- quote -


I completely agree on both topics. I have learned something new as in
the past I have invested quarterly in my and my wife's Roth IRAs
believing that doing so was preferable to investing all at once or
less frequently. Now, I will attempt to switch as much as possible to
ETFs to lower costs (not that most of mine are high, anyway, but even
small differences matter long-term) and invest perhaps twice per year
as close to the beginning of the year as possible to have the funds
invested the longest possible and also to minimize transaction costs.
These are all retirement funds for maybe 25 years from now.

On Jul 23, 11:35 am, mjgrinn...[at]gmail.com wrote:
- quote -

What is your time horizon? If you plan on leaving this money invested
for a long time, you should have much less concern for the markets
highs and lows. Even if it falls right after you invest, it should
recover given a long enough time horizon (past performance not
indicative of future performance yada yada...). Many of the tech
investors of 2000 that lost 1/2 of their investment have since
returned to their original levels (or better). If you need the money
in the near future (2-5 years) then maybe DCA is best anyway. It will
mitigate an early market downturn. But for long term investments,
don't worry about it.

As for the "forced" DCA and how frequently you should purchase topic:
As I said above, it depends on how much you plan to invest. I try to
keep transaction costs between 0.5 and 1%. For $10 transaction costs
this means I don't invest until I have between $1000 and $2000 saved
up. If you can save that much in a week, then invest every week. If it
takes six months to save up that much then purchase every six months.
There's no universal answer, its a function of investment amount and
transaction costs.

- quote -

I suppose it's again a question of the individual strategy. If one's
strategy is to buy and hold for the long-term versus trying to time
the market then I suppose it's a lot less counterintuitive to invest
as much as one has to invest regardless of where the market is at the
time. A lot of the literature talking about why not to invest when
the market is high seems to be written by and for 'professional'
investors who probably change positions more frequently in week then I
do in 2 years.

- quote -

Right Will. MJ - the link earlier in this thread contained seven further
links to articles all concluding the same thing. My acceptance of their
conclusion doesn't change based on the market being at a high, because
as Will offered, that suggests market timing. When one of my clients
found herself with a lump sum to invest, I quickly had to accept the
fact that she felt similar to hoe you appear to feel, that some how
there's more risk at such a high. Knowing when to give up, we invested
the money about 5% at a time over a two year period. Look at a long term
graph of the S&P. She missed much of the move from 1996-8. Most of the
80s and 90s had the market at a 52 week high, for whatever that's worth.

JOE
JoeTaxpayer.com

mjgrinnell[at]gmail.com wrote:

- quote -

Why not? Joe doesn't seem to be the market-timing type.

-Will

- quote -

Yes, my point really was that most people are forced to DCA as they do
not have a big pile of cash. In the end, my question (or now it's not
really question but rather a point) is what is the optimal investment
strategy for the average investor. In most cases where a workplace
account is in play there is not much choice in the matter -- the
employer routes a certain percentage of the gross income to
investments each time the individual gets paid. This is DCA whether
the investor likes it or not and is optimal in that the amount the
individual has to invest is indeed invested as soon as it becomes
available, the pros/cons of DCA notwithstanding.

Then, there is the case of an IRA/Roth IRA where there is more choice
in the frequency/amount to invest. What I suppose the article Joe
Taxpayer quoted is saying is that the best way to invest, if the
individual had the funds at the time, would be to drop the entire
maximum allowed contribution in the market at the beginning of the
year. This would minimize the cost in an ETF as there is a single
transaction. I can't imagine that Joe would advise to follow this
strategy is the market is at a 52 week high at the time, though.

On Jul 18, 9:44 pm, joetaxpayer <joetaxpa...[at]nospam.com> wrote:
- quote -



Hi Joe,

I would think that the scenario the authors of the article you site
would be seen relatively infrequently. That is, how often do people
receive a large, lump sum of money they have to invest? Most people
have some amount they have from each paycheck of periodically that
they have to invest. The other option is to not invest the small sums
they have and save them all until they are large sums and then invest
them. But, I suppose the gist is that most people who are ostensibly
dollar cost averaging are really just investing whatever funds they
have available at the time. Not sure if that makes sense or not, but
I guess that relatively speaking the average Joe is investing his lump
sum each time he gets a paycheck instead of immediately when he
receives his lump sum from his rich uncle.


======================================= 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 Jul 18, 4:08 pm, mjgrinn...[at]gmail.com wrote:

- quote -

It would help to know how large an investment we are discussing. In
your case, it sounds like you are "forced" to DCA. As you make the
money you invest it. You are not sitting a large pile of cash that you
are slowly integrating into the market. If that is the case, you have
no choice but to DCA (at least to some extent). Transaction costs are
fixed, not variable. What you should do largely depends on how much of
an impact these fixed costs have on your investment. A good goal would
be to keep transaction costs below a certain level (perhaps 1% or
0.5%). This may mean investing weekly, monthly, quarterly, etc, etc.
It is dependent on your specific case.

It helps to look at the situation in extremes. Assume a $10 trade
charge. If you are depositing $20 a week, you would lose half your
money just to transaction costs if you purchased ETFs weekly. You
would need a 100% annual return just to get back to even! On the other
hand, if you were depositing $20,000 a week, I would say go for it. In
the latter case, fixed costs only erode 0.05% of your investment. That
should be easy to recover.

- quote -

Actual Dollar cost averaging has been proven inferior to investing a
lump sum immediately.
http://www.altruistfa.com/readingroo...rCostAveraging
It feels better, and would keep one from investing too much money the
month the market peaks, but the link has multiple articles supporting
this view.

What you propose is a way of putting new money into the funds or ETFs.
I'd think one can 'do the math' given the values for the variables; $$
amount per period, commission cost for ETF, assumed market return,
current return on cash. This may turn into an "angels on the head of a
pin" discussion, as it seems to me that on $1000, MM = 5% and even with
assumed stock return of 8%, that's just $30/yr or $2.50/mo. So, you lose
$2.50 by waiting one month, but save the commission, say $10, by
grouping two months into one purchase. There's nothing wrong with going
the fund route until the dollars are more significant and maybe once
every year or so converting some amount to the ETF.
That help?
JOE
JoeTaxpayer.com

- quote -

Hi Joe,

Thanks much for the comments and expertise. I wonder if any studies
have been done on frequencies of dollar cost averaging? In other
words, if I'm only investing twice/year versus month/weekly, or even
daily, is that still dollar cost averaging? Is there any optimal
frequency? There is an obvious trade-off between advantages of dollar
cost averaging and the cost of doing so in ETFs with their obvious
advantages of low cost.

On Jul 16, 3:20 pm, joetaxpayer <joetaxpa...[at]nospam.com> wrote:
- quote -

Makes perfect sense now. As Beliavsky stated above, I was not looking
purely at index funds, but all mutual funds. Many actively managed MF
perform worse than their respective benchmarks/ETFs. Index funds do
not have that problem however.

kastnna

kastnna wrote:
- quote -

The point I was trying to make (and failed, perhaps) is this; let's
compare VFINX to SPY, for example. Above, I say 'type' to try to get as
'apples to apples' as I can. I am big on recognizing the variables and
freezing the ones I can. VFINX has .18% expense, SPY, .08%. Recall, the
OP is using retirement accounts, and therefore any discussion of tax
consequences is valid, just not to him. There's a variable or two frozen
there. Assume a $10 commission. For $1000, that represents 1%, or about
10 years to have the .10% expense savings favor the ETF. If one can
limit transactions to $5000, the break even is 2 years.

For those with this investment in a post tax account, the ETF may be
favored as it avoids the risk of cap gain distributions which may mess
up one's tax planning. Or favor the fund, as it can be used for easy
rebalancing and sales to create a loss or gain if needed.

The choice isn't clear cut, it really depends on one's intent and
detailed background info. I hope that clarifies my first remarks.

JOE

On Jul 16, 10:08 am, kastnna <kast...[at]auburnalum.org> wrote:
- quote -

Maybe the assertion was about ETFs vs. comparable INDEX funds.

On Jul 14, 2:51 pm, joetaxpayer <joetaxpa...[at]nospam.com> wrote:
- quote -

Please clarify? I've read in various places that given two similar
investments the ETF will often outperform solely based on the lack of
active management (read: market timing) that is otherwise present in
most funds (average fund turnover is 89%). According to the S&P Index
versus active report (SPIVA) active managers rarely outperform the
market over any significant time period, so why would you take the
actively managed fund over the passive ETF if expenses were similar?

I am assuming you mean that the funds are a better buy because the
lack of a bid-ask spread and no transaction costs? If so, that is most
definitely an advantage of MFs. Some would argue that being able to
buy and sell ETFs at any time during the day can compensate for this.
I'm not a big market timer so I'll stay away from this one.

- quote -

Very good point. There are plenty of low cost funds out there, but
according to morningstar the average expense ratio of ETFs in 2006 was
0.36%, while the average fund expense ratio was 1.07% (0.83% for bond
funds).

mjgrinnell[at]gmail.com wrote:
- quote -

If the type of investment is available in both fund and ETF choices, and
the expenses are similar, you are better off with the fund.

There are some reasons why in a post tax (non 401(k) or IRA, I mean)
account the ETF may have advantages. Depending how tightly you manage
your money with an eye on taxes, a fund's year end distribution may be
very unwelcome.

In your case, you mention funds with as high as 1.2% expenses. Is the
same type of investment available as an ETF with even a .7% expense? If
so, you need to do the math. .5% on $2000 is $10 which should cover the
commission on the ETF. If the saving is less, the break even is longer.
I see you making 8 transactions a year, you might consider cutting it
down to 6 or even 4 to reduce expenses. How about (in 2008) loading the
403(b) with a percentage that will let you hit your saving goal by
August, then making two deposits to your Roths at the end of the year?
Same total dollars, just deposited a bit differently to minimize
transactions. Just a thought.

JOE

I invest quarterly in my and my wife's Roth IRA's (Feb, May, Aug, Nov
for me and March, June, September, and December for my wife). I have
an index fund at .10% expense ratio and several other funds all around
1.2% or less. I am contemplating moving some of these periodic
investments to ETF funds but I am new to ETFs and wonder about how
commissions and the likes would affect periodic, but not frequent
investments. I dollar cost average for my workplace 403b and that is
rather frequent (ever other week) with relatively small sums so I
figure that is out for ETFs. But, what kind of a case can be made for
the Roth IRAs?

Thanks,

Mike