Have you hit your rebalancing bands yet?

[DoWahDaddy]

I am 75/25 and still have not hit my bands over the past couple weeks. I rebalance on a 5% absolute change in the 75/25 alloc, and im still only halfway there (72.5/27.5). Have i lost my mathematical senses or does anyone else share this baffling status.

[stevewolfe]

What are your investments?!

[bob90245]

I checked my portfolio on Friday. And two holdings are below their lower band. So I will rebalance. However, when I do perform a rebalance, I wait a little over month from the previous rebalance. And since I performed my previous rebalance last month, I'm still waiting for that monthly window to elapse.

[Boglenaut]

I've hit so many bands they can power my ruber band airplane.

[integrity]

Lets do some math.

Assume you have $75 of equities and $25 of stocks. Stocks go down by 5%. So you have (0.95*$75) = $71.25.

Bonds may have gone up by 2%. So (1.02*$25) = $25.50.

So now your stocks are ($71.25)/($71.25+$25.50) = 73.6%.

So you went down by 1.4%.

This should be obvious when you realize that the increase in value of your bond holding is actually "giving back" some of your losses in the denominator. Also there is the fact that your 5% hit only applied to 75% of your overall holdings, not 100%.

[leod]

i have and now thinking of rebalancing now or wait on 9/1 my original schedule

[DRiP Guy]

I was getting all set mentally to do a rebalance.... then I checked and remembered that I had put most of my equity/ short term bond stuff into the single Vanguard 2010 retirement fund, VTENX, instead of in separate funds.

I went ahead and checked balances anyway just now, but since I still have a lot of CDs that go to the cash side of the ledger.... nothing for me to do -- it looks like the bond/cash/equity split within the fund itself is smoothing things out just like it's supposed to.

Not excited about total return vs S&P prior to this week, but looking at the one recent week drop in the stock indexes, and the overall trend in my portfolio since I made the change.... well, everything is mighty fine to me.

I likewise, of course, will miss out on the chance to capitalize on what one would suppose will eventually be a huge equities rally, but.... I'll automatically participate, in my own proportion. And that's good enough, for me.

[JimInIllinois]

The higher your stock allocation the more stocks have to drop to hit your 5% band. If your target is 90/10 and we assume bonds don't change then to hit 85/15 your stocks would have to drop by 37%. To go from 75/25 to 70/30 the required drop is 22%, which is a pretty big correction. What really surprised me is that to go from 50/50 to 45/55 still requires stocks to drop 18%.

I'm guessing you didn't realize this when you chose 5% absolute change as a trigger in your IPS.

[DoWahDaddy]

Very good point re 90/10, i suppose it feels more like a rebal trigger should have been hit than the math reveals, and thats the purpose of the ips isnt it, to replace gut feel with a solid plan and stay the course? I will say i have smaller allocations that have breached based on my 20% move in allocation pct (1% to 0.8%), but again i question the appropriateness of rebalancing. Dont you cutoff the benefit of the 5% absolute band if you undercut it by reallocating smaller breaches bevore the 5% has a chance to be reached? Has anyone looked at that before? Even a 5% say reit alloc if it goes to 4%, arent you preventing the bigger 5% absolute Band from doing its job? My guess is that any return benefit is neglected in order to maintain the desired risk parameters.

[YDNAL]

I am 75/25 and still have not hit my bands over the past couple weeks. I rebalance on a 5% absolute change in the 75/25 alloc, and im still only halfway there (72.5/27.5). Have i lost my mathematical senses or does anyone else share this baffling status.

I agree with you Daddy. It appears lots of people are scrambling to find (make?) ways to "do something."

I always choose to sit tight until my risk-tolerance is tested (up/down).

[JustinR]

I am 75/25 and still have not hit my bands over the past couple weeks. I rebalance on a 5% absolute change in the 75/25 alloc, and im still only halfway there (72.5/27.5). Have i lost my mathematical senses or does anyone else share this baffling status.

I agree with you Daddy. It appears lots of people are scrambling to find (make?) ways to "do something."

I'm wondering this as well. I'm only at 3.7% myself.

[jddphd]

I checked my balances last night and updated my "master spreadsheet" (that I bet many of you have as well). I started off around 73/27 stocks to bonds, so I'm looking at the same math at most of you.

Truth be told I'd be happy to stay where I am now rather than fiddling with anything!

[Rodc]

Not yet, but getting close.

[richard]

I'm 25/75. Under the popular 5 percentage points or 25% of a position (which ever is less), I'd have to hit 20/80 or 30/70 to rebalance. We're not close and haven't been for quite some time.

[tetractys]

My TIPS fund went through its +25% band by 3%. So today I'm rebalancing that against my Intl. Value fund, which is out the most in the other direction. -- Tet

[jheez]

Tips have been doing awesome. Spreadsheet told me to rebalance out of that into mid cap value (hardest hit)

[zed]

I'm at 50/50 and also use the 5% rule. Neither the aggregate band nor any of the individual AAs moved out of the 5% range. I did have a TLH opportunity this morning but it was swiftly traded away before I could act on it.

zed

[tfb]

If you are still following this, I created these two charts following the insights from JimInIllinois.

You need a ~20% drop in stock prices to hit the 5% trigger on the down side.



Stocks have to be up by ~25% to hit the 5% trigger on the up side.



More info: +/- 5% Rebalancing Bands

[Jay69]

Thanks for posting tfb.

Just so happens I was on vaction last week and from what I can tell I came close to my bands but being out of town I did not worry about it until today. Looks like I may or may not have missed out on a little gain, time will tell.

My IPS for rebalancing could use some fine tuning!

[Doc]

If you are still following this, I created these two charts following the insights from JimInIllinois.

You need a ~20% drop in stock prices to hit the 5% trigger on the down side.

This thing is bugging the heck out of me.

Looking at it bassackwards.

With a 90/10 and stocks drop 5%:

Your FI increases to:

10/(85+10)= 10.53% And it is your FI that has met the 5% change trigger. Or is that supposed to be 25% relative and 5% absolute?

And if it is a >20% drop that I need to trigger the rebalance how can I buy it back when I have only 10% FI to start. (Yeah I know, I'm using fuzzy math.)

The trick is to change your IPS from 90/10 to something more conservative.

[tfb]

With a 90/10 and stocks drop 5%:

Your FI increases to:

10/(85+10)= 10.53% And it is your FI that has met the 5% change trigger. Or is that supposed to be 25% relative and 5% absolute?

It's 5% absolute. FI will have to hit 15% before a rebalance is triggered. The charts don't take into account the 25%-relative rule.

[peter71]

Hi TFB,

Looks like great work, but do you have the graph legends partly reversed?

i.e., shouldn't it read, "required drop . . . to INCREASE" and "required gain . . . to REDUCE"?

I'm not a bands guy so apologies if I'm the one who's messed up the mental inversion!

Best,
Pete

[louis c]

One other reason some may be confused may be due to the 'starting point' of the correction - were you near but not quite at the top of the band? If so, you were not quite at the bottom when this got going.

I never hit the bottom of the band because I was about 2-3% over target when the correction started.

[tfb]

Looks like great work, but do you have the graph legends partly reversed?

i.e., shouldn't it read, "required drop . . . to INCREASE" and "required gain . . . to REDUCE"?

I don't know. By reduce and increase I meant the direct result of stock price changes, not the counter actions you take. If you start with 70% in stocks, a 20% drop in stock prices will reduce the 70% number to 65% and thus trigger a rebalance. Whether you rebalance to target or to the lower or upper edge, it's up to you.

[livesoft]

Wouldn't it be great if one could determine when the momentum of dropping was about to change?

Then one wouldn't need to use a 5% band.

[DoWahDaddy]

Tfb... this is a great addition to the boglehead chart database. You may want to slip it in the wiki.

Though it leads me to wonder...

There have been some studies on the best time to rebalance, and the 20/25 pct fluctuatuion in alloc pct combined w 5 pct absolute move is a generally accepted go to, and widely used metric. However, most people implement the annual rebalance as a result of their first come first served method in the rebal portion of their IPS.

Question is, are they shortchanging their ability to earn greater long term returns by having the annual option in there when no trigger has been hit?

Has there been any analysis of this?

I would think if you are 60/40, you cant be too compelled risk wise to rebal at 64/36 or 56/34 if there is potential for greater long term benefit of waiting for a numerical trigger.

[livesoft]

...
There have been some studies on the best time to rebalance, and the 20/25 pct fluctuatuion in alloc pct combined w 5 pct absolute move is a generally accepted go to, and widely used metric. However, most people implement the annual rebalance as a result of their first come first served method in the rebal portion of their IPS.

A recent poll suggests otherwise. Many, if not most, respondents said that they took some kind of rebalancing action very recently.

Question is, are they shortchanging their ability to earn greater long term returns by having the annual option in there when no trigger has been hit?

Has there been any analysis of this?

I would think if you are 60/40, you cant be too compelled risk wise to rebal at 64/36 or 56/34 if there is potential for greater long term benefit of waiting for a numerical trigger.

Have you read Daryanani's "Opportunistic Rebalancing" paper? It has some of this analysis.

[btenny]

Yes I have too dang much cash and no where decent to invest it. Gov't treasuries bonds and TIPS are just too expensive and I have a full allotment of corporates and munis so what now? Also sold stocks earlier this year to reduce international stock exposure so what now?

Bill

[btenny]

By the way all this band rebalancing stuff has me wondering about you young guys. All the rebalancing studies and academic stuff I have ever read say to ONLY rebalance ever 1 to 2 years at the MOST. Yes rebalancing may improve your performance and reduce risk if you buy when stocks are low and bonds are high. But the issue with rebalancing every time you hit 5% error bands could result in chasing a loosing stock market down like happened in the depression and may be happening now. All the studies I read say this is the key reason not to rebalance too often. Are you guys sure you really want to rebalance this often?

Bill

[zed]

tfb,

Thanks for taking the time to do the analysis.

fwiw my admittedly recently minted IPS says that I review/rebalance annually on my birthday. Thats what I wrote done and thats the standing plan. But having said that I'm giving serious consideration to Otar's recommendation of review/rebalance each election year. I think that there is a lot to be gained by a) realizing momentum and b) anticipating possible changes in administration/policy.

zed

[Jay69]

By the way all this band rebalancing stuff has me wondering about you young guys. All the rebalancing studies and academic stuff I have ever read say to ONLY rebalance ever 1 to 2 years at the MOST. Yes rebalancing may improve your performance and reduce risk if you buy when stocks are low and bonds are high. But the issue with rebalancing every time you hit 5% error bands could result in chasing a loosing stock market down like happened in the depression and may be happening now. All the studies I read say this is the key reason not to rebalance too often. Are you guys sure you really want to rebalance this often?

Bill

Being 42 years young, DCA every month to maintain my AA, I suppose I could say I'm rebalancing every month. I will say I just pick the fund I'm shortest in and buy just that one fund. Spent a few months chasing bonds may very well spend the next few months chasing stocks. Just hope I'm not tossing it into a rat hole

[Anon1234]

Have you read Daryanani's "Opportunistic Rebalancing" paper? It has some of this analysis.

Oh no, now I have a reason to peek. I'm going to need therapy. I was 0.04% from rebalancing to small cap value, but it never tripped and I did nothing.

[Jay69]

Have you read Daryanani's "Opportunistic Rebalancing" paper? It has some of this analysis.

Great Artical, thanks.

In my IPS I'm using 5% absolute bands.

The 20% range ballance with a 10% tolerance band may work very well as per the artical.

One question I do have, why not make the tolerance band something closer to 5%? If you have to rebalance anyway why not get it closer to the target AA?

For a smaller 5% holding a bigger band may make some sense to I would think, thoughts?

[YDNAL]

By the way all this band rebalancing stuff has me wondering about you young guys. All the rebalancing studies and academic stuff I have ever read say to ONLY rebalance ever 1 to 2 years at the MOST. Yes rebalancing may improve your performance and reduce risk if you buy when stocks are low and bonds are high. But the issue with rebalancing every time you hit 5% error bands could result in chasing a loosing stock market down like happened in the depression and may be happening now. All the studies I read say this is the key reason not to rebalance too often. Are you guys sure you really want to rebalance this often?

Bill

Chasing how and how much?

Previously, it has been stated and shown graphically, that in order to hit a 5% Band at the major asset class level (Stocks/Bonds), Stocks would have to drop as shown (or higher) to necessitate a rebalance:

Code: Select all

Stocks      Drop      Stocks   Bonds   New AA   
90   --->   37%   --->   56.7   10   85%   15%
80   --->   26%   --->   59.2   20   75%   25%
70   --->   19%   --->   56.7   30   65%   35%
60   --->   19%   --->   48.6   40   55%   45%
50   --->   19%   --->   40.5   50   45%   55%
40   --->   19%   --->   32.4   60   35%   65%
30   --->   24%   --->   22.8   70   25%   75%
20   --->   30%   --->   14.0   80   15%   85%
10   --->   50%   --->   05.0   90   05%   95%

And, of course, younger investors would be invested with higher Stocks at the top of the table.

[Sammy_M]

But having said that I'm giving serious consideration to Otar's recommendation of review/rebalance each election year. I think that there is a lot to be gained by a) realizing momentum and b) anticipating possible changes in administration/policy.

If following the presidential cycle rebalancing scheme, what does one do with new cash if it is not directed toward the underweighted asset class?

[tfb]

But having said that I'm giving serious consideration to Otar's recommendation of review/rebalance each election year. I think that there is a lot to be gained by a) realizing momentum and b) anticipating possible changes in administration/policy.

If following the presidential cycle rebalancing scheme, what does one do with new cash if it is not directed toward the underweighted asset class?

Add to portfolio in proportion to target allocation?

[kenyan]

I use relative percentages rather than absolute, because I don't like the distortions (as I see them) that come from using absolute bands of various stock allocations. I'd rather have my rebalancing triggers be based upon percentages of market movement. Perhaps if my rebalancing had tax implications, I would aim to do so less frequently.

[#Cruncher]

I created these two charts following the insights from JimInIllinois ...
More info: +/- 5% Rebalancing Bands

Thanks, tfb, for the article on your blog where I first came across this thread. I'd never before realized how much stocks would have to rise or fall in order to change the allocation by 5% points. I've generalized the formula in your article to work for bands other than 5%:

x = b / (s X (1 - s - b)) where
b is the band size and direction -- e.g., -0.01 for stock alloc to fall 1% point
s is the original stock allocation -- e.g., 0.9 for originally 90% stocks
x is the change in stock value -- e.g., -0.10 or stocks would have to fall 10%

Here is a table similar to YDNAL's above, except showing how much stocks would have to fall to reach just a 1% point band:

Code: Select all

Orig     Stocks      New
Stocks    Drop      Stocks   Bonds      New AA   
------    ----      ------   -----    ---------
  90       10%       80.9      10     89%   11%
  80        6%       75.2      20     79%   21%
  70        5%       66.8      30     69%   31%

  60        4%       57.6      40     59%   41%
  50        4%       48.0      50     49%   51%
  40        4%       38.4      60     39%   61%

  30        5%       28.6      70     29%   71%
  20        6%       18.8      80     19%   81%
  10       11%        8.9      90      9%   91%

So in DoWahDaddy's case, even if his band size was only 1% point, stocks would still have to fall 5-6% to move from 75:25 to 74:26.

[tfb]

I've generalized the formula in your article to work for bands other than 5%:

x = b / (s X (1 - s - b)) where
b is the band size and direction -- e.g., -0.01 for stock alloc to fall 1% point
s is the original stock allocation -- e.g., 0.9 for originally 90% stocks
x is the change in stock value -- e.g., -0.10 or stocks would have to fall 10%

I also generalized the formula (in the comments) to include the size of bond price changes. Using t instead of b in your formula above,

x = t * (1 + b) / s / (1 – t – s) + b

where

x = stock price change, up 1% = +0.01, down 1% = -0.01
t = rebalancing threshold, up 5% = +0.05, down 5% = -0.05
b = bond price change, up 1% = +0.01, down 1% = -0.01
s = starting weight of stocks, 60% = 0.6

If bonds are up when stocks are down, stocks don't have to be down as much to hit the rebalancing band. If a run-up in stock prices took you near the upper edge of the rebalancing band, it will take a larger drop to move past the lower edge.

You basically multiply the per-1%-change number in your table by the distance to the edge of the band, then add bond price changes.

[rkhusky]

Funny, I just did the same calculation last week, because I was noticing the same thing and was curious.

[zed]

hmmm, the bogleheads webserver seems to need a reboot. It's not letting me reply to earlier posts.

In response to SammyM and tfb:

I directed the question, "what to do with cash accumulations", to Mr Otar. The delay in response is mine not his.

Jim C, Otar
I am sensing that you are talking about accumulation portfolios and not distribution portfolios.

I have observed that 4-year rebalancing protects the portfolio from the black-swan type events in the equity market a bit better than frequent rebalancing. However, when you take an average of 110-years, its effect gets lost in the average -like any other Gaussian type methodology. So, the 4-year rebalancing is like an insurance, you won't see much of its effect on a day-to-day basis other than extreme events.

Generally, my guidelines -that is if you are following the buy and hold type strategic asset allocation- for rebalancing is as following (from pages 81 and 82):
= = = = =
What is the purpose of rebalancing? Before you can answer this question, you need to remember the most important thing for a buy–and–hold portfolio:

· In an accumulation portfolio: the important thing is the volatility of returns. Thus, the purpose of rebalancing in an accumulation portfolio is to contain the volatility of returns.

· In a distribution portfolio: the important thing is the sequence of returns. Thus, the purpose of rebalancing in a distribution portfolio is to minimize the effect of a bad sequence of returns.

Having defined these two purposes of rebalancing, here are my general guidelines:

Accumulation Portfolios:

· If retirement is at least five years away:

When equity exceeds its target: rebalance whenever the equity percentage exceeds its target by more than 5%. There is no time limit for that; you may need to rebalance several times in a year if the market happens to be bullish.

When fixed income exceeds its target: when the fixed–income portion of the portfolio exceeds its target percentage by more than 5%, then rebalance on the anniversary of the most recent rebalance, never more often.

· Within five years of retirement:

Rebalance at the end of the presidential election years only.

Distribution Portfolios:

· If the withdrawal rate is 5% or less:

Rebalance at the end of the presidential election years only.

· If the withdrawal rate is over 5%:

Rebalance annually.

There is no holy grail for rebalancing. Following these guidelines can help you reduce large losses in extreme markets. In normal markets, rebalancing sells high and buys low. For a buy–and–hold portfolio, that is good enough for me.

If you want to do anything more sophisticated than that, you need to step up from the “rebalancing school” into the “market timing school”. It is important to differentiate between these two schools. Many people don’t or cannot see the difference. I have seen several studies on rebalancing that look like poor imitations of market timing strategies. Rebalancing is rebalancing. Market timing is market timing. If you confuse the two, then your returns will suffer. Whatever you do, not only in investments but in life in general, make sure that the purpose of each one of your actions is very clear to you.

And the next time you are asked to sign the form for automatic rebalancing, stop and think. You may be –unknowingly– signing away years of portfolio life or years of growth. The right way of investing is about being careful with strategies. It has absolutely nothing to do with convenience.

= = = = = =

The above reference in from (IMHO the required read) Mr Otar's book, "Unveiling the Retirement Myth"
http://www.retirementoptimizer.com/

zed

[papito23]

I have a desire to rebalance 1-2 times a year, once I figure out an IPS.

However, I might like to keep the door open to rebalance between scheduled times if I see a big move (ok, market timing via rebalancing, right?).

However, once I get off this Bogle-binge and portfolio dissection (I have OCD tendencies) I'd like to NOT be checking the news or my portfolio every week or month. Twice a year, perhaps.

Is there a way for my portfolio to communicate with me without me peeking in on it? Some kind of auto-generated email or something?

Dreaming...