Commodity Mutual Fund Correlated with S&P, Diversifying?

[Relmiw]

If I am adding a small commodity mutual fund to my portfolio with the aim of diversification, is it a problem if the commodity fund attempts to and succeeds in mirroring the growth of the S&P 500?

I thought the point of investing in commodities is to minimize risk by balancing asset allocation. Why then would a commodity mutual fund have such a tactic? I'm specifically reffering to
Credit Suisse Commodity Return Strategy (CRSOX)

https://personal.vanguard.com/us/funds/risk?FundId=9986&FundIntExt=EXT

That is an incredibly high correlation value. I like the idea of a broad commodity fund, and this is the lowest expense ratio I've seen on one. However, I fail to see the point if it is going to take hits in the same patterns as stocks.

I'd rather not invest in a concentrated commodity fund, but I can't find a broad fund with a lowish expense ratio and little correlation with the stock index.

[jpg7n16]

If it makes you feel any better, the fund has a minimum initial purchase amount of $250,000

https://personal.vanguard.com/us/funds/fees?FundId=9986&FundIntExt=EXT

So there ya go.


The point of adding commodities to the portfolio is very similar to why you add healthcare, technology, or energy sectors if you didn't have them already. It spreads the risk out to a different section of the economy. Any R^2 less than 1, no matter how less, will add to the diversification of the overall portfolio.

But commodities shouldn't be a huge portion of your account either. Maybe 5-10% or so.

[Relmiw]

If it makes you feel any better, the fund has a minimum initial purchase amount of $250,000

https://personal.vanguard.com/us/funds/fees?FundId=9986&FundIntExt=EXT

So there ya go.


The point of adding commodities to the portfolio is very similar to why you add healthcare, technology, or energy sectors if you didn't have them already. It spreads the risk out to a different section of the economy. Any R^2 less than 1, no matter how less, will add to the diversification of the overall portfolio.

But commodities shouldn't be a huge portion of your account either. Maybe 5-10% or so.

Thanks for your response. I'm aware of all that jpg, but there are commodity funds with a squared correlation coefficient lower than that value. This commodity fund aims at having the highest r^2 value as possible, and 0.96 is very close. How is that desirable for someone aiming to diversify? Is someone aware of a commodity fund not limited to a certain sector that isn't highly correlated with large cap stocks and has an expense ratio lower than 1, or even 2%?

[snafu]

Depending on your risk tolerance you might look at Commodity Exchange Traded Funds. It's a roller-coaster but if you're young and can stomach the ride, it has potential to give you good returns.

[kv968]

I take it by "commodity" you mean something like precious metals and the like. Finding a good mutual fund that's reasonably cheap with a decent initial invesment amount that holds the metals or their futures seems about impossible. The best you could probably do is one that holds the companies that are in that sector (i.e. VGPMX).

If you want to get exposure to the metals themselves, you'd be better off going the ETF route. Powershares Commodity Index might do it (DBC). It has a 3 year r^2 of 0.84 to the S&P and a one year of 0.16. If you're not totally dedicated to just metals but just want diversification, you could also look into other sectors such as utilities (XLU) which has a 0.82 r^2 for 3 years and 0.59 for a year and currently pays a 3.9% dividend. There's also Powershares Agriculture (DBA) which has is 0.64 for 3 years and -0.51 for a year. Then there's always good old gold (GLD) with a 3 year 0.77 and one year -0.37. Even a REIT should give you some decent diversification.

Finding investments that aren't tightly correlated can be somewhat tough anymore but it can be done. Plus you have to make sure what index it is they're comparing the investment to in order to get the r^2 number. It's not always the S&P 500. Sometimes it's to a similar index just to show you how tightly the ETF or fund tracks it so in that case you'd always get a number real close to 1 which means nothing so far as what you're looking for. Of course if you want REAL non-correlation you can go the inverse ETF route but I wouldn't really suggest holding them long-term.

Unfortunately, when the entire market really crashes one of the only things that really goes up is correlation.

[jpg7n16]

Thanks for your response. I'm aware of all that jpg, but there are commodity funds with a squared correlation coefficient lower than that value. This commodity fund aims at having the highest r^2 value as possible, and 0.96 is very close. How is that desirable for someone aiming to diversify? Is someone aware of a commodity fund not limited to a certain sector that isn't highly correlated with large cap stocks and has an expense ratio lower than 1, or even 2%?

Well, quick word on R^2, it's not the only thing that matters. And R^2 doesn't always match to the S&P 500.

Something I didn't really think of before, R^2 is a measure against the funds benchmark index not a measure against the overall market. I didn't really consider that it's not always against the S&P 500.

From: R-Squared Definition | Investopedia

Definition of 'R-Squared'
A statistical measure that represents the percentage of a fund or security's movements that can be explained by movements in a benchmark index. For fixed-income securities, the benchmark is the T-bill. For equities, the benchmark is the S&P 500.


Investopedia explains 'R-Squared'
R-squared values range from 0 to 100. An R-squared of 100 means that all movements of a security are completely explained by movements in the index. A high R-squared (between 85 and 100) indicates the fund's performance patterns have been in line with the index. A fund with a low R-squared (70 or less) doesn't act much like the index.

A higher R-squared value will indicate a more useful beta figure. For example, if a fund has an R-squared value of close to 100 but has a beta below 1, it is most likely offering higher risk-adjusted returns. A low R-squared means you should ignore the beta.

So now that I think about it, having a high R^2 is what any index fund tries to do.

So a high R^2 means that the fund closely matched it's benchmark index. In this case, the fund's benchmark index is not the S&P 500. Although I realize that's what the Vanguard site says, Morningstar compares it to a commodity only index and has the same value of 96.17.

Credit Suisse Commodity Ret Strat Instl (CRSOX) Fund Risk and Morningstar Rating


And by the fund's own prospectus, they are not trying to match the S&P, they are trying to match a commodity index

From: CRSOX: Credit Suisse Commodity Ret Strat Instl Fund Prospectus 485BPOS Filling

The fund is designed to achieve positive total return relative to the performance of the Dow Jones-UBS Commodity Index Total Return ("DJ-UBS Index"). The fund intends to invest its assets in a combination of commodity linked-derivative instruments and fixed income securities. The fund gains exposure to commodities markets by investing through the Subsidiary and in structured notes linked to the DJ-UBS Index, other commodity indices, or the value of a particular commodity or commodity futures contract or subset of commodities or commodity futures contracts. The value of these investments will rise or fall in response to changes in the underlying index or commodity.

Nonetheless, in the end having a portion of your portfolio in commodities is just like having a portion in the healthcare sector - it spreads the risk out to a different section of the economy (as stated above).

[kv968]

Nonetheless, in the end having a portion of your portfolio in commodities is just like having a portion in the healthcare sector - it spreads the risk out to a different section of the economy (as stated above).

Having a portion of your portfolio in commodities isn't really just like having it in the healthcare sector if its correlation is sufficiently different. Sure, investing in different sectors does offer varying degrees of diversification but its not all the same.

DBC (commodities) has a 3-year r^2 to the S&P of 0.84 and a one year of 0.16. The healthcare sector (XLV) has a 3-year of 0.92 and a one year of 0.96 (not much non-correlation there). Granted, anything below 1 isn't perfectly correlated but anything above 0.9 isn't doing much either in my opinion when looking specifically for such a non-correlation.

I'd say probably the best thing to hold for diverisfying from the S&P would the VIX (VXX). That has a 3-year of -0.88 and a one year of -0.89. Not THAT'S some non-correlation. The only problem I would have with holding something like that for any real amount of time is that it uses futures and is in contango most of the time. Which, over time, will constantly drag the value down if there isn't any significant change.

[jpg7n16]

Having a portion of your portfolio in commodities isn't really just like having it in the healthcare sector if its correlation is sufficiently different. Sure, investing in different sectors does offer varying degrees of diversification but its not all the same.

But it is sufficiently different. Have you ever compared a graph of how the fund CRSOX compares to the market? You can tell by looking at it that they don't move in tandem.

Just for kicks, compare CRSOX to SPY (stock market index) and to GSG (commodity index). When you look at a chart you'll see which one CRSOX really compares to. (Hint: Vanguard has it wrong on their site)

And you'll hopefully understand why CRSOX doesn't move just like the stock market, and has a place in your portfolio.


Just some food for thought - what is the R^2 of the Vanguard Total Bond Market Index Fund (VBMFX)??

https://personal.vanguard.com/us/funds/snapshot?FundId=0084&FundIntExt=INT#hist=tab%3A2

So are you saying that it's not beneficial for your portfolio to have bonds? They have a higher R^2 than commodities.

R^2 doesn't mean "correlation to the stock market" -- it means "correlation to the benchmark index."


Besides, my point was not "Commodities = Healthcare"

My point was, it's a different section of the economy. Like healthcare, or energy, or utilities, or consumer staples (please see my post above that I was referencing).

If healthcare laws change, commodities and energy will not be affected. If a new energy technology comes out, commodities and healthcare won't be affected. If Apple posts a bad quarter, it won't directly affect the price of commodities, healthcare, or energy.

By having a portion of your portfolio in commodities, you have a portion insulated from the economic effects to other sectors of the economy... just like healthcare, etc.

Also, if the prices of commodities start to rise, you won't benefit unless you own any.

DBC (commodities) has a 3-year r^2 to the S&P of 0.84 and a one year of 0.16. The healthcare sector (XLV) has a 3-year of 0.92 and a one year of 0.96 (not much non-correlation there). Granted, anything below 1 isn't perfectly correlated but anything above 0.9 isn't doing much either in my opinion when looking specifically for such a non-correlation.

Are you saying you shouldn't have any healthcare sector in your portfolio because it's too correlated to the market?

When you think through it logically, it makes sense to have some commodities. It's a different section of the economy that behaves differently, and doesn't move in tandem with the stock market - therefore there are diversification benefits to having a portion of your portfolio in commodities.

You're getting lost in numbers that don't mean much. R^2 isn't really that important. You could make it through life just fine and never know what the R^2 is for a fund/stock you own.

[kv968]

But it is sufficiently different. Have you ever compared a graph of how the fund CRSOX compares to the market? You can tell by looking at it that they don't move in tandem.

I wasn't saying that the commodities aren't different. I was trying to point out that the healthcare sector didn't differ very significantly from the S&P. I get that you're not saying healthcare=commodities, I'm just using that as an example of a sector you brought up. The whole thing was the OP was looking specifically to diverisfy from the S&P.

Just for kicks, compare CRSOX to SPY (stock market index) and to GSG (commodity index). When you look at a chart you'll see which one CRSOX really compares to. (Hint: Vanguard has it wrong on their site)

I have compared CRSOX and GSG to the SPY and they do offer diversification and I'd suggest holding a small portion of commodities for that reason. I just thought (as did the OP) that the non-correlation would have been higher. Granted he was looking at the wrong r^2 number but even the right one was less than what I thought it would be.

Just some food for thought - what is the R^2 of the Vanguard Total Bond Market Index Fund (VBMFX)?

So are you saying that it's not beneficial for your portfolio to have bonds? They have a higher R^2 than commodities.

R^2 doesn't mean "correlation to the stock market" -- it means "correlation to the benchmark index."

As far as bonds, BND (Vanguard's ETF of the mentioned fund) has a 3-year r^2 of 0.56 and a -0.11 for one year against the SPY. How is that higher than commodities?

Are you saying you shouldn't have any healthcare sector in your portfolio because it's too correlated to the market?

When you think through it logically, it makes sense to have some commodities. It's a different section of the economy that behaves differently, and doesn't move in tandem with the stock market - therefore there are diversification benefits to having a portion of your portfolio in commodities.

Again, I'm not saying that you shouldn't have any healthcare, or any other sector for that matter, in your portfolio for diversification. I totally agree with you and I hope I'm not coming off as saying diversification into different sectors can't work. I'm just trying to point out that they may not be as uncorrelated as someone may think.

You're getting lost in numbers that don't mean much. R^2 isn't really that important. You could make it through life just fine and never know what the R^2 is for a fund/stock you own.

I may be getting "lost in the numbers" but I would also have to disagree that, when applied right, they don't mean much. If you have 4 different sectors that have a correlation to each other of say 0.93 you could think you're well diversified because you "own different sectors". However when the market takes a turn, whether that be up or down, and they all do pretty much the same exact thing, how have you reduced your risk?

And you're totally right, one could make it through life just fine never knowing r^2, beta, alpha and all that good stuff. However, you're a financial advisor right? You don't pay at least a little attention to any of that? How do you quantify the risk of your client's portfolios? If a client asks you how much risk they're exposed to or how their portfolio will possibly act in a certain situation, what do you tell them, "Don't worry about it, I have you in a few different sectors?" I'm pretty sure you're more thorough than that and to be so you gotta use, or at least look at, some metrics that the average person doesn't.

Don't get me wrong, numbers (especially statistics) can be misleading and overused, and maybe I do the latter at times, but I also believe they can serve a purpose and give you an insight as to what's going on.

[jpg7n16]

As far as bonds, BND (Vanguard's ETF of the mentioned fund) has a 3-year r^2 of 0.56 and a -0.11 for one year against the SPY. How is that higher than commodities?

Because the R^2 is a measure of correlation to the benchmark. CRSOX correlates to its benchmark at .96, BND correlates to its benchmark at .99. Therefore it has a higher R^2.

My point was that you cannot just look at a number on a 'risk' page and know everything about what's going on. You have to consider what the number is comparing, and why it's that way.

I may be getting "lost in the numbers" but I would also have to disagree that, when applied right, they don't mean much. If you have 4 different sectors that have a correlation to each other of say 0.93 you could think you're well diversified because you "own different sectors". However when the market takes a turn, whether that be up or down, and they all do pretty much the same exact thing, how have you reduced your risk?

Because if new healthcare legislation comes out (good or bad), I can guarantee that the healthcare sector will not move in line with historical correlation. Imagine a new bill comes out that requires everyone to visit their doctor once a year by law. The healthcare industry would strongly benefit from that. Do you think the overall market will benefit to the same degree?? How about 94% of the same degree? No way.

But it's R^2 is .94... Why didn't it move the same way??

Diversification is more than just the everyday fluctuations of the market (which can be explained in part by R^2).

And you're totally right, one could make it through life just fine never knowing r^2, beta, alpha and all that good stuff. However, you're a financial advisor right? You don't pay at least a little attention to any of that? How do you quantify the risk of your client's portfolios? If a client asks you how much risk they're exposed to or how their portfolio will possibly act in a certain situation, what do you tell them, "Don't worry about it, I have you in a few different sectors?" I'm pretty sure you're more thorough than that and to be so you gotta use, or at least look at, some metrics that the average person doesn't.

If you're a normal person, and you have concerns about your portfolio, and your advisor comes at you with facts and figures and starts talking about R^2 and correlation coefficients... what would you do? Normal people would thank me for the info, get up, walk out, and go to a new advisor who they could understand.

Being an advisor isn't about talking over your client's head. It's about keeping the client on track towards their goals.

And no, I don't pay attention to the R^2 of different funds.

You don't need to know it in order to describe to a client what happened. "We've got a sizable stock portion of the account, and you've seen what's gone on with the market lately. We expect volatility along the way, but we're spread out well across the economy going forward. We're not going to bet your whole account on one segment of the economy."

Well why did you guys buy commodities in my portfolio? They've been doing terrible!

"You're right, they haven't been doing that great recently. We expect that at some point inflation will start to kick back in. Inflation, as you know, is rising prices. Well what prices are going to be rising? Food costs, gas prices, raw materials - aka commodities. We don't know when that will happen, and we're not going to bet your whole portfolio on the price of corn, but we do feel you need a portion of your portfolio in commodities to take advantage of that opportunity."

Explains it all just fine without having to ever look at an R^2 figure.

[kv968]

Because the R^2 is a measure of correlation to the benchmark. CRSOX correlates to its benchmark at .96, BND correlates to its benchmark at .99. Therefore it has a higher R^2.

My point was that you cannot just look at a number on a 'risk' page and know everything about what's going on. You have to consider what the number is comparing, and why it's that way.

I understand what r^2 is and what it's measuring and using it for an index fund such as BND is practically useless since it shouldn't be anything less than 0.95. If it is there's a serious problem with the fund/ETF or it's the wrong index.

However, if you read my post, I'm not using the given r^2's that compare the ETF's I've mentioned to its benchmark. I'm comparing them to the SPY (S&P 500) which was the benchmark the OP was looking at.

And I agree, you can't just look at one number and think you know the entire picture. However, you can collectively use the numbers to make that picture possibly a little clearer.

Because if new healthcare legislation comes out (good or bad), I can guarantee that the healthcare sector will not move in line with historical correlation. Imagine a new bill comes out that requires everyone to visit their doctor once a year by law. The healthcare industry would strongly benefit from that. Do you think the overall market will benefit to the same degree?? How about 94% of the same degree? No way.

But it's R^2 is .94... Why didn't it move the same way??

Diversification is more than just the everyday fluctuations of the market (which can be explained in part by R^2).

I'm sure the healthcare sector will move quite differently than the overall market with the healthcare legislation ruling (although I'm not sure how much it'll move since I think a "no" vote is already kind of factored in). And that could be one of the 6% of the times that it doesn't move in tandem.
A
nd you're right, diversification doesn't mitigate everyday fluctuations in the market nor does r^2 (or any number for that matter) always capture what's truly going on. However if you look at some of the numbers I've posted for a 3-yr time span you can get a better picture of how certain investments may react.

If you're a normal person, and you have concerns about your portfolio, and your advisor comes at you with facts and figures and starts talking about R^2 and correlation coefficients... what would you do? Normal people would thank me for the info, get up, walk out, and go to a new advisor who they could understand.

Being an advisor isn't about talking over your client's head. It's about keeping the client on track towards their goals.

And no, I don't pay attention to the R^2 of different funds.

You don't need to know it in order to describe to a client what happened. "We've got a sizable stock portion of the account, and you've seen what's gone on with the market lately. We expect volatility along the way, but we're spread out well across the economy going forward. We're not going to bet your whole account on one segment of the economy."

Well why did you guys buy commodities in my portfolio? They've been doing terrible!

"You're right, they haven't been doing that great recently. We expect that at some point inflation will start to kick back in. Inflation, as you know, is rising prices. Well what prices are going to be rising? Food costs, gas prices, raw materials - aka commodities. We don't know when that will happen, and we're not going to bet your whole portfolio on the price of corn, but we do feel you need a portion of your portfolio in commodities to take advantage of that opportunity."

Explains it all just fine without having to ever look at an R^2 figure.

I could definitely see a client high-tailing it out of there if you were to start talking about r^2, alpha and who knows what else Your job is to make it understandable to your clients and they may not need to know, nor care, about all the nuts and bolts of the portfolio. And I sympathize with you for having to try to explain to a client why they were in a sector or asset that's been going down. However, as an advisor I would hope that you'd know and understand the metrics behind the scenes.

Let's say a person comes in and doesn't want a very volitile portfolio but wants to be invested mainly in stocks. Do you look at the beta of the stocks or funds you put them in or do you just look at a chart and guesstimate what the volitility would be? Albeit it numbers aren't perfect by any means but looking at a chart isn't exact either. Again, all I'm saying is use as many tools as you can to get the bigger picture in focus. Try not to overdo it, but I think they are something you should incorporate. I know you probably think I just look at and crunch numbers and go by that, but I really don't. I do look at them, take them into consideration with everything else available to me and make my decisions from all the collected info.

Now I'm curious...what numbers, metrics, charts, etc...(if any) do you use to figure out risk/diversification in your or your client's portfolio's? I understand that each client's situation is unique, but what do you use to match them up investment-wise with something that would be appropriate for their personal situation?

[jpg7n16]

I'm sure the healthcare sector will move quite differently than the overall market with the healthcare legislation ruling (although I'm not sure how much it'll move since I think a "no" vote is already kind of factored in). And that could be one of the 6% of the times that it doesn't move in tandem.

And that's why diversification into sectors even with a higher R^2 is important

Let's say a person comes in and doesn't want a very volitile portfolio but wants to be invested mainly in stocks.

Then that client is in need of education

Now I'm curious...what numbers, metrics, charts, etc...(if any) do you use to figure out risk/diversification in your or your client's portfolio's? I understand that each client's situation is unique, but what do you use to match them up investment-wise with something that would be appropriate for their personal situation?

The main thing I'd look at is overall asset allocation percentages, the sector weightings, and any concentrated positions. Fixing errors in those categories can solve a lot of problems people face in their portfolio.

[kv968]

Let's say a person comes in and doesn't want a very volitile portfolio but wants to be invested mainly in stocks.

And that's why diversification into sectors even with a higher R^2 is important

Then that client is in need of education

The main thing I'd look at is overall asset allocation percentages, the sector weightings, and any concentrated positions. Fixing errors in those categories can solve a lot of problems people face in their portfolio.

That's true, the client probably does need to have a talk. However there are ways to tamper the volitility in a stock portfolio besides just being in different sectors (i.e. lower beta stocks in a sector) and I was just wondering, since you don't seem to really use the numbers, how you went about it.

Do you do any sector rotation or do you just rebalance periodically?

[Relmiw]

Thanks for the advice guys. It's good to know that Vanguard had the r^2 value wrong. I'm putting about 5% into a commodity index but haven't quite decided which one yet.

[kv968]

Thanks for the advice guys. It's good to know that Vanguard had the r^2 value wrong. I'm putting about 5% into a commodity index but haven't quite decided which one yet.

It's not that Vanguard had the value wrong, they didn't. They just weren't comparing it to the S&P which is the "standard index" for most things. They were comparing it to how closely it tracked the index it was following instead. You always have to watch how these numbers are generated.

It would be nice if they provided the numbers against the S&P though.