Rob...

Since you're stating that I'm attacking AccountSync, I'd like to
respond. I'm not attacking AccountSync. I just think approach to
allocate rounding differences into a cost item and not just float the
share price is extreme. Yes... you have a very valid point that
dividend reinvestment have hidden costs in the rounding differences.
And I agree that it's an interesting issue to point out that the number
of shares receive, when multiplied by the reported share price most
often don't equate to the cash value of the dividend. But, let's look
at the magnitude of the difference, assuming that the broker truncates
down to the thousandth of a share (0.001 shares) and the share price is
rounded to the nearest penny.

Using an extreme example, let's assume a $1000.00 dividend is received
in your DRIP, a $1000.01 share price and shares received are reported to
0.001 shares. Then, at most, the recipient would miss out on almost
0.001 shares if the shares beyond the third decimal place are truncated
and not rounded. At $1000.01/share, this amounts to $0.99 in value.
This is a worst case scenario I can come up with. With a more plausible
example, for example, a $50.01 stock, and a $100 dividend, the
difference amounts to $0.03 of rounding error (shares received at
$100/$50.01/share, and result truncated at the 4th decimal place and the
truncated remainder of 0.0006001 shares being multiplied by the share
price of 50.01 = $0.03).

Yes... when thousands of transactions are being made, there may be real
money here. But I don't have thousands of transactions... just four a
year, and I just can't get exited about trying to allocate these penny
differences as a fee, and then tracking them as such and explaining them
to my partners when few, if none of my partners, would go out of their
way to return a six pack of bottles to reclaim the $0.30 deposit.

OK.. that said, there are DRIPS where fixed fees are charged for the
DRIP transaction and these cases, the fees can be costly. For example,
if the broker charges $5 to reinvest dividends, then we have a different
scenario. Under this situation, I'd want to see the cost treated as a
trading cost, and in this case the cost is material. I would treat
these costs should be separately and not use a rounding approach, though
the rounding approach facilitates the programming involved to allow
AccountSync to be automated. And please correct me if I'm wrong, is the
real reason the software determines the expense in the way it does.

John Munn

Rob...

 From your response and Rip's discussions, I do understand what you're
doing. No need to respond. Thanks Rip and Rob for the discussion...
it was interesting, at least to me, to see your perspectives. And
Rip.... significant figures is a discussion for another day! A lost
concept from the days of slide rules!

Rob... the software does what it does so it can be universally applied
to transactions. As I understand it, incidental rounding differences
are just a part of the cost differences and is co-mingled with SEC fees
and trading costs to account between the amount that was expected from
calculations versus the amount actually received.

John


John Munn wrote:

> Rob...
>
> Since you're stating that I'm attacking AccountSync, I'd like to
> respond. I'm not attacking AccountSync. ...cut...

Rob Nagler wrote:
> Here's the data we have: REIN @ 44.6790.

Let's see if I understand this. (Obviously, it will take me
a while to earn my One Minute Treasurer certificate too!
<g>).

AccountSync downloads two transactions from the broker and
both get recorded in bivio.
(1) $9.45 cash dividend paid.
(2) 0.211 shares purchased @ $44.679.

0.211 x 44.679 = 9.427269 (exactly) which rounds to $9.43
(there can't be fractional cents). bivio knows that these
two broker transactios go together and, so, both must
account for the same dollar amount. So, bivio adds whatever
"fee" is necessary to the second transaction ($0.02 in this
case) to make the two transactions balance.

-Jim Thomas

> (1) $9.45 cash dividend paid.
> (2) 0.211 shares purchased @ $44.679.
> ...
> So, bivio adds whatever "fee" is necessary to
> the second transaction ($0.02 in this case)
> to make the two transactions balance.

Or, perhaps (I'd think more likely) the 2nd transactions is
...

(2) $9.45 for 0.211 shares (@ $44.679)

... so bivio doesn't have to infer the dollar amount of the
2nd transaction. bivio is inferring a "fee" so it can
preserve the shares price shown by the broker in the
transaction comment.

-Jim Thomas

Jim Thomas writes:
> ... so bivio doesn't have to infer the dollar amount of the
> 2nd transaction. bivio is inferring a "fee" so it can
> preserve the shares price shown by the broker in the
> transaction comment.

Correct.

Rob