Why Annuities are So Terrible (and why you might need a specific kind anyway.)


In the (somewhat mythical) “Good Old Days” you would retire from your corporate or union job with a pension. Combined with a check from social security and some of your own savings, you’d be comfortable in retirement. The biggest problem was that you might have a fixed income that didn’t keep up with inflation. But you didn’t worry about running out of money: pensions are for life.

Remember the inflation of the Carter Years? I suspect anyone reading this probably does. It increased the cost of providing pensions (“defined benefits plans”) just when the tax code was discovered to have a provision 401(k) that said an employer could define the contribution they’d give instead. Suddenly – and it was really very sudden – the world of pensions tilted away from employers buying you a pension. Now individuals had control over a pot of money that they were supposed to live on in retirement: their choice how.

The study of behavior economics is really new, but the study of how to sell things to frail, fragile, flawed humans holding a large wad of cash in their hand is an old, old profession. The insurance companies that used to be hired by employers to provide the pensions suddenly realized that they had a huge new market of newbie investors who both wanted their cake (their 401(k) balance intact) and to eat it too (to get a pension.) And so “variable annuities” became a hot commodity.

Learning about variable annuities is super hard to do. Almost no one but annuity salesmen are trained on them, and the SEC regulates them so tightly that hardly anyone, other than people who sell them, are legally allowed to advise on them. Understanding the costs and benefits is like understanding theoretical physics. I picked up an 85 page book (in a large font) to read yesterday and it was hard, hard going. I could read about 15 minutes and then had to put it down for a bit. After finishing it, and a few related articles, I had to go for a three mile walk in the woods to let it all gel into my brain.

The book is “Why Variable Annuiuties Don’t Work the Way You Think: Hidden Dangers That Can Devastate Retirees” and is written by a guy with the same credential I have, Jeffrey D. Voudrie, and is pretty succinct in his objections to them. Basically, it’s just the most expensive possible way to achieve your actual objectives. All the complexity is designed to obfuscate this truth. It is impossible to compare apples to apples in the world of annuities, with one single exception. It’s the exception that drives the engine of sales for annuities: because there actually *are* some reasons you might want an annuity.

To understand why you’d want a specific kind of annuity, you have to understand the concept of “mortality credits”. An author and speaker I admire, Michael Kitces, talks about this here. Short version: even if you don’t know how long you’ll live, actuaries have a really good idea of how long a large population will live. So let’s say you and 25 friends band together to put together an investment fund, each anteing up $4,000. The earnings of the $100,000 investment are split 25 ways the first year. Let’s say it makes 4%: you’d each get $160. Now let’s say that 1 of you die each year. The second year the money is split between the 24 remaining people: the $4,000 the investment kicks off is now split 24 ways, you each get $166.67. The year after that the grim reaper takes another, the $4,000 is split 23 ways and you each get $173.91. In the olden days there was a winner-takes-all verison of this called a “tontine”, but now actuaries sum the stream of payments up over the entire lifespan of the participants and include the return of principle that the last person to die would have gotten, so the payments become equal: some of the benefits the survivors would have gotten can be front-loaded to everyone, and you get to include the return of principle in the payment amounts, too.

This is exactly how traditional pensions worked: your employer would go to an insurance company and say, “how much do I need to pay you to give my retiring employee $25,000/year pension?” The insurance company would say, “that’ll cost $502,371” and the the employer would pay it and the employee would get a guaranteed right to the cash flow for their entire life that would end when they did. This is called a Single Premium Immediate Annuity.

Now, however, the employee gets handed $600,000 and told to have a nice retirement. And human foibles kick in. The first one is loss aversion: they absolutely hate the idea that they might lose that money, so they want to lock it up someplace tight. (But, of course, no risk = no return, so they’re foregoing earnings, but they don’t see that most of the time.) The next thing is they want to leave a legacy for their kids. No one (other than maybe Mom) expects to inherit Dad’s pension, right? But suddenly when it’s the lump of $600,000 sitting in an IRA they want to make sure their heirs get it if they die before using it. It’s a super common human instinct which insurance companies absolutely abuse. It’s important to pay for life insurance if you are financially responsible for people and might die early, but in retirement there isn’t usually a need for insurance if you die, and it’s silly to pay for insurance if there isn’t a particular catastrophe you’re covering with the insurance premium. In fact, the inverse is true: you often need to buy insurance in case you live too long! But, wait, did you just catch how hard that paragraph was to read? Yeah, insurance agents know that, too.

What people *ought* to do is pretty variable. There are some really good reasons to buy a single premium immediate annuity – the same thing your employer would have done – with much of your money. You can also make a really good case for doing a bond ladder to fund core retirement income needs. Same with just doing a drawdown formula of some sort: 4% of beginning balance indexed for inflation is popular, but all sorts of systems involving rolling averages make sense. (LOL, I just realized that Michael Kitces is either the author or interviewed in each of these articles. I frequently see him at conferences, too. I respect his opinions a lot but he’s not right 100% of the time, just as no solution is right for everyone.)

Personally, I like the idea of a bond ladder set up for about 15 years and invest the rest in a diversified passive portfolio where we rebalance annually and harvest money from the investment account to buy bond ladder rungs each year. But this means you have to meet with a professional investment advisor every year, sometimes twice a year. (One meeting would be for the annual rebalancing meeting, but more meetings are quite common: you might need to either integrate new funds – like an inheritance – or to withdraw/rebalance when you need money for a specific goal.) Managing that $600,000 is not a “set it and forget it” project. You need to have yet another trustworthy professional with whom you have a relationship, alongside your doctor and your dentist.

Not everyone gets to have someone knowledgeable and trustworthy on their team. For those people, the ones who don’t want to manage their money or hire someone to do it, for the ones who have sharks circling them trying to separate them from their money, and, yes, for the fools who would soon be parted from their money, buying a single premium immediate annuity – buying themselves a pension with at least some of their kitty – makes a lot of sense.

But the insurance salesmen don’t make money selling SPIAs. They make money selling deferred “variable” or “fixed” or “equity-indexed” annuities or life insurance to people who don’t actually NEED life insurance. And people fall for these pitches because they’re finely tuned to their human fears and wishes.

I have two more books to read on annuities. I chose the hardest one first, and read through the FINRA brochures about them, too. It’s starting to come together in my head when someone should choose a SPIA, but I really really can’t ever come up with a good reason to choose a deferred variable annuity with a death benefit. If you want a death benefit then don’t buy an annuity, go explicitly buy life insurance, it’s way cheaper!

Upcoming blog entries: how to get OUT of a variable annuity you bought (the most common question people have after discovering how badly they actually suck). (Short answer: it’s terribly punishing to try. You’re sort of stuck with it. The best thing you can do is suck it dry with as high a guaranteed living benefit as you can get out of it.)

Next in the series: “How To Not Get Ripped Off When Buying An Annuity”.