Tuesday, March 22, 2016

A Follow-Up on Confusing Issues

In a recent post, I asked, “What Do You Find Most Confusing About Retirement?” You can read the comments I received and my responses, as well as responses from other retirement planners and researchers, at that link. I also searched those comments for common threads and will consolidate my (our) responses in this post.

Social Security benefits are confusing, as is the process for identifying optimal claiming ages.

Amen. I have studied Social Security benefits extensively and have concluded that it is a topic with which most retirees and planners (including me) need help. Becoming an expert requires a lot of work and understanding the rules is only the first step. You then must apply the rules optimally for a given retiree's unique financial situation, life expectancy and tolerance toward longevity risk.

My favorite Social Security guru is Mike Piper at ObliviousInvestor.com. He has updated his book, Social Security Made Simple, to include recent changes regarding file-and-suspend and I recommend it. I am also quite fond of Laurence Kotlikoff's MaximizeMySocialSecurity software to identify optimal claiming strategies.

Trying to become a Social Security benefits expert on your own will be quite frustrating.

I have trouble finding a good retirement planner.

Join the club. I haven't found a lot of retirement planners I would trust with my money and the top planners and retirement researchers with whom I have discussed this challenge completely agree. But there are good ones out there, as I suggested in the previous post.

It isn't terribly difficult to become a Chartered Financial Analyst (CFA) or Certified Financial Planner (CFP), so those credentials alone don't guarantee you've found a good planner. I put a bit more stock in Retirement Management Analysts (RMAs) because I know their training process and I have met several of them, but it's still no guarantee. I would suggest you use such designations as a baseline starting point in your search. I really only trust financial planners I know, like Dana Anspach and Mike Lonier (see their comments on my previous post) or those referred by someone I trust.

I also suggest that you review a prospect's qualifications. You can usually tell when a CPA, insurance salesman, tax preparer or stock broker has added retirement planning to the quiver. Finding a planner with insurance or investment expertise, for example, can be a good thing unless it's merely a path to generate more insurance or stock sales.

Don't limit yourself to retirement planners who are located near you. Because there are relatively few good ones, you may have to search other cities or even states. Planning via telephone and internet is perfectly viable.

Lastly, and perhaps most importantly, I recommend a fee-only planner who has no monetary  incentive to sell you products. If the planner will manage your money or provide investment advice, confirm in writing that he or she assumes fiduciary responsibility for your account. And don’t give your adviser custody of your funds! (See A New Ponzi Scheme Every Week.)

To reiterate, most of us agree that good retirement planners are hard to find, but they're out there.

“Figuring out the most tax-efficient way to distribute from IRAs, Roth IRAs, and taxable accounts at the right time is just a few too many variables for me and Excel.”

Many planners recommend that you distribute from the taxable account first, then the traditional IRA, then the Roth last, but there will be situations that dictate otherwise. It may sometimes be optimal to withdraw from two or more types of accounts in the same year and the optimal withdrawal plan may change with time. The decision between Roth and Traditional IRA withdrawals will be affected by your opinion of future tax rates, which are impossible to predict. The outcome is also tied to your Social Security claiming strategy.

The problem is so complicated that at least one company, Retiree Inc., has been formed primarily to solve this problem for retirees. The company offers to develop an optimal withdrawal plan for a fee of $500.

I like Kotlikoff's E$Planner for this. It allows you to switch the order of withdrawal and measure the results by the consumption generated. It’s less expensive (currently $149) and is a more general purpose retirement planning tool. (I have used both Kotlikoff products, but I am unfamiliar with the Retiree Inc. product. I should also note that many find E$Planner confusing to use.)

It may not work well to decide on a withdrawal plan at the beginning of retirement and stick with it come hell or high water. Your situation may change as retirement progresses, suggesting an annual review and tweaks to your strategy may be warranted.

Another reason to dynamically manage this process is that tax laws change over time. Changes are nearly impossible to predict because they are political and they can have a significant impact on your plan. Stretch IRA's, for example, appear to be in present danger. A withdrawal plan you choose today might not make sense if tax laws change.

Retirement plan distribution choices also affect creditor protection of assets. Retirement plan assets are generally better protected than assets outside a plan. When you are spending retirement plan assets you are probably decreasing you pool of protected assets. (See Is My Retirement Plan Protected?)

Find a good retirement planner with tax expertise, or one who is willing to get help from a tax expert. If you insist on doing it yourself, try E$Planner. Optimizing your withdrawal strategy can provide significant improvements, but it is, in my opinion, a second-order problem. Getting it mostly right or a little wrong won’t make or break your plan. Spending from taxable accounts, then traditional IRAs, then Roth IRAs may get you most of the way there. The more important piece of the strategy is to adapt your plan as changes dictate.

The potential for early retirement.

I wrote a few posts on this previously beginning with The Risk of Retiring (or Being Retired) Early. Retiring early is far riskier than most people assume. We say the most important factor in the cost of retirement is how long you live, but what we really mean is "how long you live in retirement." Retiring early means increasing the largest risk factor in the cost of retirement.

Retirees who ask, “Do I have enough savings to retire?” are often frustrated when they can get a yes or no answer. The issue is far more complicated. A better question would be, “Do I have enough savings to retire and maintain my desired standard of living with an acceptable risk of not going broke before I die?”

Two of the critical pieces of that question can be answered by you alone. What is your desired standard of living? What do you consider an acceptable risk of going broke before you die?

Retiring earlier is significantly riskier than retiring later, so the correct answer to the question, "Can I retire early?" is "Maybe – we'll need to run the numbers – but it would probably be a lot safer not to."

A simple way to look at this problem is to add your expected annual Social Security benefits to the annual payout you could receive by purchasing a life annuity. According to Wade Pfau’s Retirement Researcher Dashboard, a 65-year old couple could currently receive a 3.85% pay-out, or about $3,850 a year for every $100,000 they have saved.

For example, a couple who has saved $250,000 and expects $30,000 a year in Social Security benefits could generate $39,625 of annual income.

Do you feel like the total you calculate would be adequate retirement income? Are you willing to accept the risk that this amount of income will cover future expenses including significant expense surprises? If so, you can probably retire.

Continuity of investment and household expense management if I die or become disabled and where to find emotional, decision making, and day-to-day support in very old age, when we're less sharp and mobile.

These are concerns I share, even though I am blessed with bright and reliable children who could help and a wife with an MBA. These issues are a bit outside my area of expertise, however.

Huffington Post recently provided a list of “10 Bloggers Who Make Aging A Whole Lot Easier.” Ted Carr and his wife were applauded in that article for providing excellent podcasts on a broad range of retirement issues.

When I asked Ted about the continuity issue, he told me that his “May podcast is with Professor Sharona Hoffman who has written Aging With a Plan. Her book contains a lot of information about different types of support to help with all sorts of issues. She too does not have children and shares similar concerns.” I downloaded the book to my Kindle but haven’t read it, yet. I suggest you take a look at the other nine recommendations, as well.

I thank my readers who posted their concerns. I will keep an eye out for additional concerns at that post and invite anyone to ask any retirement finance question at any time following any of my posts.


  1. cynicalanddisgustedMarch 26, 2016 at 4:58 AM

    Retiring early has significantly greater risk than normal retirement as (Obviously) You will be retired longer. But the risks is increased as the sequence of returns risk is not linear to the length of retirement. The longer time horizon for withdrawals means that you have increased your time horizon for poor early returns nearly by the number of extended years of retirement - not by the fractional extension of the retirement.

    SOR is (IMHO) the biggest risk associated with an early retirement.

    1. Under reasonable assumptions of spending (say, not more than 4%) for early retirement on the order of 5 to 10 years, sequence risk doesn’t increase much and it’s completely manageable.

      Using Milevsky’s formula with an expected real return of 5% and sigma of 12%, a 19.7 year life expectancy for a 65-year old, and Wade Pfau’s current sustainable withdrawal rate estimate of 3%, the probability of successfully funding retirement is 96%. For a 60-year old with life expectancy of 23.6 more years, the probability of success decreases to only 95%, and for a 55-year old with life expectancy of 27.8 more years, the probability of success is still 93.9%. This is a relatively small range within the margin of error for a typical retirement plan and P(ruin) doesn’t increase significantly.

      (Here’s a spreadsheet that let’s you play with Milevsky’s estimator.)

      More importantly, exposure to sequence risk is both manageable and a choice. We can reduce sequence risk by lowering the withdrawal rate. We can also reduce the magnitude of the risk by investing less in equities, especially wise early in retirement as most research suggests.

      In fact, retirees who choose to fund retirement with Social Security benefits, annuities, TIPs ladders, pensions and the like instead of equities will have no sequence risk, at all. Hence, exposure to sequence risk is a choice.

      Sequence risk can be mitigated or completely avoided. The retiree whose biggest risk is the additional sequence risk created by retiring a few years early probably just needs a better plan.

      Analogously, if your biggest concern with swimming is shark bites, either stay in the shallow water next to the beach or swim in a pool!

      Thanks for writing! I love getting comments from regulars.

  2. cynicalanddisgustedApril 4, 2016 at 8:05 AM

    Thanks for your followup comment. I couldn't get Milevsky's estimator to work - but I built my own (simpler but probably adequate to the task). Sure enough, I could not find what I expected - Milevsky is right - and I obviously was wrong. SOR is not really dependent on the length of the retirement - even when comparing 25 to 40 year retirement plans.
    Thank you for straightening out that erroneous thought.

    1. You're welcome, but I would like to tweak your conclusion just a bit more.

      SOR is actually quite dependent on the length of retirement, but it isn't linear. The increase in probability of ruin flattens out at around thirty years, so 40 or 50 years aren't much riskier than 30 years, but 30 years is a lot riskier than 5 years or 10.

      That's why I disagreed with your initial comment that adding a few years of retirement at age 65 to say, age 60 is the biggest risk of retiring early. You changed a possible 35-year retirement to a 40-year retirement. Not a big hit to P(ruin).

      Play with Milevsky's formula a little longer and you will see what I mean.

      Fun, huh? :-)