On Fri, Mar 16, 2018 at 6:38 AM, Martin Michlmayr <[email protected]> wrote:

> * Martin Blais <[email protected]> [2018-02-06 00:43]:
> > > > 2016-07-28 * "Buy some shares of retirement fund"
> > > >   Assets:Invest       45.0045 VBMPX {11.11 USD}
> > >
> > > > 2016-10-12 * "Buy some shares of retirement fund"
> > > >  Assets:Invest       54.5951 VBMPX {10.99 USD}
> > >
> > > > Should result in a single lot with the total number of units and the
> > > averaged cost:
> > > >          units ccy    cost    cost-ccy lot-date    label
> > > >        99.5996 VBMPX {11.0442 USD,     2016-07-28, None}
> > >
> > > But average cost is calculated with the second transaction so
> > > shouldn't the lot_date be 2016-10-12?
> >
> > No, it's a choice.
> >
> > The thing is in the USA it may have tax consequences, e.g. LT vs. ST
> > capital gains rates are different.
>
> I don't know much about US taxation but wouldn't it depend on the
> *last* time you bought something?
>

Actually, to be realistic, in the US, everything I know that's allowed at
average cost is treated as tax-free, e.g. in a 401k plan.
The reality is that for those lots, the date doesn't really matter, and
besides perhaps some generic marker like "VARIOUS" would be the better
treatment than making a choice.
That would introduce an inconsistency in the data type though, i.e,
"sometimes the lot date field is VARIOUS".

In the meantime, I think an option to decide which date should be selected
(oldest or newest) is totally fair game.



> Perhaps the Right Thing to do would be to provide an option to select what
> > should happen automatically.
>
> Right, that might be a good idea.
>
> Actually, I wanted to ask how the average cost method will work.
> Presumably, if I buy some shares and then buy more shares, it will
> calculate the average cost immediately?


> Personally, I prefer to keep the original holdings as long as
> possible.  i.e. I'd prefer to calculate the average cost only when
> needed (e.g. immediately before selling some).
>

IIRC that's the proposal I put forward, that lot augmentations shouldn't
merge immediately, but that lot reductions should trigger a merge before
and after reducing (to recalculate the cost, in case you reduce at another
cost than average).


Will this be possible?  e.g. can I tell beancount not to use anything
> and then easily call the average cost function from beancount code?
>

Yes, I think this way it makes it automatic enough, yet it leaves some
under your control.

Another option will also be to leave the account as e.g., FIFO but if/when
you want to force averaging, you could use the unimplemented merge marker
that's supported in the syntax, e,g. "{101.23 USD, *}".




--
> Martin Michlmayr
> http://www.cyrius.com/
>
> --
> You received this message because you are subscribed to the Google Groups
> "Beancount" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> To post to this group, send email to [email protected].
> To view this discussion on the web visit https://groups.google.com/d/
> msgid/beancount/20180316103800.atdrkt2vel2stxx2%40jirafa.cyrius.com.
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups 
"Beancount" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/beancount/CAK21%2BhOLo9%3D9dp%3DmfMX4Po%2B0XyshR45E1V36WqKYyAwgApKZ2w%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to