allocation

[bosco]

Quote:

Originally Posted by brewer12345 Ding ding ding!!! We have a winner. Yrs, you are missing the most important part of what the research says: a diversified portfolio of fully collateralized commodity futures is NEGATIVELY CORRELATED with stock and bond futures while having about the same risk premium and volatility as equities. Negatively correlated means that the futures portfolio does well when stock and bond returns are lousy, and vice versa. It is as close to a free lunch as exists in the world of investing and finance. Take a few minutes and read the research raddr has done and cobbled together. He says it a lot more eloquently than I can: http://raddr-pages.com/research/CommodityFutures.htm I find the research sufficiently convincing to maintain a minimum of 10% of my portfolio in PCRDX. If I ever choose to have a heavier bond allocation, that percentage will rise.

hmmm....from looking at this graph of PCRDX versus S&P, it seems that there may be some negative correlation on a short-term basis which would smooth out short-term volatility. But it's far from obvious that this will result in the kind of long-term hedge that people are hoping for.

[brewer12345]

That's because all the studies use decades worth of data. Your chart shows, what, a couple of years? Not statistically significant.

[Gone4Good]

brewer,
I've read the research.* I am, however, highly skeptical of research produced by people trying to sell me something (e.g. Greer/PIMCO).* The research also runs contrary to my understanding of how futures work (they are not insurance products) and my observations (most commodities under perform inflation for long periods of time), which only increases my skepticism.*

Greer completely dismisses the traditional “cost of carry” model for pricing futures contracts without so much as an explanation.* He constructs an elaborate example around cattle futures, but does not explain why the constraints in that market should apply to metals, agriculture, and energy.*

If the cost of carry model holds, then the interest accrued from “collateralization” simply compensates for the cost of carry imbedded in the futures contract.* Buying a 1 year forward contract for gold and buying an equivalent notional amount of 1 year treasuries should yield the same result as buying the gold outright.* Any other outcome means arbitrage profits are available for the taking.*

I’ve read other research suggesting that long-dated futures contracts do have some “insurance premium” built in because of a lopsided market for sellers and buyers.* The research focused on trading strategies to extract that premium.* I can only assume that whatever premium existed in the past, if one ever existed at all, is rapidly being arbitraged away.*

In any event, these sources of returns over and above the return generated by the underlying commodity must be pretty generous to offset the horrible inflation adjusted returns of commodities (as shown below).* Greer says they are, but I can't prove / observe that.

Here are some examples of commodity performance for your viewing pleasure:

In all cases the white line is the commodity and the green line is the CPI index.* The differing time periods reflect the limitations of Bloomberg's historic data.


1) Gold





2) Copper* (would have been better off burying your cash in mason jars)




3) Soy beans (not any better)





4) Oil (energy commodities are the only ones that seem to track inflation well)





5) Here is a comparison of the S&P 500, 30-Year Treasury Bond and DJAIG





Good luck to everyone!

[brewer12345]

Only problem with all those pretty charts is that they don't actually include:

A) the DJAIG plus tips returns, and

B) any environmennts other than generally falling interest rates and inflation.

A more detailed rebuttal will have to wait fort a time when I'm not holding a squirming toddler.

[wabmester]

Quote:

Originally Posted by brewer12345 A more detailed rebuttal will have to wait fort a time when I'm not holding a squirming toddler.

I can vouch for brewer. This is not a cop-out. Those little buggers are cute, but they strictly enforce a 2-minute attention span for anybody in the room house.

[Gone4Good]

Quote:

Originally Posted by brewer12345 A) the DJAIG plus tips returns, and B) any environmennts other than generally falling interest rates and inflation. A more detailed rebuttal will have to wait fort a time when I'm not holding a squirming toddler.

A) These are prompt prices. You can't just add an interest rate of return. See "cost of carry model." There is an imbedded time value of money in longer dated commodity futures. As stated in the text, adding a future contract to a bond creates a synthetic commodity which should have the same return attributes as the underlying commodity.

B) Generally true for all recent investments. But if you're going to bill something as an inflation hedge I would think that it should keep pace with inflation at some point over the past 30 YEARS. Jeesh!

[brewer12345]

Quote:

Originally Posted by . . . Yrs to Go A)* These are prompt prices.* You can't just add an interest rate of return.* See "cost of carry model."* There is an imbedded time value of money in longer dated commodity futures.* As stated in the text, adding a future contract to a bond creates a synthetic commodity which should have the same return attributes as the underlying commodity. B)* Generally true for all recent investments.* But if you're going to bill something as an inflation hedge I would think that it should keep pace with inflation at some point over the past 30 YEARS.* Jeesh!

Still holding the squirming kid, but it looks to me like the DAIG has kept up with CPI lately, although it is hard to tell since you didn't put the two on the same chart. Care to do so?

[Gone4Good]

Quote:

Originally Posted by brewer12345 Still holding the squirming kid, but it looks to me like the DAIG has kept up with CPI lately, although it is hard to tell since you didn't put the two on the same chart.* Care to do so?

I will.* But it won't be until Tuesday (Bloomberg is at work).* From my back of the envelope calcs (and recollection) it looks like DAIG underperformed CPI from about 1990-2000.* In May of 2000 DAIG started to rally and overtook CPI by late 2004.* I believe DAIG has an annualized "real" return from 1990-today of somewhere around 1% +/-.***

But almost all of that positive real return is attributable to the last few years.* We don't know what it looked like before that.* But judging from the returns of some of the other components of DAIG (energy excepted) I would expect negative real commodity returns for the last 20-30 years.*

Even if the current commodity rally continues and brings the long-run trend back in line, or better than, inflation the cycles seem ridiculously long.* We're all long-term investors but if I buy something to hedge against inflation I'd expect it to keep pace with no more than a few years lead / lag.* Otherwise it doesn't seem that useful for its intended purpose.*

I don't argue that commodities can help diversify an investment portfolio.* I hold some cash and other assets that probably don't provide positive real returns over the long run.* I'm just cynical given all the recent hype about commodities, cautious due to there recent performance, and skeptical of the claims made by some who are trying to sell securities.*


** Edit. Eyeballing the charts it looks like DJAIG may have crossed back into positive territory for a short time in early 1997. It's not clear why that would be considering the other commodities shown here were essentially flat during that time period. Maybe a rally in some other part of the index - I'll investigate but it may take a while.

Enjoy your weekend!

[greg]

My comments for what they're worth (under the KISS umbrella):

I like PM miners or a diversified PM fund such as VGPMX for a few reasons: First, the miners have some leverage over gold itself. I have heard that it is 2-3 times the price of gold, so that if gold goes up 20% then the mining stocks should go up 40-60%. It's just the volitility in between that is the killer--and the timing. I think a smaller position in PM shares, especially if one is leary of commodities in general could be a safety bet if things go haywire with inflation in the future. I think 2-3% of the total portfolio is a reasonable allocation for an apocalyptic hyper-inflation event alone. The Hussman article posted on Raddr's site does a good job of explaining how gold functions in a rising interest rate environment--well worth the look.

As mentioned before, a significant drop in the US$ could also be an upcoming event in the near to middling future, due to our increasing dependence on external financing for spending/consuming. Most economic critics agree that at some point, other countries just won't want our $$ anymore, or if they do it will have to be at a higher interest rate. If the dollar devalues, I suspect inflation will rocket up, interest rates will jump, and, of course, your portfolio will appear appear to be the same size but actually buy far fewer goodies--and necessities.

If any of the above is even a slight possibility, would you want all your assets in US financial (read "paper") instruments or in something more tangible--such as commodities.

Although I've never been in the military, I believe that strategic planning there and in other fields says you should "Plan for the worst, but hope for the best." I think this is very important for your portfolio allocation too. Portfolio focus shouldn't be limited to "highest total safe return" and a mix of assets based solely on that strategy. This can skew your results if something unexpected happens.

--Greg

[wildcat]

Yrs to go-

My argument has been for the diversificataion 1st and 2nd a free option on any potential bull market cycle in commodities. Aside from all of that (and the inflation argument) do have any charts or facts on historical supply/demand of raw commodities? I think that would look pretty interesting to observe.

[Gone4Good]

Quote:

Originally Posted by wildcat do have any charts or facts on historical supply/demand of raw commodities?* I think that would look pretty interesting to observe.*

Wildcat, I don't have any of that info on hand.* I know the Energy Information Agency (of the DOE) publishes a lot of that kind of information but it's specific to oil, gas, electricity, etc.* If you feel like plowing through it you can take a look here:

http://www.eia.doe.gov/

You can also look here for a longer dated oil price series than what is shown on the Bloomberg images I posted.*

http://www.eia.doe.gov/emeu/aer/txt/ptb0518.html

These are "well-head" prices so they are lower than what you will see quoted in the paper.* But the left most column shows oil prices in constant 2000 dollars dating back to 1949.* It's pretty interesting.* If you chart it out you will see that "real" oil prices declined consistently from 1949 to about 1972.* Then they spiked in the early 1970's and continued to rise until peaking in 1981.* Since then "constant dollar" oil prices have reset to around $20 / barrel (higher than the level before 1970 - I guess due to OPEC).* *It has fluctuated, with pretty wide variation, around that level ever since.* But in 2004 oil burst out of the trading range it had held for most of the past 20 years - recently approaching its all time high.* Whether it resets at a new level again because of growing world demand or falls back to its historic trading range (about $30 / bbl in today's dollars) is anyone's guess.

Not what you asked for, but interesting stuff (to me anyway) nonetheless.*

[wildcat]

That is what I wonder about. Does oil hit a new range where $40 is as low as it gets, i.e. is $30 oil gone for good? I have no idea. I don't know if it could just be an oil based scenario or if it could be spread across other commodities. It could be a bumpy ride (drops due to periodic slowdowns in worldwide growth) but the scenario is possible in my head.

[Gone4Good]

Quote:

Originally Posted by wildcat That is what I wonder about.* Does oil hit a new range where $40 is as low as it gets, i.e. is $30 oil gone for good?* I have no idea.* I don't know if it could just be an oil based scenario or if it could be spread across other commodities.* It could be a bumpy ride (drops due to periodic slowdowns in worldwide growth) but the scenario is possible in my head.

Beware of mean reversion

Naturally no one knows the answer to your question but oil prices have averaged about $21 / barrel over the past 20 years (excluding 2005). During that time constant $ prices have ranged from a high of $38 to a low of $11. So even if prices re-set to an average of $40 in constant dollar terms over the next 20 years, we could very well see periods of time with prices in the high teens.

[Gone4Good]

Quote:

Originally Posted by brewer12345 It looks to me like the DAIG has kept up with CPI lately, although it is hard to tell since you didn't put the two on the same chart.* Care to do so?

As promised.

[brewer12345]

Thanks, Yrs. Clearly the DJ-AIG is more volatile than CPI. If we add TIPS (or some sort of proxy, like CPI + 1.5%) to the DJ-AIG return, things actually look pretty good as far as performance versus inflation.