PP with 2x leverage (no 2x funds)

[Kbg]

#1 will be the cheapest by far and with a PP one should have plenty of cash buffer if leverage is kept to 2x...the downside is you would need around a min of $275K to run a balanced futures 2x PP

Not sure where you're coming up with such a ridiculous number. The S&P is only $4200, gold only $4500 and T-Bonds only $3650. And that will give you 5:1 margin.

Here's the math...

Well I assume as a starting point one is going to maintain a balanced PP for the three assets. If so, the long poll in the tent is the UB contract which doesn't come mini sized. Using today's price and rounding that's $177K you have to equalize to and at IB overnight margin on the UB is $5500. So rough/rounded math is:

UB 177K (Total Port not including cash is 531K (177x3) . Add 25% cash and you are now at a 708K portfolio)
ES 105K (balance with 72K of SPY...one could make this smaller with 3 SSFs/covers 63K/12,600 cash)
GC 127K (balance with 50K of GLD...one could make this smaller with 3 SSFs/covers 48K/11,600 cash)

Overnight margin of ES 5250, UB 5500, GC 5625 = 16375
Equalizing SSF margin = 24200
Leftover equalizing for ETFs SPY 9000, GLD 2000 = 11000

Min cash is 51,575 (that leaves you exactly zero cash buffer)

Assuming gearing is 2x (708K/2) = 354K in cash. This leaves around $300K as a buffer. 300K gives one the ability to absorb a 56% loss. Obviously you can do it as cheap as 51K (and your port will probably be good for a week tops), or 177K but then that is geared to a full 3x with 126K as a cash buffer and gives one the ability to absorb a 23% loss. (Loss calcs = cash / 531K)

[MachineGhost]

I think he was making the point that if you're shooting for ONLY 2x leverage, you would need $275k because of the large amounts you're controlling via futures. For instance a GC contract represents 100 oz of gold @ $1,200oz that would equate to $120k just in gold holdings.

Of course you can run a 20x portfolio with futures, but you may not last very long.

Oh, that make sesnse. If i did this right, then a 5x leveraged account would make only 13.16% CAGR with a -99.82% MaxDD on a nominal basis and is the maximum amount of leverage that can be used without wiping out. Obviously the actual CAGR would be higher on a relative basis with margin vs nominal.

Definitely the loan margin seems problematic (if it went past 2:1) and I would rank it below options. It adds a lot of risk for very little reward. Options are a major bitch to equalize but they're more or less the same as a 2:1 when buying deep ITM or generally 5:1 ATM. They are not perfect proxies, however, since they're influenced by more than just the underlying price movement. I only use them when the risk of a trade is no more than the cost of an option.

[Kbg]

Oh, that make sesnse. If i did this right, then a 5x leveraged account would make only 13.16% CAGR with a -99.82% MaxDD on a nominal basis and is the maximum amount of leverage that can be used without wiping out. Obviously the actual CAGR would be higher on a relative basis with margin vs nominal.

Those numbers don't seem right to me...and at least with futures you would have had to recapitalize with that large a DD so like AB does there is a "ruin" in there somewhere.

Another thing with futures, you really need to capture STT interest as a net to your account/CAGR results. Not a biggie right now but certainly in the 70s extra cash in STTs was a superb tailwind (and one of the reasons, if not '"the" reason, CTA returns looked so good for such a long time.). It would be very interesting to see how a futures-based PP did in the 70s on a very large account. I'll bet it was quite a bit better...but I'm pretty sure UB didn't go back that far so it's a moot question.

[MachineGhost]

Those numbers don't seem right to me...and at least with futures you would have had to recapitalize with that large a DD so like AB does there is a "ruin" in there somewhere.

I didn't run it as futures but as a margin loan. So I believe 5x was enough to recapitalize. It did not run out of money to buy trades, but 6x did.

Doing a futures backtest to 1970's wouldn't be realistic without using point-in-time margin requirements which change constantly. In fact, I recall reading about how the exchanges increased gold and silver margin requirements and that popped the bubble.

With a normal 2:1 RegT margin loan at 0% interest, I get 11.20% CAGR and -51.21% MaxDD which looks about right.

[jason]

Sure, and you could do this with options too, right? How do you think interest rate hikes will be reflected in the expenses of the leveraged ETFs?

I can't remember who is running a small options version of PP, but they post an update annually. I personally don't recommend one unless you are doing it as a speculative VP. Options are accurately priced for the most part which means that they are very tough to make any money on the long side. Higher interest rates should be a net benefit to the ETF holder not a cost as normally excess cash will be invested in very safe ST instruments and accrues to the owners. However, having noted this only a highly interesting and entertaining prospectus read will tell that story. If I can find it fast I'll post the results here. If I actually have to read a good chunk of it to find out I see no reason why I should have the only eyeballs bleeding and folks will need to send me a money order to find out the answer.

To me, in order, the best ways to do a leveraged PP are:

1. Futures
2. Leveraged ETFs
3. Margin
4. Options

#1 will be the cheapest by far and with a PP one should have plenty of cash buffer if leverage is kept to 2x...the downside is you would need around a min of $275K to run a balanced futures 2x PP
#2 I like due to better control vs margin...I'd bet this will option will be cheaper as well over #3
#3 Nothing to really add from previous posts, has the benefit of no daily reset/decay
#4 I can't backtest this, but I'd bet a burger and fries it's not a good approach long term from a performance perspective.

I don't recall reading anything about a futures-based PP. I'd really like to learn more about that. Are there any good articles or threads on that? I know pretty much nothing about futures.
Regarding to the leveraged ETFs, I also don't like the additional counter-party risk. I believe these ETFs are based on derivatives, so if the SHTF, the company underwriting the derivatives could disappear.

[MachineGhost]

Regarding to the leveraged ETFs, I also don't like the additional counter-party risk. I believe these ETFs are based on derivatives, so if the SHTF, the company underwriting the derivatives could disappear.

Leveraged ETF's just use futures, options and equity swaps which are all derivatives traded on exchanges so there's no counterparty risk.

[Kbg]

Regarding to the leveraged ETFs, I also don't like the additional counter-party risk. I believe these ETFs are based on derivatives, so if the SHTF, the company underwriting the derivatives could disappear.

Leveraged ETF's just use futures, options and equity swaps which are all derivatives traded on exchanges so there's no counterparty risk.

And a pile on...if LTT, Gold and SPX derivatives blow up in a SHTF scenario then I think the their original forms also blow up and you would want to be long farm land and guns. Seriously, these are all highly liquid, huge in global scale and standardized market contracts. Esoteric stuff, yeah worry about it. Not so much these.

[Kbg]

I don't recall reading anything about a futures-based PP. I'd really like to learn more about that. Are there any good articles or threads on that? I know pretty much nothing about futures.

Which is exactly why you should stay far away from them.

Futures are bad because people do stupid things with them. Inherently they are no better or evil than any other investment. But let's say your account was sufficiently large for an unleveraged PP consisting of only the futures mentioned earlier...there is no more efficient/lower cost method available to John/Jane Q Public to implement a PP than standard futures contracts. And in a taxable account if one is required to do a less than one year rebalance the tax rates are substantially better due to futures tax treatment. Honestly, I wish all my investments had the same tax treatment as futures. You get your Form 6781 from your broker with three lines noting total, ST (40%) and LT (60%) gains give it to your accountant/enter it into TaxWhatever Program and you are done. If all your trades were ST, 60% of profits are still regarded as LT gains.

[slk23]

How would one rebalance with futures? Is it possible to trade fractions of a contract? I assume not, which means trades are in approx. $100K blocks.

Also, the 20%-30% max. DD stated earlier for the PP sounds high. 1980 had the largest DD that I'm aware of: 20% according to http://www.peaktotrough.com/hbpp.cgi. Even 2008-2009, when things got very ugly, wasn't too bad: 14% according to http://www.etfreplay.com/combine.aspx.

[Sam Brazil]

How would one rebalance with futures? Is it possible to trade fractions of a contract? I assume not, which means trades are in approx. $100K blocks.

Also, the 20%-30% max. DD stated earlier for the PP sounds high. 1980 had the largest DD that I'm aware of: 20% according to http://www.peaktotrough.com/hbpp.cgi. Even 2008-2009, when things got very ugly, wasn't too bad: 14% according to http://www.etfreplay.com/combine.aspx.

For Gold, each e-micro MGC contract is like $13k, about 1/10th of GC normal gold futures contracts. So that makes for very granular control of gold.

For Stocks, if you went long ES ($104k/contract today) and short NQ ($89k/contract today), that leaves you net about 14k increments for you to play with. You could do it with the Dow or Russell instead for different combinations.

For long bonds, UB is about $192k/contract today and you could short ZB which is $172k/contract today, netting out to 20k increments to play with.

If you've got portfolio margin, I'm assuming that hedging in this fashion wouldn't reduce your buying power by much. Or even if you didn't you'd wind up with something like 10x leverage instead of 20x, but I'm too lazy to do the math.

[Kbg]

How would one rebalance with futures? Is it possible to trade fractions of a contract? I assume not, which means trades are in approx. $100K blocks.

Also, the 20%-30% max. DD stated earlier for the PP sounds high. 1980 had the largest DD that I'm aware of: 20% according to http://www.peaktotrough.com/hbpp.cgi. Even 2008-2009, when things got very ugly, wasn't too bad: 14% according to http://www.etfreplay.com/combine.aspx.

For Gold, each e-micro MGC contract is like $13k, about 1/10th of GC normal gold futures contracts. So that makes for very granular control of gold.

For Stocks, if you went long ES ($104k/contract today) and short NQ ($89k/contract today), that leaves you net about 14k increments for you to play with. You could do it with the Dow or Russell instead for different combinations.

For long bonds, UB is about $192k/contract today and you could short ZB which is $172k/contract today, netting out to 20k increments to play with.

If you've got portfolio margin, I'm assuming that hedging in this fashion wouldn't reduce your buying power by much. Or even if you didn't you'd wind up with something like 10x leverage instead of 20x, but I'm too lazy to do the math.

Sam...freaking awesome idea. I always am amazed at the thoughts others have that simply never occur to me. Obviously not a perfect match but definitely worth spending some time studying. Off the cuff, the Dow and the SPX minis might be a closer hedge and the spread is roughly 14k as well.

[slk23]

Interesting idea to short the smaller contracts. However, aren't some of them fairly low volume with large bid/ask spreads? I've read that's the case for the smaller gold contracts, for example. I think I'd feel more comfortable staying with the very liquid full size contracts, filling in where needed with the VTI, GLD, and TLT ETFs.

I've been playing with a paper trading account at IB. Using ES, GS, and UB futures contracts with VTI, GLD, and TLT ETFs for balancing does work quite well but since the futures contracts are large it requires a fair amount of capital to guard against a draw down inducing a margin call.

[edit]
I suppose one could be long the full size contracts and short the high liquid ETFs in order to reduce overall portfolio size. That would avoid the less liquid gold mini/micro contract.

For Gold, each e-micro MGC contract is like $13k, about 1/10th of GC normal gold futures contracts. So that makes for very granular control of gold.

For Stocks, if you went long ES ($104k/contract today) and short NQ ($89k/contract today), that leaves you net about 14k increments for you to play with. You could do it with the Dow or Russell instead for different combinations.

For long bonds, UB is about $192k/contract today and you could short ZB which is $172k/contract today, netting out to 20k increments to play with.

If you've got portfolio margin, I'm assuming that hedging in this fashion wouldn't reduce your buying power by much. Or even if you didn't you'd wind up with something like 10x leverage instead of 20x, but I'm too lazy to do the math.

[Kbg]

Interesting idea to short the smaller contracts. However, aren't some of them fairly low volume with large bid/ask spreads? I've read that's the case for the smaller gold contracts, for example. I think I'd feel more comfortable staying with the very liquid full size contracts, filling in where needed with the VTI, GLD, and TLT ETFs.

I've been playing with a paper trading account at IB. Using ES, GS, and UB futures contracts with VTI, GLD, and TLT ETFs for balancing does work quite well but since the futures contracts are large it requires a fair amount of capital to guard against a draw down inducing a margin call.

For Gold, each e-micro MGC contract is like $13k, about 1/10th of GC normal gold futures contracts. So that makes for very granular control of gold.

For Stocks, if you went long ES ($104k/contract today) and short NQ ($89k/contract today), that leaves you net about 14k increments for you to play with. You could do it with the Dow or Russell instead for different combinations.

For long bonds, UB is about $192k/contract today and you could short ZB which is $172k/contract today, netting out to 20k increments to play with.

If you've got portfolio margin, I'm assuming that hedging in this fashion wouldn't reduce your buying power by much. Or even if you didn't you'd wind up with something like 10x leverage instead of 20x, but I'm too lazy to do the math.

No doubt regular contracts are better, but that requires a very large account. One could do the math on spreads vs. annual fees and see what would be better.

I have an active trading system though that I'm definitely going to look at for the hedge method mentioned.

[Sam Brazil]

How would one rebalance with futures? Is it possible to trade fractions of a contract? I assume not, which means trades are in approx. $100K blocks.

Also, the 20%-30% max. DD stated earlier for the PP sounds high. 1980 had the largest DD that I'm aware of: 20% according to http://www.peaktotrough.com/hbpp.cgi. Even 2008-2009, when things got very ugly, wasn't too bad: 14% according to http://www.etfreplay.com/combine.aspx.

For Gold, each e-micro MGC contract is like $13k, about 1/10th of GC normal gold futures contracts. So that makes for very granular control of gold.

For Stocks, if you went long ES ($104k/contract today) and short NQ ($89k/contract today), that leaves you net about 14k increments for you to play with. You could do it with the Dow or Russell instead for different combinations.

For long bonds, UB is about $192k/contract today and you could short ZB which is $172k/contract today, netting out to 20k increments to play with.

If you've got portfolio margin, I'm assuming that hedging in this fashion wouldn't reduce your buying power by much. Or even if you didn't you'd wind up with something like 10x leverage instead of 20x, but I'm too lazy to do the math.

Sam...freaking awesome idea. I always am amazed at the thoughts others have that simply never occur to me. Obviously not a perfect match but definitely worth spending some time studying. Off the cuff, the Dow and the SPX minis might be a closer hedge and the spread is roughly 14k as well.

Thanks, I'm curious how viable it is in reality because it's just a theoretical idea I've been kicking around for a while. Speaking of not being a perfect match, I wonder if a similar idea could be used to lower PP taxes, especially on leveraged version that require more frequent rebalancing...e.g., when it's time to sell a PP asset before 1 year, if you're using leveraged ETFs, if you implement a futures hedge that's an imperfect hedge, hold until the 1 year mark, and then sell the PP asset, you can ensure long term capital gain tax rate instead of short term. But I'm sure there are all sorts of tax laws that make that a very complex issue.

[Sam Brazil]

Interesting idea to short the smaller contracts. However, aren't some of them fairly low volume with large bid/ask spreads? I've read that's the case for the smaller gold contracts, for example. I think I'd feel more comfortable staying with the very liquid full size contracts, filling in where needed with the VTI, GLD, and TLT ETFs.

I've been playing with a paper trading account at IB. Using ES, GS, and UB futures contracts with VTI, GLD, and TLT ETFs for balancing does work quite well but since the futures contracts are large it requires a fair amount of capital to guard against a draw down inducing a margin call.

[edit]
I suppose one could be long the full size contracts and short the high liquid ETFs in order to reduce overall portfolio size. That would avoid the less liquid gold mini/micro contract.

For Gold, each e-micro MGC contract is like $13k, about 1/10th of GC normal gold futures contracts. So that makes for very granular control of gold.

For Stocks, if you went long ES ($104k/contract today) and short NQ ($89k/contract today), that leaves you net about 14k increments for you to play with. You could do it with the Dow or Russell instead for different combinations.

For long bonds, UB is about $192k/contract today and you could short ZB which is $172k/contract today, netting out to 20k increments to play with.

If you've got portfolio margin, I'm assuming that hedging in this fashion wouldn't reduce your buying power by much. Or even if you didn't you'd wind up with something like 10x leverage instead of 20x, but I'm too lazy to do the math.

Yes, MGC is less liquid compared to GC, but I don't really think it's material because you're not day trading. You're holding for weeks or months. At that time scale, any spreads you pay can be thought of as the cost of leverage, which is still really cheap leverage. It's like sub 1% spread, so it's relatively high vs. ES, but it's still pretty negligible in terms of margin cost for such a huge amount of leverage.