Thursday, November 12, 2015

The Role of Xanax in Asset Allocations

It's amazing how often this happens. I publish a post and within 24 hours something pops up in the news that would have worked beautifully with it. The most recent post in this case is Homo Economicus and the something-in-the-news that works with it is an opinion piece I read in the New York Times online this morning by J.L. Cowles entitled “Defeating My Anxiety.”

Mr. Cowles' piece is actually about dealing with his anxiety. Investment results are just one cause of his angst (see his second paragraph below for a more complete inventory) but those results illustrate quite well the concepts of risk tolerance towards short-term portfolio volatility that I discussed in Homo Economicus compared to risk tolerance towards long-term volatility. The first two paragraphs from that Times piece are relevant to retirement:
When the stock market crashed in 2008, my wife and I were 70. And we saw half of our retirement funds disappear. Before the crash, we felt secure in the belief that we had enough money to last as long as we lived; after the crash, we feared that we would not, and I worried about it a great deal. I had a hard time going to sleep and an even harder time going back to sleep after getting up to go to the bathroom in the middle of the night. I came to hate going into that bathroom because I knew my demons resided there and would invade my consciousness immediately.
By the time the stock market began to recover and our savings were again at a comfortable level, I had become conditioned to associate my nightly bathroom trips with “worry time.” I would worry about everything: home repairs, trip planning, medical issues and all the vicissitudes of old age, fears of infirmity, dying and seeing my friends and loved ones die.
It was probably not Cowles' intent that my first thought after reading his article would be that his stock allocation had been too high, but it was. The stock market fell a little more than 50% from late 2007 to early 2009, so the Cowles must have been fully invested in equities. They had laid a lot of chips on the table.

I'm not a doctor, but if you can't sleep at night because you're worried about losing your retirement standard of living in a market crash, you have too much of those savings invested in stocks. If you can't sleep and need anti-anxiety medications because you're worried about your portfolio, you have way too much invested in stocks.

"When the stock market crashed in 2008, my wife and I were 70. And we saw half of our retirement funds disappear."
[Tweet this]

Cowles notes that “the stock market began to recover and our savings were again at a comfortable level,” though he doesn't say they fully recovered. His portfolio will probably eventually recover, assuming he didn't bail out at the bottom of the crash, but that's the big risk.

If Cowles decided near the bottom of the crash that in addition to finding anti-anxiety meds, religion, and meditation he needed to eliminate a source of the anxiety, he might have chosen to reduce his equity allocation (sell stocks). A lot of research shows that investors tend to buy at the top and sell at the bottom. With a lower allocation to stocks after the fall, the climb back uphill becomes even harder. Your equity allocation will already be lower because your stocks will have fallen faster than your bonds. Selling more stocks makes it worse.

As I mentioned in Homo Economicus, it's important to understand your risk tolerance before you retire. If you guess that your tolerance to short-term volatility is higher than it turns out to be, you are faced with two unattractive options near the end of your first bear market after retiring. You can sell equities to stop the bleeding, increase the sleeping, change your asset allocation to where it should have been all along, and severely hamper your prospects for fully recovering.

Or, you can fight through the pain and convince yourself, as your stockbroker tried to do, that the market always recovers. Keep in mind, of course, that your savings portfolio isn't the market.

The Cowles say they saw half their retirement savings disappear in 2008 at age 70. That's a lot of stress when you have no career to return to.

Insuring that you have a solid floor of income from assets not tied to the stock market may ease the pain of a bear market.

Perhaps you're one of the 14% of respondents to my informal survey (bottom of that page) who eats risk for breakfast, sleeps well and never takes anything stronger than vitamin C. In that case, your asset allocation should be based on factors other than short-term volatility. But if you're one of the 84% of respondents who need some bonds to help you sleep, or the 2% who want nothing to do with stocks in retirement, your tolerance to short-term volatility should be a factor in your asset allocation. That stock allocation may be smaller, for example, than one that would optimize a safe withdrawal rate over the long run.

I apologize that this post and the previous are somewhat redundant, but it's an important point and sometimes a real-life scenario is more convincing.


  1. Excellent post Dirk. I think one's comfort level with his/her allocation is one of the most important investing decisions one can make, and I think it becomes even more critical obviously as one's assets hopefully rise to their highest point near retirement. Finding this level (whatever it is) and sticking to it are critical. Either not finding the correct level or changing it midstream can be killers to one's portfolio.

    I fear for my friends who are near retirement, and have an allocation that they are unaware could set back their retirement date substantially. For example, target date funds (TDFs) which vary fairly dramatically in their recommended allocations, only take one variable into account (years to retirement). TDFs may be better than nothing, but they are not personalized to each individual's unique situation. I wish there were funds that were based on ones funded ratio (current assets compared to the PV future liabilities). As Wm Bernstein says in one of his books when you have won the game why keep playing. I couldn't agree more. Risk is a part of all aspects of life. In terms of one's financial risk I think few things are more important than recognizing and understanding the risks you are taking and deciding which risks you can accept and which you cannot. As always Dirk thanks for making me think. Brad

    1. The above is one of the few times I have seen this issue expressed intelligently (that "acceptable" risk is not simply a function of age). "Current assets compared to the PV of future liabilities" is a good way of expressing that idea. It leads to an explanation of the reason that the rich get richer-- they can accept the greater risk that in the long term produces greater returns. In this context, retirement age is perhaps the one age related issue that has some validity in terms of risk. People who are still working have a greater ability (by continuing to work) to control the risk of investment loss. There's nothing quite like losing 30% or 50% of your portfolio right after retiring.

    2. Nothing, indeed.

      I think some academic research tends to ignore the behavioral issues and makes recommendations that work for people with high risk tolerance but not for others.

      Another way of looking at what you describe is that retirees have depleted their human capital.

      Thanks for your comments!

  2. Investors can invest however they want, but if he lost 50% in 2008, one of two things is going on. He is overstating his losses - I see this all the time with clients!!! He loses 25 to 30% and characterizes it as 50%. Or, he was, clearly, far more aggressive than he should have been. If this is how he responded to what happened, then it is no way to live your life, either pre or post retirement.

    1. Well said, Mark, especially your last sentence.

      I have a client who apologizes for being too conservative, compared to the common wisdom of financial planners. But he understands his risk tolerance and invests accordingly. That's not poor planning, that's just healthy.

      It is possible that Cowles exaggerated. I see people do that, too. Heck, I probably do it unintentionally. And I see them understate losses. But if he did only lose 30% of his retirement funds at age 70, that's still too much.

      Thanks for adding to the discussion, Mark.

  3. Bernstein's booklets, “Deep Risk,” comes to mind here Dirk.

    – Risk = [size of the loss] X [duration of the loss]
    – Shallow Risk = loss of real capital that recovers relatively quickly
    – Deep Risk = permanent loss of real capital
    – A person may experience shallow risk, but put themselves into deep risk by buying high and selling low … thus, permanently losing the capital value they would have had if they had stuck it out (goes to the two kinds of risk in the first booklet).
    – You can NOT avoid deep risk no matter how disciplined or prudent you are (think 2008).
    – Gold is NOT an inflation hedge … he explains it works during DEflation (falling general price levels). [ Two other risks he discusses are confiscation and devastation].
    – Not all 4 of these risks are equally possible … he discusses probability and how this may impact allocation to address those that concern you.

    I think it is important to better understand what kind of risk one is taking and talking about and how to address both shallow and deep risk - especially how NOT to turn shallow into deep risks!

    Both of your posts on this are great discussions Dirk!

    1. Bernstein rocks! :-)

      Thanks, Larry. His comment on not turning shallow risk into deep risk always takes me back to sustainable withdrawal rates and constant-dollar spending and your discussions about dynamic withdrawals. If you want to turn shallow risk into deep risk, keep spending the same amount when your portfolio value declines.

      Nice addition to the discussion. Thanks!

  4. " ... keep spending the same amount when your portfolio value declines." A great statement Dirk!

    Indeed, the common mistake is to set spending on what I call "peak" portfolio values - and then have to adjust spending as values declines (but happily adjust it back up when values go up). This comes from the economist approach (my words describing general origin of the concept) to retirement income.

    Instead, by way of my "Ocean Analogy" (which Kitces loves too), set spending on statistical expected "trough" portfolio values ( ).

    Trough values allow spending to remain the same for most market misbehaviors. When the markets behave most of the time, it allows for some discretionary spending. When markets misbehave occasionally, a trough (or what you like to call a floor) spending is already in place by design. Thus one has important expenses covered with some flexibility on the up side instead of putting their whole plan at risk of failure (running out of money later because the set their plan up with no flexibility).

    The above combined with separation of money uses, commonly called buckets, also helps. Having money in a Distribution Reservoir with no stock market exposure allows the check book to get money each month (based on trough values above). The Long Term Portfolio is exposed to market ups and downs proportional to the exposure designed in that portfolio allocation. Good design and mathematically set decision rules for overall management help here too.

    Of course, the above relates to the portfolio component for retirement income. That income is added to Social Security and/or pension income to form the total income available. And, if one still wants more fixed income, purchasing a SPIA or DIA works. However, that income typically is subject to loss of purchasing power, especially if purchased at younger ages when more time for inflation to have its' effect.

    So often the risks are discussed as the bogie man (emotion) in the room. Rational approaches shine light to get control over emotion. It is the emotions that torpedo the success of portfolio retirement income for people who haven't taken the time to insulate their plan from emotion. The emotions are always there - we're hard wired to be that way. Actions though should be based on reason - and that is where Bernstein rocks!

  5. Am always surprised by such thoughts; and the solution proposed of reducing stock component in a 'Constant Asset Allocation' Plan.
    The media goes to great length to provide us with valuation measures.
    Higher the valuations the more 'downside risk' and the lower expected returns, the reverse being true with lower valuations.
    As Wade Pfau has written, in retirement with few years left to recover we need to look at valuations and be particularly aware of the sequence of returns risk.
    Why do so few retirees not advocate using Dynamic 'Variable Asset Allocation' Plan guided by valuations?
    We retirees need to be ulra-cautious.

    1. Yes, we do.

      I think everything should be dynamic. Your life expectancy, risk tolerance, spending needs and portfolio balance will change throughout retirement. Periodically (annually is good) you need to adjust all of those things to a new reality instead of basing your decisions on conditions that existed back when you retired.

  6. Dirk,

    I would be interested in your and your readers thoughts on the flip side of risk. Most of the financial literature I have read talks about the negative aspects of risk,and maybe rightly so But it would seem to me that risk is the source of any excess returns relative to other less risky assets, and therefore it is not all bad. I think I read, and agree with one should take the least amount of risk needed to achieve one's financial goals. I guess what I am getting at is could we discuss other ways to mitigate and possibly harness risk so as to take advantage of risk. You've mentioned dynamic updating of one's withdrawal percentage for income during retirement as one way to mitigate risk, and of course getting one's asset allocation correct is another way. I just think it is interesting that most talk about the behavioral aspects of investing, I have seen, deal with the downside of one's behavior as it relates to risk such as buying high and selling low, and how this can result in permanent portfolio loss or deep risk. Why is there not as much talk about the positive opportunities risk presents such as having a disciplined re-balancing approach that creates a positive behavioral outcome or a permanent portfolio gain or high risk (vs deep risk). Just curious what other sorts of suggestions you and your readers might have to give risk a little more favorable slant. Fire can be both bad and good. I hope this makes sense. I know it is a little disjointed. Thanks, Brad.

    1. Sorry, Brad. I should have entered it here, but my response follows.

  7. Brad, you ask the most interesting (and consequently most fun to answer) questions!

    Your question makes perfect sense, but I totally disagree with your premise. If "most of the financial literature [you] have read talks about the negative aspects of risk", then you need to get out more!

    Not only is risk not all bad, it is essential for funding retirement. It’s hard to imagine that anyone could save for retirement with only risk-free investments, fund it with risk-free investments, and maintain their pre-retirement standard of living throughout a long retirement. Our retirement system forces us to rely heavily on investment returns and, as you suggest, you don’t get that without taking risk. I’d be surprised if you haven’t read that in a few places, at least.

    That doesn’t mean all risk is good, and most retirement writers, myself included, try to steer you away from bad risk while making sure you understand that avoiding all risk is probably not going to work very well.

    What is bad risk? Bad risk would include risk that you don’t need to take (Bernstein: when you win the game, stop playing). It would include risk that is beyond your risk capacity. (Leading up to the 2007 crash, many financial institutions took far more risk than they could afford and a loss meant bankruptcy for Lehman and government bailouts for several others. It meant bankruptcy for my friend who had tremendously leveraged real estate investments.)

    Bad risk includes risk for which you don’t get compensated, for example, sequence of returns risk and non-systematic stock risk. You mention reducing risk through a diversified asset allocation. Diversification reduces non-systematic risk (the risk that a single company failing will destroy your returns) but not systematic risk. Systematic risk (the risk of the total market) is good risk for which you are compensated by the market. Since non-systematic risk is avoidable, no one will pay you to take that risk. Not good. Financial writers don’t say you should avoid market risk, they say you should diversify your investments so you can earn the rewards of good systematic risk and avoid the risk of uncompensated non-systematic risk.

    Bad risk, as I mentioned in the post, also includes taking so much risk that you can’t sleep at night (risk tolerance). I'm not suggesting that you don't take risk; I'm suggesting that you don't take so much that it makes you miserable.

    Writers say you should avoid taking more risk than you can tolerate or more than you can afford. They say you shouldn’t take risk that you don’t need. But nearly all say that you should take good risks and, in fact, that you really have to in order to succeed.

    I think most of what we write about is “other ways to mitigate and possibly harness risk so as to take advantage of risk.” If you have something specific in mind, I'd be happy to write about it.

    Sounds like you’re only seeing the “glass-half-empty” side of those articles!

    As always, thanks for writing, Brad.

  8. Thanks Dirk for your response. Maybe I exaggerated a little for effect with my comment about most of financial literature emphasizing the negative aspects of risk, I certainly did not mean you. I guess I was just thinking in addition to getting rid of risk you are not compensated for or don't have the ability to take what other sorts of suggestions are there for taking advantage of risk. What came to my mind when writing the above comment was to turn the negative effect of buying high and selling low on its head with positive behavior by being able to re-balance by buying low and selling high. Of course this gets to the issue of how low is low enough to buy. During the recent drop in August I almost pulled the trigger, but could not buy. With a 25/75 stock/bond allocation the typical re-balance trigger of 5% change in allocation percentages rarely gets triggered (20% loss), and maybe that is the way it should be. Maybe the most efficient re-balancing is every 3-4 yrs. I realize my 25% stock allocation is below the recommended for maximum withdrawal efficiency (35-40%) so I am trying to make my portfolio otherwise as efficient as possible. I must admit I do have this internal debate going on sometimes though where I intellectually realize that 35-40% is a more efficient long term stock allocation (for maximizing withdrawals), but emotionally I'm not quite there yet. Thus I am at 25% though I hope I may not always be here so timid. Thanks for your insights as always Dirk. Brad

  9. I wasn't referring to myself, either. I was speaking on behalf of retirement writers and researchers in general.

    William Bernstein once wrote that there is a strong argument that "never" is a good frequency for rebalancing. Still, I rebalance when an asset class gets an absolute 10% off target, which is to say "rarely."

    Regardless, Joe Tomlinson wrote an excellent piece last week at Advisor Perspectives explaining that our estimates of future market returns are so shaky that trying to tweak an asset allocation by even 10% is pure guesswork. I'm writing more on Joe's post and basically arguing that his estimate of our uncertainty is probably optimistic.

    Rebalancing will help buy low and sell high if you get lucky with your timing, but the deck is stacked against you after retirement when you start selling from your portfolio and effectively dollar-cost-average in reverse (sell more shares when prices drop).

    It sounds like you're searching for a way to make stock investing less risky (who isn't?), but the reality is that stock investing is always risky and the best you can do is not do anything silly to add to that risk.

    The suggestions for taking advantage of risk are all about only taking the good ones.

    Still, it's usually necessary. I don't know how to put market risk in a better light than to say that for most of us it is essential.

    Thanks, Brad.

  10. Dirk, thanks for the Joe Tomlinson article. I enjoyed it, and look forward to yours. Investing is a whole different ball game before and after you're retired. So much to learn. Brad

  11. Dirk, one other article that helped me, and maybe others. Brad

    1. Thanks. I disagree with his first characterization that diversification minimizes regret, but I think he gets it right the second time when he talks about "trade-offs." Diversification is a compromise. To minimize the regret of missing out on huge gains for example, you would invest 100% in risky stocks, maximizing possible regret for huge losses.

      And frankly, the idea that "investing comes down to regret minimization" totally eludes me, though I think asset allocation often does.

      But this behavioral aspect isn't the most important characteristic of diversification. Diversification actually reduces non-systematic risk in a portfolio, know matter how you feel about it.


  12. Dirk,
    Wonderful words of advice! Thank you.
    Chuck Burgner,
    Davenport, iowa

    Dirk Cotton: NOV 15 2015

    "I think everything should be dynamic. Your life expectancy, risk tolerance, spending needs and portfolio balance will change throughout retirement. Periodically (annually is good) you need to adjust all of those things to a new reality instead of basing your decisions on conditions that existed back when you retired."