The Annuity Puzzle by Richard Thaler

“Annuities and the Puzzle of Income”

Can anyone point out several omissions and/or fallacies in this article by the famous professor Richard Thaler? Just a sentence or two.

Prize to be emailed. (Hint: biggest bubble today)

The Annuity Puzzle


IMAGINE a set of 65-year-old identical twins who plan to retire this summer after long careers. We’ll call them Dave and Ron. They have worked for different employers and have accumulated retirement benefits worth the same amount in dollars, but the benefits won’t be paid out the same way.

Dave can count on a traditional pension, paying $4,000 a month for the rest of his life. Ron, on the other hand, will receive his benefits in a lump sum that he must manage himself. Ron has a lot of choices, but all have consequences. For example, he could put the money into a conservative bond portfolio and by spending the interest and drawing down the principal he could also spend $4,000 a month. If Ron does that, though, he can expect to run out of money sometime around the age of 85, which the actuarial tables tell him he has a 30 percent chance of reaching. Or he could draw down only $3,000 a month. He wouldn’t have as much to live on each month, but his money should last until he reached 100.

Who is likely to be happier right now? Dave or Ron?

If this question seems a no-brainer, welcome to the club. Nearly everyone seems to prefer the certainty of Dave’s pension to Ron’s complex options.

But here’s the rub: Although people like Dave who have them tend to love them, old-fashioned “defined benefit” pensions are a vanishing breed. On the other hand, people like Ron — with defined-contribution plans like 401(k)s — can transform their uncertainty into a guaranteed monthly income stream that mirrors the payouts of a traditional pension plan. They can do so by buying an annuity — but when offered the chance, nearly everyone declines.

Economists call this the “annuity puzzle.” Using standard assumptions, economists have shown that buyers of annuities are assured more annual income for the rest of their lives, compared with people who self-manage their portfolios. One reason is that those who buy annuities and die early end up subsidizing those who die later.

So, why don’t more people buy annuities with their 401(k) dollars?

Here’s one part of the answer: Some people think that buying an annuity is in some way a bad deal for their heirs. But that need not be true. First of all, a retiree can decide to set aside some portion of a retirement nest egg for bequests, either immediately or at a later date. Second, if a retiree chooses to manage his or her own money, the heirs may face the following possibilities: Either they get financially “lucky” and the parent dies young, leaving a bequest, or they are financially “unlucky,” meaning that the parent lives a long life, and the heirs take on the burden of support. If you have aging parents, you might ask yourself how much you’d be willing to pay to insure that you will never have to figure out how to explain to your spouse, or whomever you may be living with, that your mother is moving in.

There are other explanations for the unpopularity of annuities, but I think two are especially important. The first is that buying one can be scary and complicated. Workers have become accustomed to having their employers narrow their set of choices to a manageable few, whether in their 401(k) plans or in their choice of health and life insurance providers. By contrast, very few 401(k)’s offer a specific annuity option that has been blessed by the company’s human resources department. Shopping for an annuity with hundreds of thousands of dollars at stake can be daunting, even for an economist.

The second problem is more psychological. Rather than viewing an annuity as providing insurance in the event that one lives past 85 or 90, most people seem to consider buying an annuity as a gamble, in which one has to live a certain number of years just to break even. But, as the example of Dave and Ron shows, it’s is the decision to self-manage your retirement wealth that is the risky one.

The most complex and unknowable part of that risk is in predicting how long you will live. Even if there are no medical advances in the coming years, according to the Social Security Administration, a man turning 65 now has almost a 20 percent chance of living to 90, and a woman at this age has nearly a one-third chance. This means that a husband who retires when his wife is 65 ought to include in his plans a one-third chance that his wife will live for 25 more years. (A “joint and survivor” annuity that pays until both members of a couple die is the only way I know for those who are not wealthy to confidently solve this problem.)

An annuity can also help people with another important decision: when to retire. It’s hard to have any idea of how much money is enough to finance an appropriate lifestyle in retirement. But if a lump sum is translated into a monthly income, it’s much easier to determine whether you have enough put away to afford to stop working. If you decide, for example, that you can get by on 70 percent of preretirement income, you can just keep working until you have accrued that level of benefits.

IN the absence of annuities, there is reason to worry that many workers are having trouble with this decision. Over the last 60 years, the Bureau of Labor Statistics reports that the average age at which Americans retire has trended downward by more than five years, from 66.9 to 61.6. Of course, there is nothing wrong with choosing to retire a bit earlier, but over the same period, live expectancy has risen by four years and will likely continue to climb, meaning that retirees have to fund at least an additional nine years of retirement. Those who manage their own retirement assets can only hope that they have saved enough.

Annuities may make some of these issues easier to solve, but few Americans actually choose to buy them. Whether the cause is a possibly rational fear of the viability of insurance companies, or misconceptions about whether annuities increase rather than decrease risk, the market hasn’t figured out how to sell these products successfully. Might there be a role for government? Tune in next time for some thoughts on that question.

Richard H. Thaler is a professor of economics and behavioral science at the Booth School of Business at the University of Chicago. He is also an academic adviser to the Allianz Global Investors Center for Behavioral Finance, a part of Allianz, which sells financial products including annuities. The company was not consulted for this column.

July 18 Update: To the Editor:

Re “Annuities and the Puzzle of Income” (Economic View, June 5), in which Richard H. Thaler described some advantages of buying annuities:

Annuities are superficially attractive, but they have important flaws the column did not discuss. People don’t tend to buy diversified sets of annuities, and because they are issued by inherently leveraged financial intermediaries, there can be material credit risks that are hard to assess over the 20- to- 40-year horizon of their payouts.

Annuity providers also incur sales charges and operating expenses, and need capital and a cushion of economic earnings. Those costs may not be incurred by someone making their own investments, or doing so through lower-cost vehicles.

Finally, inflation is an even more important consideration. Not discussing it is like playing “Hamlet” without the prince. In 30 years, the level of consumer prices in the United States might double or sextuple. The first would be painful, and the second ruinous, to someone who relied only on a fixed annuity.

Paul J. Isaac

Manhattan, June 5

The writer is the portfolio manager at Arbiter Partners, a hedge fund.

Editor:A devastating critique of annuities: Credit risk, high costs, and inflation risk.  Right now might be the WORST TIME in the past 100 years to buy an annuity. Be careful out there in your search for returns.

For those who gave a good answer, claim your prize. This link will disappear in a day and the prize will go into the Value Vault (under spins). Remember just email me at with VALUE VAULT in the subject line (don’t write anything else). The value vault has over 200 books, videos, case studies and just plain great material for learning about investing and business analysis. Reward given ($$$) to someone who can find a better resource on the web.

12 responses to “The Annuity Puzzle by Richard Thaler

  1. The system in my country is a combination of DC and DB (I’m not an expert in the US system so..), and companies outsource their retirement plans to insurers who must ‘normally’ deliver 3,25% / 3,75%. Given their investment policy (read : you don’t get these type of returns in ‘safe’ govies anymore) insurers have difficulties delivering these returns and companies will have to jump in.

  2. Hi John,
    The author assumes that the pension provider will always be there and be solvent!
    He mentions an annuity inside of a 401(k), which is already shealtered from taxes.
    He seems to assume that folks are not capable of finding a professional that could manage the right mix of fixed income/equity to provide cash as well as some growth.
    If you would like to leave an inheritance for family members, you’d have to buy some sort of additional policy with the annuity option.

    (I thought that (one of) the biggest bubbles right now was student loans… and I don’t need one of those.)


  3. Agree with the post above about his assumption that the pension provider will always be solvent. Pension fund under funding in the US has hit record highs this year at $354.7 billion – an increase of over $100 billion from 2010 and surpassing the 2008 record of $308.4 billion. Yikes!

    There is no mention whatsoever of how long your money is tied up in an annuity. What if you need a lump sum for an emergency? Too bad – a hefty penalty awaits.

    There is no mention of the fees involved in the annuity. The S&P 500 Index has a track record dating back to January, 1926 and here are 950 historical 7-year rolling, monthly return periods since 1926. With a portfolio comprised of 60% S&P 500 Index and 40% 5-year Treasury bonds, out of the 950 7-year periods, there are only 3 times when the portfolio’s return was less than 0%. So why are we paying to insure that?!

    Furthermore, the incentives are misaligned. Typically, annuities carry the highest commission for a salesman. They can easily sell the “guarantee” and have a nice payday for themselves. There is also no mention of opportunity cost. Parking that nest egg and tying it up at fixed rates for all that time could be less than desirable. I remember clients piling into fixed annuities in early 09 because they were scared. Salesmen preyed on that fear. Its a same because they missed nearly 100% returns just by owning an index.

    • an additional note on Pension obligations and the shortfalls:

      “With three California cities electing to file for bankruptcy in the past month, in large part due to underfunded pension obligations, the last thing California needs right now is more bad pension news. Which is unfortunate, because the California Public Employees’ Retirement System, the nation’s biggest public pension fund, delivered a jolt of bad news this week when it reported a paltry 1% return on its investments in the fiscal year ended June 30.

      “The last twelve months were a challenging period for all investors as the ongoing European debt crisis and slowing global economic growth increased market volatility and reduced equity returns,” said Joe Dear, CalPERS chief investment officer, in a statement “It’s a clear reminder that we must remain focused on performance, risk and internal controls in today’s financial environment.”

      CalPERS’ 1 percent return is well below the fund’s discount rate of 7.5%, a long-term target that CalPERS lowered recently as it re-evaluated its economic assumptions in the current investing environment. Other pension plans have made similar adjustments recently. The rate is significant in that it determines the amount of money such funds need to invest now in order to meet future pension obligation needs.”

      I’m curious about the second order effects of this much money falling this far behind. Will they cave and pour into risky assets to try and gain some ground? What is the effect of that much money moving into stocks? If that were to happen, and the markets fell significantly, how would something as big as Calpers deal with market losses? Would they compound the problem by de-risking their portfolios further pushing down stock prices?

  4. The professor fails to compare the value of the annuity vs the price you have to pay. I would assume if one wishes to insurance against longevity inflation would be a major concern. A quick Google search shows a payout of 4-5% for inflation adjusted annuities, which seems rather low. If one wanted $50k/year they’d need to have saved up $2-2.5M, which seems a bit unlikely for a large majority.

    Another concern would be the likelihood of any company remaining financially stable for several decades in the event you live for a long time.

    • The article fails to mention:
      1) Annuities do not guarantee a real return. There is inflation risk.
      2) There is no guarantee that the annuity payer will be able to keep their promises.
      3) The cost of an annuity may make it uneconomic.

      FWIW, I recently reviewed a friend’s annuity which I believe is typical. She is around 65, paid a lump sum today, and would start receiving payments when she turned 75. The math (which was obscured by Nationwide) came down to this: She would effectively earn a 0% return if she lived to 86 and would need to live to 109 in order to get a 5% IRR. And she forfeits the possibility of passing money to her heirs!

      She purchased this annuity a year ago, when interest rates were higher! I’m sure the terms are worse today.

      • Excellent Post. You nailed it! I will post an addendum as to what an astute investor said about the NY Times article.

        But your answer was even better.

        Others who replied did well. You gift was emailed.

    • Dear Hunter: B (tough grader!) you missed the inflation danger. See updated post above.

      Yes, student debt is a bubble (don’t forget government debt!)

    • You got an A-. Good answer. Prize to be emailed.

  5. A Reader who shall remain Anon. couldn’t post so here it is:

    Sorry to email this to you, but I have not been able to post. Something with
    my browser or firewall. Anyway, I did want to comment on Thaler’s “The
    Annuity Puzzle”. I highly respect him and his body of work, but agree that
    this is not as big a puzzle as he makes it out to be.

    Thaler wants to frame the issue purely as “mental accounting” whereby those
    getting a lump sum view the portfolio value whereas those receiving an
    annuity stream view the monthly or annual amount. Those differences create
    differences perspectives of value.

    However, a bigger issue is that of pricing of the portfolio and future
    purchasing value. Today, any individual going out to buy an annuity is
    robbed by currently low interest rates and future inflation. Taking the lump
    sum and accepting a less certain cash flow might be the wiser choice. One
    has the option to invest in ways that will hold value against future
    inflation which the annuity doesn’t.

    Even ignoring the issue of inflation, you cannot compare defined benefit
    pensions to a lump sum. It’s apples to oranges. Thaler himself describes the
    difference. The lump summer cannot buy equivalent income. Either the retiree
    takes a 25% haircut in return for the security of lifetime income or accepts
    a 30% probability of outliving the savings. Thaler is unclear whether he
    believes that the retiree could purchase an annuity that is equivalent to
    the pension, but in the real world we know that the retiree cannot.

    DBPs typically determine annual payout based on ending salary and years
    employed. Some even have built in inflation adjustments like social
    security. In many ways DBPs act like Ponzi schemes requiring additional
    contributions from the employer or working employees to remain solvent. Very
    rarely are they self-sustaining. Meanwhile, annuities are priced off current
    interest rates and relate to the probable life span of the individual. As an
    insurance product annuities also contain pretty significant fees because
    insurance companies as opposed to pension plans are for-profit enterprises.

    There exists a very strong argument that the decision to forego the comfort
    of a fixed income stream in retirement is rational. By accepting the lump
    sum, one might gain an edge against inflation — a higher probability of
    avoiding the foreseeable decline in purchasing power. reduced income
    received to begin because of the fees and low interest rate peg of the
    annuity stream.

    Finally, business has a very great incentive to abandon DBPs. They are
    hugely expensive. The uncertainty of retirement income is taken off the
    pensioner’s shoulders and onto the employer. However, the uncertainty and
    burden still exists and can be quantified by reading the footnotes to the
    financial statements of any old industrial company.

    Just saw the Leon Cooperman post pass by. I love the last line about the
    note on his monitor. It’s really comforting to know that even the guys who
    have been doing this for forty years need to keep reminding themselves of
    the basics and guarding against making the foreseeable mistakes. Nothing is
    worse than making the same error twice!

    Thanks for all the great thoughts and information. I enjoy it a lot.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.