The Annuitization Puzzle

One of my favorite memories of visiting my grandparents over the holidays was working on jigsaw puzzles. These were generally large, complicated affairs — whose construction was spread over days, as various family members would stop by to work on a section, to build on a border, or sometimes contribute a single piece they would spot as they drifted by on their way to another activity.

Perched in a prominent place throughout would be the puzzle lid with that all-important picture of what we were working toward to help keep all those individual (and sometimes fleeting) efforts in the proper perspective — making it possible, for example, to differentiate the blue of what would appear to be sky from what would turn out to be an important but obscure section of mountain stream.

In retirement plans, one of the more intransigent concerns for policy makers, providers, advisors and plan sponsors alike is what has been called the “annuity puzzle” — the reluctance of American workers to embrace annuities as a distribution option for their retirement savings. What economists call “rational choice theory” suggests that at the onset of retirement, individuals will be drawn to annuities because they provide a steady stream of income and address the risk of outliving their income. And yet, given a choice, the vast majority aren’t.

Over the years, a number of explanations have been put forth to explain this reluctance: the fear of losing control of finances; a desire to leave something to heirs; discomfort with entrusting so much to a single insurer; concern about fees; the difficulty of understanding a complex financial product; or simple risk aversion. All have been studied, acknowledged and, in many cases, addressed, both in education and in product design, with little impact on take-up rates.

Yet today the annuity “puzzle” remains largely unsolved. And, amid growing concerns about workers outliving their retirement savings, a key question — both as a matter of national retirement policy and understanding the potential role of plan design and education in influencing individual decision-making — is how many retiring workers actually choose to take a stream of lifetime income, versus opting for a lump sum.

As outlined in a new EBRI Issue Brief, the evidence on annuitization in workplace pension plans has been mixed. But the EBRI report provides an important new perspective to a 2011 paper, “Annuity Puzzles,” by Shlomo Benartzi, Alessandro Previtero, and Richard H. Thaler. In that paper, the authors analyzed 112 different DB plans provided by a large plan administrator and, focusing on those who retired between ages 50 and 75 with at least five years of job tenure and a minimum account balance of $5,000, and noted that, “…virtually half of the participants (49 percent) selected annuities over the lump sum,” further observing that, “When an annuity is a readily available option, many participants who have non-trivial account balances choose it.”

The study’s authors went on state that, “The notion that consumers are simply not interested in annuities is clearly false,” explaining that “… the common view that there is little demand for annuities even in defined benefit plans is largely driven by looking at the overall population of participants, including young and terminated employees and others with small account balances who are either required to take a lump-sum distribution or simply decide to take the money.”

Mixed Behaviors?

In essence, Benartzi’s “Annuity Puzzles” study says that much of the existing research draws inaccurate conclusions by mixing the behaviors of younger workers with smaller balances with those who might, based on their age and financial status, be expected to choose an annuity. However, as noted in the EBRI Issue Brief, that study failed to take into consideration what has long been known to be a key element in retirement plan behaviors: retirement plan design. In effect, the Benartzi study blends the behaviors of participants who have the ability to choose an annuity with those who have either no choice, or one restricted by their plan.

What kind of difference might this make? Well, taking into account the same types of filter on tenure, balance and age employed by the Benartzi study, as well as a series of plan design restrictions, EBRI found that for traditional defined benefit plans that imposed no restriction on doing so, fewer than a third of those with balances greater than $25,000 opted for an annuity, as did only about one in five whose balances were between $10,000 and $25,000. (Participants in cash balance plans were even less likely to choose annuities than those in “traditional” final-average-pay defined benefit plans.) Those with balances between $5,000 and $10,000 were even less likely to do so. (The Benartzi study filtered only on individuals with a balance greater than $5,000.) In sum, even filtering to focus on the behaviors of individuals seen as most likely to choose an annuity distribution, we found that, given an unfettered choice, the vast majority do not.

Ultimately, the EBRI analysis shows that, to a large extent, plan design drives annuitization decisions. 1 We know that plan design changes have been successful in influencing participant behavior in DC plans via auto-enrollment and auto-escalation, and the new EBRI study suggests that plan design can also play a critical role in influencing distribution choices.

It also shows the importance of taking into account the whole picture when you’re trying to solve a puzzle.


  1. The EBRI analysis also found that annuitization rates increase steadily with account balance for older workers (but not for younger workers), and that annuitization rates also increase with tenure. See “Annuity and Lump-Sum Decisions in Defined Benefit Plans: The Role of Plan Rules.”

Add Your Comments

One Comment

  1. Elmer Rich Iii
    Posted February 26, 2013 at 12:53 pm | Permalink

    The fact is that there is no money for the Boomers to retire – anywhere. No one has enough money for this. Not companies, governments or Boomers themselves.

    Since retirement costs are actuarially certain and predictable some type of pre-funding will have to occur for future generations — globally. Extended, healthier life spans are a demographic sea-change that all societies and individuals are going to have to fund.

    Annuitization is the only method of this pre-funding. However, it is likely not commercially viable for the population as a whole so will have to be a government utility.

    We are marketers and pro communicators and we all need to accept that annuities cannot be sold to the vast numbers that need them. Our brains just don’t “get it.”

    We challenge this view:

    “Ultimately, the EBRI analysis shows that, to a large extent, plan design drives annuitization decisions. 1 We know that plan design changes have been successful in influencing participant behavior in DC plans via auto-enrollment and auto-escalation, and the new EBRI study suggests that plan design can also play a critical role in influencing distribution choices.”

    In fact, we have studied the behavioral approaches in plans and other applications and simply – they don’t work and often cause “blowback.” The British research on this is much better than the US. In the US there is a unwarranted acceptance of beh econ ideas.

    The idea that “nudges” will effect serious problems is appealing, and has made a great marketing/sales campaign – but it is very naive.

Post a Comment

Your email is never published nor shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Send this to a friend