Explain why Rebalancing isn't Market Timing?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
rbaldini
Posts: 900
Joined: Mon Mar 23, 2015 3:20 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by rbaldini » Fri Dec 09, 2016 9:52 am

Oicuryy wrote: Rebalancing when the portfolio is out of balance is market timing. (OP's scenario 2)
No, it isn't. Again, by any common definition of market timing, putting your portfolio back in balance when it is unbalanced is not market timing.

protagonist
Posts: 5368
Joined: Sun Dec 26, 2010 12:47 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by protagonist » Fri Dec 09, 2016 10:14 am

avalpert wrote:
sambb wrote:obviously a lot of justification herein for reblanacing not being timing.
That's funny, what I think is obvious is a lot of people here trying to conflate two distinct concepts (what are labeled here rebalancing and market timing) in an effort to dismiss one (rebalancing) without addressing it directly because everyone is ok dismissing the other (market timing).
I have nothing against rebalancing. I think it is a strategy that makes a lot of sense for a lot of people, and if nothing else, it forces discipline, which is a good thing. Without rebalancing one's portfolio could easily eventually become "stock-heavy". There is also historical data (for what that is worth) that it is a profitable strategy.

I just think it is largely an argument of semantics. Technically, by the Wikipedia definition, rebalancing is not market timing, but I think it is fair to view it as a form of market timing (since there is much overlap- you are buying low and selling high), that has worked historically.

Whether it is a wise strategy or not is unknown, since we have not seen the sort of protracted slow meltdown of the USA market that has happened elsewhere (Japan being the best known example), and a mere 100 years or so of USA data is not statistically reliable enough to make assumptions about the next 20 or 30 or 50, even if you believe that the past is predictive of the future.

"Market timing" (by Wikipedia definition) tends to be a dirty word , and it has been shown to usually fail...often it is motivated by fear and panic or "irrational exuberance". "Rebalancing" has, to date, worked, and it is a disciplined approach, not an emotional one. Yet whether or not to call it a form of "market timing" that might work is, to me, splitting hairs. It doesn't really matter what you call it, unless the phrase "market timing" leaves a really bad taste in your mouth.
Last edited by protagonist on Fri Dec 09, 2016 10:15 am, edited 1 time in total.

tphp99
Posts: 330
Joined: Sun Jan 03, 2010 8:47 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by tphp99 » Fri Dec 09, 2016 10:15 am

TomatoTomahto wrote:You say tomato, I say tomahto. Call me anything you want, except late to dinner.
They're just words right?

Fact is no one knows what the market will do. I never understood the importance of rebalancing. AA is at best arbitrary, no? There's no guarantee that your nest egg will grow X% in 5, 10...30 yrs.

"Wait and Hope" - Count of Monte Cristo

Da5id
Posts: 2035
Joined: Fri Feb 26, 2016 8:20 am

Re: Explain why Rebalancing isn't Market Timing?

Post by Da5id » Fri Dec 09, 2016 10:22 am

tphp99 wrote:
TomatoTomahto wrote:You say tomato, I say tomahto. Call me anything you want, except late to dinner.
They're just words right?

Fact is no one knows what the market will do. I never understood the importance of rebalancing. AA is at best arbitrary, no? There's no guarantee that your nest egg will grow X% in 5, 10...30 yrs.
AA is not arbitrary, it is a means of risk management. The distinction for someone in/nearing retirement between have 100% stocks and losing half their money in a 50% stock decline and having 50% stocks and losing 25% of their money is pretty stark (ignoring what bonds might be doing for simplicity). My asset allocation spreadsheet has a little table in the corner that shows results for a 10-50% stock slide. I look at it occasionally to restrain any irrational exuberance I might be feeling (and yes, stocks could fall more than 50% and stay down forever ala Japan, etc).

Yes, you don't know what will come. And over the long haul stocks are better, so more stocks is better. But we each get to live one life and for those who were nearing retirement 2008 was quite a traumatic event (reading the threads at the time, I wasn't near retirement then though I may be now).

rbaldini
Posts: 900
Joined: Mon Mar 23, 2015 3:20 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by rbaldini » Fri Dec 09, 2016 10:33 am

protagonist wrote: I just think it is largely an argument of semantics. Technically, by the Wikipedia definition, rebalancing is not market timing, but I think it is fair to view it as a form of market timing (since there is much overlap- you are buying low and selling high), that has worked historically.
To repeat: this assumes that market timers are always trying to buy low and sell high. That's just false. Some market timing strategies say to buy more when the market is above a 10-month moving average - which means to buy more when the stock market has recently *risen*. There are many kinds of market timing. The thing they all have in common is that they try to use market signals to make predictions about future returns. Rebalancing doesn't.

And the purpose of rebalancing is not to buy low and sell high. It's to maintain a risk-reward tradeoff that you are comfortable with. If you didn't rebalance, you wouldn't be able to do this - your portfolio would drift to ever high stock allocation, completely out of your control. If you want control over how risky your portfolio is, you need to rebalance.

User avatar
oldzey
Posts: 987
Joined: Sun Apr 13, 2014 8:38 pm
Location: Land of Lincoln

Re: Explain why Rebalancing isn't Market Timing?

Post by oldzey » Fri Dec 09, 2016 11:08 am

Rebalancing is adjusting your portfolio based upon what you have stated in your IPS.

Market timing is adjusting your portfolio based on speculation of what Mr. Market might do in the future.
"The broker said the stock was 'poised to move.' Silly me, I thought he meant up." ― Randy Thurman

User avatar
Oicuryy
Posts: 1236
Joined: Thu Feb 22, 2007 10:29 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by Oicuryy » Fri Dec 09, 2016 11:09 am

rbaldini wrote:
Oicuryy wrote: Rebalancing when the portfolio is out of balance is market timing. (OP's scenario 2)
No, it isn't. Again, by any common definition of market timing, putting your portfolio back in balance when it is unbalanced is not market timing.
Others have posted the common definition of market timing. Here is Investopedia's definition of rebalancing.
Rebalancing is the process of realigning the weightings of a portfolio of assets. Rebalancing involves periodically buying or selling assets in a portfolio to maintain an original desired level of asset allocation.
And here is Merriam-Webster's definition of periodically.
1: at regular intervals of time
2: from time to time : frequently
Investopedia goes on to say (emphasis added),
For example, say an original target asset allocation was 50% stocks and 50% bonds. If the stocks performed well during the period, it could have increased the stock weighting of the portfolio to 70%. The investor may then decide to sell some stocks and buy bonds to get the portfolio back to the original target allocation of 50/50.
Rebalancing means realigning the weightings at regular intervals of time. Realigning the weightings whenever the market moves them out of alignment is a form of market timing.
Money is fungible | Abbreviations and Acronyms

rbaldini
Posts: 900
Joined: Mon Mar 23, 2015 3:20 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by rbaldini » Fri Dec 09, 2016 11:15 am

Oicuryy wrote:
rbaldini wrote:
Oicuryy wrote: Rebalancing when the portfolio is out of balance is market timing. (OP's scenario 2)
No, it isn't. Again, by any common definition of market timing, putting your portfolio back in balance when it is unbalanced is not market timing.
Others have posted the common definition of market timing. Here is Investopedia's definition of rebalancing.
Rebalancing is the process of realigning the weightings of a portfolio of assets. Rebalancing involves periodically buying or selling assets in a portfolio to maintain an original desired level of asset allocation.
And here is Merriam-Webster's definition of periodically.
1: at regular intervals of time
2: from time to time : frequently
Investopedia goes on to say (emphasis added),
For example, say an original target asset allocation was 50% stocks and 50% bonds. If the stocks performed well during the period, it could have increased the stock weighting of the portfolio to 70%. The investor may then decide to sell some stocks and buy bonds to get the portfolio back to the original target allocation of 50/50.
Rebalancing means realigning the weightings at regular intervals of time. Realigning the weightings whenever the market moves them out of alignment is a form of market timing.
Nope.

From wiki:
"Market timing is the strategy of making buy or sell decisions of financial assets (often stocks) by attempting to predict future market price movements."

Rebalancing at non-fixed intervals does not attempt to predict future market price movements. Therefore it is not market timing.

This is not difficult. And yet I've said this maybe ten times now?

Remember, the purpose of rebalancing is to regain balance. It makes sense to do this when you are off balance; that is the whole idea. I don't understand how this is controversial. To say that doing this irregularly somehow makes it market timing is ridiculous.
Oicuryy wrote: If the stocks performed well during the period
Nowhere in that example does it say that the period has to be predetermined and eternally fixed.
Last edited by rbaldini on Fri Dec 09, 2016 11:26 am, edited 3 times in total.

Da5id
Posts: 2035
Joined: Fri Feb 26, 2016 8:20 am

Re: Explain why Rebalancing isn't Market Timing?

Post by Da5id » Fri Dec 09, 2016 11:19 am

Oicuryy wrote: Rebalancing means realigning the weightings at regular intervals of time. Realigning the weightings whenever the market moves them out of alignment is a form of market timing.
I have to say this is an odd thread. You are citing Investopedia as a source. Here is what they say about market timing:
Market timing is the act of moving in and out of the market or switching between asset classes based on using predictive methods such as technical indicators or economic data. Because it is extremely difficult to predict the future direction of the stock market, investors who try to time the market, especially mutual fund investors, tend to underperform investors who remain invested.
What "predictive method" is involved here specifically, given that your asset allocation isn't changing. There is no input of technical indicators or economic data, no predictions that stocks will do better over the coming year than bonds.

MathWizard
Posts: 2959
Joined: Tue Jul 26, 2011 1:35 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by MathWizard » Fri Dec 09, 2016 11:39 am

HomerJ wrote:
Dirghatamas wrote:Based on this thesis (which is not time variant), I invested in a passive world/ global capitalized stocks only portfolio and have done precisely the same come rain or shine in this time frame. The investment philosophy requires no market timing or rebalancing or any such active strategies. Simply buy every month at the price stocks are at. Don't try to market time or guess the future or worry about whether stocks are cheap vs. bonds or US stocks are expensive vs. International or whatever. Focus energy on career/life and just buy the same way every month. All these arguments about rebalancing being different from market timing have never made much sense to me, so I ignore that stuff.
Heh, an Asset Allocation of 100% stocks doesn't require any rebalancing. That is indeed one solution.
Hey, You stole my solution! :wink:

More seriously, the OP stated somewhere along the line that he should have asked whether rebalancing based on AA with bands
hurts returns. I think that I remember seeing something about infrequent rebalancing being better (historically) than frequent
rebalancing even ignoring trading costs, but I don't remember the source.

Does anyone have a reference for a study backtesting returns when
rebalancing based on AA versus rebalancing based on time.
rebalancing based on AA versus never changing your AA
rebalancing based on time versus never changing your AA.

To me, the reason for rebalancing is to control your behavior.
E.g. Avoiding selling when stocks drop, because you panic because of a huge drop in your portfolio
due to having too high of the more volatile asset.

That is how rebalancing help returns, it's not to make you more money, it's to help you to "Not lose money."
Last edited by MathWizard on Fri Dec 09, 2016 11:53 am, edited 1 time in total.

User avatar
midareff
Posts: 5712
Joined: Mon Nov 29, 2010 10:43 am
Location: Biscayne Bay, South Florida

Re: Explain why Rebalancing isn't Market Timing?

Post by midareff » Fri Dec 09, 2016 11:47 am

aboose wrote:Midareff, thanks for your input. That said, my last paragraph covers that. Just because mentally you are doing it for different reasons doesn't mean that rebalancing per "balance bands" isn't mechanically the same as buying and selling in reaction to market movements.

Re-balancing per balance bands is a purely mathematical exercise done on a scheduled interval. There isn't an emotional component which brings one's own judgement to the forefront. ... and that judgement is a component that has been shown to compromise maximum performance.

rbaldini
Posts: 900
Joined: Mon Mar 23, 2015 3:20 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by rbaldini » Fri Dec 09, 2016 12:01 pm

MathWizard wrote: More seriously, the OP stated somewhere along the line that he should have asked whether rebalancing based on AA with bands
hurts returns. I think that I remember seeing something about infrequent rebalancing being better (historically) than frequent
rebalancing even ignoring trading costs, but I don't remember the source.
I haven't looked at the data, but some thought suggests that this basically *must* be true. Stocks have historically had a better return than bonds. Therefore, most rebalancings involve selling stock and buying bonds - which means getting rid of a higher returning asset in exchange for a lower performing one. This *should* lead to lower long term return, in the same way that holding bonds at all should lead to lower long term return.

This fact clearly demonstrates that, indeed, the point of rebalancing isn't to maximize return. If you wanted to do that, just hold 100% stock. No need to rebalance.

Similarly, the purpose isn't exclusively to minimize risk, because sometimes you're selling bonds to buy more stock. If you wanted to minimize returns, put all your money in a savings account.

The purpose is to *maintain your portfolio at the risk-return balance you are comfortable with*. If you didn't rebalance, your portfolio would slowly get more stock heavy, in a way that is dictated by the market - i.e. out of your control. If you want a 60/40 allocation, then the only way to maintain that is to rebalance from time to time - market forces alone will push you away from that.
Last edited by rbaldini on Fri Dec 09, 2016 12:08 pm, edited 1 time in total.

User avatar
Oicuryy
Posts: 1236
Joined: Thu Feb 22, 2007 10:29 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by Oicuryy » Fri Dec 09, 2016 12:03 pm

Da5id wrote:What "predictive method" is involved here specifically, given that your asset allocation isn't changing. There is no input of technical indicators or economic data, no predictions that stocks will do better over the coming year than bonds.
But there is a technical indicator - a measure of the recent relative performance of two asset classes. The prediction is that the trade will enable the portfolio to achieve the desired risk-adjusted return.

Ron
Money is fungible | Abbreviations and Acronyms

rbaldini
Posts: 900
Joined: Mon Mar 23, 2015 3:20 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by rbaldini » Fri Dec 09, 2016 12:07 pm

Oicuryy wrote:
Da5id wrote:What "predictive method" is involved here specifically, given that your asset allocation isn't changing. There is no input of technical indicators or economic data, no predictions that stocks will do better over the coming year than bonds.
But there is a technical indicator - a measure of the recent relative performance of two asset classes. The prediction is that the trade will enable the portfolio to achieve the desired risk-adjusted return.

Ron
To rebalance is to return to the desired risk-adjusted return that you decided you wanted in the past. It doesn't involve any changing predictions about future market returns - your estimate (if you have one) of the future needn't change to justify rebalancing. That is the case for me: I basically see the market as an uncorrelated random walk, meaning that the distribution of future returns isn't informed by recent behavior. So it's not market timing.

Da5id
Posts: 2035
Joined: Fri Feb 26, 2016 8:20 am

Re: Explain why Rebalancing isn't Market Timing?

Post by Da5id » Fri Dec 09, 2016 12:10 pm

Oicuryy wrote:
Da5id wrote:What "predictive method" is involved here specifically, given that your asset allocation isn't changing. There is no input of technical indicators or economic data, no predictions that stocks will do better over the coming year than bonds.
But there is a technical indicator - a measure of the recent relative performance of two asset classes. The prediction is that the trade will enable the portfolio to achieve the desired risk-adjusted return.
IMHO making the choice to not re-balance is by your logic likewise market timing. Perhaps eating lobster instead of a hamburger is market timing too, as is buying a car. By the argument here, you are engaging in market timing continuously by purchasing a Vanguard LifeStrategy fund and holding it forever. If it pleases you to think so, great. But it is not the common meaning of market timing most folks here use.

aboose
Posts: 76
Joined: Wed Jun 08, 2016 11:54 am

Re: Explain why Rebalancing isn't Market Timing?

Post by aboose » Fri Dec 09, 2016 12:15 pm

Da5id wrote:IMHO making the choice to not re-balance is by your logic likewise market timing. Perhaps eating lobster instead of a hamburger is market timing too, as is buying a car. By the argument here, you are engaging in market timing continuously by purchasing a Vanguard LifeStrategy fund and holding it forever. If it pleases you to think so, great. But it is not the common meaning of market timing most folks here use.
I actually agree with you, but in my mind the distinction comes from the desired result.

We buy a car because we desire a nice thing. However, we rebalance because we desire to reduce risk while maximizing returns. The point is that rebalancing via these % band allocations necessarily reduces long term returns, as opposed to the scenario 1 of time balanced portfolios or even not rebalancing at all.

I think I would be satisfied if I could understand the logical trade off of returns and risk between not rebalancing, rebalancing based on time (market agnostic), and rebalancing on % bands (market reaction / timing).

To me, not rebalancing should > rebalancing based on time which should > rebalancing based on % bands from a risk/return standpoint.

rbaldini
Posts: 900
Joined: Mon Mar 23, 2015 3:20 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by rbaldini » Fri Dec 09, 2016 12:24 pm

aboose wrote: We buy a car because we desire a nice thing. However, we rebalance because we desire to reduce risk while maximizing returns. The point is that rebalancing via these % band allocations necessarily reduces long term returns, as opposed to the scenario 1 of time balanced portfolios or even not rebalancing at all.

I think I would be satisfied if I could understand the logical trade off of returns and risk between not rebalancing, rebalancing based on time (market agnostic), and rebalancing on % bands (market reaction / timing).
So what, in practice, is the difference between scenario 1 and 2? It depends.

If scenario 1 says "rebalance weekly", well then you are going to rebalance often - probably more often than scenario 2 with common bands. This means scenario 1 will more actively maintain your allocation at a risk-return tradeoff that you like. Scenario 2 will probably make you slightly more money in the long run, because you will more often run up higher stock allocation... at the cost of slightly greater risk during those times.

If instead scenario 1 says "rebalance every 5 years", then the opposite is probably true. Scenario 2 would probably rebalance more often than Scenario 1, meaning it would keep you closer to the risk-reward tradeoff that you like. Scenario 1 would probably end up with more money, at greater risk.

The purpose of rebalancing is to maintain the risk-reward tradeoff you like. Do it whenever you feel you are too far off. How far is "too far"? That's up to you.

Copernicus
Posts: 322
Joined: Thu May 16, 2013 4:38 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by Copernicus » Fri Dec 09, 2016 12:30 pm

HomerJ wrote: Reacting to what has already happened is rebalancing.

Market-timing is making changes based on what you THINK will happen in the future.
+ 1

boglephreak
Posts: 441
Joined: Fri Apr 22, 2016 5:16 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by boglephreak » Fri Dec 09, 2016 12:43 pm

aboose wrote:
Da5id wrote:IMHO making the choice to not re-balance is by your logic likewise market timing. Perhaps eating lobster instead of a hamburger is market timing too, as is buying a car. By the argument here, you are engaging in market timing continuously by purchasing a Vanguard LifeStrategy fund and holding it forever. If it pleases you to think so, great. But it is not the common meaning of market timing most folks here use.
I actually agree with you, but in my mind the distinction comes from the desired result.

We buy a car because we desire a nice thing. However, we rebalance because we desire to reduce risk while maximizing returns. The point is that rebalancing via these % band allocations necessarily reduces long term returns, as opposed to the scenario 1 of time balanced portfolios or even not rebalancing at all.

I think I would be satisfied if I could understand the logical trade off of returns and risk between not rebalancing, rebalancing based on time (market agnostic), and rebalancing on % bands (market reaction / timing).

To me, not rebalancing should > rebalancing based on time which should > rebalancing based on % bands from a risk/return standpoint.
why do you keep saying people rebalance to maximize returns? many people have said that is not the purpose, and it certainly is not the purpose for my rebalancing. i rebalance to keep my holdings in line with what i think is the most risk i can handle, not to maximize returns.

Da5id
Posts: 2035
Joined: Fri Feb 26, 2016 8:20 am

Re: Explain why Rebalancing isn't Market Timing?

Post by Da5id » Fri Dec 09, 2016 12:48 pm

aboose wrote: We buy a car because we desire a nice thing.
I meant as in the decision to buy a car is a decision that you can afford it. Which makes assumptions about future results/returns. Which is market timing. By a logical extension of an odd definition of market timing that I'm not buying into.

User avatar
HomerJ
Posts: 11574
Joined: Fri Jun 06, 2008 12:50 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by HomerJ » Fri Dec 09, 2016 12:58 pm

tphp99 wrote:AA is at best arbitrary, no?
AA is not arbitrary. Why would you think that?

Let's say I retire today, and tomorrow the market starts a 50% decline, and doesn't recover for 5 years.

Do you really think AA doesn't matter in that scenario?

User avatar
HomerJ
Posts: 11574
Joined: Fri Jun 06, 2008 12:50 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by HomerJ » Fri Dec 09, 2016 1:00 pm

rbaldini wrote:
protagonist wrote: I just think it is largely an argument of semantics. Technically, by the Wikipedia definition, rebalancing is not market timing, but I think it is fair to view it as a form of market timing (since there is much overlap- you are buying low and selling high), that has worked historically.
To repeat: this assumes that market timers are always trying to buy low and sell high. That's just false. Some market timing strategies say to buy more when the market is above a 10-month moving average - which means to buy more when the stock market has recently *risen*. There are many kinds of market timing. The thing they all have in common is that they try to use market signals to make predictions about future returns. Rebalancing doesn't.

And the purpose of rebalancing is not to buy low and sell high. It's to maintain a risk-reward tradeoff that you are comfortable with. If you didn't rebalance, you wouldn't be able to do this - your portfolio would drift to ever high stock allocation, completely out of your control. If you want control over how risky your portfolio is, you need to rebalance.
This. It's really this simple. Market-timers CHANGE their AA. Rebalancers MAINTAIN their AA.
Last edited by HomerJ on Fri Dec 09, 2016 1:06 pm, edited 1 time in total.

User avatar
HomerJ
Posts: 11574
Joined: Fri Jun 06, 2008 12:50 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by HomerJ » Fri Dec 09, 2016 1:06 pm

aboose wrote:However, we rebalance because we desire to reduce risk while maximizing returns.
Why do you keep saying that when we keep correcting you? Rebalancing is ONLY about managing risk, not about maximizing returns.
The point is that rebalancing via these % band allocations necessarily reduces long term returns
Yes, you are correct. But we already know that. You're the one stating something that no one here has said, and then proving that what YOU said is wrong.

FactualFran
Posts: 712
Joined: Sat Feb 21, 2015 2:29 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by FactualFran » Fri Dec 09, 2016 1:46 pm

aboose wrote: We buy a car because we desire a nice thing. However, we rebalance because we desire to reduce risk while maximizing returns. The point is that rebalancing via these % band allocations necessarily reduces long term returns, as opposed to the scenario 1 of time balanced portfolios or even not rebalancing at all.
I don't know who the "we" are here, but it does not include me nor, I suspect, many others. I buy a car because I have transportation needs that are met by having a car, not because I desire a nice thing.

I rebalance for a reason given in the Vanguard Best Practices For Portfolio Rebalancing document.
The primary goal of a rebalancing strategy is to minimize risk relative to a target asset allocation, rather than to maximize returns. Over time, asset classes produce different returns that can change the portfolio’s asset allocation. To recapture the portfolio’s original risk-and-return characteristics, the portfolio should therefore be rebalanced.
I, and I suspect many others, do not rebalance to maximize returns.

The effect of rebalancing on returns depends on the long-term returns of the investment being rebalanced. If the investments have similar long-term returns, it is possible for there to be a Rebalancing Bonus, as described in the The rebalancing bonus: Theory and Practice article by William J. Bernstein.

KarenC
Posts: 72
Joined: Mon Apr 27, 2015 7:25 am

Re: Explain why Rebalancing isn't Market Timing?

Post by KarenC » Fri Dec 09, 2016 1:54 pm

HomerJ wrote:This. It's really this simple. Market-timers CHANGE their AA. Rebalancers MAINTAIN their AA.
Not so simple. I change my AA on a monthly basis but I am not a market-timer. (I track a target-date fund.)

(Edited to correct the attribution.)
Last edited by KarenC on Sun Dec 11, 2016 7:57 am, edited 1 time in total.
"How much you know is less important than how clearly you understand where the borders of your ignorance begin." — Jason Zweig

rbaldini
Posts: 900
Joined: Mon Mar 23, 2015 3:20 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by rbaldini » Fri Dec 09, 2016 2:04 pm

KarenC wrote:
HomerJ wrote:
rbaldini wrote: This. It's really this simple. Market-timers CHANGE their AA. Rebalancers MAINTAIN their AA.
Not so simple. I change my AA on a monthly basis but I am not a market-timer. (I track a target-date fund.)
(That's a misquote of me... HomerJ said it, not me.)

Indeed. One can change their AA without making predictions about the future of the market. By definition, this would not be market timing.

User avatar
HomerJ
Posts: 11574
Joined: Fri Jun 06, 2008 12:50 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by HomerJ » Fri Dec 09, 2016 2:14 pm

KarenC wrote:
HomerJ wrote:
rbaldini wrote: This. It's really this simple. Market-timers CHANGE their AA. Rebalancers MAINTAIN their AA.
Not so simple. I change my AA on a monthly basis but I am not a market-timer. (I track a target-date fund.)
Oh come on. Yes, investing 100% in a single Vanguard Target Date Fund technically means one is constantly changing their AA.

It's very gradual change based on age, though. Just like most of us do even with our three fund portfolios. Do we really need to account for every exception?

Is it really that hard to tell the difference between someone maintaining their portfolio and someone attempting to change their portfolio because market conditions are different in some way than yesterday.

People are focused on the fact that the rebalancer DID something (buy and sell) based on what the market just did, but what they are really doing is keeping things the same (i.e. NOT making a change in their AA). Market timers also react to what the market just did, but they CHANGE their AA.
Last edited by HomerJ on Fri Dec 09, 2016 2:44 pm, edited 3 times in total.

User avatar
Que1999
Posts: 97
Joined: Mon May 30, 2016 10:27 am

Re: Explain why Rebalancing isn't Market Timing?

Post by Que1999 » Fri Dec 09, 2016 2:15 pm

This thread is making my head hurt. :|

User avatar
goingup
Posts: 3186
Joined: Tue Jan 26, 2010 1:02 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by goingup » Fri Dec 09, 2016 2:38 pm

Que1999 wrote:This thread is making my head hurt. :|
Yes. The term circular firing squad comes to mind. :D

aboose
Posts: 76
Joined: Wed Jun 08, 2016 11:54 am

Re: Explain why Rebalancing isn't Market Timing?

Post by aboose » Fri Dec 09, 2016 2:38 pm

If you were in a situation where your portfolio risk was higher than you prefer, and you were provided with two scenarios:

Rebalance based on scenario 1, to return your risk profile to what you want based on a time-criteria so therefore not market timing and therefore not hurting your returns

Rebalance based on scenario 2, to return your risk profile to what you want based on a % allocation band criteria, therefore market timing (buy low sell high) and therefore potentially hurting your returns

Why would you prefer the second one?

In a discussion about investing and timing, risk and return go hand in hand. To say you don't try to optimize your returns strategy is fallacious, because in trying to optimize your risk strategy you are also affecting your returns.

Thanks! :sharebeer

aboose
Posts: 76
Joined: Wed Jun 08, 2016 11:54 am

Re: Explain why Rebalancing isn't Market Timing?

Post by aboose » Fri Dec 09, 2016 2:42 pm

rbaldini wrote:So what, in practice, is the difference between scenario 1 and 2? It depends.

If scenario 1 says "rebalance weekly", well then you are going to rebalance often - probably more often than scenario 2 with common bands. This means scenario 1 will more actively maintain your allocation at a risk-return tradeoff that you like. Scenario 2 will probably make you slightly more money in the long run, because you will more often run up higher stock allocation... at the cost of slightly greater risk during those times.

If instead scenario 1 says "rebalance every 5 years", then the opposite is probably true. Scenario 2 would probably rebalance more often than Scenario 1, meaning it would keep you closer to the risk-reward tradeoff that you like. Scenario 1 would probably end up with more money, at greater risk.

The purpose of rebalancing is to maintain the risk-reward tradeoff you like. Do it whenever you feel you are too far off. How far is "too far"? That's up to you.
Rbaldini, this is a really good point.

Here's my thoughts -- so if we could rebalance instantaneously, such that risk would never diverge from our desired risk allocation, in my opinion that would be an perfectly optimal situation -- completely market agnostic, with no delay and no risk drifting as you mention/

However since we can't do that, we are left with the decision of picking scenario 1 and 2 -- time criteria based versus % allocation band based.

The reason they are different is because time criteria based is market agnostic. It happens at a set factor every single day/week/month/year/5 years/ whatever set time. Therefore, you are never "buying high selling low" necessarily, you are simply buying and selling to reset to your desired amount regardless of market performance of bonds and stocks.

Scenario 2 is reactive to market. It rebalances unevenly -- based on market performance. That is why it is market timing, because the criteria to trigger it are market events. As we've discussed in here, inaction / not rebalancing is also a form of market timing. The inaction and the action are the yin and yang of market timing in this situation. Whereas in scenario 1, there is no market timing, as time is the controlled variable.

User avatar
HomerJ
Posts: 11574
Joined: Fri Jun 06, 2008 12:50 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by HomerJ » Fri Dec 09, 2016 2:47 pm

aboose wrote:If you were in a situation where your portfolio risk was higher than you prefer, and you were provided with two scenarios:

Rebalance based on scenario 1, to return your risk profile to what you want based on a time-criteria so therefore not market timing and therefore not hurting your returns

Rebalance based on scenario 2, to return your risk profile to what you want based on a % allocation band criteria, therefore market timing (buy low sell high) and therefore potentially hurting your returns

Why would you prefer the second one?
Because the second one returns me to my desired risk profile right now. If I wait, the risk may appear. How is this hard to understand?

I don't want more than 50% in stocks... If it drifts up a little, I'm not going to bother fixing it. If it drifts up a lot, I'm going to fix it right then, because what if the crash I'm trying to protect myself from happens right then?

Risk is what I care most about when picking an AA, not maximizing returns.

User avatar
HomerJ
Posts: 11574
Joined: Fri Jun 06, 2008 12:50 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by HomerJ » Fri Dec 09, 2016 2:49 pm

aboose wrote:That is why it is market timing, because the criteria to trigger it are market events. As we've discussed in here, inaction / not rebalancing is also a form of market timing.
Your definitions are useless. If action and non-action is market timing, then everything is market-timing, and there's no point to even defining the term.

If you want to make up your own definitions for words, we won't be able to agree. Best wishes.

Da5id
Posts: 2035
Joined: Fri Feb 26, 2016 8:20 am

Re: Explain why Rebalancing isn't Market Timing?

Post by Da5id » Fri Dec 09, 2016 2:52 pm

aboose wrote: Scenario 2 is reactive to market. It rebalances unevenly -- based on market performance. That is why it is market timing, because the criteria to trigger it are market events. As we've discussed in here, inaction / not rebalancing is also a form of market timing. The inaction and the action are the yin and yang of market timing in this situation. Whereas in scenario 1, there is no market timing, as time is the controlled variable.
You say "it is market timing". I say "it is not market timing".

I believe that if I have a tolerance for stock of 50% +/- 5%, and rebalance when it leaves those bounds, I am not "market timing". I'm making no predictions whatsoever about the FUTURE returns. I'm blindly restoring to my target asset allocation because it has strayed outside the desired target range. I don't predict that if I sell stocks that stocks will do worse than bonds, or that if I sell bonds bonds will do worse than stock. If I changed allocation due to interest rates, or CAPE 10, etc, that to me would be market timing.

I believe that calling rebalancing market timing is intentionally obfuscatory and will make it harder to have rational discussions where the two terms are used.
Last edited by Da5id on Fri Dec 09, 2016 2:54 pm, edited 1 time in total.

dore
Posts: 107
Joined: Tue Jan 08, 2008 6:27 pm
Location: Boise, ID

Re: Explain why Rebalancing isn't Market Timing?

Post by dore » Fri Dec 09, 2016 2:53 pm

HomerJ wrote:
aboose wrote:That is why it is market timing, because the criteria to trigger it are market events. As we've discussed in here, inaction / not rebalancing is also a form of market timing.
Your definitions are useless. If action and non-action is market timing, then everything is market-timing, and there's no point to even defining the term.

If you want to make up your own definitions for words, we won't be able to agree. Best wishes.
This. Rebalancing is not market timing in the same way that an elephant is not a mailbox.

aboose
Posts: 76
Joined: Wed Jun 08, 2016 11:54 am

Re: Explain why Rebalancing isn't Market Timing?

Post by aboose » Fri Dec 09, 2016 2:54 pm

If being reactive to the market isn't market timing, then sure, I guess I agree with your incensed comments.
HomerJ wrote:Because the second one returns me to my desired risk profile right now. If I wait, the risk may appear. How is this hard to understand?

I don't want more than 50% in stocks... If it drifts up a little, I'm not going to bother fixing it. If it drifts up a lot, I'm going to fix it right then, because what if the crash I'm trying to protect myself from happens right then?

Risk is what I care most about when picking an AA, not maximizing returns.
You can easily achieve this by rebalancing more frequently without being a market timer.
Da5id wrote:You say "it is market timing". I say "it is not market timing".

I believe that if I have a tolerance for stock of 50% +/- 5%, and rebalance when it leaves those bounds, I am not "market timing". I'm making no predictions whatsoever about the FUTURE returns. I'm blindly restoring to my target asset allocation because it has strayed outside the desired target range. I don't predict that if I sell stocks that stocks will do worse than bonds, or that if I sell bonds bonds will do worse than stock. If I changed allocation due to interest rates, or CAPE 10, etc, that to me would be market timing.

I believe that calling rebalancing market timing is intentionally obfuscatory and will make it harder to have rational discussions where the two terms are used.
I agree with having a set tolerance. I don't agree with rebalancing when it leaves the bounds, and I do think that is market timing as you are being market reactive. If you rebalance on a set schedule, for instance, every day or every week, then you achieve the same goal without being market reactive. In fact, it is superior, because you are reducing drift even more.

I'm not trying to intentionally obfuscate at all. I truly do believe that what you are describing is market timing, as you are reacting to the market and trying to adjust your risk, which goes hand in hand with returns, based on a market-performance criteria. I truly believe that the only way not to be a market timer is to rebalance / buy / sell based on time, fully agnostic from market performance.
Last edited by aboose on Fri Dec 09, 2016 2:57 pm, edited 1 time in total.

rbaldini
Posts: 900
Joined: Mon Mar 23, 2015 3:20 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by rbaldini » Fri Dec 09, 2016 2:57 pm

aboose wrote: Here's my thoughts -- so if we could rebalance instantaneously, such that risk would never diverge from our desired risk allocation, in my opinion that would be an perfectly optimal situation -- completely market agnostic, with no delay and no risk drifting as you mention/
It's not market agnostic - the infinitesimal changes of each rebalance are determined by what the market did in that time.
aboose wrote: The reason they are different is because time criteria based is market agnostic.
Again, the changes you make - how much you buy and sell - will depend on what the market did over the time period you chose, whether you predetermined the interval or not. This is still not market timing, by definition.
aboose wrote: Therefore, you are never "buying high selling low" necessarily, you are simply buying and selling to reset to your desired amount regardless of market performance of bonds and stocks.
This is, indeed, the purpose of rebalancing, whether you go by scenario 1 or 2. It's the point: to get back to what you had before. The nice thing about scenario 2 is that you only go back if you've actually left. There's no sense in going back home if you haven't left home; there's no sense in rebalancing if you're not off balance. Or, as my past analogy said: the gymnast should rebalance whenever she is off balance, not every 5 seconds. Otherwise she falls.
aboose wrote: Scenario 2 is reactive to market. It rebalances unevenly -- based on market performance. That is why it is market timing, because the criteria to trigger it are market events.
If "buying and selling in a way that is reactive to the market" were the definition of market timing, you would be right. But it isn't, so you're not. Scenario 2 doesn't make predictions about future market returns, so it's not market timing.
aboose wrote: As we've discussed in here, inaction / not rebalancing is also a form of market timing.
Not if you're not making new predictions about future market returns. Which is the definition of market timing.

Pop quiz, aboose: What is the definition of market timing?
Last edited by rbaldini on Fri Dec 09, 2016 2:59 pm, edited 1 time in total.

aboose
Posts: 76
Joined: Wed Jun 08, 2016 11:54 am

Re: Explain why Rebalancing isn't Market Timing?

Post by aboose » Fri Dec 09, 2016 2:59 pm

I think the only way to illustrate my point will be to somehow do the math calculation to show what the result of scenario 1 vs 2 would be.

Thank you everyone for your input. I've learned a lot.

Da5id
Posts: 2035
Joined: Fri Feb 26, 2016 8:20 am

Re: Explain why Rebalancing isn't Market Timing?

Post by Da5id » Fri Dec 09, 2016 2:59 pm

aboose wrote: I'm not trying to intentionally obfuscate at all. I truly do believe that what you are describing is market timing, as you are reacting to the market and trying to adjust your risk, which goes hand in hand with returns, based on a market-performance criteria. I truly believe that the only way not to be a market timer is to rebalance / buy / sell based on time, fully agnostic from market performance.
I believe you are flat out wrong. But I don't believe I could possibly persuade you of that.

aboose
Posts: 76
Joined: Wed Jun 08, 2016 11:54 am

Re: Explain why Rebalancing isn't Market Timing?

Post by aboose » Fri Dec 09, 2016 3:00 pm

Da5id wrote: I believe you are flat out wrong. But I don't believe I could possibly persuade you of that.
You could if there were empirical proof that scenario 2 doesn't damage returns while maintaining similar risk profiles.

avalpert
Posts: 6313
Joined: Sat Mar 22, 2008 4:58 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by avalpert » Fri Dec 09, 2016 3:01 pm

aboose wrote:If you were in a situation where your portfolio risk was higher than you prefer, and you were provided with two scenarios:

Rebalance based on scenario 1, to return your risk profile to what you want based on a time-criteria so therefore not market timing and therefore not hurting your returns

Rebalance based on scenario 2, to return your risk profile to what you want based on a % allocation band criteria, therefore market timing (buy low sell high) and therefore potentially hurting your returns
Your 'market timing' here is a red herring - you present no evidence that scenario two potentially hurts returns in a way different from scenario one and that evidence is not going to be based on whatever you call it or imply based on presumed similarities to other strategies.

It seems you just want to save yourself the trouble of demonstrating your assertion is actually true and instead use guilt by association. If anything, the data I have seen shows the opposite - method #2 produces superior returns to method #1 (see for examplethis 2007 paper in the Journal of Financial Planning).

rbaldini
Posts: 900
Joined: Mon Mar 23, 2015 3:20 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by rbaldini » Fri Dec 09, 2016 3:03 pm

aboose wrote:I think the only way to illustrate my point will be to somehow do the math calculation to show what the result of scenario 1 vs 2 would be.
What would you be able to prove by this? Would your calculations be able to prove that, in fact, Scenario 2 secretly *does* make updated predictions about future returns? No, you wouldn't. Which means you wouldn't be able to prove that it is market timing. Because that's what market timing is.

We've already discussed that scenario 1 and 2 will give different results. And that, depending on what interval you choose for scenario 1, either scenario 1 or 2 will give higher risk and higher return. This is irrelevant to the question of whether scenario 2 is market timing. So what are you trying to achieve?

sid hartha
Posts: 122
Joined: Fri Dec 09, 2016 2:17 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by sid hartha » Fri Dec 09, 2016 3:04 pm

I think it is market timing but in a good and logical way. You are using pre established rules / guidelines to do it rather than on a hunch or gut instinct.

aboose
Posts: 76
Joined: Wed Jun 08, 2016 11:54 am

Re: Explain why Rebalancing isn't Market Timing?

Post by aboose » Fri Dec 09, 2016 3:04 pm

avalpert wrote:Your 'market timing' here is a red herring - you present no evidence that scenario two potentially hurts returns in a way different from scenario one and that evidence is not going to be based on whatever you call it or imply based on presumed similarities to other strategies.

It seems you just want to save yourself the trouble of demonstrating your assertion is actually true and instead use guilt by association. If anything, the data I have seen shows the opposite - method #2 produces superior returns to method #1 (see for examplethis 2007 paper in the Journal of Financial Planning).
I was sent this PDF, linked before, however I want to bring up two points.

1) If the above is true, and that you can use % band allocations to improve returns therefore beating the market, then why doesn't every algo trader exploit this "hack" and get infinite returns?
2) Wealth managers do not beat the market.

And I have no real agenda. If I'm wrong, I'm wrong. I just haven't been convinced.

aboose
Posts: 76
Joined: Wed Jun 08, 2016 11:54 am

Re: Explain why Rebalancing isn't Market Timing?

Post by aboose » Fri Dec 09, 2016 3:05 pm

rbaldini wrote: What would you be able to prove by this? Would your calculations be able to prove that, in fact, Scenario 2 secretly *does* make updated predictions about future returns? No, you wouldn't. Which means you wouldn't be able to prove that it is market timing. Because that's what market timing is.

We've already discussed that scenario 1 and 2 will give different results. And that, depending on what interval you choose for scenario 1, either scenario 1 or 2 will give higher risk and higher return. This is irrelevant to the question of whether scenario 2 is market timing. So what are you trying to achieve?
What I'm trying to achieve is disproving the fallacy that scenario 2 somehow provides superior risk/return to scenario 1, which seems to be logical and main reason people employ this strategy. The reason it cannot provide superior risk/return is because of the general truths we accept that risk and return are inversely correlated and also because on average, people cannot beat the market / improve their return consistently by shuffling their portfolio around.

Da5id
Posts: 2035
Joined: Fri Feb 26, 2016 8:20 am

Re: Explain why Rebalancing isn't Market Timing?

Post by Da5id » Fri Dec 09, 2016 3:08 pm

aboose wrote:
Da5id wrote: I believe you are flat out wrong. But I don't believe I could possibly persuade you of that.
You could if there were empirical proof that scenario 2 doesn't damage returns while maintaining similar risk profiles.
You don't seem to get it. What does that have to do with whether it is "market timing". Again, mechanical, pre-determined rebalancing, whether with bands or continuously/daily/monthly/yearly, doesn't act on predictions about the future.

All done, others can keep up the good fight...

boglephreak
Posts: 441
Joined: Fri Apr 22, 2016 5:16 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by boglephreak » Fri Dec 09, 2016 3:10 pm

aboose wrote:
You can easily achieve this by rebalancing more frequently without being a market timer.

. . .

I agree with having a set tolerance. I don't agree with rebalancing when it leaves the bounds, and I do think that is market timing as you are being market reactive. If you rebalance on a set schedule, for instance, every day or every week, then you achieve the same goal without being market reactive. In fact, it is superior, because you are reducing drift even more.

I'm not trying to intentionally obfuscate at all. I truly do believe that what you are describing is market timing, as you are reacting to the market and trying to adjust your risk, which goes hand in hand with returns, based on a market-performance criteria. I truly believe that the only way not to be a market timer is to rebalance / buy / sell based on time, fully agnostic from market performance.
are you arguing that if you rebalance based on a set schedule (daily, weekly, monthly, annually) then you are not market timing? but, if you rebalance based on "bands" (e.g., 5% above desired AA), you are market timing?

aboose
Posts: 76
Joined: Wed Jun 08, 2016 11:54 am

Re: Explain why Rebalancing isn't Market Timing?

Post by aboose » Fri Dec 09, 2016 3:10 pm

Thanks, I appreciate all the input. :sharebeer

@boglephreak:

Precisely

rbaldini
Posts: 900
Joined: Mon Mar 23, 2015 3:20 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by rbaldini » Fri Dec 09, 2016 3:11 pm

aboose wrote:
rbaldini wrote: What would you be able to prove by this? Would your calculations be able to prove that, in fact, Scenario 2 secretly *does* make updated predictions about future returns? No, you wouldn't. Which means you wouldn't be able to prove that it is market timing. Because that's what market timing is.

We've already discussed that scenario 1 and 2 will give different results. And that, depending on what interval you choose for scenario 1, either scenario 1 or 2 will give higher risk and higher return. This is irrelevant to the question of whether scenario 2 is market timing. So what are you trying to achieve?
What I'm trying to achieve is disproving the fallacy that scenario 2 somehow provides superior risk/return to scenario 1, which seems to be logical and main reason people employ this strategy.
Okay, so you're not trying to show that scenario 2 is market timing. Glad we could clear that up.

As for the claim that "scenario 2 somehow provides superior risk/return to scenario 1": we've discussed this. Whichever scenario exposes you to higher stock allocation more often will be higher reward *and* higher risk. Which scenario will that be? Depends on the timing interval of scenario 1 and the band size of scenario 2. So, scenario 2 doesn't give you "superior" risk/return. It will either give you higher risk and higher return, or lower risk and lower return, depending on the details. It's not magic.

tphp99
Posts: 330
Joined: Sun Jan 03, 2010 8:47 pm

Re: Explain why Rebalancing isn't Market Timing?

Post by tphp99 » Fri Dec 09, 2016 3:11 pm

avalpert wrote: If anything, the data I have seen shows the opposite - method #2 produces superior returns to method #1
How's that credible data when we all agree that we don't know what the market will do in the next ___ yrs?

Post Reply