Parameter "_AMT_CG_rt4" wasn't equiped with correct reference

[GoFroggyRun]

The _AMT_CG_rt4 parameter in the current_law_policy.json file does not have the correct reference and name in the way that the parameter has nothing to do with capital gain rates (which only have three groups). This parameter of value 0.25 (25%) is likely to be the 25 percent described in the section "Netting capital gains and losses" on P640 in U.S. Master Tax Guide, 98th Ed. The exact source of the parameter will be further investigated.

[martinholmer]

@GoFroggyRun said on 06-Jan-2016:

The _AMT_CG_rt4 parameter in the current_law_policy.json file does not have the correct reference and name in the way that the parameter has nothing to do with capital gain rates (which only have three groups). This parameter of value 0.25 (25%) is likely to be the 25 percent described in the section "Netting capital gains and losses" on P640 in U.S. Master Tax Guide, 98th Ed. The exact source of the parameter will be further investigated.

If this is a mistake, can you prepare a pull request that fixes the mistake? I guess that pull request would involve some edits of the current_law_policy.json file, is that correct?

@MattHJensen @feenberg

[GoFroggyRun]

@martinholmer:

If this is a mistake, can you prepare a pull request that fixes the mistake?

Yes. I believe the name of that parameter wasn't properly assigned. I'm, however, also not sure what's the best name for it. This 0.25 (25%) should be the exact same amount appear hear on capital gain, where the capital gain 25% is being hard-coded.

I guess that pull request would involve some edits of the current_law_policy.json file, is that correct?

And yes, if an appropriate name were found, I'll edit and incorporate in the current_law_policy.json.

[martinholmer]

@MattHJensen and Dan @feenberg, Sean and I are confused by the the _AMT_CG_rt4 policy parameter in the current_law_policy.json file. The description of this parameter says this:

"long_name": "Long term capital gain and qualified dividend rate 4",
"description": "The capital gain and qualified dividend (stacked on top of regular income) above threshold 2 are taxed at this rate.",

But that is exactly the same description as for _AMT_CG_rt3, which says:

"long_name": "Long term capital gain and qualified dividend rate 3",
"description": "The capital gain and qualified dividend (stacked on top of regular income) above threshold 2 are taxed at this rate.",

And our confusion is deepened when we look at the functions.py file to see where _AMT_CG_rt4 is being used, which is only in the last line of this code fragment:

    # Capital Gain for AMT
    _tamt2 = 0.
    _amt5pc = 0.0
    _line45 = max(0., AMT_CG_thd1[MARS - 1] - c24520)
    _line46 = min(_alminc, c62720)
    _line47 = min(_line45, _line46)
    _line48 = min(_alminc, c62720) - _line47
    _amt15pc = min(_line48, max(0., AMT_CG_thd2[MARS - 1] - c24520 - _line45))
    if _ngamty != (_amt15pc + _line47):
        _amt20pc = _line46 - _amt15pc - _line47
    else:
        _amt20pc = 0.
    if c62740 != 0:
        _amt25pc = max(0, _alminc - _ngamty - _line46)
    else:
        _amt25pc = 0.
    c62747 = AMT_CG_rt1 * _amt5pc
    c62755 = AMT_CG_rt2 * _amt15pc
    c62760 = AMT_CG_rt3 * _amt20pc
    c62770 = AMT_CG_rt4 * _amt25pc

It looks to us that the 62770 variable has not been part of the of the PUF or the tax law for some years (at least this is what the spreadsheet of e-codes seems to suggest). And it is not part of the IRS 2008 PUF file.

Did there used to be, in years before 2008, higher AMT tax rates for capital gains? That's what it looks like to us, but we are not that well versed in tax law history. If our hunch is correct (that _AMT_CG_rt4 has not been part of the tax code, at least, since 2013), why is _AMT_CG_rt4 in the current_law_policy.json file? And, if there is a reason for its inclusion in the current_law_policy.json file, why is its description so confusing?

[GoFroggyRun]

@martinholmer I think including _AMT_CG_rt4 in current_law_policy.json file is my mistake in #438. Previously, before PR #438 being merged, we have

c62770 = 0.25 * _amt25pc,

and then the 0.25 was parameterized as _AMT_CG_rt4, which, now, is considered inappropriate. The reason why I did so might be that the name of the variable, _amt25pc, somehow confused me.

[martinholmer]

Sean said:

I think including _AMT_CG_rt4 in current_law_policy.json file is my mistake in pull request #438 [which was merged into the master branch by me on 05-Nov-2015]. Previously, before PR #438 being merged, we have c62770 = 0.25 * _amt25pc and then the 0.25 was parameterized as _AMT_CG_rt4, which, now, is considered inappropriate. The reason why I did so might be that the name of the variable, _amt25pc, somehow confused me.

@MattHJensen and Dan @feenberg, So you can see that our confusion is long-standing and still remains. Sean and I can now add to our list of questions: what is _amt25pc? This variable name seems to suggest that some part of AMT taxable income is taxed at 25%. Has this been part of current-law policy since 2013? As you can see, we need some help on this issue.

[feenberg]

On Sun, 14 Feb 2016, Martin Holmer wrote:

Sean said:

  I think including _AMT_CG_rt4 in current_law_policy.json file is my
  mistake in #438 [merged into the master branch by me on 05-Nov-2015].
  Previously, before PR #438 being merged, we have
  c62770 = 0.25 * _amt25pc
  and then the 0.25 was parameterized as _AMT_CG_rt4, which, now, is
  considered inappropriate. The reason why I did so might be that the name
  of the variable, _amt25pc, somehow confused me.

@MattHJensen and Dan @feenberg, So you can see that our confusion is long-standing and still remains. Sean and I can now add to our list of questions: what is _amt25pc? This variable name seems to suggest that some part of AMT taxable income is taxed at 25%. Has this been part of current-law policy since 2013? As you can see, we need some help on this issue.

I think there is some cg taxed at 25% but not available in the PUF. I'll check this afternoon.

dan

— Reply to this email directly or view it on GitHub.[AHvQVS6FKuZ62JiI0X0kc-YTJIQbUlctks5pkRJ9gaJpZM4HAC80.gif]

[martinholmer]

dan @feenberg said:

I think there is some cg taxed at 25% but not available in the PUF. I'll check this afternoon.

Thanks for looking into this, Dan.

[feenberg]

As far as I can tell the 25% rate applies to "Unrecaptured Schedule E Gain" and isn't something that would be in the PUF, but which is significant in the Insole files, so we should continue to carry it.

dan

On Sun, 14 Feb 2016, Martin Holmer wrote:

Sean said:

  I think including _AMT_CG_rt4 in current_law_policy.json file is my
  mistake in #438 [merged into the master branch by me on 05-Nov-2015].
  Previously, before PR #438 being merged, we have
  c62770 = 0.25 * _amt25pc
  and then the 0.25 was parameterized as _AMT_CG_rt4, which, now, is
  considered inappropriate. The reason why I did so might be that the name
  of the variable, _amt25pc, somehow confused me.

@MattHJensen and Dan @feenberg, So you can see that our confusion is long-standing and still remains. Sean and I can now add to our list of questions: what is _amt25pc? This variable name seems to suggest that some part of AMT taxable income is taxed at 25%. Has this been part of current-law policy since 2013? As you can see, we need some help on this issue.

— Reply to this email directly or view it on GitHub.[AHvQVS6FKuZ62JiI0X0kc-YTJIQbUlctks5pkRJ9gaJpZM4HAC80.gif]

[martinholmer]

Dan @feenberg said:

As far as I can tell the 25% rate applies to "Unrecaptured Schedule E Gain" and isn't something that would be in the PUF, but which is significant in the Insole files, so we should continue to carry it.

Dan, Thanks for looking into this matter.

So, maybe we should undo the confusion that Sean and I experienced months ago when we changed the hardwired 0.25 tax rate to a Policy parameter in current_law_policy.json. I notice that the parameter we added is not referenced by TaxBrain, so I'll fix this up later this afternoon.

@MattHJensen @GoFroggyRun

[GoFroggyRun]

This issue has been resolved in #599 . Thanks @martinholmer .