Asset dependent rebalancing bands

[LazyInvestor]

Based on some data I've seen in previous posts, it seems that the optimal rebalancing bands are 15/35. However, it seems also that bonds and cash are very rarely reaching these rebalance bands. So this got me thinking that maybe for equity and gold rebalancing bands should be 15/35 and for cash and bonds 20/30 or maybe even narrower? Maybe even some more refined bands for each asset?

[PP67]

I believe TennPaGa investigated this subject a while back and concluded that the 15/35 rebalancing bands (+/10% of 25%) were fairly optimal for the time period he investigated...

http://gyroscopicinvesting.com/forum/pe ... /#msg16933

[barrett]

Not sure if he is still active on this forum but Edsanville posted some fascinating rebalancing data that looked at everything from extremely narrow bands up to bands that were at least 40% (not sure about the low end... maybe 10%). What I got from those data is that if you factor out transaction costs and tax consequences (if any), it really doesn't matter all that much when you rebalance the PP. Much of my PP is in tax-deferred accounts so I don't worry about tax consequences too much.

I know that Craig and others have written about how an asset class can have a lot of momentum and that you can be limiting your gains by rebalancing too early. But there is the other side of the coin and that is that there are small gains to be harvested all the time. For example, gold punched higher a couple of times this year (up about 15% back in March if I am not mistaken) & stocks were on sale last week relative to today.

I am not advocating that people rebalance every two seconds but I do think that you can rebalance if you are thinking about something all the time. Dualstow mentioned in a recent stock post that he had been waiting for stocks to hit 35% of his PP but that they never quite made it that high. I know my personality (obsessive!) and I wouldn't want to spend a lot of time and energy just waiting for a band to be hit. Though 15/35 has been labelled as the "sweet spot", I think it is only barely so.

And let's not forget Kshartle's posts about there being no advantage to rebalancing... ever.

As always, feel free to shoot holes in my posts.

[bedraggled]

I believe Craigr mentioned the problem of paying taxes if the rebalance bands are tighter.  As I, a newcomer, have been counseled, the system has been around a long time.  The bands at 35/15 help us avoid capital gains taxes.  As I have thought while reading Craigr's book, seeing one of the 4 categories approach 35% then recede may create anxiety but it saves a person a tax bill.  Not bad.

I may ponder adjusting % in the near future but the 35/15 bands seem good.  And I am sure I will be dissuaded from playing with the %, anyway.

[bedraggled]

TennPaGa seems to have it correct about driving oneself crazy.  Aren't we supposed to be check the PP only occasionally?

After I read Craigr's book, I arranged a PP on paper and it appears to be up 5+% since !/2/14 with no tax bill yet.

[Xan]

Remember, the volatility capture aspect of the rebalancing bands is just a bonus, not the main reason for them.  (And that bonus may or may not even exist.)

The real reason for rebalancing is to keep from having too many or too few of your eggs in one basket.

[barrett]

The bands at 35/15 help us avoid capital gains taxes. 

I was specifically talking about rebalancing more often in tax-deferred accounts. Sorry if that wasn't clear.

As for how often we are "supposed" to check, well, a truly diversified portfolio gives us the option of not checking frequently. Some people actually enjoy incessantly looking at their balances.

My general opinion is that one should do whatever enables them to sleep at night.

I may have posted this before but...

A friend of mine cracked years ago "Everyone seems to be concerned about how I am sleeping. I'm sleeping fine but I'm losing my ass in the stock market!"

[bedraggled]

Barrett,

Rebalancing more often in a tax deferred account does clarify that.

Thanks.

[Mark Leavy]

Based on some data I've seen in previous posts, it seems that the optimal rebalancing bands are 15/35. However, it seems also that bonds and cash are very rarely reaching these rebalance bands. So this got me thinking that maybe for equity and gold rebalancing bands should be 15/35 and for cash and bonds 20/30 or maybe even narrower? Maybe even some more refined bands for each asset?

I like to think of the rebalancing bands in terms of Shannon's daemon and the Kelly Criterion.

In that scenario, each rebalancing event is similar to placing a new bet on the table.  If you lose, you lose half of your money.  If you win, you double your money.  Then you rebalance and try again.

Obviously, half and double are just examples and you can use different percentages - but I think it is useful to have the upper band be the inverse of the lower band.

The erstwhile and ever lucid poster, Melveyr, provided a good overview on one of his personal blog posts.
http://www.stableinvesting.com/2013/04/ ... demon.html

In truth, the rebalancing bands don't have that much effect.  In general, a bit wider is better, as the reduced transaction costs and taxable events tend to work in your favor.  But iff you are a bit nervous in carrying too much or too little of any one asset, then just rebalance.  No big deal.

Personally, I like the idea of thinking of each rebalance scenario as placing a new bet.  And then I wait and wait and wait until I find out if I have halved or doubled my money.  And then I rebalance and place another bet...  But then again, I am a big fan of Shannon.

[barrett]

Mark, Thanks for posting that. As a humorous side note, I met Claude Shannon a few times ages ago but it was in relation to one of his side hobbies & for years I didn't know who the heck he really was.

I often preface stuff that I post on here with the caveat that I am new to the PP. That being said, I find that using actual numbers to think about possible future scenarios is often helpful. For example when thinking about rebalancing using those wide 15/35 bands, picture a PP with a starting value of $400,000. Say stocks take a 40% hit and LTTs react nicely with a 20% move on the upside while gold and cash stay flat. You then have $60,000 in stocks, $100,000 each in cash and gold, and $120,000 in bonds. Stocks don't go down 40% all that often (say once a decade) so they would seem to be on sale at that point. But, using the above scenario, you'd still be at 15.8% stocks. I would personally be rebalancing at that point because I like buying stuff cheap when I can. It's the buy low-ish, sell high-ish philosophy.

Obviously I am just creating a scenario to make a point and any number of variations are possible.

[Mark Leavy]

As a humorous side note, I met Claude Shannon a few times ages ago but it was in relation to one of his side hobbies & for years I didn't know who the heck he really was.

Wow.  I am seriously jealous.  Were you hitting the roulette tables in Vegas?

What fun it would have been to know Shannon and Thorpe and Kelly in their heyday.  Running amok over everything the gambling and investment communities could ever imagine.

[blackomen]

Cash will hit rebalancing bands more often than you think if you make regular contributions..  or what I usually do is just buy the most lagging asset each time.