A recent column on variable annuities provoked the predictable torrent of hate mail from those who sell the product. I was, they declared, wrong, mis-informed, pre-judiced, dumb, ignorant or some combination of all those qualities.

The sales force is still long on bombast and short on analysis. Three planner/salesmen suggested that I take the time to become informed by reading a book on variable annuities. They all suggested the same scholarly tome, "Investing with Variable Annuities: Fifty Reasons Why Variable Annuities Are Better Long-Term Investments Than Mutual Funds" by John P. Huggard, proving that they applaud material that is positive about variable annuities but are critical (rather than curious) about material that is negative.

Mr. Huggard's book, as some readers will recall, was never sold in retail outlets, was written for the industry, and is filled with errors. Miraculously, 100 percent of his errors work in favor of variable annuities.

The most surprising communication was from an actuary in Minneapolis. For reasons that will soon be obvious, we'll call him Mr. Schmidt (with apologies to Jack Nicholson). To prove that I was wrong and variable annuities were right, the actuary sent along an Excel spreadsheet. It purported to compare the discounted after-tax values of a variable annuity and a mutual fund after a 20-year accumulation period. Both earned the same gross return, which is a way to test the value of the annuity "wrapper" and its tax deferred compounding.

Guess what?

The Schmidt model showed the annuity beat the taxable mutual fund.

Then I examined the innards of the spreadsheet. In his comparison, Schmidt made three important errors. All three favored the variable annuity:
  • First, he neglected to subtract the insurance cost from the gross return. Since 88 percent of the 631 variable annuity contracts being sold have annual insurance expenses of 1 percent or more, this is a significant omission.
  • Second, he assumed the same 25 percent tax rate for both investments. No matter that the "Jobs & Growth Tax Relief Reconciliation Act of 2003" cut the tax on dividends and capital gains to 15 percent. He simply dismissed current tax law.
  • Third, he assumed that the taxable mutual fund realized every penny of its return in each year. While this happens with some funds, it is quite easy for a prudent investor to select managed funds that are highly tax-efficient. (A fund is tax efficient when it losses little of its annual pre-tax return to taxes.) A simple alternative is to invest in large cap index funds because they are tax efficient by their nature.
Now guess what happened when the spreadsheet was corrected?

The results changed.

A taxable mutual fund was a better deal than a tax deferred annuity. It was better after one year. It was better after five years. It was better after 10 years, 15 years, and 20 years. Under current tax law the typical variable annuity is a profoundly poor investment choice. That's why I called their sale "financial malpractice."

To give you an idea of just how bad variable annuities are today, here are the results of a comparison that still has assumptions loaded against the taxable fund--- it assumes that income from the annuity is taxed at only 25 percent (many buyers are in higher tax brackets). It assumes that 100 percent of all income from the taxable fund is taxed each year. And it assumes that the fund has no tax-loss carry-forwards. (Recall from the earlier column that more than half of all equity mutual funds still have tax loss carry-forwards and the average loss is 32 percent of net asset value. In terms of taxes, this is the best time to purchase a taxable mutual fund I've seen since I began writing a personal finance column in 1977.)

The disadvantage of a $10,000 variable annuity investment is $175 at the end of the first year. This does not include the early withdrawal fees you would incur if you decided to liquidate your investment at that time. By the tenth year, when you are clear of such fees in most contracts, the variable annuity disadvantage is $2,355. By the 20th year the variable annuity disadvantage is $6,587.

The taxable mutual fund shows net after-tax growth of $12,610 after ten years. The variable annuity trails by $2,355, a growth reduction of 18.7 percent. After 20 years the after-tax growth of the mutual fund is $41,120. The variable annuity trails by $6,587, a growth reduction of 16 percent. The fact that the growth reduction declines over time indicates there is some benefit to tax-deferral--- but it is not large enough to overcome the insurance expense and the higher tax burden.

The 20 Year Variable Annuity Disadvantage
This chart compares and investment of $10,000 in a variable annuity and taxable mutual fund with a gross return of 10 percent annually. The variable annuity return is reduced by a 1 percent annual insurance expense but all taxes are deferred. The mutual fund is assumed to distribute all of its 10 percent return each year. As a consequence, taxes are paid each year and the mutual fund has no deferred tax liability.
Year VA net of insurance expense Less future VA tax liability VA after 25% tax Mutual Fund, all taxes paid Mutual Fund Advantage
1 $10,900 $     225 $10,675 $10,850 $   175
5    15,386    1,347    14,040    15,037        997
10    23,674    3,418    20,255    22,610 2,355
15    36,425    6,606    29,819    33,997 4,179
20    56,044 11,511    44,533    51,120 6,587
Source: Authors' Spreadsheet Calculations

Query: What would it take to make the VA a better deal than the taxable fund?

Answer: If you cut the insurance expense to 0.20 percent the variable annuity will have an advantage of a whopping $32 in the 20th year. In the entire universe of over 600 contracts only a handful come close to that expense, notably TIAA-CREF at 0.30 percent.

But what if tax rates on investment income rise again?

Well, they might. With a one percent insurance expense a variable annuity won't catch up to a taxable fund for over 20 years if tax rates on investment income went to 20 percent. If the tax rates were equal at 25 percent, it would take tax deferred compounding 14 years to overcome the disadvantage of the 1 percent insurance expense.

As I've written many times, variable annuities are a sold product. The sales people may calculate their commission but their research, curiousity, and due diligence appears to stop there.

Related columns:

The "Financial Malpractice" column (Thursday, 7/31/03)

William Reichenstein examines Huggard's book (Tuesday, July 23, 2002)

Variable Annuity Watch columns