What do you think of our site upgrade?
Hosting an industry conference? Ask us about including it in this ticker?

Asset Allocation’s Greatest Failure: Short-Term Investing

June 23
00:03 2015

(The following is the third of a special five part series meant to be shared by professionals and non-professionals alike. This particular series covers
only one of the 7 Deadly Sins Every ERISA Fiduciary Must Avoid.)

If one fails to use asset allocation, is one left with only “hunches and sure fire bets”? This sounds like the makings of a hypothesis, and if there’s one thing we know about hypotheses, they can be tested via experiment.

Earthrise_Usual“The test of the validity of any idea is experiment.” So said Nobel laureate Richard Feynman, in his famous Lectures on Physics at Caltech (Sept. 29, 1963). More fully, Feynman explained how scientific laws and theories are developed in this way: “First, we make a guess. Then we compute the consequences of the guess. Then we compare the consequences to experiment. If it disagrees with experiment, it’s wrong. In that simple statement is the key to science. It doesn’t make a difference how beautiful your guess is.  It doesn’t make a difference how smart you are, who made the guess, or what his name is. If it disagrees with experiment, it’s wrong.”

Is Asset Allocation Best Used for the Short-Term or the Long-Term?
It’s no wonder there’s confusion when it comes to asset allocation. Is asset allocation a long-term (a.k.a. “strategic”) tool or a short-term (a.k.a. “tactical”) tool? Or is it both? To find the answer, one need look no further than the definitive text for those studying to be Certified Financial Planners:

“Strategic Asset allocation is a method of allocating portions of a portfolio to various asset classes based on long-term capital market expectations… This strategy, therefore, is not a buy-and-hold strategy, since rebalancing will become necessary as the portfolio allocation differs from the strategic allocation over time.”1

“Tactical asset allocation is a more dynamic approach than strategic asset allocation. It involves the continual change between asset classes based on perceived investment opportunities in certain asset classes. In effect, it is a form of timing the market, but within the setting of a portfolio.”2

While asset allocation is taught as a tool for both the short and the long term, studies remain inconclusive as to its actual impact on investment performance. It is clear asset allocation does not explain “93% of investment performance,” a claim falsely attributed to the BHB study as we discussed in the previous installment. “That particular quote gets filtered by the common focus on return,” says David M. Williams, Director of Planning Services at Wealth Strategies Group in Cordova, Tennessee. “Investors and Investment Managers concentrate on return received, and feel something is wrong if they receive something lower than the average annual return. The success or failure is judged by 1-year periods.”

Recent history (q.v., 2002 and 2008/2009) has shown us the danger of relying on asset allocation in too short a time period (e.g., one year). Many cite overconfidence in long-term data gave rise to the disappointment in asset allocation following these last two economic crashes. Froilan Rellora, Chief Investment Officer at Catalina Asset Management in Mesa, Arizona, says, “Regardless of market conditions, asset allocation strategies focus on long-term returns, which didn’t take the market crashes of 2002 and 2008/2009 into consideration. Over reliance on the long term meant that in order to break even from the 37% loss in the 2008 calendar year, investors needed a 59% increase due to the asymmetry of gains and losses.”

It is now commonly understood asset allocation cannot be relied upon for extremely short-term periods. On the other hand, some see ignoring the practical realities of short-term volatility to also represent a danger. “We have found that asset allocation can provide a predictable range of returns,” say Williams. “While asset allocation is often viewed based on 30-year statistics for average return, standard deviation, and correlation, we feel that on shorter terms – 3-10 year rolling time periods – these statistics vary significantly.”

Therein lies our first hypothesis that we can test experimentally. But first, let’s make sure we all understand the nuts and bolts of the asset allocation process.

How Asset Allocation is Supposed to Work
According to a popular CFA text, “We can view the process of asset allocation as consisting of the following steps: 1) Specify asset classes to be included in the portfolio…; 2) Specify capital market expectations…; 3) Derive the efficient portfolio frontier…; and, 4) Find the optimal asset mix.”3 Among the specific tasks in these steps, the CFA applicant is told to use “both historical data and economic analysis to determine your expectations of future rates of return over the relevant holding period on the assets to be considered for inclusion in the portfolio,”4 find “portfolios that achieve the maximum expected return for any given degree of risk,”5 and select “the efficient portfolio that best meets your risk and return objectives while satisfying the constraints you face.”6

Consistent with BHB, this process does not guarantee the outcome will yield the best investment returns. Unfortunately, over the years, whether through practitioners or retail investors themselves, there’s a common misconception asset allocation can be used to produce the best investment returns and, at the same time, dampen return volatility. It’s really intended to operate in the reverse fashion – it tries to find the best investment return given a specified volatility.

Nonetheless, given the popularity of this fallacy regarding asset allocation, wouldn’t it be convenient if one could easily summon up an illustrative example that proves the point. Such a case study would be especially helpful when trying to explain the myth to those who insist asset allocation can be used to produce better invest returns. The challenge, unfortunately, is trying to find that particular scenario which most convincingly exposes this fairytale.

Rest easy. did the work for you. Without getting into the details of the mathematics (i.e., “mean-variance optimization”), we carried out an experiment in the spirit of Feynman’s earlier description. In practice, believers of asset allocation have a choice: either use historical data to project future returns or use their own estimates. Since the latter reeks of the alchemy of market timing (which most serious investors and nearly all academics agree has all the credibility of a perpetual motion machine – it just doesn’t work), this limits the believer of asset allocation to using historical data. In fact, most of the asset allocation software used by professionals and non-professionals (and, dare we say, robo-advisors) employs this approach.

Does Asset Allocation Work in the Short-Term?
We even concede to the use of pie charts to explain our test. The three pie charts shown here represent the three different asset allocations we must choose from. In the real world, we would not be limited to only three choices, but for the sake of brevity, this article will focus on only these three choices. In the case described below, we used data provided by Ibbotson Associates – the industry’s favorite source for data like this. Solely for the purpose of adding a bit of fun, we’ll relate the test in terms of a hypothetic story from the point of view of a recent college graduate during the late 1960s. (If it will make you feel better, we will refer to our subject as “Ben.”)


Here’s the test: On December 31, 1968, perhaps inspired by the Christmas Eve reading of the first ten verses of Genesis by the three astronauts of Apollo 8, Ben concludes America is great and he must invest in it. Ben just graduated from a leading academic institution and he’s learned all about the capital asset pricing model and the efficient frontier. He also knows there’s this new movement to apply the concept not merely to individual securities, but to entire asset classes. As a result, he decides to allocate his assets among the three standard asset classes – stocks, bonds, and cash. We see these options represented by the three portfolios in the above graphic: Portfolio A, Portfolio B, and Portfolio C.

Ben’s not quite sure where he’ll be in his thirties, (you see, there’s this girl Elaine,… but that’s another story). In either case, just so he doesn’t have to commit to tying up his money forever, he decides to invest with a ten year time frame. At the close of the markets on December 31, 1968, he runs his mean-variance optimization for the previous ten years and discovers Allocation “C” offers the best performance, After all, thinks Ben, that’s the ultimate goal of asset allocation – superior long-term performance – otherwise, no one would buy into it.

So, on January 1, 1969, Ben invests his portfolio as guided using Allocation “C.” A decade later, on December 31, 1978, Ben looks at his return only to learn Allocation “C” was the worst performer for the decade. Disgusted, he goes back to his optimizer and it reveals Allocation “A” actually performed the best during the last ten years. Ever the obedient one, on January 1, 1979, Ben reallocates his assets per the instructions of Allocation “A.” What does he find out on December 31, 1988?

Unfortunately for Ben, Allocation “A” performed the worst in the ten years ending December 31, 1988. So, as a result of using his optimization software, he ended up choosing the worst performing asset allocation in each of the two decades. Never the one to be discouraged, though, Ben runs the software again, this time on his brand new IBM PC. Again, it was Allocation “C” that had the best performance while he was busy investing in Allocation “A.” Still, he diligently reallocates and waits another ten years.

On the cusp of a new millennium, Ben receives great news. When the final bell rings to close the year of 1998, he sees Allocation “C” topped them all! Confident he’s finally got this asset allocation thing down, he thrusts full throttle into its use for another decade. Adding fuel to his trajectory is the BHB study, which everyone who’s anyone is saying has proved that asset allocation is more important than security selection.

Alas, December 31, 2008 does not bring good news to Ben. Once again, Allocation “A” out-performed all the others. Worse, with a maximum exposure in equities, his investments suffer incredibly. Contrary to the advice given to him immediately after college graduation to invest everything in “plastics,” Ben knew the wisdom of diversification as preached on the altar of asset allocation. He sits forlornly in January 2009, the market continuing to drop like a knife, wondering what could have gone wrong.

He turns one more desperate time to his formally trustworthy optimization software. In this effort, however, he doesn’t use that past ten years historic data. Instead, he goes all the way back to when he started – a full 40 years’ worth of data. He’s using a web-based application now. He enters in his parameters and waits for the answer (the internet appears to be slower than normal on this day).

When the screen finally reveals its answer, Ben’s heart sinks. Until this moment, the optimizer never even told him to consider the best performing portfolio allocation over that entire forty year span: Allocation “B.”

Asset Allocation Flunks the Short-Term Test, So Now What?

There are two types of experiments: one which proves a hypothesis can work sometimes; and, one which proves a hypothesis works all the time. Our test falls in the latter category. As a result, to reject the hypothesis, we need find only one case, no matter how constrained, that shows the hypothesis fails. It does no good to say “what about the other asset allocation possibilities,” for the idea of asset allocation implies, no matter what given set of portfolios – finite or infinite – the ability to identify their place on the (unchanging) efficient frontier will determine whether one allocation is better than another allocation. The failure occurs, oddly, not because of the formulae, but because of the data. It turns out historical data changes over different (especially short) time periods. That produces different efficient frontiers; hence, different optimal asset allocations.

There’s a good reason why we call tactical (i.e., short-term) asset allocation “market timing” – Neither one works reliably. There’s also a good reason why the SEC demands portfolio managers and mutual funds, when discussing investment performance, must always add the disclaimer “past performance does not guarantee future results.” That’s precisely what this experiment proves, and why short-term asset allocation is doomed to disappoint.

Perhaps our imaginary Ben would have fared better had spent some time studying under Yale Emeritus Professor Roger Ibbotson. Ibbotson, who is also Chairman and CIO at Zebra Capital and Founder of Ibbotson Association, says, “Optimizers can be dangerous in practice because they’re so sensitive to input. The standard deviations and correlations are more predictable, but the standard error goes down with time, not the number of observations. That’s why it’s better to use longer term data. If you take individual decades as an input, they can be very distorted.”

Ben learned that lesson first hand. Today, having been burned several times by using asset allocation for short-term optimization, he only looks upon it as a long-term tool. And the Yale professor would agree with Ben. “It’s very hard to estimate short-term asset allocation trends,” says Ibbotson, “but I would still use long-term inputs.”

We haven’t given up on asset allocation yet. Ibbotson’s comment suggests yet another test – the long-term test. Will that prove the savior of asset allocation? Or will it provide the final nail in the coffin of asset allocation?

You’ll find the answer in Part IV of this series.

Part I: 7 Deadly Sins Every ERISA Fiduciary Must Avoid: The 5th Deadly Sin – Asset Allocation
Part II: How’d an Innocent Fiduciary Like You End Up Asset Allocating?
Part III: Asset Allocation’s Greatest Failure: Short-Term Investing
Part IV: Why Asset Allocation Doesn’t Matter In The Long Run
Part V: The Hows, Whys, and Right and Wrong Way to Use Asset Allocation

Are you interested in discovering more about issues confronting 401k fiduciaries? If you buy Mr. Carosa’s book 401(k) Fiduciary Solutions, you’ll have at your fingertips a valuable reference covering the wide spectrum of How-To’s (including information on the new wave of plan designs) every 401k plan sponsor and service provider wants and needs to know. Alternatively, would you like to help plan participants create better savings strategies? You can buy Mr. Carosa’s latest book Hey! What’s My Number? How to Improve the Odds You Will Retire in Comfort right now at your favorite on-line or neighborhood book store.

Mr. Carosa is available for keynote speaking engagements, especially in venues located in the Northeast, MidAtantic and Midwestern regions of the United States and in the Toronto region of Canada.

1Dalton, Michael A. and James F. Dalton, Personal Financial Planning Theory and Practice, Dalton Publications, 2000, p. 412
2Ibid., p. 412
3Bodie, Zvi; Alex Kane; and, Alan J. Marcus, Investments, Richard D. Irwin, Inc., 1989, p. 814
4Ibid., p. 814
5Ibid., p. 814

About Author

Christopher Carosa, CTFA

Christopher Carosa, CTFA


No Comments Yet!

There are no comments at the moment, do you want to add one?

Write a comment

Only registered users can comment. Login is sponsored by…

Order Your 401k Fiduciary Solutions book today!

Vote in our Poll


The materials at this web site are maintained for the sole purpose of providing general information about fiduciary law, tax accounting and investments and do not under any circumstances constitute legal, accounting or investment advice. You should not act or refrain from acting based on these materials without first obtaining the advice of an appropriate professional. Please carefully read the terms and conditions for using this site. This website contains links to third-party websites. We are not responsible for, and make no representations or endorsements with respect to, third-party websites, or with respect to any information, products or services that may be provided by or through such websites.