"FranksPlace2" <FranksPlace2[at]gmail.com> wrote in message
news:1173105722.335033.90990[at]j27g2000cwj.googlegroups.com...

- quote -

Usually we think of highly aggressive portfolio allocations causing loss of
sleep. But this is a good example of an allocation that is probably too
conservative. We should all be striving for an allocation that allows us to
sleep at night.

Elizabeth Richardson

- quote -

A question regarding that strategy - using the 4% rule, 5 years is about
20% in cash/bonds. Do you reduce this by the cash dividends of your
stock portfolio? 2% (for example) yield would provide half the 4% you'd
need for spending, and allow you to have 10% in cash/bonds to make up
the difference.
JOE

On Mar 4, 10:00 am, darknes...[at]yahoo.com wrote:
- quote -

However I occasionally toss and turn at turn at night thinking about
the gain foregone by having this much in bonds.

Frank

On Mar 2, 4:50 pm, "FranksPlace2" <FranksPla...[at]gmail.com> wrote:
- quote -

Remember the risk (volatility) of holding *any* equity fund is much
greater than the risk of holding most fixed income (perhaps with the
exceptions of emerging market and high yield bonds or convertible
bonds).

Annual volatility of returns on bond funds are on the order of +/- 10%
(worst year in the last 20). Annual returns on equity funds are on
the order of +/- 30% (c. -40%, if you took the worst 18 month period
since 1985, I believe)-- in practice although there have been -20%
days in the stock market (and there would be again, if, say, Al Quaida
manages to detonate a radiological bomb in central London or NYC, or a
major hedge fund implodes), markets usually rebound a bit after that.

So aggressive stock funds or not, your real risk comes from holding
stocks/equities. Whether a fund is 10 or 20% riskier than the index
as a whole is somewhat academic.

To hedge what I have said a bit, funds focusing on high risk sectors
(Nasdaq, Emerging Markets) can have volatility which is a multiple of
that of your average equity index fund. In 1994 during the Mexican
crisis, and again in 1997during the SE Asian crisis, emerging markets
globally fell by something like 50%. And Nasdaq bottomed at less than
1/3rd of its peak 2.5 years earlier. (I'm pulling those numbers out
of my head, without googling them).

Interestingly, when those 'high risk'/ bubble sectors and funds blow
up, they take years to recover, whereas the broader markets can
recover more quickly. What happens with the funds is they usually get
closed/ merged with other funds, so the performance record is erased
from the Fund Management company's track record.

On Mar 2, 11:38 am, "kastnna" <kast...[at]auburnalum.org> wrote:
- quote -

kastnna,

Theoretically you are correct. But it hasn't happened yet because I
am only at half my limit and my agressive portfolio does not perform
that much better than my market portfolio.

Frank

beliavsky[at]aol.com wrote:
- quote -

B- you left out the verb in that last sentence. If you meant
"understand" - c'mon man! If you meant "noticed", you're right -- I
missed on first read which correlations you're talking about...one can
envision combinations where rebalancing lowers returns, though
practically speaking, I don't think of them as likely to occur in a
well-designed portfolio (begging the question a bit in defining
"well-designed"). If you meant "would be able to elaborate further given
the short-post leash we're on" my answer is -- "no, I can't."

-Tad

On Mar 2, 4:17 pm, beliav...[at]aol.com wrote From your reply I wonder if you the distinction between cross
- quote -

I'm pretty sure I don't know, so lay the knowledge on me brotha!

PS - laymans terms please.

Vanguard Institutional Investors site has an informative paper titled
"Portfolio rebalancing in theory and practice" at

institutional.vanguard.com/VGApp/iip/Research?IIP_INF=ZZ

Do note that the most effective way to implement your chosen
rebalancing strategy is with new money (avoiding sales and
shence avoiding tax costs).

David

On Mar 2, 4:37 pm, Tad Borek <bore...[at]pacbell.net> wrote:
- quote -

Assets returns could have low SIMULTANEOUS or cross- correlations to
each other, but if individual asset returns
have positive (negative) SERIAL correlations, rebalancing will tend to
reduce (increase) returns. The cross-correlations and serial
correlations are distinct quantities.

These points may seem academic, but they are relevant to the question
at hand. From your reply I wonder if you the distinction between cross
and serial correlations. In the time series literature the latter are
called autocorrelations.

beliavsky[at]aol.com wrote:
- quote -

B- that's sort of an academic point...if the investments you hold are
highly + correlated then implicitly, you won't have a reason to
rebalance. If kastna holds a Wilshire 5000 index fund and a Russell 3000
index fund he can rebalance in 2027 when the negligible difference in
returns finally takes its toll.

-Tad

On Fri, 2 Mar 2007 09:25:52 -0600, beliavsky[at]aol.com wrote:


- quote -

snip

We have not relaxed our guidelines with respect to lengthy posts.
This one got by us because the poster was on the fast approval list at
the time.

-------------
As usual, please do not respond to this post. Our posting guidlines
can be found in the weekly post, "Posting to
misc.invest.financial-plan" on the newsgroup.

Thank you.


-HW "Skip" Weldon
Columbia, SC

On Mar 2, 10:50 am, "FranksPlace2" <FranksPla...[at]gmail.com> wrote:

- quote -

I don't understand. Suppose you've got 20K in "aggressive MFs" and 80K
in all other investments. Hypothetically the agressive funds double in
value and the others stay the same. You just went from a 20% to a 33%
allocation of aggressive holdings. So one of your rules is violated:
1) you have greater than 20% agressive holdings or 2) you have to
rebalance.

Following the risk/return trade-off you also increase the "reasonable
risk" of your portfolio as the % of agressive holdings increases.

Sorry for the confusion, but your statement lost me.

On Mar 1, 12:47 pm, "kastnna" <kast...[at]auburnalum.org> wrote:
- quote -


My goal is to maximize growth subject to reasonable risk. To protect
against excessive risk, I invest only in mutual funds, limit my
portfolio of agressive mutual funds to less than 20% of my portfolio
and maintain five years of income in bonds or near bonds.

My agressive mutual funds are about 10% and an increase in the value
of my equities does not automatically increase the amount I maintain
in bonds.

On Mar 1, 6:49 pm, Tad Borek <bore...[at]pacbell.net> wrote:
- quote -

I think this advice is oversimplified to the point that it is more
wrong than right. IF, and only if asset returns have NEGATIVE serial
correlation at a certain time scale can one be confident that
rebalancing at that time scale can increase returns. This should be
studied empirically. I have seen studies saying that at the 1 year
time horizon, there is some positive serial correlation in asset
returns, in which case I would not expect yearly rebalancing to boost
returns.

- quote -

Yes, except for the "fixed cost" of the time that one would incur for
daily monitoring. I work for a fund that will probably move to daily
rebalancing when our systems are advanced enough that the time cost of
monitoring and doing small trades is close to zero. Right now we
monitor weekly.

In the presence of proportional transaction costs, due to either bid-
ask spreads or capital gains taxes, some research supports the
existence of a "no-trade region" with only enough trading done to
bring the portfolio within the band. For example, if one wants 60/40
stocks bonds with a no-trade region half-width of 5, and one monitored
the allocation monthly, if the allocation moved to 70/30 because of a
big rise in the stock market one would shift money from stocks to
bonds to reach 65/35. Trading to the nearest edge of the band rather
than all the way to the target has been found to be optimal (requires
less trading without sacrificing too much risk-adjusted return) under
certain assumptions, as discussed in a paper

http://www.phildybvig.com/papers/tcost1per2.pdf
Mean-Variance Portfolio Rebalancing with Transaction Costs
Philip H. Dybvig
Washington University in Saint Louis
Summary
Transaction costs can make it unprofitable to rebalance all the way to
the ideal portfolio. A single-period analysis using mean-variance
theory provides many interesting insights. With fixed or variable
costs, there is a non-trading region within which trading does not
pay. With only variable costs, any trading is to the boundary of the
non-trading region, while fixed costs induce trading to the interior.
With costly trading in futures and underlying, it might be optimal to
use a synthetic equity strategy or an asymmetric futures overlay
strategy that takes better advantage of extra expected return than
traditional futures overlays.

Another paper

"Optimal Rebalancing Strategy Using Dynamic Programming for
Institutional Portfolios"
by Sun et al.
http://ssg.mit.edu/~waltsun/docs/rebalancingSSRN05.pdf

says the same thing:

"Previous research on dynamic strategies for asset allocation [17] has
established the existence of a notrade
region around the optimal target portfolio weights [9]. If the
proportions allocated to each asset at
any given time lie within this region, trading is not necessary.
However, if current asset ratios lie outside
the no-trade region, Leland has shown that it is optimal to trade but
only to bring the weights back to
the nearest edge of the no-trade region rather than to the target
ratios. The optimal strategy has been
shown to reduce transaction costs by approximately 50%. However, the
full analytical solution involves
a complicated system of partial differential equations in multiple
dimensions."

- quote -

smartmoney did a magazine article on this years ago... back tested to
1929 or something like that.

The conclusion was "best" returns did let winners ride for 2-3 year
periods. Not sure if magazine articles are available via the web...

On Mar 2, 12:58 am, "BeachBum" <bbgh...[at]yahoo.com> wrote:
- quote -

Looking for investment in Mutual Funds in India-Please contact [at]
sreejankumar[at]gmail.com or call [at] 91 9847147171


======================================= MODERATOR'S COMMENT:
Please trim the post to which you are responding. "Trim" means that except for a FEW lines to add context, the previous post is deleted.

On Mar 1, 3:49 pm, Tad Borek <bore...[at]pacbell.net> wrote:

- quote -

I often see this cited as an advantage of doing rebalancing. Yet, if
one starts with a portfolio that contains a mix of stocks and bonds,
one will usually achieve a higher return without ever rebalancing
(when considering long periods of time 20-30 years).

While I can't cite anything specific right now, I remember asking
this question during the seminar and getting the answer that
the portfolio without rebalancing would do better than one that
balanced annually. Intuitively this makes sense because the
"risky" part of the portfolio would tend to grow faster and the
effect would compound if left to itself.

However, if we are in the business of managing risk, then the
objective is different. In that case rebalancing would make
sense.

Anoop

kastnna wrote:
- quote -


Well first off if you don't rebalance you're probably leaving money on
the table. You miss out on the benefit that can result from the natural
"sell high, buy low" process that comes with rebalancing.

I believe the principal reason for time-based rebalancing seen in
studies of asset allocation is...it's easy to run the data. I think
that's why you see models showing quarterly, annual, etc., rather than
"any asset class more than X% off of target." Not that the latter is
impossible to model, but it's a lot harder!

In practice I think the "X% off" method makes the most intuitive sense.
The point is to take money off the table from your above-target asset
classes and shift into your below-target ones. Logically the more
off-target you are, the more important it is to do this. There's really
no significance to time, given this rationale.

You're right though that tax effects factor into it -- you should assess
all investment changes on a net-of-cost basis, and alongside
commissions, taxes are a cost. It's a lot easier to avoid the problem
when there's new money coming in to invest, and when there's qualified
money to shift around. When big sales are required you do them in a 401k
or IRA, to the extent possible. If it's all taxable money, a solution is
NOT reinvesting dividends and distributions so you have free cash
available to do this kind of rebalancing. I find that a combination of
these often allows for zero-taxable-gain rebalancing.

Last point...advanced investors only =)...the easiest method of tax
accounting for mutual funds is "average cost," and I think most people
use it because it's what the custodians report. If you use it once,
you're required to use it forever for that holding. BUT..if rebalancing
is a big part of your strategy, and you're willing to do the work,
consider using specific-ID with all of your mutual fund sales. That way
you can flag your highest-basis lots for sales when doing rebalancing.
You can even generate losses while doing this, it depends on the
portfolio mix and frequency of investing and of distributions.
Specific-ID has the most flexibility; even if you want "average cost" as
your basis you can accomplish that with careful lot selection.

Here I have to put in a rare jab at Vanguard. I've heard through a
client that they give a big "huh?" when you ask for written confirmation
of a specific-ID sale request...even after trying to escalate the
question to a manager. This, despite the IRS guidance requiring
documentation of specific-ID sales. If you're a pro you can probably
accomplish this through your order-entry system or some other means. I
don't know how to solve the problem if you're a retail investor at
Vanguard, or any other firm that doesn't provide this.

-Tad

"kastnna" <kastnna[at]auburnalum.org> wrote in message
news:1172774777.461922.278580[at]s48g2000cws.googlegroups.com...
- quote -

First, most of our assets are in tax-deferred accounts, and there are no
costs associated with rebalancing. Some of our funds are in Vanguard's
LifeStrategy series, bought specifically so I don't have to think about
rebalancing. BUT, about half of our retirement funds are in an account with
about 6 funds. I have consciously not done any rebalancing for several
years. While things are getting further from their target allocations, the
fund furthest from its target percentage is a bond fund. While I monitor
this account regularly, I have consciously chosen not to take from the
winners, just to plunk it into this bond fund. I never thought I'd be one
for market timing, BUT taking money out of the stock market and putting it
into the bond market seems pretty stupid, and has seemed so for the past 3
years.

Elizabeth Richardson

On Mar 1, 1:47 pm, "kastnna" <kast...[at]auburnalum.org> wrote:
- quote -

I rebalance twice. June/July timeframe and Dec/Jan timeframe.

In June/July I tweak contribution percentages. My typical allocation
is 45% large cap, 15% mid cap, 15% small cap, 15% international large
cap and 10% small cap. If I think something needs tweaking (less
money or more), I might switch percentages by 5%.

In December/January I look to see if contributions affected the
"allocation" and buy/sell as needed to meet the requirements.

A few points, some funds in 401ks charge short term redemption fees,
so it's important to watch that. By doing twice a year, I know I
think carefully about rebalancing twice each year. I could not tell
you my allocation "today", but this strategy forces me to look deeply
twice a year.