Help me understand how rebalancing affects compounding

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
Ryan2390
Posts: 17
Joined: Mon Sep 16, 2013 12:07 pm

Help me understand how rebalancing affects compounding

Post by Ryan2390 »

Okay, so we have all heard about the "power of compounding".

My question is, how does rebalancing affect it?

My strategy is to use rebalancing bands of 5% for the major asset classes I own (US small cap, US total, International Total, Emerging Markets Small Cap ETF).

So say for example, starting in 2010, when US small caps did really well (and have continued until now), my small cap allocation would probably have exceeded 5% of the target, and I would have moved that money at the end of the year elsewhere.

Doesn't that hurt the power of compounding though? I'm guessing yes it does, but it's also a far too risky approach. Obviously had I not touched small cap money starting in 2010, my total return in small caps would now be close to 100%. I know I'm wrong about this as a prudent strategy, but can someone provide some simple data to show how re-balancing always wins out in the long run?

Thanks!!
User avatar
momar
Posts: 1359
Joined: Sun Nov 13, 2011 12:51 pm

Re: Help me understand how rebalancing affects compounding

Post by momar »

Rebalancing doesn't win in the long run. You are by definition moving money from higher return to lower return asset classes, on average.

It may or may not win; it depends on what everything does.

Rebalancing DOES allow you to maintain a risk profile appropriate for your situation.
"Index funds have a place in your portfolio, but you'll never beat the index with them." - Words of wisdom from a Fidelity rep
Topic Author
Ryan2390
Posts: 17
Joined: Mon Sep 16, 2013 12:07 pm

Re: Help me understand how rebalancing affects compounding

Post by Ryan2390 »

What do you mean by "winning"?

My basic understanding is that it is a reliable method to manage risk at the expense of the additional returns from letting winning asset classes compound. Can anyone expand on this?

If one has a well-diversified portfolio to begin with, is it a feasible strategy to never rebalance??

This article is interesting:

http://money.cnn.com/2013/10/01/investi ... .moneymag/

In the long run, if one does not need the money immediately, is it okay to never rebalance?
User avatar
bertilak
Posts: 7938
Joined: Tue Aug 02, 2011 5:23 pm
Location: East of the Pecos, West of the Mississippi

Re: Help me understand how rebalancing affects compounding

Post by bertilak »

A mind experiment.

You are invested in two asset classes, stocks and bonds, at a 60/40 ratio which (we are assuming) exactly matches the overall market's asset allocation.

Now time goes by and your allocation drifts, along with the total market's, to 55/45.

Question: Has your risk/reward ratio changed?
Last edited by bertilak on Sat Oct 19, 2013 3:35 pm, edited 1 time in total.
May neither drought nor rain nor blizzard disturb the joy juice in your gizzard. -- Squire Omar Barker (aka S.O.B.), the Cowboy Poet
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Help me understand how rebalancing affects compounding

Post by larryswedroe »

Basically what happens if you rebalance is that the portfolio's return will exceed the weighted average return of the components. By how much depends on correlations (the lower the larger the diversification) and the volatility.
Larry
Drew31
Posts: 222
Joined: Mon Oct 22, 2012 1:18 pm

Re: Help me understand how rebalancing affects compounding

Post by Drew31 »

There are three very good articles at the bottom of the wiki that help explain the theory and practice on rebalancing. They really helped me on this very topic. I'd recommend checking those out.


http://www.bogleheads.org/wiki/Rebalancing
User avatar
nedsaid
Posts: 13955
Joined: Fri Nov 23, 2012 12:33 pm

Re: Help me understand how rebalancing affects compounding

Post by nedsaid »

For me, rebalancing will most likely dampen the volatility of my portfolio and also the returns. I do not expect a "rebalancing bonus" or a "rebalancing premium." Why? Because at age 54, I am stock heavy at 69% stocks and 31% bonds and cash. It is highly likely that my rebalancing will all be one way from stocks to bonds.

If on occasion, I rebalanced from bonds to stocks I might capture a "bonus" or a "premium." I am unlikely however to do that. The last time the stock market fell a lot back in 2008-2009, I was too scared to sell bonds to buy stocks. What I did do however was shift all my new monies for investment into stocks for about a year. Now being 4 years older, if the same thing happened again, I would be even less willing to sell bonds to buy stocks.

A younger person might be more able to get a rebalancing bonus.
A fool and his money are good for business.
Topic Author
Ryan2390
Posts: 17
Joined: Mon Sep 16, 2013 12:07 pm

Re: Help me understand how rebalancing affects compounding

Post by Ryan2390 »

Thanks for the replies so far.

So is it possible to just not rebalance? Would that be a major mistake, even if the portfolio is well diversified?
User avatar
Kalo
Posts: 543
Joined: Sat May 25, 2013 1:01 pm

Re: Help me understand how rebalancing affects compounding

Post by Kalo »

Ryan2390 wrote:Thanks for the replies so far.

So is it possible to just not rebalance? Would that be a major mistake, even if the portfolio is well diversified?
It would be a mistake to not rebalance and think that the risk profile of your portfolio would necessarily remain the same. It would probably increase since riskier assets tend to outperform less risky assets.

If your current AA is right for you today, you want to stick with it. If it drifts you want to get it back to what is right for you at that time.

I think only a person who is so wealthy that they could never run out of money could keep even a portion of their portfolio all in stocks. Even they may not want to though, because of natural loss aversion.

Kalo
"When people say they have a high risk tolerance, what they really mean is that they are willing to make a lot of money." -- Ben Stein/Phil DeMuth - The Little Book of Bullet Proof Investing.
User avatar
VictoriaF
Posts: 19500
Joined: Tue Feb 27, 2007 7:27 am
Location: Black Swan Lake

Re: Help me understand how rebalancing affects compounding

Post by VictoriaF »

Ryan2390 wrote:Okay, so we have all heard about the "power of compounding".

My question is, how does rebalancing affect it?
Rebalancing affects the risk profile of the portfolio and its overall returns, but it has little to do with compounding. Think about it this way. You have a portfolio that contains S amount of stocks and B amount of bonds. When you rebalance you move some money "d", e.g., from S to B. Immediately after ebalancing you'll have (S - d) in stocks and (B + d) in bonds. Observe that:
1. The total size of your portfolio has not changed. It's still S + B.
2. The money you have moved (d) can be viewed as a small piece of your portfolio that used to compound like "S" and now will be compounding like "B". It will not stop compounding.

Victoria
WINNER of the 2015 Boglehead Contest. | Every joke has a bit of a joke. ... The rest is the truth. (Marat F)
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Help me understand how rebalancing affects compounding

Post by larryswedroe »

one more thought, for a stock/bond portfolio rebalancing should definitely lower volatility because if nothing else on average you are selling stocks to buy bonds, which have lower volatility
User avatar
JoMoney
Posts: 9919
Joined: Tue Jul 23, 2013 5:31 am

Re: Help me understand how rebalancing affects compounding

Post by JoMoney »

If you believed "risk premiums" were constant and forever, and that one asset would always earn more then another, the best bet would be to put your money in whatever asset would earn the most and let it be. Over the past 15 years, U.S. Small-Caps (and Real Estate) have been the place to be. The best bet would have been to have all your money in those areas. Using MPT and various regression techniques you may have been able to find some optimal mix with something else that would have achieved more by rebalancing a particular percentage at a particular interval. The problem is, the formula for that rebalancing percentage and timing is not known in advance, it can only be calculated in hind-sight.
The 15 years prior to the most recent 15 year period would show Large-Cap stocks (and Telecom) as being the place to be, but again, only in hind-sight would you be able to see this....
If you combine the 30 years what you find is a "Return To the Mean" where the various asset classes show very similar performance. This isn't to say that over any prolonged period of time everything will find an equilibrium, even over the 30 year period Mid-Caps and Health Sectors had even stronger performance then the others. The problem is this is all only known in hind-sight. I've given up on trying to find the needle, and bought the hay-stack. Buy and hold the total-market and be done with it, you'll be guaranteed the markets performance, and based on many long-term empirical studies outperform 80% of the money that's trying to beat it.
As far as rebalancing to improve performance, I view it like all the gambling strategies that claim you can beat casino by following a certain process, it may show results from time to time, but over the long run it doesn't change the negative expected value of the casino game. The stock market has positive expected value, but gaming the system trying to achieve more then the sum of the parts doesn't work unless you manage to get lucky enough to get ahead, then quit while you're ahead.
(The performance I used to look at this is the data on Morningstar charts of real funds and benchmarks over the time period)
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham
Topic Author
Ryan2390
Posts: 17
Joined: Mon Sep 16, 2013 12:07 pm

Re: Help me understand how rebalancing affects compounding

Post by Ryan2390 »

Thanks again for the replies. So it seems like the money is still being compounded, albeit not at the highest rate possible as if you had left the money in the riskier asset classes. Of course, this also mitigates volatility. I think I understand...

Does anyone re-balance every 2 years? Or is 1 year generally agreed upon as the better option? I'm guessing that the longer the period before rebalancing, the higher return/higher risk. Is 2 or 3 years considered too much?
dbr
Posts: 33859
Joined: Sun Mar 04, 2007 9:50 am

Re: Help me understand how rebalancing affects compounding

Post by dbr »

Ryan2390 wrote:Thanks for the replies so far.

So is it possible to just not rebalance? Would that be a major mistake, even if the portfolio is well diversified?
It would be a major mistake if the result is that the portfolio evolves to a position that is too risky. If you start at 50/50 and find yourself at 80/20 twenty years later and then stocks crash and you sell out in a panic, then not rebalancing would have been the thing you didn't do that would have saved your skin, assuming you could have tolerated a stock crash at 50/50. That is all rebalancing really is.

PS I am not ignoring the correlation effects mentioned by Larry, just starting with first order effects.
Last edited by dbr on Sun Oct 20, 2013 9:04 am, edited 1 time in total.
dbr
Posts: 33859
Joined: Sun Mar 04, 2007 9:50 am

Re: Help me understand how rebalancing affects compounding

Post by dbr »

Ryan2390 wrote:Thanks again for the replies. So it seems like the money is still being compounded, albeit not at the highest rate possible as if you had left the money in the riskier asset classes. Of course, this also mitigates volatility. I think I understand...

Does anyone re-balance every 2 years? Or is 1 year generally agreed upon as the better option? I'm guessing that the longer the period before rebalancing, the higher return/higher risk. Is 2 or 3 years considered too much?
A better and more sensible choice than periodic is to rebalance when your asset allocation gets too far away from what you plan for it to be. A tolerance of about 5% of assets away from plan is one good suggestion. In practice you probably rebalance less than once a year that way. It takes some pretty big upsets to throw most asset allocations out of whack. Again the concept is the long term position.

You should also recognize that if you are making contributions or taking withdrawals you can use those transactions to continually push your allocation toward plan.
User avatar
nisiprius
Advisory Board
Posts: 42243
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Help me understand how rebalancing affects compounding

Post by nisiprius »

This is what I think.

a) Rebalancing reduces expected return because you are constantly rebalancing out of the assets with the highest expected return.

b) IF it is actually true that there is mean reversion and IF your rebalancing strategy happens to be in accord with the period over which mean reversion occurs--for example, rebalancing on a predetermined schedule on the order of "end of year in every even-numbered year" with or without rebalancing bands--then it MIGHT increase risk-adjusted return. A lot of people smarter and more knowledgeable than I think it does. If it does, it's not by a whole lot and I'm skeptical.

There are various backtested explorations of various rebalancing strategies but as always it's just backtesting and I doubt there's enough data or will ever be enough data to pull out statistically significant differences.

c) Whatever reason you had for having some specific asset allocation is a reason for rebalancing.

d) In terms of pure risk, measured in almost any way, the difference between, say 50/50 and 60/40 is awfully small, This is shown by the fact that Vanguard offers only four LifeStrategy funds. Rebalancing to hold portfolio risk steady isn't something that needs to be done very precisely or very often.

e) Rebalancing can increase risk. For example, during 2008-2009 my Balanced Index fund dropped a little more than it would have dropped with an initial 60/40 mix of Total Stock and Total bond, and then climbed back a little more. The reason is that when a volatile asset class keeps dropping, rebalancing keeps pulling money out of the stable assets and putting them into the falling assets. There was some discussion of whether that constitutes some kind of dangerous black hole. It certainly can feel that way if you are doing the rebalancing manually and consciously. In reality, the effect doesn't become very large unless you assume truly catastrophic declines in the more volatile asset. They are, of course, balanced by gains if the asset recovers, but in the thought experiment where the asset goes to zero, or close to zero, and stays there, it could be bad.
Last edited by nisiprius on Sun Oct 20, 2013 9:29 am, edited 1 time in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
VictoriaF
Posts: 19500
Joined: Tue Feb 27, 2007 7:27 am
Location: Black Swan Lake

Re: Help me understand how rebalancing affects compounding

Post by VictoriaF »

nisiprius wrote:d) In terms of pure risk, measured in almost any way, the difference between, say 50/50 and 60/40 is awfully small, This is shown by the fact that Vanguard offers only four LifeStrategy funds. Rebalancing to hold portfolio risk steady isn't something that needs to be done very precisely or very often.
I'd also add that if you are rebalancing by price, it's much more difficult to do just when you need it the most. For example, the best time to rebalance was March 2009, but at that time even the most weathered investors were reexamining their strategies.

Victoria
WINNER of the 2015 Boglehead Contest. | Every joke has a bit of a joke. ... The rest is the truth. (Marat F)
User avatar
nisiprius
Advisory Board
Posts: 42243
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Help me understand how rebalancing affects compounding

Post by nisiprius »

VictoriaF wrote:
nisiprius wrote:d) In terms of pure risk, measured in almost any way, the difference between, say 50/50 and 60/40 is awfully small, This is shown by the fact that Vanguard offers only four LifeStrategy funds. Rebalancing to hold portfolio risk steady isn't something that needs to be done very precisely or very often.
I'd also add that if you are rebalancing by price, it's much more difficult to do just when you need it the most. For example, the best time to rebalance was March 2009, but at that time even the most weathered investors were reexamining their strategies.

Victoria
Notice, however, that if you were rebalancing at the end of the year, well, neither year-end 2008 nor year-end 2009 happened to fall anywhere near the bottom. There is a very dangerous mental trap in which, in thinking about rebalancing, one imagines that it magically occurs at the bottom.

If you rebalance very frequently, e.g. almost continuously as in Balanced Index or a target-date fund, you are guaranteed to be rebalancing at the bottom... but in fact you get almost no "rebalancing bonus" because there isn't any mean reversion to speak of over periods on the order of a month; there may even be momentum.

Strategies involving combinations of calendar dates and rebalancing bands have an element of market timing formula to them, and work only to the effect that the manage to rebalance near the bottom without also managing to rebalance everywhere else, too.

The huge huge problem with investing is that we constantly fall back into bad mental models of price behavior in which we imagine them as smooth, continuous curves with much more predictability to them than they really have. In reality, you can't find a rule that reliably trips on the big movements without tripping on the small ones, too.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
swimirvine
Posts: 371
Joined: Tue May 28, 2013 7:51 am

Re: Help me understand how rebalancing affects compounding

Post by swimirvine »

I believe rebalancing can increase return in the long run. This is my VERY simple under standing of why:

Let's say you start off with a portfolio of 80% Large Cap and 20% small cap and over a two year period small cap does poorly and your threshold for rebalancing is 5%. you end up with 85% large cap and 15% small cap so you sell some large cap and buy small cap. By definition you have sold high and bought low. You have purchased small cap at a "discounted" price and you have also put your portfolio back to risk level that you wanted.

In one of the Bogle, Solin books it gives an example similar to this but it uses real historical numbers and shows how over 30 years or something you would have ended up with a higher return by rebalancing once a year.
The way I invest my money is not the right way to invest, it's the right way for ME to invest.
User avatar
midareff
Posts: 7234
Joined: Mon Nov 29, 2010 10:43 am
Location: Biscayne Bay, South Florida

Re: Help me understand how rebalancing affects compounding

Post by midareff »

and then again..... when the market is making new highs almost daily, and you have drifted several % away from your target, is really is tough not to sell a bit of equities to buy more bonds.

There is more than one approach..... there have been papers on balance bands.. Google "Opportunistic Rebalancing - A New Paradigm for Wealth Management", Jack Bogle seems to have been quoted as saying it doesn't matter much, Jim Otar recommends every fourth year after the Presidential cycle, and there are others.
Last edited by midareff on Sun Oct 20, 2013 9:51 am, edited 1 time in total.
User avatar
VictoriaF
Posts: 19500
Joined: Tue Feb 27, 2007 7:27 am
Location: Black Swan Lake

Re: Help me understand how rebalancing affects compounding

Post by VictoriaF »

nisiprius wrote:
VictoriaF wrote:
nisiprius wrote:d) In terms of pure risk, measured in almost any way, the difference between, say 50/50 and 60/40 is awfully small, This is shown by the fact that Vanguard offers only four LifeStrategy funds. Rebalancing to hold portfolio risk steady isn't something that needs to be done very precisely or very often.
I'd also add that if you are rebalancing by price, it's much more difficult to do just when you need it the most. For example, the best time to rebalance was March 2009, but at that time even the most weathered investors were reexamining their strategies.

Victoria
Notice, however, that if you were rebalancing at the end of the year, well, neither year-end 2008 nor year-end 2009 happened to fall anywhere near the bottom. There is a very dangerous mental trap in which, in thinking about rebalancing, one imagines that it magically occurs at the bottom.

If you rebalance very frequently, e.g. almost continuously as in Balanced Index or a target-date fund, you are guaranteed to be rebalancing at the bottom... but in fact you get almost no "rebalancing bonus" because there isn't any mean reversion to speak of over periods on the order of a month; there may even be momentum.

Strategies involving combinations of calendar dates and rebalancing bands have an element of market timing formula to them, and work only to the effect that the manage to rebalance near the bottom without also managing to rebalance everywhere else, too.

The huge huge problem with investing is that we constantly fall back into bad mental models of price behavior in which we imagine them as smooth, continuous curves with much more predictability to them than they really have. In reality, you can't find a rule that reliably trips on the big movements without tripping on the small ones, too.
I agree with everything you wrote, but I'd like to clarify that I specifically mentioned "if you are rebalancing by price" (i.e., not by calendar), and my point is that, paradoxically, the more it is appropriate to rebalance the more difficult it is to do.

Victoria
WINNER of the 2015 Boglehead Contest. | Every joke has a bit of a joke. ... The rest is the truth. (Marat F)
User avatar
VictoriaF
Posts: 19500
Joined: Tue Feb 27, 2007 7:27 am
Location: Black Swan Lake

Re: Help me understand how rebalancing affects compounding

Post by VictoriaF »

swimirvine wrote:I believe rebalancing can increase return in the long run. This is my VERY simple under standing of why:

Let's say you start off with a portfolio of 80% Large Cap and 20% small cap and over a two year period small cap does poorly and your threshold for rebalancing is 5%. you end up with 85% large cap and 15% small cap so you sell some large cap and buy small cap. By definition you have sold high and bought low. You have purchased small cap at a "discounted" price and you have also put your portfolio back to risk level that you wanted.

In one of the Bogle, Solin books it gives an example similar to this but it uses real historical numbers and shows how over 30 years or something you would have ended up with a higher return by rebalancing once a year.
Hi swimirvine,

Your summary is what drives the idea of rebalancing. Theoretically, it's simple and straightforward. In practice, the question of when to rebalance dominates the decision and it's far from being simple.

- Do you rebalance every several minutes as large funds and endowments do? It's too expensive and impractical for individual investors.
- Do you rebalance when stocks rise or decline by some percentage? How do you determine that percentage? There is some momentum in the market movements and by focusing on percentages you would lose some potential gains.
- Do you rebalance by calendar, e.g., on 5 January every year? Your rebalancing bonus may be very small because by that time your assets will have moved back towards the mean.
- Should rebalancing approach be the same whether you do it between domestic large-caps and small-caps, or between U.S. and international stocks, or between stocks and bonds? What if something fundamentally different going on in the world or in the fixed-income domain?
- Importantly, mean reversion is not a natural law.

Victoria
WINNER of the 2015 Boglehead Contest. | Every joke has a bit of a joke. ... The rest is the truth. (Marat F)
The Wizard
Posts: 13356
Joined: Tue Mar 23, 2010 1:45 pm
Location: Reading, MA

Re: Help me understand how rebalancing affects compounding

Post by The Wizard »

I've said it before, but "compounding" is really the wrong word to use for an investment portfolio containing a good percentage of stocks. Your $100K portfolio could easily be worth $90K in 12 months, compounding or not.
Compounding is more relevant to fixed income securities: CDs and bonds.

And yes, rebalancing tends to reduce total gains in a year like the present when stocks are doing well.
But in a year like 2009, rebalancing INTO stocks can have positive long term benefits...
Attempted new signature...
User avatar
swimirvine
Posts: 371
Joined: Tue May 28, 2013 7:51 am

Re: Help me understand how rebalancing affects compounding

Post by swimirvine »

VictoriaF wrote:
swimirvine wrote:I believe rebalancing can increase return in the long run. This is my VERY simple under standing of why:

Let's say you start off with a portfolio of 80% Large Cap and 20% small cap and over a two year period small cap does poorly and your threshold for rebalancing is 5%. you end up with 85% large cap and 15% small cap so you sell some large cap and buy small cap. By definition you have sold high and bought low. You have purchased small cap at a "discounted" price and you have also put your portfolio back to risk level that you wanted.

In one of the Bogle, Solin books it gives an example similar to this but it uses real historical numbers and shows how over 30 years or something you would have ended up with a higher return by rebalancing once a year.
Hi swimirvine,

Your summary is what drives the idea of rebalancing. Theoretically, it's simple and straightforward. In practice, the question of when to rebalance dominates the decision and it's far from being simple.

- Do you rebalance every several minutes as large funds and endowments do? It's too expensive and impractical for individual investors.
- Do you rebalance when stocks rise or decline by some percentage? How do you determine that percentage? There is some momentum in the market movements and by focusing on percentages you would lose some potential gains.
- Do you rebalance by calendar, e.g., on 5 January every year? Your rebalancing bonus may be very small because by that time your assets will have moved back towards the mean.
- Should rebalancing approach be the same whether you do it between domestic large-caps and small-caps, or between U.S. and international stocks, or between stocks and bonds? What if something fundamentally different going on in the world or in the fixed-income domain?
- Importantly, mean reversion is not a natural law.

Victoria
I see rebalancing the same as asset allocation. You choose what you want or in this case how you want to do it and you stick to your guns. I decided to rebalance once per year or at anytime a portion or my portfolio is out of balance by more than 5%. That's my plan and I'm sticking to it! :D
The way I invest my money is not the right way to invest, it's the right way for ME to invest.
Topic Author
Ryan2390
Posts: 17
Joined: Mon Sep 16, 2013 12:07 pm

Re: Help me understand how rebalancing affects compounding

Post by Ryan2390 »

Interesting replies.

So the 5% band seems to be pretty popular. Is this for each asset class or stocks as a whole?

I divide my stocks into 4 asset classes- Large US, Small US, Total International (mostly developed markets), and Emerging Markets.

Should I be re-balancing when one of those asset classes exceeds 5% of its original weight, or when my entire equity portion exceeds 5% of my portfolio?
User avatar
swimirvine
Posts: 371
Joined: Tue May 28, 2013 7:51 am

Re: Help me understand how rebalancing affects compounding

Post by swimirvine »

There's no right answer just like there's not right answer for asset allocation

If you have your portfolio divided into 4 parts then the 5% rule might air sense but if you slice and dice more and your international small cap allocation is only 5% to begin with then it wouldn't make much sense. You could set ranges for each:

Large Cap - 30% (rebalance if out of this range: 26-34%)
Small Cap - 15% (rebalance if out of this range: 13-17%)

Come up with a plan that makes sense to you and makes sense for your AA and then stick to it even if it means selling some shares of a holding that has sky rocketed recently.
The way I invest my money is not the right way to invest, it's the right way for ME to invest.
dbr
Posts: 33859
Joined: Sun Mar 04, 2007 9:50 am

Re: Help me understand how rebalancing affects compounding

Post by dbr »

Ryan2390 wrote:Interesting replies.

So the 5% band seems to be pretty popular. Is this for each asset class or stocks as a whole?

I divide my stocks into 4 asset classes- Large US, Small US, Total International (mostly developed markets), and Emerging Markets.

Should I be re-balancing when one of those asset classes exceeds 5% of its original weight, or when my entire equity portion exceeds 5% of my portfolio?
A common suggestion that is a little more nuanced is that one rebalances the smaller of when an asset allocation is more than 5% of portfolio value away from target or when an asset is more than 25% away from its own target.

For example, if you target emerging markets at 10% of portfolio, then you would rebalance if above 12.5% or below 7.5%. If equities are 60% of portfolio, you would rebalance if below 55% or above 65%.

Most commonly asset allocation is a portfolio level exercise. That means one assigns asset weights as a fraction of the whole portfolio. If you want International to be 40% of stocks and stocks 60% of assets, then you just say international is 24% of the portfolio. A more complicated scheme of rebalancing is to run a hierarchical allocation where certain fractions are assigned to stocks and bonds, etc., and then within these certain fractions are assigned to subclasses. That becomes complicated as one must calculate balancing of sub-classes and then balancing of classes. I doubt there is anyone that goes that far, but maybe people do.

If you would rather use the prime number philosophy of investment management, then 7% can be used rather than 5%, and perhaps 23% rather than 25%, for rebalancing. As far as asset allocation, it is admittedly difficult to find four prime numbers that add to 100% to divide up your four stock asset classes. 3, 7, 43, and 47 work, so 47% large US, 43% Total International, 7% Small US, and 3% EM is a nice portfolio, if speaking of stocks only. Not everyone by a long shot follows this method.

There are people who devise bands in proportion to volatility so that highly volatile asset classes are not rebalanced too frequently.

It is more important that rebalancing is a plan so that one does not respond irrationally to market conditions than it is important what plan it is exactly.
umfundi
Posts: 3361
Joined: Tue Jun 07, 2011 5:26 pm

Re: Help me understand how rebalancing affects compounding

Post by umfundi »

VictoriaF wrote: - Do you rebalance every several minutes as large funds and endowments do? It's too expensive and impractical for individual investors.
- Do you rebalance when stocks rise or decline by some percentage? How do you determine that percentage? There is some momentum in the market movements and by focusing on percentages you would lose some potential gains.
- Do you rebalance by calendar, e.g., on 5 January every year? Your rebalancing bonus may be very small because by that time your assets will have moved back towards the mean.
- ...

Victoria
Or:

- Do you invest in a Vanguard LifeStrategy fund and cross rebalancing off the list of things to worry about?

Here is a previous thread on rebalancing started by Larry Swedroe:
http://www.bogleheads.org/forum/viewtop ... 7#p1812147

It seems to me that allowing your Asset Allocation to drift by not rebalancing is somewhat illogical. How and why did you decide on the beginning allocation?

Timing your rebalancing by whatever means (bands, etc.) is, in my book, timing. But, in any event, set a plan and stick to it.

It was mentioned at BH2013 that rebalancing across similar assets (like classes of equities) is less the point than rebalancing across dissimilar less-correlated assets like equities and bonds. It is a simple thought experiment to understand that rebalancing dampens fluctuations in the total portfolio, and therefore reduces risk.

Keith
Déjà Vu is not a prediction
staythecourse
Posts: 6993
Joined: Mon Jan 03, 2011 9:40 am

Re: Help me understand how rebalancing affects compounding

Post by staythecourse »

umfundi wrote:It was mentioned at BH2013 that rebalancing across similar assets (like classes of equities) is less the point than rebalancing across dissimilar less-correlated assets like equities and bonds. It is a simple thought experiment to understand that rebalancing dampens fluctuations in the total portfolio, and therefore reduces risk.
Diversification benefit/ rebalancing bonus is most effective with assets that have 1. Low correleations and 2. High volatility. Large cap U.S. vs. small cap have neither so not so much benefit. U.S. equity vs. bonds has low correlation, but one has low volatility so you will be (over the long run) rebalancing to mantain a static volatility level, but will be decreasing return. Stocks and gold or commodities or timber would be ideal as they have low correlation and high volatility. The diversification benefit would be maximized in this situation. Now you have to believe in gold, commodities, timber, etc... to want to do it.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle
umfundi
Posts: 3361
Joined: Tue Jun 07, 2011 5:26 pm

Re: Help me understand how rebalancing affects compounding

Post by umfundi »

staythecourse wrote:
umfundi wrote:It was mentioned at BH2013 that rebalancing across similar assets (like classes of equities) is less the point than rebalancing across dissimilar less-correlated assets like equities and bonds. It is a simple thought experiment to understand that rebalancing dampens fluctuations in the total portfolio, and therefore reduces risk.
Diversification benefit/ rebalancing bonus is most effective with assets that have 1. Low correleations and 2. High volatility. Large cap U.S. vs. small cap have neither so not so much benefit. U.S. equity vs. bonds has low correlation, but one has low volatility so you will be (over the long run) rebalancing to mantain a static volatility level, but will be decreasing return. Stocks and gold or commodities or timber would be ideal as they have low correlation and high volatility. The diversification benefit would be maximized in this situation. Now you have to believe in gold, commodities, timber, etc... to want to do it.

Good luck.
That depends on why one chooses to rebalance.

I, for one, choose to rebalance to maintain my overall AA, reduce total portfolio volatility and therefore manage and reduce risk.

Sure, there is a potential bonus if you rebalance between fluctuating investments that are not correlated. That is not my goal, though it is a nice side benefit.

Keith
Déjà Vu is not a prediction
Topic Author
Ryan2390
Posts: 17
Joined: Mon Sep 16, 2013 12:07 pm

Re: Help me understand how rebalancing affects compounding

Post by Ryan2390 »

So it seems what some of you are saying is that the benefits of rebalancing are maximized when rebalancing between uncorrelated asset classes.

So of my asset classes, what if I only rebalanced between US equities, Emerging Markets, and High Yield/Intermediate Munis? (in other words, don't rebalance between US large/small cap or Developed Markets Large Cap which is most of TISM).
staythecourse
Posts: 6993
Joined: Mon Jan 03, 2011 9:40 am

Re: Help me understand how rebalancing affects compounding

Post by staythecourse »

Ryan2390 wrote:So it seems what some of you are saying is that the benefits of rebalancing are maximized when rebalancing between uncorrelated asset classes.

So of my asset classes, what if I only rebalanced between US equities, Emerging Markets, and High Yield/Intermediate Munis? (in other words, don't rebalance between US large/small cap or Developed Markets Large Cap which is most of TISM).
Just to be clear so we haven't debated so much as to confuse the basics: Rebalancing is an intentional market timing move for the purposes of MANTAINING a set volatility to one's portfolio. The reason I say intentional market timing is it is based on the belief that prices are not so random as not to have a reversion to mean otherwise one would keep rebalancing out of the highest returning asset class into a perrennial loser.

The in depth discussion is regarding the diversification benefit/ rebalancing bonus which is the thought of trying to increase returns by ghe act of rebalancing. That is maximized with low correlation assets AND/OR high volatility assets.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle
User avatar
dratkinson
Posts: 5102
Joined: Thu Jul 26, 2007 6:23 pm
Location: Centennial CO

Re: Help me understand how rebalancing affects compounding

Post by dratkinson »

VictoriaF wrote: - Importantly, mean reversion is not a natural law.

Victoria
Why not?

My understanding is that reversion to the mean (RTM) is natural. Why? Because, it is my belief:

--Markets grow uniformly exponentially based on human consumption---the more people, the more consumption. (Predicted consumption = need for food, clothing, housing,... / mo * population).
--In a disturbed market environment, market consumption grows erratically.
--*Market noise disturbs the market by accelerating/depressing consumption by segment or total market. (Severe or long term market disturbances can result in panic buying/hoarding/shortages, reduced population/refugee population influx: 1906 San Francisco earthquake, WWI/II, 1980 Mariel boatlift,...)
--When market noise reverses (crisis solution/overcome/forgotten), then the market returns to normal predicted consumption levels for new population. (Market need temporarily depressed until stored/hoarded supplies consumed: Y2K preparations.)
--RTM can happen quickly (artificial budget crisis: most recent and Clinton's), or over many years (California's '73 and '79 gasoline shortages, '77 Community Redevelopment Act/housing crisis).
  • *In thinking about this I am coming to understand that market noise is any disaster (natural/man made, real/imaginary, anticipated/actual, large/small) that disturbs normal population consumption patterns.

    Any population changes resulting from the disaster reset consumption levels and baseline for market trend line.


In summary, this is my rough understanding of market RTM:

--Normal human (exponential growth) consumption creates the market trend line.
--Market trend line + market noise = market deviations from trend line and possible population changes.
--Market deviation - market noise (which caused it) = restores normal consumption patterns for new population baseline.
--Restoration of normal consumption patterns restores normal market trend line (RTM).
--Index funds (total/segment) capture market consumption patterns and therefore also responds to RTM.


Barring any flaws in my logic, it seems to me that RTM is as natural as market growth based on human consumption/population growth.

I am willing to be shown the error in my logic.
Or you may develop this idea and cite me as a reference when you accept your Nobel. :)

Question: Could the BH forum win a Nobel prize for service to investors?



Edit: Added consideration for effects of population change on consumption baseline.
Last edited by dratkinson on Mon Oct 21, 2013 8:34 am, edited 2 times in total.
d.r.a., not dr.a. | I'm a novice investor, you are forewarned.
YDNAL
Posts: 13774
Joined: Tue Apr 10, 2007 4:04 pm
Location: Biscayne Bay

Re: Help me understand how rebalancing affects compounding

Post by YDNAL »

nisiprius wrote:
VictoriaF wrote:
nisiprius wrote:d) In terms of pure risk, measured in almost any way, the difference between, say 50/50 and 60/40 is awfully small, This is shown by the fact that Vanguard offers only four LifeStrategy funds. Rebalancing to hold portfolio risk steady isn't something that needs to be done very precisely or very often.
I'd also add that if you are rebalancing by price, it's much more difficult to do just when you need it the most. For example, the best time to rebalance was March 2009, but at that time even the most weathered investors were reexamining their strategies.
Notice, however, that if you were rebalancing at the end of the year, well, neither year-end 2008 nor year-end 2009 happened to fall anywhere near the bottom. There is a very dangerous mental trap in which, in thinking about rebalancing, one imagines that it magically occurs at the bottom.
This comes up regularly. If we rebalance to maintain risk, the *calendar* has nothing to do with it.

If we held ONLY Vanguard S&P 500 and Total Bond (for simplicity), rebalancing would take place anytime a 5% cumulative move takes place (or 10%, or what floats your boat).

Code: Select all

DATE	     VBMFX	   VFINX	  CHANGE
4/1/2008	 $10.21 	$126.12 	
6/26/2008	$10.01 	$118.12 	 -6.34%
7/15/2008	$10.06 	$111.94 	-11.24%
9/29/2008	 $9.99 	$101.85 	-19.24%
10/7/2008	 $9.96 	 $91.74 	-27.26%
10/9/2008	 $9.78 	 $83.84 	-33.52%
11/19/2008	$9.77 	 $74.58 	-40.87%
3/2/2009	 $10.03 	 $64.83 	-48.60%
Landy | Be yourself, everyone else is already taken -- Oscar Wilde
User avatar
JoMoney
Posts: 9919
Joined: Tue Jul 23, 2013 5:31 am

Re: Help me understand how rebalancing affects compounding

Post by JoMoney »

Rebalancing is critical to maintaining your risk profile. The stock market is very risky. To mitigate that risk it's prudent to have a portfolio that includes things not in the stock market... since most of us consider money the main thing at risk, the stock portfolio should probably be hedged with cash/bonds/CD's as opposed to timber or gold unless we have some expectation of needing those commodities.
As far as "Return To the Mean" and using correlations to rebalance and achieve a higher return, I don't believe it happens with any kind of calculable precision, certainly nothing that you could time or build a formula around... which may be problematic if you're aiming for a Nobel prize.
The Nobel laureates running LTCM made some pretty big bets expecting that the market would mean-revert in an arbitrage gap that had to happen in an "efficient market", unfortunately that gap didn't close in a time frame convenient for them.
Other types of Hedge Funds attempt to use statistical arbitrage to profit from mean-reversion, but their success is questionable. A lot of really smart people out in the market trying to game it for profit, but one thing is certain: the end result over any period of time is that the aggregate sum of all investors will not earn more from the market then what the market itself returns. The "bonus" isn't being created by the market return, so it must be achieved by skillfully out-trading your market trading competitors, and I question the outcome.
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham
Post Reply