Friday, February 5, 2016

Retirement Plan or Investment Plan?

My last few posts have explained that outliving retirement savings as a result of sequence of returns risk isn’t the worst thing that can happen to a retiree. Becoming insolvent as the result of unexpected and uncontrollable expenses is significantly worse, though less likely – less than half a percent of Americans over age 65 declare bankruptcy.

I also recommended that a comprehensive retirement plan address all known financial risks and not simply the risk of outliving our savings due to market volatility and overspending. In this post, I want to suggest what those other risks might be (by reviewing reasons that people over age 65 typically file for bankruptcy and adding a couple of risks of my own), and some potential ways in which planning might mitigate (lessen the severity) or even insure against them.

Becoming insolvent (unable to pay your debts) and declaring bankruptcy are not the same thing. Declaring bankruptcy is a legal action that might help deal with insolvency, depending on the specifics of your situation. Whether or not it is the best alternative is something you would want to discuss with a bankruptcy attorney, should you ever need to consider it.

(In this post, I am discussing the risk of insolvency. In previous posts, I used bankruptcy statistics from the U.S. Court system. It's important to distinguish between the two.)

Far too many retirement planning discussions focus primarily on market risk.
[Tweet this]

Given that I described spending crises as chaotic, synergistic (in a bad way – they combine to have a greater negative impact that the sum of their parts), unpredictable, and difficult to stop once they begin to spiral downward, it might seem that there would be little to be gained from planning, but that isn't the case.

Certainly, there are some spending crises that would completely overwhelm nearly anyone's personal finances (a spinal cord injury, for example, could cost millions), but there are many crises that can be mitigated with the right planning. Let's start by recalling the 10 most common causes of elder bankruptcies:
  • Credit Card Interest and Fees (67%)
  • Illness and Injury (65%)
  • Income Problems (41%)
  • Aggressive Debt Collection (35%)
  • Housing Problems (27%)
  • Divorce (15.1%)
  • Birth or adoption of child (9.7%)
  • Death of family member (7.5%)
  • Retirement (6.7%)
  • Identity Theft (1.9%)
Recall that the major causes were self-reported and that most respondents to the survey reported multiple causes, so the percentages total more than 100%.

In Retirement and Chaos Theory, I suggested that insolvency behaves like a fixed point attractor. Our finances can develop a positive feedback loop and spiral downward to that point if they enter into insolvency's “gravitational field.” Sometimes, it's nearly impossible to avoid ruin once that process begins.

One way to mitigate that risk would be to stay as far away from insolvency's region of influence as possible, avoiding the risk of going broke like avoiding the gravitational field of a black hole. Consumer credit, inadequate insurance, and financial dependence on a spouse, for example, move you closer to insolvency's gravitational pull.

Let's look at those risks, chaotic though they may be, and show that there are ways we can plan for them.

Credit card interest can be deadly and research shows that typical credit balances continue to increase throughout retirement. Retirees can plan to pay off all consumer (non-mortgage) debt before they stop working and to control credit use after retiring by sticking to a solid retirement spending plan.

Most retirees will be eligible for Medicare when they turn 65, but Medicare doesn’t pay all of our medical costs. Fidelity Investments has been tracking retiree health care costs since 2005 and estimates that a 65-year-old couple retiring in 2015 will need $245,000 to cover future medical costs, not including the cost of long-term care.

That's more than most households save for retirement. It will devour a large portion of Social Security benefits for many. Retirement plans shouldn't ignore this risk.

We might consider planning to retire to a state with low health insurance costs, lower long-term care costs, and lower medical costs, but we should also consider the quality of care available.

Income problems cited in the studies were primarily the result of either age discrimination leading to unplanned early retirement or medical problems leading to unplanned early retirement. In some cases, unplanned early retirement was necessary to take care of a family member with a medical problem.

A retirement plan can assess the risk of unplanned retirement in our profession. Optometrists can work longer than oil field roughnecks. We can also assess our own health risks and the health risks of others who might need our care. These observations might lead us to the conclusion that we need to plan to find a job where we are more likely to be able to work to an older age, even if it pays less, or that we need to increase our savings, or lower our retirement income expectations.

Foreclosures soared in 2007 and they are particularly damaging to seniors. Foreclosure risk can be reduced by relocating to an area with less expensive housing, for example, or by planning to pay off (or pay down) a mortgage before retirement.

Believe it or not, one can purchase divorce insurance, though that doesn’t seem like a great way to deal with the devastating risk of elder divorce. A comprehensive retirement plan will model our financial risk of a divorce at various ages and can show the impact that a break-up would have on each spouse. E$Planner software can provide such contingency plans, for example. This analysis may provide insight into ways our retirement plan might be adjusted to improve outcomes for both spouses.

As with elder divorce, the impact of the death of a spouse at various ages can be modeled in a retirement plan. Life insurance, however, works much better than divorce insurance and software like E$Planner can calculate the amount of life insurance that will be needed at various ages. It can also model the impact of that death on Social Security benefits.

The final cause for bankruptcy cited by the Institute for Financial Literacy is identity theft (1.9%). I have previously recommended placing a freeze on your credit report.

Although lawsuits don’t appear to be major contributors to elder bankruptcy, I’m still fond of umbrella liability insurance.

Do you have a child or parent with health issues? Legal or medical costs could break your retirement plan and need to be considered.

Lastly, if you can’t avoid bankruptcy, there are steps you can take to protect assets. Holding assets in a retirement account will usually protect them from creditors. Future Social Security benefits are also protected. Even Social Security benefits that you have already received can be protected in bankruptcy if you hold them in a separate account that is only funded by those benefits. Commingling them with other funds can expose them to creditors.

The following list of insolvency risks is not comprehensive (though all but the last two risks were identified in a study as the top causes cited for elder bankruptcy). Nor is my list of possible strategies to mitigate them by any means complete. The key point is that it is important to identify the major financial risks in your retirement plan, to quantify them, and to identify ways to mitigate them, if possible. If mitigation is not possible, it is important to understand the risks if for no other reason than to avoid overconfidence in your plan. Far too many retirement discussions focus primarily on market risk.

Risk Potential Mitigation or Insurance
Credit Card Interest and Fees Pay off consumer debt before retiring
Illness and Injury Relocate for lower medical costs, Medicare, Health Insurance
Income Problems Change to a more sustainable job, evaluate your household's health risks
Aggressive Debt Collection Understand your rights
Housing Problems Pay off mortgage before retiring, relocate
Divorce Develop retirement financial contingency plans for divorce
Birth or adoption of child
Death of family member Develop financial contingency plans, life insurance
Retirement Evaluate your prospects for continued employment
Identity Theft Educate yourself, keep current on identity theft scams and freeze your credit report
Fraud Educate yourself and keep current on elder fraud
Financial needs of parents and children
Legal liability Umbrella liability insurance

A good place to start would be to list the risks to which you have significant exposure from the table above, add any other risks that might apply specifically to you, and then check to see if they are considered in your current retirement plan. If not, then you (or your advisor) have more work to do.

A plan that only addresses asset allocation and withdrawals isn't a retirement plan – it's an investment plan.
[Tweet this]

Does your retirement plan consider these common causes of insolvency and plan for them? If it mostly tells you how to invest your savings and how much of your portfolio you can safely spend each year then you don’t really have a retirement plan.

You have an investment plan.


  1. Unfortunately, that $245K Fidelity figure has all kinds of embedded assumptions that Fidelity does not seem to share, even on their web site. For instance, what inflation rate are they assuming for Medicare Parts B & D? What are they assuming that your current funds reserved for health care will earn until they are spent? What are they assuming about average annual increases in out-of-pocket health care costs (including co-pays & deductibles)? What general inflation rate are they assuming? Personally, I think that they've significantly understated the costs. For one thing, they're assuming an AVERAGE joint life expectancy, which implies that there'll be a 50% chance of additional years of health care spending required. They want to challenge people to invest more (with Fidelity, of course), but not freak them out with too high a number. The devil's in the details...

    1. Is there a reason you believe the cost is understated? (I have no idea whether it is or isn't. As you say, they don't state their assumptions and I'm not aware of any similar studies. If you know of others, please share.)

      Not sure I understand your point on the average life expectancy. An average will understate the spending for about half of the population, yes, and it will overstate it for the other half. I must be missing your point.


    2. I think Fidelity's costs are understated because they are based more on historic data than forward looking trends. For instance, drug prices -- even for generics -- have begun to accelerate at an unprecedented rate relative to the general (CPI) inflation rate. Also, medical spending and cost increases were likely restrained somewhat during the Great Recession and its aftermath, yet these data enter prominently into Fidelity's calculations.

      The point about life expectancy is that it's risky to create a financial plan assuming you'll only live to an average life expectancy. The risk of outliving your assets would be unacceptably high and this is a point you've made yourself a number of times. So why does Fidelity tout a figure that will be too low for about 50% of retired couples? To put it another way, their $245K figure isn't sufficient for many couples' retirement plans.

    3. You could be correct about the cost being even higher in the future. As K. K. Steincke once said, it is difficult to make predictions, especially about the future. I suspect they won't be much lower than that.

      Completely agree that you should never plan for retirement based on historical averages (though they are helpful). I don't think I suggested that. If I implied it, I apologize.

  2. The Fidelity numbers may be well-publicized but I believe they can be misleadingly low, too. Many people do not recognize that Fidelity's numbers exclude dental and vision costs, not just any provision for LTC costs. And of course, there are also the much higher Medicare B and D rates for people with MAGIs of over $170K. IMHO, just like with one's investment plans typically one thinks about each major asset class, it is not hard to do a line-by-line cost estimate that anticipates each key health cost area: Medicare B, MediGap, Part D, Vision, Dental and out-of-pocket. Then one can pick historical or their own guess at healthcare inflation. When I do that for myself. I find Fidelity's numbers are low by about 50%, even assuming no LTC and that healthcare costs end at Social Security's predicted age of death.

    1. Apphound, I believe you are saying (correct me if I misunderstand) that when you estimate your own future health care costs, you come up with an estimate that is larger than the average that Fidelity estimates for all retirees.

      Because Fidelity's is an average, that should be true for about half of retirees. But, it should be an overestimate for the other half.

      Statistical inference can't predict how an individual's financial future will turn out (nothing can). The purpose of these studies is to provide more information for creating an individual's estimate than would be provided by simply saying "we can't predict it." They can only tell you what will happen most often, and when combined with a standard deviation (Fidelity didn't publish one to my knowledge), they can tell us the probabilities for different levels of expenses than just the average.

      You might find the Fidelity information useful in creating your own estimate. You might use it instead of creating your own estimate, because it is the best possible guess if the only information you have is that you're a 65-year old couple. Or, you might ignore this information altogether and create your own estimate.

      No matter which approach you choose, you will still have only a guess of what your future medical expenses might cost. There is no way to predict which guess is best for an individual household, so choose the one with which you are most comfortable. In theory, your guess should be better because you know a lot more about your personal situation than Fidelity does – they assumed only that you were 65, married and lived somewhere in the U.S.

      Fidelity's guess is that we could measure the medical expenses for the remainder of the lives of all American 65-year old couples today until their deaths, average those expenses, and come up with something near $245,000.

      Your household's costs could indeed turn out to be $368,000 (50% higher) and Fidelity's expected average cost could turn out to be precisely $245,000. Neither of you would be incorrect.

      The three things to take away from the Fidelity data are, first, that you should expect to pay a lot for medical expenses during retirement, probably somewhere around $250,000. Second, your personal expenses could be much lower or tremendously higher than Fidelity expects will be typical. And third, you need to plan, as best you can, for not only "expected" levels of medical expenses, but for the possibility that they will be much higher.

      Thanks for writing!

  3. Yes, my estimates for required healthcare spending -- even my optimistic ones -- are much higher than Fidelity's "average" number. But the point I guess I failed to highlight is that it is not difficult to develop a more realistic expectation that using Fidelity's average. Other studies and sources provide more detail than Fidelity's single number and liekly make more sense IMHO. For instance, I found the analysis at the link below interesting and much more in line with my own analysis content:

  4. "it is not difficult to develop a more realistic expectation that using Fidelity's average."

    I did miss that point and you are correct. Thanks!

    1. AppHound, I apologize for the lengthy response. I receive similar questions constantly and was trying to write a response that broadly addresses the issue. I may write a post on it.

      Thanks again for writing.

  5. If Fidelity and others wanted to provide more useful numbers for financial planning, they could give estimates in percentile terms other than the median point (e.g., 95% of today's 65-year-old couples will need no more than $xxx in today's dollars for their future healthcare expenses). With all of the caveats, of course, about what isn't included in the figure.

    1. That would indeed be more useful.

      My cynical side suspects their interests are more along the line of selling more products than helping planners. Nonetheless, I will see if we can get that information from them and I will get back to you.

      Still, I would rather see the information from someone who isn't trying to sell me something, so I'll see if I can find another source while I'm at it.

      Full disclosure, I've been a Fidelity customer for decades and I am happy with their service. I am also a customer of Schwab and Vanguard. I like them all for different reasons and I have bones to pick with all of them, as well.

      Thanks for writing!