Are Commodities Still a Good Portfolio Diversifier?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
matjen
Posts: 2128
Joined: Sun Nov 20, 2011 11:30 pm

Are Commodities Still a Good Portfolio Diversifier?

Post by matjen » Fri Jan 06, 2017 11:11 am

I found this overview and its conclusions pretty persuasive. Thought others would be interested.

Are Commodities Still a Good Portfolio Diversifier?
Gary Antonacci
The three portfolio studies above, as well as the Levine et al. results over the past 40 years, may be a more accurate view of what to expect in the future. All four show that including commodities in a stock and bond portfolio is no longer beneficial.
http://www.dualmomentum.net/2017/01/are ... folio.html
A man is rich in proportion to the number of things he can afford to let alone.

asif408
Posts: 1836
Joined: Sun Mar 02, 2014 8:34 am
Location: Florida

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by asif408 » Fri Jan 06, 2017 12:01 pm

Thanks for the article link. The part about commodity futures was heavily discussed here a few years ago in an excellent post that included input from Larry Swedroe and Bill Bernstein. I am wary of the commodity futures as a good diversifier but am more convinced that the stocks of commodities producers can provide some level of diversification.

I disagree with the part on correlations. If you look at the correlations of precious metals funds and energy funds (i.e., commodities producers) and compare them to a total stock market fund, you'll notice correlations were lower before 2003. Then between 2004-2011 the correlations increased dramatically. If you look at correlations from 2012-present you'll see that they have fallen again to pre-2004 levels. I think this shows that when there is a financial crisis worldwide like there was in 2007-2009 all equity or equity like assets, including commodities/commodities futures, will perform poorly. However, in more "normal" times I am not so sure there is not some diversification benefit when there are more "normal", localized crises.

Wagnerjb
Posts: 7203
Joined: Mon Feb 19, 2007 8:44 pm
Location: Houston, Texas

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by Wagnerjb » Fri Jan 06, 2017 12:48 pm

I have struggled with the notion of commodities offering both a) equity like returns and b) low correlation to equities. When the 2005 Gorton/Rouwenhorst paper came out, I was suspicious of the conclusions. First of all, their results flew in the face of economics, suggesting that a magic asset class existed with both equity-like returns and low correlation to equities. This is like the fountain of youth for asset classes. Second, I noted that they had constructing returns for historical periods when ordinary investors did not trade commodity futures (and certainly could not own commodity futures funds).

The past dozen years have shown that this asset class probably doesn't have these magic qualities, as demonstrated by the paper linked by the OP. Maybe commodities did indeed offer these great characteristics in the past, and then the market reacted by eliminating the free lunch through aggressive investing in commodities. Or maybe the artificial history of the 2005 paper was the problem. It doesn't matter which, the free lunch isn't here.

Best wishes.
Andy

User avatar
nisiprius
Advisory Board
Posts: 39723
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: Are Commodities Still a Good Portfolio Diversifier?

Post by nisiprius » Fri Jan 06, 2017 1:21 pm

Wagnerjb wrote:...their results flew in the face of economics, suggesting that a magic asset class existed with both equity-like returns and low correlation to equities. This is like the fountain of youth for asset classes...
The article by Gary Antonacci led me to that paper, which I hadn't seen.

Gorton and Rouwenhorst don't talk about low correlation. They say "negative correlation." And notice how their abstract suffers from verb-tense disease!
...Fully-collateralized commodity futures have historically offered the same return and Sharpe ratio as equities. While the risk premium on commodity futures is essentially the same as equities, commodity futures returns are negatively correlated with equity returns and bond returns.
Shame on them!
Fully-collateralized commodity futures have historically offered the same return and Sharpe ratio as equities. While the risk premium on commodity futures is has been essentially the same as equities, commodity futures returns are have been negatively correlated with equity returns and bond returns.
'There, fixed that for ya'."
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
unclescrooge
Posts: 4160
Joined: Thu Jun 07, 2012 7:00 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by unclescrooge » Fri Jan 06, 2017 1:30 pm

I have a small allocations to a commodity ETN, gold ETF and gold miners ETF in my portfolio.

2016 returns were 13%, 10% and 50% (or close to that).

So while the correlation to stocks may be lower than previously reported, the dispersion of returns is still rather high.

Based on that, I think I'll keep them for now.

User avatar
nisiprius
Advisory Board
Posts: 39723
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: Are Commodities Still a Good Portfolio Diversifier?

Post by nisiprius » Fri Jan 06, 2017 2:21 pm

asif408 wrote:I disagree with the part on correlations. If you look at the correlations of precious metals funds and energy funds (i.e., commodities producers) and compare them to a total stock market fund, you'll notice correlations were lower before 2003. Then between 2004-2011 the correlations increased dramatically. If you look at correlations from 2012-present you'll see that they have fallen again to pre-2004 levels. I think this shows that when there is a financial crisis worldwide like there was in 2007-2009 all equity or equity like assets, including commodities/commodities futures, will perform poorly.
There is a lot of sampling error in correlations. It takes a lot of data to pin down a correlation as "really" being in some range (with 1% significance). And there aren't too many non-overlapping periods to look at.
However, in more "normal" times I am not so sure there is not some diversification benefit when there are more "normal", localized crises.
All my investments are terrific during "normal times," because whenever they aren't terrific, I complain about a "ten-sigma event" and throw out the data.
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.

Theoretical
Posts: 1481
Joined: Tue Aug 19, 2014 10:09 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by Theoretical » Fri Jan 06, 2017 2:29 pm

I think Larry Swedroe has made the point that commodities alone don't provide much, but harmonize well with extending bond duration to take more term risk.
Last edited by Theoretical on Fri Jan 06, 2017 2:41 pm, edited 1 time in total.

lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 11:55 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by lack_ey » Fri Jan 06, 2017 2:30 pm

A lot of the "post-financialization" analysis is based on a pretty small sample that's furthermore heavily biased by the effects of the financial crisis and what specifically happened then. This particularly shows up for statistics like Pearson's product-moment correlation over the data, which is a little "wrong" when examining data with non-constant volatility. (The figures get pulled around too much by extreme values, which in this case were extreme downwards values with almost everything falling together.)

So I don't buy as much into inferences along the lines of before vs. after. The "after" data is not especially robust and it's difficult from the data to draw strong conclusions.

Now, the fact that the data corroborates economic sense and intuition, particularly about a crowding of money chasing a limited risk premium and changing relationships, and especially data about correlation between individual commodities... that is more compelling.

You always have to be skeptical of return sources and where they come from, particularly of more opaque sources, even more especially when positive past results of a strategy come significantly from rebalancing.

I don't really know what to expect for future spot return, contango/backwardation, etc. It's really hard to estimate long-term mean return, I would think. And with commodities investing, it's difficult to determine what actually makes sense for relative weightings of individual commodities. All the indexes disagree. Though for that matter, if you're worried about index frontrunning, you can use one of the active funds instead. It's not like the index funds are cheap either.

I'm seeing a lot—but far from everything—in contango now, including oil.

All in all, not terribly encouraging I guess but I wouldn't overreact the other way either.

asif408
Posts: 1836
Joined: Sun Mar 02, 2014 8:34 am
Location: Florida

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by asif408 » Fri Jan 06, 2017 2:40 pm

nisiprius wrote:
asif408 wrote:I disagree with the part on correlations. If you look at the correlations of precious metals funds and energy funds (i.e., commodities producers) and compare them to a total stock market fund, you'll notice correlations were lower before 2003. Then between 2004-2011 the correlations increased dramatically. If you look at correlations from 2012-present you'll see that they have fallen again to pre-2004 levels. I think this shows that when there is a financial crisis worldwide like there was in 2007-2009 all equity or equity like assets, including commodities/commodities futures, will perform poorly.
There is a lot of sampling error in correlations. It takes a lot of data to pin down a correlation as "really" being in some range (with 1% significance). And there aren't too many non-overlapping periods to look at.
True. It will be interesting to see if the pre-2003 correlations continue or are an anomaly. Only time will tell.

User avatar
nisiprius
Advisory Board
Posts: 39723
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: Are Commodities Still a Good Portfolio Diversifier?

Post by nisiprius » Fri Jan 06, 2017 2:56 pm

A challenge. This table is presented in Gorton and Rouwenhorst's original paper, downloadable here.
Image

Since they comment that "the negative correlation of Commodity Futures with Stocks and Bonds tends to increase with the holding period" I will take it that they regard it as valid to examine correlations over 5-year overlapping intervals, so I will use 5-year overlapping intervals in my challenge.

Look carefully at this chart, which represents rolling 5-year correlations between two simulated asset classes, whose mean and standard deviation at the beginning of the time period were similar to stocks. I guarantee that there are no more than three times at which the basic characteristics of the two classes change, and that there was no change during years 0-10.

Looking at this chart and using your best judgement, answer these questions:

a) Did their characteristics ever change? If so, when do you think the changes occurred?

b) What do you believe is their true underlying correlation during years 0-10?

c) What do you believe is their true underlying correlation during years 20 through 24?

d) For "overlapping 5-year correlations," what is an appropriate ballpark number for a typical ± margin of error? For example, complete this statement by filling in the blank.

"In year 15, the true correlation between the two assets was -0.3 with a margin of error of ± __________"


Image
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.

lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 11:55 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by lack_ey » Fri Jan 06, 2017 3:28 pm

Ooh, challenge time.
nisiprius wrote:a) Did their characteristics ever change? If so, when do you think the changes occurred?
Hm, looks like the entire thing is not terribly inconsistent with a constant true underlying of around 0. Also not inconsistent with a few different regimes. It's obviously hard to distinguish because of the noise, limited samples, etc. unless the regimes are drastically disparate.

Though if the periods have fairly similar characteristics then it's not very wrong to assume they have the same underlying behavior and generally introducing more complexity in the model can get you in trouble. Unless something adds a lot of explanatory power you should err towards simplicity and keeping it out.

That said, it'd be boring if your up-to-3 changes was actually 0 so let's go with one modest change upwards around year 24.
nisiprius wrote:b) What do you believe is their true underlying correlation during years 0-10?
If I had to guess for this specifically I'd say 0.05. But if you're requiring all my answers to be consistent with each other, I say 0. Actually, let me just say 0.
nisiprius wrote:c) What do you believe is their true underlying correlation during years 20 through 24?
If I had to guess for this specifically I'd say -0.05. But if you're requiring all my answers to be consistent with each other, I say 0. Actually, let me just say 0.
nisiprius wrote:d) For "overlapping 5-year correlations," what is an appropriate ballpark number for a typical ± margin of error? For example, complete this statement by filling in the blank.

"In year 15, the true correlation between the two assets was -0.3 with a margin of error of ± __________"
Would anybody seriously guess -0.3 for year 15? That's obviously not the best choice because it never got lower. Given the whole series, it's more likely that -0.3 resulted from some randomness/noise from a true underlying not quite as negative.

So margin of error meaning within say 95%?

I don't have much intuition on this calculation. Maybe, um... -0.15% +/- 0.25%? You generated monthly data, right?
Last edited by lack_ey on Fri Jan 06, 2017 4:23 pm, edited 1 time in total.

heyyou
Posts: 3595
Joined: Tue Feb 20, 2007 4:58 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by heyyou » Fri Jan 06, 2017 4:14 pm

When boomers with memories of the 1980s were worried about inflation during upcoming retirements, Wall Street offered Collateralized Commodities Futures (CCF) funds based on a historical period, that has not repeated.

When the student is ready, the teacher will appear, but the student may not like the lesson he is about to receive.

Long term, equities have out grown inflation. What is the purpose of having commodities exposure? When diversifying into alternatives, there is also the risk of less overall returns from having less exposure to the positive but variable returns of the usual assets.

PCRIX, a CCF fund, recently had a 2 to 1 split to keep the price from looking so low. Note that the issue price on the chart is now $5, not the original and usual $10. That is not dishonest, but could be deceptive to some.

Trader/Investor
Posts: 261
Joined: Wed Dec 02, 2015 9:35 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by Trader/Investor » Fri Jan 06, 2017 4:28 pm

heyyou wrote:When boomers with memories of the 1980s were worried about inflation during upcoming retirements, Wall Street offered Collateralized Commodities Futures (CCF) funds based on a historical period, that has not repeated.

When the student is ready, the teacher will appear, but the student may not like the lesson he is about to receive.

Long term, equities have out grown inflation. What is the purpose of having commodities exposure? When diversifying into alternatives, there is also the risk of less overall returns from having less exposure to the positive but variable returns of the usual assets.

PCRIX, a CCF fund, recently had a 2 to 1 split to keep the price from looking so low. Note that the issue price on the chart is now $5, not the original and usual $10. That is not dishonest, but could be deceptive to some.
Someone who gets it! Great post albeit I recall more the inflationary period being the 70s when I was a futures broker. I feel for the kool-aid drinkers who have been seduced into buying the AQR managed futures fund QMHIX/QMHNX. Ouch! Down over 13% the past 12 months. When you see anyone recommending managed futures run for the exits!

User avatar
Clever_Username
Posts: 1405
Joined: Sun Jul 15, 2012 12:24 am
Location: Southern California

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by Clever_Username » Fri Jan 06, 2017 4:36 pm

nisiprius wrote:Gorton and Rouwenhorst don't talk about low correlation. They say "negative correlation." And notice how their abstract suffers from verb-tense disease!
...Fully-collateralized commodity futures have historically offered the same return and Sharpe ratio as equities. While the risk premium on commodity futures is essentially the same as equities, commodity futures returns are negatively correlated with equity returns and bond returns.
Shame on them!
Fully-collateralized commodity futures have historically offered the same return and Sharpe ratio as equities. While the risk premium on commodity futures is has been essentially the same as equities, commodity futures returns are have been negatively correlated with equity returns and bond returns.
'There, fixed that for ya'."
Nice catch. The first claim of the result is what surprised me the most until I saw this.
"What was true then is true now. Have a plan. Stick to it." -- XXXX, _Layer Cake_ | | I survived my first downturn and all I got was this signature line.

lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 11:55 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by lack_ey » Fri Jan 06, 2017 4:55 pm

Trader/Investor wrote:
heyyou wrote:When boomers with memories of the 1980s were worried about inflation during upcoming retirements, Wall Street offered Collateralized Commodities Futures (CCF) funds based on a historical period, that has not repeated.

When the student is ready, the teacher will appear, but the student may not like the lesson he is about to receive.

Long term, equities have out grown inflation. What is the purpose of having commodities exposure? When diversifying into alternatives, there is also the risk of less overall returns from having less exposure to the positive but variable returns of the usual assets.

PCRIX, a CCF fund, recently had a 2 to 1 split to keep the price from looking so low. Note that the issue price on the chart is now $5, not the original and usual $10. That is not dishonest, but could be deceptive to some.
Someone who gets it! Great post albeit I recall more the inflationary period being the 70s when I was a futures broker. I feel for the kool-aid drinkers who have been seduced into buying the AQR managed futures fund QMHIX/QMHNX. Ouch! Down over 13% the past 12 months. When you see anyone recommending managed futures run for the exits!
I know you don't like either, but how does a post and thread about (long) commodities exposure relate to a specific managed futures fund? Managed futures is in general long/short, primarily trend following as a primary investment strategy, with significantly different properties over time. Relevantly, the quoted post references inflation, where long commodities exposure may help in the case of some unexpected inflation shocks (which we haven't had lately, so this is something akin to a case of buying insurance that didn't pay out over a given period), whereas managed futures funds don't have the persistent long exposure there and shouldn't be expected to do anything really with regards to inflation.

Furthermore, the specific fund highlighted roughly keeps only about 25% of its risk budget in commodities, the rest in currencies, equities, and fixed income all long and short depending on the trend. By notional dollar exposure that'd be less than 25% as the commodities are usually going to be more volatile than fixed income and currencies, probably equities too.

Seems like a non sequitur.

Standard disclaimer: lack_ey does not own shares in any of the securities mentioned above. (CCFs or managed futures) :wink:

Trader/Investor
Posts: 261
Joined: Wed Dec 02, 2015 9:35 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by Trader/Investor » Fri Jan 06, 2017 5:08 pm

Seems like a non sequitur.


You are correct. I keep forgetting that as an actual practitioner I can't compete intellectually with the academics, theoreticians, and researchers on this board. Another non sequitur? Morningstar better get their act together as they have that "groupthink" fund QMHIX/QMHNX in their managed futures category.

User avatar
Kenkat
Posts: 5471
Joined: Thu Mar 01, 2007 11:18 am
Location: Cincinnati, OH

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by Kenkat » Fri Jan 06, 2017 5:18 pm

I have a commodities fund and I love it. When stocks go up, it goes down and when stocks go down, it goes down. A true pleasure to own over the past - meh, 10 years? (I've owned it about 14).

/ \
||

Gallows humor above.

I think I am back to zero return on my holding after a (finally) decent return this year, the first in awhile. It is a small percentage of my portfolio - and getting smaller every day - bazinga, I'm on fire today, folks! I plan to hold what I've got but it feels like I will have to wait a long time to see a positive diversification benefit, with a long time possibly including the word never.

lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 11:55 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by lack_ey » Fri Jan 06, 2017 5:26 pm

Trader/Investor wrote:Seems like a non sequitur.


You are correct. I keep forgetting that as an actual practitioner I can't compete intellectually with the academics, theoreticians, and researchers on this board. Another non sequitur? Morningstar better get their act together as they have that "groupthink" fund QMHIX/QMHNX in their managed futures category.
Oh, wait, you're not making a point about investment behavior or ideas, just about "expert" recommendations and thinking. That's the commonality, not the economic behavior/theory, right? i.e. "Experts are bad and have recommended both commodities and managed futures." Not "commodities are bad for portfolios and have bad returns in the real world so managed futures are bad for portfolios and have bad returns in the real world."

Sorry, I mean I was legitimately confused how you got from point A to point B. I'm interested in hearing different points of view and lines of thought but couldn't follow at first.


Just for reference, in general it matters in commodities funds the weighting and strategy used, and similarly for managed futures there are a range of metrics and underlying investment universes that can be used. So you expect differences between funds in a category, generally larger than differences between say two different large cap stock funds. But let me just say that here are three commodity funds and three managed futures funds and I think you can tell the groups apart:
Image
http://quotes.morningstar.com/chart/fun ... A%5B%5D%7D

Trader/Investor
Posts: 261
Joined: Wed Dec 02, 2015 9:35 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by Trader/Investor » Fri Jan 06, 2017 5:50 pm

lack_ey

Actually great comeback. But please don't get me going on experts. There are none - just in their own minds and the minds of the lemmings who can't think for themselves. Funny, but just in the past hour I was messaging with a trader about a few of the books in my vast (and mostly worthless) library on investing and trading. More specifically the ones that have hardened my view that no one knows anything. Once we realize that (most never will) we can go about developing our own unique and individualistic methodology based on our particular risk tolerances, goals, time frames, trading vehicles, and most importantly levels of commitment. It's that later (commitment) that really separates the very successful from the not so. But I digress.

User avatar
nisiprius
Advisory Board
Posts: 39723
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: Are Commodities Still a Good Portfolio Diversifier?

Post by nisiprius » Fri Jan 06, 2017 7:00 pm

lack_ey wrote:Ooh, challenge time... Hm, looks like the entire thing is not terribly inconsistent with a constant true underlying of around 0. Also not inconsistent with a few different regimes.... That said, it'd be boring if your up-to-3 changes was actually 0 so let's go with one modest change upwards around year 24.
nisiprius wrote:b) What do you believe is their true underlying correlation during years 0-10?
If I had to guess for this specifically I'd say 0.05. But if you're requiring all my answers to be consistent with each other, I say 0. Actually, let me just say 0.
"In year 15, the true correlation between the two assets was -0.3 with a margin of error of ± __________"
Would anybody seriously guess -0.3 for year 15? That's obviously not the best choice because it never got lower. Given the whole series, it's more likely that -0.3 resulted from some randomness/noise from a true underlying not quite as negative.

So margin of error meaning within say 95%?

I don't have much intuition on this calculation. Maybe, um... -0.15% +/- 0.25%? You generated monthly data, right?
You are correct on all points. There is no change, the number of changes is zero. I generated monthly data in which the monthly return for each asset was a random number with the uniform distribution, scaled and shifted so that it had µ = 0.797%, sigma = 1.531%, corresponding to 10% and 20% annually which is about the same as the stock market. There was no serial correlation within either time series. There was no correlation between the two, but even over the full 360-sample period, there was enough sampling error on the correlation so that the correlation within the sample was +.05.96.

I don't have a real number on the margin of error.

I was hoping that someone was going to think there was a real difference if some kind between years 0-14 and years 15-24.

Certainly, I read real writers writing about real asset correlations in ways that assume that "the correlations really changed" between two consecutive five-or-ten year periods of time, rather than "nothing changed, there's just that much sampling error in correlations."
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
CULater
Posts: 2340
Joined: Sun Nov 13, 2016 10:59 am

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by CULater » Sat Jan 07, 2017 10:20 am

Investing in Commodity Futures/Gold = "Skating where the puck was" ~ courtesy of B. Bernstein.

It's amazing how much money you can make in the past....
The future is a little more tricky....
On the internet, nobody knows you're a dog.

Angst
Posts: 2088
Joined: Sat Jun 09, 2007 11:31 am

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by Angst » Sat Jan 07, 2017 10:44 am

Larry,
Any chance you would want to weigh in with your current opinion on commodities? I haven't read you new book yet.

User avatar
Portfolio7
Posts: 695
Joined: Tue Aug 02, 2016 3:53 am

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by Portfolio7 » Sat Jan 07, 2017 11:01 am

Trader/Investor wrote:
heyyou wrote:When boomers with memories of the 1980s were worried about inflation during upcoming retirements, Wall Street offered Collateralized Commodities Futures (CCF) funds based on a historical period, that has not repeated.

When the student is ready, the teacher will appear, but the student may not like the lesson he is about to receive.

Long term, equities have out grown inflation. What is the purpose of having commodities exposure? When diversifying into alternatives, there is also the risk of less overall returns from having less exposure to the positive but variable returns of the usual assets.

PCRIX, a CCF fund, recently had a 2 to 1 split to keep the price from looking so low. Note that the issue price on the chart is now $5, not the original and usual $10. That is not dishonest, but could be deceptive to some.
Someone who gets it! Great post albeit I recall more the inflationary period being the 70s when I was a futures broker. I feel for the kool-aid drinkers who have been seduced into buying the AQR managed futures fund QMHIX/QMHNX. Ouch! Down over 13% the past 12 months. When you see anyone recommending managed futures run for the exits!
Great Post by Hey You. I'm not a fan of commodity funds, my real life experience with them and the theoretical underpinnings that I comprehend are less than convincing. Playing with portfolio backtests, iffy as that can be, also suggests they have little value for a diversified investor. I've read that there is a fair amount of indirect commodities exposure to be had in an emerging markets allocation, and would be interested in hearing a wider perspective on this from anyone who has any insight to offer.
"An investment in knowledge pays the best interest" - Benjamin Franklin

User avatar
nisiprius
Advisory Board
Posts: 39723
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: Are Commodities Still a Good Portfolio Diversifier?

Post by nisiprius » Sat Jan 07, 2017 11:43 am

lack_ey wrote:...here are three commodity funds and three managed futures funds and I think you can tell the groups apart:
Image
http://quotes.morningstar.com/chart/fun ... A%5B%5D%7D
You've convinced me that commodities and managed futures behave differently, and thus griping about managed futures holds no lessons about commodities. And this is off on a tangent. But, never having looked at managed futures funds before, looking at your chart my reaction is "and what on earth are those things supposed to be good for?"

AQMIX had the highest return and is very nearly as old as MHFIX, so lets use it as our example. What happens if we take a three-fund portfolio of 45% Total Stock, 15% Total International, and 40% of Total Bond (portfolio 1) and first replace half of it (portfolio 2) and then all of it (portfolio 3) with AQMIX?

Answer: 1) Nothing much happens... 2) and none of the things that do happen are good.

As we move from portfolio 1 with Total Bond--Total Bond in an era of handwringing about low bond returns--all that happens is the return drops, the standard deviation increases, the worst year gets worse, the maximum drawdown gets bigger, and the Sharpe ratio goes down. And, to my eye, starting from the first data PortfolioVisualizer has, if investors A and B had invested in portfolios without and with AQMIX, there is no point in the whole history at which the investor with AQMIX would have been ahead of the one without AQMIX.

Source
Image

So what is it all supposed to be about? Sophistication for the sake of sophistication?

P.S. Apologies, I misspoke when I said "there is no point in the whole history at which the investor with AQMIX would have been ahead of the one without AQMIX."
Image
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.

lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 11:55 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by lack_ey » Sat Jan 07, 2017 12:49 pm

nisiprius wrote:So what is it all supposed to be about? Sophistication for the sake of sophistication?
First of all, when 75/25 US/ex-US stocks have a Sharpe ratio of 0.81 and bonds have a Sharpe ratio of 0.96, it generally ain't gonna look good if you replace either of those with any kind of alts. That is, ex-post we know that reducing exposure to really great investments (over the period) is not a good thing. That's more of the difference you're seeing than anything particular to managed futures itself.

If you leveraged your portfolio in order to add other things without reducing your stock/bond exposures then your conclusion might be a little different. You can kind of simulate this to a limited degree by just deleveraging the non-managed futures reference allocation with cash, and taking managed futures for the comparison out of the cash allocation. Now, with actual leverage you'd be borrowing at above the cash rate (roughly let's say 50 bp or so above 3-month T-bills at large institutional levels, which is effectively available to individuals via futures), so no, this comparison is not entirely fair. But I think it is illustrative:

Image
Portfolio 1: 36% US stock, 12% ex-US stock, 32% bonds, 20% cash
Portfolio 2: 36% US stock, 12% ex-US stock, 32% bonds, 10% managed futures, 10% cash.
https://www.portfoliovisualizer.com/bac ... tion5_2=10

Managed futures is an investment strategy category primarily focused on trend following, based on the very simple concept of "the trend is your friend" (it's a bit ironic to call it "sophisticated"). That is, time-series momentum. Generally, this means investing long in whatever has been going up lately and shorting whatever has been going down lately, hoping that it'll keep doing the same thing. Usually, as implied by the name, this is implemented via futures contracts, which allow you to cheaply go long and short the underlying investments, be they individual commodities, equity indexes, bonds (well, Treasury bonds), currencies, etc. Futures also allow for leverage, obviously. Most implementations should be net over a business cycle around zero betas on average.

Managed futures is an old strategy, with decades of history from live money trading, so old there's a well-known index that tracks managers in the category, the BTOP50, with an inception of 1987. Commodities trading advisors (CTAs) get lumped in with managed futures because they're both doing the same thing, though CTAs at least traditionally more explicitly via commodities futures as opposed to other futures.

Nowadays hundreds of billions of dollars are invested in these strategies and I think it's reasonable to expect that they're kind of competing for and crowding out the same alpha. You can find more details in a number of papers, including a couple by AQR themselves, which are obviously bullish on them:
https://www.aqr.com/~/media/files/paper ... utures.pdf
https://www.aqr.com/~/media/files/paper ... esting.pdf

That said, one of the objections to the work is that a significant amount of the backtested and historical return seems to come from the rebalancing assumed in the above. And as always, today's markets are not the same as yesterday's. Still, those numbers are kind of eye popping.

One of the potential benefits sometimes touted with managed futures is that the strategy inherently switches to shorting risk assets during a bear market where they've been declining, so in a protracted bear market where presumably you're overall getting hammered, this is something of a hedge, perhaps being negatively correlated to stocks (because it's actively shorting stocks). Yes, I made a mythical negative-correlation-at-least-at-times-with-stocks-for-a-strategy-historically-with-positive-returns argument, and I'm sure your alarm bells are ringing.

Of course if what you see is a bunch of zigzagging in underlying assets with no clear trends, managed futures will dutifully churn its wheels losing money.

Trader/Investor
Posts: 261
Joined: Wed Dec 02, 2015 9:35 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by Trader/Investor » Sat Jan 07, 2017 1:57 pm

Managed futures is an old strategy, with decades of history from live money trading, so old there's a well-known index that tracks managers in the category, the BTOP50, with an inception of 1987. Commodities trading advisors (CTAs) get lumped in with managed futures because they're both doing the same thing, though CTAs at least traditionally more explicitly via commodities futures as opposed to other futures.

Correct me if I am wrong but as I recall the defect in that index is losing managers that go bust simply don't report their results thus falling out of the database. What is that called? Survivorship bias? I have known my share of CTAs who started out with a bang trying to raise money and diligently reported their stellar results only to crash and burn and were never heard from again. I could be thinking of another database of advisors and CTAs (Barclays or Vann maybe) but survivorship bias was a huge problem.

As an aside, back in the 80s there was a real money trading contest that ran for many years. At first it was just 4 month championships and then expanded to one year championships. It was the U.S. Trading Championship overseen by Norm Zadeh. It was broken down into several categories ala stocks, futures, options etc. The contestants were not your Average Joes as there was an entry fee. Some high profile traders including a few that were profiled in Schwager's Market Wizard books. Anyway, bottom line is that in a conversation I had with Mr Zadeh he said that overall (meaning all the years taken as an aggregate) it was something like 99% were net losers in the commodity futures division.

lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 11:55 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by lack_ey » Sat Jan 07, 2017 2:07 pm

Trader/Investor wrote:Managed futures is an old strategy, with decades of history from live money trading, so old there's a well-known index that tracks managers in the category, the BTOP50, with an inception of 1987. Commodities trading advisors (CTAs) get lumped in with managed futures because they're both doing the same thing, though CTAs at least traditionally more explicitly via commodities futures as opposed to other futures.

Correct me if I am wrong but as I recall the defect in that index is losing managers that go bust simply don't report their results thus falling out of the database. What is that called? Survivorship bias? I have known my share of CTAs who started out with a bang trying to raise money and diligently reported their stellar results only to crash and burn and were never heard from again. I could be thinking of another database of advisors and CTAs (Barclays or Vann maybe) but survivorship bias was a huge problem.

As an aside, back in the 80s there was a real money trading contest that ran for many years. At first it was just 4 month championships and then expanded to one year championships. It was the U.S. Trading Championship overseen by Norm Zadeh. It was broken down into several categories ala stocks, futures, options etc. The contestants were not your Average Joes as there was an entry fee. Some high profile traders including a few that were profiled in Schwager's Market Wizard books. Anyway, bottom line is that in a conversation I had with Mr Zadeh he said that overall (meaning all the years taken as an aggregate) it was something like 99% were net losers in the commodity futures division.
Oh yeah, for sure survivorship bias is a huge problem for that and most any indexes of this sort, and it's not really representative of the whole category. I didn't mean to take that index seriously as what you actually saw across the board, in terms of average performance stats.*

*though what you do see from the index is that the returns do regress significantly on mechanical trend following backtests, i.e. yes, that explains much of the typical behavior and effective or net strategy in common with a lot of the funds used in the index

That was just referenced in support of it being around for decades, that such investments were available. You can't have survivors and cherry-picked entrants in a category that doesn't exist. A lot of times a common complaint with some touted strategy X is that it's based solely on backtests and nobody actually ran real money with it for a while. This is not really the case here.

afan
Posts: 4465
Joined: Sun Jul 25, 2010 4:01 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by afan » Sun Jan 08, 2017 10:48 am

Among the many problems with backtesting commodities is the lack of an obvious standard. There are many candidates, but all are selected retrospectively to make the advertised strategy look good. There is no equivalent of a cap weighted total stock market. The costs of implementation of the various indexes vary widely and many are simply not investable at all. Depending on what the markets have done lately, some putative indexes may do much better than others.

A lot of smoke and mirrors.
We don't know how to beat the market on a risk-adjusted basis, and we don't know anyone that does know either | --Swedroe | We assume that markets are efficient, that prices are right | --Fama

User avatar
nisiprius
Advisory Board
Posts: 39723
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: Are Commodities Still a Good Portfolio Diversifier?

Post by nisiprius » Sun Jan 08, 2017 11:13 am

afan wrote:Among the many problems with backtesting commodities is the lack of an obvious standard. There are many candidates, but all are selected retrospectively to make the advertised strategy look good. There is no equivalent of a cap weighted total stock market. The costs of implementation of the various indexes vary widely and many are simply not investable at all. Depending on what the markets have done lately, some putative indexes may do much better than others.

A lot of smoke and mirrors.
A very good point.

One of the virtues of the DJIA, and, to a lesser extent, the S&P 500 and its backfilled extensions back to 1870, is that they are old and generally accepted as "the" measure of something. They represent a single generally accepted point of reference... and the DJIA to some extent validates the S&P-500-and-predecessor-data because they aren't terribly different.

What is the history of "established, accepted commodities indexes?" Are there any old indexes, widely respected and maintained according to consistent rules for a long time, or does everyone basically "invent" their own index for backtesting? One moment while I go look at the paper... (Gorton and Rouwenhurst) "We construct an equally-weighted index of commodity futures monthly returns over the period between July of 1959 and December of 2004 in order to study simple properties of commodity futures as an asset class." Home-brew.
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.

lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 11:55 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by lack_ey » Sun Jan 08, 2017 11:51 am

I mentioned it upthread how all the indexes disagree and it's a problem determining which to look at. Also, before you go looking, don't forget to distinguish between commodity price indexes and commodity futures indexes.

The Bloomberg Commodity Index, recently the Dow Jones-UBS Commodity Index, formerly the Dow Jones-AIG Commodity Index, is kind of well known, started in 1998. The Thomson Reuters/Jefferies CRB Index dates to 1957, and what's what Rick Ferri cites in All About Asset Allocation.

Many papers seem to equal weight, especially if they're taking data from a long time ago. But others don't.

Keep in mind that many advocate for weighting by production or economic importance. Note that this generally tends to emphasize energy at the expense of other components, which looks terrible backtested the last decade. Others put arbitrary caps on sectors, equal weight, or who knows what else. Like I said, widespread disagreements in methodology and weightings.

A commodities futures index also has to worry about how and when the futures contracts are rolled and which are used. Most assume front-month contracts but many actual funds will use longer-dated contracts at times if this looks like it will suffer less/benefit more from roll yield. In fact, some commodity index funds follow smart-beta-esque indexes that try to optimize in this actual way (baking in some active, rules-based strategy into the index).

TonyDAntonio
Posts: 475
Joined: Thu Mar 03, 2016 8:32 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by TonyDAntonio » Sun Jan 08, 2017 12:05 pm

Your experience sounds like mine. On the writings of Larry Swedroe I bought PCRIX about 8 years ago. Luckily it is not a large part of my portfolio. I really don't want to sell it when it is lower than when I bought it and more importantly when I have not given it a chance during a rising inflation environment. Threads like these cause me pause.
kenschmidt wrote:I have a commodities fund and I love it. When stocks go up, it goes down and when stocks go down, it goes down. A true pleasure to own over the past - meh, 10 years? (I've owned it about 14).

/ \
||

Gallows humor above.

I think I am back to zero return on my holding after a (finally) decent return this year, the first in awhile. It is a small percentage of my portfolio - and getting smaller every day - bazinga, I'm on fire today, folks! I plan to hold what I've got but it feels like I will have to wait a long time to see a positive diversification benefit, with a long time possibly including the word never.

Trader/Investor
Posts: 261
Joined: Wed Dec 02, 2015 9:35 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by Trader/Investor » Sun Jan 08, 2017 12:15 pm

On the writings of Larry Swedroe I bought PCRIX about 8 years ago.

Use that as a learning experience.

TJSI
Posts: 365
Joined: Thu Jan 27, 2011 4:03 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by TJSI » Sun Jan 08, 2017 12:27 pm

TonyDAntonio, Trader/Investor,

You are confusing strategy with outcome!

TJSI

User avatar
Kenkat
Posts: 5471
Joined: Thu Mar 01, 2007 11:18 am
Location: Cincinnati, OH

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by Kenkat » Sun Jan 08, 2017 12:29 pm

TonyDAntonio wrote:Your experience sounds like mine. On the writings of Larry Swedroe I bought PCRIX about 8 years ago. Luckily it is not a large part of my portfolio. I really don't want to sell it when it is lower than when I bought it and more importantly when I have not given it a chance during a rising inflation environment. Threads like these cause me pause.
kenschmidt wrote:I have a commodities fund and I love it. When stocks go up, it goes down and when stocks go down, it goes down. A true pleasure to own over the past - meh, 10 years? (I've owned it about 14).

/ \
||

Gallows humor above.

I think I am back to zero return on my holding after a (finally) decent return this year, the first in awhile. It is a small percentage of my portfolio - and getting smaller every day - bazinga, I'm on fire today, folks! I plan to hold what I've got but it feels like I will have to wait a long time to see a positive diversification benefit, with a long time possibly including the word never.
I initially acquired PCRIX - PIMCO Commodity Real Return in 2004. I just ran my numbers and my total annualized return 2004-present is -2.43%. So, it's a bit worse than I thought. I made smaller rebalancing purchases in 2008 and 2013. It's probably due for another rebalance but honestly can't quite bring myself to risk flushing more money right now. It's a small part of my portfolio anyway.

What has really killed this fund is this run:

2013: -14.90%
2014: -18.05%
2015: -25.69%

I believe this is largely a result of the historic collapse in oil prices from over $100 bbl to under $30 at the bottom. I do think this fund will contribute something positive someday but there is a lesson in patience and expected returns not always equaling actual returns over your time period.

lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 11:55 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by lack_ey » Sun Jan 08, 2017 12:41 pm

kenschmidt wrote:What has really killed this fund is this run:

2013: -14.90%
2014: -18.05%
2015: -25.69%

I believe this is largely a result of the historic collapse in oil prices from over $100 bbl to under $30 at the bottom. I do think this fund will contribute something positive someday but there is a lesson in patience and expected returns not always equaling actual returns over your time period.
I'm curious. What do you estimate to be the long-run expected returns (including fund fees, costs)?

FWIW if you extended bond duration while adding commodities—here specifically looking at PIMCO's fund, which did better than some others—then it wouldn't have been bad in 2007-present:
https://www.portfoliovisualizer.com/bac ... ion5_3=100
That's a little bit over 1 extra year of duration while adding 5% overall commodities.

The argument there being you can extend duration because you're less exposed to unexpected inflation, in which nominal bonds of increasing duration will do increasingly poorly but commodities might do better. Turns out we've had deflationary forces, a collapse in some commodity prices, which did the reverse by rewarding term risk and killing commodities (well, the roll yield also did too).

User avatar
nedsaid
Posts: 12964
Joined: Fri Nov 23, 2012 12:33 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by nedsaid » Sun Jan 08, 2017 12:47 pm

Trader/Investor wrote:On the writings of Larry Swedroe I bought PCRIX about 8 years ago.

Use that as a learning experience.
I have read your posts here and other places and have concluded that you are the real deal. Particularly when a moderator found a presentation you had made at a conference. So I read what you have to say with great interest.

From my layman's experience as an investor, I have concluded that markets will do what markets will do. As human beings, we like to have control or at least the illusion of control over our investment performance. Investing is worthwhile because it is tied in the long run to real businesses and to a real economy. As such, we should reap the long term benefits of owning the means of production in a capitalist economy. What gives us pause is that while we know what is pretty likely to happen in the long term, during shorter periods of time almost anything can happen. Even long term, I said "pretty likely" because even that we don't know that 100% for sure. We just have faith that human civilization will advance, businesses and the economy will grow, and those things should allow our investments to grow too. We know from history that entire civilizations fall. We save and invest anyway.

This whole exercise of portfolio theory and the search for that asset class that will a perfect diversifier is an attempt to take the volatility out of the upward climbs of our portfolios. But markets are very turbulent, and are that way despite our very best efforts. I have thrown a lot of asset classes and sub-classes at this problem and my portfolio was still volatile. Commodities might work after I have thrown everything else and the kitchen sink at the volatility problem. What I find is that markets inevitably will do the very thing that you and the very smart people never thought of.

Commodities would help a portfolio in the case of the 1973-74 oil shocks and bear market on Wall Street. The infamous period of stagflation, a stagnant economy mixed with higher inflation. In fact, commodities might have been the only thing that would have "worked" during that period. Unfortunately, just as the 1930's Great Depression never repeated itself, I think that those of us who arm our portfolios against a 1973-74 repeat, are likely to be disappointed.

Stagflation, inflation spikes, and the oil shocks were tough on stocks. Higher inflation was bad for bonds too. As I have posted many times, it took investors about 10 years or so to get their inflation adjustment from the US Stock Market. Stocks struggled in this environment that I described and it took the falling interest rates and falling inflation of the 1980's to get stocks roaring again. I am not sure that we should construct our portfolios around a scenario that we may never see again in our lifetimes.

As far as PCRIX (Pimco Commodity Real Return), I am sure that Larry would say something like you shouldn't expect fire insurance to reimburse your loss in an auto crash. Commodities are insurance against inflation spikes, and since the 2008-2009 financial crisis and bear market, the big worry has been about deflation. Larry would also say that owning something like PCRIX would allow you to lengthen the maturities and duration of your bonds.

The idea of commodities as "portfolio insurance" is an interesting one, but so far I have not wanted to pay the premiums.
A fool and his money are good for business.

Trader/Investor
Posts: 261
Joined: Wed Dec 02, 2015 9:35 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by Trader/Investor » Sun Jan 08, 2017 1:06 pm

I am not sure that we should construct our portfolios around a scenario that we may never see again in our lifetimes.

Could not agree more. Be it a repeat of the Great Depression, the inflationary 70s, or the Great Recession/Credit Crisis of 2008. The first one, the Great Depression scared a generation of investors off stocks for two and one half decades. The second saw the rise of the gold and precious metal bugs and the mindset that inflation is always around the corner. The later has been the most interesting with the rise of alternative funds. Great funds for those with a vested interest in marketing such. All the while those staying the course with the simple Boglehead portfolio would have accumulated the most wealth.

Edit: Be wary of the real deals of the 80s and 90s where anyone could have become a real deal. A lot have crashed and burned since 2000, albeit so far that tragedy has not befallen me. A real deal although not a big deal as I am just a simple small time Mom and Pop trader who made good.

User avatar
Kenkat
Posts: 5471
Joined: Thu Mar 01, 2007 11:18 am
Location: Cincinnati, OH

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by Kenkat » Sun Jan 08, 2017 1:27 pm

lack_ey wrote:I'm curious. What do you estimate to be the long-run expected returns (including fund fees, costs)?
I really bought this as a portfolio diversified than for straight up return. However, with the caveat that returns are notorious hard to predict, I would expect that the underlying commodities would return 0% real, the TIPS that the excess funds are invested in to be around 1-2% real and perhaps a diversification benefit of 1-2% real, so maybe 2-4% real best case scenario, with the actual results being much worse, of course.

lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 11:55 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by lack_ey » Sun Jan 08, 2017 1:32 pm

kenschmidt wrote:
lack_ey wrote:I'm curious. What do you estimate to be the long-run expected returns (including fund fees, costs)?
I really bought this as a portfolio diversified than for straight up return. However, with the caveat that returns are notorious hard to predict, I would expect that the underlying commodities would return 0% real, the TIPS that the excess funds are invested in to be around 1-2% real and perhaps a diversification benefit of 1-2% real, so maybe 2-4% real best case scenario, with the actual results being much worse, of course.
Yeah, I know, we want to know how the entire portfolio fits. Anyway, that's interesting, thanks.

Actually, by "diversification benefit" do you mean internal to the fund, from shifting weights of underlying exposures? Or are you talking about a whole-portfolio-level rebalancing bonus? Scaled to the weight of... wait a sec, how large is your weighting to the fund, anyway?


nedsaid wrote:From my layman's experience as an investor, I have concluded that markets will do what markets will do. As human beings, we like to have control or at least the illusion of control over our investment performance. Investing is worthwhile because it is tied in the long run to real businesses and to a real economy. As such, we should reap the long term benefits of owning the means of production in a capitalist economy. What gives us pause is that while we know what is pretty likely to happen in the long term, during shorter periods of time almost anything can happen. Even long term, I said "pretty likely" because even that we don't know that 100% for sure. We just have faith that human civilization will advance, businesses and the economy will grow, and those things should allow our investments to grow too. We know from history that entire civilizations fall. We save and invest anyway.
IMHO it's a misplaced faith. It's more than likely we'll see advances and markets working as mostly in the past but far from guaranteed even outside of the doomsday scenarios. Many people are significantly too confident in prosperity and work off of a heuristic of a high probability being treated as a virtual certainty.
nedsaid wrote:This whole exercise of portfolio theory and the search for that asset class that will a perfect diversifier is an attempt to take the volatility out of the upward climbs of our portfolios. But markets are very turbulent, and are that way despite our very best efforts. I have thrown a lot of asset classes and sub-classes at this problem and my portfolio was still volatile. Commodities might work after I have thrown everything else and the kitchen sink at the volatility problem. What I find is that markets inevitably will do the very thing that you and the very smart people never thought of.
No, it's not all about volatility, and to begin with this language already assumes an upward climb, which is likely but not the only possibility.

Vol is what we measure based on fluctuations over time. Here's a made-up illustration showing returns (vertical axis) over time (horizontal axis):
Image

We're at the gray dividing line. I've graphed three possible future routes out of the infinite possible continuations with three different colors. What you measure with vol based on actual data is the fluctuations along the given line. What you don't see, and what's hard to determine, is which kind of path we'll see going forward, including very importantly the kind of long-run average we'll get out there. If we get the blue line, everything is great. If we get red, it was a hell of a ride but turned out okay. If we get orange, that sucks, even though vol was low. Obviously things could also be a lot worse or better.

Part of the reason to diversify (say, have both stocks and bonds) is because we don't know which paths we'll get on for anything. For one, we don't really know what to guess for expected value.

Now if you just want to reduce vol, that's just about reducing exposure to volatile assets and/or finding assets with legitimately low enough correlation without a lot higher vol of its own.
Last edited by lack_ey on Sun Jan 08, 2017 1:33 pm, edited 1 time in total.

TonyDAntonio
Posts: 475
Joined: Thu Mar 03, 2016 8:32 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by TonyDAntonio » Sun Jan 08, 2017 1:33 pm

Not totally. I knew that PCRIX was basically unexpected inflation insurance. What gives me pause are threads like this where more knowledgeable people than myself theorize that commodity futures funds aren't all they are cracked up to be. I'll admit, I don't delve into the details of my investments. I read forums like this, used to listen to merriman podcasts (some of which included Larry Swedroe) and then take my best guess and pretty much stay the course.
TJSI wrote:TonyDAntonio, Trader/Investor,

You are confusing strategy with outcome!

TJSI

TJSI
Posts: 365
Joined: Thu Jan 27, 2011 4:03 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by TJSI » Sun Jan 08, 2017 1:51 pm

Sorry Tony, meant that as humor. Guess it was not funny.

TJSI

User avatar
nedsaid
Posts: 12964
Joined: Fri Nov 23, 2012 12:33 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by nedsaid » Sun Jan 08, 2017 2:09 pm

Trader/Investor wrote: I am not sure that we should construct our portfolios around a scenario that we may never see again in our lifetimes.

Could not agree more. Be it a repeat of the Great Depression, the inflationary 70s, or the Great Recession/Credit Crisis of 2008. The first one, the Great Depression scared a generation of investors off stocks for two and one half decades. The second saw the rise of the gold and precious metal bugs and the mindset that inflation is always around the corner. The later has been the most interesting with the rise of alternative funds. Great funds for those with a vested interest in marketing such. All the while those staying the course with the simple Boglehead portfolio would have accumulated the most wealth.

Edit: Be wary of the real deals of the 80s and 90s where anyone could have become a real deal. A lot have crashed and burned since 2000, albeit so far that tragedy has not befallen me. A real deal although not a big deal as I am just a simple small time Mom and Pop trader who made good.
As far as being the "real deal", what I meant is that you appear to be exactly what you claim to be. You are not a household name but you did well enough with your investments that your are now comfortable. You evidently succeeded at doing something at which most people fail. As the late Dizzy Dean would say, "If you have done it, it ain't bragging." The same reason I have a lot of respect for Jim Kramer, he succeeded in a very, very tough business. May not agree with much of what he says but I will listen. Kramer has his TV show and his millions and I do not.

Almost any success in life can be attributed to luck by skeptics; but I think there was hard work, willingness to take risks, and good old fashioned stick-to-itiveness or perseverance. Pretty much, I am an avatar on an anonymous web forum who in real life has invested money with some success. But I am not worth millions. I don't recommend to ordinary folks to try to make a living by trading as you did but at the same time I know there are professional traders who make a career out of it.

I also think you are saying that times change. The economy changes, the markets change. What worked in the 1980's and 1990's won't necessarily work today. There is something in us that always wants to go back to those first successes that we had. We have to be realistic enough to realize that we have to be willing to be flexible and to adjust to conditions. The world is a dynamic and not a static environment. Thus we have to be dynamic too.

Pretty much, it is an exercise of knowing yourself and knowing why you do the things that you do. For me, I want to own the means of production in a capitalist economy and reap the benefits of business and real estate ownership without the headaches. My REIT ETF and REIT funds won't call me at 3:00 in the morning to tell me that a toilet plugged up. My view is longer term, thinking of myself as a long term owner and a long term beneficiary of capitalism and the economic growth that it affords. Personality wise, I am just not a trader, I have zero interest in it.

As far as your comment about experts, I have found myself returning to my roots as an investor. My interest in running my own individual stock portfolio is on the rise again and I may modestly increase my financial commitment there. For example, I purchased shares of Gilead Sciences. Couldn't believe that such a quality firm could be had for a P/E ratio of 6! There are risks in a risky company in a risky industry but I thought these were risks well worth taking. I also wanted to own Coca-Cola shares for a long time, and I sold trimmed shares of a company that recently did well to buy Coke and to increase my position in Gilead Sciences.

I get Bogleheadism and index funds. I am not going to return to the days where my individual stocks were 45% of my retirement portfolio. But I might go from 12-13% or so to maybe 15%. It doesn't seem 100% logical to me or 100% rational but I am comfortable with it. I would like to own 20-25 stocks in both retirement accounts and taxable accounts. I think now that I own 21 stocks.

Part of the reason I like my individual stock portfolio is that I like to know what I own and I like to collect the dividends. I also like that dividend payments tend to increase over time. A stream of income that has a good chance of growing faster than inflation is very appealing to me. Despite getting beaten with wet noodles in the various dividend threads, I still like them. Something about seeing those dividend payments hitting my account that I like.

So I look at what I am and who I am and invest accordingly. I index a lot of my money, still own actively managed mutual funds, and still own individual stocks. I am comfortable with that as my active funds are actually employed to capture factors like Value and Momentum. I am also in a minority in that I still work with a person who is essentially a stock broker with a portion of my investments. I am what I am and thus don't necessarily advise that others do what I do. My suspicion is that you don't tell individual investors to trade like you did, you probably give acquaintances mostly standard advice.

This is why have not invested in Commodities. First, I am an investor by nature and temperament and not a trader. Second, I really don't understand the commodity markets very well and my interest is limited. Third, the case for including them in a portfolio is pretty sketchy. So far, they don't fit in with who I am, what I believe about markets and investments, and my circle of competence.
A fool and his money are good for business.

User avatar
nedsaid
Posts: 12964
Joined: Fri Nov 23, 2012 12:33 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by nedsaid » Sun Jan 08, 2017 2:26 pm

lack_ey wrote:
nedsaid wrote:From my layman's experience as an investor, I have concluded that markets will do what markets will do. As human beings, we like to have control or at least the illusion of control over our investment performance. Investing is worthwhile because it is tied in the long run to real businesses and to a real economy. As such, we should reap the long term benefits of owning the means of production in a capitalist economy. What gives us pause is that while we know what is pretty likely to happen in the long term, during shorter periods of time almost anything can happen. Even long term, I said "pretty likely" because even that we don't know that 100% for sure. We just have faith that human civilization will advance, businesses and the economy will grow, and those things should allow our investments to grow too. We know from history that entire civilizations fall. We save and invest anyway.
IMHO it's a misplaced faith. It's more than likely we'll see advances and markets working as mostly in the past but far from guaranteed even outside of the doomsday scenarios. Many people are significantly too confident in prosperity and work off of a heuristic of a high probability being treated as a virtual certainty.

Nedsaid: You have to believe in something or you will be too scared to even roll out of bed in the morning. Yes, I have faith in certain things even knowing that there is a chance that my faith might be misplaced. There are no guarantees with investing, virtually anything can happen. If I 100% knew that everything would work out for me in the long run, then it wouldn't be faith. For all I know, the we could get hit by a killer asteroid tomorrow and it would all be over. But I have to go with what is likely to happen and invest. I can't be paralyzed with the fear that something horrible might go wrong. So invest I must.
nedsaid wrote:This whole exercise of portfolio theory and the search for that asset class that will a perfect diversifier is an attempt to take the volatility out of the upward climbs of our portfolios. But markets are very turbulent, and are that way despite our very best efforts. I have thrown a lot of asset classes and sub-classes at this problem and my portfolio was still volatile. Commodities might work after I have thrown everything else and the kitchen sink at the volatility problem. What I find is that markets inevitably will do the very thing that you and the very smart people never thought of.
No, it's not all about volatility, and to begin with this language already assumes an upward climb, which is likely but not the only possibility.

Nedsaid: Yes, but when you talk to investors, they might intellectually believe there are other things at work, and perhaps other definitions of risk; but deep down in our hearts what we are worried about is that our investments will fluctuate, fluctuate down significantly, and fluctuate down significantly at the worst possible time. Risk of loss, and risk that the loss will be permanent rather than temporary.

Vol is what we measure based on fluctuations over time. Here's a made-up illustration showing returns (vertical axis) over time (horizontal axis):
Image

Nedsaid: Intellectually, you are probably right. But I can tell you that investors are most concerned about the price their investments can fetch in the markets at any given time. If the value of your portfolio goes down 40%, nearly all investors won't give a hoot if the losses occurred with thin trading volume or heavy trading volume. Almost no one, in their heart of hearts, lays awake at night over trading volumes. What they worry about is the market value of their investments.

We're at the gray dividing line. I've graphed three possible future routes out of the infinite possible continuations with three different colors. What you measure with vol based on actual data is the fluctuations along the given line. What you don't see, and what's hard to determine, is which kind of path we'll see going forward, including very importantly the kind of long-run average we'll get out there. If we get the blue line, everything is great. If we get red, it was a hell of a ride but turned out okay. If we get orange, that sucks, even though vol was low. Obviously things could also be a lot worse or better.

Part of the reason to diversify (say, have both stocks and bonds) is because we don't know which paths we'll get on for anything. For one, we don't really know what to guess for expected value.

Nedsaid: This is why I diversify. I literally have thrown factors, asset classes, and the kitchen sink at my portfolio and it still fluctuates at times I don't like in directions I don't like. I love volatility on the way up but emotionally volatility on the way down is hard to take. Unlike the quants, I realize that I am making educated guesses and that my approach is less scientific than it appears. Pretty much a lot of eyeballing. Or at least I am more honest about my doubts and uncertainties.

Now if you just want to reduce vol, that's just about reducing exposure to volatile assets and/or finding assets with legitimately low enough correlation without a lot higher vol of its own.
A fool and his money are good for business.

User avatar
Kenkat
Posts: 5471
Joined: Thu Mar 01, 2007 11:18 am
Location: Cincinnati, OH

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by Kenkat » Sun Jan 08, 2017 2:27 pm

lack_ey wrote:Actually, by "diversification benefit" do you mean internal to the fund, from shifting weights of underlying exposures? Or are you talking about a whole-portfolio-level rebalancing bonus? Scaled to the weight of... wait a sec, how large is your weighting to the fund, anyway?
Whole portfolio level.

The original plan was 10% hard assets made up of REIT @ 5% and commodities @ 5%. The REIT has done great since 2000 of course, so it is around 6%, while commodities has dropped under 3%. Rebalancing is somewhat complicated by the fact that the account that has access to PCRIX also holds an international fund and I'm under target there also. Admittedly there's also an aspect of finding it difficult to stick to a so far losing plan re: commodities.

lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 11:55 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by lack_ey » Sun Jan 08, 2017 2:55 pm

nedsaid wrote:Nedsaid: Intellectually, you are probably right. But I can tell you that investors are most concerned about the price their investments can fetch in the markets at any given time. If the value of your portfolio goes down 40%, nearly all investors won't give a hoot if the losses occurred with thin trading volume or heavy trading volume. Almost no one, in their heart of hearts, lays awake at night over trading volumes. What they worry about is the market value of their investments.
Question: did you think I wrote anything that has anything to do with trading volumes (maybe the example was even more unclear than I thought and I gave you the wrong impression), or was that just an off-the-cuff remark about something people don't care about relative to something like sharp drops?

User avatar
nedsaid
Posts: 12964
Joined: Fri Nov 23, 2012 12:33 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by nedsaid » Sun Jan 08, 2017 3:11 pm

lack_ey wrote:
nedsaid wrote:Nedsaid: Intellectually, you are probably right. But I can tell you that investors are most concerned about the price their investments can fetch in the markets at any given time. If the value of your portfolio goes down 40%, nearly all investors won't give a hoot if the losses occurred with thin trading volume or heavy trading volume. Almost no one, in their heart of hearts, lays awake at night over trading volumes. What they worry about is the market value of their investments.
Question: did you think I wrote anything that has anything to do with trading volumes (maybe the example was even more unclear than I thought and I gave you the wrong impression), or was that just an off-the-cuff remark about something people don't care about relative to something like sharp drops?
No, I think you thought through your comments very well. You are correct, trading volumes are important and investors don't give enough thought to it. You would expect that a market crash would be associated with very heavy trading volumes, race to the exit and all of that.

Yet, the bond correction in 2008-2009 was due to lack of volume and lack of liquidity, the complete opposite of what you would think. There literally was no market for anything in the credit markets but for US Treasuries and US Government Agency Bonds. Even TIPS had a liquidity issue. General Electric had a AAA Credit rating at the time and couldn't roll over its commercial paper. GE itself had a terrible liquidity problem and would have been temporarily insolvent if Warren Buffett had not stepped in. There was a period of time that Uncle Sam and Warren Buffett were in effect the only buyers in most segments of the bond market. Not a great situation to be in.

Your graph of three possible outcomes was very helpful to visualize what could happen. But we did have a situation of great volatility in most bonds with little volume. This is what I mean that the markets will inevitably do what even the brightest people didn't anticipate. Doesn't do any good to hit the "sell" button if there are no buy orders. In my wildest imagination, I would never have guessed that investors would be so scared that they would consider AAA rated short term commercial paper to be too risky to purchase. I also never imagined that there would be a run on money market accounts which are regarded (or at least used to be regarded) as safe.
A fool and his money are good for business.

lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 11:55 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by lack_ey » Sun Jan 08, 2017 3:47 pm

Oh, well it was actually nothing about trading volumes or liquidity.

I was attempting to make a point about stochastic processes, that what we observe is just variability over time along a sample path (varying t but not omega), which does not exactly describe the variability among possible sample paths.

I don't know what a good explanation would look like for a general audience.

User avatar
nedsaid
Posts: 12964
Joined: Fri Nov 23, 2012 12:33 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by nedsaid » Sun Jan 08, 2017 4:08 pm

lack_ey wrote:Oh, well it was actually nothing about trading volumes or liquidity.

I was attempting to make a point about stochastic processes, that what we observe is just variability over time along a sample path (varying t but not omega), which does not exactly describe the variability among possible sample paths.

I don't know what a good explanation would look like for a general audience.
Let me take a stab it at. Our mind assigns patterns and trends to things that are in fact random. We might see a pattern or a trend in a dataset, and perhaps assign a causation to what we are seeing, when in fact what we are seeing is just random noise. No trend, no pattern, no causation, but just random movement. We think we are seeing something but in fact we are not. Perception but not reality. This is a criticism of technical analysis.

Edit: Did some reading on Wikipedia and found that a stochastic process is an attempt to explain what is seemingly random, the opposite of what I described above. So we could look at something that we think is utterly random and find that indeed there is some pattern and some power to explain what we are seeing.

It is sort of like the difference between Newtonian physics and Quantum Mechanics. The physics of large objects vs the physics of much smaller objects. In Newtonian physics, the movement of large objects is 100% predictable. Hence the term, plug and chug, in physics. We know the equation, we just need the proper number to plug into the variable so that we can calculate the correct answer.

In quantum mechanics, we found that the behavior of sub-atomic particles wasn't so predictable. It seems that the rules that govern large objects are different for much, much smaller objects. Plug and chug just doesn't work or at least the underlying math is much more complex. There is an uncertainty principal at work. The more we know about the location of a particle the less we know about its momentum. The more we know about the momentum of a particle, the less we know about its location. In other words, as we studied light and as we studied sub-atomic particles, there was another level of complexity that we didn't understand before.

I have read that heartbeats are sort of like what I am describing. They are mostly regular and predictable but there is a seeming randomness in the process as well. I remember watching the vital signs monitor display in the hospital while my father was there. I would get really uneasy seeing stuff fluctuate all over the place including his heart rate. But the nurses and doctors went about their business completely unconcerned about what was happening. I was particularly alarmed because my father has a rather high tech pacemaker which is supposed to keep his heartbeat regular. But his heart rate and the graph on the screen fluctuated and I had to stop watching the blasted thing. I didn't like what I was seeing but in fact what I was seeing was perfectly normal, even with a pacemaker. As long as the lines wiggle, I don't worry. It is the flat lines that cause the medical teams to scurry.

I suppose as we learn more, the seeming randomness and variability will be perceived as less random. We will further develop math and physics theory to explain what seems to a layman to have no explanation. The motion of sub-atomic particles and I suppose the variation that I could see in my father's vital signs, particularly the heartbeat. What seemed random wasn't so random after all. In fact, the perceived randomness is part of what makes it all work.
A fool and his money are good for business.

User avatar
nedsaid
Posts: 12964
Joined: Fri Nov 23, 2012 12:33 pm

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by nedsaid » Sun Jan 08, 2017 4:54 pm

The discussion above is why there is a lot of skepticism about back testing and about academic research. Are we really seeing patterns and trends and coming up with explanations of what causes them? Or is it sort of like the Rorschach test where you are interpreting inkblots? Is it reality or is it perception? Are we really seeing something or are we just seeing what we want to see to confirm our biases? The human mind has the habit of screening things out, in other words we don't perceive everything that our five senses take in. Certain things just don't make it do our perception.

I like to use the analogy of the Subaru Outback Station Wagon. Didn't know the darned things existed. Probably saw them but it just never registered with me. One day, my neighbor had one parked next to my parking spot as I pulled in. After that, I saw Subaru Outback Station Wagons all over the place. I was tuned into them, before they were invisible and I didn't know even what they were or if they existed and afterwards they were all over the place! Nothing changed but my perception.
A fool and his money are good for business.

Fundhunter
Posts: 181
Joined: Sun Mar 11, 2007 9:11 pm
Location: Atlanta

Re: Are Commodities Still a Good Portfolio Diversifier?

Post by Fundhunter » Sun Jan 08, 2017 5:03 pm

OK, I don't own any commodities and never have. I don't claim to have analyzed all the technical stuff posted in this thread, nor to be any expert on commodities, but skimming what is posted here, I haven't seen anything that would convince me that this type of investment needs to be part of anything my wife or I own. I know that commodities behave differently than the stock market (so do mutual funds that short the market), but that does not mean I should own them. Are there people who think that this type of investment SHOULD be a part of a well diversified portfolio (whether you "like" your particular fund or not)?

Post Reply