Rebalancing Hypothesis

General Discussion on the Permanent Portfolio Strategy

Moderator: Global Moderator

Post Reply
Jack Jones
Executive Member
Executive Member
Posts: 522
Joined: Mon Aug 24, 2015 3:12 pm

Rebalancing Hypothesis

Post by Jack Jones » Wed Apr 12, 2017 3:30 pm

If we accept that rebalancing locks in some gains due to momentum or volatility, perhaps we should reconsider the standard advice of setting all assets back to 25%. Has anyone backtested a strategy where you only rebalance the assets that are out of the bounds? Rebalancing long bonds from 20% to 25% just because the stock market is on a tear (> 35%) seems somewhat arbitrary.

Basically, I'm thinking of a strategy where you rebalance as little as possible, while still staying within the bands. This has the added benefit of reducing transaction costs and taxes. Also, as a mostly uninformed investor, I believe making as few moves as possible to be closer to optimal. Sort of like, when you walk into a casino, the optimal number of games to play is as few as you can manage.
Kbg
Executive Member
Executive Member
Posts: 2815
Joined: Fri May 23, 2014 4:18 pm

Re: Rebalancing Hypothesis

Post by Kbg » Thu Apr 13, 2017 9:46 pm

J-J,

Vanguard has an excellent study on portfolio rebalancing effects. Well worth the read as the general principles discussed are applicable to any portfolio. Highly recommend a google/read of it.

I've posted this numerous times...backtesting can help one understand but it can not predict. And I will boldly say anyone who tells you that rebalancing will do X or Y is lying or clueless. What we can say safely and truthfully is the following:

1. Rebalancing's performance efficacy is solely path dependent meaning...

- Strong trends will be performance punished if one rebalances
- The opposite is true if the market is choppy and or has good mean reversion moves from time to time

2. Rebalancing impacts performance negatively due to trading costs and taxes. The former isn't such a big deal anymore, the latter can be huge toward the bottom line if bullet 2 above isn't helping out sufficiently.

3. Rebalancing keeps you close to the portfolio's expected risk profile...you can't expect to allow, for example, extreme drift in the PP's asset mix and expect the same smooth ride.

If you go the VP section and look up the 20% returns thread, I've written about rebalancing and its effects in what relative to a PP would be an extreme scenario. This isn't an endorsement of what I'm doing/trading/investing in on that thread but it will make crystal clear with actual results the impact rebalancing has. With the PP the impact is going to be more subtle, over there it kinda shouts loudly into your face...in short a more concentrated/visible example of the principles noted above.
User avatar
mathjak107
Executive Member
Executive Member
Posts: 4456
Joined: Fri Jun 19, 2015 2:54 am
Location: bayside queens ny
Contact:

Re: Rebalancing Hypothesis

Post by mathjak107 » Fri Apr 14, 2017 3:08 am

excellent article about rebalancing by kitces

he found it generally reduces returns but helps keep the risk profile you want . it all depends what your goal is you hope to achieve from investing .

https://www.kitces.com/blog/how-rebalan ... nt-anyway/
Pet Hog
Executive Member
Executive Member
Posts: 257
Joined: Tue May 28, 2013 4:08 pm

Re: Rebalancing Hypothesis

Post by Pet Hog » Fri Apr 14, 2017 5:53 am

Jack, I don't understand what you mean by only rebalancing assets that are beyond their bands. In your stocks 35%/bonds 20% example, would you sell stocks down to 25% and leave the rest as cash? What if the original allocation was 35% stocks, 25% cash, 20% bonds, and 20% gold? You'd rebalance to 25/35/20/20? Then have to rebalance cash into ... bonds and gold? In the end it's a typical PP rebalance to 25/25/25/25!
Jack Jones
Executive Member
Executive Member
Posts: 522
Joined: Mon Aug 24, 2015 3:12 pm

Re: Rebalancing Hypothesis

Post by Jack Jones » Fri Apr 14, 2017 7:18 am

Pet Hog wrote:Jack, I don't understand what you mean by only rebalancing assets that are beyond their bands. In your stocks 35%/bonds 20% example, would you sell stocks down to 25% and leave the rest as cash? What if the original allocation was 35% stocks, 25% cash, 20% bonds, and 20% gold? You'd rebalance to 25/35/20/20? Then have to rebalance cash into ... bonds and gold? In the end it's a typical PP rebalance to 25/25/25/25!
Yeah, in that case you have to put it somewhere, so yes, it would be split between bonds and gold. I'm thinking more of scenarios like this: 35/15/20/30.
Kbg
Executive Member
Executive Member
Posts: 2815
Joined: Fri May 23, 2014 4:18 pm

Re: Rebalancing Hypothesis

Post by Kbg » Fri Apr 14, 2017 7:42 am

Same principles I mentioned above always apply...so in this case the following:

If you ride a good momentum asset then that will increase your returns while increasing risk should that asset mean return, and your risk profile will have changed.

I'm not understanding why you wouldn't allocate back to whatever portfolio mix you had chosen though (if I'm understanding you are not balancing to your original allocation...if I misread disregard). If you don't then it's just a "new" portfolio with whatever risk characteristics your new allocation levels have inherently.

In some aggressive trading systems a rebalance to but never from cash method is used, but the point here is the person is allocating a specific amount to a volatile system and that system must prove itself by not blowing and adding to its cash allocation if the system goes bad or hits a rough patch. Usually once the cash position has been doubled from the original amount invested then it is removed as the system is now "risk" free as measured against the original amount.
bedraggled
Executive Member
Executive Member
Posts: 705
Joined: Sat Sep 13, 2014 4:20 am

Re: Rebalancing Hypothesis

Post by bedraggled » Fri Apr 14, 2017 5:25 pm

Do the 40/10 rebalancing bands allow an asset to run favorably?

Also, in rebalancing less, I would think there are fewer taxable events.

The CAGR on peaktotrough.com looks pretty good when reviewing the 4x25 HBPP.

Do I present a reasonable case? Am I missing something?
Last edited by bedraggled on Fri Apr 14, 2017 9:12 pm, edited 2 times in total.
Kbg
Executive Member
Executive Member
Posts: 2815
Joined: Fri May 23, 2014 4:18 pm

Re: Rebalancing Hypothesis

Post by Kbg » Fri Apr 14, 2017 8:35 pm

Try here...do a band, annual and no rebalance. 10/40 would be somewhere between the band rebalance and no rebalance. Look at max dd, worse year and CAGR. If you are up on portfolio stats, there is more advance info.

https://www.portfoliovisualizer.com/bac ... sisResults
User avatar
Hal
Executive Member
Executive Member
Posts: 1349
Joined: Tue May 03, 2011 1:50 am

Re: Rebalancing Hypothesis

Post by Hal » Sat Apr 15, 2017 5:26 am

I think it is equally important to decide when you will "NOT" rebalance...

If a country's credit rating is not good would you really rebalance and buy more of their bonds?

In the attached link they talk about how Italy's 90 day bills were converted into 10 year bonds without the bond holders consent

Just something to think about . Would be interested to know what criteria other forum members would use to halt reblalancing?

https://www.youtube.com/watch?v=rWjL1WcD8Do
barrett
Executive Member
Executive Member
Posts: 1982
Joined: Sat Jan 04, 2014 2:54 pm

Re: Rebalancing Hypothesis

Post by barrett » Sat Apr 15, 2017 8:40 am

Hal wrote: In the attached link they talk about how Italy's 90 day bills were converted into 10 year bonds without the bond holders consent
Looks like the link you posted is about philosophy. Only watched the first 30 seconds so maybe the guy does a major pivot and starts talking about sovereign debt!

I've been yammering on about how I don't believe wide bands can currently capture the rebalancing value of LTTs with rates being this low. Yes, it could happen mathematically but it's extremely unlikely with, say, 35/15 bands. I can easily see rebalances happening due to stocks or gold running up to 35%, but not with bonds.

Any thoughts on this? Seems to me that low interest rates have been so good for stocks that they tend drive stock prices up at the same time as bond prices.

It bothers me to hold an asset when is has a low expected medium-term return unless I am periodically taking some gains off the table. In practice what this has meant for me is a kind of bond market timing where I've been lucky to sell some bonds at short-term bottoms only to not be able to pull the trigger and buy back in when yields go back up. The result is that I am lowish on bonds (though still a bit more that 20%, I think).

Or does this belong in new PP Therapy thread?
Post Reply