Friday, January 13, 2017

The Opening Game

In my previous post, The Opening, the Middle Game and the Endgame, I compared retirement to a game of chess that can be broken down into three distinctive game phases. Like the three stages of chess, the three stages of retirement have different risks and opportunities. It is very difficult in either chess or retirement to predict what your position will be at the beginning of the next phase until you are almost there.

The opening game typically begins around age 65 to 70, depending on when the worker retires. It can begin earlier if the retiree is one of the 50% or so who reports that they were forced to retire earlier than planned (see The Risk of Retiring (or Being Retired) Early ). About 7 in 10 Americans born in 1952 have lived to reach age 65 in 2017, according to the Social Security Administration [1].

The opening game lasts until about age 75, though this is my own somewhat-arbitrary break point and depends on individual household situations. The beginning and end of the three games are fuzzy ranges, not specific ages.

The table below shows the median remaining life expectancy in 2017 for a male, female or at least one spouse of a couple, all age 65. The second column shows the 5% probability that these will live even longer. For example, about half of females age 65 today will live to age 86 and about 5% will live to age 99.

Remaining Life Expectancy (Years) at Age 65
Median 5% Probability
Male 18 30
Female 21 34
Either 24 35

The opening phase of retirement is when retirees tend to be most physically active. For example, as I have mentioned previously, airlines report that Americans stop traveling internationally around age 70 and domestically around age 80. This decline in activity may also mean a decline in spending.

David Blanchett and Sudipto Banerjee have reported in multiple studies that spending typically declines about 1.5% per year throughout retirement and that spending late in retirement is typically lower than that at the beginning of retirement even when large end-of-life expenses are experienced. So, the opening of retirement may be the most expensive of the three games. Not all households will experience this decline, however, and it is very difficult to predict for an individual household.

Key risks of the Opening Game include forced retirement, the “Tax Torpedo”, and sequence of returns risk. (For other risks, see Why Retirees Go Broke).

The three stages of retirement have different opportunities and different risks.
[Tweet this]

About half of recent retirees report that they were forced to retire sooner than they had planned to take care of a relative, as a result of age discrimination, as a result of layoffs or for other reasons. The decade before retirement has a large impact on retirement finances and retiring sooner than planned can be devastating.

The “tax torpedo” [4] is the name given to the scenario in which other income, like Required Minimum Distributions (RMDs) from your retirement accounts, forces more taxation of your Social Security benefits. Depending on your income and marital status, up to 85% of your Social Security benefits can become taxable. These RMDs must begin at age 70½ and this tax increase can take a substantial bite out of your Social Security retirement benefits.

Sequence of returns (SOR) risk is the risk that a series of poor market returns early in retirement can result in a retiree's savings being depleted earlier than planned. Poor returns later in retirement have less impact on the survivability of invested retiree savings and this risk declines exponentially throughout retirement. That means that the Opening Game is the phase of retirement when a series of poor portfolio returns can have the greatest negative impact on future retirement finances.

A retiree who experiences these poor returns early in retirement isn't likely to deplete her savings during the Opening Game. In fact, my SWR studies show that portfolio depletion with a reasonable spending rate of about 4% or less rarely occurs before the retiree's median life expectancy at age 65. (See Death and Ruin.) In other words, a series of poor returns that occurs in the Opening Game won't deplete savings until the Middle Game, assuming spending is somewhat reasonable. Spend 10% of your savings annually, however, and you can deplete your portfolio pretty quickly.

Sequence risk is also very high during the decade preceding retirement. Many households who believed they were all set for retirement in 2000 to 2006 encountered a dramatic setback when the Great Recession hit in late 2007. Many had to postpone or scale back their retirement plans. In my previous post, I mentioned the transition into the Opening Game (obviously not a feature of chess). That's a good time to reduce equity exposure to mitigate the risk of losing capital with little time to recover before retirement – you probably should have reduced your equity exposure before the Opening Game.

Several key decisions in the opening game also affect later games including claiming Social Security benefits, spending a reverse mortgage, purchasing a fixed annuity, and purchasing long-term care.

Claiming Social Security benefits is an excellent example. Claim benefits early and you increase Opening Game income but decrease income in later games – should you live that long. Likewise, borrowing from a reverse mortgage in the Opening Game will increase spending immediately but you might have an even greater need for that spending in later games – should you live that long.

Purchasing a fixed annuity will cost more in the Opening Game because you have a longer life expectancy. On the other hand, delay too long and you may have significantly less capital with which to purchase an annuity. At some age approximately halfway through the Middle Game, buying a fixed annuity may no longer be economically sound.

Long-Term Care insurance policies require that you qualify medically. Wait too long to purchase one and you may no longer qualify.

Health  care expenses typically increase after we retire. According to health economist Austin Frakt [5],
"Older people need more health care, and they spend more. Compared with the working-age population (people 19 to 64 years old), those 65 to 74 spend two times as much; those 75 to 84 spend four times as much; and those 85 and older spend six times as much. And the growth in health care spending is faster for retirees than for younger Americans."
These decisions have both “expiration dates” and impacts on future games that are both good and bad in addition to the immediate impacts they have on the Opening Game. Decisions need to consider the impacts on all three games in a good retirement plan.

The following summarizes the salient characteristics of retirement's Opening Game:

The Opening Game has its own risks and rewards and decisions made in early retirement can have a dramatic impact on later games. A good retirement strategy will not only consider the impacts of decisions on the Opening Game but also impacts on later games. Plan with the understanding that the three games are different, that each may require its own strategy, that decisions may affect more than the current game, and that you won't know what pieces are still in play until you almost reach the next game.

In my next post, I'll move on to the Middle Game.


[1] Life Tables for the United States Social Security Area 1900-2100. SSA,gov.

[2] Blanchett, D. (2013, November 5). Estimating the True Cost of Retirement. Morningstar.

[3] Banerjee, S. (2012). Expenditure patterns of older Americans, 2001-2009. EBRI Issue Brief, (368).

[4] The Tax Torpedo explained, Kiplinger.

[5] Austin Frakt, Blame Technology, Not Longer Life Spans, for Health Spending Increases.


  1. An excellent post Dirk, as always! I have one disagreement which relates to Sequence Risk. Sequence Risk exists at any and all ages – just ask those later in retirement if they worry about how much their portfolios fluctuate. The misperception that it is only more prevalent during the early years of retirement comes from two sources: 1) the prevalence of research that focuses on those starting years (notice most start with the premise of the retirees being in their 60’s; and 2) the application of the rule-of-thumb or conservative spending approaches when starting out (mostly as a response to managing sequence risk).

    So what happens, and Kitces mentions this occasionally, is the conservative spending results in the portfolio balance growing to “a gazillion” dollars. In other words, there’s more money available than needed for spending! Of course, due to this excess from the planning approach, it creates a decrease in exposure to market sequences. Clinically, from actual observations from clients who are in their 70’s and 80’s two things do happen: 1) their spending on a dollar basis does go down a little over time from not doing activities they did when younger retired, and 2) due to the growth in their everyday costs (utilities are higher now than before, phone and TV costs are higher now than before, etc. (in other words, inflation), their spending relative to their portfolio balances hasn’t gone down (or stated differently, their portfolio balances haven’t gone up faster than their spending). Thus, they remain just as exposed to sequence risk as before.

    So if becomes more important to develop decisions rules and a process to manage things with both spending needs, flexibility in those needs, and retaining shares when markets misbehave so they’re not selling into continually declining share prices.

    As to RMDs, I find that for most people who haven’t over saved to begin with, or focused savings exclusively on tax qualified retirement accounts, RMDs simply become a subset of the total they actually need. In other words, without a little tax diversification before 70 ½ (e.g., Roth conversions and saving in non-tax qualified accounts as well), their spending needs exceed the RMDs to begin with.

    Of course, specific situations are different in subtle ways from others, and that is where individual financial planning and strategy development is important.

    Great post, and look forward to the Middle and End Game posts!

    1. Larry, you and I continue to disagree on the nature of sequence risk and I suppose we always will, given that we have probably discussed it for a hundred hours. Sequence risk is a function of the spending rate as a percent of portfolio size, variance of portfolio returns and remaining years in retirement.

      Since the latter constantly declines as we age, sequence risk declines, assuming we hold the other two factors constant. Poor returns when you are 98 years old simply don't have time to cause the damage that poor returns at 65 do.

      I agree with most of what you say here and I certainly respect your opinion, but I have to go with Moshe Milevsky on this one: sequence risk is always present but declines exponentially with age, all other factors held constant.

      Thanks for writing. There is no one I enjoy arguing with more! :-)

    2. Sorry, left out the obvious -- expected portfolio return. There are four parameters, not three.

  2. I would agree "all other factors held constant." However, this is another area of contention - rising or declining glidepaths, which means those other factors are also stochastic in real life (not static as a research parameter).

    Any decline in portfolio value regardless of age hurts. It's a combination of time down and time to recover compared to how much time a retiree has remaining.

    BTW, declining glidepaths (portfolio allocations slowly become less exposed to stock fluctuation through use of more shorter term bonds) continues to show itself in the research method where each year is recalculated, what I've come to call multi-casting. Shawn Brayman and I are continuing a research project and this is one area we intend to sharpen the pencil on.

    Now it's 101 hours of discussion (I don't view it as an argument)!

    1. I mean "argument" in a good way. But, I don't believe in glide paths, either, except for research and mutual funds. Makes no sense for a household to determine an asset allocation before they have to.

      I'll make you a deal. Name just one retiree who has ever outlived his/her portfolio primarily as a result of a series of bad returns experienced after the age of 80 and I'll rejoin this "argument."

      Cheers. :-)

    2. By the way, elders running out of savings is not the same as elders running out of savings primarily as the result of a poor sequence of returns after age 80. Know any of those?

  3. Actually - not predetermined as to must make a change every pre-determined age. I mean a slow adjustment taking risk off the table (risk being defined as the magnitude of the decline). The greater the decline of one more aggressive allocation compared to a lesser decline of a less aggressive allocation (of the same holdings to be fair too) results in a longer recovery for the more aggressive portfolio. I have plenty of elder clients who are outliving their portfolios as a result of NOT being exposed to as great a decline as they once may have when they were younger. I think we're using similar terms for different concepts maybe?

  4. Probably terminology, because I totally agree with your first three sentences. I was referring to predetermined AA's.

    I'm confused about your comments concerning some of your clients, though. Are they over-spending after reducing equity exposure?

    Thanks for sharing this. I have readers who need to understand that while people who fund retirement largely from investments have a reasonably small chance of outliving their savings, it does happen. I received an email recently from an 80-year old couple who have depleted their savings with a pretty low floor. It isn't pretty.

    And this is why I love arguing with you. Or debating. Or whatever you want to call it. I always end up learning something. :-)

  5. As well you should be confused (your second sentence). I left out the word NOT. It should read:

    I have plenty of elder clients who are NOT outliving their portfolios as a result of NOT being exposed to as great a decline as they once may have when they were younger. Thanks for pointing out the confusion - I'd be confused too!