Wisdom of Jack Bogle on rebalancing and on bond funds

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
Lbill
Posts: 4997
Joined: Thu Mar 13, 2008 11:25 pm
Location: Somewhere between Up and Down

Wisdom of Jack Bogle on rebalancing and on bond funds

Post by Lbill » Tue Jun 08, 2010 7:27 pm

On rebalancing: it don't make no difference:
We’ve just done a study for the NYTimes on rebalancing, so the subject is fresh in my mind. Fact: a 48%S&P 500, 16% small cap, 16% international, and 20% bond index, over the past 20 years, earned a 9.49% annual return without rebalancing and a 9.71% return if rebalanced annually. That’s worth describing as “noise,” and suggests that formulaic rebalancing with precision is not necessary.

We also did an earlier study of all 25-year periods beginning in 1826 (!), using a 50/50 US stock/bond portfolio, and found that annual rebalancing won in 52% of the 179 periods. Also, it seems to me, noise. Interestingly, failing to rebalance never cost more than about 50 basis points, but when that failure added return, the gains were often in the 200-300 basis point range; i.e., doing nothing has lost small but it has won big.

My personal conclusion. Rebalancing is a personal choice, not a choice that statistics can validate. There’s certainly nothing the matter with doing it (although I don’t do it myself), but also no reason to slavishly worry about small changes in the equity ratio. Maybe, for example, if your 50% equity position grew to, say, 55% or 60%.
On bond funds - Vanguard IT bond index fund is better than the Total Bond Market Fund:
I like the (taxable) IT bond index fund because it provides more stability than the LT index fund, and more income than the ST index fund. The Total Bond Market Index Fund is fine, but I vaguely wonder about a bond fund that has 35% of its portfolio in non-bonds (i.e., GNMA securities, with their risk of being prepaid early, when interest rates tumble).

That said, TBMF happens to have a maturity profile that is intermediate-term on balance, and so differs from IT largely in its holdings of GMNAs and Treasurys. Their ten-year records are similar, based on the tabulation I’m sending separately (IT 6.49%, TBM 5.96%, which included a single year–2002–in which we sort of forgot to stick to index principles, costing 2.00%, or about 0.20% per year. I’m assured by management that such an aberration will not recur.)
Full text is available on the Bogle eBlog
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard | | "You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs

livesoft
Posts: 69566
Joined: Thu Mar 01, 2007 8:00 pm

Post by livesoft » Tue Jun 08, 2010 7:38 pm

Rebalancing can be tricky. Why should one expect rebalancing on a date to outperform anything? Many of the studies just study a simple rebalancing (e.g., 'we re-balanced once at the beginning of each year').

Rebalancing can bring your risk set by your asset allocation back to what you previously picked. So without rebalancing you get virtually the same performance, but at a higher risk than you would get with rebalancing.

See also: http://www.retailinvestor.org/why.html#bonus and this link therein: http://www.wiseradvisor.com/university- ... quency.asp

I'd like to see a rebalancing study where they tested rebalancing bands over a wide range of possibilities, much like that "Opportunistic Rebalancing ..." study.

User avatar
bob90245
Posts: 6511
Joined: Mon Feb 19, 2007 8:51 pm

Post by bob90245 » Tue Jun 08, 2010 7:44 pm

Over long periods, 50/50 not rebalanced (which likely then becames closer to 75/25) should have higher terminal value than 50/50 rebalanced for the simple fact that stocks typically outperform bonds over the long term. When you don't rebalance, you allow your winners, which are usually stocks, run.
Ignore the market noise. Keep to your rebalancing schedule whether that is semi-annual, annual or trigger bands.

Call_Me_Op
Posts: 7423
Joined: Mon Sep 07, 2009 2:57 pm
Location: Milky Way

Post by Call_Me_Op » Tue Jun 08, 2010 7:53 pm

I will remind you that the primary reason for rebalancing is not to enhance returns - but to reduce volatility.
Best regards, -Op | | "In the middle of difficulty lies opportunity." Einstein

jshaffer740
Posts: 49
Joined: Mon May 03, 2010 7:21 pm

Post by jshaffer740 » Tue Jun 08, 2010 7:53 pm

Right. I think the key point to be made is that by rebalancing you are keeping risk in check, based on your pre-defined AA. Perhaps a study comparing not only return, but measures of risk as well, could be useful?

Jason

User avatar
Topic Author
Lbill
Posts: 4997
Joined: Thu Mar 13, 2008 11:25 pm
Location: Somewhere between Up and Down

Post by Lbill » Tue Jun 08, 2010 7:58 pm

Rebalancing can bring your risk set by your asset allocation back to what you previously picked. So without rebalancing you get virtually the same performance, but at a higher risk than you would get with rebalancing.
IMO, the conventional wisdom about controlling risk with rebalancing is wrong. Consider a portfolio that is 50/50 TSM- Bonds. The stock market crashes and loses 50% while the bond market remains steady. Your allocation is now 33% stocks and 67% bonds. The markets have repriced the value of stocks relative to bonds. If I now rebalance to 50/50 by selling bonds to buy stocks, I'm actually increasing the risk of my portfolio, not lowering it. The market has said that stocks are not worth the premium against bonds that they were before the crash. By rebalancing I'm betting that the market is wrong. Rebalancing is more about betting on reversion to the mean than about risk management. In this case, you are taking on more risk in order to bet that stocks will mean-revert and go up following a 50% loss. Sometimes you win, but sometimes you lose. Bogle found that you lose that bet about as often as you win it, and the result is a wash.
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard | | "You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs

User avatar
rustymutt
Posts: 3779
Joined: Sat Mar 07, 2009 12:03 pm
Location: Oklahoma

Post by rustymutt » Tue Jun 08, 2010 8:03 pm

Call_Me_Op wrote:I will remind you that the primary reason for rebalancing is not to enhance returns - but to reduce volatility.
exactly, thank you.
I'm amazed at the wealth of Knowledge others gather, and share over a lifetime of learning. The mind is truly unique. It's nice when we use it!

User avatar
Topic Author
Lbill
Posts: 4997
Joined: Thu Mar 13, 2008 11:25 pm
Location: Somewhere between Up and Down

Post by Lbill » Tue Jun 08, 2010 8:50 pm

I will remind you that the primary reason for rebalancing is not to enhance returns - but to reduce volatility.
But, where's the beef? For example, over the 10-year period of 2000-2009 a portfolio of 60% TSM (total stock market) and 40% TBM (total bond market) that was rebalanced annually had an annualized volatility (SD) of 12.69%, but a non-rebalanced 60/40 portfolio had an SD of 10.22%, which was smaller by 24.2%. Like many commonly believed truisms, this one turns out to be false also when subject to scrutiny. As Mark Twain said: "It's not what we believe that gets us in trouble, as much as what we believe that ain't so."
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard | | "You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs

jshaffer740
Posts: 49
Joined: Mon May 03, 2010 7:21 pm

Post by jshaffer740 » Tue Jun 08, 2010 9:06 pm

Lbill wrote:
I will remind you that the primary reason for rebalancing is not to enhance returns - but to reduce volatility.
But, where's the beef? For example, over the 10-year period of 2000-2009 a portfolio of 60% TSM (total stock market) and 40% TBM (total bond market) that was rebalanced annually had an annualized volatility (SD) of 12.69%, but a non-rebalanced 60/40 portfolio had an SD of 10.22%, which was smaller by 24.2%. Like many commonly believed truisms, this one turns out to be false also when subject to scrutiny. As Mark Twain said: "It's not what we believe that gets us in trouble, as much as what we believe that ain't so."
I'm not disputing what you've said in any way, because honestly I haven't researched this area enough. I've held the belief that rebalancing was to take you back to your AA risk level. But then what is the point of rebalancing? If we are lowering risk, and (per Mr. Bogle) increasing potential upswing by *not* rebalancing, what do we gain by rebalancing?

Although I may have just answered my own question: is it to avoid the potential of taking too *little* risk? That is, we define our risk tolerance and AA, and by not rebalancing we are "risking" not having enough risk for our needs. Does this make sense?

Jason

dbr
Posts: 31155
Joined: Sun Mar 04, 2007 9:50 am

Post by dbr » Tue Jun 08, 2010 9:27 pm

Rebalancing is just the idea of returning to a fixed allocation when market changes have carried things away. Naturally if the returns and standard deviations of the parts drift around, then that allocation will not have a constant risk (as SD) and constant expected return. I suppose it is true that this only makes sense if one believes that said risk and return is stationary in the long run. That means we do operate as if there will be return to an already expected mean.

I suppose, on the other hand, that one can adopt a philosophy of rebalancing to maintain a fixed commitment to some a-priori concept of diversification without assuming anything quantitative about risks and returns. This would be an approach assuming ignorance regarding all of the risk, the return, and the correlation of return among the chosen assets.

It is possible to constantly allocate to a targeted risk and return based on following whatever the current just past data is, perhaps on an annual basis. It is even possible to run MVO models to maximize efficiency at a given risk, based on immediately past data. In such a case the AA will be constantly adjusting to follow the ex post facto optimum.

User avatar
Topic Author
Lbill
Posts: 4997
Joined: Thu Mar 13, 2008 11:25 pm
Location: Somewhere between Up and Down

Post by Lbill » Tue Jun 08, 2010 10:24 pm

The reason that annual rebalancing didn't reduce portfolio risk (defined as volatility) during the 10 years ending in 2009 is that this was a poor period for stocks - the nominal returns have been essentially zero.

As a result, if you didn't rebalance your portfolio it drifted from a 60/40 allocation at the beginning of 2000 to a 45/55 allocation at the end of 2009. On average, the non-rebalanced portfolio was less volatile because it became more tilted toward bonds.

If you rebalanced annually to reset the allocation to 60/40, you had a more volatile portfolio over this period than if you hadn't rebalanced. Rebalancing during this period mostly involved selling bonds to buy more stocks. As I said in my earlier post, this was effectively a "bet" at each rebalancing point that stocks would rebound and you would be "buying low." The cost of this bet was increasing the "riskiness", or volatility, of your portfolio by repeatedly increasing the weight of stocks and thereby decreasing the weight of bonds. With the benefit of hindsight, we now know that it did increase riskiness but it didn't increase returns correspondingly. On average throughout the 10-year period, your rebalancing "bets" on equities didn't pay off; although it did pay off big in 2009.

The only time that rebalancing actually limits portfolio volatility is during periods in which stocks outperform bonds; and, consequently, the portfolio becomes more tilted toward stocks. By selling stocks to buy bonds, you are reducing overall portfolio volatility. Once again, however, rebalancing represents a "bet" that stock returns are likely to revert and that you will be able to capitalize; i.e. you will be "selling high." If you think about it this way, rebalancing really represents a kind of closet market timing strategy. It might work or might not work over any given period of time. You just don't know in advance.
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard | | "You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs

User avatar
mdrileynyc
Posts: 104
Joined: Thu Sep 27, 2007 9:31 pm
Location: New York

rebalancing

Post by mdrileynyc » Tue Jun 08, 2010 11:45 pm

I think it is was Sharpe that noted that everyone can hold the market portfolio but everyone can't rebalance to yesterday's portfolio (at today's changed prices). Rebalancing to a fixed allocation can't be a universal recommendation - the market at times gets riskier and everybody can't unload that added risk. Not sure what this mathematical truth says about what an individual should do though.

User avatar
anthau
Posts: 146
Joined: Wed Feb 21, 2007 5:29 pm
Location: Laramie, WY

Post by anthau » Wed Jun 09, 2010 5:13 am

Having read Geoff Considine's Rethinking Rebalancing: A Risk/Reward Analysis, it occurs to me that the difference between Bogle's study and others' work on rebalancing (e.g., William J Bernstein & Daryanani [.PDF]) is that Bogle's analysis is of a "lumper's" portfolio, while the others' analyses are of "splitter's" portfolios.

While I lack both the data and the mathematical horsepower, it does make me wonder: How many slices does it take for the "rebalancing bonus" to present as more than statistical noise? Three funds (US, Int'l, TBM)? "The Core Four?" More?

Edit: As I re-read Bernstein, his conclusion, "Only when long term return differences among asssets exceed 5 percent do nonrebalanced portfolios provide superior returns, and then only at the cost of increased risk." seems to be the crux of things. Rebalancing between stocks and bonds doesn't enhance returns, but rebalancing within stocks (or bonds?) does.
Best, | | Anth

User avatar
nisiprius
Advisory Board
Posts: 39720
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

Post by nisiprius » Wed Jun 09, 2010 7:11 am

I think rebalancing probably does make sense as risk control if you intended to be at 50/50 and see that you're at 75/25, as might have actually happened at the start and finish of the Great Bull Market of the 1990s--don't have time to check the numbers right now.

I think that rebalancing probably makes more sense in the context of slice-and-dice, because a slice-and-dicer is probably holding a fair number of different slices that are as volatile or more volatile than stocks but (hopefully) are moving in different directions. If your plan called for holding (say) 10% REIT Index Fund I think you could find that in a single year it had dropped to 5% or risen to 20%.
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.

Ping Pong
Posts: 512
Joined: Fri Feb 20, 2009 12:51 pm

Post by Ping Pong » Wed Jun 09, 2010 8:19 am

I think rebalancing only applies if you do everything in percentages, which is what most people here seem to do. If you've got your liabilities matched, then fluctuations in stocks won't affect your risk at all. (assuming you're not defining risk as percentage fluctuations in portfolio value, which a lot of people seem to do)

YDNAL
Posts: 13774
Joined: Tue Apr 10, 2007 4:04 pm
Location: Biscayne Bay

Re: Wisdom of Jack Bogle on rebalancing and on bond funds

Post by YDNAL » Wed Jun 09, 2010 8:24 am

Lbill wrote:IMO, the conventional wisdom about controlling risk with rebalancing is wrong. Consider a portfolio that is 50/50 TSM- Bonds. The stock market crashes and loses 50% while the bond market remains steady. Your allocation is now 33% stocks and 67% bonds. The markets have repriced the value of stocks relative to bonds. If I now rebalance to 50/50 by selling bonds to buy stocks, I'm actually increasing the risk of my portfolio, not lowering it.
One sided.

You are correct in that Equity drops make the portfolio less risky.

Rebalancing to "control risk" relates largely to the other side of the equation - the one you ignored.
Lbill wrote:
I like the (taxable) IT bond index fund because it provides more stability than the LT index fund, and more income than the ST index fund. The Total Bond Market Index Fund is fine, but I vaguely wonder about a bond fund that has 35% of its portfolio in non-bonds (i.e., GNMA securities, with their risk of being prepaid early, when interest rates tumble).
I thought TBM was Mr. Bogle's favorite Bond fund! :)
Landy | Be yourself, everyone else is already taken -- Oscar Wilde

Beagler
Posts: 3442
Joined: Sun Dec 21, 2008 7:39 pm

Re: Wisdom of Jack Bogle on rebalancing and on bond funds

Post by Beagler » Wed Jun 09, 2010 8:45 am

YDNAL wrote:
Lbill wrote:IMO, the conventional wisdom about controlling risk with rebalancing is wrong. Consider a portfolio that is 50/50 TSM- Bonds. The stock market crashes and loses 50% while the bond market remains steady. Your allocation is now 33% stocks and 67% bonds. The markets have repriced the value of stocks relative to bonds. If I now rebalance to 50/50 by selling bonds to buy stocks, I'm actually increasing the risk of my portfolio, not lowering it.
One sided.

You are correct in that Equity drops make the portfolio less risky.

Rebalancing to "control risk" relates largely to the other side of the equation - the one you ignored.
Lbill wrote:
I like the (taxable) IT bond index fund because it provides more stability than the LT index fund, and more income than the ST index fund. The Total Bond Market Index Fund is fine, but I vaguely wonder about a bond fund that has 35% of its portfolio in non-bonds (i.e., GNMA securities, with their risk of being prepaid early, when interest rates tumble).
I thought TBM was Mr. Bogle's favorite Bond fund! :)
Maybe since TBM became so full (over a third!) of non-bonds he's changed his mind.
“The only place where success come before work is in the dictionary.” Abraham Lincoln. This post does not provide advice for specific individual situations and should not be construed as doing so.

tms
Posts: 400
Joined: Mon Dec 15, 2008 11:44 am

Post by tms » Wed Jun 09, 2010 8:54 am

How can Bogle be in favor of the "100-age rule", and be against "rebalancing". Doesn't he have to rebalance, at least occassionally, to maintain his target allocation.

And on the time period from 2000-2009, this is what I found:

Take a diversified portfolio made up of one-third US stocks (as measured by the Russell 3000 index), one-third international stocks (as measured by the MSCI EAFE USD index), and one-third US investment grade bonds (as measured by the BarCap US Aggregate Bond index).

Assuming no rebalancing, this portfolio resulted in a cumulative return of 24.54% over the ten year period ending March 31, 2010. The portfolio at the end of the ten year period looks very different from the original portfolio, with US investment grade bonds make up more than 50% of the ending portfolio.

If, on the other hand, the portfolio had been rebalanced annually, the cumulative return increased to 32.44%, and more importantly, the portfolio allocation at the end of the ten year period looks very similar to the original target allocation.

I don't have the data on the risk, but the non-rebalancing option has to be less, because bonds make up more of the portfolio. In most decades, when stocks outperform bonds, it would be the opposite.

User avatar
nisiprius
Advisory Board
Posts: 39720
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

Post by nisiprius » Wed Jun 09, 2010 9:21 am

tms wrote:How can Bogle be in favor of the "100-age rule", and be against "rebalancing".
I like to think that Mr. Bogle might be a rough-n-ready satisficer.

I'm not sure what the definition of "rebalancing" is.

As used in this forum, it refers to a fairly structured discipline, such as balancing on a schedule, with 6 months, 1 year, and 2 years being intervals that are frequently suggested; or, by a rule such as Larry Swedroe's: "I like to take a 5/25 approach. That means if an asset class has moved an absolute 5% or a relative 25%, you should probably rebalance." Anyone know how frequently that typically is? I imagine it's around once a year.

Rebalancing seems to mean tweaking about once a year and keeping your asset allocations rather tightly on track.

But what do we call it if you only reset your asset allocation every decade or two--say, at significant life events like retiring?

The assumption that you should be constantly making minute adjustments, that a 55-year-old needs a different asset allocation from a 50-year-old, is something new. And, of course, patently bogus. Every mutual fund company's target-date funds imply that it's really important for a 55-year-old to have a shade more conservative allocation than a 50-year old, yet they differ wildly in what a 55-year-old's allocation should be. And in Vanguard's case, has changed course by as much as 15% during the rather short period of time--less than a decade--they've been in existence. If Vanguard doesn't know the right number to within 15%, then it can't be important to do 1% annual tuning adjustments.

In Ye Olde Days, say ten years ago, the various advice books and such illustrations didn't draw minute distinctions in asset allocation. You could live your whole life happily with just four or so different pie charts.

I think Burton Malkiel's "A Random Walk Down Wall Street" has four different model portfolios for different age ranges.

The old Vanguard LifeStrategy mutual funds have a name mplies that they are related to lifecycle investing. They don't self-adjust like target retirement, and they aren't named in terms of ages or years, but the working assumption is that you shift from one to another over time. And there are only four. Growth, Moderate Growth, Conservative Growth, and Income. Four sizes fit all ages.
Last edited by nisiprius on Wed Jun 09, 2010 9:39 am, edited 4 times in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
Doc
Posts: 9295
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Post by Doc » Wed Jun 09, 2010 9:27 am

nisiprius wrote:
As used in this forum, it refers to a fairly structured discipline, such as balancing on a schedule, with 6 months, 1 year, and 2 years being intervals that are frequently suggested; or, by a rule such as Larry Swedroe's: "I like to take a 5/25 approach. That means if an asset class has moved an absolute 5% or a relative 25%, you should probably rebalance." Anyone know how frequently that typically is? I'll bet it's around once a year.
It depends on your AA. The higher the stock allocation the more frequent the rebalancing. It's just arithmetic.
A scientist looks for THE answer to a problem, an engineer looks for AN answer and lawyers ONLY have opinions. Investing is not a science.

User avatar
nisiprius
Advisory Board
Posts: 39720
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

Post by nisiprius » Wed Jun 09, 2010 9:41 am

Doc wrote:
nisiprius wrote:That means if an asset class has moved an absolute 5% or a relative 25%, you should probably rebalance." Anyone know how frequently that typically is? I'll bet it's around once a year.
It depends on your AA. The higher the stock allocation the more frequent the rebalancing. It's just arithmetic.
Of course, but would you hazard a guess at an actual ballpark number? Is it once a month, once a year, or once a decade?
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.

Bobby Ingersoll
Posts: 59
Joined: Tue Oct 20, 2009 6:03 pm

Post by Bobby Ingersoll » Wed Jun 09, 2010 10:07 am

Is re-balancing from existing assets the only technique studied? I've heard the risk of this type re-balancing described in scenarios where a continuously falling asset consumes the portfolio itself.

However, most people spend at least half of their investing lives in the accumulation phase. I'd like to see a continuous re-balance method studied (especially within tax advantaged such as 401k where many people have a fixed amount going in every other week), only using new accumulation funds to continuously maintain the selected AA as closely as possible.

User avatar
Topic Author
Lbill
Posts: 4997
Joined: Thu Mar 13, 2008 11:25 pm
Location: Somewhere between Up and Down

Post by Lbill » Wed Jun 09, 2010 10:16 am

There's another inconsistency that troubles me. Despite controversy about Ayres & Nalebuff's lifecycle investment approach, I'm convinced that that their research substantiates the value of "time diversification." According to their findings, a constant percentage allocation to equities over intermediate to long time periods is mean-variance efficient; that is, results in superior risk-adjusted portfolio returns compared to alternative strategies. If this is true, then implementation would require periodic rebalancng to a policy target equity allocation. I'm not sure how to reconcile their research findings (which are quite robust because they are based on backtesting in U.S. markets since 1872, cross-validation in foreign markets, and Monte Carlo simulations) with the idea that rebalancing is not very important.
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard | | "You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs

Beagler
Posts: 3442
Joined: Sun Dec 21, 2008 7:39 pm

Post by Beagler » Wed Jun 09, 2010 1:08 pm

rcasement wrote:
Call_Me_Op wrote:I will remind you that the primary reason for rebalancing is not to enhance returns - but to reduce volatility.
exactly, thank you.
You won't have to search too far to find the term "rebalancing bonus" so I don't think that the presumption that rebalancing is for control of risk factors alone is considered a universal.
“The only place where success come before work is in the dictionary.” Abraham Lincoln. This post does not provide advice for specific individual situations and should not be construed as doing so.

User avatar
Mel Lindauer
Moderator
Posts: 30039
Joined: Mon Feb 19, 2007 8:49 pm
Location: Daytona Beach Shores, Florida
Contact:

Re: Wisdom of Jack Bogle on rebalancing and on bond funds

Post by Mel Lindauer » Wed Jun 09, 2010 2:49 pm

YDNAL wrote:quote="Lbill"]]I thought TBM was Mr. Bogle's favorite Bond fund! :)
Actually, IT has been his personal favorite for a long time. I remember moderating a Q&A with Jack (I think it was either in Denver in 2004 or Las Vegas in 2005) where he said he preferred the IT because of the high concentration of MBS in the Total Bond fund. So this really isn't something new; perhaps it's just that the press hasn't asked him about it before, just assuming that TBM was his personal favorite.
Best Regards - Mel | | Semper Fi

User avatar
House Blend
Posts: 4645
Joined: Fri May 04, 2007 1:02 pm

Post by House Blend » Wed Jun 09, 2010 2:52 pm

Doc wrote:
nisiprius wrote:
As used in this forum, it refers to a fairly structured discipline, such as balancing on a schedule, with 6 months, 1 year, and 2 years being intervals that are frequently suggested; or, by a rule such as Larry Swedroe's: "I like to take a 5/25 approach. That means if an asset class has moved an absolute 5% or a relative 25%, you should probably rebalance." Anyone know how frequently that typically is? I'll bet it's around once a year.
It depends on your AA. The higher the stock allocation the more frequent the rebalancing. It's just arithmetic.
Actually it's a 50/50 portfolio that is most likely to trigger rebalancing.

Depends on exactly how you want to formulate the question, but let's say you've got x% stocks and (1-x)% bonds, and that's your target AA. Your plan is to rebalance whenever stocks reach (x+5)% or (x-5)%.

Question: how much must the stock market move (up or down), percentage-wise, to trigger one of your rebalancing bands, and which AA target value for x requires the least amount of stock movement to trigger?

I'll leave out the algebra, but it's easiest to hit the (x+5)% band when your equity target is x = 47.5% (this requires the stock market to move up only 22.16%), and it's easiest to hit the (x-5)% band when your equity target is x = 52.5% (this requires equities to drop 18.14%).

By contrast, with an 80/20 portfolio, you'll hit the upper band only if equities jump 41.66%, and the lower band if they drop 25%.

I'm surprised Nisiprius didn't point this out in his response, because he brought it up in a thread some years ago.

As to how often this going to happen, you probably need to restrict the discussion to retirees, or at least non-accumulators. A young accumulator will almost never need to rebalance (at least not by selling/exchanging), because new money overwhelms investment returns.

Come to think of it, decumulators can probably stay in balance in many situations just by strategic withdrawals. Only when the market really moves far and fast would you ever need to actually make exchanges.

Beagler
Posts: 3442
Joined: Sun Dec 21, 2008 7:39 pm

Re: Wisdom of Jack Bogle on rebalancing and on bond funds

Post by Beagler » Wed Jun 09, 2010 3:27 pm

MelLindauer wrote:
YDNAL wrote:quote="Lbill"]]I thought TBM was Mr. Bogle's favorite Bond fund! :)
Actually, IT has been his personal favorite for a long time. I remember moderating a Q&A with Jack (I think it was either in Denver in 2004 or Las Vegas in 2005) where he said he preferred the IT because of the high concentration of MBS in the Total Bond fund. So this really isn't something new; perhaps it's just that the press hasn't asked him about it before, just assuming that TBM was his personal favorite.
In his Little Book, does he recommend TBM or IT bond?
“The only place where success come before work is in the dictionary.” Abraham Lincoln. This post does not provide advice for specific individual situations and should not be construed as doing so.

User avatar
Mel Lindauer
Moderator
Posts: 30039
Joined: Mon Feb 19, 2007 8:49 pm
Location: Daytona Beach Shores, Florida
Contact:

Re: Wisdom of Jack Bogle on rebalancing and on bond funds

Post by Mel Lindauer » Wed Jun 09, 2010 3:39 pm

Beagler wrote:
MelLindauer wrote:
YDNAL wrote:quote="Lbill"]]I thought TBM was Mr. Bogle's favorite Bond fund! :)
Actually, IT has been his personal favorite for a long time. I remember moderating a Q&A with Jack (I think it was either in Denver in 2004 or Las Vegas in 2005) where he said he preferred the IT because of the high concentration of MBS in the Total Bond fund. So this really isn't something new; perhaps it's just that the press hasn't asked him about it before, just assuming that TBM was his personal favorite.
In his Little Book, does he recommend TBM or IT bond?
I don't recall, Beagler, and one of the problems with the "Little Book" series is that there's no index for easy reference.
Best Regards - Mel | | Semper Fi

User avatar
Rick Ferri
Posts: 8983
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri » Wed Jun 09, 2010 3:46 pm

During the 1990s, rebalancing lowered allocations to stock going into the market crash from 2000-2002. Over the past 10 years, rebalancing added over 1% excess return to diciplined investors who held balanced portfolios. So, rebalancing reduces risk in a bull market and increases return in a volatile market. I think those are two good reasons to rebalance.

Rick Ferri

dbr
Posts: 31155
Joined: Sun Mar 04, 2007 9:50 am

Post by dbr » Wed Jun 09, 2010 5:07 pm

Rick Ferri wrote:During the 1990s, rebalancing lowered allocations to stock going into the market crash from 2000-2002. Over the past 10 years, rebalancing added over 1% excess return to diciplined investors who held balanced portfolios. So, rebalancing reduces risk in a bull market and increases return in a volatile market. I think those are two good reasons to rebalance.

Rick Ferri
Rick, I am curious if, when you manage client portfolios, you attempt MVO or some type modeling to optimize the efficiency of said portfolios, or do you merely operate to stay on an AA target in a cost and tax efficient manner?

User avatar
Rick Ferri
Posts: 8983
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri » Wed Jun 09, 2010 5:40 pm

MVO is wonderful if you know what the risks, returns and correlations of asset classes will be in the future. That's not possible to know, or even probable to guess.

Our strategy is NOT to rely on MVO or low correlations, and to expect all asset classes have a correlation of 1.0 with each other (that's not realistic, but it's better to expect the worst). Based on this, we allocate, stay on an AA target, keep cost low, and be tax efficient.

We do know that certain asset classes are fundamentally different than one another in structure, and we do know that these fundamentally different asset classes have varying correlations with each other over time. We just don't know how much or for how long.

We also know that there are premiums paid over inflation on some asset classes (stocks and bonds) and not on others (commodities). So, we only stick with the asset classes that pay a premium over inflation. This ensures that the portfolios earn a real "eat-able" return over time.

That's the best anyone can do aside from getting lucky or being super human (and a lot of advisors sell their services by making believe they're super human!). All this is in my All About Asset Allocation book, the 2nd edition will be out within the month.

Rick Ferri

User avatar
stratton
Posts: 11083
Joined: Sun Mar 04, 2007 5:05 pm
Location: Puget Sound

Post by stratton » Thu Jun 10, 2010 3:22 am

Rick Ferri wrote:That's the best anyone can do aside from getting lucky or being super human (and a lot of advisors sell their services by making believe they're super human!). All this is in my All About Asset Allocation book, the 2nd edition will be out within the month.
What's new/different in this version?

Paul

User avatar
Topic Author
Lbill
Posts: 4997
Joined: Thu Mar 13, 2008 11:25 pm
Location: Somewhere between Up and Down

Post by Lbill » Thu Jun 10, 2010 8:32 am

Rick said:
During the 1990s, rebalancing lowered allocations to stock going into the market crash from 2000-2002. Over the past 10 years, rebalancing added over 1% excess return to diciplined investors who held balanced portfolios. So, rebalancing reduces risk in a bull market and increases return in a volatile market. I think those are two good reasons to rebalance.
I used Simba's spreadsheet to look at the returns for a portfolio invested 50/50 in Vanguard's Total Stock and Total Bond index funds.

For the decade of the 1990s (1990-1999) the CAGR for an annually rebalanced portfolio was 12.65% (rebalanced) and 13.46% (non rebalanced) and the average annual StDev were 9.28% (rebalanced) and 9.48% (non rebalanced). You would have ended up with 71% stocks and 29% bonds had you not rebalanced over that period.

How much did it hurt you if you continued to not rebalance through the 2000-2002 market crash? Well, for the period of 1990-2002, the CAGR of the annually rebalanced portfolio was 8.97% with an average annual StDev of 10.60%, while the non-rebalanced portfolio had a CAGR of 8.61% with an SD of 12.05%. If you didn't rebalance you would have ended up with an allocation of 54% stocks and 46% bonds - just about where you started in 1990.

Over the last 10 years (2000-2009), the CAGR for a rebalanced 50/50 portfolio was 3.60% and it was 3.38% for a non-rebalanced portfolio. The relative average annual SDs were 10.44% and 8.01% respectively.

I just don't see a difference that makes a difference in these data for either the 1990's or the 2000's. In the words of John Bogle:
That’s worth describing as “noise,” and suggests that formulaic rebalancing with precision is not necessary.

My personal conclusion. Rebalancing is a personal choice, not a choice that statistics can validate.
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard | | "You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs

User avatar
bob90245
Posts: 6511
Joined: Mon Feb 19, 2007 8:51 pm

Post by bob90245 » Thu Jun 10, 2010 9:04 am

Lbill wrote:I just don't see a difference that makes a difference in these data for either the 1990's or the 2000's. In the words of John Bogle:
That’s worth describing as “noise,” and suggests that formulaic rebalancing with precision is not necessary.

My personal conclusion. Rebalancing is a personal choice, not a choice that statistics can validate.
I agree that over the short term like 12 years, you likely won't see much of a difference. However, I agree with Bogle that it is much more useful to look at long term data. Bogle used 25 years.

So go back to Simba's database and see how much difference it would make for something like 25 to 30 years. Or what ever is the maximum the data would allow.
Ignore the market noise. Keep to your rebalancing schedule whether that is semi-annual, annual or trigger bands.

User avatar
bob90245
Posts: 6511
Joined: Mon Feb 19, 2007 8:51 pm

Post by bob90245 » Thu Jun 10, 2010 9:16 am

As you may guess, the results for rebalancing versus not rebalancing are very much period dependent. If you want to study this in more detail, I have already done the heavy lifting for you with the spreadsheet at this link:

http://www.bobsfinancialwebsite.com/dow ... #Rebalance
bobsfinancialwebsite.com wrote:To Rebalance or Not Rebalance

This Excel file has a “mountain” chart showing the growth of $1000 so you can compare the results of rebalancing each year or not rebalancing. View the combined growth of the S&P 500, Small-Cap Stocks, Intermediate-Term Government Bonds and 90-Day Treasury Bills by varying the asset allocation. You may vary the timeframe from 5 to 50 years. Data is from 1926 to 2009.
And for a look at how asset allocation changes with and without rebalancing, use the "Growth of $1000" spreadsheet at the following link:

http://www.bobsfinancialwebsite.com/dow ... tml#Growth
Ignore the market noise. Keep to your rebalancing schedule whether that is semi-annual, annual or trigger bands.

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

Post by Rodc » Thu Jun 10, 2010 9:30 am

Lbill wrote:Rick said:
During the 1990s, rebalancing lowered allocations to stock going into the market crash from 2000-2002. Over the past 10 years, rebalancing added over 1% excess return to diciplined investors who held balanced portfolios. So, rebalancing reduces risk in a bull market and increases return in a volatile market. I think those are two good reasons to rebalance.
I used Simba's spreadsheet to look at the returns for a portfolio invested 50/50 in Vanguard's Total Stock and Total Bond index funds.

For the decade of the 1990s (1990-1999) the CAGR for an annually rebalanced portfolio was 12.65% (rebalanced) and 13.46% (non rebalanced) and the average annual StDev were 9.28% (rebalanced) and 9.48% (non rebalanced). You would have ended up with 71% stocks and 29% bonds had you not rebalanced over that period.

How much did it hurt you if you continued to not rebalance through the 2000-2002 market crash? Well, for the period of 1990-2002, the CAGR of the annually rebalanced portfolio was 8.97% with an average annual StDev of 10.60%, while the non-rebalanced portfolio had a CAGR of 8.61% with an SD of 12.05%. If you didn't rebalance you would have ended up with an allocation of 54% stocks and 46% bonds - just about where you started in 1990.

Over the last 10 years (2000-2009), the CAGR for a rebalanced 50/50 portfolio was 3.60% and it was 3.38% for a non-rebalanced portfolio. The relative average annual SDs were 10.44% and 8.01% respectively.

I just don't see a difference that makes a difference in these data for either the 1990's or the 2000's. In the words of John Bogle:
That’s worth describing as “noise,” and suggests that formulaic rebalancing with precision is not necessary.

My personal conclusion. Rebalancing is a personal choice, not a choice that statistics can validate.
I have come up with similar numbers in the past. It is certainly possible that over 30 or 40 years the differences are larger, but then over that time period as you monitor your desire and ability to take risk you will in all likelihood have made adjustments for other reasons, so this is all moot on that time scale as well.

I rebalance with bounds, but all in all this looks like one of those situations which arises so often: it just does not matter very much (or has not in the past at least, future could be different), so pick a plan, any more or less reasonable plan, stick with it, and don't spend a lot of time or energy on it.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

User avatar
Rick Ferri
Posts: 8983
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri » Thu Jun 10, 2010 9:31 am

stratton wrote:
Rick Ferri wrote:That's the best anyone can do aside from getting lucky or being super human (and a lot of advisors sell their services by making believe they're super human!). All this is in my All About Asset Allocation book, the 2nd edition will be out within the month.
What's new/different in this version?

Paul
There must be SOMETHING new because it's 40 pages longer!

There's a new chapter on when to change your allocation. More information on different ways to set asset allocation including the addition of a modified age weighted method (your age in bonds modified). Expanded areas in foreign bonds, commodities, and currencies. More information on dynamic correlations and the greater emphasis on the fact that the optimal asset allocation cannot be known in advance.

Rick Ferri
Last edited by Rick Ferri on Thu Jun 10, 2010 9:32 am, edited 1 time in total.

User avatar
Topic Author
Lbill
Posts: 4997
Joined: Thu Mar 13, 2008 11:25 pm
Location: Somewhere between Up and Down

Post by Lbill » Thu Jun 10, 2010 9:32 am

I agree that over the short term like 12 years, you likely won't see much of a difference. However, I agree with Bogle that it is much more useful to look at long term data. Bogle used 25 years.
I studied shorter periods because Rick cited these data to document the benefits of re-balancing. I agree that longer periods would provide more meaningful statistical results. One problem with this, however, is that shorter periods such as 10 years may be more representative of what investors do in real life. How many investors are out there who hold a fixed allocation for 25 years or longer and actually rebalance using the same rebalancing rule over that period of time? I'd guess 0
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard | | "You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs

User avatar
Rick Ferri
Posts: 8983
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri » Thu Jun 10, 2010 9:35 am

Lbill wrote:How many investors are out there who hold a fixed allocation for 25 years or longer and actually rebalance using the same rebalancing rule over that period of time? I'd guess 0
The data suggests that most individual investors lack the discipline to hold an asset allocation for even 3 years. This is why advisors exist. :D

Rick Ferri

dbr
Posts: 31155
Joined: Sun Mar 04, 2007 9:50 am

Post by dbr » Thu Jun 10, 2010 10:23 am

Rick Ferri wrote:MVO is wonderful if you know what the risks, returns and correlations of asset classes will be in the future. That's not possible to know, or even probable to guess.

Our strategy is NOT to rely on MVO or low correlations, and to expect all asset classes have a correlation of 1.0 with each other (that's not realistic, but it's better to expect the worst). Based on this, we allocate, stay on an AA target, keep cost low, and be tax efficient.

We do know that certain asset classes are fundamentally different than one another in structure, and we do know that these fundamentally different asset classes have varying correlations with each other over time. We just don't know how much or for how long.

We also know that there are premiums paid over inflation on some asset classes (stocks and bonds) and not on others (commodities). So, we only stick with the asset classes that pay a premium over inflation. This ensures that the portfolios earn a real "eat-able" return over time.

That's the best anyone can do aside from getting lucky or being super human (and a lot of advisors sell their services by making believe they're super human!). All this is in my All About Asset Allocation book, the 2nd edition will be out within the month.

Rick Ferri
Thanks

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

Post by Rodc » Thu Jun 10, 2010 10:49 am

Rick Ferri wrote:
Lbill wrote:How many investors are out there who hold a fixed allocation for 25 years or longer and actually rebalance using the same rebalancing rule over that period of time? I'd guess 0
The data suggests that most individual investors lack the discipline to hold an asset allocation for even 3 years. This is why advisors exist. :D

Rick Ferri
Good advisers.

The other 99.4% (give or take) exist to make as money off clients as they can.

:)
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

User avatar
Topic Author
Lbill
Posts: 4997
Joined: Thu Mar 13, 2008 11:25 pm
Location: Somewhere between Up and Down

Post by Lbill » Thu Jun 10, 2010 2:49 pm

Good advisers.

The other 99.4% (give or take) exist to make as money off clients as they can.
Rick being one of the good advisors, I think most would agree. :thumbsup
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard | | "You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs

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

Post by Rodc » Thu Jun 10, 2010 2:54 pm

Lbill wrote:
Good advisers.

The other 99.4% (give or take) exist to make as money off clients as they can.
Rick being one of the good advisors, I think most would agree. :thumbsup
Yes
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

Call_Me_Op
Posts: 7423
Joined: Mon Sep 07, 2009 2:57 pm
Location: Milky Way

Post by Call_Me_Op » Fri Jun 11, 2010 3:58 pm

Lbill wrote:
I will remind you that the primary reason for rebalancing is not to enhance returns - but to reduce volatility.
But, where's the beef? For example, over the 10-year period of 2000-2009 a portfolio of 60% TSM (total stock market) and 40% TBM (total bond market) that was rebalanced annually had an annualized volatility (SD) of 12.69%, but a non-rebalanced 60/40 portfolio had an SD of 10.22%, which was smaller by 24.2%. Like many commonly believed truisms, this one turns out to be false also when subject to scrutiny. As Mark Twain said: "It's not what we believe that gets us in trouble, as much as what we believe that ain't so."
You know what they say about statistics, right? Why pick that 10-year period? The relationship should hold in general, but not necessarily during any particular 10-year interval.
Best regards, -Op | | "In the middle of difficulty lies opportunity." Einstein

simplecd
Posts: 158
Joined: Tue Dec 15, 2009 10:04 pm

Post by simplecd » Fri Jun 11, 2010 6:37 pm

IMO, the conventional wisdom about controlling risk with rebalancing is wrong. Consider a portfolio that is 50/50 TSM- Bonds. The stock market crashes and loses 50% while the bond market remains steady. Your allocation is now 33% stocks and 67% bonds. The markets have repriced the value of stocks relative to bonds. If I now rebalance to 50/50 by selling bonds to buy stocks, I'm actually increasing the risk of my portfolio, not lowering it.


When stocks loose 50% of their value i would argue that they become less risky... so rebalansing at that point is a decent approach... although you've got to consider your time horizon...

User avatar
stratton
Posts: 11083
Joined: Sun Mar 04, 2007 5:05 pm
Location: Puget Sound

Post by stratton » Sat Jun 12, 2010 1:37 am

Rick Ferri wrote:
Lbill wrote:How many investors are out there who hold a fixed allocation for 25 years or longer and actually rebalance using the same rebalancing rule over that period of time? I'd guess 0
The data suggests that most individual investors lack the discipline to hold an asset allocation for even 3 years. This is why advisors exist. :D
Tax loss harvesting has mine somewhat drifted from my IPS. It may be low enough I'll tip it back at some point. I do need to make sure I don't drift too far away though.

One of the things I've noticed as I've become more knowledgable my ideas about risk and what I want in a portfolio have changed. Slightly simpler is better than too many moving parts. More expensive funds if there isn't an index don't bother me either, but it has to be for a *small* amount of the asset allocation.

Paul

User avatar
spam
Posts: 907
Joined: Tue Jun 10, 2008 9:47 am

Post by spam » Sat Jun 12, 2010 5:56 am

Maybe the problem is with Jacks asset allocation. With the exception of the bonds, the other asset classes tend to be pretty strongly correltated.

Image

Adding a REIT and Commodity index to the S&P and International index did much better. Another difference would be that these indexes (above) were held in equal amounts as opposed to Jack's tested portfolio. On the chart above, portfolio AB would be his, but the return would have been less due to his larger allocation to the S&P 500.

This table was taken from the book "Asset Allocation 4th Edition" written by Roger C Gibson.

User avatar
Rick Ferri
Posts: 8983
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri » Sat Jun 12, 2010 9:19 am

If only Gibson (or anyone) could show us this chart from 2010-2040. That information would be very helpful.

Here's my guess. It's as good (or bad) as anyone else.

Rick Ferri

User avatar
Topic Author
Lbill
Posts: 4997
Joined: Thu Mar 13, 2008 11:25 pm
Location: Somewhere between Up and Down

Post by Lbill » Sat Jun 12, 2010 9:34 am

Rick - I'd like to bet you another Starbucks on those 30-year asset return projections. See you in 30. :)
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard | | "You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs

Lucio
Posts: 286
Joined: Wed Apr 08, 2009 10:31 pm

Post by Lucio » Sat Jun 12, 2010 9:45 am

Rick Ferri wrote:
Here's my guess. It's as good (or bad) as anyone else.

Rick Ferri
Rick,

Why did you choose to use the Fama-French 5 factor model for the risk premia of the equity asset classes instead of the 3 factor model? I am referring, in particular, to the last chart on the web page that you linked to.

Lucio

Post Reply