By Dr. James M. Dahle, WCI Founder
Every few years a financial organization, often a life insurance company, commissions a “study” that shows whole life insurance is awesome. The motive behind these studies is usually pretty obvious. The most recent one going around is done by Ernst & Young, a company that consults for insurance companies. Here's what their website says:
“Our global team of professionals combines industry knowledge and technical experience to help with your most pressing issues. Whether through our tax and audit advice or our innovative advisory services, we help insurers explore M&A strategies, adopt new business models, develop new products, embrace technology, optimize customer experience and address shifting workforces.”
Conflict of interest? Of course. The most “pressing issue” for companies selling whole life insurance is that people are wising up to the fact that almost no one needs these policies and once they understand how they really work, few people want them.
As expected, the “study” comes out with some very pro-insurance recommendations. The study purports to determine whether adding whole life insurance and a deferred income annuity will “add value” to an investment-only strategy. They compare five strategies to one another:
- Investments only
- Investments plus term life insurance
- Whole life insurance plus investments
- Deferred income annuity plus investments
- Deferred income annuity plus whole life plus investments
If you've seen a few of these, you won't find the conclusions surprising. Here is the overall conclusion:
“Our analysis shows that integrating insurance products into a financial plan provides value to retirement investors. Insurers can use these products to strengthen their relationships with investors and seize upon the possibilities in a marketplace that has proved challenging.”
Yea, the retirement savings marketplace is challenging to insurance companies because they're trying to convince people to buy their policies as retirement accounts, and they simply do not work as well as traditional investments inside tax-protected accounts.
Do You Believe the Study's Conclusion?
Call me cynical, but I've run the numbers enough times myself that I'm a little skeptical of their conclusions. Maybe it's because the “study” itself basically looks like a brochure for a whole life insurance policy. I mean, who puts a picture of a family collecting pumpkins into a real study?
While most people only read the conclusions of any study, anyone who has spent any time at all studying evidence-based medicine knows the real information is in the process and assumptions sections of a paper. Let's take a look at it and see what we find. Luckily, it doesn't take that long to read the entire 18-page study since it has so many glossy photos, fancy charts, and case studies any prospective buyer could relate to in it. It helps if you start at the end and just look for the fine print, which is really the only part of this sort of thing worth reading. Let's list out all the problems with the study.
18 Problems with the Ernst & Young Insurance Products in Retirement Study
#1 Ridiculous Advisory/Investment Management Fee
The best way to make whole life insurance look good is to tack a bunch of fees onto the investments you are going to compare it to. In this case, they used a 1.25% advisory and investment management fee. Now the industry average is 1%, but no informed consumer with enough money to consider whole life insurance should be dumb enough to pay that. And they certainly don't want you to compare whole life insurance to investments without a fee. How much of a difference does 1.25% a year make? Well, let's say one investor makes 8% a year and another makes 9.25% a year, both investing $50K a year for 30 years and then letting it run for another 30 years. The 8% investor ends up with $5.66M after 30 years and $57M after 60 years. The 9.25% investor ends up with $7.14M (26% more) after 30 years and $101M (78% more) after 60 years. That's a pretty big difference and obviously a pretty significant assumption. But wait, there's more.
#2 Ridiculous Equity Turnover Ratio
The study also used an equity turnover ratio of 25% in their assumptions. We're not talking about the inherent annual equity turnover in a mutual fund (which is < 5% in a total market index fund). We're talking about selling the entire taxable portfolio every four years and realizing the capital gains on it. Now I know that due to our charitable giving that I can invest our taxable account in a VERY tax-efficient way (we NEVER pay a realized gain, actually deduct losses every year, and flush capital gains out of the accounts by gifting appreciated shares), but 25% a year is ridiculously tax-inefficient. Of course, that's going to make the investment comparison look bad. But wait, there's more!
#3 Ridiculous Equity Dividend Rate
Dividends make a taxable investment less tax-efficient, they basically function as a mandatory, taxable return of capital. The “study” uses a dividend rate of 2.5%. That doesn't seem crazy until you realize the dividend yield of a US total stock market index fund as I write this is 1.37%. Yes, almost doubling the assumed yield is going to have an effect on your results. Yes, that effect is also to make the investing comparison look bad. But wait, there's more!
#4 Initial Taxable Equity Basis
Here's another hilarious finger on the scale. The assumption is that you are starting with a taxable investment account that already consists of 50% gains. That's right, before it grows at all it's somehow already had 100% in gains. That means when you sell it, you'll pay more in taxes, again making the investment comparison look bad. The basis in my taxable account is far closer to 100% of the portfolio value than 50%, and I'm already financially independent! But wait, there's more!
#5 High State Tax Assumptions
In the study, they assumed a state income tax of 6%. Now you'll obviously pay more than that in California and New York, but you'll pay a lot less in many states, including 0% in seven to eight states (NH only taxes interest and dividends). Only 18 states tax ANY income at more than 6%, and since only nine states have a flat income tax, due to the way tax brackets work, most people in those states don't have an effective tax rate of > 6% on their state income taxes. Again, this makes investments look worse than insurance by comparison.
#6 Weird Tax Assumptions
I find it bizarre that the people in their examples only pay 15% on their capital gains but somehow have enough money to have an estate tax problem where they owe 40%! None of the case studies have anywhere near enough money to have an estate tax problem under current law. Why are we even talking about estate taxes? If you're going to include a tax assumption, how about you tell us what tax rate you applied to the retirement account withdrawals? Nope, wouldn't want to do that. Because that makes a huge difference in a study like this. I'll bet dollars to donuts it was too high.
#7 Black Box Insurance Products
So what insurance products do Ernst & Young compare these investments to? They don't tell you! They call it an “industry-representative whole life illustration” and an “industry-representative product” for a deferred income annuity. Basically, they're saying “just trust us”. Forgive me for wondering if those are the products most frequently sold to consumers using this “study” to convince them to buy. They certainly don't include the illustration or even the name of these products in the study. Maybe they don't even exist. But of course, they are using the healthiest person they can find to buy the policy. No tobacco use, no risky hobbies, no medical problems. Wouldn't want to do anything that would make insurance look worse, like using the average policy sold.
#8 The Whole Life Policy Is Managed
In the appendix, we learn that “Premiums are level until age 65 when the policy goes paid up, lowering the base face amount to what is supported by the cash value”. This isn't a bad idea for someone using their policy as a retirement account, but in my experience, very few permanent life insurance purchasers are serviced in this way by agents. The agents aren't even in business anymore by the time those who bought policies from them need to make these changes in retirement. If you go to a new one to help you, how are they going to get paid for that? They're not; they're going to try to sell you a new policy!
#9 Some Wacky Stuff with Projected Policy Performance
One of the more interesting things they do is use a dividend interest rate model to adjust the projections in the illustrations. What good are the illustrations if we can't even use them to project future returns from the policy? Why do they need to be adjusted? Now I'm no actuary, but I'll bet this adjustment makes the insurance product look better. They also start with the “projected” column in the illustration, not the “guaranteed” column. In my experience, the actual returns usually fall between those two numbers, so using the projected numbers is another thumb on the scale.
#10 Weird Reinvestment of Income Annuity Proceeds
You can't make this stuff up. “In retirement, we assume the investor takes 50% of the dividend for retirement income and allocates the remaining 50% to purchase more DIA with IIP.” So despite buying an income annuity, presumably for retirement income, you're not actually going to spend all that income. You're going to buy even more income annuity with half of the income. You'd have to ask an actuary how that affected the study, but I've got a feeling it makes the insurance look better.
#11 Only Replaced Fixed Income with Insurance Products
As you dive into the “case studies”, you see that they didn't replace the equity investments with any insurance products. They only replaced bonds with insurance products. Obviously, they're going to compare a lot better when you only compare them against the lowest returning part of the portfolio.
#12 The Investor Pays Back the Whole Life Policy
You can't make this up, but it's obvious why it is in the fine print. “The investor is assumed to repay the policy loan once their portfolio recovers sufficiently from the down market.” The study is both counting the “income” (money borrowed against the policy) and the full face value of the whole life insurance as part of the legacy, because you're going to pay the money back somehow despite the fact that you already spent it on living expenses.
#13 Bizarre Investment Choices
The first case study is a 25-year-old couple making $80K a year and saving 20% of it, or $16K a year. For some bizarre reason, they put $14,400 into a 401(k) and spend $1,600 a year on a whole life policy instead of putting more money into the 401(k), potentially getting a match, or even using the Roth IRAs available to them. I've run the numbers six ways to Sunday and never, ever, has whole life insurance policy performance come anywhere close to what you would get with a tax-protected and asset-protected retirement account. It's not even close. Choosing to buy whole life over maxing out available retirement accounts is dumb. This couple is supposedly going to have $61K in today's dollars in retirement income. In 40 years. Once you apply inflation for 40 years to those tax brackets, the tax rate on that income will be minuscule. Their standard deduction will probably be $61K in 40 years.
But wait it gets worse, supposedly the best results come from putting 50% of their savings into whole life insurance. I'd love to see the work behind this one, but of course it isn't included in the “study”. I guess there wasn't enough room left after they put all the glossy photos in. But seriously, a 25-year-old couple is supposed to put 50% of their savings into whole life insurance? And that is supposed to provide the most income and the most legacy 70 years later? Give me a break. And besides, up above they said they were only putting “bond money” into whole life. Are you now telling me you're recommending this 25-year-old couple have a 50% bond portfolio? This is all getting more bizarre as we go.
#14 70-Year Time Horizon
The first case study also uses a 70-year time horizon. You have to use these long time horizons to make whole life look good, because it looks absolutely horrible for the first decade or two when returns are negative. But the truth is that 70%-80% of whole life policies are surrendered within the first 30-40 years, many at a loss. In fact, 1/3 are surrendered within 5 years. Seems silly to do 70-year projections on something that on average people only keep for 20 years.
#15 No Discussion of Cancelling Term
A major problem with most of these “studies” is that they assume the person who buys term and invests the rest keeps that term life policy until they die at age 80 or 90. The whole point of buying term and investing the rest is to save money on insurance and use that money to invest or spend. So when you hit financial independence, you cancel the life insurance. This “study” never indicates when the term life is canceled, which leads me to believe they ran their figures with it never being canceled. Obviously, the cost of term life at age 85 is going to eat up an awful lot of that retirement income.
#16 “Whole Life Beats Bonds”
They make the outlandish claim that whole life insurance performance beats bonds. That's not actually true. If you look back to the “golden age” of whole life insurance, policies bought in the early 1980s when interest rates were sky high, you can see that people had returns of 7-8% on their policies over the next few decades. But 30-year treasury bonds bought in the early 1980s were yielding 11-14%. Even in 1994, you could still buy an 8% 30-year treasury bond. Whole life didn't outperform bonds and it still doesn't especially in the first decade when they have negative returns. They just expect you to believe them when they tell you it does. They even try to explain why:
- Participating insurance products tend to outperform fixed income because mutual life insurance companies, as institutional investors, have access to asset classes that individual investors do not. (This is a lie. Look at the portfolios of insurance companies—they're mostly filled with the same boring old bonds Vanguard bond funds buy on your behalf. The vast majority of what they buy you can buy, too, without any issue at all and avoid losing their cut of the return.)
- These companies also have professionals managing their assets, which has been proven to provide value for fixed income. (Not true, index funds outperformed actively managed funds, even in fixed income. See the chart below for details.)
- Whole life tends to provide superior returns over fixed income in long-run scenarios due to the combined effect of the guaranteed growth of cash value and dividends. (No, that's not really true either as we discussed above with the best whole life policies ever sold.)
- Using whole life as a volatility buffer improves returns because the investor does not have to sell and realize losses on their investments. (During a market downturn, you can sell appreciated bonds instead of temporarily decreased stocks just as easily as you can borrow against whole life.)
Lie, lie, lie, lie. None of it is true. Whole life does not beat bonds. It certainly does not beat them in a retirement account. And it doesn't even come close to beating them except in the very long term in a taxable account of someone in high brackets.
#17 No Discussion of SPIAs
Want guaranteed income in retirement? Buy a Single Premium Immediate Annuity (SPIA). No reason to buy a deferred income annuity at age 55 if you're not going to start spending for another decade or more. Better to leave that money in higher returning investments until you need to start getting guaranteed income. Want to solve for the highest possible income? Invest aggressively until age 70 and then put all that money into SPIAs. You'll have dramatically more money to spend.
#18 No Significant Difference
Despite all of the advantages given to insurance products in this study, if you look at the results they all look the same. Retirement spending for the first case study couple ranged from $61K-$66K. Amazingly the less you spent on insurance products, the less money you ended up with. But if you think your life is significantly better and worth dealing with the hassle of not one but two complicated insurance products for 70 years for an extra $400 a month, I've got a bridge to sell you. I can't believe with all the help they gave the insurance products in this “study” that they couldn't make a larger difference than that. I mean, look at the effect of that 1.25% advisory fee alone on the $16K this couple invested per year for 40 years. It's a $1.6 million difference. That's $65K a year in additional retirement spending. And we're bragging about winning by $5K a year? The difference in their study was not significant, but you'd never know that unless you combed through the results with a fine-toothed comb. The conclusion does not match the results.
If they had explained more of what they were doing, I'm sure there would be even more to criticize about this sales brochure masquerading as an academic study. They have truly tortured the data until it confessed. But given we're now approaching 3,000 words in this blog post, it's time to wrap it up.
The bottom line is that this EY Brochure is not a study at all and it should not enter into your consideration of buying whole life insurance or an annuity at all. If you need or want whole life insurance, and truly understand how it works, then go buy some. Same with annuities. But don't let an insurance company-funded study convince you that you are likely to come out ahead of a buy term and invest the rest approach to investing. If you will do that, you are likely to reach retirement age with FAR more money, and I assure you that having more money at the time of retirement will allow you to generate more retirement income and/or leave more money behind to your heirs. Buy insurance for the guarantees when you need or want the guarantees. Otherwise, avoid it.
What do you think of the paper? Why do companies put these easily debunked studies out? How many people do you think were convinced to buy whole life insurance by this paper? Comment below!