Rebalancing is profitable?

Have a question about your personal investments? No matter how simple or complex, you can ask it here.
Post Reply
jjmaddison
Posts: 7
Joined: Sat Oct 21, 2017 3:52 am

Rebalancing is profitable?

Post by jjmaddison » Sat Oct 28, 2017 3:38 am

Hi,

One of the things that I read regularly on this forum is: "rebalancing is good, because you sell high buy low (stocks go up you rebalance sell stocks buy bonds and vise versa)".
Or, like this: "having bonds is good, because you rebalance to stocks (buy cheap) in case of a crash".

Is that really true?
Smells like timing the market.

Rebalancing should happen yearly? Monthly? On sharp dips?

Did anyone test things like "100% equity" vs "80/20 plus yearly rebalancing" using a portfolio modelling tool or something?

P.S. I tried https://www.portfoliovisualizer.com/mon ... simulation, it shows 100% equity is ~25% better, not sure how they handle rebalancing though.

P.P.S. This exact question refers to rebalancing for profits (as mentioned in other posts on this forum), not keeping bonds to reduce the risk.
Last edited by jjmaddison on Sat Oct 28, 2017 6:50 am, edited 3 times in total.

selters
Posts: 597
Joined: Thu Feb 27, 2014 9:26 am

Re: Rebalancing is profitable?

Post by selters » Sat Oct 28, 2017 6:32 am

Proftable if there is mean reversion between the asset classes or sub-asset classes you're rebalancing between. Unprofitable if not.

Call_Me_Op
Posts: 7028
Joined: Mon Sep 07, 2009 2:57 pm
Location: Milky Way

Re: Rebalancing is profitable?

Post by Call_Me_Op » Sat Oct 28, 2017 6:48 am

The purpose of rebalancing is not to increase returns - although it sometimes does. Its purpose is to control risk.
Best regards, -Op | | "In the middle of difficulty lies opportunity." Einstein

AlohaJoe
Posts: 3744
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Rebalancing is profitable?

Post by AlohaJoe » Sat Oct 28, 2017 6:49 am

People have tested lots of things. I would say that the current consensus (which doesn't mean everyone agrees) is that rebalancing offers little to no benefit when it comes to "profitability". It may offer some benefits related to risk control but to a large extent that depends on one's definition of risk.

McGilicutty
Posts: 149
Joined: Tue Dec 13, 2016 5:24 pm

Re: Rebalancing is profitable?

Post by McGilicutty » Sat Oct 28, 2017 6:58 am

Rebalancing has not been profitable recently as the stock market has gone up too far too fast. I think last year the S&P 500 returned something like 15% and this year it has already returned over 16%. You will need a big drop in the stock market to make up for the profits lost by being in bonds during just that small time period.

There are folks on this board who dream of another 50%-2008-like-crash, so that they can rebalance out of their high bond allocations, but those types of drops don't happen very often.

User avatar
nisiprius
Advisory Board
Posts: 36637
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: Rebalancing is profitable?

Post by nisiprius » Sat Oct 28, 2017 7:06 am

This is a very frequently debated point. I am going to state what I personally believe without trying to defend it or support it.

1) Rebalancing is not a magic market-timing-like formula that works. Rebalancing does not create extra return without extra risk. Rebalancing does not work by "forcing you to buy low and sell high."

2) Rebalancing cannot manufacture extra return out of pure volatility. If two assets are following random walks, rebalancing will not create extra return.

3) Rebalancing does create extra return if a) the assets exhibit mean reversion, and the mean reversion is really consistent, and b) the rebalancing interval is tuned to the mean reversion period. For example, it has been widely found--starting with an academic paper published in 1937!--that stocks show momentum over short periods of time, e.g. less than a year, and mean reversion over longer periods, e.g. around five years. If an asset shows momentum between rebalancing times, then rebalancing is going to hurt, not help. Nearly-continuous rebalancing, as done by balanced mutual funds, will neither hurt nor help because the time between rebalancing isn't long enough for either mean reversion or momentum to have much effect. The fact that it neither hurts nor helps was seen personally by me in the actual history of the Vanguard Balanced Index Fund around 2008-2009, and you can see this in the chart below under point 7.

4) Spreadsheet exercise and simulations that purport to show benefit from rebalancing with pure volatility and no mean reversion are like diagrams of perpetual motion machines; it is often hard to see the problem, but mean reversion has usually been smuggled in, in some way. Arguments about them are like arguments about the Monty Hall problem--or arguments back in 1998 about whether or not the year 2000 was going to be a leap year. The people you are trying to convince always come back with clever observations you've overlooked, and it is rarely possible to come to an agreement.

5) Innumerable backtests have been made on various rebalancing schemes, and various claims have been made. It is similar to most claims for small departures from buying-and-holding cap-weighted index funds. There's a difference, but it's hard to show that it is large, robust, and not just an artifact of the specific time period chosen.

6) Our mental model of rebalancing is often bad. It's complicated. When someone talks about "rebalancing during 2008-2009," the first picture that flashes through my mind is that of a single profitable purchase of stocks somewhere reasonably close to the bottom. In reality, time-based rebalancing rules fire at random times and are just as likely to have you buying before the crash, or selling into the crash. Band-based rebalancing rules are likely to fire multiple times during both decline and recovery. To assess overall profit you have to actually match purchase during the decline with sales during the recovery, and you have a collection of multiple sales and purchases that don't pair up as neatly as you'd like.

7) Rebalancing deepens losses during a long, large, decline. The slogan is that rebalancing is a formula for "buy low, sell high." But it is just as true to say that while the decline is in progress, rebalancing is a formula for "throwing good money after bad." Everybody acknowledges this and advocates of rebalancing note that yes, it is hard to do for that reason.

I can relate to this personally because I actually saw it happen in the Balanced Index Fund during 2008-2009. The decline was noticeably more than "decline in S&P 500 times 60%." We can see this in PortfolioVisualizer. The blue line, portfolio 1, is the Vanguard Balanced Index Fund, which is simply a two-fund portfolio of Total Stock and Total Bond continuously rebalanced to 60/40. Portfolio 2 is 60% Total Stock, 40% Total Bond, without rebalancing. Notice that overall there was very little benefit if any to the rebalancing in Balanced Index, but it did deepen the decline just noticeably; -32.47% versus -28.91% if not rebalanced.

Source
Image

There is also an extreme case; I've convinced myself that the extreme case is so extreme that it is not worth worrying about, but it does illustrate the problem. I'll call it "rebalancing suckage." Imagine two investors, both with 60/40 portfolios, and imagine that one rebalances and one doesn't... and imagine that the decline in stocks goes on and on and on without limit. The most that the first investor can lose is 60%; when the stocks go to zero, he has 40% left. The investor who rebalances can actually go to zero. Rebalancing into stocks keeps depleting the 40% bonds, and there is no limit to how far it can go. I repeat: I don't think this is of practical importance because even during a severe downturn the numbers aren't that extreme. Again, I can relate to this personally because during 2008-2009 I really observed that my portfolio, which at that time had a heavy allocation to the Balanced Index Fund, was declining annoyingly more than back-of-the-envelope calculations suggested it should. It was just about the same as in the PortfolioVisualizer chart. "Why am I down 33% instead of only 30%?" And the fact that it was only that much during a very severe downturn is why I think it is just annoying, not serious.
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
Earl Lemongrab
Posts: 5403
Joined: Tue Jun 10, 2014 1:14 am

Re: Rebalancing is profitable?

Post by Earl Lemongrab » Sat Oct 28, 2017 12:03 pm

In general the market goes up. Rebalancing is a way to keep a relatively constant allocation. For bonds versus stocks, that means that most of the time you are reducing the amount of stocks in the portfolio, which reduces your expected return (along with the expected volatility).
This week's fortune cookie: "Your financial life will be secure and beneficial." So I got that going for me, which is nice.

fedbogle
Posts: 29
Joined: Sat Mar 18, 2017 12:30 am

Re: Rebalancing is profitable?

Post by fedbogle » Sat Mar 24, 2018 11:49 pm

nisiprius wrote:
Sat Oct 28, 2017 7:06 am
This is a very frequently debated point. I am going to state what I personally believe without trying to defend it or support it.

1) Rebalancing is not a magic market-timing-like formula that works. Rebalancing does not create extra return without extra risk. Rebalancing does not work by "forcing you to buy low and sell high."

2) Rebalancing cannot manufacture extra return out of pure volatility. If two assets are following random walks, rebalancing will not create extra return.

3) Rebalancing does create extra return if a) the assets exhibit mean reversion, and the mean reversion is really consistent, and b) the rebalancing interval is tuned to the mean reversion period. For example, it has been widely found--starting with an academic paper published in 1937!--that stocks show momentum over short periods of time, e.g. less than a year, and mean reversion over longer periods, e.g. around five years. If an asset shows momentum between rebalancing times, then rebalancing is going to hurt, not help. Nearly-continuous rebalancing, as done by balanced mutual funds, will neither hurt nor help because the time between rebalancing isn't long enough for either mean reversion or momentum to have much effect. The fact that it neither hurts nor helps was seen personally by me in the actual history of the Vanguard Balanced Index Fund around 2008-2009, and you can see this in the chart below under point 7.

4) Spreadsheet exercise and simulations that purport to show benefit from rebalancing with pure volatility and no mean reversion are like diagrams of perpetual motion machines; it is often hard to see the problem, but mean reversion has usually been smuggled in, in some way. Arguments about them are like arguments about the Monty Hall problem--or arguments back in 1998 about whether or not the year 2000 was going to be a leap year. The people you are trying to convince always come back with clever observations you've overlooked, and it is rarely possible to come to an agreement.

5) Innumerable backtests have been made on various rebalancing schemes, and various claims have been made. It is similar to most claims for small departures from buying-and-holding cap-weighted index funds. There's a difference, but it's hard to show that it is large, robust, and not just an artifact of the specific time period chosen.

6) Our mental model of rebalancing is often bad. It's complicated. When someone talks about "rebalancing during 2008-2009," the first picture that flashes through my mind is that of a single profitable purchase of stocks somewhere reasonably close to the bottom. In reality, time-based rebalancing rules fire at random times and are just as likely to have you buying before the crash, or selling into the crash. Band-based rebalancing rules are likely to fire multiple times during both decline and recovery. To assess overall profit you have to actually match purchase during the decline with sales during the recovery, and you have a collection of multiple sales and purchases that don't pair up as neatly as you'd like.

7) Rebalancing deepens losses during a long, large, decline. The slogan is that rebalancing is a formula for "buy low, sell high." But it is just as true to say that while the decline is in progress, rebalancing is a formula for "throwing good money after bad." Everybody acknowledges this and advocates of rebalancing note that yes, it is hard to do for that reason.

I can relate to this personally because I actually saw it happen in the Balanced Index Fund during 2008-2009. The decline was noticeably more than "decline in S&P 500 times 60%." We can see this in PortfolioVisualizer. The blue line, portfolio 1, is the Vanguard Balanced Index Fund, which is simply a two-fund portfolio of Total Stock and Total Bond continuously rebalanced to 60/40. Portfolio 2 is 60% Total Stock, 40% Total Bond, without rebalancing. Notice that overall there was very little benefit if any to the rebalancing in Balanced Index, but it did deepen the decline just noticeably; -32.47% versus -28.91% if not rebalanced.

Source
Image

There is also an extreme case; I've convinced myself that the extreme case is so extreme that it is not worth worrying about, but it does illustrate the problem. I'll call it "rebalancing suckage." Imagine two investors, both with 60/40 portfolios, and imagine that one rebalances and one doesn't... and imagine that the decline in stocks goes on and on and on without limit. The most that the first investor can lose is 60%; when the stocks go to zero, he has 40% left. The investor who rebalances can actually go to zero. Rebalancing into stocks keeps depleting the 40% bonds, and there is no limit to how far it can go. I repeat: I don't think this is of practical importance because even during a severe downturn the numbers aren't that extreme. Again, I can relate to this personally because during 2008-2009 I really observed that my portfolio, which at that time had a heavy allocation to the Balanced Index Fund, was declining annoyingly more than back-of-the-envelope calculations suggested it should. It was just about the same as in the PortfolioVisualizer chart. "Why am I down 33% instead of only 30%?" And the fact that it was only that much during a very severe downturn is why I think it is just annoying, not serious.
One of your many great posts, Nisiprius. Clear insights on a question that doesn't yet have a clear answer. Thank you for sharing your knowledge with us.

longinvest
Posts: 2927
Joined: Sat Aug 11, 2012 8:44 am

Re: Rebalancing is profitable?

Post by longinvest » Sun Mar 25, 2018 8:51 am

Nisiprius,
nisiprius wrote:
Sat Oct 28, 2017 7:06 am
I can relate to this personally because I actually saw it happen in the Balanced Index Fund during 2008-2009. The decline was noticeably more than "decline in S&P 500 times 60%." We can see this in PortfolioVisualizer. The blue line, portfolio 1, is the Vanguard Balanced Index Fund, which is simply a two-fund portfolio of Total Stock and Total Bond continuously rebalanced to 60/40. Portfolio 2 is 60% Total Stock, 40% Total Bond, without rebalancing. Notice that overall there was very little benefit if any to the rebalancing in Balanced Index, but it did deepen the decline just noticeably; -32.47% versus -28.91% if not rebalanced.

Source
Image
I think that this is the wrong way to look at it.

Extending the original chart until the end of February 2018 reveals that the non-rebalanced portfolio would have drifted to 70% stocks / 30% bonds by the end of last month*. I don't think that it would be wise for someone who can't tolerate a 33% drop to be invested into such a high stock allocation due to a fear of rebalancing.

* I'm writing this on March 25, 2018.

The problem here isn't that the rebalanced portfolio lost more than the non-rebalanced one; the problem is that the allocation of the non-rebalanced portfolio drifted far away from target. Clicking on the "Allocation Drift" tab in Portfolio Visualizer reveals this:
Image

I think that if one doesn't have the emotional ability for a 33% drop, preferring to let the bond allocation jump to 53% during a crisis, one should immediately adopt a 47/53 stock/bond allocation from the start. Here's the result:
  • Source: Portfolio Visualizer
  • Portfolio 1:
    • VBINX Vanguard Balanced Index Inv 100.00%
  • Portfolio 2:
    • VTSMX Vanguard Total Stock Mkt Idx Inv 47.00%
    • VBMFX Vanguard Total Bond Market Index Inv 53.00%
    • Rebalanced monthly
Image

In other words, it's better to choose one's asset allocation according to one's volatility tolerance from the start, in good times before a crisis, instead of freezing like a deer in headlights during a crisis and letting one's asset allocation drift.
nisiprius wrote:
Sat Oct 28, 2017 7:06 am
There is also an extreme case; I've convinced myself that the extreme case is so extreme that it is not worth worrying about, but it does illustrate the problem. I'll call it "rebalancing suckage." Imagine two investors, both with 60/40 portfolios, and imagine that one rebalances and one doesn't... and imagine that the decline in stocks goes on and on and on without limit. The most that the first investor can lose is 60%; when the stocks go to zero, he has 40% left. The investor who rebalances can actually go to zero.
Again, let me suggest a different view.

Let me, first, pick an extreme case to illustrate that there is a real problem, in general, with rebalancing to a fixed allocation. Imagine that I decided to allocate 50% of my portfolio to Apple stock and 50% to Enron stock, rebalancing my portfolio daily. It's sad that Portfolio Visualizer doesn't recognize the ENE ticker symbol. But, it should be obvious that rebalancing would have killed such a portfolio.

This is one of the main reasons, actually, for investing into a cap-weighted total-market index fund. It avoids internally** rebalancing to $0. (Unless the entire market goes to $0, like Russia and China).

** Yes, cap-weighted index funds have to mildly rebalance to market cap weightings, internally, because of things like stock buybacks, new issues, and so on.

This brings me to the second point: It's crucial to build a rebalanced portfolio from broad cap-weighted index funds for which we have a reasonable confidence they won't go to $0.

Taylor Larimore's Three-Fund Portfolio is an awesome illustration of such a rebalance-ready portfolio. Let's consider each of its three components:
  • Cap-weighted total U.S. market stock index fund: If the entire U.S. stock market collapses, a U.S. investor will have much bigger problems than his portfolio problems.
  • Cap-weighted total U.S. market nominal bond index fund: A total default of the U.S. government and corporations is similar to the problem of a collapse of the entire U.S. stock market. As for hyperinflation, it also happens in periods where portfolio problems are the least of an investor's problems.
  • Cap-weighted total international stock index fund: If all non-U.S. stock markets collapse, or all international countries confiscate all U.S. investments, it's likely one has much bigger problems than portfolio problems.
Lastly, to mitigate the hypothetical*** problem of a total collapse of one of the three components of the Three-Fund Portfolio, one simply needs to adopt a sensible rebalancing approach. The simplest is this:
  • Regular partial rebalancing with contributions only (during accumulation) or withdrawals only (during retirement). In other words, invest new contributions into assets lagging their target allocation during accumulation, and take withdrawals from assets above their target allocation during retirement.
  • Infrequent, once a year, full rebalancing (e.g. sell and buy) on a fixed date, like one's birthday. This gives ample time for a portfolio component to fully collapse, if it has to, before having rebalanced into it, avoiding the catastrophe of continuously rebalancing into a collapsing asset.
Unsurprisingly, this happens to be the typical Bogleheads rebalancing recommendation.

*** Even though the problem is more hypothetical than real, it's important to think such things through to avoid letting our emotions take control during a crisis.

Note that trigger rebalancing, like 5/25 and other methods which let the market decide when to rebalance, and portfolio slicing-and-dicing across many smaller portfolio components, actually increase the risk of rebalancing to $0. All risks are not measured by volatility. It's not surprising that the confessed objective of trigger rebalancing and slicing-and-dicing is to increase returns (which is an ill-advised objective, in my opinion). It's an insidious risk/reward relation.
Last edited by longinvest on Thu Apr 12, 2018 6:44 am, edited 19 times in total.
Bogleheads investment philosophy | Lifelong Portfolio: 25% each of (domestic/international)stocks/(nominal/inflation-indexed)bonds | VCN/VXC/VLB/ZRR

MrMatt2532
Posts: 448
Joined: Sun Mar 15, 2009 11:58 am

Re: Rebalancing is profitable?

Post by MrMatt2532 » Sun Mar 25, 2018 8:59 am

William Bernstein analyzed this, and he demonstrates that the return of a regularly rebalanced portfolio is usually in excess of the weighted average return of the assets, which would mean the rebalancing bonus is real:
http://www.efficientfrontier.com/ef/996/rebal.htm

However, the effect is probably minor and the primary purpose of rebalancing should be to control risk. However, potentially there are some learnings about the frequency of rebalancing that may maximize chances of getting the bonus.

alex_686
Posts: 3830
Joined: Mon Feb 09, 2015 2:39 pm

Re: Rebalancing is profitable?

Post by alex_686 » Sun Mar 25, 2018 9:10 am

Call_Me_Op wrote:
Sat Oct 28, 2017 6:48 am
The purpose of rebalancing is not to increase returns - although it sometimes does. Its purpose is to control risk.
I would second this.

To complicate matters, there are convex and concave rebalancing strategies and these offer different rebalancing bonuses depending on the market. Concave strategies - or holding a constant AA, gets you a rebalancing strategies in volilital markets, but pentalizes you in calm markets. For the past 10 years we have had calm markets, so rebalancing actually underperform. What the next 10 years will bring, who knows??

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

Re: Rebalancing is profitable?

Post by dbr » Sun Mar 25, 2018 11:21 am

The simple minded statement that you make money from rebalancing by buying low and selling high is not correct. The process is more complex than that and produces variable "profitability" depending on circumstances but on average very small or not at all. The purpose of rebalancing is to keep risk under control by maintaining a selected asset allocation. It is true that over time a portfolio that is not rebalanced will eventually drift higher and higher in stock allocation and therefore have greater return. That is not a reason to avoid rebalancing as the risk of the portfolio will have also increased. Naturally the highest returning portfolio would be 100% stocks and such a portfolio does not have to be rebalanced. In the basic case we are talking about allocation between stocks and "bonds."

fedbogle
Posts: 29
Joined: Sat Mar 18, 2017 12:30 am

Re: Rebalancing is profitable?

Post by fedbogle » Sun Apr 01, 2018 2:23 am

longinvest wrote:
Sun Mar 25, 2018 8:51 am
Nisiprius,
nisiprius wrote:
Sat Oct 28, 2017 7:06 am
I can relate to this personally because I actually saw it happen in the Balanced Index Fund during 2008-2009. The decline was noticeably more than "decline in S&P 500 times 60%." We can see this in PortfolioVisualizer. The blue line, portfolio 1, is the Vanguard Balanced Index Fund, which is simply a two-fund portfolio of Total Stock and Total Bond continuously rebalanced to 60/40. Portfolio 2 is 60% Total Stock, 40% Total Bond, without rebalancing. Notice that overall there was very little benefit if any to the rebalancing in Balanced Index, but it did deepen the decline just noticeably; -32.47% versus -28.91% if not rebalanced.

Source
Image
I think that this is the wrong way to look at it.

Extending the original chart until the end of February 2018 reveals that the non-rebalanced portfolio would have drifted to 70% stocks / 30% bonds by the end of last month*. I don't think that it would be wise for someone who can't tolerate a 33% drop to be invested into such a high stock allocation due to a fear of rebalancing.

* I'm writing this on March 25, 2018.

The problem here isn't that the rebalanced portfolio lost more than the non-rebalanced one; the problem is that the allocation of the non-rebalanced portfolio drifted far away from target. Clicking on the "Allocation Drift" tab in Portfolio Visualizer reveals this:
Image

I think that if one doesn't have the emotional ability for a 33% drop, preferring to let the bond allocation jump to 53% during a crisis, one should immediately adopt a 47/53 stock/bond allocation from the start. Here's the result:
  • Source: Portfolio Visualizer
  • Portfolio 1:
    • VBINX Vanguard Balanced Index Inv 100.00%
  • Portfolio 2:
    • VTSMX Vanguard Total Stock Mkt Idx Inv 47.00%
    • VBMFX Vanguard Total Bond Market Index Inv 53.00%
    • Rebalanced monthly
Image

In other words, it's better to choose one's asset allocation according to one's volatility tolerance from the start, in good times before a crisis, instead of freezing like a deer in headlights during a crisis and letting one's asset allocation drift.
nisiprius wrote:
Sat Oct 28, 2017 7:06 am
There is also an extreme case; I've convinced myself that the extreme case is so extreme that it is not worth worrying about, but it does illustrate the problem. I'll call it "rebalancing suckage." Imagine two investors, both with 60/40 portfolios, and imagine that one rebalances and one doesn't... and imagine that the decline in stocks goes on and on and on without limit. The most that the first investor can lose is 60%; when the stocks go to zero, he has 40% left. The investor who rebalances can actually go to zero.
Again, let me suggest a different view.

Let me, first, pick an extreme case to illustrate that there is a real problem, in general, with rebalancing to a fixed allocation. Imagine that I decided to allocate 50% of my portfolio to Apple stock and 50% to Enron stock, rebalancing my portfolio daily. It's sad that Portfolio Visualizer doesn't recognize the ENE ticker symbol. But, it should be obvious that rebalancing would have killed such a portfolio.

This is one of the main reasons, actually, for investing into a cap-weighted total-market index fund. It avoids internally** rebalancing to $0. (Unless the entire market goes to $0, like Russia and China).

** Yes, cap-weighted index funds have to mildly rebalance to market cap weightings, internally, because of things like stock buybacks, new issues, and so on.

This brings me to the second point: It's crucial to build a rebalanced portfolio from broad cap-weighted index funds for which we have a reasonable confidence they won't go to $0.

Taylor Larimore's Three-Fund Portfolio is an awesome illustration of such a rebalance-ready portfolio. Let's consider each of its three components:
  • Cap-weighted total U.S. market stock index fund: If the entire U.S. stock market collapses, a U.S. investor will have much bigger problems than his portfolio problems.
  • Cap-weighted total U.S. market nominal bond index fund: A total default of the U.S. government and corporations is similar to the problem of a collapse of the entire U.S. stock market. As for hyperinflation, it also happens in periods where portfolio problems are the least of an investor's problems.
  • Cap-weighted total international stock index fund: If all non-U.S. stock markets collapse, or all international countries confiscate all U.S. investments, it's likely one has much bigger problems than portfolio problems.
Lastly, to mitigate the hypothetical*** problem of a total collapse of one of the three components of the Three-Fund Portfolio, one simply needs to adopt a sensible rebalancing approach. The simplest is this:
  • Regular partial rebalancing with contributions only (during accumulation) or withdrawals only (during retirement). In other words, invest new contributions into assets lagging their target allocation during accumulation, and take withdrawals from assets above their target allocation during retirement.
  • Infrequent, once a year, full rebalancing (e.g. sell and buy) on a fixed date, like one's birthday. This gives ample time for a portfolio component to fully collapse, if it has to, before having rebalanced into it, avoiding the catastrophe of continuously rebalancing into a collapsing asset.
Unsurprisingly, this happens to be the typical Bogleheads rebalancing recommendation.

*** Even though the problem is more hypothetical than real, it's important to think such things through to avoid letting our emotions take control during a crisis.

Note that trigger rebalancing, like 5/25 and other methods which let the market decide when to rebalance, and portfolio slicing-and-dicing across many smaller portfolio components, actually increase the risk of rebalancing to $0. All risks are not measured by volatility. It's not surprising that the confessed objective of trigger rebalancing and slicing-and-dicing is to increase returns (which is an ill-advised objective, in my opinion). It's an insidious risk/reward relation.
Interesting perspective.

Longinvest or anyone else who cares to chime in: what do we think the chances are of the market actually going to zero? Should we factor that in as a concern?

JustinR
Posts: 689
Joined: Tue Apr 27, 2010 11:43 pm

Re: Rebalancing is profitable?

Post by JustinR » Sun Apr 01, 2018 6:06 am

jjmaddison wrote:
Sat Oct 28, 2017 3:38 am
Hi,

One of the things that I read regularly on this forum is: "rebalancing is good, because you sell high buy low (stocks go up you rebalance sell stocks buy bonds and vise versa)".
Or, like this: "having bonds is good, because you rebalance to stocks (buy cheap) in case of a crash".

Is that really true?
Smells like timing the market.

Rebalancing should happen yearly? Monthly? On sharp dips?

Did anyone test things like "100% equity" vs "80/20 plus yearly rebalancing" using a portfolio modelling tool or something?

P.S. I tried https://www.portfoliovisualizer.com/mon ... simulation, it shows 100% equity is ~25% better, not sure how they handle rebalancing though.

P.P.S. This exact question refers to rebalancing for profits (as mentioned in other posts on this forum), not keeping bonds to reduce the risk.
No, the "rebalancing bonus" is a myth. You actually lose money if you rebalance.

Use the Portfolio Visualizer backtest tool and try it yourself using the different Rebalancing options at the top. You'll find that "No rebalancing" will give you the greatest return.

This makes sense because if you just leave your investments as is, over time your asset allocation will drift towards the higher return, higher risk asset.

The point of rebalancing is to maintain your pre-planned asset allocation and risk, not to get higher returns.

Check out Vanguard's white paper of rebalancing: https://www.vanguard.com/pdf/ISGPORE.pdf

The very first thing it says is:
The primary goal of a rebalancing strategy is to minimize risk relative to a target asset allocation, rather than to maximize returns

Then they find that never rebalancing will give you the highest return:
If a portfolio is never rebalanced, it tends to gradually drift from its target asset allocation as the weight of higher-return, higher-risk assets increases. Compared with the target allocation, the portfolio’s expected return increases, as does its vulnerability to deviations from the return
of the target asset allocation. To illustrate this, we compared two hypothetical portfolios, each with a target asset allocation of 50% global stocks/50% global bonds for the period 1926 through 2014; the first portfolio was rebalanced annually, and the second portfolio was never rebalanced. As the figure shows, and consistent with the risk-premium theory, the never-rebalanced portfolio’s stock allocation gradually drifted upward, to a maximum of 97%, and was 81% on average for the period. As the never-rebalanced portfolio’s equity exposure increased, the portfolio displayed higher risk (a standard deviation of 13.2%, versus 9.9% for the rebalanced portfolio) and a higher average annualized return (8.9% versus 8.1%, respectively).

longinvest
Posts: 2927
Joined: Sat Aug 11, 2012 8:44 am

Re: Rebalancing is profitable?

Post by longinvest » Sun Apr 01, 2018 7:30 am

fedbogle wrote:
Sun Apr 01, 2018 2:23 am
Longinvest or anyone else who cares to chime in: what do we think the chances are of the market actually going to zero? Should we factor that in as a concern?
Fundamentally, financial securities are promises which rely on laws and courts. I personally don't try to protect myself from an entire market going to zero other than by adopting a slow rebalancing method to mitigate the consequences of such an unlikely thing. Such an event tends to be caused by a collapse of society (Russia and China in the 1900's); in such times, one has much bigger concerns like simply staying alive.
Last edited by longinvest on Sun Apr 01, 2018 8:15 am, edited 1 time in total.
Bogleheads investment philosophy | Lifelong Portfolio: 25% each of (domestic/international)stocks/(nominal/inflation-indexed)bonds | VCN/VXC/VLB/ZRR

longinvest
Posts: 2927
Joined: Sat Aug 11, 2012 8:44 am

Re: Rebalancing is profitable?

Post by longinvest » Sun Apr 01, 2018 8:11 am

JustinR wrote:
Sun Apr 01, 2018 6:06 am
Check out Vanguard's white paper of rebalancing: https://www.vanguard.com/pdf/ISGPORE.pdf

The very first thing it says is:
The primary goal of a rebalancing strategy is to minimize risk relative to a target asset allocation, rather than to maximize returns

Then they find that never rebalancing will give you the highest return:
If a portfolio is never rebalanced, it tends to gradually drift from its target asset allocation as the weight of higher-return, higher-risk assets increases. Compared with the target allocation, the portfolio’s expected return increases, as does its vulnerability to deviations from the return
of the target asset allocation. To illustrate this, we compared two hypothetical portfolios, each with a target asset allocation of 50% global stocks/50% global bonds for the period 1926 through 2014; the first portfolio was rebalanced annually, and the second portfolio was never rebalanced. As the figure shows, and consistent with the risk-premium theory, the never-rebalanced portfolio’s stock allocation gradually drifted upward, to a maximum of 97%, and was 81% on average for the period. As the never-rebalanced portfolio’s equity exposure increased, the portfolio displayed higher risk (a standard deviation of 13.2%, versus 9.9% for the rebalanced portfolio) and a higher average annualized return (8.9% versus 8.1%, respectively).
The thing is that the portfolio's asset allocation was 81/19 stocks/bonds on average (see the part I put in red above). It shouldn't be compared to a rebalanced 50/50 stocks/bonds portfolio, but to a rebalanced 81/19 stocks/bonds portfolio.

Let me try to do a fair comparison using Portfolio Visualizer, using 2000 as the end date to maximize the higher return of stocks. I'll use two asset classes for which we have historical returns since 1972: US Stock Market and 10-year Treasury. (Total Bond Market return history starts in 1987).

Here are the returns and allocation drift of a 50/50 stocks/bonds portfolio which was never rebalanced:

Source: Portfolio VIsualizer
Image
Image

We see that stocks drifted from 50% to 78%. I'll estimate the average stock allocation as ((50% + 78%) / 2) = 64%.

Here are the returns of a 64/36 stocks/bonds portfolio which was rebalanced monthly:

Source: Portfolio Visualizer
Image
We see that both portfolios had similar returns, overall. I see no surprise, there.

The point is that an investor willing to let his portfolio drift from 50/50 to 78/22 stocks/bonds is actually investing into a portfolio which averages 64/36 stocks/bonds. He would be best to invest into a 64/36 stocks/bonds portfolio from the start and stay the course by rebalancing his portfolio. Starting with 50/50 and letting the portfolio drift leaves him with a 78/22 stocks/bonds portfolio in 2000, just before a big stock market drop!
Last edited by longinvest on Thu Apr 12, 2018 6:43 am, edited 3 times in total.
Bogleheads investment philosophy | Lifelong Portfolio: 25% each of (domestic/international)stocks/(nominal/inflation-indexed)bonds | VCN/VXC/VLB/ZRR

Marketman
Posts: 196
Joined: Sun Feb 04, 2018 2:24 pm
Location: Texas

Re: Rebalancing is profitable?

Post by Marketman » Sun Apr 01, 2018 8:21 am

One difficulty I have in the "slow rebalancing" portfolio (say once a year or even every several years) is that the arbitrary dates you choose to rebalance may or may not be good dates to rebalance (buy stocks cheap or sell stock high). IF there is a momentum factor at work and IF the market has momentum for some period of time how do you know when the "momentum period" starts and ends?

Assume you can show stocks have momentum for say 6 months and then revert to some mean valuation in 10 years. I still don't see how you could capture the momentum if you don't know when a new trend begins and ends. For example, if the stock market has up momentum for the first 6 months of a calendar year and then down momentum of the last 6 months of the calendar year and you rebalance April 1 and October 1 you have not profited by the momentum. Or if you rebalanced every January 1 you also would not profit from the momentum. It seems one would need to know in advance when to rebalance to profit from the momentum.

Perhaps some trend following program could capture the momentum but I can't visualize how arbitrary rebalance dates could, even if you knew momentum existed and how long it lasts.

I have the same difficulty visualizing how a slow to rebalance portfolio could better protect you from market crashes. You are just as likely to rebalance the day before a crash as the day after. Back tests may show slow to rebalance portfolios better, but for the life of me, I can't visualize how.
Last edited by Marketman on Sun Apr 01, 2018 8:25 am, edited 1 time in total.

longinvest
Posts: 2927
Joined: Sat Aug 11, 2012 8:44 am

Re: Rebalancing is profitable?

Post by longinvest » Sun Apr 01, 2018 8:24 am

Marketman wrote:
Sun Apr 01, 2018 8:21 am
One difficulty I have in the "slow rebalancing" portfolio (say once a year or even every several years) is that the arbitrary dates you choose to rebalance may or may not be good dates to rebalance (buy stocks cheap).
Rebalancing isn't about increasing returns. One can shoot for higher returns by increasing the allocation to stocks.
Bogleheads investment philosophy | Lifelong Portfolio: 25% each of (domestic/international)stocks/(nominal/inflation-indexed)bonds | VCN/VXC/VLB/ZRR

Marketman
Posts: 196
Joined: Sun Feb 04, 2018 2:24 pm
Location: Texas

Re: Rebalancing is profitable?

Post by Marketman » Sun Apr 01, 2018 8:27 am

I agree that rebalancing is to control risk. Some say it is best to rebalance at infrequent intervals, but I can't see why. (I'm assuming we are talking about a tax free account.)

longinvest
Posts: 2927
Joined: Sat Aug 11, 2012 8:44 am

Re: Rebalancing is profitable?

Post by longinvest » Sun Apr 01, 2018 8:41 am

Marketman wrote:
Sun Apr 01, 2018 8:21 am
I have the same difficulty visualizing how a slow to rebalance portfolio could better protect you from market crashes. You are just as likely to rebalance the day before a crash as the day after. Back tests may show slow to rebalance portfolios better, but for the life of me, I can't visualize how.
The problem case is that of the unlikely event of an asset going to almost zero over a few months. Let's imagine that an asset loses 50% per month for 6 months. This represents a cumulative loss of (1 - (1 - 50%)^6) = 98%.

Monthly rebalancing a 50/50 portfolio would lead to a cumulative loss of 82%. Slow yearly rebalancing would reduce the loss by almost half to 49%. In other words, rebalancing monthly when an asset loses 50% every month reduces the portfolio to 18% of its initial value in 6 months! Yearly rebalancing reduces the portfolio to 51% of its initial value.

Slow rebalancing isn't about completely avoiding the loss. That can only be done through not being invested into the asset in the first place*. Slow rebalancing mitigates extreme (unlikely) outcomes. It helps with mental preparation for times when everybody thinks that the sky is falling, like during a market crash.

* That's why I don't invest into speculate with things like Bitcoins.
Last edited by longinvest on Sun Apr 01, 2018 9:07 am, edited 6 times in total.
Bogleheads investment philosophy | Lifelong Portfolio: 25% each of (domestic/international)stocks/(nominal/inflation-indexed)bonds | VCN/VXC/VLB/ZRR

Freefun
Posts: 356
Joined: Sun Jan 14, 2018 3:55 pm

Re: Rebalancing is profitable?

Post by Freefun » Sun Apr 01, 2018 8:55 am

Great post and responses. FWIW I've found this article helpful:

https://www.kitces.com/blog/best-opport ... hresholds/
Remember when you wanted what you currently have?

Marketman
Posts: 196
Joined: Sun Feb 04, 2018 2:24 pm
Location: Texas

Re: Rebalancing is profitable?

Post by Marketman » Sun Apr 01, 2018 9:34 am

Great article thanks!

longinvest
Posts: 2927
Joined: Sat Aug 11, 2012 8:44 am

Re: Rebalancing is profitable?

Post by longinvest » Sun Apr 01, 2018 9:42 am

Freefun wrote:
Sun Apr 01, 2018 8:55 am
Great post and responses. FWIW I've found this article helpful:

https://www.kitces.com/blog/best-opport ... hresholds/
Rebalancing bands would lead to a 82% loss reducing a 50/50 portfolio to 18% of its initial value, instead of a 49% loss reducing the portfolio to 51% of its initial value, in a doomsday scenario where one of the two assets lost 50% per month for 6 months. That's what I've been trying to explain in my last post: viewtopic.php?p=3858543#p3858476.

I've also discussed this same problem, and why (slow) annual rebalancing can be a good tool to help stay the course with rebalancing (along with investing into broad cap-weighted index funds), in another post: viewtopic.php?p=3858543#p3846674.
Bogleheads investment philosophy | Lifelong Portfolio: 25% each of (domestic/international)stocks/(nominal/inflation-indexed)bonds | VCN/VXC/VLB/ZRR

Marketman
Posts: 196
Joined: Sun Feb 04, 2018 2:24 pm
Location: Texas

Re: Rebalancing is profitable?

Post by Marketman » Sun Apr 01, 2018 9:51 am

Longinvest, yes I see your point in a doomsday scenario. Frequent rebalancing starting at Japan's bubble peak would have been very bad I bet. Do you see infrequent rebalancing catching potential market momentum? I still don't see how it could even if you know how long the momentum lasts if you don't know the start and end dates.

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

Re: Rebalancing is profitable?

Post by dbr » Sun Apr 01, 2018 9:55 am

If a person is truly worried about a death spiral of rebalancing, that can be nipped in the bud by implementing a rule of never rebalancing into stocks but only out. That is arguably practical for a retiree withdrawing money but not so much during accumulation. Even Larry Swedroe once suggested that as an option.

longinvest
Posts: 2927
Joined: Sat Aug 11, 2012 8:44 am

Re: Rebalancing is profitable?

Post by longinvest » Sun Apr 01, 2018 10:07 am

dbr wrote:
Sun Apr 01, 2018 9:55 am
If a person is truly worried about a death spiral of rebalancing, that can be nipped in the bud by implementing a rule of never rebalancing into stocks but only out. That is arguably practical for a retiree withdrawing money but not so much during accumulation. Even Larry Swedroe once suggested that as an option.
Who said that only stocks can experience a death spiral? I also consider the history 1923-24 German-style hyperinflation (bonds down 99.99%) and Venetian prestiti.

Here's how William Bernstein describes the history of Venetian prestiti in his awesome book The Four Pillars of Investing:
In The Four Pillars of Investing, William Bernstein wrote: The best-known early annuity was the Venetian prestiti, used to finance the Republic’s wars. [...] the Venetian treasury did allow owners to sell their prestiti to others—that is, to change the name registered at the central office. Prestiti soon became the favored vehicle for investment and speculation among Venetian noblemen and were even widely held throughout Europe. This “secondary market” in prestiti provides economic historians with a vivid picture of a medieval bond market that was quite active over many centuries.

[...] Defined in its most basic terms, risk is the possibility of losing money.

A fast look at Figure 1-3 shows that prestiti owners were certainly exposed to this unhappy prospect. For example, in the tranquil year of 1375, prices reached a high of 92 1/2. But just two years later, after a devastating war with Genoa, interest payments were temporarily suspended and vast amounts of new prestiti were levied, driving prices as low as 19; this constituted a temporary loss of principal value of about 80%. Even though Venice’s fortunes soon reversed, this financial catastrophe shook investor confidence for more than a century, and prices did not recover until the debt was refinanced in 1482.
People bought these bonds on the secondary market, expecting them to pay the stated coupon (5%). As expected from bonds, their prices fluctuated with interest rates. No surprise there. It doesn't appear that they had a maturity date, but that's not really important for my argument.

What happened is an unexpected risk in 1377. War, default, levies. Only then did people realize that there were other risks to these bonds. It took over 100 years for their price to recover. In other word, any 1375 buyer didn't see a price recovery during his lifetime.

Let me repeat this: a person who put all of his money into prestiti, in 1375, was taking a huge risk, most probably without being aware of it. Holding the investment from 1375 until his death didn't pay off. (Later investors made an awesome investment, though!)

For me, one of the most important (hidden) lessons of this chapter of The Four Pillars of Investing is that one doesn't always perceive the risks one is taking.

Adopting a sensible slow rebalancing method mitigates the consequences of an unforeseen big asset loss.
Bogleheads investment philosophy | Lifelong Portfolio: 25% each of (domestic/international)stocks/(nominal/inflation-indexed)bonds | VCN/VXC/VLB/ZRR

fedbogle
Posts: 29
Joined: Sat Mar 18, 2017 12:30 am

Re: Rebalancing is profitable?

Post by fedbogle » Fri Apr 06, 2018 12:49 am

longinvest wrote:
Sun Apr 01, 2018 7:30 am
fedbogle wrote:
Sun Apr 01, 2018 2:23 am
Longinvest or anyone else who cares to chime in: what do we think the chances are of the market actually going to zero? Should we factor that in as a concern?
Fundamentally, financial securities are promises which rely on laws and courts. I personally don't try to protect myself from an entire market going to zero other than by adopting a slow rebalancing method to mitigate the consequences of such an unlikely thing. Such an event tends to be caused by a collapse of society (Russia and China in the 1900's); in such times, one has much bigger concerns like simply staying alive.
Agree completely. It seems to me that if you are in it for the longish run (20+ years?) then you will weather almost any crash or collapse.

This has me thinking about long-term declines. I guess Japan had a pretty long one. Looking at Japan, it looks like about a 25 year decline to me, from about 1989 to 2012. That would be a hard one to weather. But with a standard Bogleheads-type globally diversified portfolio, I would think we would be able to survive even that.

Perhaps you could even view the Russia and China examples you mention as temporary, as they recovered at some point. Or perhaps most or all investors lost their hats first.

But I wonder, what are the relevant examples of long, slow, declines, aside from Japan?

longinvest
Posts: 2927
Joined: Sat Aug 11, 2012 8:44 am

Re: Rebalancing is profitable?

Post by longinvest » Fri Apr 06, 2018 7:53 am

fedbogle wrote:
Fri Apr 06, 2018 12:49 am
But I wonder, what are the relevant examples of long, slow, declines, aside from Japan?
U.S. nominal bonds, in real* terms, from 1940 to 1980.

* Inflation-adjusted.
Bogleheads investment philosophy | Lifelong Portfolio: 25% each of (domestic/international)stocks/(nominal/inflation-indexed)bonds | VCN/VXC/VLB/ZRR

MinhN
Posts: 31
Joined: Sun Mar 11, 2018 11:57 pm

Re: Rebalancing is profitable?

Post by MinhN » Fri Apr 06, 2018 8:05 am

Rebalancing doesn't give a free lunch. The purpose of rebalancing is to target a specific historical return and volatility. Whether or not the return and volatility will hold in the future, it's anyone's guess.
Last edited by MinhN on Fri Apr 06, 2018 8:13 am, edited 1 time in total.

longinvest
Posts: 2927
Joined: Sat Aug 11, 2012 8:44 am

Re: Rebalancing is profitable?

Post by longinvest » Fri Apr 06, 2018 8:11 am

A different shorter-term example of a negative market is international stocks, in real terms, from November 2007 to March 2018, a period of more than 10 years with an inflation-adjusted CAGR of -0.62% (source: Portfolio Visualizer).
Bogleheads investment philosophy | Lifelong Portfolio: 25% each of (domestic/international)stocks/(nominal/inflation-indexed)bonds | VCN/VXC/VLB/ZRR

Post Reply