Two Questions--Commodity Index Fund--how much / what index

Have a question about your personal investments? No matter how simple or complex, you can ask it here.
User avatar
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri »

There may be no implied risk in an arbitrage trade, just like there is no implied risk in a T-bill / TIPS trade over the long-term. Nonetheless, it is interesting to note the large number of hedge funds that specialize in so-called 'risk free' arbitrage that lost money and have gone out of business, including Long Term Capital Management.

I have no more to add. Thanks for the conversation.

Rick Ferri
User avatar
retiredjg
Posts: 54082
Joined: Thu Jan 10, 2008 11:56 am

Post by retiredjg »

Enough!

You two are not ever going to agree. The only choice you have is to agree to disagree.

I'm interested in hearing what each position is. I'm not interested (or impressed) when one of you is pointing out how wrong the other is.

The early parts of this thread were somewhat helpful. It has now turned into nothing more than "right-fighting" in my opinion.

As another poster said earlier, losing respect....
Roy
Posts: 970
Joined: Wed Sep 10, 2008 9:34 am

Post by Roy »

retiredjg wrote:Enough!

You two are not ever going to agree. The only choice you have is to agree to disagree.

I'm interested in hearing what each position is. I'm not interested (or impressed) when one of you is pointing out how wrong the other is.

The early parts of this thread were somewhat helpful. It has now turned into nothing more than "right-fighting" in my opinion.

As another poster said earlier, losing respect....
I respect your opinion, Retiredjg, but disagree. Refutation is an important part of the process, especially if an unsupported, or specious claim is made. But it often takes many words to refute properly, and lots of work on our part to read through this complex issue. Now, both seem well-intended men, to me, but I say again that the quality of the arguments made—as supported by evidence provided or questionable statements made—is decidedly not the same, in my opinion. And I say this even though I choose not to own CCFs, at this time, and I've read enough on CCFs for now too!

Roy
User avatar
retiredjg
Posts: 54082
Joined: Thu Jan 10, 2008 11:56 am

Post by retiredjg »

Roy wrote:I respect your opinion, Retiredjg, but disagree. Roy
Well, I'm glad we can disagree without getting personal. :D

I'm all for a good honest debate, but this one has taken on a surly tone and devolved into just 2 guys trying to be right. That is different from two people debating an issue. Did that make sense?

That is simply my opinion. I'm sure not everyone agrees. But I also feel sure that many folks out there are just shaking their heads and saying "Oh, no, there they go again...." :roll:

I would really like to read a discussion of the pros and cons of commodities without all the childish personal attacks. I know nothing of commodities and this kind of "discussion" has not been helpful. At least not for me. Maybe someday....
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Post by larryswedroe »

Retiredjg
I have to disagree. First how can you decide who is right if one makes false statements but you don't know it is false?

As I have said, people can look at the same facts and sometimes come to different conclusions because this is not an exact science.

As I noted, Robert T came to different conclusion but his facts were right and his logic good. So got no argument from me. In the other case there are repeated false statements and errors of various kinds. And the problem is that because the person making the error is an author and advisor people give it credibility--unless it is corrected. So the false statements need to be exposed as wrong so people can in fact make informed decisions based on the right set of facts. That is all I have done. Shown why the statements are simply wrong on their face. Examples of recent errors in this thread alone: EM can act as substitute for commodities--so I showed that to be false. I also showed that because you have 15% exposure to commodity producers that you don't need CCF--that too is wrong because you have large exposures to companies that have negative exposure to commodities. And there are other errors of fact including even basic economic definitions like arbitrage. Errors of all kinds should be exposed. Note I have not used words like arrogant in my statements. Just presented facts and provided the correct lens through which one should view things.

This is no different than when any of us point out the errors in analysis made by the media, active managers, etc. If I make an error I hope it will be correct --that is the way you learn.

you can decide for yourself in the debate who presents the facts and logic and who makes statements that cannot be backed up by the facts or the logic.

BTW-I have lot better things to do with my time that have to correct misstatements so I wish they would cease. Certainly make me happy
User avatar
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri »

This debate has a lot of history behind it. If you wish to see more information, go to:

The Great Debate Part I

The Great Debate Part II

The moderated debate took place in February, 2008, before the commodities crash.

Here are some classic lines from the debate worth remembering:

[Swedroe]: Futures, or more specifically, CCFs, are one of the rare asset classes that have negative correlation to both stocks and bonds. That makes them excellent risk diversifiers.

Ouch! That one must hurt, Larry.

[Swedroe]: The negative correlation is easily explained and quite logical.

Uh-huh. Except that, how do you explain the extremely high correlation last year when stocks AND commodities came crashing down?

[Swedroe]: Another consideration is that CCFs tend to produce very poor returns for a very long time and then have very short bursts of spectacular returns.

This we can agree on. CCFs definitely are expected to produce very poor returns.

[Swedroe]: It's true that commodities themselves have no expected real return. That's a pretty good reason to avoid investing in them.

We can agree on this also!

Okay, Larry. Take you shots....

Rick Ferri

PS. If this conversation style is truly annoying some people on the board, then we will stop it.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Post by larryswedroe »

It does truly amaze me how someone can continue to make the SAME mistakes over and over. In each of the examples he repeats the same error. So you judge as the audience
Futures, or more specifically, CCFs, are one of the rare asset classes that have negative correlation to both stocks and bonds. That makes them excellent risk diversifiers.

Ouch! That one must hurt, Larry.


AS I have explained many times this statement shows the lack of understanding of the term negative correlation. No one said the correlation was negative 1. Just negative. The term negative correlation is a tendency--tendency means not always. And I had presented here and in my books the data showing CCF did not always produce positive returns in negative years for stocks, nor should that be expected. So nothing, repeat nothing that happened last year should have surprised anyone. As my daughter says EOD.
The negative correlation is easily explained and quite logical.

Uh-huh. Except that, how do you explain the extremely high correlation last year when stocks AND commodities came crashing down?
See above--as i have explained MANY times and shown the data, CCF hedges SOME risks (inflationary type shocks) but not all (deflationary shocks). And since there is a tendency for there to be more inflationary than deflationary shocks there is a negative correlation. EOD.
It's true that commodities themselves have no expected real return. That's a pretty good reason to avoid investing in them.


Perhaps this is the most amazing of them all, but they all amaze me., No one is recommending investing in commodities, but CCF which are ENTIRELY different. And this has been gone over ad infinitum
EOD

Now one can conclude not to invest in CCF (As Robert Did) but by now I am virtually certain that everyone else on this board could have provided the comebacks to Rick's challenge. Also note that Robert basically made the same points I made.
User avatar
retiredjg
Posts: 54082
Joined: Thu Jan 10, 2008 11:56 am

Post by retiredjg »

Hi Larry.

Well, when I pulled out my little bitty soap box, I certainly did not expect to be getting a reply from you or Rick. Since you have taken the time to reply, I will do the same.

If you go back and read your post to me, you'll realize that almost all of it says, one way or the other, "I'm right! I'm right! He's wrong! I'm right!" You have spent a lot of time defending how "right" you are. And you have an incredible amount of energy tied up in being "right" about this.

You think you are right. Rick thinks he is right. It is nothing more than a disagreement and all this right-fighting is not helpful.

Obviously, this is a subject on which reasonable people can disagree. So disagree. It would be much more helpful to learners like me to just present your position and let me figure it out for myself. All of this "rightness" is just raising blood pressure and producing stress. It is not conducive to learning. As far as I know, this forum is a platform for learning, not a platform for being right.

Yes, I understand that you believe you are correcting the errors that Rick has made. I'm sure he feels he is doing the same. That's why it is called "a dis-agreement".

Lack of agreement is an everyday event around here. Every subject has its supporters and opponents. No subject (except maybe keeping costs low) is agreed on by everyone. But it is usually handled with respect and courtesy. Somehow, when you and Rick are on opposite sides of an issue, respect and courtesy go out to door. Well, when that happens, I lose respect myself. I'm sure that is not the result you are trying to achieve.

So, that's my thoughts. I still know nothing more than I did about commodities. It would be oh-so-much more helpful if you each could just learn to say "Rick (Larry) and I disagree on this issue. Here are the facts as I see them."

Learning to argue an issue well is quite a skill. It is not personal and it relies on each listener to determine the truth of the matter for him/herself. Some of us will get it right. Some will not. But the rightness of an issue is for the listener to decide, not the debaters. Present your own truth and let it go.

Someday, I may learn to be more concise. Not making progress very fast though. I see Rick has replied while I was typing...guess I'll go read it now.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Post by larryswedroe »

retiredjg

I have to disagree. No where did I say anything about my being "right"

Note I show that Rick's statements are wrong. The statements are not opinions of mine but factual evidence demonstrating his comments/statements are wrong. I cite the data and the academic literature. Rick cites his own opinions unsubstantiated by anything and often simply incorrect, even by definition. Even the statement about arbitrage was wrong. A simple definition of a common term. And his statements about negative correlation. That is not a debatable issue. Rick is saying the equivalent of 1+1 =3, Check for yourself the definition of negative correlation if you doubt me. That should end this whole discussion right now.

Note that nowhere has anyone shown what I have said is wrong.. Just come to different conclusion. So it has nothing do with my being right. Don't know how you can come to that conclusion when I specifically state in my review of Robert T's conclusion that he was perfectly right and his conclusion is different than mine. But his conclusion was based on facts and logic, not what false statements totally unsupported by either logic or fact.

So this is not about whether to invest in CCF or not. But to make sure people are making informed decision based on facts, not falsehoods.

It is not about my being "right"--but about FACTS being right. Note I don't recall anyone showing that any of the statements I have made have been incorrect. Just opiniois about whether to invest in them or not. Which we can agree to disagree, this is not an exact science and each of us values some attributes more than others

I will add one more thing--in my entire career I have never heard such repeated errors of facts. Never. Once shown their statements are incorrect rational people stop repeating the statements.

Just look at three of Ricks quotations. All repeat the same mistakes--of FACTS, not opinions
User avatar
retiredjg
Posts: 54082
Joined: Thu Jan 10, 2008 11:56 am

Post by retiredjg »

Larry, I didn't really expect that you would agree. :wink:
User avatar
Sammy_M
Posts: 1935
Joined: Sun Nov 25, 2007 7:30 am

Post by Sammy_M »

I find the back-and-forth to be quite good and reasonably civilized. As I said in a thread between Larry and Henry on SV premium not long ago, debate brings out info and provokes thought in a way like no other. As long as Rick and Larry don't take it personal and stop posting here (as has been threatened in the past), please continue. Seems you actually enjoy it. Also is funny that the extent to which you agree on things far exceeds that which you disagree on, but those agreements rarely come out in the discussion boards.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Post by larryswedroe »

sammy
I don't mind pointing out that something is incorrect, or IMO incorrect, but having to waste time repeating the same thing, things that are not debatable like definitions even or historical evidence showing the statements are wrong is a total waste. And if it was just another poster I would not bother. But most people here have no way of knowing something is wrong so they assume an "expert" must be right.
WileECoyote
Posts: 84
Joined: Thu Jun 18, 2009 10:55 pm

Post by WileECoyote »

There is quite a bit of arguing but I do appreciate hearing different parts and sides of the argument. Part of the back and forth reveals aspects of CCFs that may not have come up otherwise.

I've been trying to dig into CCFs a bit more and was thinking that maybe the mechanical aspect of GSCI and DBC might lead to front running\arbitrage that occurs in the Russell 2k for instance that bleeds away the returns. It was interesting to hear from people more knowledgable than myself to lay out their arguments so I can make my own decision.
User avatar
Robert T
Posts: 2803
Joined: Tue Feb 27, 2007 8:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Post by Robert T »

.
Few more thoughts

1. Historically the average compound excess return of INDIVIDUAL commodities futures has been close to zero (in analysis of 36 individual commodities futures Gorton and Rouwenhorst show 18 had geometric excess returns [above the risk free rate] and 18 below).

2. But…. historically the average annual excess return of an equally weighted, monthly rebalanced, simulated PORTFOLIO of commodity futures has been positive and significant (often shown to be ‘equity-like’)

3. How so? The geometric return of a portfolio can significantly exceed the weighted average geometric return if the constituent securities in the PORTFOLIO have low correlation with one another and the securities have high average standard deviation – often called the diversification return. The SIZE of the diversification return historically seems unique to a PORTFOLIO of COMMODITIES (i.e. idiosyncratic risk drives low correlations across commodities as there doesn’t appear to be any common risk factors (beta, size, value, term, default) driving the variation in return – unlike both stocks and bonds).

4. In addition to correlations, standard deviations, and number of securities, the weighting given to each security also seems to determine the size of the diversification return.

5. The DJ AIG commodity index weights primarily on futures contract liquidity data, supplemented with production data. The S&P GSCI weighting is based on the level of worldwide production for each commodity (as mentioned earlier...unlike stocks there is no commodity futures market capitalization, outstanding value of long and short futures is exactly offsetting) – hence different weights often referred to as strategies (as in Erb and Harvey).

6. The future excess returns of a PORTFOLIO of commodities futures will likely be driven by two factors (i) the roll return, and (ii) the diversification return. If history is a guide the roll return on the GSCI may be larger than on the DJ AIG commodity index by the simple fact that oil accounts for the dominant share of the index, and historically has, for two-thirds of the time since 1982, been in backwardation (resulting in a positive roll return). Not the case for all commodities (e.g. gold has historically been in contango). However the DJ AIG commodity index appears more diversified likely resulting to a higher rebalancing return (which will likely the more dominant share of future CCF returns, as Erb and Harvey suggest “For a broadly diversified portfolio of commodities futures a risk averse investor might very well want to assume a future roll return of zero (or less)”)

7. So it seems we are then left with the diversification returns as the main driver of future (real) excess CCF returns (historically estimated at around 3 to 4 percent per year). As above the main drivers of the diversification return are average correlation and standard deviation (assuming a constant number of securities, and security weighting). Historically the average correlation coefficient of the commodities in the GSCI has been 0.2. and average standard deviation of individual commodities has been about 30%. If average correlation rise to 0.3 returns expected return declined by about 15%, and if volatility declines from 30% to 20% expected return declines by 60% (as demonstrated in simple simulations in Erb and Harvey).

8. Is expected return higher or lower than the past? As mentioned the earlier post, agriculture and oil markets have likely become more integrated (correlated) through ethanol base biofuels - at least higher that the 0.01 correlation coefficient from 1982-2004 (in simulated backtests). In addition, with more integrated commodity markets average volatility likely declines. So IMO expected returns are lower.

9. Does it matter? Expected returns matters for allocation decisions. Erb and Harvey find that even with an excess return of 1% and ‘optimal allocation” has historically been about 3% (high returns lead to much higher allocation). So still shows up with lower excess return.

The above draws heavily on the Erb and Harvey paper. Essential reading for anyone considering adding CCFs IMO. Will post a few more thoughts on event risk later...

Robert
.
Tramper Al
Posts: 3665
Joined: Thu Oct 18, 2007 11:42 am

Post by Tramper Al »

larryswedroe wrote:And if it was just another poster I would not bother. But most people here have no way of knowing something is wrong so they assume an "expert" must be right.
Just my opinion, but I think perhaps that all of this bickering about commodities and CCFs effectively reduces the expert down to the level of just another poster. What is ultimately lost is credibility. It doesn't say much for the average Boglehead reader either if it is thought we will swallow any old expert's assertion without question.

Sure, I may have learned a few things about CCFs along the way, but honestly very little after the 2nd or 3rd thread. Any decent article or paragraph or two in a good book can cover this fairly well. Surely there have been 50 commodity threads by now, all eventually spiraling downward with the same two opposite positions. A great debate? Maybe once.

It's getting so no one else can discuss commodities or CCFs. Heck, I'm not sure you can even ask a peripheral question like "CCFs in tIRA or Roth?" without the same old scrap taking center stage, again.

But then, I'm just another poster.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Post by larryswedroe »

tramper
simple solution. People should stop posting false statements or statements as if they are fact that are in fact not so. Then no need to waste time exposing false statements. But at any rate this thread is my last on CCF. Everyone by now should be able to differentiate fact from fiction.

Note also Robert T's responses. Which are basically the same as mine in content. We do disagree on our own solutions but we agree on the facts and don't make stuff up. So there is nothing to disagree about here
User avatar
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri »

retiredjg wrote:It would be oh-so-much more helpful if you each could just learn to say "Rick (Larry) and I disagree on this issue.
I agree to disagree for about the 10th time.

The funny thing about this non-sense phesdo-debate is that in the long-term, it very likely not going to make much difference if you buy a CCF or not. What are we talking about, maybe 10% portfolio allocation? That amount is not going to affect your long-term return by more than 0.2% (on the down-side, of course). That is not a meaningful number. So, if you think that Larry makes a good argument about owning CCFs, then follow the herd and go for it.

Rick Ferri
barefootjan
Posts: 459
Joined: Mon Feb 19, 2007 5:45 pm

Post by barefootjan »

retiredjg:
Obviously, this is a subject on which reasonable people can disagree. So disagree. It would be much more helpful to learners like me to just present your position and let me figure it out for myself.

Retiredjg, you said that you're a "learner," so maybe if you look up the definition of "negative correlation" you'll understand what Larry means about making arguments based on the facts.

I understand your frustration, but I don't see how you can "figure it out for yourself" if you don't understand the arguments, and you can't understand the arguments unless you first have a decent understanding the terms being used.

If I were an interior decorator, and I told you I don't think it's a good idea to paint your bedroom orange and red, in part because "contrasting colors are much too harsh," would you call that a reasonable argument? ~ Because - as far as I know!- orange & red are not contrasting colors. :roll:

Since Larry & Rick apparently don't agree on what negatively correlated means, maybe Rick can give his definition/understanding of it (Larry has given his many times already). I don't think the term is controversial, though, since to the best of my knowledge it's simply a statistical measurement.



PS You might find this article helpful...?

http://www.investopedia.com/articles/fi ... cation.asp
Last edited by barefootjan on Tue Feb 16, 2021 2:59 am, edited 2 times in total.
DanHess
Posts: 8
Joined: Fri May 30, 2008 2:05 pm
Location: North Carolina

Post by DanHess »

Rick and Larry

I have been reading the debate on the use of commodities place in a portfolio with great interest. I view it healthy to learn of the differing viewpoints of two great authors I admire. I also do not think there needs to be one right answer for everyone and investors can choose based on their own personal views. Thus I do not view this debate trying to determine a winner or loser but rather to provide facts and information upon which each of us can learn.

I tend to be more in Larry's camp to include commodities in my portfolio. Some of the key points I view influencing me are:

1) I expect the emerging markets such as China and India will be contining to improve their standards of living and that in turn will place an increasing demand on commodities overall and eventually higher prices.. This trend will accelerate once the world wide recession is over.

2) The Geopolitical situation in the world with North Korea, Iran, Latin America, Nigeria, Terrorists, Pirates, etc is in my view like a powder keg ready for someone to ignite. It is unclear what may happen but in my view one needs insurance to protect against this and I see commodities providing some of that insurance.

3) The US government actions to stimulate, spend, borrow, print money, etc will eventually lead to increasing inflation and a weaker US dollar. While this may not happen during the recession and high unemployment it seems to me to be fairly certain to occur longer term. In addition politicians tend to use inflation to pay off debts with dollars of lesser value. Thus with my portfolio of many US equities in US dollars it seems appropriate to hedge against this inflation and I see commodities as a better way to accomplish this than by using TIPS.

I do recognize all will not agree with the three scenarios I have shown but in my view if they occur including commodities in my portfolio seems like worthwhile insurance. Am I off track in my thinking?

Dan
User avatar
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri »

barefootjan wrote:Since Larry & Rick apparently don't agree on what negatively correlated means, maybe Rick can give his definition/understanding of it (Larry has given his many times already). I don't think the term is controversial, though, since to the best of my knowledge it's simply a statistical measurement.
What does negative correlation mean to commodities? It means that when stocks go down, commodities go up. But that is not how commodities behave. In my All About Asset Allocation book that I wrote in 2005 and published in 2006, there is a very nice chart on page 181 that shows the rolling 36-month correlation between commodities, stocks and bonds. Commodities are negatively correlated at times, positively correlated at times, and non-correlated at other times. The questions is, WHEN were they negatively correlated and WHEN were they positively correlated? That question makes a big difference to investors.

I looked at all the time periods, and some surprising events jumped out. During periods of extreme equity market stress, commodities were POSSITIVELY correlated with equities. In other words, 2008 was not the first time stocks AND commodities fell together. It occurred in 2002, 1998, and 1982. By doing my own research, and studying the details of specific time periods, I knew how commodities could act in a bear stock market.

I am not against commodities, but I am not for the either. Remember, back in the 1990s, I was head-over-heels in love with having commodities in a portfolio for all the reasons Larry states today. And I included commodities in all client accounts. But the results were disappointing. Commodities did not act as the research inferred.

So, I did my own number crunching, my own research, and this is how I became a dis-believer. And a few years ago chose not to believe the conclusions of much of that research because 1) my own experiences with the asset class, 2) the writers of some research are bias based on their jobs or products they were trying to sell, and 3) the research conclusions did not always match my own conclusions drawn from my own analysis (please give me credit for being a full-fledged financial analyst by profession. i.e. a CFA charterholder).

I am not trying to be un-civil in this debate. I never said Larry was lying or making false claims. In fact, he makes some very good points about commodities and CCFs based on the research reports he has read. But we come from different experiences and different backgrounds, and I have a different viewpoint. Unfortunately, my happy-go-lucky method of persuasion does not always come across in print as it does in person. If anyone took offense, particularly Larry, then I apologise.

Rick Ferri
WileECoyote
Posts: 84
Joined: Thu Jun 18, 2009 10:55 pm

Post by WileECoyote »

There are a couple of things I would like to ask while we have everyone on the subject.

IF one were to decide to make an allocation to CCFs:

1. They would have to reside in a tax sheltered account unless they are an ETN correct? The tax implications of something like PCRIX look onerous to say the least. As Tramper Al asked, is there any advantage to Roth vs tIRA?

2. Could you shoot holes in the LSC ETN? It is able to go long\short and although it has a short history do you see any glaring flaws with their general strategy?
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Post by larryswedroe »

This is EXACTLY what I mean

It now must be obvious that Rick does not understand the term negative correlation since his explanation is simply wrong, as wrong as saying 1+1 =3, Negative correlation has nothing to do with one going up and other going down.

This is a very common and unfortunate error while it is understandable to make the mistake once, there is no excuse for a professional to keep repeating it
What does negative correlation mean to commodities? It means that when stocks go down, commodities go up
But here is the correct definition as I have explained numerous times:

Negative Correlation When one asset experiences above average returns the other tends to experience below average returns, and vice versa.
Note the word TENDS to and note there is nothing here about moving in opposite directions.

Here is example. Let's assume two assets have average return of 6% and when one returns 8 the other returns 4 and vice versa. They would be negatively correlated yet there would never be a period when they moved in opposite directions--when one was up the other was down.

Thus if you cannot understand the simple terms needed to make an appropriate recommendation, how is it possible to make a correct analysis of the risks? You cannot.
Last edited by larryswedroe on Tue Jun 30, 2009 10:54 am, edited 1 time in total.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Post by larryswedroe »

wileycoyote
I would only recommend CCF in a tax advantaged account unless one is in lowest brackets

Second I would never recommend an ETN because you are taking the incremental credit risk and not getting paid to take it (unless the ER was so much lower to compensate for the credit risk). Example, if you buy a Barclay's ETN with say a 10 year term you should add in the spread of Barclay's debt over Treasuries as an implied cost
barefootjan
Posts: 459
Joined: Mon Feb 19, 2007 5:45 pm

Post by barefootjan »

Rick thank you, I greatly appreciate your reply.

To increase my understanding of your position, I have a few follow-up questions, if you don't mind:

1) In your reply you refer almost exclusively to "commodities." Is that your short-hand way of referring to "collateralized commodity futures" (CCFs), or did you really mean commodities? If so, did you research both? Did CCFs behave the same as commodities?

2) Maybe I should've asked this first: What differences if any do you see between commodities and CCFs?

3) What is your understanding of the risks that CCFs might hedge against? Do you believe they're an effective hedge against event risk, for instance? If so, what did your research show about their effectiveness in that particular instance?

THANK YOU so much for hanging in there with this discussion, and no I for one haven't been offended by what you or anyone else has said in this thread.
Last edited by barefootjan on Tue Feb 16, 2021 3:00 am, edited 1 time in total.
WileECoyote
Posts: 84
Joined: Thu Jun 18, 2009 10:55 pm

Post by WileECoyote »

I think maybe you're being a bit too symantical. What I take from Rick is that using his analysis he doesn't see a strong enough negative correlation during weak equity markets to warrant CCFs use. That's a valid argument to me.

Your argument seems to be in an overall portfolio CCFs can be helpful over time as the correlation waxes and wanes with the other asset classes, which also seems valid.

Would I be summing those up correctly or am I missing something?
barefootjan
Posts: 459
Joined: Mon Feb 19, 2007 5:45 pm

Post by barefootjan »

Larry wrote:
"Negative Correlation When one asset experiences above average returns the other tends to experience below average returns, and vice versa. "

OK well then I didn't understand it fully either, because I was more focused on the word "tends." In other words, if you can find an instance where an asset class doesn't do what it tends to do, you haven't proved that it doesn't still have that tendency.

Can someone re-word that so it's clearer? :roll:
Last edited by barefootjan on Tue Feb 16, 2021 3:00 am, edited 1 time in total.
WileECoyote
Posts: 84
Joined: Thu Jun 18, 2009 10:55 pm

Post by WileECoyote »

Ok good info here - thanks!. Two more things if you guys wouldn't mind:

What kind of trading costs does a fund like PCRIX incur? I haven't the slightest what the trading costs are for commodities vs something like equities.

Second, I found this over the weekend and the charts at the bottom of the first page show some interesting correlations between TSM, LBAG, GSCI, DJ AIG and TIPS during the worst periods for TSM and LBAG. I honestly have no research to validate those claims, does it seem accurate to anyone in the know?

http://www.hardassetsinvestor.com/compo ... l?Itemid=4
WileECoyote
Posts: 84
Joined: Thu Jun 18, 2009 10:55 pm

Post by WileECoyote »

And as a quick follow up, if the above link seems to be accurate, does anyone think the additional returns from GSCI during the worst periods for TSM and LBAG stem from its having a higher exposure to oil? IE, maybe it's more a specific portion of the index that performs well vs all of the components?
User avatar
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri »

barefootjan wrote:Rick thank you, I greatly appreciate your reply.

To increase my understanding of your position, I have a few follow-up questions, if you don't mind:

1) In your reply you refer almost exclusively to "commodities." Is that your short-hand way of referring to "collateralized commodity futures" (CCFs), or did you really mean commodities? If so, did you research both? Did CCFs behave the same as commodities?
CCFs behave like futures products rather than commodities in the short run and those returns can be very different due to shifts between contago and backwardation. In the long run, the backwardation/contango swings tend to null out and one should expect that returns between spot and futures to very close before costs. Many of the new indexes being created play the backwardation/contango movement. See these new Morningstar Indexes.
2) Maybe I should've asked this first: What differences if any do you see between commodities and CCFs?
First, CCFs are much easier to invest in that physical commodities (except for a few gold and silver ETFs in which you own the actual asset rather than futures contracts). Second, not much in the long-term performance as I stated above. That said, there are significant difference in the strategies among CCFs based on the indexes they each follow. Consequently, there have been and will be significant differences in returns at times. Some indexes have a passive strategy (S&P GSCI), some follow an active allocation (Rogers index), and some are quasi-active in that there are caps on sectors that are rebalanced to fixed weights annually (CRB and DJ-UBS, formally DJ-AIG).

The benefit of rebalancing is questionable. CRB runs the oldest equal weighted index and it has been rebalanced continuously for more than 50 years. CRB found only scant incremental benefit from equal weighting. On the other hand, one set of researchers are claiming a 4% rebalancing benefit in the equal weighted back-tested index that they created and are trying to license. So, it depends what you read. I prefer to be skeptical.
3) What is your understanding of the risks that CCFs might hedge against? Do you believe they're an effective hedge against event risk, for instance? If so, what did your research show about their effectiveness in that particular instance?
There is occasional non-correlation and negative correlation with commodity products against commodity shortages, inflation shocks and other dollar shocks. There is a 3-5 year lag between shortages and surpluses from new production and substitutes. TIPS and stocks also hedge those risks, although the timing is different. So, when deciding how to hedge, you will need to decide what your needed time frame for the hedge is (does it need to be immediate or are you a long-term investor), how much you are going to pay for the hedge (CCF fees and taxes are much higher than equity index fund fees and taxes), and what you are giving up to hedge (in the long-term you would be taking away from equity returns to receive lower commodity returns ). So there is a lot to consider. Again, I am not against CCFs, but I am not for them either.

One more issue, Larry wrote:
It now must be obvious that Rick does not understand the term negative correlation since his explanation is simply wrong, as wrong as saying 1+1 =3, Negative correlation has nothing to do with one going up and other going down.
Anyone who has read my books knows that I very much understand correlation, variance, R square and all the other regression equations and have written extensively on these subject. If someone wishes to read a better written explanation, please see any of my books or go to my website for information.

Rick Ferri
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Post by larryswedroe »

Jan

Re negative correlation and TENDS to

In the 9 years of negative stock returns since 1970 CCF have produced positive returns 6 times with the average return during those years ABOVE the long term average return of CCF.

But there were three years (81, 01 and last year) when that was not true.

In the case of bonds every year that stocks had negative returns CCF had positive returns, and on average the returns were very high.

You see this in the correlation data. Now Rick has stated that CCF is not negatively correlated with stocks with last year as the "proof". First his definition is simply wrong as I have showed. Second here is the data.

1970-2008 Annual Correlation of S&P to GSCI was -.07 and quarterly through March was -.13. Clearly negative. Rick is simply declaring that something that is white is black.

Now for bonds, as you should expect given the examples I gave the negative correlation is even stronger. For long term government bonds the annual correlation to the GSCI was -.24 and the quarterly was -.21.

The important point is that the correlations are not negative 1, but they are negative.

So not only do CCF diversify the risks of equities but also the risks of longer bonds. And note that there has never been a year when all three were negative. Not once.
barefootjan
Posts: 459
Joined: Mon Feb 19, 2007 5:45 pm

Post by barefootjan »

Rick thanks for the explanation and the link. You are IMO a very good writer, by the way. :D

On your website I found the following:

"Asset allocation analysis should include a correlation study between investment types. Correlation analysis shows how the price of one investment has historically moved in relation to the price of another. If two asset classes moved in the same direction at the same time they had positive correlation. If the returns had moved in different directions at the same time, they had negative correlation."

I have to ask: Do you agree or disagree, then, with the definition of negative correlation that Larry gave above?
("Negative Correlation When one asset experiences above average returns the other tends to experience below average returns, and vice versa. ")

If I'm following Larry (big IF), an asset class can be negatively correlated with another asset class even if they both go up at the same time (ie move in the same direction)--just so long as one is going up less than expected and one is going up more than expected. Same with going down, I guess.

Actually, wait: what does "average returns" mean? The average returns of that asset class???

You also wrote:
"Correlation is not a fixed number. It changes over time and in unpredictable ways. It would be ideal if two asset classes had positive real returns expectations and consistent negative return correlation with each other. Unfortunately, there are no such pairs of investments. A rolling correlation study shows that the correlation between any two asset classes tends to shift over time and in an unpredictable way. Based on past data, we can only loosely predict what the correlation between two asset classes might look like."

That's the only part I can ever remember. LOL.

Thanks again.
Last edited by barefootjan on Tue Feb 16, 2021 3:00 am, edited 1 time in total.
User avatar
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri »

barefootjan wrote:On your website I found the following:

"Asset allocation analysis should include a correlation study between investment types. Correlation analysis shows how the price of one investment has historically moved in relation to the price of another. If two asset classes moved in the same direction at the same time they had positive correlation. If the returns had moved in different directions at the same time, they had negative correlation."

I have to ask: Do you agree or disagree, then, with the definition of negative correlation that Larry gave above?("Negative Correlation When one asset experiences above average returns the other tends to experience below average returns, and vice versa. ")
That is the correct definition. Take the following series:

Investment #1: +10, +5, +10, +5
Investment #2: +5, +10, +5, +10

The two investments are perfectly negatively correlated and neither return is negative. I would argue, if all investments are expected to rise, you do not need negatively correlated asset classes. We seek negatively correlated asset classes because stocks FALL at times. That is the negatively correlated context in which commodities are sold, as a hedge against a falling equity market, and that is the one I stated in the simple example.
If I'm following Larry (big IF), an asset class can be negatively correlated with another asset class even if they both go up at the same time (ie move in the same direction)--just so long as one is going up less than expected and one is going up more than expected. Same with going down, I guess.
Yes, that is correct.
Actually, wait: what does "average returns" mean? The average returns of that asset class???
It is a straight average of each asset class; year 1 + year 2 + year 3 divided by 3. It is not the annualized return.

Rick Ferri
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Post by larryswedroe »

Jan

There are two statements that cannot both be correct.

First, there are many mathematicians on this site. Any one can confirm that my definition is the correct one. Or you can google the term and find definitions--such as this one: http://www.businessdictionary.com/defin ... ation.html

A 'negative correlation' means association of high values of one with the low values of the other(s). Correlation can vary from +1 to -1. Values close to +1 indicate a high-degree of positive correlation, and values close to -1 indicate a high degree of negative correlation.

Now, my statement on the definition of negative correlation and Rick's statement that 2008 demonstrates that stocks and CCF are not negatively correlated are incompatible--one is incorrect.

In addition, the data I presented also demonstrates that Rick's statements are wrong. The two asset classes are negatively correlated. So you can draw your own conclusions about what one knows or doesn't know
User avatar
Russell
Posts: 285
Joined: Fri Feb 23, 2007 1:26 pm
Location: on the Chesapeake Bay
Contact:

Post by Russell »

barefootjan wrote: If I'm following Larry (big IF), an asset class can be negatively correlated with another asset class even if they both go up at the same time (ie move in the same direction)--just so long as one is going up less than expected and one is going up more than expected. Same with going down, I guess.
In any given year, they can even both go up more than expected (or both drop more than expected, see 2008) -- correlations that aren't +/- 1 measure a tendency in sometimes messy data.

Suppose you're comparing the returns of two different assets (A and B) over some time period. You can put together a scatter plot with the return of Asset A on the horizontal axis and the return of Asset B on the vertical axis. For example, suppose one month, Asset A returned 7% and Asset B returned 3%. You would plot a single point seven ticks to the right and three ticks up. When you plot many years worth of data on this same graph, you may see a big fuzzy cloud of points. Fit the best line you can through this cloud. If the line goes up, you have positive correlation; if it goes down, the correlation is negative. The tighter and more focused the cloud, the stronger (closer to 1 or -1) the correlation.

I grabbed this image from google, showing stocks vs bonds.

Image

Since the line goes up to the right, the correlation coefficient is positive. However, since the cloud is somewhat fuzzy (and not tightly fit to the line), the coefficient is somewhat close to zero.
The best material model for a cat is another, or preferably the same, cat. - A. Rosenblueth and N. Wiener (1945).
User avatar
Russell
Posts: 285
Joined: Fri Feb 23, 2007 1:26 pm
Location: on the Chesapeake Bay
Contact:

Post by Russell »

larryswedroe wrote:The two asset classes are negatively correlated. So you can draw your own conclusions about what one knows or doesn't know
I imagine this depends strongly on what data set you are looking at. If you're looking at annual numbers over the last thirty-six years you will probably see different things than if you're looking at monthly numbers over the last three (which, I believe, is closer to what Rick is considering).
The best material model for a cat is another, or preferably the same, cat. - A. Rosenblueth and N. Wiener (1945).
User avatar
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri »

Here are the most up-to-date correlation measurements between stocks and commodities you can get.

Using daily price data:

1) The correlation between the GSCI and the S&P 500 from July 1, 2008 to June 30, 2009 is +0.95.

2) The correlation between the DJ-UBS (formally AIG) and the S&P 500 from July 1, 2008 to June 30, 2009 is +0.95.

3) The correlation between oil prices and the S&P 500 from July 1, 2008 to June 30, 2009 is +0.94.

Those are all extremely high positive correlations. It goes without saying that commodity investments have been no help to investors during the recent bear market in equities.

Rick Ferri
User avatar
Russell
Posts: 285
Joined: Fri Feb 23, 2007 1:26 pm
Location: on the Chesapeake Bay
Contact:

Post by Russell »

Rick Ferri wrote:Here are the most up-to-date correlation measurements between stocks and commodities you can get.

Using daily price data:

1) The correlation between the GSCI and the S&P 500 from July 1, 2008 to June 30, 2009 is +0.95.

2) The correlation between the DJ-UBS (formally AIG) and the S&P 500 from July 1, 2008 to June 30, 2009 is +0.95.

3) The correlation between oil prices and the S&P 500 from July 1, 2008 to June 30, 2009 is +0.94.

Those are all extremely high positive correlations. It goes without saying that commodity investments have been no help to investors during the recent bear market in equities.

Rick Ferri
Careful with these calculations. What we care about is not the correlations of the price (or index level), but rather the correlation of the daily returns. When I run the daily returns of PCRIX versus VFINX over the last year*, I get +0.36 -- which is positive, but not nearly so close to 1....

Best, Russell

* per yahoo data, with close price adjusted for dividends and splits.
Last edited by Russell on Tue Jun 30, 2009 3:37 pm, edited 1 time in total.
The best material model for a cat is another, or preferably the same, cat. - A. Rosenblueth and N. Wiener (1945).
barefootjan
Posts: 459
Joined: Mon Feb 19, 2007 5:45 pm

Post by barefootjan »

Larry

In regards to the definition of negative correlation, I see your point.

But I disagree with your later suggestion about determining who knows what. Speaking in general and not about anyone or any conversation in particular, isn't it fair to say that one can -- for example-- make an imprecise or incomplete statement, yet still know all the facts? or take liberties with the facts to make a point, yet still know what those facts are?

Again I'm not talking about any of the points made here, I'm just suggesting that there are other conclusions that can be drawn rather than merely who knows what.

FWIW it seems to me that what Rick has written in his summary above has IMO backed up some of what you've said about CCFs. That the more risk-averse you are the more you should consider them, for instance.

"There is occasional non-correlation and negative correlation with commodity products against commodity shortages, inflation shocks and other dollar shocks. There is a 3-5 year lag between shortages and surpluses from new production and substitutes. TIPS and stocks also hedge those risks, although the timing is different. So, when deciding how to hedge, you will need to decide what your needed time frame for the hedge is (does it need to be immediate or are you a long-term investor), how much you are going to pay for the hedge (CCF fees and taxes are much higher than equity index fund fees and taxes), and what you are giving up to hedge (in the long-term you would be taking away from equity returns to receive lower commodity returns ). So there is a lot to consider. Again, I am not against CCFs, but I am not for them either."

In any event, you both know a hell of a lot more than I do...Appreciate the time you've taken to explain your positions.
Last edited by barefootjan on Tue Feb 16, 2021 3:01 am, edited 1 time in total.
User avatar
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri »

Russell wrote:When I run the daily returns of PCRIX versus VFINX over the last year*, I get +0.36 -- which is positive, but not nearly so close to 1....

Best, Russell

* per yahoo data, with close price adjusted for dividends and splits.
Something is not right with your correlation calculation. I just did the same analysis from July 1, 2008 to June 30 2009 using PCRIX versus SPY (SPRD S&P 500) and IVV (iShare S&P 500) and came up with a 12 month correlation of +0.95.

Rick Ferri
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Post by larryswedroe »

Jan
The statement from Rick that he is not against commodities--that you cited--is in total contrast to everything he has ever said before where he has ranted on about why anyone would ever want to consider them;
Here is just one quotation from this very thread
Why do you want to buy an asset class that has returned less than the inflation rate? That is dead money
Second point--Rick cites correlations for the recent past. That only shows what we all already knew--there are periods of high correlation--the negative correlation is a tendency and the long term average--Also if you know anything about correlations you know you need a lot more data points than shown to have any meaning at all.

Third, as I have pointed out, CCF hedges some risks, not all risks. That is obvious as I have cited, well before 2008 that in 81 and 01 CCF produced negative returns. So should have been no surprise to anyone and it doesn't change the fact that over the long run they have been negatively correlated/

Best
barefootjan
Posts: 459
Joined: Mon Feb 19, 2007 5:45 pm

Post by barefootjan »

Larry

"The statement from Rick that he is not against commodities--that you cited--is in total contrast to everything he has ever said before where he has ranted on about why anyone would ever want to consider them"

Yes, but for one brief moment he has seen the light. :lol: Let us revel in that moment.

God I love this forum. 8)
Last edited by barefootjan on Tue Feb 16, 2021 3:01 am, edited 1 time in total.
User avatar
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri »

larryswedroe wrote:The statement from Rick that he is not against commodities--that you cited--is in total contrast to everything he has ever said before where he has ranted on about why anyone would ever want to consider them
Actually, I provide CCF recommendations for investors who want CCFs in three books, All About Asset Allocation, All About Index Funds, and The ETF Book. I don't want any in my portfolio, but they are in the books for people who do.
Second point--Rick cites correlations for the recent past. That only shows what we all already knew--there are periods of high correlation--the negative correlation is a tendency and the long term average--Also if you know anything about correlations you know you need a lot more data points than shown to have any meaning at all.
As I wrote earlier, what is important about positive and negative correlation is WHEN it occurs. That is why I track rolling correlations over time between asset classes rather than using a single number based on the entire period.
Third, as I have pointed out, CCF hedges some risks, not all risks. That is obvious as I have cited, well before 2008 that in 81 and 01 CCF produced negative returns. So should have been no surprise to anyone and it doesn't change the fact that over the long run they have been negatively correlated/
I agree that CCF hedge some risks, not all risks. It is an expensive hedge, and not a tax-efficient one. I also believe those risks can also be hedged using global stocks and TIPS, although the timing of the hedge may be different. On the correlation argument "over the long run they have been negatively correlated", again, what matters is when the correlation is high and when it is low. You want low correlation during bear equity markets. Unfortunately, that is not always the case. In fact, during the worst equity bear markets there have been high CCF correlation because the commodities market is tanking also.

Rick Ferri
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Post by larryswedroe »

Jan
LOL

Let me repeat what I have already said. Until fairly recently I did not recommend commodities for a variety of reasons. They included
a) there was no good academic papers on the subject
b) there were no good low cost and well designed vehicles to implement the strategy

It was only after the publication of several papers and the creation of the PIMCO fund which used TIPS as the collateral, making it a significantly better investment vehicle to use to implement the strategy, that I decided to invest myself and of course recommend it to others. My first purchase was in 2003.

So smart people change their minds when they are presented with what they believe to be sufficiently compelling evidence. They don't get tied down trying to defend old positions simply to show that what they had recommended in the past is still correct. They are not afraid to say something like: The advice I gave you was based on the best information that was available at the time. Now there is new evidence and I am adapting my position to that evidence.


Hopefully this was helpful.
User avatar
Russell
Posts: 285
Joined: Fri Feb 23, 2007 1:26 pm
Location: on the Chesapeake Bay
Contact:

Post by Russell »

Rick Ferri wrote:Something is not right with your correlation calculation. I just did the same analysis from July 1, 2008 to June 30 2009 using PCRIX versus SPY (SPRD S&P 500) and IVV (iShare S&P 500) and came up with a 12 month correlation of +0.95.
My calculation is of the daily returns of the indices (for example, for SPY today I would use -0.81%). If we use the daily prices (for example, using 91.95 for SPY today), we get numbers like +0.95, but I don't believe that's as useful a computation to perform....
The best material model for a cat is another, or preferably the same, cat. - A. Rosenblueth and N. Wiener (1945).
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Post by larryswedroe »

As I wrote earlier, what is important about positive and negative correlation is WHEN it occurs.


This is exactly the right way to analyze things. As I have pointed out many times. Which is why assets like junk bonds and preferreds and convertibles and all hybrids should be avoided--the correlations are not only positive with equities but significantly positive and the correlations tend to turn high at the exactly wrong time

Now with CCF you have the opposite --that is exactly what negative correlation means--when one produces below average results the other TENDS to produce above average results (and vice versa). So CCF tends to produce its best returns just when they are needed most--Now this doesnt happen all the time, but the majority of the time.
Examples are the bear markets of 73 and 74 when the GSCI rose 75 and 40 percent. And 1990 when it rose 29% and 2000 when it rose 50% and 2002 when it rose 32%. Note that when you take all the negative years of stock returns CCF produces well above average returns.

But it is even better because adding CCF allows you to add duration risk to bonds, gaining the risk premium and they perform the best when both stocks and CCF get hit--see 2008 for example. So in the case of last year, since the right place to take CCF from is the equity allocation and CCF fell about the same as a globally diversified equity portfolio and long bonds did the best, those adding CCF and adding duration actually did better than if they had done neither, in the very period citing as the reason not to own CCF.
User avatar
BlueEars
Posts: 3968
Joined: Fri Mar 09, 2007 11:15 pm
Location: West Coast

Re: Two Questions--Commodity Index Fund--how much / what ind

Post by BlueEars »

dldetrick wrote:...
Question 2: which fund index would you use fot tracking--Goldman Sachs (GSCI), Dow Jones (DJ-AIGCI), Rogers International Commodity Index (RICI, Standard & Poors (SPCI), or ANOHTER ONE??
Regarding question 2 in the OP, what Yahoo quote is the best for just watching the commodity indexes?

Is it iPath S&P GSCI Total Return Index ETN (GSP)? I used to watch ^DJC but apparently that has been changed or just been obsoleted. GSP has only about 3 years of chart history on Yahoo.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Post by larryswedroe »

There may be no implied risk in an arbitrage trade, just like there is no implied risk in a T-bill / TIPS trade over the long-term. Nonetheless, it is interesting to note the large number of hedge funds that specialize in so-called 'risk free' arbitrage that lost money and have gone out of business, including Long Term Capital Management.
I forgot to comment on this statement

First, LTCM never engaged in risk-free arbitrage or any arbitrage. Their basic strategy was to make bets on convergence--reversion to mean of historical pricing relationships. Like the spread between the on the run and the off the run Treasury. But that is not arbitrage--that is making a bet that can go wrong.


Second, if someone is engaging in arbitrage but taking price risk then it is not arbitrage by definition. It is not that there is no implied risk, there is no risk because arbitrage is the engaging in SIMULTANEOUS transactions that eliminate risk--just "arbing" a price discrepancy that is existing--and of course in the process making markets more efficient. The classic example used to be buying and selling the same commodity simultaneous on two different exchanges, say London and NY or Chicago and NY.
User avatar
Russell
Posts: 285
Joined: Fri Feb 23, 2007 1:26 pm
Location: on the Chesapeake Bay
Contact:

Post by Russell »

larryswedroe wrote: First, LTCM never engaged in risk-free arbitrage or any arbitrage. Their basic strategy was to make bets on convergence--reversion to mean of historical pricing relationships. Like the spread between the on the run and the off the run Treasury. But that is not arbitrage--that is making a bet that can go wrong.
I should know better than to use Wikipedia as a source, but their statement is
Long-Term Capital Management (LTCM) was a U.S. hedge fund which used trading strategies such as fixed income arbitrage, statistical arbitrage, and pairs trading, combined with high leverage.
At any rate, wouldn't it have been easier to initially say "Your explanation, Rick, was spot on, but I believe the strategy you're describing is more accurately refered to as a carry trade rather than an arbitrage."
The best material model for a cat is another, or preferably the same, cat. - A. Rosenblueth and N. Wiener (1945).
User avatar
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri »

larryswedroe wrote: First, LTCM never engaged in risk-free arbitrage or any arbitrage. Their basic strategy was to make bets on convergence--reversion to mean of historical pricing relationships. Like the spread between the on the run and the off the run Treasury. But that is not arbitrage--that is making a bet that can go wrong.
I have to disagree with Larry on the definition of arbitrage. Risk-free arbitrage only occurs in theory. Arbitrage is not risk-free. Any arbitrage is subject to risk. First there is model weakness. That is the quantitative assumptions underpinning the arbitrage are wrong. They are also subject to trading liquidity issues, thus although on paper a trade may appear to be risk free, there needs to be adequate liquidity to earn a risk-free profit. There is also counter-party risk, for example, assume Lehman Brothers was on the other side of your trade when they went bankrupt. Then there are settlement issues of you are trading across global markets. So, there are risks in "risk-free" arbitrage strategies.
larry, at any rate, wouldn't it have been easier to initially say "Your explanation, Rick, was spot on, but I believe the strategy you're describing is more accurately refered to as a carry trade rather than an arbitrage."
I would have agreed with that.

Rick Ferri
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Post by larryswedroe »

Sorry but because people use a term incorrectly doesnt make it right

The definition of arbitrage is the entering into of two contracts with exactly offsetting positions. So this issue about liquidity and models is totally wrong. You buy say an ounce of gold in NY and sell it at the same time in Zurich. That is arbitrage. There is no model and no liquidity issue.

Now people use the term in the wrong way, but that doesnt make it correct

Risk free arbitrage does happen not in theory but in practice. That is again a fact, not an opinion.

"arbing" the on the run vs the off the run is not arbitrage. It is speculating. Taking a position.

Now there can sometimes be counterparty risk, depending on the trade, but that is not the kind of risk at issue here. It was the price risk
Post Reply