Wednesday, May 9, 2012

Reichenstein and Meyer’s “Social Security Strategies: How to Optimize Retirement Benefits”


Though I won’t be posting this until arriving at the hotel in Palo Alto, I am writing from somewhere over the Pacific Ocean, as I fly from Tokyo to San Francisco to attend the Stanford University Center on Longevity’s “Retirement Planning in the Age of Longevity” conference. I’m looking forward to that, and I’m planning to “borrow” Bob Seawright’s format (he is providing excellent real time notes from the CFA Institute annual conference in Chicago) and provide real time summaries of the discussions at the conference. I realize, as perhaps Bob figured out, it makes a lot more sense to provide stream of thought notes, as otherwise I might never get around to write up complete essays about the event.

Though I am not a big fan of these cross-ocean flights, I do find they provide a good opportunity to get some work done. That is easier to do with an empty seat next to me. I’ve developed a theory for how to accomplish this, and it paid dividends on this flight. Just choose an aisle seat in the center part of the plane near the back. For whatever reason, plane seats tend to be assigned from front to back, and so in the case that there are empty seats on the plane, they tend to be found in the back center parts of planes.

On the flight I’ve been reading William Reichenstein and William Meyer’s Social Security Strategies: How to Optimize Retirement Benefits.
My dissertation was about Social Security reform. As such, I’ve read more of the Social Security Administration Handbook than any human being should be forced to read (just as an aside, how did the Social Security rules ever become so complicated in the first place? It is not like the tax code where special interest groups are always pushing for exemptions. It seems like Social Security rules could be vastly simplified by sticking with a few principles, such as people with similar situations should be treated similarly no matter if they are married or single. The problem is, these rules were developed a long time ago when it was still the norm that only one spouse works… Social Security isn’t really designed to deal with the complications of having both spouses work). I’ve written programs to calculate AIMEs, PIAs, adjustments for the age of initial receipt, family maximum benefits, child benefits, survivor benefits, and so on. But time flies, and I wrote those programs about 10 years ago and felt I’ve become rather rusty on these matters in the intervening years. 

So I am glad that I did at least understand the basic idea of the book. I found I was confused about one particular rule, as when I was trying to explain to some relatives that it probably isn’t a good idea for both spouses, each with a full career, to begin their own benefits at age 62, I was wrong when I told them that, fine let the husband begin his benefit at 62, but the wife (who I think may have a long life span) should just take the spousal benefit from her husband’s record from 62 until 70, and then begin her own benefit. It turns out that I am wrong about this… this strategy can only be done when the female spouse reaches her full retirement age (somewhere around 66 now), as prior to that you cannot apply for a spousal benefit without also applying for your own work-history benefit. If this makes no sense to you, then you are in for a real treat when reading Chapter 4.

And so, the objective of the book is how to strategize to get the most out of your Social Security benefits. The book is relatively straightforward through the end of Chapter 3. It covers the basics of Social Security and also how a single individual can maximize Social Security. The book does get vastly more complicated in Chapter 4, which is about couples. The problem is, for couples in which both spouses work, each spouse is potentially eligible for benefits based on their own work record (plus your spouse is also eligible for a benefit based on your record), plus a spousal benefit based on their living spouse’s work record, plus a survivor benefit based on a deceased spouse’s record. Divorcees can get something too, with different rules, if the marriage lasted at least 10 years, but those with multiple ex’s or young kids should watch out for that family maximum benefit based on one earner’s record)

I’m not going to try to explain Chapter 4 in detail, but this matter is important to me as the curriculum director for the Retirement Management AnalystSM designation. Especially for lower and middle income Americans, Social Security may end up providing the vast majority of retirement income. And so it is very important to get Social Security claiming decisions correct. To be clear, Social Security claiming decisions are defined as: “a decision as to when a single individual will begin their own benefit or when each partner of a couple will begin their own benefits and, when applicable, their spousal benefits.”

As the authors show, a wrong decision on Social Security could end up costing retirees a couple hundred-thousand dollars worth of lost benefits. But as I think the authors also clearly show, becoming a factotum on Social Security matters is really complicated. I think it is asking too much for advisors to become deeply knowledgeable about these Social Security rules. But at least I think they should have a solid understanding of the basic principles and be able to figure out approximately correct solutions, and also have a working knowledge about how to use software that will allow them to show retirees the outcomes from different claiming strategies, and to narrow in on a good claiming strategy for both spouses. I should note that the authors of the book do offer software as a part of the website, www.SocialSecuritySolutions.com, but I haven’t had an opportunity yet to test it for myself.

What are the basic principles?

First of all, there is no universal answer for anyone. That is because retirees can make their decisions using two different criteria, and there is a tradeoff between which strategy performs best for the different criteria. The first criterion is:

1.   Retirees should pick the Social Security strategy which maximizes the net present value of their lifetime income from Social Security. And given the fact that current real interest rates are quite close to zero, this is more or less the same as choosing the strategy that maximizes the total lifetime amount (in real terms) of benefits received from Social Security

And the second criterion is:.

2.   Retirees should take advantage of the fact that Social Security provides an inflation-adjusted annuity which increases its payments when delaying the claiming decision through age 70. This helps protect against longevity risk, and it can help extend the life of one’s financial assets as well. The idea being, even if you retiree at 62, you can delay starting Social Security until age 70, such that you spend down more of your financial portfolio between ages 62 and 70, and then spend from your portfolio at a lower rate after that. They show the benefit of delay using deterministic assumptions for asset returns and it is not a sure bet (a particularly bad sequence of returns around age 62 could possibly leave you better off with starting Social Security earlier) but their point is fairly persuasive.  More risk averse retirees should delay claiming Social Security until later, because the longer they live, the more they will end up benefiting from the annuity properties of Social Security benefits.

Retirees must think about their life expectancy, and how important it is to protect one’s income level in the event of living longer than expected. In thinking about these two criteria, Reichenstein and Meyer develop several lessons to help summarize the Social Security rules.

Lesson 1: For a single individual living to age 80, it does not really matter what age to begin Social Security, 62 to 70. The total lifetime benefits received will be about the same, since Social Security is meant to be “actuarially fair” with benefit reductions and increases calibrated to life expectancies. But for retirees living “well beyond” (this term is defined and discussed in the text) 80, it becomes increasingly attractive to wait until 70. This is how Social Security provides protection against the risk of outliving one’s assets.  

Then, for the vastly more complicated case of couples, we get two more general lessons:

Lesson 2: For the higher earning spouse, the relevant life expectancy for the claiming decision is the life expectancy of the second spouse to die (which is higher than either spouse alone). The relevant life expectancy for the lower earning spouse is that for the first spouse to die.

And:

Lesson 3: If at least one of the spouses lives “well beyond” the age when the higher earning spouse turns 80, then the couple can usually maximize their joint lifetime benefits by having the higher earning spouse wait until 70 to claim his/her own benefits.

As a penultimate observation, as the authors note, for people aged at least 55, it is not a good argument to start benefits as soon as possible because you want to get what you can before Social Security “goes bankrupt.” Even if the Trust Fund empties, Social Security is not finished. It is just that benefits would have to be reduced equally for everyone to match incoming contributions to the system. Your benefit would be reduced equally no matter when you started, and there is no chance to gain from starting earlier for this reason.

Finally, is the book worth reading? I didn’t even get to the issues of benefits taxation and the earnings test in Chapter 5, and it does seem like you should want to do your homework to develop a good Social Security claiming strategy. While the book is complex, it is probably worth working through carefully to figure out what will be best for your own case. I can’t summarize it all, but the authors describe a variety of scenarios based on factors such as: the ratio of PIAs between the low-earning and high earning spouse, the relative ages of the two spouses, and the projected life expectancy for each spouse. Decisions also relate to the relative importance a retiree places on maximizing lifetime benefits (when death is expected before around 80) vs. obtaining longevity protection for a lengthy lifetime. As suggested, these efforts could have major implications for making sure you don’t unnecessarily leave oodles of money on the table. The authors describe these matters in probably about as clear of manner as can possibly be expected given the complexity of these issues. 

11 comments:

  1. There is a relatively straightforward discussion of how a couple optimizes claiming in most circumstances available at http://crr.bc.edu/briefs/why-do-women-claim-social-security-benefits-so-early/

    ReplyDelete
    Replies
    1. Thanks, that is a good point. I should try to put together a collection of some of the good work the CRR has been doing to explain Social Security basics.

      Delete
  2. Wade --

    Do you agree with their "first criterion"?

    Seems to me that in all your work on withdrawal rates, and also on investment choices during retirement, your emphasis is heavily on being OK in worst cases, and that's the right approach, and it oughta be the main consideration in SS strategy too.

    Do you agree?

    Dick Purcell

    ReplyDelete
    Replies
    1. Dick,

      Good question. I suppose the answer may partly depend on the amount of other resources you have available to help you hold off starting Social Security.

      They are not saying criteria 1 is superior, but just that this is usually what people think of first.

      The issue here is that you are leaving money on the table and have less income for a long life, which isn't exactly the same as the "failure" that must be avoided. So I can understand that people may want to lean toward this.

      I think the case is pretty strong about seeking the longevity protection benefits of Social Security, but I suppose I do not have any strong opposition to the criteria 1.

      What do you think?

      Delete
    2. Wade –

      I sure agree with your first sentence, but it seems to me that situation pre-empts the question. If you NEED the SS money before age 70, then PV versus longevity doesn’t even get considered – just start getting the money you need now.

      At the other extreme, there are a few of us rolling in so much money no SS is ever needed, it’s just a little extra. For those few, the PV approach makes sense. (I don’t even know what/if the super-rich get SS, from which admission a discerning reader will deduce that I am not there yet.)

      But for everyone else, who can get along without SS until 70 and is not just rolling in wealth, I think it is no contest. The PV approach is wrong-track thinking, longevity/worst-case is the basis for the decision.

      I just glanced at the Michael Kitces post on this topic that you linked us to at the bottom of your post. Based on just glancing and reading his ending questions, I suspect he agrees.

      Dick Purcell

      Delete
    3. Dick,

      You are probably right.

      Now the next question is, why do so many people start Social Security as soon as they can at 62?

      Thanks.

      Delete
    4. Wade, Dick

      The same deferral structure exists in Canada. If you compare the value obtained via deferral with the cost of a commercial annuity providing the same incremental benefit you will typically find that government benefits are a good deal relative to commercial annuities. To some extent this is related to difference in mortaility expectations resulting from adverse selection.

      Delete
  3. On the concept of family maximums, thankfully for all involved, amounts paid to ex-spouses based on disability or age are not included in the calculation. :)

    Regarding actuarial fairness, I think that's only very-roughly-sometimes-applicable. For example, for a single person with an average (i.e., half male, half female) life expectancy, delaying Social Security should be actuarially neutral -- if the legislatively-baked-in discount rate is appropriate given current market interest rates.

    ReplyDelete
  4. No simple adjustment can be exactly appropriate given that both life expectancies and interest rates both change frequently. But the adjustment still seems to be very close to actuarily correct; see http://crr.bc.edu/briefs/can-the-actuarial-reduction-for-social-security-early-retirement-still-be-right/

    ReplyDelete
  5. Thanks all.

    One point that came up at the conference today was that the actuarial fairness calculations have a built in implicit real interest rate of 2.9%. Since real rates are so much lower than this now, it is another reason that makes deferral more attractive at the present time.

    ReplyDelete
  6. Wonderful, just what a blog it is! This blog has provided the helpful data to us continue the good work.
    http://www.best-5-home-security-companies.com/vivint-security/

    ReplyDelete