Lowell Aronoff is the CEO of CANNEX, one of the largest providers of annuity pricing data to the financial services industry and an expert on the theoretical value of a life annuity. In a recent conversation, I asked him what I ask a lot of annuitization scholars: Does your mother own an annuity?
“I was never able to convince her to buy an annuity, which is very embarrassing,” says Aronoff. “A lot of it had to do with things like ‘what happens if I die tomorrow, what happens if the insurance company disappears, is it a good investment?’ All of which are essentially irrelevant objections but were primary in her mind.”
My mother doesn’t own an annuity, either. I’ve traveled to numerous conferences explaining to advisors the value of incorporating annuitization into a retirement income plan in order to hedge the risk that clients will outlive their assets in old age. But most people would rather not lay down a large sum of money in exchange for a pretty modest income. So what’s the problem?
Economists are often annoyed when normal people don’t act in a manner consistent with their models. Annuities provide a perfect example of a product that every economically rational retiree should own and few actual retirees buy. For decades, the infrequent use of annuities in the United States has been considered a puzzle that many economists have gone to great lengths to explain through even more complex models. The likely explanation is that most people aren’t economists.
What’s so great about annuities? To an economist, life is a game with two goals—spend about the same amount each year (because spending a lot this year and a little next year makes us less happy), and spend the highest amount each year. So the best spending plan will be one that is high and smooth. If you want to leave a bequest or prepare for an emergency, buy a product or set money aside for each non-spending goal.
Annuities fit the bill by both smoothing spending and maximizing the amount we can spend each year. Mortality credits provide the spending boost while the guaranteed lifetime income ensures smooth spending. No other financial product provides the same level of expected lifetime happiness in retirement.
One of the most important benefits of an annuity is its protection against running out of money in retirement. Economists see the big drop in spending caused by outliving one’s assets (superannuating) as the equivalent of rolling snake-eyes—you want to do everything you can to avoid it. How do we efficiently protect ourselves against an unexpected significant loss? We buy insurance. Annuities insure against this risk.
If the risk is so significant, why aren’t annuities more popular? Some annuities are very popular. People love their Social Security, which is essentially the mandatory purchase of an inflation-protected annuity. Questions inserted into the Health and Retirement Study in 2008 asked respondents how much they would accept in a lump sum to get them to give up $500 in monthly Social Security payments. The answers revealed two very important insights into how people value annuities.
First, you’d have to pay them a ton of money to get them to give up Social Security income. The average was literally over $250,000, or an amount that would be appropriate if your expected longevity was over 130 years with a 0% discount rate. Second, nobody has any idea how to value a stream of income. The majority of college-educated respondents weren’t able to come even close to an actuarially fair estimate.
Our unwillingness to sell something for a fair price once we own it is known as the endowment effect. That’s why politicians become ex-politicians when they try to cut Social Security benefits. It’s also why employee unions are so resistant to the mere suggestion that the pensions need to be readjusted. Experiments show that even when a subject is given something that they might not have even wanted (like a coffee mug) they suddenly place a much higher value on the object.
So even though pensions and Social Security are both annuities, people love them. But they don’t love buying annuities. That’s because they have to trade something they have (a lump sum of money) for something that’s a lot more abstract. It shouldn’t be surprising that a 2005 study by Professors Keith Bender and Natalia Jivan found that retirees with annuitized income through defined benefit plans were significantly more satisfied than retirees with non-annuitized retirement assets. The problem is that people aren’t willing to give up their 401(k) stash to get one because of the endowment effect.
One way to solve this part of the puzzle is to get people to reframe their retirement assets as income instead of wealth. The Department of Labor is considering a method that plan sponsors will use to illustrate retirement assets in terms of lifetime income instead of a lump sum. Since our research shows that most workers have no idea how to translate a lump sum into an actuarially fair income stream, it gives them a much more realistic idea of how well they can live in retirement (and how much more they need to save). It also changes workers’ expectations about what they’ll do with their qualified savings when they retire.
Switzerland already defaults employees’ retirement plan assets into a life annuity. Since it’s a default, it isn’t mandatory. A default can be a powerful tool because employees see it as a stamp of approval—known as the endorsement effect. Because Swiss workers generally expect that retirement assets will turn into retirement income, and because annuitization represents the path of least resistance, about 80% of Swiss workers annuitize. Framing assets as income and defaulting at least a portion of retirement assets into annuities could solve the puzzle without taking away freedom of choice.
Without the ability to translate a lump sum into income, sitting on a large defined contribution balance at retirement creates a powerful behavioral challenge. Buying an annuity means giving up a big number. And what do you get in return? You might live 30 more years. Or you might live two months. Many people focus on the worst case scenario and do whatever they can to avoid it. They don’t want to make a decision they’ll regret, and most of us suffer from regret aversion.
Aronoff notes that “people have a hard time separating their need for lifetime income from their investment.” Employees are used to seeing their retirement savings as an investment whose purpose is accumulation. The balance of this investment becomes a reference point—an amount that we start from when judging decumulation strategies. That makes annuitization seem more risky. According to Aronoff, retirees feel that “if I get hit by a bus, I lose all my money.”
Jason Hull, a financial planner in Crowley, Texas, agrees that the primary barrier to annuitization is fear of regret. Rather than focusing on the fear of running out of money in the distant future, retirees seem to care more about the possibility that they won’t get their money’s worth. This makes no sense to Hull. “As they’re floating up to heaven, they’re going to say ‘you know, I should have bought that annuity’? We are projecting a state that is never going to happen. Nobody is ever going to look down and have an out of body experience and say ‘stupid, you never should have bought that annuity.’”
That’s little consolation to financial service professionals who are trying to sell annuities to a reluctant consumer. Their choice is either to educate consumers to think like an economist or to give people what they want. Barry Mulholland, an assistant professor at Texas Tech and former insurance agent, agrees that the right product sometimes needs to appeal to the rational and the behavioral side of the customer. Characteristics such as a period certain guarantee or a cash refund, which promise to return part of the initial investment if the client dies early, are particularly appealing.
“It is about loss aversion,” says Mulholland. “If you have clients that indicate they are loss averse, then adding a period certain to a life annuity makes them feel much better about handing over $100,000 to an insurance company. While it is pointed out to clients that this will lower the monthly income due to the added cost of insuring the income stream, a decrease of $25, $50 or even $100 per month seems insignificant compared to the possibility of ‘losing’ $100,000 next month if they die in a car accident.”
Of course, the reason why annuities work is because half of the annuitants will die before their average life expectancy. Pulling money out of the pot to pay their heirs means less money is available to pay those remaining annuitants who live longer. If the retirement income game really is about spending the most money every year, a cash refund makes you a loser. Or does it?
If it reduces the anxiety of a retiree, isn’t that part of getting the most out of the retirement income game? And annuitization, even imperfect annuitization, may be better than the alternative. Aronoff sees that a number of financial firms “are pushing cash refund annuities because they can tell the client that they are guaranteed never to lose money.” Is this good because it gets people to buy an annuity or is it bad because it’s expensive? “I think it’s both” says Aronoff. “On a fundamental level, it works.”