Handling DIST_D type trans in Generic Transaction loads.

[modelerMike]

Good day Mark;
In Generic Transaction File portfolio loads when I choose
NOT to use the “have FM do the corresponding default cash account transactions” option,
how do i get a stock’s DIST_D transaction over to default cash account , FCASH below ?
I tried to do this with the following two transactions;

SYMB CUSIP MM/DD/YYYY TRAN ANUM BNUM CNUM NAME

stk4 444444444 01/17/2010 DIST_D 100 0 0.00 stk4
FCASH 999999999 01/17/2010 BUY 1 100 0.00 FCASH

the resulting investment performance report and portfolio value graph both look good;
that is stk4’s performance reflects the dividend distribution and portfolio value does not
get double counted; that is portfolio value goes up by amount of the dividend distribution
on the date from the distribution due to the FCASH , default cash account’s BUY.

Is this the way you would handle it ? Am i missing something ?

The reason I am taking this approach is to ensure cash outflows occur on the date the
transactions settle, not when they trade. I do this by using a SYNCASH security
for SELLS : ( in examples below ; first two dates = trade date and second two are the settlement date)

stk2 222222222 01/08/2010 SELL 20 50 0.00 stk2
SYNCASH 111111111 01/08/2010 BUY 1 1000 0.00 SYNCASH
FCASH 999999999 01/13/2010 BUY 1 1000 0.00 FCASH
SYNCASH 111111111 01/13/2010 SELL 1 1000 0.00 SYNCASH

for BUYS :

stk7 777777777 01/26/2010 BUY 50 20 0.00 stk7
SYNCASH 111111111 01/26/2010 SELL 1 1000 0.00 SYNCASH
FCASH 999999999 01/29/2010 SELL 1 1000 0.00 FCASH
SYNCASH 111111111 01/29/2010 BUY 1 1000 0.00 SYNCASH

Using the syncash dummy security in this way ensures cash in the portfolio over time
is exactly what it would be in the REAL portfolio over time and the value of the overall
portfolio reflects the REAL value of the portfolio day-by-day over time.

[Mark]

Hi modelerMike,

This all looks perfectly correct to me. Recording a distribution and a buy of cash is exactly what would happen if you used the default cash account, and only recorded the distribution.

The "SYNCASH" methodology you've described also looks perfectly fine.

[modelerMike]

Thanks for the reality check Mark.
Good to know.

Regards,
Mike