[ic] Re: Negative Tax - bug?

Peter peter at pajamian.dhs.org
Tue Oct 30 20:54:23 EST 2007

On 10/30/2007 03:43 PM, Dan Bergan wrote:
> On 10/30/07, Mike Heins <mike at perusion.com> wrote:
>> Quoting Dan Bergan (danb at berganconsulting.com):
>>> On 10/30/07, Dan Bergan <danb at berganconsulting.com> wrote:
>>>> On 10/30/07, Mike Heins <mike at perusion.com> wrote:
>>>>> Quoting Dan Bergan (danb at berganconsulting.com):
>>>>>> On 10/16/07, Dan Bergan <danb at berganconsulting.com> wrote:
>>>>>>> I'm having an issue where negative sales tax amounts are computed.
>>>>>>> The issue arises when there is a discount on the entire order, and the
>>>>>>> products in the cart are nontaxable.
>>>>>>> A detailed description and potential solution is here:
>>>>>>> http://www.icdevgroup.org/pipermail/interchange-users/2007-May/047410.html
>>>>>>> This issue was also found here (with no resolutions):
>>>>>>> http://www.icdevgroup.org/pipermail/interchange-users/2005-September/043923.html
>>>>>>> Is this a bug?  It doesn't seem like a negative sales tax should be valid.
>>>>>> Back in May, Carl had what looks to be a good solution for this
>>>>>> problem, which I am going to implement for my installation (see
>>>>>> below).
>>>>> It is unclear if it should be a bug at all. If you wanted to implement
>>>>> refunds, you would need negative tax.
>>>>> The bigger problem would seem to be the negative amounts.
>>>>> In any case, I have set up the ability to exclude negative tax
>>>>> amounts with:
>>>>>         Pragma  no_negative_tax
>>>>> That is in the latest CVS.
>>>> Mike -
>>>> Ahh... refunds -- I hadn't thought of that!  :-)
>>>> For refunds, the taxable sales amount would be negative, so the tax
>>>> would calculate as a negative.  In the case above (non-taxable
>>>> products with a discount), the taxable amount should still be zero
>>>> (not negative).
>>>> I currently don't use Interchange for refunds, but that doesn't mean I
>>>> never will!  So, I think I'll have to ponder this some more and dig
>>>> deeper into the code.
>>> I think this might be more of a question about sales tax law.  Should
>>> a ENTIRE_ORDER discount reduce the taxable amount?
>> Without doubt.
>>> The only mention I have found when searching sales tax documents are
>>> coupons for individual items, and percent off of all items.  In the
>>> case of ENTIRE_ORDER the actual sale prices of the individual items do
>>> not change, so my client reports that number to their state as the
>>> taxable amount.  My client treats this type of discount like a gift
>>> certificate -- it is applied after the tax is calculated.
>> That is simply wrong. The user is over-paying taxes.
> in Ohio, that may not be the case:
> (google cache of a tax commissioner judgment pdf...)
>>> Of course, this is just my client, in one state and in the US.
>>> Would it work to have a catalog variable to determine if an
>>> ENTIRE_ORDER discount should be applied pre-tax or post-tax?
>> It won't ever be applied post-tax on my watch.
>> If the order total is reduced, the tax is reduced. A gift-certificate
>> is pre-purchased, and is paid for.
> I haven't yet determined how Iowa (my client's state) applies their
> law.  But, Connecticut's formula makes a lot of sense:
> http://www.ct.gov/drs/cwp/view.asp?A=1529&Q=395904
> The ENTIRE_ORDER discount is split proportionally among each item,
> then each item's taxablility is determined to figure the total sales
> tax.  This would fix the problem of a negative tax being calculated
> for nontaxable items.  The consumer is neither overpaying nor
> underpaying taxes.

It would seem to make sense to me to allow this to be configurable, even
state-by-state (a column in the state table) so that IC will have the
ability to better adapt to different tax laws in different areas.

I would propose allowing the use of special subs that can be picked from
the state and/or country tables with a few defaults.  That would allow
customization for whatever weird taxation laws someone may come across
(that we don't anticipate).


More information about the interchange-users mailing list