To Be or Not To Be: Orthodox or Unorthodox Asset Allocation

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
cnh
Posts: 19
Joined: Mon Dec 09, 2013 7:27 pm

To Be or Not To Be: Orthodox or Unorthodox Asset Allocation

Post by cnh » Mon Dec 09, 2013 7:39 pm

I’m a strong proponent and practitioner of index investing, and over the years I’ve adhered to the standard portfolio allocation orthodoxy. That said, I’ve been reading the investing literature much more extensively since 2008-2009 than in the past and, using multiple tools, back-testing several different approaches to portfolio allocation. This has caused me to question both the standard, age-based and the own-the-whole market approaches to portfolio allocation.

For example, using multiple tools one can compare an orthodox 3-fund 30/70 index portfolio (21% total US stock market, 9% total international stock market and 70% total us bond market) to an unorthodox 4-fund index 30/70 portfolio (10% small-cap value, 10% REIT, 10% emerging market and 70% 5-yr treasuries), and the unorthodox portfolio (a bit like the Larry Swedroe Minimize FatTails Portfolio) looks superior over 10-, 20- and 30-yr time frames and multiple rolling periods. Over the period 1972-2012, the unorthodox portfolio produces the following results: CAGR, 10.3%; SDEV, 7.2%; worst year, -3.67%; and Sharpe Ratio, 0.73. Over the same period, the comparable orthodox “total market” portfolio produces these results: CAGR, 8.8%; SDEV, 7.2%; worst year, -8.21%; and Sharpe Ratio, 0.52.

I’m well aware of the cautions against “past performance” and relying on back-testing; still, the proponents of indexing and total market ownership, including Bogle, use historical data and implicitly back-testing to bolster the argument.

So my discussion prompt is this: Why own the whole market, when owning more uncorrelated slices of the market appears to produce better results (i.e. higher returns, comparable volatility, lower draw-downs and better compensation for risk)?

User avatar
G-Money
Posts: 2867
Joined: Sun Dec 09, 2007 7:12 am

Re: To Be or Not To Be: Orthodox or Unorthodox Asset Allocat

Post by G-Money » Mon Dec 09, 2013 10:27 pm

1. Because correlations shift.
2. Because past performance is not a guarantee of future results.
3. Because you may be tempted to switch to a more conventional (or, at least, a different) allocation when your unconventional portfolio underperforms.
Don't assume I know what I'm talking about.

User avatar
Index Fan
Posts: 2555
Joined: Wed Mar 07, 2007 12:13 pm
Location: The great Midwest

Re: To Be or Not To Be: Orthodox or Unorthodox Asset Allocat

Post by Index Fan » Mon Dec 09, 2013 10:33 pm

I'm convinced that past performance is historical- particular outcomes of discrete events in the past, and we rely upon it as a window to the future at our own peril.
"Optimum est pati quod emendare non possis." | -Seneca

User avatar
baw703916
Posts: 6679
Joined: Sun Apr 01, 2007 1:10 pm
Location: Seattle

Re: To Be or Not To Be: Orthodox or Unorthodox Asset Allocat

Post by baw703916 » Mon Dec 09, 2013 10:37 pm

Index Fan wrote:I'm convinced that past performance is historical- particular outcomes of discrete events in the past, and we rely upon it as a window to the future at our own peril.
If you have a better window on the future then by all means use it.
Most of my posts assume no behavioral errors.

User avatar
Index Fan
Posts: 2555
Joined: Wed Mar 07, 2007 12:13 pm
Location: The great Midwest

Re: To Be or Not To Be: Orthodox or Unorthodox Asset Allocat

Post by Index Fan » Mon Dec 09, 2013 11:32 pm

baw703916 wrote: If you have a better window on the future then by all means use it.
I don't. That's why I diversify across multiple asset classes and forget about it. But by all means, drive looking through a rear-view mirror if you like ;)
"Optimum est pati quod emendare non possis." | -Seneca

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: To Be or Not To Be: Orthodox or Unorthodox Asset Allocat

Post by SimpleGift » Tue Dec 10, 2013 1:11 am

cnh wrote:Why own the whole market, when owning more uncorrelated slices of the market appears to produce better results (i.e. higher returns, comparable volatility, lower draw-downs and better compensation for risk)?
Incorporating "uncorrelated slices of the market" into one's portfolio has been the holy grail of portfolio design for decades. Many of us were first persuaded by Roger Gibson's classic book, Asset Allocation, when it was first published in 1990, and have been investing in "alternative asset classes" (REITs, small caps, emerging market stocks, commodities, etc.) ever since. The problem today is that, due to globalization and financial integration of the world economy, many of these alternative assets classes have become much more highly correlated within the last decade (chart below). As a result, they no longer appear to be providing the degree of diversification today that backtested results would indicate they should.

Image
Source: Capital Spectator

In my view, a much more intriguing approach to portfolio diversification today for investors who are just starting out is diversification by "risk factors," such as size, value, momentum, profitability, etc. Larry Swedroe has been writing and posting about this approach for some time. These risk factors seem to be fairly independent of one another and, so far at least, unaffected by global financial integration (chart below).

Image
Source: French Global Factors

PS. Welcome to the Forum!
Cordially, Todd

Call_Me_Op
Posts: 7029
Joined: Mon Sep 07, 2009 2:57 pm
Location: Milky Way

Re: To Be or Not To Be: Orthodox or Unorthodox Asset Allocat

Post by Call_Me_Op » Tue Dec 10, 2013 7:07 am

Back-testing can be hazardous to your wealth.
Best regards, -Op | | "In the middle of difficulty lies opportunity." Einstein

User avatar
nisiprius
Advisory Board
Posts: 36640
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: To Be or Not To Be: Orthodox or Unorthodox Asset Allocat

Post by nisiprius » Tue Dec 10, 2013 10:32 am

The problem is that financial data is terribly intractable, and estimates just don't want to settle down quickly over reasonable periods of time. Thus, in my post Endpoint sensitivity on "historic" data, even something as basic as "the total return of the U.S. stock market" over a period as long as 80+ years can vary by two percentage points--9% to 11%--with seemingly negligible changes in the choice of endpoints.

Image

Otar's chart is good to keep in mind. Now, I think Otar actually believes in the existence of consecutive straight-line segments, "secular" markets one after another. I am very skeptical about that. But his chart does underline the nature of market fluctuations, and all market data is like that. One time is not like another, "times" can last 10 years, and your 20- and 30-year averages mean very little--they are not samples of 30 independent things, they are samples of a couple-three of those line segments.

The odd thing is that there really is convincing data that investors in a fund consistently underperform the fund--that is to say, the fund's performance data, which is the results what would have been gotten by an investor who stayed the course. ALL of the historical average data we use assumes that we stay the course.

Of course, the double whammy is that once you go back more than a few decades it becomes increasingly hard to believe that you are looking at the same thing. The stock market of the 1920s, with retail investors just coming into it for the first time, no SEC, no Investment Company Act of 1940, no mutual funds, no 401(k)'s... stock tickers and quoted in eighths of dollars... this is the same thing as the stock market of today? Quantitatively, numerically, do-math-on-it the same thing? Really?

So I don't know what to say. There's very little point in using any kind of past data unless you are prepared to take your best shot, do something sensible, and then FREEZE. Yes, sit out the next decades NOT taking advantage of improvements in this, that, and the other thing. That's humanly impossible to do, I think, so I hope that very slow, gradual changes don't do too much harm...
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
baw703916
Posts: 6679
Joined: Sun Apr 01, 2007 1:10 pm
Location: Seattle

Re: To Be or Not To Be: Orthodox or Unorthodox Asset Allocat

Post by baw703916 » Tue Dec 10, 2013 11:49 am

Index Fan wrote:
baw703916 wrote: If you have a better window on the future then by all means use it.
I don't. That's why I diversify across multiple asset classes and forget about it. But by all means, drive looking through a rear-view mirror if you like ;)
And how do you know if the risk level is appropriate without basing the conclusion on any past data?
Most of my posts assume no behavioral errors.

IlliniDave
Posts: 2294
Joined: Fri May 17, 2013 7:09 am

Re: To Be or Not To Be: Orthodox or Unorthodox Asset Allocat

Post by IlliniDave » Tue Dec 10, 2013 12:18 pm

I don't see anything necessarily wrong with the "unorthodox" portfolio approach. I'm not sure you can say with any certainty that it will enjoy the same performance edge over the "orthodox" going forward, but it probably wouldn't be disastrous either. Starting with the unorthodox, If you take 30% out of the bond pot and put it in total stock market(s) you're close to what many "tilters"/"slice-and-dicers" do. I guess they would be semi-unorthodox (or semi-orthodox) in your vocabulary.
Don't do something. Just stand there!

heyyou
Posts: 3130
Joined: Tue Feb 20, 2007 4:58 pm

Re: To Be or Not To Be: Orthodox or Unorthodox Asset Allocat

Post by heyyou » Tue Dec 10, 2013 1:48 pm

Less obvious here are the messages of Live Below Your Means for career periods of time, Invest Your Savings, Stay the Course (do not sell out during down markets), and to carefully watch investing expenses. Total Market investing (with a mix of bonds) is simple, cheap, and easy for everyone, requiring little knowledge of statistics. When offering suggestions to the public on how to acquire a retirement nest egg, plain and simple is the best because it is adequate for all applicants.

Your proposal is only better for a very narrow slice of the population. As Taylor has wisely said, "There are many roads to Dublin." Successful entrepreneurs often have a history of first failing, but continuing to learn, then finding success. Actively managing a business is not for everyone. Some workers acquire rental properties until they can retire on that income. Having a side job of being a handyman is not for everyone. To own your choice of volatile equity assets that do not track the broad markets, takes some depth of knowledge and great patience. Unorthodox AA is fine for those who are suitable, but not for many.

Note that you have mentioned the expected higher returns. That desire for more than what is easily available has often caused problems in the past. Orthodox is good enough.

sdrone
Posts: 108
Joined: Wed Jun 10, 2009 11:17 pm

Re: To Be or Not To Be: Orthodox or Unorthodox Asset Allocat

Post by sdrone » Tue Dec 10, 2013 4:26 pm

"an unorthodox 4-fund index 30/70 portfolio (10% small-cap value, 10% REIT, 10% emerging market and 70% 5-yr treasuries"

Thanks for the portfolio idea. I love plugging this stuff into spreadsheets and playing with the results.

cnh
Posts: 19
Joined: Mon Dec 09, 2013 7:27 pm

Re: To Be or Not To Be: Orthodox or Unorthodox Asset Allocat

Post by cnh » Tue Dec 17, 2013 9:45 pm

G-Money wrote:1. Because correlations shift.
Agreed, but over 10-, 20- and 30-year periods with correlations shifting back and forth, multiple market cycles completing and the results being what they were, it seems a little flippant to dismiss the analysis and the question. Apologies, but I was hoping for dialogue and not dogma.

cnh
Posts: 19
Joined: Mon Dec 09, 2013 7:27 pm

Re: To Be or Not To Be: Orthodox or Unorthodox Asset Allocat

Post by cnh » Tue Dec 17, 2013 9:54 pm

IlliniDave wrote:I don't see anything necessarily wrong with the "unorthodox" portfolio approach. I'm not sure you can say with any certainty that it will enjoy the same performance edge over the "orthodox" going forward, but it probably wouldn't be disastrous either. Starting with the unorthodox, If you take 30% out of the bond pot and put it in total stock market(s) you're close to what many "tilters"/"slice-and-dicers" do. I guess they would be semi-unorthodox (or semi-orthodox) in your vocabulary.
Great point! Still, running the numbers on a 30/10/10/10/40 portfolio as you suggest yields CAGR, 10.9%; SDEV, 11.3%; worst year, -21.3%; and Sharpe, 0.55. Targeting a comparable return with an index portfolio of 13% SCV, 13% REIT, 13% EM and 61% 5-year Treasuries, we get CAGR, 10.9%; SDEV, 8.4%; worst year, -7.7%; and Sharpe, 0.72. I'm still left wondering why the current portfolio orthodoxy persists.

lazyday
Posts: 3297
Joined: Wed Mar 14, 2007 10:27 pm

Re: To Be or Not To Be: Orthodox or Unorthodox Asset Allocat

Post by lazyday » Wed Dec 18, 2013 12:43 am

Bogle and maybe Sharpe might say something about how everybody together owns the market. What if everyone tried to get risk factor diversification, by overweighting S, V, etc?

Maybe a lot of money has been trying over the last decade or so. Maybe some risk factors are now more expensive than the market. For example, some think S is expensive today.

Larry has posted about a paper, I think from DFA, maybe Davis, showing IIRC that it hasn't made sense to attempt to time V or S. You might switch when expensive, but still likely to regret the switch. I was unsuccessful finding a copy of this paper online.

I think both sides of this debate have some good cases.
What I do personally is change my investing strategy every so often, based in large part on what I think is cheap at the time. There were times I favored SV. I probably will again. I've never done total market.
I do not suggest this approach for anyone else. Just giving one person's answer to the dillema, and apologise that this and my posts on this forum in general are a bit hypocritical. For an argument against thinking like I do, see the relevant section of W Bernstein's Investor's Manifesto, or any papers or books on investor behavioral errors. Normal people tend to buy high and sell low. So far, my abnormal level of social need to join the crowd has outweighed my mediocre abilities to evaluate valuations or expected returns.

Post Reply