a random sample of underlying mutual funds. There is a clear attempt to use "better" funds. Thus, on average, comparing the performance of variable products with the performance of mutual funds in the same category is analogous to comparing the pick of the litter with the whole litter.

For example, in the category of non-U.S. large-cap blend there were 769 mutual funds, which had an average performance of -8.05%. By comparison, there were 783 VUL/VL sub-accounts that used non-U.S. large-cap blend funds as the underlying investment, but among them there were only 69 unique funds. Among VA products, there were 2,013 sub-accounts using non-U.S. large-cap blend as the underlying investment, but only 133 unique funds. Based on average performance, the VUL/VL and VA sub-accounts picked through the litter and tended to use better funds as the underlying investment. Likewise in the intermediate bond category, there were far fewer unique mutual funds used as sub-accounts for VUL/VL products and for VAs. (There were 1,061 mutual funds and 1,144 VUL/VL sub-accounts, but only 102 were unique. Among 2,319 VA sub-accounts, only 191 were unique.)

The pick-of-the-litter theory likely explains much of why variable products sometimes compare favorably with their mutual fund counterparts. One could argue that all a mutual fund investor has to do is pick through the same litter and select a winning mutual fund. How to successfully do that is the trick. One could look through variable product sub-accounts for the mutual funds that repeatedly show up. They are not guaranteed winners, but are a consensus starting point for selecting funds.

In summary, the performance of vari-able products is often closer to the performance of comparable mutual funds than would be expected, given the higher expense ratios of variable products. This may be encouraging to the sellers and users of variable annuities and variable life insurance products. However, it's important to remember that keeping expenses low (in mutual funds and variable products) is still vital-particularly in times when equity markets are struggling to produce positive returns. BIC

Craig L. Israelsen is an associate professor at Brigham Young University, a principal at Target Date Analytics (www.TDBench.com) and the designer of the 7Twelve Portfolio (www.7TwelvePortfolio.com).

HOW VARIABLE PRODUCTS DIFFER

Variable annuities differ from mutual funds in several ways: First, VAs make periodic payments for the rest of a policyholder's life, the life of a spouse or any other designated person. Second, variable annuities have a death benefit. If the policyholder dies before the insurer has started making payments, beneficiaries are guaranteed to receive a specified amount-typically at least the amount of the purchase payments. Third, owners pay no taxes on the income and investment gains from their annuity until they withdraw their money. See (http://www.sec.gov/investor/pubs/varannty.htm)

Essentially, variable universal life and variable life both allow owners to invest premiums paid in stock mutual funds. Morningstar Principia defines VUL and VL policies this way: VUL combines the concepts of mutual fund investing with universal life insurance. A VUL policy allows the policy owner to invest premiums in various mutual fund sub-accounts. The performance of these sub-accounts adds to the death benefit of the policy.

Universal life insurance is unlike traditional cash-value policies known as "whole life." Universal life policy returns were freed from long-term, fixed-rate contracts and replaced with policies whose returns were tied to short-terminterest rates and periodically adjusted. In addition, premiums and death benefits can be changed by the policyholder. So when and how much is paid in premiums is flexible.


February 3, 2009

Copyright © 2009 LexisNexis, a division of Reed Elsevier Inc. All Rights Reserved.
Terms and Conditions Privacy Policy