20% annual returns over 40 years...interested?

[Kbg]

I've been wanting to do this for a while...so here we go, a 3xPP extended back to 2005. Excel is a 1x simulation taken from daily percentage change values using Norgate data down to the 5th decimal point. Excel 3x is the same only multiplying daily returns by 3 to simulate a 3x ETF. Norgate is a backtest of GLD, SPY and TLT at 33.3333% each with annual holds. You will see there is some variation between Excel and Norgate, but I think this is close enough to verify my simulation was done properly. I also summed up the returns, deleveraged the 3x returns and got 82% vs. 79% meaning that over this time period you got 3% (x3 = 9%) less than a full 3x daily return. 9% over seven years is pretty good. Very specifically, perfect tracking would have given you 247% and the return was 238%.




.
Year     Excel Excel 3x Norgate
2005 11.20% 33.87% 12.30%
2006 13.03% 35.19% 13%
2007 15.30% 46.67% 15.30%
2008 0.69%   6.48%   0.60%
2009 9.52% 24.61%   8.40%
2010 17.78% 52.25% 17.90%
2011 15.15% 39.55% 15.20%

[Kbg]

Kbg, isn't it true that if the asset classes are keeping to a trend, then the (internally daily adjusted) 3xETFs + cash (so deleveraged to be back to 1x but not rebalanced every day) will out perform the 1xETF whilst if the asset classes are whipsawing, then the 3xETF + cash will suffer?

Was mid 2011 a time when the 3xETF + cash strategy would have performed badly due to all of the assets whipsawing to and fro with no consistent trend?

Sorry if I'm in a muddle / being stupid etc -I haven't followed this thread.

Unbeknownst to me I was working on an answer to your question with actual data. 

But if I read your question correctly, the answer is yes. A trend will outpeform and chop will under perform. The magnitude is simply unknown and unknowable because we have absolutely no idea what each day's return is going to be into the future. I don't know a person on the planet who could tell you with exact precision what tomorrow's return on the S&P 500 is going to be...now to know for certain for the PP here is what must be accomplished;  for three assets you have to know the exact return each and every day for however long you want to peak into the future. Thus, we can only guess using history and we know we are going to be really happy with a great strong long term trend and really unhappy with chop. However, historically the PP has enabled you "to hook at least one good ride" a year.

[Kbg]

I've been wanting to do this for a while...so here we go, a 3xPP extended back to 2005. Excel is a 1x simulation taken from daily percentage change values using Norgate data down to the 5th decimal point. Excel 3x is the same only multiplying daily returns by 3 to simulate a 3x ETF. Norgate is a backtest of GLD, SPY and TLT at 33.3333% each with annual holds. You will see there is some variation between Excel and Norgate, but I think this is close enough to verify my simulation was done properly. I also summed up the returns, deleveraged the 3x returns and got 82% vs. 79% meaning that over this time period you got 3% (x3 = 9%) less than a full 3x daily return. 9% over seven years is pretty good. Very specifically, perfect tracking would have given you 247% and the return was 238%.




.
Year     Excel Excel 3x Norgate
2005 11.20% 33.87% 12.30%
2006 13.03% 35.19% 13%
2007 15.30% 46.67% 15.30%
2008 0.69%   6.48%   0.60%
2009 9.52% 24.61%   8.40%
2010 17.78% 52.25% 17.90%
2011 15.15% 39.55% 15.20%

And to extend the above using UGLD, SPXL and TMF (33.3333% each)

2012 15.2%
2013  1.2%
2014 38.3%
2015 16.2% (to date)

Max DD 26.2%

Looks like a nice little portfolio for the aggressive part of your VP. Since 2005 it has absolutely trashed a stock index portfolio with probably less volatility (I could do a max DD for 2005-2011 but that would be really painful. I may eventually but not today.)

[stone]

Kbg, I'm especially impressed at how the whipsaw chop of 2011 didn't do anything worse than simply make the 3x version less than 3x the 1xETF version (and so only "disapointing" if a better than 3x the 1xETF result was envisioned).

But, do (would) the 3xETFs  in reality manage to track anything like as well as your pre-2012 simulations when it really matters? I guess they have internal trading expenses and such like that are very period specific (having to buy and sell a lot during choppy markets).

[Kbg]

Stone,

Read my swap references. If you look at these ETFs they are almost all based on swaps and I'm pretty sure the three listed in the backtests are exclusively swaps. Easy enough to check via the prospectus. From memory I think the expense rate on them is .95% a year.

I'm not saying expenses aren't important but I'd probably put them at the bottom of my worries for a 3xPP. It's not like we are worried about .15% vs. .13% on two STT funds that have slightly different durations and are returning around 1%. In the latter case the relative impact is proportionately large...not so much here.

If you want to worry about something, worry about a correlated move down by all three assets.

[jason]

Another good one...this one gets into the mechanics a little better. I don't think I would invest in an ETF that did swaps for obscure stuff, but I'm not particularly concerned with swaps on US LTTs, the S&P500 or gold.

http://www.horizonsetfs.com/Pdf/Educati ... d_ETFs.pdf

More 3x ETF performance vs. underlying

http://seekingalpha.com/article/2854476 ... -past-year

Given how the 3x ETFs hold a bunch of swaps, doesn't that create tremendous counter-party risk?  Keep in mind that we are not talking about the counter-party risk involved with Vanguard or Fidelity holding your VTI shares in their name.  Vanguard and Fidelity could go bankrupt, and there is still a very good chance nothing bad will happen to your VTI holdings.
Here are the top holdings for UPRO 3x S&P ETF:
S&P 500 Index Swap Citibank, N.A. N/A 61.39
S&P 500 Index Swap Morgan Stanley & Co. International Plc N/A 36.65
S&P 500 Index Swap Deutsche Bank Ag N/A 33.42
S&P 500 Index Swap Societe Generale N/A 31.71
S&P 500 Index Swap Ubs Ag N/A 23.78
S&P 500 Index Swap Credit Suisse International N/A 17.40
S&P 500 Index Swap Goldman Sachs International N/A 9.36
E-mini S&P 500 Index Future Mar15 N/A 6.77
S&P 500 Index Swap Bank Of America, Na N/A 6.18
Apple Inc. AAPL 2.51

Now, if Morgan Stanley, Deutsche Bank, UBS, or one of the other banks backing up the swaps goes bankrupt, a fat chunk of your money is going to disappear overnight, right?  And in a SHTF economic collapse scenario where many big banks go bankrupt, you could lose a huge portion of your entire 3x ETF investment, right?  If so, this appears to be a very big counter-party risk.
Instead of doing 3x ETFs, why not do a regular PP with 3x leverage?  Margin rates at Interactive Brokers are extremely low, close to 1%.  Check out this chart:
http://investorjunkie.com/12389/best-margin-rates/
What are the risks involved with investing on margin (only a little bit of margin, all in the VP), and how do those risks compare to the counter-party risk involved with 3x ETFs that use swaps?

[Kbg]

Ok folks, this is the last time I answer a question on any of this stuff unless someone posts a new question I haven't thought of at which point it becomes beneficial to myself (possibly). Let's just say I get a bit annoyed when I detect people are using me as their research assistant and probably not reading what I have posted for them to read (or not understanding it which is a completely different problem). However, since I didn't actually post this link I will do this one last post. Read the entire thing, carefully. I particularly recommend pages 10 to the end of the document. Also be aware of the fact that US law is still being hashed out as Dodd-Frank is being implemented, but it appears swaps are pretty much going to have margin requirements like futures. How much risk the ETF has is completely addressed in the publication...

https://pressroom.vanguard.com/content/ ... c_ETFs.pdf

And finally I am most certainly not a lawyer and we are into questions that only a competent and qualified securities lawyer could answer. I recommend you seek legal advice if of sufficient concern. I'm invoking the "investing education" website caveat here formally and suggest anyone who needs additional information seek professional legal and tax advice nor should anything I've posted be considered as investing advice and all that has been posted was for education purposes only.

Jason,

With regard to futures or margin vs. leveraged ETFs, not answering that question. You have all you need to figure it out for yourself. I suggest some additional reading and some Excel work. And if you aren't understanding what I'm posting then please take HB's sage advice I posted earlier.

[Kbg]

Back to our regularly scheduled programming.

Jan performance

3xPP: 9.36/-2.85dd
1xPP: 3.22/-.86dd

3x w/ XIV Twist: 4.42/-4.09dd

[ozzy]

Kbg, how did you calculate those January performance numbers? 

ETFreplay.com shows the following returns for January 2015:

(1x) PP: TLT/GLD/SHY/VTI = +4.15%
(3x) PP: UPRO/UGLD/TMF = +16.1%

[Kbg]

Amibroker with Northgate data. The replay numbers look high to me, but I haven't hand checked.

[Kbg]

A couple of things.

I'm getting 16.57 now and 16.68 using Yahoo data.  Two issues with my original post.

- I was using close to close...open to close closes the gap

- Also my posts include 25% cash which accounts for most of the difference...the 16% numbers are 33% each

[dragoncar]

Another good one...this one gets into the mechanics a little better. I don't think I would invest in an ETF that did swaps for obscure stuff, but I'm not particularly concerned with swaps on US LTTs, the S&P500 or gold.

http://www.horizonsetfs.com/Pdf/Educati ... d_ETFs.pdf

More 3x ETF performance vs. underlying

http://seekingalpha.com/article/2854476 ... -past-year

Given how the 3x ETFs hold a bunch of swaps, doesn't that create tremendous counter-party risk?  Keep in mind that we are not talking about the counter-party risk involved with Vanguard or Fidelity holding your VTI shares in their name.  Vanguard and Fidelity could go bankrupt, and there is still a very good chance nothing bad will happen to your VTI holdings.
Here are the top holdings for UPRO 3x S&P ETF:
S&P 500 Index Swap Citibank, N.A. N/A 61.39
S&P 500 Index Swap Morgan Stanley & Co. International Plc N/A 36.65
S&P 500 Index Swap Deutsche Bank Ag N/A 33.42
S&P 500 Index Swap Societe Generale N/A 31.71
S&P 500 Index Swap Ubs Ag N/A 23.78
S&P 500 Index Swap Credit Suisse International N/A 17.40
S&P 500 Index Swap Goldman Sachs International N/A 9.36
E-mini S&P 500 Index Future Mar15 N/A 6.77
S&P 500 Index Swap Bank Of America, Na N/A 6.18
Apple Inc. AAPL 2.51

Now, if Morgan Stanley, Deutsche Bank, UBS, or one of the other banks backing up the swaps goes bankrupt, a fat chunk of your money is going to disappear overnight, right?  And in a SHTF economic collapse scenario where many big banks go bankrupt, you could lose a huge portion of your entire 3x ETF investment, right?  If so, this appears to be a very big counter-party risk.
Instead of doing 3x ETFs, why not do a regular PP with 3x leverage?  Margin rates at Interactive Brokers are extremely low, close to 1%.  Check out this chart:
http://investorjunkie.com/12389/best-margin-rates/
What are the risks involved with investing on margin (only a little bit of margin, all in the VP), and how do those risks compare to the counter-party risk involved with 3x ETFs that use swaps?

Most people don't use their VP to protect against SHTF.  This approach definitely belongs in your VP. 

[Kbg]

Most people don't use their VP to protect against SHTF.  This approach definitely belongs in your VP.

Indeed, I agree. I will however throw out a couple of thoughts. Gold, US-Ts and the S&P 500 are about as bedrock as you can get for financial instruments. Swaps on them, short of Armageddon, are likely not a lot more risky. If Doomsday comes, I think you are much better served owning your own house, land, food and guns (if so inclined) than any of the other three in real or swap form. This place gets very inconsistent when it comes to this stuff in my view. You either think the US financial system is going to hold or you do not. And if you do not, why would you not plow your life's work into actual tangible things with no debt obligations? This thread has discussed a lot of the finer points of going with 3xETFs and frankly I've learned a ton just due to questions I'd never really put a whole lot of thought to.

However, all said and done, if anyone is remotely considering this you really need to focus on the most dangerous and the most obvious issue...being on the wrong side of 3x leverage. It is entirely possible, in fact likely, that during the course of a year one of the assets could go down to just about zero. If you don't have a corresponding zig to counterbalance the zag you will be wearing a hurt monkey. In my view the rest are minor and pale in comparison to the most dangerous and obvious one.

[jason]

Another good one...this one gets into the mechanics a little better. I don't think I would invest in an ETF that did swaps for obscure stuff, but I'm not particularly concerned with swaps on US LTTs, the S&P500 or gold.

http://www.horizonsetfs.com/Pdf/Educati ... d_ETFs.pdf

More 3x ETF performance vs. underlying

http://seekingalpha.com/article/2854476 ... -past-year

Given how the 3x ETFs hold a bunch of swaps, doesn't that create tremendous counter-party risk?  Keep in mind that we are not talking about the counter-party risk involved with Vanguard or Fidelity holding your VTI shares in their name.  Vanguard and Fidelity could go bankrupt, and there is still a very good chance nothing bad will happen to your VTI holdings.
Here are the top holdings for UPRO 3x S&P ETF:
S&P 500 Index Swap Citibank, N.A. N/A 61.39
S&P 500 Index Swap Morgan Stanley & Co. International Plc N/A 36.65
S&P 500 Index Swap Deutsche Bank Ag N/A 33.42
S&P 500 Index Swap Societe Generale N/A 31.71
S&P 500 Index Swap Ubs Ag N/A 23.78
S&P 500 Index Swap Credit Suisse International N/A 17.40
S&P 500 Index Swap Goldman Sachs International N/A 9.36
E-mini S&P 500 Index Future Mar15 N/A 6.77
S&P 500 Index Swap Bank Of America, Na N/A 6.18
Apple Inc. AAPL 2.51

Now, if Morgan Stanley, Deutsche Bank, UBS, or one of the other banks backing up the swaps goes bankrupt, a fat chunk of your money is going to disappear overnight, right?  And in a SHTF economic collapse scenario where many big banks go bankrupt, you could lose a huge portion of your entire 3x ETF investment, right?  If so, this appears to be a very big counter-party risk.
Instead of doing 3x ETFs, why not do a regular PP with 3x leverage?  Margin rates at Interactive Brokers are extremely low, close to 1%.  Check out this chart:
http://investorjunkie.com/12389/best-margin-rates/
What are the risks involved with investing on margin (only a little bit of margin, all in the VP), and how do those risks compare to the counter-party risk involved with 3x ETFs that use swaps?

Most people don't use their VP to protect against SHTF.  This approach definitely belongs in your VP.

This is for my VP, but I still wanted to be protected if the SHTF, so I figured the 3x gold would save the day. My thinking is, if I trust the PP enough to go up over time that I am putting my life savings into it, then I should trust a 3x PP with 2% of my money to go up over time, as well. Of course, there is the never know factor. If the PP has a historically bad year and drops 25%, it would be ugly for the 3x PP.
I read the article that KBG posted and it does seem like the counterparty risk is reasonably low for UPRO and TMF because they are collateralized, but UGLD, on the other hand, is an ETN, not an ETF, so if the issuer goes bust, so does the ETN, apparently. Is UGLD the only option for 3x gold?  I can't find any others that actually track the price of gold.

[dragoncar]

Given how the 3x ETFs hold a bunch of swaps, doesn't that create tremendous counter-party risk?  Keep in mind that we are not talking about the counter-party risk involved with Vanguard or Fidelity holding your VTI shares in their name.  Vanguard and Fidelity could go bankrupt, and there is still a very good chance nothing bad will happen to your VTI holdings.
Here are the top holdings for UPRO 3x S&P ETF:
S&P 500 Index Swap Citibank, N.A. N/A 61.39
S&P 500 Index Swap Morgan Stanley & Co. International Plc N/A 36.65
S&P 500 Index Swap Deutsche Bank Ag N/A 33.42
S&P 500 Index Swap Societe Generale N/A 31.71
S&P 500 Index Swap Ubs Ag N/A 23.78
S&P 500 Index Swap Credit Suisse International N/A 17.40
S&P 500 Index Swap Goldman Sachs International N/A 9.36
E-mini S&P 500 Index Future Mar15 N/A 6.77
S&P 500 Index Swap Bank Of America, Na N/A 6.18
Apple Inc. AAPL 2.51

Now, if Morgan Stanley, Deutsche Bank, UBS, or one of the other banks backing up the swaps goes bankrupt, a fat chunk of your money is going to disappear overnight, right?  And in a SHTF economic collapse scenario where many big banks go bankrupt, you could lose a huge portion of your entire 3x ETF investment, right?  If so, this appears to be a very big counter-party risk.
Instead of doing 3x ETFs, why not do a regular PP with 3x leverage?  Margin rates at Interactive Brokers are extremely low, close to 1%.  Check out this chart:
http://investorjunkie.com/12389/best-margin-rates/
What are the risks involved with investing on margin (only a little bit of margin, all in the VP), and how do those risks compare to the counter-party risk involved with 3x ETFs that use swaps?

Most people don't use their VP to protect against SHTF.  This approach definitely belongs in your VP.

This is for my VP, but I still wanted to be protected if the SHTF, so I figured the 3x gold would save the day. My thinking is, if I trust the PP enough to go up over time that I am putting my life savings into it, then I should trust a 3x PP with 2% of my money to go up over time, as well. Of course, there is the never know factor. If the PP has a historically bad year and drops 25%, it would be ugly for the 3x PP.
I read the article that KBG posted and it does seem like the counterparty risk is reasonably low for UPRO and TMF because they are collateralized, but UGLD, on the other hand, is an ETN, not an ETF, so if the issuer goes bust, so does the ETN, apparently. Is UGLD the only option for 3x gold?  I can't find any others that actually track the price of gold.

We must have different definitions of SHTF because in mine, you certainly aren't getting anything out of a brokerage account.

[jason]

Most people don't use their VP to protect against SHTF.  This approach definitely belongs in your VP.

This is for my VP, but I still wanted to be protected if the SHTF, so I figured the 3x gold would save the day. My thinking is, if I trust the PP enough to go up over time that I am putting my life savings into it, then I should trust a 3x PP with 2% of my money to go up over time, as well. Of course, there is the never know factor. If the PP has a historically bad year and drops 25%, it would be ugly for the 3x PP.
I read the article that KBG posted and it does seem like the counterparty risk is reasonably low for UPRO and TMF because they are collateralized, but UGLD, on the other hand, is an ETN, not an ETF, so if the issuer goes bust, so does the ETN, apparently. Is UGLD the only option for 3x gold?  I can't find any others that actually track the price of gold.

We must have different definitions of SHTF because in mine, you certainly aren't getting anything out of a brokerage account.

LOL, true.  I meant severe debasement of the dollar/inflation and very poor economy (I guess like what happened to Iceland).  I agree that in a true SHTF scenario, there will likely be capital controls in place and brokerage accounts probably will be frozen.

[dragoncar]

A couple of things.

I'm getting 16.57 now and 16.68 using Yahoo data.  Two issues with my original post.

- I was using close to close...open to close closes the gap

- Also my posts include 25% cash which accounts for most of the difference...the 16% numbers are 33% each

When you say 25% cash, do you mean 25% of total exposure?  E.g.

16.5% 3x stock = 50% exposure
16.5% 3x gold = 50% exposure
16.5% 3x bond = 50% exposure
50% cash = 50% exposure

thus, equal exposure to each asset class?  Or are you doing:

25% 3x stock = 75% exposure
25% 3x gold = 75% exposure
25% 3x bond = 75% exposure
25% cash = 25% exposure

[Kbg]

The second is what I post, the first is what I do.

[dragoncar]

OK, folks, I'm on this 3x ride (50% cash) in my Roth as a VP.  Mark well this day, inevitably the secular top for the PP.

[Kbg]

Too late, I cursed it first :-)

[Kbg]

Through Feb performance

3xPP: 5.22/-9.12dd
1xPP: 2.02/-2.87dd

3x w/ XIV Twist: 4.48/-8.22dd

February wasn't as fun as January was.  A good example of bad decay as well.

[Kbg]

Through Feb performance

3xPP: 5.22/-9.12dd
1xPP: 2.02/-2.87dd

3x w/ XIV Twist: 4.48/-8.22dd

February wasn't as fun as January was.  A good example of bad decay as well.

A 16.7% SPXL/TMF, 16.6% UGLD, 50% SHY was 3.55/-6.38dd

Now here is the part I keep pounding on after two months of live 2015 returns...path dependency

Perfect 1.5x tracking would be a 3.03% return and 4.30dd; however, what we got was 1.75x leverage and 2.25x dd. The reason that happened was (using a 100K port) because we got to just shy  of 109K and ended up dropping below 102 before climbing back out a bit.

[Mark Leavy]

Just a quick thank you Kbg for your honest and insightful updates.

Reviewing, dissecting, and understanding your posts is an education in the fundamental economics of the permanent portfolio and of leveraged ETFs.

[Kbg]

Just a quick thank you Kbg for your honest and insightful updates.

Reviewing, dissecting, and understanding your posts is an education in the fundamental economics of the permanent portfolio and of leveraged ETFs.

Thank you for the kind words.

[Kbg]

Monthly/Annual UPRO Gains Map Almost Perfectly To The S&P 500 $UPRO

http://www.seekingalpha.com/article/2966666


A picture is worth a thousand words...the last graph perfectly illustrates what I've written about the nature of these 3x ETFs. Note upper right and lower left blue line...about as close to investing nirvana as it gets. Better when it is good and less bad when it is bad. However, the author points out the negative intercept (going nowhere is bad).