Value added by continuous rebalancing?

[charlie]

Hi all,

We all know that rebalancing at least once per year
adds value to a portfolio by forcing you to "buy low
sell high".

My question for the group is: Is there addidional
value added by a fund like Vanguard's Balanced
Index Fund which rebalances continuously? One
could argue, I suppose, that you would miss out
a little on the years when stocks do really well.
On the other hand, when stocks tank you are
buying the dips.

Comments anyone?

Cheers!

Charlie

[Brat]

Post the question on the Fund Alarm BB, many of the folks who frequent know their stuff.

[wabmester]

The answer appears to be yes. There was an empirical study on rebalancing frequency sited in this newsletter from Bill Bernstein:

http://www.efficientfrontier.com/ef/996.pdf

The article also explains how to calculate the rebalancing bonus from SD and asset covariance, which also answers the question about rebalancing posted by Cut-throat in another thread.

[wabmester]

Argh, Bernstein revisited this question and came up with a different answer here:

http://www.efficientfrontier.com/ef/100/rebal100.htm

Basically, I wouldn't sweat the frequency. One thing he doesn't touch upon is the tax impact and trading expenses associated with more frequent rebalancing.

Although it's somewhat apples and oranges, compare the perform of Vanguard's Balanced Index to their Tax-Managed Balanced fund, for example.

[Ted]

Rebalancing continuously would really be taking a good thing to an extreme that would generate huge trading costs for any individual who tried to do it, and for the overall financial system if it became common practice.

In theory, rebalancing may either increase returns or decrease them, depending on whether a person is adding to their assets or drawing them down. Continuous rebalancing would presumably accentuate the gains or losses.

Basically, the down side to rebalancing is this. In the absence of rebalancing, the assets in a portfolio will shift to the asset class that has the highest long-term return. Historically, this has been stocks, and presumably will continue to be in the future. Therefore, if a person is accumulating assets and rebalances, they are on the average lessening the percentage of their portfolio in stocks, and realizing a lesser rate of return as the result. The benefit of this is a reduction in the volatility of the portfolio's returns, and this benefit translates into a higher sustainable withdrawal rate for people (retirees) who are drawing down their assets.

[charlie]

Thanks for your thoughtful comments!

I have some observations on this subject
myself.......

1. Unless I am mistaken, a balanced index fund like
Vanguard's rebalances continuously but there
is no taxable event consequence to the fund
investors. Ted is right in saying that the tax
consequences would be awful if one tried to
rebalance separate taxable funds too often.

2. The main advantage of rebalancing appears to
be reduction of portfolio risk. If you like the
asset classes and mix of a low cost balanced
index fund like Vanguard's Balanced Index or
Target Retirement or Life Strategy, then go
for it. The automatic rebalancing takes away
one more worry. .

3. If I can get up the energy, I am thinking about
doing a long term comparison of the returns of
Vanguard's Balanced Index (60/40 Total Stock
Market/Total Bond Market) vs. holding each
fund separately and rebalancing annually.

Cheers!

Charlie

[ats5g]

1. Unless I am mistaken, a balanced index fund like
Vanguard's rebalances continuously but there
is no taxable event consequence to the fund
investors. Ted is right in saying that the tax
consequences would be awful if one tried to
rebalance separate taxable funds too often.

I believe you're mistaken. Try comparing the "tax-cost ratio" of Vanguard's Balanced fund and Vanguard's Tax Managed Balanced fund over at Morningstar. Everytime the Balanced fund takes a capital gain, it has to pass it onto fund shareholders. Just like it has to pass on dividends to shareholders.

I totally agree w/ #2. Vanguard's Balanced and hybrid funds are excellent for those not that interested in managing their portfolio that much. My mom uses the Balanced index in her 403(b).

#3 sounds interesting.

- Alec

[ats5g]

Hey Charlie,

Return data on the Total Stock and Balanced Index only go back to 1992. That's when they were created. Using the full year 1993-2003 data, here's what I got.

VTSMX:

Annualized return: 10.49%

VBMFX:

Annualized return: 6.96%

VBINX:

Annualized return: 9.36%

60 VTSMX /40 VBMFX mix (rebalanced annually):

Annualized return: 9.52%

60/40 weighted average mix of 1993-2003 annualized returns of VTSMX (60%) and VBMFX (40%):

Return: 9.08%

Rebalancing Bonus:

VBINX: 0.29%

60/40 rebalanced annually: 0.44%

- Alec

[fire5soon]

I can't find the reference right now, but I believe I read where Bill Bernstein said rebalancing about every 2 years would work best for the average investor. I'm going from pure memory here so **insert disclaimer here**

[charlie]

Hi ats5g,

Thanks for running the numbers. I agree with your
results within rounding error except for the case
of a 60/40 mix without rebalancing. In that case,
I got a total annual return of about 9.21%.

It was an interesting exercise, but does not prove
anything since another 11 year period might tell
another story. The message of this period is that
rebalancing helped (a little) in the bad years.

The year by year numbers show that the "no rebalance"
approach leads the pack With Balanced Index bringing
up the rear (slightly) through 2000. By the end of
2003, my numbers show that for a $100,000 initial
investment, the end result was:

Rebalance Annually = $271,917
Balanced Index = $267,730
No rebalance = $263,640

I plan to call Vanguard tomorrow to confirm
your comment on the tax consequences of
holding a Balanced Index fund. Thanks for your
heads up.

Cheers!

Charlie

[Ted]

If you folks really want to make this exercise complicated, you can also investigate Vanguard's other "balanced" funds, such as the STAR fund and all of the Life Strategy funds.

Intuitively, I believe that the (1) the differences in the return/risk ratio will be small and that (2) even if these differences are not small based on historical results, it is not safe to assume that the same relative performance will be repeated in the future.

As JWR45 says, Have Fun.

[charlie]

Alec,

I checked the prospectus for Vanguard's Balanced
Index Fund and found the following on capital
gains distribution:

1998 ....... $.14, NAV = $18.48
1999......... $.14, NAV = $20.22
2000 ........ $.10, NAV = $19.08
2001......... $.025, NAV = $17.86
2002......... $ .00, NAV = $15.65

I don't know how they do it, but daily rebalancing
does not appear to cause excessive capital distributions.

I have not verified this yet, but I suspect that you would
have seen similar capital gains distributions with just
the Total Stock Market Index Fund.

Regards,

Charlie

[AltaRed]

I believe it is because daily rebalancing often involves losses almost as much as it involves gains. At the end of the year, most of the positives and negatives balance out leaving only the residual capital gains (in a positive year). The net effect wouldn't be much (if any) different than infrequent re-balancing.

I have absolutely nothing to base this on, but I could see an argument against a benefit of frequent rebalancing. By continuously selling off and reducing your stake in the hot funds/stocks, you would not be able to fully take advantage of the gains during their hot streak. I don't know if their would be a corresponding balance on the other end though.

[unclemick]

That's the classic momentum argument - backed with stop/loss points or other sell discipline.

It's as old as Ford versus Chevy pickups - both have committed partisans.