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

A place to talk about speculative investing ideas for the optional Variable Portfolio

Moderator: Global Moderator

User avatar
Kbg
Executive Member
Executive Member
Posts: 1553
Joined: Fri May 23, 2014 4:18 pm

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

Post by Kbg » Wed Nov 16, 2016 2:26 pm

I debated rebalancing last night but I'm going to hold for a bit. We are in the sweet spot of the investments year right now for equity.

With regard to going to zero the reverse occurs as well. I have some lots up 150%ish with a hold measured in months. Just keep an eye on the account balance not the position balances. Day to day big picture the ride is pretty mellow I think considering what is under the hood.
jay
Junior Member
Junior Member
Posts: 22
Joined: Fri Nov 11, 2011 10:53 am

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

Post by jay » Sat Nov 26, 2016 6:23 pm

this is a very interesting thread, but too long to read every post so I apologize if this question is already addressed.

What is a good rebalancing strategy for different flavours of this leveraged combo of Cash/TMF/SPXL/UGLD? like

75/8.33/8.33/8.33
50/16.67/16.67/16.67
25/25/25/25
0/33.33/33.33/33

With 3X ETFs I suppose one cannot wait too long before rebalancing due to time decay. It seems to me that bands and/or quarterly might be a good option. But what bands should one use? A good band for 50/16.67/16.67/16.67 might be 7.5% but not sure.

One can also keep track of the corresponding paper 1X (expanded) version and balance using 35/15 bands and use that as a model for rebalancing. For instance, if one has 100K in 50/16.67/16.67/16.67, that is equivalent to having 200K in a traditional PP. Restore the weights to their targets when the model indicates so. Not sure if this would work as it can be 2-3 years without rebalancing. Might work well in a trending market but not so much in a side market.

How do you guys re-balance?
User avatar
Kbg
Executive Member
Executive Member
Posts: 1553
Joined: Fri May 23, 2014 4:18 pm

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

Post by Kbg » Sat Nov 26, 2016 11:25 pm

I don't get too spun up about rebalancing...though for 3xETFs doing it is critical. I do somewhere between +\- 33-50% of the target (e.g. for 16.67 @ 11.11 or 8.33 and etc. on the upside). I'm more a believer in bands than time hacks for rebalancing. My basic method is to hit a band and then apply judgement as to when to actually pull the trigger. I'm there on stocks but plan on waiting for either the year end due to stock market seasonal positive tendencies and/or LTTs to stabilize/base. If stocks stall I will pull the trigger immediately since I'm right below the high side threshold.

In the perfect world you rebalance at a trend reversal point...but who can do that?

Just pick something you will stick with and remember rebalancing is about risk NOT performance.
jay
Junior Member
Junior Member
Posts: 22
Joined: Fri Nov 11, 2011 10:53 am

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

Post by jay » Mon Nov 28, 2016 11:58 am

With a BTB (by the book) conventional PP, Harry's recommendation was based on a 10% band. An example would be one asset moving +40%, another losing 4%, and the other 2 remaining flat. Resulting portfolio would be 25/25/35/15. This happens once every 2-3 years or so.

I total agree that for a such a conventional PP, one should not be too picky with their rebalancing strategy, as long as they stick to it. One can use a 5% band, 10% band, 15% band, yearly or even *never rebalance* and still do comparably OK.

However, I think that rebalancing is very critical to the success of a strategy that uses 3X leveraged funds. If I were to deploy such a strategy, I would use bands for the most part. But I would also make it a rule to rebalance once / year (if no band was hit that is) so as to not let the 3X drift too much. A 3X ETF moves +50% or -50% fairly frequently (once in less than a year I would say) so I would use a band that corresponds with such a move.
User avatar
Kbg
Executive Member
Executive Member
Posts: 1553
Joined: Fri May 23, 2014 4:18 pm

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

Post by Kbg » Mon Nov 28, 2016 9:10 pm

I'm not so sure about end of year if one is rebalancing with bands. Seems like extra cost for no real solid reason to me, particularly if there isn't much of a spread between elements of the port. If one is kinda close to a band rebalance, sure why not if it helps you sleep better.
jay
Junior Member
Junior Member
Posts: 22
Joined: Fri Nov 11, 2011 10:53 am

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

Post by jay » Tue Nov 29, 2016 11:45 am

@Kbg, do you mind sharing what combo you run and how long you have been doing it?

I personally use a no cash PP (retirement account, no cash needed for a while). I am currently a third in each of TLT, SPY and GLD. Given PP's poor performance in the last 3 and 5 years, I am expecting better returns in the next few years and hence my interest in leverage, such as as a 50/16.67/16.67/16.67 Cash/TMF/SPXL/UGLD. That would be equivalent to a 2X conventional PP and its CAGR since Oct 17, 2011 be around +5%. Not too bad given that a traditional PP would've had a ~3.5% CAGR.

What in your opinion and experience should one be concerned about when implementing such a strategy?
dragoncar
Executive Member
Executive Member
Posts: 1111
Joined: Wed Aug 10, 2011 7:23 pm

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

Post by dragoncar » Tue Nov 29, 2016 2:45 pm

Pulled the trigger on a rebalance. My 3xSPY was bouncing around 34-35% of exposure for a while. I was way up for a while, but now back to square one (very modest gain).
User avatar
Cortopassi
Executive Member
Executive Member
Posts: 1834
Joined: Mon Feb 24, 2014 2:28 pm
Location: Illinois

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

Post by Cortopassi » Tue Nov 29, 2016 3:05 pm

dragoncar, I am curious, so the 3x SPY about doubled in the timeframe, how did the 3x gold and bond ETFs do?
dragoncar
Executive Member
Executive Member
Posts: 1111
Joined: Wed Aug 10, 2011 7:23 pm

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

Post by dragoncar » Tue Nov 29, 2016 10:43 pm

Cortopassi wrote:dragoncar, I am curious, so the 3x SPY about doubled in the timeframe, how did the 3x gold and bond ETFs do?
These numbers will be a bit off since I already rebalanced, but:

SPXL +39.07%
TMF -10.52%
UGLD -12%

All acquired 03/01/2016


(so the above would make SPXL 33.4% exposure, but as I said it was bouncing around and probably a bit higher intradray)

If I'm doing something wrong, let me know. I basically started with 50% cash and 1/6 of the 3x ETFs
User avatar
Kbg
Executive Member
Executive Member
Posts: 1553
Joined: Fri May 23, 2014 4:18 pm

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

Post by Kbg » Tue Nov 29, 2016 11:02 pm

jay wrote:@Kbg, do you mind sharing what combo you run and how long you have been doing it?

I personally use a no cash PP (retirement account, no cash needed for a while). I am currently a third in each of TLT, SPY and GLD. Given PP's poor performance in the last 3 and 5 years, I am expecting better returns in the next few years and hence my interest in leverage, such as as a 50/16.67/16.67/16.67 Cash/TMF/SPXL/UGLD. That would be equivalent to a 2X conventional PP and its CAGR since Oct 17, 2011 be around +5%. Not too bad given that a traditional PP would've had a ~3.5% CAGR.

What in your opinion and experience should one be concerned about when implementing such a strategy?
My specific mix is: 45% SHY, 16.67 SPXL/TMF, 16.66 UGLD, 5% XIV

Total Rtn 28.9%
2014: 11.98%
2015: -9.99%
2016 YTD: 25.61

These figures are net of all expenses. A pristine backtest from 1/1/14 to today annually rebalanced is 7.13% CAGR/-18.25% Max DD. I have no idea what my actual Max DD is. I could figure it out, but I don't want to expend the energy. This compares to a backtested regular PP of 4.34/-6.93 during the same time period using SHY, SPY, TLT and GLD. Running the port since first possible (2012) the stats are 8.37/-18.25 vs. 3.28/-7.95. I've posted this as well...please note the impact of path dependency. From 2014 we have not achieved 2x leverage. Since 2012 we have achieved 2.55x leverage on the upside and 2.29 on the downside. And in 2016 we are pretty close to right at 2x on both sides. My "use bands with judgement approach" has outperformed the straight backtest: 28.9% vs. 22.2%.

I would say the below are major concerns.

1. The future is not like the past though the past has been pretty crazy and the PP has held up as advertised.

2. Getting slaughtered by a big multi-asset drawdown while being leveraged up.

3. #2 happens and I bail.

4. Getting too "predictive" and messing with the PP portfolio mix in a dangerous way. Categorically, I do not recommend increasing leverage beyond 16.67% ish. If you can't stand an asset reduce it or turn it into whatever you are using for cash.

5. Long lasting positive correlation between bonds and stocks (2-3 years or so).

6. I have mentioned repeatedly that if one can not ignore a single asset's performance then they absolutely should in no way invest in a 3x ETF leveraged PP. Assuming an annual rebalance, the history is clear that one will experience an asset going to zero not too infrequently. I don't have this problem personally as I'm able to keep my eyes focused at the portfolio level. Others may find this hard to do emotionally and should not underestimate their ability to deal with it.

My personal expectations are stock like returns with less "major event" volatility and pretty close to 2x of whatever the standard PP will give us going forward. Since 2012 we have definitely been kicked in the junk compared to 100% stocks and we are pretty close to a 50/50 stocks/cash mix but with triple the drawdown. If anyone has paid attention, basically the way a PP works is it gets stomped during a good bull market for stocks and then makes it up during crisis and/or a bear market. I don't expect anything different going forward. If your starting point is 2012 this portfolio sucks. If your starting point is 2008 it is looking pretty good....slightly better performance half the drawdown of an all stock buy and hold.

To repeat this is an aggressive growth portfolio, not grandma's stable retirement withdrawal PP...and of course, all the above is for educational purposes only and folks should seek the assistance of a professional for any real advice they may need.
jay
Junior Member
Junior Member
Posts: 22
Joined: Fri Nov 11, 2011 10:53 am

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

Post by jay » Wed Nov 30, 2016 2:23 pm

Thanks for your reply Kbg. Good discussion.

I am leaning towards a 50/16.67/16.67/16.66 version and will later consider XIV if/when a correction in equity markets occurs.

Going back 10 years, and including today's drop, the current DD of the conventional PP is ~7.5%. There are very few (5-6) drawdowns (depending on starting time and rebalancing options) that are 7.5% in the last 20 years. The PP is definitely in oversold territory unless a repeat of the ~15% drawdown we had in 2008 is around the corner. Still, the odds are in one's favour to be long the PP or a leveraged version of it.

Let's see what the six months will bring.
User avatar
Kbg
Executive Member
Executive Member
Posts: 1553
Joined: Fri May 23, 2014 4:18 pm

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

Post by Kbg » Sun Dec 04, 2016 2:33 pm

Jay,

It is good you are looking at the numbers hard. I definitely recommend folks run 3x simulations which assuming you have the data is easy to do in Excel. You get a good feel for how bad the portfolio can behave as well as individual elements of it. One also understands very very quickly why it is important to have a big wad of cash. It is imperative to be able to recharge and take advantage of mean reversion.
Post Reply