Global Macro Trading Journal

Quote from Daal:

I just calculated the implied interest rate in the mini Gold contracts from NYSELIFFE(Essentially the size of the premium you pay for contracts further out). From June 2012 to Dec 2013 the rate came out at 0.09%, this is about ~0.22% for yearly financing of a $55K gold position with about $3K margin(Little less actually). Commissions for rollover costs are quite small($1.62 per contract I believe, 2 times a year should be enough, bid ask spreads are tight even going out many months)

Compared to 0.4% expense ratio for the ETF and something similar to that for putting gold in banks.
Margin requirements of course are much worse and this is where the a LOT of percentage points are given up if you don't use the futures. With the ETF you are forced to pay up 50% in cash(It's actually 100% because borrowing from your broker can quite expensive) and physical gold 100%. The futures require little less than 7% and it doesn't have to be cash, can be stocks, gov bonds, some fixed income ETFs, etc

This should add 2-3% a year for conservative fixed income portfolio for the next 10 years. In my case, since I live in Brazil. Its even more, real rates are quite high. Giving up 2-3% a year in order to satisfy some paranoia about having to have the gold in front of you is just insane in my view. At 2-3% one is overpaying a LOT for that Mad Max option

Plus its always nice to have positions that don't use up cash so I don't have to pay IB their 1.5% a year
More...

I agree you should generally use the most efficient way of getting exposure. I've not advocated favouring GLD over futures or physical, I would choose a mix of whichever provided the best characteristics. It's useful to have some physical, especially living in a South American country where a government raid on your brokerage accounts is a genuine risk. And futures have the drawback of incurring capital gains every time you roll over, whereas physical and GLD can be held indefinitely with no taxable gains.

I'm not sure where you get the 2-3% a year figure from. Let's say you hold 10% in gold, you are giving up 0.1x t-bill rates. At the moment that's virtually nothing. Even in somewhere with 10% t-bill rates, that's 1% per annum, and such a country definitely will have credit risk. As for other things riskier than t-bills (commercial paper, medium-term government bonds etc) - you shouldn't go into a 'conservative fixed income portfolio' with the majority of the gold portion because then you increase your exposure to capital losses from rate rises. Any yield-chasing is taking on extra risk, which you are already doing in the stock and bond portions. If you want higher return for higher risk, then increase your stock allocation, that's the most efficient way to do it.

Finally, I don't consider it rational to risk total loss of all life savings in return for an extra few basis points. Disasters happen and keeping 10-25% of your net worth in such a scenario is well worth a few dozen basis points per annum.
 
Quote from Ghost of Cutten:

I agree you should generally use the most efficient way of getting exposure. I've not advocated favouring GLD over futures or physical, I would choose a mix of whichever provided the best characteristics. It's useful to have some physical, especially living in a South American country where a government raid on your brokerage accounts is a genuine risk. And futures have the drawback of incurring capital gains every time you roll over, whereas physical and GLD can be held indefinitely with no taxable gains.

I'm not sure where you get the 2-3% a year figure from. Let's say you hold 10% in gold, you are giving up 0.1x t-bill rates. At the moment that's virtually nothing. Even in somewhere with 10% t-bill rates, that's 1% per annum, and such a country definitely will have credit risk. As for other things riskier than t-bills (commercial paper, medium-term government bonds etc) - you shouldn't go into a 'conservative fixed income portfolio' with the majority of the gold portion because then you increase your exposure to capital losses from rate rises. Any yield-chasing is taking on extra risk, which you are already doing in the stock and bond portions. If you want higher return for higher risk, then increase your stock allocation, that's the most efficient way to do it.

Finally, I don't consider it rational to risk total loss of all life savings in return for an extra few basis points. Disasters happen and keeping 10-25% of your net worth in such a scenario is well worth a few dozen basis points per annum.
More...

If the intention is to hold gold for the long-run I see no significant risk at all in having a 5-8 year gov bond portfolio(maybe with some agencies in it) that will be held to maturity. The fluctuations in the P&L due interest rate risk can be safely ignored(except for the agencies) and there is no additional credit risk given that the alternative are shorter-term bonds who carry the same risk
 
Problem with using Post-Nixon data is that you capture 2 gold bull markets and 1 bear. This makes it look better than it really is. A more fair way perhaps would be from WW2 and on
 
Good point about the tax difference though. Its a not a problem in my particular situation which is why I didn't took that into account but most US investors would have that issue
 
Quote from Daal:

Problem with using Post-Nixon data is that you capture 2 gold bull markets and 1 bear. This makes it look better than it really is. A more fair way perhaps would be from WW2 and on
More...

How does it make it look better than it really is? I am not talking just about the 40-year CAGR from Bretton Woods collapse up to today, I am talking also about the CAGR/max DD during ANY holding period during the last 4 decades. I.e. the 5 year returns, the 10 year returns, the 15 or 20 year returns.

For example, 1981-2000 saw a massive secular bear market for gold, falling from over $800 to $250, yet the Permanent Portfolio had only 2 losing years (down 4.9% in 1981, and down 0.9% in 1994) and made a compound return of 8%. The worst rolling 5 year return over that whole period was a CAGR of 5.9%.

It also works for all countries, even 3rd world banana republics, places that suffered foreign defeat and occupation, hyperinflation, depressions and collapses of the banking system, massive currency devaluations etc. In each case the diversified conservative portfolios massively outperform in terms of risk control (due to the gold and foreign stock holdings), yet still earn surprisingly respectable returns during good times.

There's no need to take my word for it - go back and test the data in periods like 1929-32, 1937, WWII, the panic of 1907, or Brazil for the last 50 years, or Germany during the hyperinflation or WWII, Japan in WWII or 1990-2008. Compare the returns and the drawdowns with a typical 50/50 stock/bond portfolio (which lost 100% in many countries over the last century), or your own suggested portfolio. What you will find is that having substantial foreign holdings protects against currency collapse, and having a decent chunk in gold protects against monetary debasement, or systemic failure.

Economic or systemic disaster is a real risk, especially outside the 1st world. If the USA can see stocks fall 89% or 60%, if the UK can see stocks fall 75%, then somewhere like Brazil can definitely have a much worse outcome, as it has had more than once in the past.
 
Quote from Daal:

If the intention is to hold gold for the long-run I see no significant risk at all in having a 5-8 year gov bond portfolio(maybe with some agencies in it) that will be held to maturity. The fluctuations in the P&L due interest rate risk can be safely ignored(except for the agencies) and there is no additional credit risk given that the alternative are shorter-term bonds who carry the same risk
More...

No, if you have 8 year government bonds yielding 2%, and then rates go to 20% and inflation hits 15%, then you lose big compared to someone who has their cash in t-bills, whose rates will go up significantly, allowing them to reinvest at higher rates every 3 months and keep pace with inflation.

If you have 25% of capital in gold, cash paid, then you have no credit risk on that portion. If you are long gold futures and have that 25% in Brazilian treasuries, then Brazil does a Greece or Argentina, then you lose a large chunk of that money. Furthermore, in a scenario of potential default, your capital losses on t-bills will be far less than on 5-8 year bonds, and you can exit the position with no losses at all if default does not occur in the next 3 months. With intermediate government bonds you have to wait 5-8 years to get paid back your principal, or alternatively sell at the market incurring a giant loss.

As I said, yield-hogging comes at the cost of increased risk. The purpose of gold in a portfolio is to reduce total portfolio risk. Therefore it makes little sense to yield hog to earn a bit more cash using margin, and thus massively increasing tail risk, in what is meant to be the disaster-hedge portion of your portfolio. Your approach is dangerously complacent about risks.
 
Quote from Daal:

Problem with using Post-Nixon data is that you capture 2 gold bull markets and 1 bear. This makes it look better than it really is. A more fair way perhaps would be from WW2 and on
More...

You do realize gold was fixed at $35/ounce during this time.
 
Quote from jj90:

Anyone have any comment on Jordan's statement today? Sounds more aggressive than previous speeches, he included the ready to act clause, not just the usual stand behind the peg phrase.

And on shorting GBP, I tried that in Q4 11, when the UK data was absolutely horrendous, and that didn't work. To ralph: you have some insight on the BOE or UK data that we don't know about?
More...

CFTC: Merc traders leaning heavily to one side of the boat on GBP, CAD, and CHF.

Short EURGBP seems to be the new fad.

http://www.efxnews.com/story/12301/speculative-investors-aggressively-betting-more-pounds-gains-cftc
 
Quote from Ghost of Cutten:

How does it make it look better than it really is? I am not talking just about the 40-year CAGR from Bretton Woods collapse up to today, I am talking also about the CAGR/max DD during ANY holding period during the last 4 decades. I.e. the 5 year returns, the 10 year returns, the 15 or 20 year returns.

For example, 1981-2000 saw a massive secular bear market for gold, falling from over $800 to $250, yet the Permanent Portfolio had only 2 losing years (down 4.9% in 1981, and down 0.9% in 1994) and made a compound return of 8%. The worst rolling 5 year return over that whole period was a CAGR of 5.9%.

It also works for all countries, even 3rd world banana republics, places that suffered foreign defeat and occupation, hyperinflation, depressions and collapses of the banking system, massive currency devaluations etc. In each case the diversified conservative portfolios massively outperform in terms of risk control (due to the gold and foreign stock holdings), yet still earn surprisingly respectable returns during good times.

There's no need to take my word for it - go back and test the data in periods like 1929-32, 1937, WWII, the panic of 1907, or Brazil for the last 50 years, or Germany during the hyperinflation or WWII, Japan in WWII or 1990-2008. Compare the returns and the drawdowns with a typical 50/50 stock/bond portfolio (which lost 100% in many countries over the last century), or your own suggested portfolio. What you will find is that having substantial foreign holdings protects against currency collapse, and having a decent chunk in gold protects against monetary debasement, or systemic failure.

Economic or systemic disaster is a real risk, especially outside the 1st world. If the USA can see stocks fall 89% or 60%, if the UK can see stocks fall 75%, then somewhere like Brazil can definitely have a much worse outcome, as it has had more than once in the past.
More...

There is no disagreement here, a certain portion on real assets will protect against this. Question is whether an index is better than gold or vice-versa. In the 70's recession gold fell by 50% while CRB dropped by 20%(And a certain portion of the CRB is gold), in 2008 it was the opposite

The 30's can be an unfair comparison given that gold prices were not set by market forces. Given that gold has shown the ability to go down in recessions(Sometimes by a lot like in the 70's) I'd say is far from clear whether it would have done well there
 
Quote from Ghost of Cutten:

No, if you have 8 year government bonds yielding 2%, and then rates go to 20% and inflation hits 15%, then you lose big compared to someone who has their cash in t-bills, whose rates will go up significantly, allowing them to reinvest at higher rates every 3 months and keep pace with inflation.

If you have 25% of capital in gold, cash paid, then you have no credit risk on that portion. If you are long gold futures and have that 25% in Brazilian treasuries, then Brazil does a Greece or Argentina, then you lose a large chunk of that money. Furthermore, in a scenario of potential default, your capital losses on t-bills will be far less than on 5-8 year bonds, and you can exit the position with no losses at all if default does not occur in the next 3 months. With intermediate government bonds you have to wait 5-8 years to get paid back your principal, or alternatively sell at the market incurring a giant loss.

As I said, yield-hogging comes at the cost of increased risk. The purpose of gold in a portfolio is to reduce total portfolio risk. Therefore it makes little sense to yield hog to earn a bit more cash using margin, and thus massively increasing tail risk, in what is meant to be the disaster-hedge portion of your portfolio. Your approach is dangerously complacent about risks.
More...

This argument would be almost the same as saying that float in the insurance industry is bad. Yes you COULD be hit by a lighting but the vast majority of that time you will be fine. I simply won't accept that idea that one can't build a conservative portfolio with positive expectation with that cash

You could hold a basket of develop world gov bonds with 1-3 year maturies or basket of currencies earning money market rates,etc. But even in T-Bills, they won't be 0% forever and your portfolio is supposed to be about the very long-term. Its extremely likely that the next 30y returns in T-Bills will be something nice, probably the inflation rate minus taxes or something along those lines
 
You must log in or register to reply here.
Back
Top