commodities and multi-asset class portfolios

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

commodities and multi-asset class portfolios

Post by larryswedroe »

new study on subject

http://www.etf.com/sections/index-inves ... -risk.html

I'd add this, commodities have tended to perform best when they have tail wind of backwardation, which has now been the case for about the last year in the important oil futures, which for years were in contango--some saying that it was permanently doomed to be that way because of the institutional demand. That obviously turned out to be incorrect and there is now a big tailwind in oil.

Best wishes
Larry
heyyou
Posts: 4461
Joined: Tue Feb 20, 2007 3:58 pm

Re: commodities and multi-asset class portfolios

Post by heyyou »

Seems to be a couple of more hurdles for retail investors before they can benefit from the known theory of commodities investing.

There is another step of looking at the funds available to retail investors. I cannot choose among specific commodities, only among specific commodities funds each with their own spectrum of expense ratios.

Are there commodities funds with passive management that is not disadvantageous (specific futures sell and buy dates is one example)?

PIMCO makes a hundred million a year in fees from their commodities fund, regardless of the commodities market and with economy of scale on its expenses. I wish that I could invest in that division of PIMCO as my exposure to commodities.
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: commodities and multi-asset class portfolios

Post by larryswedroe »

heyyou
Well we used to use the PIMCO fund until DFA finally came out with a lower cost version with a lot less active management and risk on the collateral side. Fortunately DFA has also been lowering its ER which went from 54 to now 34bp. One could say that's Vanguard like, if not quite Vanguard.
There are plenty of ETFs that one might also consider.

Larry
Calm Man
Posts: 2917
Joined: Wed Sep 19, 2012 9:35 am

Re: commodities and multi-asset class portfolios

Post by Calm Man »

This all sounds wonderful. That said I think I"ll stick with Taylor's 3 fund portfolio and let the chips fall where they may. They will probably fall better than adding any of these new ideas which need to be discovered because otherwise professionals wouldn't have anything to interest others in.
afan
Posts: 8195
Joined: Sun Jul 25, 2010 4:01 pm

Re: commodities and multi-asset class portfolios

Post by afan »

Intersting paper, but hardly a strong argument in favor of commodities. When Vanguard (or I suppose some other firm, but probably Vanguard) starts offering commodities for <10 basis points, I might get interested in investing. Until then, if ever, it is interesting to think about, but I will leave my money in broad index funds at low prices.

More gems from the paper.
Overall, our empirical results suggest that an aggregate commodity index improves the risk-return profile of a stock-bond-portfolio for almost all analyzed investment strategies. However, the out-of-sample portfolio benefits of commodities are much smaller than indicated by the in-sample analysis and vary for different commodity groups.
Our empirical results provide evidence that the benefits of commodities critically depend on the individual commodities included in the portfolio.
Our analysis of sub-periods suggests that, consistent with the literature (Cao et al., 2010), commodities do not improve the portfolio performance of a stock-bond-portfolio during the most recent financial crisis period (2008-2012), independent of the asset allocation strategy employed.
As in Bessler, Opfer and Wolff (2013), we find that very long estimation windows of 48 month and more reduce the out-of-sample performance of all asset allocation models in most cases, as the models react too slowly to structural breaks. For very short estimation windows, transaction costs increase substantially, thereby lowering performance net of transaction costs, as well.
Overall, our analysis of the different sub-periods reveals that the benefits of commodities are time depended. In support of the existing literature (Cao et al., 2010) we find that for the most recent time period from 2008 to 2012, the portfolio benefits of commodities diminished relative to earlier sub-periods.
This has led to an integration of the price behavior of commodity markets with that of stock and bond markets (Silvennoinen and Thorp, 2010), resulting in higher correlations of commodities with the traditional asset classes (Tang and Xiong, 2010), and consequently reducing the positive diversification effects. As a result, the benefits of investing in commodities as part of a multi-asset allocation strategy might be substantially lower in the future.
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
wesmouch
Posts: 32
Joined: Wed Apr 30, 2014 4:45 pm

Re: commodities and multi-asset class portfolios

Post by wesmouch »

Most interesting was that commodities improved portfolio performance during a LOW INFLATION period.
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: commodities and multi-asset class portfolios

Post by Browser »

afan - Dr. Bernstein has suggested that commodities were a good idea if you'd thought of it in the past, before the bozos discovered it and Wall Street accommodated them with the invention of dozens of bozo commodity ETFs. Are they a good idea now?
We don't know where we are, or where we're going -- but we're making good time.
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: commodities and multi-asset class portfolios

Post by Random Walker »

Fwiw, I just added 4% CCFs to my heavily tilted 80/20 portfolio by taking 4% from the equities side. I'm really intrigued by the portfolio effects. I'm swayed by Gibson's book and Larry's writings. William Bernstein, from what I've read, is a CCF fan but has not recommended them because of costs and the conversion to contango when the CCFs became popular. Well now DFA offers a reasonable ER and contango doesn't appear to have been a permanent shift.
It is so easy to focus on ER exclusively and not other costs. ER is easy to know and is a certainty. Other costs/benefits such as tax efficiency, portfolio efficiency, turnover, increased potential returns, benefits of patient trading are hard to know, hard to quantify, and not a certainty.

Dave
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: commodities and multi-asset class portfolios

Post by Browser »

How much difference is 4% going to make, particularly taking into consideration the higher costs associated with this type of investment?
We don't know where we are, or where we're going -- but we're making good time.
afan
Posts: 8195
Joined: Sun Jul 25, 2010 4:01 pm

Re: commodities and multi-asset class portfolios

Post by afan »

Some models do suggest benefits from very small allocations to commodities. Of course, these are backward looking, and the value, if any, depends a lot on exactly when you look.
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
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: commodities and multi-asset class portfolios

Post by Random Walker »

Well the ER on the DFA fund is I think 0.34%, which is about the same as the average ER of the rest of portfolio. So cost did not increase. As to the second question, that's the point of CCFS. A small dose has a much bigger impact than one would expect because of the low/negative correlation and the great volatility.
The big reason I'm an asset class junkie, is that good portfolio additions bring annualized return closer to weighted average annual return of the portfolio components. So the CCFs have serious potential for improving portfolio efficiency. I have no idea how big an effect that can be, but I wouldn't be surprised if it overshadows a 0.34% ER.

Dave
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: commodities and multi-asset class portfolios

Post by Browser »

Random Walker wrote:Well the ER on the DFA fund is I think 0.34%, which is about the same as the average ER of the rest of portfolio. So cost did not increase. As to the second question, that's the point of CCFS. A small dose has a much bigger impact than one would expect because of the low/negative correlation and the great volatility.
The big reason I'm an asset class junkie, is that good portfolio additions bring annualized return closer to weighted average annual return of the portfolio components. So the CCFs have serious potential for improving portfolio efficiency. I have no idea how big an effect that can be, but I wouldn't be surprised if it overshadows a 0.34% ER.

Dave
I dunno. Comparing 80% TSM/20% Treasurys to 76% TSM/ 4% Commodities/ 20% Treasurys from 1972-2013 using Simba, the needle barely moves -- well with the range of "noise" IMO. Sharpe ratios are 0.41 and 0.42 respectively. If these things are such great diversifiers, why not go the Permanent Portfolio route and put 25% in? Now you get a Sharpe of 0.60 for 25/25/50. Gold is just as useful as commodities, despite the "story" about commodities. Sharpe is 0.43 for 76/4/20, and 0.56 for 25/25/50. I prefer actual data to stories.
We don't know where we are, or where we're going -- but we're making good time.
normaldude
Posts: 803
Joined: Tue Jan 27, 2009 3:41 am

Re: commodities and multi-asset class portfolios

Post by normaldude »

larryswedroe wrote:Wolfgang Bessler and Dominik Wolff, authors of the April 2013 study “Do Commodities Add Value in Multi-Asset-Portfolios?” analyzed the portfolio benefits of commodity investments.
I downloaded the PDF.

1) Their default stock & bond portfolio was the S&P500 index and Barclays US aggregate government bond index.

2) Most commodity funds don't actually invest in commodities. They trade derivatives. And most bogleheads would probably prefer a set-it-and-forget-it approach, rather than worrying about contango & backwardation periods.

3) Stock market crashes like 2008 are what bother most investors. In 2008..

- Vanguard S&P 500 Index Fund: –36.97%
- PIMCO CommodityRealReturn Strategy Fund: -43.33%
- PIMCO Foreign Bond Fund (Unhedged): -4.05%
- Vanguard Inflation-Protected Securities Fund: –2.78%

My Conclusion: Assuming their default portfolio of "S&P500 index and Barclays US aggregate government bond index", it doesn't surprise me that adding a commodity fund could reduce risk through diversification. However, I think that instead of adding a commodity fund to the default portfolio, it would have been better to add an unhedged foreign bond fund and inflation bond fund. That way, you gain a hedge against inflation, and your hedge doesn't completely implode along with the stock market (like in 2008).
User avatar
Rick Ferri
Posts: 9708
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: commodities and multi-asset class portfolios

Post by Rick Ferri »

It's been my experience in over 25 years of researching and providing advice that commodities story sounds much better in theory than it actually works in practice. The whole thing is a Rube Goldberg machine; lot's of bells, whistles and moving parts just to perhaps get you where you would be with a simple portfolio of low-cost stock and bond index funds.

Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: commodities and multi-asset class portfolios

Post by larryswedroe »

normaldude
The problem with that line of thinking is that it helps in DEMAND shocks but not in supply shocks. As I have pointed out many times CCF hedges some type of risks (supply shocks and inflation shocks) but not others. It is no surprise as many seem to want to believe that a 2008 happened. It happened before in 2001 for example and in 1981. That's why one should think of a portfolio in the whole. So if going to add CCF first it should be with a safer bond portfolio with no call risks or credit risks (which don't combine well with demand shocks) and then consider adding bit more duration. The three work well together. Never been a year when all three had losses. Anyone who followed that DID NOT see the portfolios harmed in 2008 as many critics point out. Now you did not get protection, but CCF not designed to protect against demand shocks. That's like complaining about your home owner's policy because it didn't protect you against the flood---you want that buy flood insurance
And now that you have a low cost vehicle in DFA commodities fund, and one that avoids the negatives of indexing CCF (rolling on the set dates being the big one, and shifting maturities to some degree being the other) IMO it is certainly worth considering.
Larry
wesmouch
Posts: 32
Joined: Wed Apr 30, 2014 4:45 pm

Re: commodities and multi-asset class portfolios

Post by wesmouch »

normaldude wrote:
larryswedroe wrote:Wolfgang Bessler and Dominik Wolff, authors of the April 2013 study “Do Commodities Add Value in Multi-Asset-Portfolios?” analyzed the portfolio benefits of commodity investments.
I downloaded the PDF.

1) Their default stock & bond portfolio was the S&P500 index and Barclays US aggregate government bond index.

2) Most commodity funds don't actually invest in commodities. They trade derivatives. And most bogleheads would probably prefer a set-it-and-forget-it approach, rather than worrying about contango & backwardation periods.

3) Stock market crashes like 2008 are what bother most investors. In 2008..

- Vanguard S&P 500 Index Fund: –36.97%
- PIMCO CommodityRealReturn Strategy Fund: -43.33%
- PIMCO Foreign Bond Fund (Unhedged): -4.05%
- Vanguard Inflation-Protected Securities Fund: –2.78%

My Conclusion: Assuming their default portfolio of "S&P500 index and Barclays US aggregate government bond index", it doesn't surprise me that adding a commodity fund could reduce risk through diversification. However, I think that instead of adding a commodity fund to the default portfolio, it would have been better to add an unhedged foreign bond fund and inflation bond fund. That way, you gain a hedge against inflation, and your hedge doesn't completely implode along with the stock market (like in 2008).

The really bad times for investors are periods like the 1970's with high inflation. Worse than the 1930's on a real return basis.
staythecourse
Posts: 6993
Joined: Mon Jan 03, 2011 8:40 am

Re: commodities and multi-asset class portfolios

Post by staythecourse »

normaldude wrote:My Conclusion: Assuming their default portfolio of "S&P500 index and Barclays US aggregate government bond index", it doesn't surprise me that adding a commodity fund could reduce risk through diversification. However, I think that instead of adding a commodity fund to the default portfolio, it would have been better to add an unhedged foreign bond fund and inflation bond fund. That way, you gain a hedge against inflation, and your hedge doesn't completely implode along with the stock market (like in 2008).
There are many good and/ or bad reasons to add CCF to a portfolio, but it does not pass the "sniff" test of common sense to think commodities will protect you in a financial downmarket any more then foreign bond funds or TIPS. Just think about your feelings back in 2008 when the world was falling apart did you think, "Boy I am scare, but I would feel better if I was in pork belly contracts or investing in other countries governments or even government bonds protecting against unexpected inflation."

There will always be something that does well or not as bad when stocks go down the tube that is just because of random chance. Harry Browne wrote in his book "Best laid plans..." more then 30 yrs. ago now and seems to be the most accurate (paraphrasing) when there is stress in the market folks flee to reserve currency in the world, i.e. U.S. dollar (cash and treasuries especially on the long end) unless the stress is caused by concern of the U.S. dollar itself which causes folks to flee to the 2nd reserve currency in the world which is gold".

One has to be careful of looking at the outcome of small sample sizes and extrapolating a cause and effect.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle
steve_14
Posts: 1507
Joined: Wed Jun 20, 2012 12:05 am

Re: commodities and multi-asset class portfolios

Post by steve_14 »

Commodities speculations, in my experience, are mostly a marketing tool used by the investment advisory industry to make their recommendations look more "sophisticated" than simple do it yourself stock/bond mixes.

If you legitimately need to hedge a commodities position, do use commodity futures. But speculating by individual investors here is just plain dumb.
Roy
Posts: 970
Joined: Wed Sep 10, 2008 9:34 am

Re: commodities and multi-asset class portfolios

Post by Roy »

Rick Ferri wrote:It's been my experience in over 25 years of researching and providing advice that commodities story sounds much better in theory than it actually works in practice. The whole thing is a Rube Goldberg machine; lot's of bells, whistles and moving parts just to perhaps get you where you would be with a simple portfolio of low-cost stock and bond index funds.

Rick Ferri
Rick: I've looked at this for years now. I now agree with you in the main, as they are not my preference.

But I still think the correlative part of the commodities story does work—hedging both some risks of stocks and longer, quality bonds, such that the portfolio as a whole has a smoother multi-decade ride—be it with Gold or CCFs, providing the commodities are drawn from the equity side. And one must take a long view of them and not look just at 2008 or a short time frame. I respect that you have and may disagree with all of this.

But let's say they do work. Even though they are volatile (a good thing), it does seem one needs to have much more of them in the portfolio to protect the portfolio as a whole to a meaningful degree, in my opinion. (And I've looked at the backtest results that showed what a small amount does, and am not that impressed, even as the correlative story does work.)

This is where things like the Permanent Portfolio come into it (with a meaningful 25% commitment, regardless of what one thinks of the concept), though I'm not suggesting that approach is better than traditional, conservative portfolios. And the good commodities funds, except for Gold ETFs, are still pricey or have reasonable expense but need advisors that amp the cost. But Harry did realize a significant commitment was needed to do what he envisioned; again, not saying that is the way to go. But the same is true with TIPS for the protective quality they provide. You can't just have a little bit to protect the whole portfolio significantly from unexpected inflation risk—they must dominate the portfolio composition to do that, in my view.

By now I know they aren't for me, but they may have some utility for others who have particular risk aversions, if purchased at reasonable costs/fees and comprise a meaningful allocation in the portfolio.
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: commodities and multi-asset class portfolios

Post by Browser »

As I said before, the tiny fraction of assets that Larry recommends for commodities are a pretty slim insurance policy even disregarding whether the insurance will actually pay out when it's supposed to. Seems like a lot of trouble for nothing. Besides, the best commodities have done is to help defray short term losses. Over time, they're a drag on portfolio returns IMO.
We don't know where we are, or where we're going -- but we're making good time.
User avatar
midareff
Posts: 7711
Joined: Mon Nov 29, 2010 9:43 am
Location: Biscayne Bay, South Florida

Re: commodities and multi-asset class portfolios

Post by midareff »

Browser wrote:How much difference is 4% going to make, particularly taking into consideration the higher costs associated with this type of investment?
Exactly........ a 2% outperformance by 4% of a portfolio is a .08% gain...... and that is assuming the 2% is a given....which it certainly is not. A 1% outperformance is .04% and realistically is no better than a rounding error. Again, the 1% gain is a big if. I think I, and most others, have much larger fish to fry.
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: commodities and multi-asset class portfolios

Post by larryswedroe »

All this analysis misses the point. The impact is not on the average return but on cutting the TAILS, but of them
I would add this, it's interesting when people recommend adding say 3-5% of Real Estate as good diversfier, but if recommend considering commodities of similar amount that is "pooh-poohed"
FWIW, diversifying across as many factors as possible--where they can add value ---is the most prudent strategy as we don't know what factors will do well when, so adding exposure to factors that have low/no correlation is a good idea. That's why invested in the AQR multi-asset class strategy

Best wishes
Larry
normaldude
Posts: 803
Joined: Tue Jan 27, 2009 3:41 am

Re: commodities and multi-asset class portfolios

Post by normaldude »

larryswedroe wrote:normaldude
The problem with that line of thinking is that it helps in DEMAND shocks but not in supply shocks. As I have pointed out many times CCF hedges some type of risks (supply shocks and inflation shocks) but not others. It is no surprise as many seem to want to believe that a 2008 happened. It happened before in 2001 for example and in 1981. That's why one should think of a portfolio in the whole. So if going to add CCF first it should be with a safer bond portfolio with no call risks or credit risks (which don't combine well with demand shocks) and then consider adding bit more duration. The three work well together. Never been a year when all three had losses. Anyone who followed that DID NOT see the portfolios harmed in 2008 as many critics point out. Now you did not get protection, but CCF not designed to protect against demand shocks. That's like complaining about your home owner's policy because it didn't protect you against the flood---you want that buy flood insurance
And now that you have a low cost vehicle in DFA commodities fund, and one that avoids the negatives of indexing CCF (rolling on the set dates being the big one, and shifting maturities to some degree being the other) IMO it is certainly worth considering.
Larry
Larry,

Ok, one thing that I'm concerned about with your ETF.com article is that you talk about investing in commodities in general, without warning about indexed commodities funds. In fact, the 2nd page of the article repeatedly talks about the benefits of the commodities index.

This is a bit of a problem because bogleheads have been telling people that index funds are generally better than actively managed funds, and not to bother with managers who try to beat the index. With commodities funds, it's the opposite (indexed commodities funds are terrible, and actively managed commodities funds might be better).

Most commodity funds invest in derivatives. The exception are gold ETFs like GLD, which hold physical gold in vaults. So here are the main ways that retail investors could invest in commodities..

a) Commodities Fund, Indexed, trading in derivatives. Like USO. The problem is that we all know that these commodities index funds can get wrecked, like in periods of heavy contango (2009). In that period, USO holders got destroyed, while Morgan Stanley, Goldman Sachs, Citicorp stored massive amounts of physical oil in oil tankers, and booked risk-free profits. They rented 1 in 12 of the largest oil tankers in the world, that if lined up end to end, would have stretched out for 26 miles.

Image

http://www.businessweek.com/magazine/co ... 970461.htm

http://en.wikipedia.org/wiki/Oil-storage_trade

It should also be noted that USO started trading on 4/10/06 at $68.82, and has FALLEN to $37.23. In the same time period, the price of oil has RISEN from $84.07 to $102.18. This is an absolute disaster for anyone who bought USO, thinking it would somewhat track the price of oil.

b) Commodities Fund, Actively-Managed, trading in derivatives. This is basically paying a fund manager to trade derivatives all day, hoping he can beat an index. Choosing these funds is not obvious to bogleheads or the public, since we've been telling everyone the exact opposite with stock & bond funds (that index funds are usually better than actively-managed funds).

DFA Commodity Strategy Portfolio (DCMSX) seems to have done well, but it's fairly new (founded 11/2010), so it didn't capture the above brutal 2009 contango period, when USO got obliterated. It's also not available to the general public.

c) Physical Commodities. I don't think people want to store barrels of oil in their house. The stuff smells, leaks, and can blow up the house.

d) Gold ETF. Like GLD. Essentially a gold index fund, holding physical gold in vaults. Since they're not rolling futures contracts, in this case, following an index is perfectly fine. However, the tax treatment is a bit annoying, since GLD regularly sells off bits of gold to pay their bills. When you do your taxes in TurboTax, most Vanguard funds take mere seconds since all the cost basis is provided. But with the GLD distributions, you're supposed to manually calculate them with a complicated formula. There are online tools to help, but it can still be annoying. And unlike options "e" and "f" below, you can never take physical delivery of your gold.

e) Gold Accounts. Like GoldMoney.com, SGPMX.com, etc. A company buys gold for you, stores it in vaults around the world, supposedly audited, and with most firms you can theoretically ask for physical delivery.

f) Physical Gold. Like buying gold bullion coins (Canadian Gold Maples, etc), storing them in a home safe, safety deposit box, overseas private vault, etc.

You're saying that option "b" is something to consider. But I'm really hesitant about sending retail investors down this road. Another financial advisor once said that if a person can't explain the difference between contango and backwardation, then they shouldn't invest in commodity funds/ETFs. Most retail investors are in this category.

In the past, I've mentioned that option "e" and "f" are something to consider, since they can also add sovereign diversification (sometimes having offshore assets can be useful, like when my grandparents had to flee their home country because of civil war), and tangibility diversification (you can hold physical gold in your hand, put a million dollars of gold in a suitcase, trunk of a car, drive across the country, etc). But "e" and "f" have big risks (fraud, theft, violence), so it's only for people who understand this market very well, and how to mitigate those risks.

For most retail investors, I still think that adding unhedged foreign bond funds (like PFUIX) and inflation bond funds (like VIPSX / VAIPX) are the most sensible way to hedge against inflation risks. Someone like my mom can set it, forget it, and get on with their life. And I don't have to complicate their life by saying "I've been saying index funds are generally better than actively managed funds, but now let's complicate that by adding commodities funds that trade derivatives, where the actively managed funds are better than index funds.."
Last edited by normaldude on Sun May 18, 2014 2:30 pm, edited 1 time in total.
dl7848
Posts: 440
Joined: Tue Mar 25, 2014 12:46 am

Re: commodities and multi-asset class portfolios

Post by dl7848 »

staythecourse wrote:
Just think about your feelings back in 2008 when the world was falling apart did you think, "Boy I am scare, but I would feel better if I was in pork belly contracts or investing in other countries governments or even government bonds protecting against unexpected inflation."
Comfort level is a personal thing that everyone has to eventually figure out. I have found, through trial and error, that the one thing that makes me reasonabily happy (or at least not horribly unhappy) during a major downturn, is getting monthly investment income. That's like a reward for holding on and not panic-selling. If I had no periodic income, and especially no monthly income, and I had commodities in my portfolio in addition to stocks during 2008, I would have been an extremely unhappy camper.
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: commodities and multi-asset class portfolios

Post by larryswedroe »

DL
If you took the CCF from the equity side it would have made very little difference to your portfolio in that period, and if you had done as I suggested, adding duration to the bond side and done so likely you would have been better off

normal dude
First, just like ANY asset, valuations matter. With stocks it's P/E, with bonds yield and with commodities it's contango vs backwardation.
Now you can be totally passive in any asset or you can be mostly passive and decide that extremes should lead to different decisions

With stocks for example I personally sold all but value in 98 when valuations got excessive and then developed and stayed with the "Larry Portfolio"

With TIPS I have loaded up when yields were very high and gotten completely out of them when they fell well below historical real yields

With CCF, IMO it's no different. The evidence is that when you have backwardation that tailwind leads to generally good returns and vice versa. There are even funds that play that long-short strategy, just as they do with the "carry trade" or shift maturities as yield curves shift. Personally I sold my own commodity position when oil went into big contango (of course gave up my unexpected inflation hedge, but I shortened bond maturity at same time---the yield curve had flattened greatly) and while ago when curve steepened again and oil went into backwardation and stayed there (for last year now has been backward basically, and pretty large for last several months) I went longer with my bonds and added CCF again.

The PIMCO fund does shift maturity fully, based on the cheapest part of the futures curve, "unfortunately" I was only able to convince DFA to shift maturities on 40% of the portfolio, but they like PIMCO fully avoid the five trade dates when futures roll (that alone has added significantly to returns).

so like anything else you can stay passive all the time (right IMO for most people, otherwise they end up confusing strategy and outcome and then lose discipline and make mistakes) or take actions when valuations reach extremes.

Having said that, and even despite basically 20 years of declining inflation and no real supply shocks, and the period of extended contango, even with that CCF being added hasn't hurt portfolios (of course that may not be the case in the future). What might have happened if we had supply shocks or unexpected inflation?

I hope that is helpful
Larry
dl7848
Posts: 440
Joined: Tue Mar 25, 2014 12:46 am

Re: commodities and multi-asset class portfolios

Post by dl7848 »

larryswedroe wrote:DL
If you took the CCF from the equity side it would have made very little difference to your portfolio in that period, and if you had done as I suggested, adding duration to the bond side and done so likely you would have been better off
Larry,

Agreed on duration on the bond side, which I do have.

But wrt to commodites vs TSM, I would be looking at more than just the absoulte losses.

This chart shows that the commodity losses (DBC) would hurt more than the TSM losses because the DBC trajectory was basically straight up, then straight down. There would enormous regret at not having sold at the top. These types of parabolic rises followed by a crash make the pain far worse. The TSM decline was more gradual until the final phases, i.e, the terminal plunge. There was more time to get used to it. Put differently, DBC was more volatile on both the upside and the downside and it makes a difference in how you experience the crash.

Obviously, if DBC (or similar) is only a small part of one's portfolio, it won't make that much of a difference, but during a market plunge, people aren't rational and are likely to compound their sense of self-loathing by examining every mistake they have made. In the case of DBC, it would have been: "If only I had sold at the top while I had the chance."
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: commodities and multi-asset class portfolios

Post by larryswedroe »

dl
The losses were about the same for a globally diversified portfolio, depending on how you tilted, but that is ONLY if you look in ISOLATION, meaning that if you don't extend bond maturity.
If you added CCF and extended bonds you came out ahead because CCF did about as bad as the stocks but longer bonds did better than shorter bonds and now you own more of them

And this is the big benefit, cutting the tails. Of course you'll give up some right tail as well. For the retire for whom sequence and worst cases losses matter, this is overlooked benefit IMO of CCF.

Larry
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: commodities and multi-asset class portfolios

Post by Browser »

larryswedroe wrote:
DL
If you took the CCF from the equity side it would have made very little difference to your portfolio in that period, and if you had done as I suggested, adding duration to the bond side and done so likely you would have been better off
This has been pointed out before in another thread. If you had added duration on the bond side, spun around 3 times and said "abracadabra" it would have helped too. Commodities had nothing to do with it.
We don't know where we are, or where we're going -- but we're making good time.
Roy
Posts: 970
Joined: Wed Sep 10, 2008 9:34 am

Re: commodities and multi-asset class portfolios

Post by Roy »

Browser wrote:
larryswedroe wrote:
DL
If you took the CCF from the equity side it would have made very little difference to your portfolio in that period, and if you had done as I suggested, adding duration to the bond side and done so likely you would have been better off
This has been pointed out before in another thread. If you had added duration on the bond side, spun around 3 times and said "abracadabra" it would have helped too. Commodities had nothing to do with it.

For me, you are correct, Browser, though perhaps absent the magic!

But if I hear you right, I think having longer maturities at the right time is really what mattered—and having them in the first place as part of the plan. Clearly, one could have extended maturities at the wrong time, had another reality than 2008 played out, where in retrospect it is easy to say one should have gone longer—because it worked so well that time. Lot of moving parts here that are very imperfect to suss out, at least to my paygrade.

My question is if one had a portfolio dominated by High Quality bonds—say 70% or more, say Short Term high quality, how many years of maturity does one add for 3-5% of CCFs? and what was the maturity in the first place, 3-5 years? And how long does one really go? Begs lots of questions. I'm doubtful that 3-5% of anything can confer all this protective quality, above and beyond the proper maturity (short-intermediate, sometimes in TIPS) and having a low beta portfolio, which typifies the "Larry", for example.
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: commodities and multi-asset class portfolios

Post by Random Walker »

If someone is adding CCFs now, would you recommend starting to lengthen bond duration with new portfolio additions at this time? Also, does adding CCFs affect the ratio nominal bonds to TIPs one chooses? Thanks

Dave
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: commodities and multi-asset class portfolios

Post by larryswedroe »

Browser
It's been "pointed out" and simply WRONG, The data shows if you did both you benefited, Now one would not have extended duration if you did not own the commodities as a hedge ---so the statement is obviously wrong
Larry
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: commodities and multi-asset class portfolios

Post by Browser »

So, let's look at the period 1972-1981, which was characterized by rising interest rates as well as some demand shock events (OPEC). A simple portfolio of 50% TSM / 50% 5 Yr Treasurys returned a compound annual growth rate of 6.7% with an SD of 11.5%, Sharpe ratio = 0. How did we do if we put 5% from stocks into commodities and lengthened the bond duration? An allocation of 45% TSM, 5% CCF, 25% 5 Yr Treasurys, 25% 20-year Treasurys returned 6.4% annually with an SD of 10.6%, Sharpe = -0.1. Virtually identical risk-adjusted returns for the two allocations during a period of secular rising interest rates -- which is what everybody is all nervous about now. If you hadn't lengthened bond duration you did a little better, but not significantly. A simple portfolio of stocks plus intermediate term Treasurys is a pretty hard bogey to beat. Not worth the fiddle, IMO.
We don't know where we are, or where we're going -- but we're making good time.
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: commodities and multi-asset class portfolios

Post by larryswedroe »

RandomWalker
It's all relative. If you add CCF you can take more duration risk as bonds and CCF historically have been strongly negatively correlated and there is very good logic behind it as well.
And the yield curve is still relatively steep, which is historically when term risk has been best rewarded, and thus when I tend to go longer. Which is what I have done and again bought CCF to hedge that risk
Now of course if you own TIPS don't need that hedge, but most higher net work people hold most of their bonds in munis, which means you do need the hedge.
Larry
afan
Posts: 8195
Joined: Sun Jul 25, 2010 4:01 pm

Re: commodities and multi-asset class portfolios

Post by afan »

larryswedroe wrote:RandomWalker
If you add CCF you can take more duration risk as bonds and CCF historically have been strongly negatively correlated and there is very good logic behind it as well.
Larry
Over the past 7 years, the correlation of PowerShares DB Commodity Index Tracking Fund, DBC, with AGG (iShares Core Total Aggregate US Bond ETF) has bee 0.01, positive, effectively zero, and hardly strongly negative.

Correlation with VWITX, Vanguard Intermediate Term Tax Exempt bond fund has been -0.09. One could debate whether this is effectively zero, I would say yes, but again hardly "strongly negative"

With Vanguard Long Term Tax Exempt bond fund (VWLTX) correlation is -0.03. Same interpretation.

Investing plans that are based on strong negative correlation of commodities with bonds do not seem to be based on a strong foundation.


How high is an 0.34% expense ratio? For some, this is low cost. For me, and I suspect many Bogleheads, that is unconscionably high. I have not put money in a fund with a ratio that high for many years. Can't see a reason to start now. If they get the ratio down to a respectable 10 basis points or less then it would not be ruled out on cost alone. There would still have to be a reason for it, but at least it would be possible.
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
afan
Posts: 8195
Joined: Sun Jul 25, 2010 4:01 pm

Re: commodities and multi-asset class portfolios

Post by afan »

Five year correlations of DBC with
AGG -0.14
VWITX -0.12
VWLTX -0.15

Of course, this includes 5 of the same years as the 7-year results

3 year correlations
AGG -0.11
VWITX -0.06
VWLTX -0.12

Of course, this includes 3 of the same years as the 5- and 7-year results.

I don't have data for longer term.
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
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: commodities and multi-asset class portfolios

Post by larryswedroe »

afan
7 years is virtually meaningless in correlations
70-13 annual correlation of GSCI and LT governments -0.2

There was also no single year till last year when both had negative returns, and last year CCF lost just 1%
And stocks did well. The three work together, with no single year where all three lost money

Larry
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: commodities and multi-asset class portfolios

Post by larryswedroe »

browser
73-4 we had the supply shock and CCF certainly helped by taking it from equities even if lengthened the bond side some.
That cut the tail risk, just as stated

Using your example of 50% S&P 500 and 50% 5 year and then adding 5% GSCI from equity and splitting bonds between 5 and 20 year cut loss from -7.8% a year to 4.7%
I would say that is not bad for a small allocation of 5%, while also more than doubling duration risk (which I would not suggest)

And it certainly helped cut losses in 2008 as well

And it would have helped in the third big drop in 200-02,
Larry
afan
Posts: 8195
Joined: Sun Jul 25, 2010 4:01 pm

Re: commodities and multi-asset class portfolios

Post by afan »

Larry,

But 7 years is all the data I could find on real world performance of real world portfolios. If you have longer, then great, please cite it.

I thought the purpose of commodities in this context was to reduce overall volatility while more or less maintaining returns. Would not 7 years be long enough to see the effect of controlling volatility? Particularly the last 7 years when we have had periods of record volatility? If not, then how long a period does one need to wait to see this lower volatility/higher Sharp ratio materialize?

Do you consider -0.2 correlation, 0.04 R2, "strongly negative"? I might have placed the threshold for that description more at -0.7 R, 0.49 R2. I would call -0.2 something more than "near zero" but far from stongly negative.
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
afan
Posts: 8195
Joined: Sun Jul 25, 2010 4:01 pm

Re: commodities and multi-asset class portfolios

Post by afan »

larryswedroe wrote:afan
7 years is virtually meaningless in correlations
Larry
I think that overstates the case. It is not meaningless for "strong" correlations. One only needs larger sample sizes when the effect sizes are small.

The 7 year correlations of AGG with
VWLTX 0.39
VWITX 0.43

7-year DFSCX (DFA microcap stock) with VTI (total stock market) 0.94

VWITX with VWLTX 0.97

Now that is strong.
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
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: commodities and multi-asset class portfolios

Post by Browser »

larryswedroe wrote:browser
73-4 we had the supply shock and CCF certainly helped by taking it from equities even if lengthened the bond side some.
That cut the tail risk, just as stated

Using your example of 50% S&P 500 and 50% 5 year and then adding 5% GSCI from equity and splitting bonds between 5 and 20 year cut loss from -7.8% a year to 4.7%
I would say that is not bad for a small allocation of 5%, while also more than doubling duration risk (which I would not suggest)

And it certainly helped cut losses in 2008 as well

And it would have helped in the third big drop in 200-02,
Larry
Illustrations of the benefits of commodities seem to focus on short-term, episodic returns. But rarely on returns over longer horizons of 10 years or longer. I agree with Asness that there is a difference:
Short-term crashes can be painful, but long-term returns are far more important to wealth creation and destruction
When I look at alternative investments like commodities, risk-adjusted returns over the long run are not noticeably improved. We do see that portfolio volatility is reduced somewhat, but at the cost of compound returns. If you want a smoother ride to lower returns, then commodities should help.
We don't know where we are, or where we're going -- but we're making good time.
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: commodities and multi-asset class portfolios

Post by larryswedroe »

Browser

I have shown the benefits over the longest periods we have data for, as well as the KEY periods, when shortfalls are greatest, so your statement about short period is just incorrect.

I would add this, if you think there is no significant improvement, you'll also note no harm, and the data covers mostly a period of falling inflation, basically since 1981 we have had bond rally. Now imagine an alternative universe where another outcome had happened and we had the unexpected inflation. In other words even though for most of the period the insurance wasn't needed it did no harm. When it was needed the most it helped cut the tails, and thus reduced the risks/odds of running out of money while still alive. So no harm in long term, helped cut tail risk. And I'll note that live funds run well would have outperformed the passive indices --very easily in fact. Just as it was easy to beat the R2k. Thus the data historically understates, not overstates, IMO what would have happened.

If you conclude it isn't worth it, fine, but the data doesn't show that, at least IMO.
That's why I have owned CCF for my own portfolio. I drew a different conclusion

Best wishes
Larry
User avatar
packer16
Posts: 1488
Joined: Sat Jan 04, 2014 1:28 pm

Re: commodities and multi-asset class portfolios

Post by packer16 »

One of the issues I see is commodities have the risk of stocks but the returns of bonds. They also are not productive assets. As Warren Buffett has laid out in the 2011 Letter to Shareholders the asset classes of productive assets are stocks and real estate. Commodities are one of the assets where buyers hope another buyer will buy there asset for a higher price. The only exception I can think of is commodities sell below extraction value. I would think you can get many of the advantages of commodities and still get productive returns from real estate. Commodities I also think are in a long-term decline as more productive ways of using the commodities are developed. So with commodities you have periodic increases in value in a long-term secular bear market. It would be interesting to see if you increased the allocation to real estate in place of commodities if that would not provide higher returns with the same amount of risk.

Packer
Buy cheap and something good might happen
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: commodities and multi-asset class portfolios

Post by Browser »

Larry -
Respectfully, I guess we'll have to agree to disagree. There's no magic about cutting left tail risk -- the cost of doing that is cutting right tail gains as well; at least that's what the data look like to me. The net result is lower long term compound returns with somewhat lower portfolio volatility, not enough to be significant to long term investors, IMO.

I compared a portfolio of (1) 50% TSM / 50% Int Treasurys to a portfolio with (2) 40% TSM / 10% Commodities / 50% Int Treasurys for all rolling return periods of 3, 5, & 10 years beginning from 1972 using Simba's data. For 3 year rolling return periods, the average return of Port #2 (the commodities portfolio) was higher than for Port #1. But for the 5 year periods, the average returns were about the same, and for the 10 year periods, the average returns of Port #1 were higher than for Port #2. The average volatility of Port #2 was lower for all the rolling return periods. My conclusion: the longer the time period, the more likely that compound returns will be higher without commodities; the main effect of commodities is to reduce (somewhat) the expected volatility of the portfolio. If that's important to you, commodities could be a useful addition; but if higher compound returns are what's important, they probably aren't because the volatility reduction is not great enough to make much difference. Commodities have a mixed record in helping investors "stay the course" during equity bears. Most notably they lost even more than equities in the 2007-08 meltdown, so you can't always count on pork bellies to save your bacon.
We don't know where we are, or where we're going -- but we're making good time.
afan
Posts: 8195
Joined: Sun Jul 25, 2010 4:01 pm

Re: commodities and multi-asset class portfolios

Post by afan »

Browser,

Did you calculate the Sharp ratios and skewness of the portfolios?
Did you calculate the effect of taking an incremental amount from stocks and adding it to bonds, rather than to commodities? For example shifting from 50% stock, 50% int term treasuries to 40% stocks and 60% int term treasuries? This would also reduce volatility, is available at very low cost and has positive expected returns. Did you look at what it would do to the risk and return of the portfolio as opposed to adding commodities? If you found lower returns and lower volatility then your portfolios might have had higher Sharp ratios with commodities.

Larry suggests that the Sharp ratios of the portfolios will be higher with commodities than without. He does not offer an opinion about skewness, but investors should and do value positive skewness.

Larry,

If we are interested in "KEY periods, when shortfalls are greatest", then we should be interested in correlations over short terms, should we not?

Everyone,

In the Bessler and Wolff article Larry cited, data from 1986 to 2012, the US aggregate government bond index correlation with GSCI was -0.13, significant, but again I would not characterize it as "strong". GSCI had a 0.13 positive correlation with T bills, not significant. The correlation of GSCI with a blend of T bills and bonds would be somewhere in between.

Again, this paper suggested that the commodities contribution was modest at best, varied widely depending on the time period and the mix of commodities included and that the optimal weights of commodities were very strongly dependent on the time period. When they compared results with weights based on in-sample data (which would never be available to a real world investor) to out-of-sample data (which is all anyone would ever have), they found the out-of-sample performance was not nearly as good. Using weights based on in-sample data is the equivalent to calculating the historical returns to active funds by picking an optimal portfolio of such funds based on past performance and assuming this will predict returns to that mix going forward. In other words, meaningless.
Last edited by afan on Mon May 19, 2014 6:56 am, edited 1 time in total.
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
normaldude
Posts: 803
Joined: Tue Jan 27, 2009 3:41 am

Re: commodities and multi-asset class portfolios

Post by normaldude »

larryswedroe wrote:The problem with that line of thinking is that it helps in DEMAND shocks but not in supply shocks.
It should be noted that physical gold can serve as a hedge in both scenarios (demand shock, and supply shock).

- In the 1970s inflation period, gold price went from $40.80 to $612.56 (+1401% over 10 years, 1971-1980 price avgs)

- In the 2000-2002 crash period, gold price went from $282.05 to $347.20 (+23% over 3 years, 2000-2002)

- In the 2008 crash year, gold price went from $846.75 to $869.75 (+3% over the year)

http://www.kitco.com/charts/historicalgold.html
Roy
Posts: 970
Joined: Wed Sep 10, 2008 9:34 am

Re: commodities and multi-asset class portfolios

Post by Roy »

afan wrote: Did you calculate the effect of taking an incremental amount from stocks and adding it to bonds, rather than to commodities? For example shifting from 50% stock, 50% int term treasuries to 40% stocks and 60% int term treasuries? This would also reduce volatility, is available at very low cost and has positive expected returns.
I see this as the first consideration for someone risk averse and perhaps the only one needed. It's like getting the big rocks in a container before considering adding gravel or sand. Since 1972 this approach showed lower Standard Deviation and smaller Drawdowns at just a bit lower CAGR, lower cost and complexity (not that more complex is necessarily bad). And had this been done with a fully diversified portfolio (say, across factors and asset classes), the returns would be higher.
Roy
Posts: 970
Joined: Wed Sep 10, 2008 9:34 am

Re: commodities and multi-asset class portfolios

Post by Roy »

normaldude wrote:
larryswedroe wrote:The problem with that line of thinking is that it helps in DEMAND shocks but not in supply shocks.
It should be noted that physical gold can serve as a hedge in both scenarios (demand shock, and supply shock).

- In the 1970s inflation period, gold price went from $40.80 to $612.56 (+1401% over 10 years, 1971-1980 price avgs)

- In the 2000-2002 crash period, gold price went from $282.05 to $347.20 (+23% over 3 years, 2000-2002)

- In the 2008 crash year, gold price went from $846.75 to $869.75 (+3% over the year)

http://www.kitco.com/charts/historicalgold.html
In the discussions here on commodities over the years, Gold seemed to confer similar benefits, is readily available these days in ETFs, for those who want to go that way (I don't use any commodities but respect the views of those who do). In some of those discussions, Larry stated the reasons for his preference for CCFs over the shiny metal.

This was one of many discussions where poster Lbill looked at Gold :

http://www.bogleheads.org/forum/viewtop ... st=1305954
Topic Author
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: commodities and multi-asset class portfolios

Post by larryswedroe »

Re Gold
I disagree on gold, though I do think it provides a hedge against SOME things, like geopolitical risks
But it won't help you necessarily against a supply shock to energy, as the price of gold may not go up as its supply isn't impacted--I would not look at 73-4 at all for this as cannot separate the freeing up of gold price after 40 years of being fixed.

And gold certainly is not a hedge against inflation, that idea is without question wrong. It cannot be if you can have a 24 year period when it loses 85% of its real value. That's why I don't prefer gold. Though I do agree it has some hedging value

Re shorter correlations, CCF still negatively correlated to bonds even at monthly level, though I would not care at that short horizon. You care about period likely will be rebalancing, so for most people annual will be fine

Browser, and of course it's perfectly fine to disagree on conclusions-
We both agree that there has been little difference in long term results---My conclusion is that given the left tail was cut, and MOST investors are risk averse (care much more about left tail than right tail) then they should prefer a portfolio with less skewness and less kurtosis, especially left tail risk. Only those who are risk lovers should not care, or risk indifferent. I don't know any such person. I would also add, and I think every financial advisor I know would agree, that cutting left tail risk increases the odds that the investor will stay disciplined, adhering to their plan, and more likely to also rebalance. Those are IMO very important issues. And finally again, we have had a 30+ year period of declining bond yields and falling inflation basically. As Nicholas Taleb pointed out in his first book, only fools don't consider what would have happened if alternative universes had shown up. In this case adding CCF didn't hurt at all and did reduce left tail risks (and improved odds of not running out of money in withdrawal phase, an important issue)---now what would have happened if we had the unexpected and rising inflation, the environment when CCF produces it' best returns? Given the above, that's why I think most investors, especially conservative ones, should consider adding small allocation, diversifying their risks. You can draw different conclusions of course.

Larry
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: commodities and multi-asset class portfolios

Post by Browser »

I'll just restate that if CCF are the magic sauce, a small allocation doesn't seem like enough to make a difference that makes a difference. Seems like much ado about playing around the edges. Considering the higher ER and costs of rebalancing, it's not for everyone.
We don't know where we are, or where we're going -- but we're making good time.
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: commodities and multi-asset class portfolios

Post by Browser »

I agree that gold seems to have played a similar role to CCFs as a portfolio diversifier, maybe even better. For example, since 1972 there have been 10 annual periods in which TSM had negative returns. In 7 of those CCF had positive returns and in 6 gold had positive returns. Over the 3/10 years that CCFs had negative returns the average loss was -27%, while over the 4/10 years gold had negative returns in two of those the loss was barely negative and overall the average loss was -11%.
We don't know where we are, or where we're going -- but we're making good time.
Post Reply