Lump Sum or Annuity?

By James Kwak

Usually the New York Times gives reasonably good financial advice—or, at least it avoids giving really bad advice. Today, however, Paul Sullivan’s column borders on the latter. The question is whether to take a pension payout as a lump sum or as an annuity (a guaranteed, fixed amount per year until you die).

Sullivan’s column isn’t all bad. He talks about the importance of being able to manage your money and the need to be comfortable with risk if you take the lump sum. He also points out the annuity (in this case, based on what GM workers are being offered) isn’t indexed to inflation, which is an important consideration. And he doesn’t come down on one side or the other, although he says he would take the lump sum because, he says, “I would rather control the money myself.”

But there are some serious problems with his discussion. One is the way he discusses control as a good thing. Reams of empirical research have shown that, for the typical individual investor, control is a bad thing. Here’s just one fact: from 1991 through 2010, investors in stock mutual funds earned an annual return of 3.8 percent, while the S&P 500 earned an annual return of 9.1 percent. That’s because investors tend to buy high and sell low (based on past performance), and because they generate transaction fees through excessive trading.

This is a case where individuals are better off tying themselves to the mast, like Odysseus before the Sirens, and making it impossible for themselves to make dumb mistakes. Unfortunately, human beings suffer from optimism bias, so most people think they can do better than the market.

The bigger thing that’s missing, however, is a serious discussion of risk. For the purposes of comparison, start by assuming that the conversion between lump sum and annuity is actuarially fair. (If it isn’t—if, for example, the expected value of the lump sum is much higher than the expected value of the annuity—then that will be a major factor. But that’s not something that the typical individual is going to be able to figure out on her own. And more on that later) In that case, you are being offered two things with the same expected value. But one has much higher risk than the other: your average annual retirement income is highly volatile with the lump sum, while it’s known with certainty with the annuity. Basic finance dictates that between two things with the same expected value, the one with lower risk is better.

Now, some people think that the “risk” is that you will die young and you will lose on the annuity; the lump sum, by contrast, exists today, so it has no risk. But that’s the wrong way to think about it. When it comes to retirement income, the most important priority is making sure you don’t run out of it. The thing you really care about is not cash in hand today, but either your average income in retirement or your minimum income in retirement. An annuity guarantees you won’t run out of money; a lump sum doesn’t, and that’s a lot of risk you’re taking on.

So it only makes sense to take the lump sum if the expected value of the annuity is lower than the expected value of the lump sum—and the more risk-averse you are, the bigger the gap has to be. One reason the EV of the annuity could be low is if you know you are likely to die early. Leaving that aside, however, it’s hard to see what rational basis any retiree could have for thinking that the EV of the lump sum is much higher than the EV of the annuity.

If you want to do the calculations, you can’t assume a 10% return, or an 8% return, or even a 4% return on investing your lump sum. You have to assume a risk-free return (since the annuity is risk-free), meaning the yield on Treasuries of appropriate maturity (ten years is probably about right as an average for a retiree in her sixties, who is likely to live twenty more years), meaning about 1.5%. At that yield, I suspect that most GM retirees could not convert their lump sums into the annuities that they are being offered.

Besides, we know what happens when people take lump sums at retirement: just look at West Virginia (although the problems there went beyond just lump sum payouts). It isn’t pretty.

20 responses to “Lump Sum or Annuity?

  1. Reblogged this on Underwriting Solutions LLC and commented:
    A nice article on the topic.

  2. Speaking of risk, you also have to consider the risk of the pension fund going belly up. Even though there is a government agency that “guarantees” private company pensions, you usually don’t get your full pension amount if the pension fund fails. No idea what the risk is for GM’s pension, but it’s something you need to consider.

  3. Purchasing a life annuity when rates on offer are at an all-time low is rational if you believe that rates will stay low for the duration of the annuity. Some rational people would take the lump sum and invest short term hoping that annuity rates would improve before making a long term committment. Monthly annuity payouts are at least 1/3 larger if the annuity pays 6% rather than 2%. Past volatility of annunity rates may lead a rational investor to wait several years for rates to increase before making a one-time committment to an annuity.

  4. maynardGkeynes

    My distinct impression is that Prudential has assumed borderline insane corporate bond interest, in which case it makes sense to take the annuity. Unfortunately, if Prudential is insane on this score, that is evidence that its entire business model is now insane, especially given the extent to which the Fed seems out to savage the long term bond market, on which insurers rely. In that case the annuitant assumes the risk of Prudential going under, which is not exactly a great position to be in when you are 85. Sports fans, you make the call.

  5. Inflation risk is a consideration on anything as long as an annuity as even average inflation will lower the value by half. Solvency risk is also. Liquidity risk is another. How constant are your expenses going to be? Will inflation and infirmities increase them? Will age and incapacity decrease them? Will unforeseen events change the amount and timing of preferred distributions? Annuities can provide an income but that isn’t quite the same as not running out of money.

    They already have a good and sizable indexed annuity in Social Security. How fixed an income do they really want? Do they want to consider consuming more up front to delay taking Social Security? Most are probably not informed or equipped enough to handle these decisions and probably are better off with an annuity but there isn’t one answer.

  6. Moses Herzog

    I’m curious if Professor Kwak is familiar with NYU’s Systemic risk list??? Not as funny as Letterman’s Top 10, but perhaps more useful to retaining your humor in your golden years. If Mr. Kwak will notice, at least 3 of the top 10, their main line of business is the insurance business. And sometimes AIG is quite confusing as they seem to have some type of long-term identity crisis, behaving like an investment bank, but some might call them the 4th insurance company on this list. A lot of insurance companies basically are involved in kickbacks for employers to accept a limited list of crappy choices of funds for their employees’ 401k plans. Which insurance companies do this and how many do it??? The baselinescenario reader must decipher on their own, I can only say “where there is one cockroach there are……”
    See the NYU Systemic Risk List here, just to the right numbered 1 through 10:

    http://vlab.stern.nyu.edu/welcome/risk/

    Here is a breakdown (basically paraphrasing original journalism by Stephane Fitch at Forbes) on what I think of your average 401k:

    http://grahambrokethemold.blogspot.com/2010/02/is-401k-really-good-for-you-or-just.html

    You can also read here about something economists often refer to as “the agency problem” or the “principal–agent problem”.

    http://www.auburn.edu/~johnspm/gloss/agency_problem

    One of the best academic papers I have ever read in my life, by Jennifer Taub, examines “the agency problem” as it relates to mutual funds thoroughly (the paper is slightly dated as to specifics but most of the general conditions of the mutual fund industry still hold very true to now):

    http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1066831

    Mr. Kwak says “Reams of empirical research have shown that, for the typical individual investor, control is a bad thing.” Yeah, really??? Ask the Bernie Madoff investors about that.

    I’ll take the lump sum, thanks anyway Bernie.

  7. ‘Yeah, really??? Ask the Bernie Madoff investors about that.’

    They gave up control to Madoff. Sort of bought an annuity from him.

  8. Tax consequences need to be considered too, as well as what the distributions (whether lump sum or over the years) will do to increase your Medicare premiums. I also prefer the lump sum, but paying a lot of tax up-front isn’t such an appealing situation.

  9. @bhans, yes, you took the words right out of my mouth. : )

    NO more Ten Year Averaging for lump sum distributions, so caveat emptor on the tax bite.

  10. “Control” can mean many things; you are reading it as “the ability to day-trade my account and chase hot performers.” Control for me means the ability to move my money into very low-cost index funds at, say, Vanguard.

    I think your interpretation of “risk” is similarly off. The counterparty risk associated with a private pension annuity is substantial. Look around! And as such, I don’t think the appropriate comparison is to risk-free Treasuries. Maybe forty years ago, but not today.

  11. Patty Craven

    Very good article, James. My father took out an annuity, and he died young (age 60). My sister and I inherited the benefit; we each have been receiving a small monthly sum for 35 years. During that time, the amount paid out to us only went down a few dollars one time.

  12. Not too long ago the company I worked for (very large global company with > $30B in annual revenue) closed the site where I worked putting >500 people on the street. Of those who were eligible for retirement 90% took the lump sum. I asked each of those that I knew – Why? Every answer was the same:
    1. Current low interest rates make the lump sum unusually large.
    2. Don’t trust the company or a new owner post-merger/takeover to continue to pay the annuity for 30+ years.
    Some of them will poorly manage the money I’m sure but are they wrong to think that at least they had a chance that way?

  13. A somewhat happy medium might be to subdivide your money into various pots (say, 5-6 pots), and layer each pot such that you annuitize one after another over a few decades, about every 5-8 years (this is an example only). You’ve locked in the first pot with a level and certain monthly payment, but the rest of your money is still growing until you annuitize the next one, etc. The money reserved for later is in effect your inflation hedge as it keeps growing till you tap it. This layering strategy seems to utilize both growth and certainty.

  14. * Credit risk. I’m always concerned about the credit risk of the pension fund. Given the underfunding of both government private pension plans and the shaky state of the PBGC, I’d be inclined to take the lump sum.

    That doesn’t mean I have to individually invest it, however. I could take that money to three different insurance companies and buy life annuities with COLAs–that would diversify my credit risk. Or I could self-annuitize for a fixed 15 year period while purchasing a deferred life annuity, which would cover. A proxy for that would be to defer social security payments, which go up the later that they are taken so that they are actuarially constant.

  15. Moses Herzog

    I’m wondering if the thought has ever occurred to the very intelligent (but sometimes very naive) Professor Kwak, that the greed of your average insurance company in Connecticut, is really not that far removed/differentiated from the greed of Wall Street. Judging from his Twitter, Mr. Kwak’s “basic assumption” about the safety of insurance companies’ annuities vs. lump sums, if “challenged”, is not likely to be “absorbed”.

    “@JacobMGerber Counterparty risk is an issue, but (a) I’d take Prudential over the average person’s investing skills); (b) there’s the PBGC.”

    May I remind Professor Kwak, although AIG was behaving in a much more risky fashion than many investment banks, they presented/represented themselves to the public as an insurance company. As far as Prudential Insurance company is concerned, well, I don’t know what could possibly go wrong there…..

    http://www.nytimes.com/1993/11/11/business/prudential-fraud-scandal-reaches-to-parent-company.html?_r=1

    http://articles.latimes.com/1996-10-07/business/fi-51492_1_audit-system

    http://www.bloomberg.com/news/2010-08-30/prudential-veterans-insurance-lawsuit-is-expanded-with-new-fraud-claims.html

    Now here is also an interesting one from Credit Slips blog, making reference to Prudential and MetLife paying VA life insurance. Now this is a “lump sum” type payment, but it’s not a lump sum you can receive payment of immediately, it’s a lump sum that “by definition” pays at death. That means in all that time that you are paying for the insurance the funds are held in the insurers’ general corporate account. I’m going to liberally lift an entire long paragraph here from Katie Porter’s writing at Credit Slips blog, and I pray Miss Porter doesn’t get angry for that:

    “Apparently many of the VA life insurance companies, including Met Life and PRUDENTIAL, do not give beneficiaries a check when the policy is payable as a lump sum. Instead, grieving family are mailed a checkbook and told that the payout was ready for use in an ‘convenient interest-bearing account.’ But here are the catches: 1) The money is not in an FDIC-0insured account, meaning these beneficiaries’ money could disappear if the insurers went under. This fact would be hard for consumers to discern given that the checks themselves bear the names of large banks like JPMorgan Chase. The money actually resides in the insurers’ general corporate account, earning investment income. 2) The VA was under the impression that the insurers were patriotic volunteers, earning no profit on the checkbook option. In fact, the insurers earned about a 5 percent return, while the beneficiaries received 1 percent interest. When the NY Times explained this, the VA spokesperson said ‘Maybe I didn’t ask enough questions.’ My former colleague at UNLV’s William S. Boyd School of Law, Jeff Stempel, put a finer point on it. ‘[T]his is a scheme to defraud by inducing the policyholder’s beneficiary to let the life insurance company retain assets they are not entitled to. It’s turning death claims into a profit center.’ “
    Miss Porter’s Credit Slips Post:

    http://www.creditslips.org/creditslips/2010/07/financial-risks-follow-to-the-grave.html

    And the New York Times/Bloomberg July 2010 breaking story here:

    http://www.nytimes.com/2010/07/29/business/29insure.html?_r=1

  16. Moses Herzog – thanks for your very knowledgeable critique.
    It is somewhat surprizing after all the financial shennanigans Mr. Kwak has written about that he has such a blind eye to the manipulations of insurance companies.
    If investors are naive about investing, which I would agree with,they are even more naive aboutreading their annuity statements and discerning what fees have been added.

  17. There is a risk I hardly ever see debated. And in these times, when there are a lot of quesitons about the stability to these organizations that are offering annuities, it would seem to be an important question. What is the risk that an annuity will not be paid in full because the company went our of business. Didn’t a lot or people lost a large portion of their annuities because the companies that they had them with went belly up?

  18. I will echo the previous posts re. the safety of the insurance companies. In these times I would really *want* to be able to trust the insurers, and whatever government regulation is behind them. I would like to hand off the problem and just cash in whatever annuity will come in. But I agree – in a world of negative risk free interest rates, of excessive leverage in the system, of widespread corporate greed (insurers are just as greedy as wall-street traders) protected (backed?) by legislation, I guess that my wife and I shall have to take the risk in our own hands. Diversify into a range of options – zero bonds, corporate paper, gold, and yes one or two policies issued by different companies (assuming they won’t merge, at some later stage). I don’t expect to beat the market, but I don’t want to be screwed either….

  19. I generally cannot advise saving for retirement. It may be the worst investment you can make: http://andreasmoser.wordpress.com/2012/08/05/most-important-investment-advice-3/

  20. Hey saving for retirement is really necessary, as when you get retired there is no income source and at that age there are more expenses related to hospitals, medical, kids education etc, so if you plan it right now it would be easy to survive in future.