Insights into the annuity puzzle

Why do most people love pensions but hate annuities? That, in essence, is the “annuitization puzzle” spoken of by Franco Modigliani, a Nobel laureate in economics. He noted that other than in the form of group insurance through pension systems, annuity contracts are extremely rare. Twenty-five years later, the puzzle remains. Rational choice theory predicts that at the onset of retirement, households will find annuities attractive because they address the risk of outliving one's income. But relatively few people choose to annuitize a substantial portion of their wealth. Adding some behavioral factors only deepens the puzzle, since annuities (in this case, immediate-payout annuities, not variable annuities used as an investment vehicle in the accumulation phase) have the potential to solve some of the complex problems with which individuals struggle, such as when to retire and how much can be spent each year in retirement, and thus might be expected to be attractive for that reason. There is now substantial literature on the behavioral economics of retirement saving, which has stressed that both behavioral and institutional factors play an important role in determining saving accumulations. Some of these factors also are important in explaining why there seems to be so little demand for products that annuitize wealth at retirement.

LACK OF SAVINGS

One simple reason is that many people simply have not saved up enough to make buying an annuity a viable option. The sizable portion of households with little or no wealth at retirement is, in essence, completely annuitized since their only source of income is Social Security. For this segment of society, there is no annuity puzzle, although there is probably a savings puzzle. But even among households that do accumulate enough in their retirement accounts to make an annuity feasible, one is rarely chosen. Is the low annuitization rate a reflection of underlying preferences or features of the choice environment? This question has important practical implications, because in most retirement plans, when an employee stops working and is interested in investing part or all of the balance in an annuity, he or she has to shop actively. Remember, few defined-contribution plans offer annuities. Owners of individual retirement accounts are in the same boat; they have to look actively for annuity products if they want to ensure lifetime income. Researchers have looked at these situations to see whether retirees become more interested in annuity products when that choice is easy to select. Some defined-benefit pension plans require retirees to make an active choice between a lifetime-income option — which we will loosely refer to as an “annuity” — and a lump-sum distribution. The plan sponsor offers the annuity directly and chooses the specific form it will take, saving the employee the possibly overwhelming task of selecting an annuity from the many options offered in the market. After analyzing three studies, and additional research we've undertaken, we found that when an annuity is a readily available option, many participants who have nontrivial account balances choose it. This finding suggests that the low annuitization rates in the case of defined-contribution plans might not reflect retirees' underlying assumptions about lifetime-income products but rather a mix of other institutional factors regarding the availability of annuities within their existing retirement plan. However, these institutional factors are not the entire story. We know from many studies in psychology that minor differences in wording can create large differences in behavior. For example, describing beef as “75% lean” sounds healthier than presenting it as “25% fat.” Similarly, an income replacement ratio of 70% at retirement seems much more palatable than a spending reduction of 30%. Framing issues can be particularly powerful in the case of annuitization because the concept is complicated and most people have not thought very much about the question before they approach retirement age. This also is not a domain where there can be much learning from experience. You only reach age 65 once per lifetime. One recent study illustrated the potential importance of framing in the context of annuities. The authors conducted an Internet survey of adults at least 50 in which subjects were asked to rate the attractiveness of an annuity in one of two conditions: a “consumption frame” in which the annuity was described as providing $650 of monthly spending for life, or an “investment frame” in which the annuity offered a guaranteed monthly return of $650 for life. In both conditions, there was no residual income or wealth after death. Consistent with the authors' predictions, the percentage of subjects choosing to annuitize was just 21% in the investment frame, versus 70% in the consumption frame.

LOSS AVERSION

Another issue is how the product is presented. While economists tend naturally to think about annuitization as a risk-reducing strategy, many consumers feel that they are taking a considerable sum of money and putting it at risk — the risk being that the consumer will die young, making the purchase a bad deal. The phenomenon of loss aversion also comes into play here. As Daniel Kahneman, a Nobel laureate in psychology, and others have found, losses hurt about twice as much as equivalent gains give pleasure, both in uncertain contexts and cases without uncertainty. Thus, a key issue is whether a consumer considers the purchase of an annuity as a cost, and thus a loss, or as an “investment.” Like many other economists who have studied this problem, we conclude that many households would benefit by increasing the share of their retirement wealth that is annuitized. While at this point in the history of defined-contribution plans, it is only households in the top half of the income distribution that have enough financial resources to be able to purchase a meaningful annuity, this is likely to change over time, both because the plans will have been in existence longer and because plan sponsors have learned how to encourage employees to increase their savings rates. If these arguments for encouraging more annuitization are found persuasive, it becomes interesting to ask whether there are steps the government might take to facilitate, though not compel, more annuitization. We think reforms in two general areas are worth considering: modifications to the Social Security system and changes in the regulations of defined-contribution plans.

END THE CONFUSION

The changes to Social Security would be aimed at increasing the demand for annuitized assets. The easiest way to increase the amount of annuity income that families have is to delay the age at which people start claiming Social Security benefits. And a good place to start would be to end the use of the confusing term “full retirement age,” currently 65 or 66, which may be inadvertently influencing participants' decisions about when to retire. Other policy changes should involve increasing the supply of easy-to-find annuity options for those of retirement age with 401(k) and other defined-contribution plans. The goal should be to emulate the progress that has taken place in the design of plans for the accumulation phase: specifically, the widespread adoption of “automatic” features, including automatic enrollment, automatic escalation and default investment strategies such as target date funds that re-balance a portfolio for a decreasing level of riskiness as the participant ages. Clearly, the notion that consumers simply are not interested in annuities is false. Social Security remains a wildly popular federal program, and those workers who still have defined-benefit pension plans typically choose to retain the annuity, rather than switching to a lump-sum distribution. Today, annuities often are considered a gamble instead of being viewed as a risk-reducing strategy. Perhaps additional research can help change that perception. This article was adapted from “Annuitization Puzzles” (investmentnews.com/puzzle), a working paper co-written by Shlomo Benartzi, a professor at the UCLA Anderson School of Management and chief behavioral economist at the Allianz Global Investors' Center for Behavioral Finance.