Rebalancing Band and Rebalancing Frequency

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
teacher
Posts: 988
Joined: Sun Oct 05, 2008 5:45 pm
Location: California

Rebalancing Band and Rebalancing Frequency

Post by teacher » Sat Sep 26, 2009 1:32 pm

Our rebalancing band is 10% on a 50/50 AA. This is in our rough draft IPS, but I am beginning to think that may be too restrictive. I know this is personal risk tolerance decision, but I wonder if 10% is considered tight by most on this forum.

Also, the longer one waits to rebalance, the more likely it will be that the dynamic allocation would bypass the band. Conversely, it seems the tighter the band, the more frequent rebalancing would need to occur. I am thinking in order to not cheat on the IPS, the band size and rebalance frequency needs to be decided in tandem, and I wonder what would be the best way to do that.

Any thoughts?

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Post by Rodc » Sat Sep 26, 2009 1:50 pm

No way to know.

I think at the stock vs bond level 5% absolute is more commonly used. That is hit 55/45 and rebalance.

The question is how big is the next bubble going to be? Pick a band that keeps you up to, but not over the peak. :)
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

dbr
Posts: 27207
Joined: Sun Mar 04, 2007 9:50 am

Post by dbr » Sat Sep 26, 2009 7:24 pm

In my understanding of rebalancing a person using bands does not have a frequency. More exactly the frequency is simply what is determined by when the bands indicate rebalancing.

Mention of dynamic allocation suggests you are applying two different rules to control asset allocation. Can you explain this in more detail?

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Post by Rodc » Sat Sep 26, 2009 7:31 pm

I missed the frequency AND bands, thinking you simply meant that tighter band would naturally trigger more often than looser bands.

As dbr notes generally one picks one or the other.

But, in part of my portfolio I pay a $75 fee to buy, nothing to sell. So I never make a buy in less than 6 months of the last buy. Completely arbitrary. A year might be even better too keep from buying falling knives.

I'm not sure why one would do this, but I don't think you'd find any real harm in rebalance if you hit a band or every one year, whichever comes first, or for that matter whichever comes last.

Rebalancing details just are not all that important.

Pick something, anything reasonable, and be done with it.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

EyeDee
Posts: 1270
Joined: Tue Feb 20, 2007 12:15 am

Rebalancing Band and Frequency

Post by EyeDee » Sat Sep 26, 2009 8:46 pm

.
Teacher,

You might consider only checking your bands quarterly (or semiannually or annually - whatever mix you feel comfortable with) - as Rodc said ”anything reasonable” that you can maintain comfortably.
Randy

User avatar
fishnskiguy
Posts: 2595
Joined: Tue Feb 27, 2007 1:27 pm
Location: Sedona, AZ

Post by fishnskiguy » Sat Sep 26, 2009 9:12 pm

Seems to me there was a recent thread that had a study that indicated that the "sweet spot" was right around 20% for each asset class, which intuitively seemed about right to me.

The study was directed to financial advisers, and suggested frequent looks at the portfolio and added that costs might be an issue for frequent rebalancers.

Since "looking" costs me nothing and neither does moving between our Vanguard funds (PCRIX being the exception) I rebalance whenever any of the following asset classes grow or shrink by 20%:

TSM
Value Index
Small Cap Value index
REIT Index
Emerging Markets
Precious Metals and Mining
All World ex-US
CCF (PCRIX)

We have set a floor below which we will not let our bond holdings drop. That floor goes up a little each year.

So far the strategy has served us well.

Chris
Trident D-5 SLBM- "When you care enough to send the very best."

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Post by Rodc » Sat Sep 26, 2009 9:30 pm

Seems to me there was a recent thread that had a study that indicated that the "sweet spot" was right around 20% for each asset class, which intuitively seemed about right to me.
Based on the 12 years leading up to and over the Great Tech Bubble, hardly enough time or a representative time, IMHO, to draw much of a conclusion.

Even so, I like 20%. Seems about right, don't rebalance all that often, but often enough.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

asset_chaos
Posts: 1345
Joined: Tue Feb 27, 2007 6:13 pm
Location: Melbourne

Post by asset_chaos » Sat Sep 26, 2009 10:15 pm

I think the bands should be set by two rational factors and one emotional factor. The volatility of an asset class and the cost to buy and sell the asset class are the rational factors. E.g., for a low volatility asset in a taxable account you'd want to set wider bands than for a high volatility asset in a tax deferred, no transaction fee account. The emotional factor is to set the bands so you're comfortable with them, whether that means 5%, 10% or whatever, and will act on them consistently.
Regards, | | Guy

User avatar
teacher
Posts: 988
Joined: Sun Oct 05, 2008 5:45 pm
Location: California

Post by teacher » Sat Sep 26, 2009 11:23 pm

dbr wrote:
In my understanding of rebalancing a person using bands does not have a frequency. More exactly the frequency is simply what is determined by when the bands indicate rebalancing.

Mention of dynamic allocation suggests you are applying two different rules to control asset allocation. Can you explain this in more detail?
If one shouldn't use both bands and a rebalancing schedule, that answers my question. I just finished revamping our portfolio in August, and for the last 30 or so years I just bought and held and never rebalanced...pretty much 100% equities most of those years. Now that the portfolio is structured to be 50/50 AA and is well diversified, I want to stay the course. I saw a conflict with the idea of bands and a rebalance schedule. When we are in a bull or bear market, a prescribed schedule may not maintain the AA band. By 'dynamic allocation,' I just meant a shifting allocation.

Now I need to decide whether I want to use a band or schedule. I was planning to use a 10% band or 55/45 equity limit. I am comfortable allowing more either side. A more generous band would make it more likely to stay within the guidelines of the IPS. If I use a 20% band or 60/40 equity limit, I know I will not let emotion rule me, and I can abide by the IPS. I had planned on rebalancing annually in the summer months back to 50/50 AA dropping one percent to bonds each year. I think the market is typically sluggish in the summer (with the exception of this summer :) ) , and I may be able to tax loss harvest more at that time of the year. But if I choose a band over a rebalance schedule, I can peak more often to see if I am within the band guidelines. Please set me straight if any of this does not sound reasonable.

asset_chaos wrote:
I think the bands should be set by two rational factors and one emotional factor. The volatility of an asset class and the cost to buy and sell the asset class are the rational factors. E.g., for a low volatility asset in a taxable account you'd want to set wider bands than for a high volatility asset in a tax deferred, no transaction fee account. The emotional factor is to set the bands so you're comfortable with them, whether that means 5%, 10% or whatever, and will act on them consistently.
My funds are almost all index, VTSMX, VFWIX, VBMFX & TIPS and 90% of it is in tax deferred so rebalancing will not be costly. I feel comfortable starting with the more generous band initially and after time, I may want to tighten it a bit, but 20% seems a good place to start.

Thanks everyone for your input. I have a different perspective on this now as a result of your comments. I know if I don't have workable guidelines, I will go with equities because that is my track record, and I am getting too long in the tooth to keep up that nonsense.

User avatar
Dale_G
Posts: 3082
Joined: Tue Feb 20, 2007 5:43 pm
Location: Central Florida - on the grown up side of 81

Post by Dale_G » Sun Sep 27, 2009 12:42 am

I don't think there is any "sweet spot" for rebalancing. People looking at the past 3, 5, 10, 15, 20, or 30 years of market performance will be able to come up with an optimum, but data mining the past rarely predicts the future.

None-the-less, I favor rebalancing with some sort of mechanical system whether based on the calendar or bands.

I use bands - and I use fairly narrow bands because the number of dollars I need to move end up being relatively small - and this improves my comfort level. Call it a crutch if you will, but it is easier for me to move 10K than it is to move 100K.

From 11/08/07 to 03/05/09 I moved well into 6 figures from bonds to stocks at an average TSM price of $24.25 to maintain my 50% equity asset allocation. From 06/01/09 to date I've sold mid-five figures to get back to 50% equities.

At the moment, I am very close to target. If bonds do nothing and equities go up by about 4%, I'll be selling equities. If bonds do nothing and equities go down by 4%, I'll be buying equities. For the past 30 years or so, volatility has been my friend, rather than my enemy.

Dale
Volatility is my friend

User avatar
ObliviousInvestor
Posts: 3354
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Post by ObliviousInvestor » Sun Sep 27, 2009 7:27 am

In Larry's What Wall Street Doesn't Want You to Know, he suggests a combination of bands & frequency.
"I suggest using a 5%/25% rule in an asset class's allocation before rebalancing. That is, rebalancing should only occur if the change in an asset class's allocation is greater than either an absolute 5% or 25% of the original percentage allocation....The portfolio should undergo the 5%/25% test on a quarterly basis."
He then notes,
"The 5%/25% test is just a guideline. You can create your own rebalancing guideline. The discipline this process provides is far more important than the ratios used."
Mike Piper, author/blogger

livesoft
Posts: 62278
Joined: Thu Mar 01, 2007 8:00 pm

Post by livesoft » Sun Sep 27, 2009 7:52 am

Dale_G wrote:...
I use bands - and I use fairly narrow bands because the number of dollars I need to move end up being relatively small - and this improves my comfort level. Call it a crutch if you will, but it is easier for me to move 10K than it is to move 100K.
...
That's sounds like a good system, but why not have wider bands and move smaller amounts? That is, once your asset allocation is out-of-range, there doesn't have to be a rule that says move $100K to get back. The rule could be "move $100K in $10K to $20K increments over the next few weeks" to get back.

I am intrigued by narrow bands because I think that with a volatile market that moves up or down by 3% to 4% in a single day, that there opportunities to rebalance on those days.

Another mental crutch is to look at the amount as a percent-of-portfolio and not as a dollar amount. Indeed, I have the opposite problem: I know that moving small amounts will really have no affect on the portfolio performance, so one has to move largish amounts.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Post by Rodc » Sun Sep 27, 2009 8:43 am

ObliviousInvestor wrote:In Larry's What Wall Street Doesn't Want You to Know, he suggests a combination of bands & frequency.
"I suggest using a 5%/25% rule in an asset class's allocation before rebalancing. That is, rebalancing should only occur if the change in an asset class's allocation is greater than either an absolute 5% or 25% of the original percentage allocation....The portfolio should undergo the 5%/25% test on a quarterly basis."
He then notes,
"The 5%/25% test is just a guideline. You can create your own rebalancing guideline. The discipline this process provides is far more important than the ratios used."
That is not really both.

He is suggesting bands, but only look quarterly.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Post by Rodc » Sun Sep 27, 2009 8:49 am

asset_chaos wrote:I think the bands should be set by two rational factors and one emotional factor. The volatility of an asset class and the cost to buy and sell the asset class are the rational factors. E.g., for a low volatility asset in a taxable account you'd want to set wider bands than for a high volatility asset in a tax deferred, no transaction fee account. The emotional factor is to set the bands so you're comfortable with them, whether that means 5%, 10% or whatever, and will act on them consistently.
That is not a bad idea. I use the cost issue in my brokerage account (tax advantaged) to set a maximum buy frequency. I thought about using bands based on volatility, but decided that given the fairly unimportant nature of the rebalancing details and the fact that with just redirecting inflows to lagging funds I don't actively rebalance very often, it was better to just stick with something simple. Still, not a bad idea.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

User avatar
ObliviousInvestor
Posts: 3354
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Post by ObliviousInvestor » Sun Sep 27, 2009 9:25 am

Rodc wrote:
ObliviousInvestor wrote:In Larry's What Wall Street Doesn't Want You to Know, he suggests a combination of bands & frequency.
"I suggest using a 5%/25% rule in an asset class's allocation before rebalancing. That is, rebalancing should only occur if the change in an asset class's allocation is greater than either an absolute 5% or 25% of the original percentage allocation....The portfolio should undergo the 5%/25% test on a quarterly basis."
He then notes,
"The 5%/25% test is just a guideline. You can create your own rebalancing guideline. The discipline this process provides is far more important than the ratios used."
That is not really both.

He is suggesting bands, but only look quarterly.
Fair enough. I'd still say it's certainly different from somebody who intends to rebalance any time that the bands are exceeded and who checks, say, daily or weekly.
Mike Piper, author/blogger

User avatar
Peter Foley
Posts: 4497
Joined: Fri Nov 23, 2007 10:34 am
Location: Lake Wobegon

Post by Peter Foley » Sun Sep 27, 2009 9:39 am

I think the recent thread that fishnskiguy referred to was: "What does rebalancing do for you? Not much "

My take away from that thread was that there was not a lot to be gained from frequent rebalancing. However, rebalancing periodically so that your AA does not drift too much from your ideal is warranted.

Based on advice from the forum I changed from annual rebalancing to bands about two years ago. My bands are roughly 5%. I do not slice and dice and a high percentage of my assets are in tax deferred. The only buy and sell rebalancing I do is related to equities versus bonds. Foreign versus domestic I handle by redirecting new contributions.

While my plan calls for rebalancing at 5%, I am not constantly recalculating my holdings to see where I'm at. I basically have been recalculating AA percentages every time the S&P 500 moves by 100 points. [Since 100/1400 S&P = 7.1% and 100/700 S&P= 14.2%, my reviews are not equally spaced and so my AA is sometimes off more than 5%. During the recent recovery this was to my advantage as it let my equities run a little longer.] As the market was falling it took about a 200 point drop in the S&P 500 for me to hit my rebalancing bands.

User avatar
teacher
Posts: 988
Joined: Sun Oct 05, 2008 5:45 pm
Location: California

Post by teacher » Sun Sep 27, 2009 11:05 am

Rodc wrote:
So I never make a buy in less than 6 months of the last buy. Completely arbitrary. A year might be even better too keep from buying falling knives.
You articulated a concern I have here. Instinctively, I felt there may be a need to wait a bit to let the chips fall before rebalancing when the market is as fast paced as it is now. I feel the timing of the rebalance is important for this reason. Timing the rebalance is not the same as a scheduled rebalance. I am beginning to think the band size is most important and frequency should occur when the bands are compromised and when knives are not falling so much, but all this is gut feeling on my part. I am just sort of thinking out loud here, formulating my thoughts with your input. As a novice, I appreciate your thoughts a great deal.

Also, thanks Peter Foley, for the hefty thread.
http://www.bogleheads.org/forum/viewtop ... ebalancing
I should be able to glean a couple nuggets from it.

walkinwood
Posts: 72
Joined: Wed Sep 05, 2007 10:57 am
Location: Denver, CO

Post by walkinwood » Sun Sep 27, 2009 12:03 pm

There is an article in the Jan 08 issue of Journal of Financial Planning called "Opportunistic Balancing : A new Paradigm for Wealth Managers".

http://www.irebal.com/docs/Opportunisti ... ancing.pdf

The conclusion is that
a) Use rebalancing and 'tolerance' bands. Tolerance bands are 50% of the rebalance bands.
b) Use wide rebalancing bands. (20% seems to be the sweet spot)
c) Inspect often, but rebalance only when bands are crossed.
d) Rebalancing benefits outweigh tax & transaction costs.

Tolerance bands are the values to which you rebalance your portfolio when it crosses a rebalance band. They advocate this to reduce the number of rebalancing transactions. Eg if an asset moves out of the rebalance band, rebalance the portfolio so that all assets are within the tolerance band, rather than trying to get them all at the target allocation.

I found the paper very insightful and am using the approach.

dbr
Posts: 27207
Joined: Sun Mar 04, 2007 9:50 am

Post by dbr » Sun Sep 27, 2009 12:13 pm

How wide bands are and what frequency than dictates depends on the volatility of the assets involved and the AA. A 40/60 allocation with a 5% band to rebalance stock/bond will still not rebalance very frequently. Someone rebalancing an 80/20 with a 5% band will see more frequent reblancing. Note there is a logical disconnect in that 5% applied to 80/20, meaning 75/25 is the edge, works differently than 5% applied to 50/50, meaning 55/45 is the edge. Someone concerned about rebalancing small value or emerging market may want much wider bands.

Even in the recent market situations a 50/50 AA with 5% bands has not needed to rebalance that often.

Rob't
Posts: 195
Joined: Thu Apr 19, 2007 8:33 pm

fishnskiguy's bond base

Post by Rob't » Sun Sep 27, 2009 12:15 pm

Fishnskiguy said:
We have set a floor below which we will not let our bond holdings drop. That floor goes up a little each year.
One of the things that's scary about catching the falling knife is wondering whether you will excessively deplete your "safe" fixed allocation buying into a prolonged bear. I fear some very well intentioned re-balancing plans (especially my own?!) might not survive a perception that you are feeding the entire content of your nest egg to an asset that seems intent on going to zero. Thus this "adjustable bond floor" is interesting.

Could your share how you determined its construction and value?

dbr
Posts: 27207
Joined: Sun Mar 04, 2007 9:50 am

Post by dbr » Sun Sep 27, 2009 1:12 pm

It is a possible plan that one should never rebalance from bonds to stocks, but only vice-versa. It is also a possible plan that one should not rebalance at all.

I have always wondered why an investment plan that includes an at least minimum balance to a "safe" asset would not in fact commit the entirety of the balance to the safe asset. After all, if one's financial position is assured, then shouldn't one be even happier to have a no risk assurance of having more than the minimum? This is the part of Plan B and of other thoughts about the riskiness of equities that logically leads to the conclusion that in many circumstances one should not be invested in equities at all.

Is it possible that there are no risk free assets, just that different assets have different types of risks and different consequences of risks materializing? Is not diversification and maintaining diversification the all purpose risk management tool?

User avatar
Rick Ferri
Posts: 8506
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri » Sun Sep 27, 2009 1:38 pm

dbr wrote:It is a possible plan that one should never rebalance from bonds to stocks, but only vice-versa. It is also a possible plan that one should not rebalance at all.
I have crunched countless numbers on rebalancing and it is a 'risk control' mechanism that does seem to work to control investor emotions. If you compare the risk of a 50/50 portfolio each decade, assuming the portfolio is rebalanced annually, the risk as measured by standard deviate is in a very tight range with one standard deviate between 8 and 9 percent. It did not matter which decade you study, the risk of a 50/50 portfolio is always about the same. Rebalanced portfolio returns have predictable long-term risk.

Not rebalancing creates stealth risk that creeps into your portfolio and may cause you to react to downturns the wrong way. If you let the portfolio ride and never rebalance, your return will be higher over most decades and the risk will be increasing higher as you go along. During a bull market, you are increasing portfolio risk as stocks move higher. When a correction comes, you have more risk in the portfolio than when you started, and that will cause bigger losses, which may cause an emotional knee-jerk reaction to sell. That is not in your best interest.

Rick Ferri

JW-Retired
Posts: 6877
Joined: Sun Dec 16, 2007 12:25 pm

Post by JW-Retired » Sun Sep 27, 2009 2:55 pm

Are you guys sure you have thought through this mechanistic tolerance band rebalancing? Personally, I want to be be prepared for a worst case stock market. By worst I mean a hypothetical calamity like a 1929-32 type bear market that never recovers much at all in the rest of my lifetime. Japan's stock market of the last 20 years is a real example of this, but I think not quite as bad.

The depression bear market lost 89% from the DOW peak in 1929. If I have a 50/50 AA and that calamity happens to us I would still have 55.5% of my money left, but only if I do not rebalance.... i.e., all my original bonds are still there and I'm assuming are worth roughly the same, and the stocks are now worth 11% of their original value. The bear would shift my AA to 89/11 at the bottom. On the other hand, a steadfast Boglehead tolerance band rebalancer would have also poured the majority of his bond assets into the bear market money shredder by selling bonds and buying more stocks during the trip down. I did a spreadsheet exercise to see exactly how much this would amount to. If a 50/50 AA rebalancer used 5% bands and rebalanced every time the band edge was crossed, then by the time stocks had reached the 89% loss point he would be down to 34.6% of his original stock/bond portfolio value.

So at the bottom, if you rebalance you lose 65.4%. If you don't rebalance you lose 44.5%. Big difference. Rebalancing is risky. Now if the market fully recovers both strategies get back to even again if they keep doing what did on the way down, so no problem there. The problem is what if it takes 20 years for the market to recover and the Boglehead needs income to live on? He would need to liquidate his now emaciated 50/50 portfolio. Ouch!

I'm not going to get into that position. I didn't rebalance last fall and don't intend to do so into a bear market when it happens again. Likewise into a bull market, where I will just let my AA run up. However, you obviously need to get back to your target AA at some time. So instead of tolerance bands or arbitrary time intervals I'm using a market momentum based method to pick the time. In particular, a 50/200 moving average market timing (gasp!) method. That method times your rebalancing so you won't be doing it in a long falling market, you wait until it bottoms and starts up again or at least until it has gone a long way down. The time signal came on August 3rd by this method, 2 years into the bear. Likewise, in a long rising market this method will usually wait until after a top. I can't see how it could do anything but help if I time my rebalancing like this. Sometimes the timing is particularly good and you buy significantly low and sell significantly high. If it often is not particularly good then so what?...., all you did was rebalance back to your preferred AA anyway.

If some set of Boglehead blessed completely arbitrary time intervals are OK, how could this not be still better? IMO most importantly, it protects me from my feared worst-case calamity, which is my objective.

Those Bogleheads who see pitfalls in this please explain to me what they are. I can't seem to find any.
JW

User avatar
gotherelate
Posts: 845
Joined: Wed May 28, 2008 6:57 pm
Location: Texas

Post by gotherelate » Sun Sep 27, 2009 3:53 pm

Not really pitfalls, but a couple of questions. If you obviously need to get back to your target AA only in bull markets, this sounds a lot like a one-directional rebalancing plan, i.e., rebalance only out of but not into equities.

Also, what if the signal is wrong or you misinterpret the signal? That doesn't seem to be a problem with band or calendar based rebalancing.

What do you think?

-Grandpa
-Grandpa | I'd rather see where I'm going than see where I've been.

JW-Retired
Posts: 6877
Joined: Sun Dec 16, 2007 12:25 pm

Post by JW-Retired » Sun Sep 27, 2009 4:42 pm

gotherelate wrote: Not really pitfalls, but a couple of questions. If you obviously need to get back to your target AA only in bull markets, this sounds a lot like a one-directional rebalancing plan, i.e., rebalance only out of but not into equities.
Also, what if the signal is wrong or you misinterpret the signal? That doesn't seem to be a problem with band or calendar based rebalancing.

I must have been unclear. It's not one-directional or only in bull markets. I just rebalanced into more equities on Aug 3. My bonds fraction had grown because of the bear market. Since Aug 3 the stock fraction is growing. Whenever the timing signal comes that the present bull market is over and it's a bear market, I will rebalance (once) to 50/50 equities (assuming I am indeed above my 50/50 stock AA.) I just won't keep on doing it over and over throughout the bear market.

If the signal is "wrong" it has the same effect as if I happened to rebalance on that date based on the calender. The day after the signal day I'm back to my 50/50. Why is one rebalancing method any more of a "problem" than the other?
JW

Rob't
Posts: 195
Joined: Thu Apr 19, 2007 8:33 pm

Post by Rob't » Sun Sep 27, 2009 4:59 pm

Just to see if the timing police have their radar on, JW's idea is half appealing to me.

The top half is not. I would fear allowing equity to run up too far as it would just add way more risk than I am comfortable with. This time it worked great, but sometimes that first signal to rebalance into fixed comes a tad late ('87, '00) and your 60/40 that you let go to 70/30 is 50/50 by the time you catch it.

On the bottom end, though, it might make it less likely that you buy an asset on its way to zero, or hold it during a 20 year sideways period.

User avatar
Rick Ferri
Posts: 8506
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri » Sun Sep 27, 2009 5:25 pm

Whenever the timing signal comes that the present bull market is over and it's a bear market, I will rebalance (once) to 50/50 equities (assuming I am indeed above my 50/50 stock AA.) I just won't keep on doing it over and over throughout the bear market.
"If this then that, but if that then this, and if another than Plan B." Talk about complicated.

Don't play mind games with the market. If you are not willing to rebalance back to 50% stock/ 50% bond position, then you do should start with a lower allocation to stocks. Try 30% stocks and 70% bonds and see if you have the stomach to rebalance each time you should.

Rick Ferri

JW-Retired
Posts: 6877
Joined: Sun Dec 16, 2007 12:25 pm

Post by JW-Retired » Sun Sep 27, 2009 5:52 pm

Whenever the timing signal comes that the present bull market is over and it's a bear market, I will rebalance (once) to 50/50 equities (assuming I am indeed above my 50/50 stock AA.) I just won't keep on doing it over and over throughout the bear market.
Rick Ferri wrote: "If this then that, but if that then this, and if another than Plan B." Talk about complicated.
Don't play mind games with the market. If you are not willing to rebalance back to 50% stock/ 50% bond position, then you do should start with a lower allocation to stocks. Try 30% stocks and 70% bonds and see if you have the stomach to rebalance each time you should.
Sorry to mislead. There is no "if this then that" stuff. Just a simple signal to follow. I should have said something like, "throughout the serious bear market, the system is very unlikely to have me rebalancing over and over like a tolerance band system inevitibly would dictate."
JW

User avatar
fishnskiguy
Posts: 2595
Joined: Tue Feb 27, 2007 1:27 pm
Location: Sedona, AZ

Re: fishnskiguy's bond base

Post by fishnskiguy » Sun Sep 27, 2009 6:22 pm

Rob't wrote:Fishnskiguy said:
We have set a floor below which we will not let our bond holdings drop. That floor goes up a little each year.
One of the things that's scary about catching the falling knife is wondering whether you will excessively deplete your "safe" fixed allocation buying into a prolonged bear. I fear some very well intentioned re-balancing plans (especially my own?!) might not survive a perception that you are feeding the entire content of your nest egg to an asset that seems intent on going to zero. Thus this "adjustable bond floor" is interesting.

Could your share how you determined its construction and value?
My wife and I get along nicely on my Navy pension and social security. Our portfolio exists to help out my wife if I prematurely go The Great Ski Hill in the Sky, since her share of my pension is quite small.

I set the first floor about 2003 as enough to barely keep her out of Alpo at 2.5% withdrawal. It's now high enough to allow her the occasional sirloin steak. :D

Chris
Trident D-5 SLBM- "When you care enough to send the very best."

User avatar
Rick Ferri
Posts: 8506
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri » Sun Sep 27, 2009 7:54 pm

JW wrote:Sorry to mislead. There is no "if this then that" stuff. Just a simple signal to follow. I should have said something like, "throughout the serious bear market, the system is very unlikely to have me rebalancing over and over like a tolerance band system inevitably would dictate."
JW
How do you know it is a 'serious bear market' and not just a bear market or perhaps not bear market at all? Here is what I believe. If you are not willing to rebalance when you should, then you have the wrong asset allocation to start with.

Rick Ferri

JW-Retired
Posts: 6877
Joined: Sun Dec 16, 2007 12:25 pm

Post by JW-Retired » Sun Sep 27, 2009 8:08 pm

How do you know it is a 'serious bear market' and not just a bear market or perhaps not bear market at all?
I don't but I do the same thing regardless. I am rebalancing when I should.
JW

asset_chaos
Posts: 1345
Joined: Tue Feb 27, 2007 6:13 pm
Location: Melbourne

Post by asset_chaos » Sun Sep 27, 2009 9:10 pm

Rodc wrote:
asset_chaos wrote:I think the bands should be set by two rational factors and one emotional factor. The volatility of an asset class and the cost to buy and sell the asset class are the rational factors. E.g., for a low volatility asset in a taxable account you'd want to set wider bands than for a high volatility asset in a tax deferred, no transaction fee account. The emotional factor is to set the bands so you're comfortable with them, whether that means 5%, 10% or whatever, and will act on them consistently.
That is not a bad idea. I use the cost issue in my brokerage account (tax advantaged) to set a maximum buy frequency. I thought about using bands based on volatility, but decided that given the fairly unimportant nature of the rebalancing details and the fact that with just redirecting inflows to lagging funds I don't actively rebalance very often, it was better to just stick with something simple. Still, not a bad idea.
Thanks. I suspect for most people the emotional factor swamps the rational factors. Even though one can derive an "optimum" rebalancing band based on cost and volatility, the math still has a free coefficient that essentially chooses the relative importance of the rebalancing cost now with the potential dollar losses if the portfolio is at the upper end of the band just as the market drops. So, even the math says that, in the end, you set the bands where you're comfortable.
Regards, | | Guy

asset_chaos
Posts: 1345
Joined: Tue Feb 27, 2007 6:13 pm
Location: Melbourne

Post by asset_chaos » Sun Sep 27, 2009 9:25 pm

ObliviousInvestor wrote:
Rodc wrote:
ObliviousInvestor wrote:In Larry's What Wall Street Doesn't Want You to Know, he suggests a combination of bands & frequency.
"I suggest using a 5%/25% rule in an asset class's allocation before rebalancing. That is, rebalancing should only occur if the change in an asset class's allocation is greater than either an absolute 5% or 25% of the original percentage allocation....The portfolio should undergo the 5%/25% test on a quarterly basis."
He then notes,
"The 5%/25% test is just a guideline. You can create your own rebalancing guideline. The discipline this process provides is far more important than the ratios used."
That is not really both.

He is suggesting bands, but only look quarterly.
Fair enough. I'd still say it's certainly different from somebody who intends to rebalance any time that the bands are exceeded and who checks, say, daily or weekly.
I expect that nearly every non-professional investor that uses a band implicitly has a frequency too because hardly any non-professional would tally up their portfolio daily. Maybe, I'm wrong on that, or maybe someone who knows what a rebalancing band is and would think about what theirs should be is self-selecting to also tally their portfolio more often than the general population. But you're right that the results should be somewhat different between infrequently checking if the bands have been breached versus checking continuously. I suspect the difference, though, is small between daily and quarterly checking.
Regards, | | Guy

User avatar
artthomp
Posts: 327
Joined: Tue Feb 20, 2007 4:44 pm
Location: Ballwin Missouri

Rebalancing

Post by artthomp » Mon Sep 28, 2009 9:27 am

At the end of 2008 my portfolio allocation was 40% equities/60% income. This was more or less based upon the rule of thumb that one should hold the fixed income portion in proportion to one’s age and I was 68 years old.

I normally rebalanced in December and, up till recently, this meant moving assets from equities into bonds. Last fall during the decline the portfolio shrank 22% at the lowest point and the portfolio allocation became even less than 30% equities versus more than 70% fixed income. Because of the extremely unsettled market in December and because I am approaching 70 years old I decided not to rebalance from fixed income but to reset my desired portfolio allocation to a new ratio reflecting my age better.

Since the rebound, the portfolio is back within 8% of the all time high but the allocation is approximately 32% equities/ 68% fixed income. I plan to rebalance to my new 30% equities/ 70% fixed target when I take my first IRS Required Minimum Distribution (RMDS) next year. I also plan to reinvest some of the taxable proceeds back in the equities portion of the portfolio.

My planned allocations for next year are:
Vanguard Money Market – 2%
Vanguard Total Bond Index – 34%
Vanguard Inflation-protected Bond Fund – 34%
Vanguard Total Stock Market Index Fund – 20%
FTSE All-World ex-US Inv – 10%
Art

Post Reply