Bogleheads

[thirdcamper]

I am a longtime investor by the methods advocated by this forum. I am considering commodities as a possible diversifier. I would like people's notions about this:

First, is there a substantial advantage to a general commodities basket fund over gold? (I already have GLD at 2 percent of my total holdings.)

Second, is this a poor time to diversify with a commodities fund? (Not out of desire to market time but to avoid chasing past returns and buy at a high point given recent inflation in gold and oil prices.)

Third, what vehicles for this do people recommend? I'm a little confused by commodities ETFs that seem somehow to build in futures markets; since I don't believe in investing in things I don't understand, I'd like the specific ETFs explained to me in layman's terms, if possible.

Fourth, am I correct that Vanguard offers no appropriate fund for this?

Fifth, what percentage of total holdings would those who believe in this as a diversifier place in commodities? I was thinking no more than 2 percent, but I see sample portfolios here with far more. I wonder if this reflects recent market trends and fads rather than sound diversifying.

Thank you.

Thirdcamper

[SpringMan]

Here is a link that may be useful.
http://www.altruistfa.com/commodityfunds.htm
I don't own them but like you I have been pondering the idea. Larry S. likes PCRIX, which concurs with the link above. Vanguard brokerage allows purchase of PCRIX for a minimum $25K. Another reason I am probably going to pass on it is that I am trying to reduce my total number of funds, not increase it. The asset class has had a good run. I realize there is a difference between CCFs and VG Precious Metals and Mining fund, which I own, but there seems to be positive correlation recently. I guess I am in the not to decide, is to decide camp.
Best,
_________________
SpringMan

[DaveTH]

[stratton]

[thirdcamper]

Thanks for these replies. Actually my 2 percent figure for gold in my total portfolio is a mere target. Right now my GLD shares are closer to 3 percent, after gold's runup. I plan to hold those shares, not sell them. I am putting new funds into other asset classes until the proportion is reduced back to around 2 percent.

Taken as a percentage of equities alone, my gold holdings are higher, of course--4.3 percent--but I don't really think of gold's function in my portfolio in relation to equities alone. I prefer to think of this commodity in relation to the total portfolio.

I can see the virtue of going from one commodity to a diverse group of commodities, though I wonder if all the others (wheat, corn, etc.) operate as variantly from equities as gold, and therefore I wonder if the diversification value is actually reduced by adding more commodities. I'd like to hear some thinking on this.

I would also very much like to hear from others about the wisdom or setup of these funds rooted in commodities futures. GLD may be limited to one commodity, but at least the firm running it buys the commodity and holds it, making its price directly correlative to the price of gold proper, and its effect is to function as if one actually owns gold. Futures seem speculative to me and I am dubious about the value of adding them to the portfolio.

As for the smug idea that it's "wasting your time" to have any asset class below 5 percent in one's portfolio, that obviously overlooks that asset allocation is highly personal, based on one's own estimation of risk. There is no law of nature dictating 5 percent of any given asset class. For me I would consider it, especially now, to be foolhardy to have 5 percent in gold or oil. But I wouldn't impose my own caution on anyone else. Comfort levels in asset allocation simply can't be dictated universally.

[Rick Ferri]

Over the last 5 years I have benefited significantly from my investment in companies that are in the commodities markets.

My precious energy stock holdings have achieved annual compounded returns in excess of 35% over the past 5 years. My precious metals & mining stock holdings have achieved annual compounded returns in excess of 40% over the 5 past years.

Guess what? I achieved those return just by global stock market index funds (US, Europe, and Emerging Markets). About 15% of my global index funds are in stocks of natural resource companies.

Who needs a high cost direct investment in commodities when natural resource companies that a part of a low cost index fund earn those returns?

No one. Stick with real companies that have real operating earnings and can make money even in bad commodities markets.

Rick Ferri

[thirdcamper]

Rick - Thanks for yours. I'm sure ours were being written simultaneously. To explain my own reasoning, I hold gold not to seek super-phenomenal returns, but because the price of gold tends to vary from equities. This has proven true in the past year, most definitely. It is nice to have it as a diversifier, because it assists me in identifying where to put new money to restore asset allocation and thus in theory helps guide me to buy low in the comparative asset classes. This is why I think it helps my portfolio to have gold. I am not necessarily sure about whether I should hold all other commodities, which is why I raised my original question. I remain uncertain about that. Yours, Thirdcamper

[DaveTH]

[DaveTH]

[rkbrashear]

I struggle with whether to add a CCF as well, but it seems like you should either do it or not do it. I don't see the value in adding some small percentage of the portfolio where the results will not be significant.

David Swenson in Unconventional Success book,

[thirdcamper]

Well, my total holdings have now reached a level where the 2 percent counted out in actual dollars seems a great deal to me. So I'm simply not comfortable with gold going too far above that percentage, and that's why I keep it as the target.

I don't consider it a "waste of time" -- it takes hardly any time at all, certainly no more than 5 percent would take out of my day. Granted, it's small in the big scheme, but that's the point. I want it there, and I want it small. It plays a useful diversifying role, without having too much riding on gold, which I don't want. I have no transaction costs. There are no adverse consequences for my small allocation. So I diverge from everyone else. I don't have a problem with that.

There's a possibility I'd increase the percentage if I were going to a more diversified commodity ETF, because then the asset class would be a broader-based thing, but... I'm a little dubious about that.

I'd still be interested if some other reader out there, someone who advocates these commodity ETFs might answer my questions above that are still unaddressed (esp. about the wisdom of dabbling indirectly in futures, and about the relative diversification of total commodities v. gold).

[craigr]

[docneil88]

Agricultural Commodities

[thirdcamper]

Craigr, useful data -- exactly the kind of thing I was hoping would come. Very, very true about going forward, which is exactly why futures make me nervous. I think I'll stick with gold in my plan, not go to commodities more generally. And I will stay at 2 percent of total portfolio. That may strike others as too small, but it works for me. If it plays a role at 5 percent, it also works (less extensively) at 2 percent. Here I'm talking about 2 percent of all mine and my wife's retirement accounts combined, a large dollar amount, and I'm just not interested in going higher on it, no matter how many people post here that 5 percent is necessary. Gold is my only asset class that small -- deliberately so because I'm not persuaded it merits more. In the case of other classes I'd agree with you guys and all my other asset classes fit the 5-35 percent rule of thumb.

[thirdcamper]

I gather these funds tied to commodities future indexes are seen as reliable gauges of agricultural commodity prices. But a question: why don't they trace actual commodity prices rather than futures?

It's possible I could add in the fund you just bought at 2.5 percent of my portfolio, raise my gold target to 2.5 percent, and between the two of them hit the 5 percent everyone else seems to think vital. That I'd be comfortable with, since it wouldn't all be staked on gold.

But I like GLD better than the other commodities indexes I've seen, because GLD actually holds real gold to support its mirroring of the price of gold, and thus it's the equivalent of holding gold.

The futures stuff makes me nervous and seems a bit speculative -- even if it'd be in the form of this index fund and not involve margins trading. I'd like to see a counterargument.

It seems more likely to me that the price of fuel (for tractors, etc.) and the price of fertilizer rising, as well as further pressure on corn to be used as ethanol, will be more important than land in increasing agricultural commodity prices, by the way. There's a lot of rural land whose value has fallen. Cf. North Dakota. So for land to increase to the point where it would actually begin to increase ag prices in this country, a long way. Not so long for rising petroleum prices to put further upward pressure.

[larryswedroe]

Doc
Unfotunately Rick's advice goes counter to mostl of the academic literature and is not backed up by the facts--He just data mined--looking at recent period.

As I have pointed out in my books and here many times, the stocks of producers of natural resources have higher correlation to equities in general than they do to the returns of commodities futures. So they are not a good substitute. That is the long term evidence. Not just last few years.

Also CCF has negative correlation to equities themselves, as well as bonds, making them an excellent diversifier.

And do not make the common error of thinking of an asset class in isolation but in terms of how the asset's addition impacts risk and expected return of the portfolio. The high volatlity of CCF combined with negative correlation to other portfolio assets results in very large diversification return assuming you rebalance. That is the conclusion reached by such academic papers as the ones by Ibbotson, Erb and Harvey and Rouwenhorst----

The historical evidence is that adding CCF to portfolios has improved Sharpe Ratios quite a bit. Also reduced downside risks due to the negative correlations. This year in particular a good example as CCF has provided higher than average returns while stocks have provided in general below average returns (that is what negative correlation means)

[thirdcamper]

Larry,

I understand your case for function of CCF in a total portfolio, but I still am unsure about commodities futures, precisely because they are futures.

For one thing, regarding the long-range data studies: how did they manage to weigh long-term returns in rebalanced portfolios, given that the exchange trading vehicles are relatively new?

For another, I'm still a little perplexed by how the things actually function and why they aren't highly speculative given that they are futures rather than commodities proper. Your position is that risk-averse portfolios should hold these, but on the face of them they seem fairly risky.

I understand that a bit more risk in asset allocation decreases overall risk, but there are some investment vehicles anyone would shy away from. Why are these wise?

Thirdcamper

[Valuethinker]

[Rick Ferri]

Larry said,

[thirdcamper]

Valuethinker, Exactly right - that's why I treat gold in relation to total portfolio rather than in relation to equities. And it's exactly why I only hold a small portion of it, partly as hedge against inflation, partly as rebalancing mechanism.

[SpringMan]

Valuethinker wrote:

[DaleMaley]

I know Rick Ferri and Larry Swedroe greatly disagree about commodities per their other postings on this subject.

Back in early 2003, I wanted to add an investment that gave me some protection from inflation and was based upon natural resources. At that time I had never even heard of funds based upon commodities like we often hear about today (PCRIX actually started on 6/28/02).

My choice back in 2003 was the Rowe Price New Era fund ticker PRNEX. It owns gold mining companies, petroleum companies, aluminum companies, etc. It has a 0.66% ER.

This fund has more than doubled in value with the recent run-up in natural resources. It is now 7% of my portfolio.

From my viewpoint, my portfolio already has 7% in commodities....so why would I bother with adding PCRIX....since for me.....7% is a high enough allocation to commodites (or natural resources).

If I had a 0% allocation to commodities (or natural resources) today, it would be a tough decision on whether to pick PCRIX or PRNEX. According to Morningstar, PCRIX has a 0.74% ER.

It is fun to go to the gas station every week.........at first I curse the high gasoline prices.......then I'm happy because my PRNEX has probably gone up in value.
_________________
Dale
____________________
Most investors, both institutional and individual, will find that the best way to own common stocks is through an index fund that charges minimal fees. – Warren Buffett

[SpringMan]

Dale,
I feel the same way about VGENX (VG Energy) and VGPMX (VG Precious Metal & Mining). I do understand there is a difference between owning equities and owning hard assets. I am working on simplifying my complex portfolio so I have decided on passing on PCRIX. I have to admit that Larry's endorsement makes PCRIX tempting.
Best,
_________________
SpringMan

[thirdcamper]

Three different kinds of commodity funds are now being discussed here:

1) Funds like VGENX (VG Energy) and VGPMX (VG Precious Metal & Mining). These are equities sector funds that invest in the stocks of commodities producers.

2) Funds that are based on the actual commodity (GLD, for example, an exchange-traded fund that is virtually equivalent to owning gold and whose price is almost exactly correlative to price of gold).

3) Funds that are collateralized commodities futures funds (such as PCRIX and the other possibilities of that kind, including the ag-specific fund mentioned by a poster above).

I don't think we should mix these up although they are all somewhat related.

I still remain a bit skeptical about the third option, which is what I've been asking about all along, and hope Larry or someone else expert in this might answer my questions.

[market timer]

Thirdcamper,

Here is a summary of a recent academic paper on the role of CCFs in one's portfolio.

[Robert T]

.
Larry,

The period 1985-2006 – the longest period of data that I am aware of for an energy fund (the Vanguard Energy fund), suggests that it was not a bad substitute – not perfect but not bad.

1. Energy fund correlations with CCF (GSCI) and TSM

[thirdcamper]

Market Timer: Thanks -- That's very useful, and I went and read the academic paper itself just now. I may actually add a CCF fund

As a general note for those considering considering these matters, here are some management/expense fee notes:

streetTRACKS Gold Shares (GLD) = .40%
PowerShares DB Agriculture ETF (DBA) = .75%
PIMCO CommodityRealReturn Strategy Fund A (PCRAX) = 1.24%

As for me, I am going to stick with GLD at 2 percent. I am now seriously considering adding DBA at 2 percent. (I know, I know, between the two they will still not reach 5 percent. So shoot me.) While PCRAX includes gas, oil, etc., I'm not persuaded that the portfolio diversification, rebalancing benefit, and inflation hedging aren't accomplished just as well by a combination of DBA and GLD -- while at a vastly reduced cost.

I'm still not sure I'm going to go in for the CCF (DBA), by the way. I'm going to meditate about this a while.

[SpringMan]

Does anyone know what Jack Bogle thinks regardings CCFs? My guess is that he is not a big fan.
Regards,
_________________
SpringMan

[thirdcamper]

Bogle is against commodities, period. Rick Ferri is expressing the true Boglehead philosophy, if being a Boglehead means following Bogle down the line. From a July 6, 2007, interview of Jack Bogle by Jason Zweig of Money magazine:

[larryswedroe]

Robert--from my upcoming book

A study covering the period 1962–2003 found that the correlation between the returns on CCF and the returns of commodity producers was just 0.40 while the correlation of commodity company stocks to the S&P 500 Index was 0.57. Contrast the 0.57 correlation of commodity producing equities with the negative 0.28 correlation of CCF with equities.

Yanbo Jin and Philippe Jorion, “Firm Value and Hedging: Evidence from the U.S. Oil and Gas Producers,” Journal of Finance (June 2006).

[tokyoleone]

I understand the idea of CCF as a diversifier and their low correlations to stocks. But wouldn't those correlations to stocks increase in a recession? I mean, falling stocks often indicate a poor economy and less money to buy oil, wheat, orange juice, etc.

CCFs seem really popular now because they are doing so well. Being as volatile as they are, it will be interesting to see what supposedly risk-averse investors do when they see their CCF funds falling 30-50%.

CCfs as diversifiers sounds appealing on paper, but I am not sure people could hang on during their long periods of underperformance.

[larryswedroe]

tokyo
The answer depends
IT depends on what causes the 'recession" or bear market-If it is inflationary recession or one that is caused by event risks that causes commodities to rise in value (see 73-4) then CCF will rise while stocks fall

On other hand could have deflationary type recession (see 2000-2) then CCF correlation to stocks rises

But correlations to other assets--the bonds in portfolio--are also important. And here you get very strong and persistent negative correlations.

It is the impact on whole portfolio that matters

[Robert T]

.
Larry,

As I understand, the 1962-2003 correlation numbers come from “Facts and Fantasies About Commodities Futures”.

The paper uses an equally-weighted commodities futures index then compares this to returns of companies ‘closely’ matched with the commodity of interest (equally weighting the returns of 17 of the 36 commodities in the futures index to form a commodity-producing firms stock index). Now if I am reading the construction methodology correctly there were only three commodities in the ‘producing companies’ index in 1962 while there were 12 in the CCF. This gradually increased to about half in 1983 (12 of 25), then 17 of 36 by the end of the period. To me this doesn’t seem like an ‘apples to apples’ comparison ie. comparing the correlation of CCF returns with stocks of commodity producing companies of just under half the commodities in the index (and as the Erb and Harvey paper show there is low correlation among commodities in the CCF, the same of which may be true of commodity producing companies). So its probably not surprising that the correlations came out low (0.4).

IMO the comparison of the Vanguard Energy fund with the GSCI (which is over 70% energy) is a closer apples to apples comparison. This is not to deny the low correlation of the GSCI with TSM, but that in a portfolio context the Vanguard energy fund was not a terrible substitute for GSCI over the 1985-2006 period - or at least as bad as suggested by some of the academic papers (the 1985-2006 period was about the period natural gas, crude oil, unleaded gas etc were included in the equally weighted index in the facts and fantasies paper).

Robert
.

[docneil88]

CCFs; DBA's & DBC's Index Returns

[subrosa]

Hello Robert

[larryswedroe]

Robert
Dont have time to go back and check the papers but that data came from this study
Yanbo Jin and Philippe Jorion, “Firm Value and Hedging: Evidence from the U.S. Oil and Gas Producers,” Journal of Finance (June 2006).

(Now they may have used the G&R data)

Doc
That is not necessarily so re oil producers--examples are that many of them get large % of profits from refineries--and that is different business. Also you could easily see things like an excess profits tax on oil companies or expropriation of their assets, or simply bad management. That is why you get an equity risk premium for investing in those company's stocks.

[subrosa]

The paper is here:

http://som.xjtu.edu.cn/uploadf....734748.pdf

[thirdcamper]

Um, while you are all off in Numbers Land, is there anyone out there willing and able to answer all the more basic questions set forth above?

In addition to above, I now notice that DBC (the general PowerShares commodities future index) is .75% expense ratio, making it too substantial cheaper than the Pimco option. I'm thinking about that rather than pure ag, and despite my suspicions that energy prices won't stay so high. Does the difference in bonds used actually make the Pimco's higher expense ratio worth it?

I'm still wondering about the highly speculative quality of these instruments. Even their prospectuses say "watch out--speculative."

[Prokofiev]

" I now notice that DBC (the general PowerShares commodities future index) is .75% expense ratio, making it too substantial cheaper than the Pimco option"


No. Not cheaper than PCRIX which I believe has a .74% ER. From everything I have read, I would favor PCRIX as recommended by Larry S.
_________________
Everything should be made as simple as possible, but not simpler - Einstein

[thirdcamper]

Unless I'm missing something, PCRIX is the institutional option. It has a minimum investment of $5 million.

PCRAX is what's available for regular investors. It has a load of 5.5% and then an expense ratio of 1.24%. It's expensive.

I'd love to be wrong about the above. Tell me if I am.

[Robert T]

.
Subrosa,

Thanks for the data.

Interestingly the longer time series shows GSCI to have a more significant portfolio impact than the T.Rowe Price fund (relative to the shorter period with the Vanguard fund). The key difference comes in 1973-74 when the GSCI returns were +75% and +40% respectively while the T.Rowe Price fund returns were –2% and –26%. These two years seem to drive much of the difference in impact relative to the shorter period with the Vanguard fund highlighted above.

What’s the reason for the difference? My best guess is that the 1973-74 oil price shock was caused by a restriction in supply and an embargo on oil shipments from the Middle East to the US, and other developed countries. So while international oil prices were high may ‘oil’ companies (particularly service companies which are a larger share of the T.Rowe Price fund) had nothing (or relatively little) to sell so revenues dropped dramatically. If these types of events happen in the future – CCFs will likely outperform a fund of commodity companies i.e. CCFs are a better hedge against event risk.

Neil,

My sense is that the agricultural commodity index is an ‘energy play’ at least in the short term. Much of the price increase of the last year was driven by demand for corn (and to a lesser extent sugar) based ethanol (corn and sugar comprise 50% of the index). This may be the dominant driver of price change in future – at least in the short-term. Some of the energy companies may be investing significantly in this area. Just my take – could be wrong.

Larry,

The extract from the G&R paper –

[Prokofiev]

You can buy PCRIX thru VBS. Minimum purchase is only $25K. Total cost is $35 with no load and ER of .74%. I have done it and I am glad I did . . . although never confuse strategy with outcome. Had it gone down, I would also be happy for the insurance coverage. Just not quite as happy . . .

Cheers, -P
_________________
Everything should be made as simple as possible, but not simpler - Einstein

[thirdcamper]

Okay, that's very useful -- I guess someone else had mentioned it earlier but it hadn't quite registered with me. Vanguard must be buying institutionally, then chipping off bits to its clients?

Prokofiev, by the way, is an excellent moniker. Maybe I'll come back as Purcell.

I do look forward to the day someone unpacking my other questions above. I'm off to read in bed now, though.

[subrosa]

Hello Robert

[stratton]

Let's make this more readable:

[docneil88]

Corn & Sugar for Energy

[heyyou]

3rdcamper-
Gold vs. other commodities: Rebalancing among different commodities is added diversification. As an indexer, I consider that an advantage even if the total return happens to be less, and it will be, since gold and oil are today's hottest commodities. The internal rebalancing of a multi-commodity fund is at low cost to me (part of ER), since I'm not selling and buying among ETFs.

today's pricing: It won't matter in ten years, but the price may dip sometime after your purchase. That is how it is with investments, today looks risky, and no one knows about the future prices. Someone has already suggested a set allocation, rebalanced as necessary. Gold has gone so high, you should rebalance your holding. Buy some other commodities with your profits.

Commodity Futures vs. Commodity-based Equities: A hotly debated topic with no one backing down. I own PCRIX and Large Cap index funds with cap-weighted big oil companies, so I'm covering both camps. To learn of the various ETFs, start your reading with the alturistfa.com link in the second post. At that site, there may be links to papers written on commodities. Buy a Rick Ferri book and Larry Swedroe's "the only guide to a winning investment strategy" (not the bond book with a similar title). In Larry's book, Appendix H on page 267, discusses commoditiy futures in depth.

VG has some equity sector funds including Energy and Precious Metals, but no direct commodity fund.

You and only you have to live with your allocations. My fixed pension needs inflation protection so my 5% of portfolio in PCRIX is the bare minimum for me.

[docneil88]

PCRIX, DBC, DBA, JJA, & RJA

[Robert T]

.
Neil,

Here is a historical corn price chart and a related article.

Robert
.

[thirdcamper]

Thanks, HeyYou, much obliged. I agree about taking some gold profits now and putting them into other commodities. And I appreciate the point about the internal rebalancing of the general commodities funds, which I agree is a significant implicit advantage.

I still have a nagging suspicion I'll be buying high, and I still am a little nervous about the speculative character of futures markets. I've ordered Larry's and Rick's books from the library and will study them first.

I'm not in general a market timer and am interested in this purely for diversification and as an inflation hedge. But I am wary of getting in after a big runup. So it goes -- and anyway, we may be in for much bigger energy and food runups in the future, so my fears about consequences for my portfolio if I buy now may turn out to be unwarranted.

I most likely will keep GLD at 2 percent and add a CCF at 2 percent for a commodities total of 4 percent. I'm leaning toward a general CCF rather than ag-only, but I'm going to think over that and read this debate between Doc and others as it unfolds while thinking about it.