GAIN Act of 2017 EITC reform file

[martinholmer]

This pull request is being submitted on behalf of @MattHJensen, who analyzed the reform bill and prepared the JSON reform file. This pull request replaces the incorrect and unmerged pull request #1553 and resolves issue #1552.

Using the new BrownKhanna.json reform file with puf.csv input produces the following ten-year static results:

2017 -128.3
2018 -131.7
2019 -135.4
2020 -139.1
2021 -142.8
2022 -146.5
2023 -150.3
2024 -154.2
2025 -158.2
2026 -162.4

The ten-year cost of the EITC reform is estimated by Tax-Calculator to be about $1449 billion, which is close to the TPC rounded estimate of $1400 billion.

The Tax-Calculator estimate of the cost of the reduction in the minimum eligibility age for childless filing units is about $91 billion, which is close to the TPC rounded estimate of $100 billion.

For the first year of the reform (2017), Tax-Calculator produces these static results:

Weighted Tax Reform Totals by Baseline Expanded-Income Decile
    Returns    ExpInc    IncTax    PayTax     LSTax    AllTax
       (#m)      ($b)      ($b)      ($b)      ($b)      ($b)
 0    17.00    -119.4      -6.0       6.4       0.0       0.5
 1    17.00     152.4     -28.8      14.4       0.0     -14.4
 2    17.01     273.9     -56.2      24.5       0.0     -31.7
 3    17.00     400.6     -47.7      35.8       0.0     -11.8
 4    17.01     558.3     -25.4      52.6       0.0      27.3
 5    17.01     754.0      10.0      72.0       0.0      82.0
 6    17.01    1008.8      55.4      97.0       0.0     152.4
 7    17.00    1379.3     116.0     143.2       0.0     259.1
 8    17.01    2054.4     215.0     233.3       0.0     448.3
 9    17.01    6272.4    1339.3     413.3       0.0    1752.7
 A   170.05   12734.6    1571.7    1092.6       0.0    2664.3

Weighted Tax Differences by Baseline Expanded-Income Decile
    Returns    ExpInc    IncTax    PayTax     LSTax    AllTax
       (#m)      ($b)      ($b)      ($b)      ($b)      ($b)
 0    17.00    -119.4      -3.4       0.0       0.0      -3.4
 1    17.00     152.4     -16.7       0.0       0.0     -16.7
 2    17.01     273.9     -31.3       0.0       0.0     -31.3
 3    17.00     400.6     -30.4       0.0       0.0     -30.4
 4    17.01     558.3     -24.5       0.0       0.0     -24.5
 5    17.01     754.0     -16.3       0.0       0.0     -16.3
 6    17.01    1008.8      -5.6       0.0       0.0      -5.6
 7    17.00    1379.3      -0.1       0.0       0.0      -0.1
 8    17.01    2054.4      -0.0       0.0       0.0      -0.0
 9    17.01    6272.4      -0.0       0.0       0.0      -0.0
 A   170.05   12734.6    -128.3       0.0       0.0    -128.3

@MattHJensen @feenberg @Amy-Xu @andersonfrailey @hdoupe @GoFroggyRun

[codecov-io]

Codecov Report

Merging #1555 into master will not change coverage.
The diff coverage is n/a.

@@          Coverage Diff           @@
##           master   #1555   +/-   ##
======================================
  Coverage     100%    100%           
======================================
  Files          37      37           
  Lines        2644    2644           
======================================
  Hits         2644    2644

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2f755a1...f268846. Read the comment docs.

[MattHJensen]

Thanks a lot @martinholmer. LGTM, although as far as RELEASES.md goes, this feels more like a joint effort.

[martinholmer]

@MattHJensen said about pull request #1555:

LGTM, although as far as RELEASES.md goes, this feels more like a joint effort.

OK, but I'm not sure what I've done on this except make a big mistake in #1553.

[martinholmer]

Using the new earnings response option added to the Behavior class in pull requests #1554 and #1556, we can show how earnings might change under the EITC expansion specified in the Brown-Khanna GAIN Act.

We have no firm expectations about how responsive earnings are to changes in marginal tax rates on earnings or to changes in after-tax income. So, what is shown here is just arbitrary examples of how the new earnings response capability works.

To illustrate the earnings response analysis capabilities, we show results for three cases:

1. Changes in marginal tax rates on earnings lead to changes in hours worked, and therefore, earnings. The arbitrarily assumed substitution elasticity is 0.3, which means that the proportional change in earnings is 0.3 times the proportional change in the after-tax marginal rate on taxpayer earnings (i.e., 1-MTR). So, for example, if the reform caused the marginal tax rate for a filing unit to drop from 0.10 to -0.05 (see MTR graph above), then the proportional change in the after-tax marginal rate is 1.05 divided by 0.90 minus one, which is about 1.16667 minus one, or 0.16667. Multiplying that by 0.3 produces an assumed proportional change in earnings of 0.05. In other words, the reduction in the marginal tax rate on earnings for this filing unit is assumed to cause a five percent increase in hours and earnings for this filing unit.

2. Changes in after-tax income lead to changes in hours worked, and therefore, earnings. The arbitrarily assumed income elasticity is -0.5, which means that the proportional change in earnings is -0.5 times the proportional change in the after-tax income of the filing unit. So, for example, if a filing unit experiences a six percent increase in after-tax income under this reform, then multiplying that 0.06 by -0.5 produces an assumed proportional change in earnings of -0.03. In other words, the increase in after-tax income for this filing unit is assumed to cause a three percent reduction in earnings for this filing unit.

3. Changes described in 1. and 2. both happen. This case combines the substitution elasticity assumed in case 1 with the income elasticity assumed in case 2.

Here are the static results for 2017 and the results for these three arbitrary sets of earnings response assumptions in 2017, the first year of the GAIN Act reform:

STATIC RESULTS:
Weighted Tax Differences by Baseline Expanded-Income Decile
    Returns    ExpInc    IncTax    PayTax     LSTax    AllTax
       (#m)      ($b)      ($b)      ($b)      ($b)      ($b)
 0    17.00    -119.4      -3.4       0.0       0.0      -3.4
 1    17.00     152.4     -16.7       0.0       0.0     -16.7
 2    17.01     273.9     -31.3       0.0       0.0     -31.3
 3    17.00     400.6     -30.4       0.0       0.0     -30.4
 4    17.01     558.3     -24.5       0.0       0.0     -24.5
 5    17.01     754.0     -16.3       0.0       0.0     -16.3
 6    17.01    1008.8      -5.6       0.0       0.0      -5.6
 7    17.00    1379.3      -0.1       0.0       0.0      -0.1
 8    17.01    2054.4      -0.0       0.0       0.0      -0.0
 9    17.01    6272.4      -0.0       0.0       0.0      -0.0
 A   170.05   12734.6    -128.3       0.0       0.0    -128.3

CASE 1 RESULTS:
Weighted Tax Differences by Baseline Expanded-Income Decile
    Returns    ExpInc    IncTax    PayTax     LSTax    AllTax
       (#m)      ($b)      ($b)      ($b)      ($b)      ($b)
 0    17.00    -119.4      -3.7       0.1       0.0      -3.6
 1    17.00     152.4     -17.9       0.5       0.0     -17.4
 2    17.01     273.9     -31.6       0.2       0.0     -31.5
 3    17.00     400.6     -31.8      -0.8       0.0     -32.6
 4    17.01     558.3     -27.4      -1.5       0.0     -29.0
 5    17.01     754.0     -18.0      -0.8       0.0     -18.8
 6    17.01    1008.8      -6.8      -0.5       0.0      -7.3
 7    17.00    1379.3      -0.2      -0.0       0.0      -0.2
 8    17.01    2054.4      -0.0      -0.0       0.0      -0.0
 9    17.01    6272.4      -0.0      -0.0       0.0      -0.0
 A   170.05   12734.6    -137.5      -2.9       0.0    -140.4

CASE 2 RESULTS:
Weighted Tax Differences by Baseline Expanded-Income Decile
    Returns    ExpInc    IncTax    PayTax     LSTax    AllTax
       (#m)      ($b)      ($b)      ($b)      ($b)      ($b)
 0    17.00    -119.4      -3.0      -0.2       0.0      -3.1
 1    17.00     152.4     -14.5      -0.8       0.0     -15.3
 2    17.01     273.9     -30.6      -1.5       0.0     -32.1
 3    17.00     400.6     -32.6      -1.8       0.0     -34.4
 4    17.01     558.3     -27.4      -1.7       0.0     -29.1
 5    17.01     754.0     -18.8      -1.2       0.0     -19.9
 6    17.01    1008.8      -6.4      -0.4       0.0      -6.8
 7    17.00    1379.3      -0.1      -0.0       0.0      -0.1
 8    17.01    2054.4      -0.0      -0.0       0.0      -0.0
 9    17.01    6272.4      -0.0      -0.0       0.0      -0.0
 A   170.05   12734.6    -133.4      -7.6       0.0    -141.0

CASE 3 RESULTS:
Weighted Tax Differences by Baseline Expanded-Income Decile
    Returns    ExpInc    IncTax    PayTax     LSTax    AllTax
       (#m)      ($b)      ($b)      ($b)      ($b)      ($b)
 0    17.00    -119.4      -3.2      -0.1       0.0      -3.3
 1    17.00     152.4     -15.9      -0.3       0.0     -16.3
 2    17.01     273.9     -31.0      -1.4       0.0     -32.4
 3    17.00     400.6     -33.8      -2.6       0.0     -36.4
 4    17.01     558.3     -30.4      -3.2       0.0     -33.6
 5    17.01     754.0     -20.4      -2.0       0.0     -22.3
 6    17.01    1008.8      -7.6      -0.9       0.0      -8.6
 7    17.00    1379.3      -0.2      -0.0       0.0      -0.2
 8    17.01    2054.4      -0.0      -0.0       0.0      -0.0
 9    17.01    6272.4      -0.0      -0.0       0.0      -0.0
 A   170.05   12734.6    -142.6     -10.5       0.0    -153.1

The case 3 results are generated using the tc CLI to Tax-Calculator and the following reform and assumption files:

$ cat BK.json
// Title: Brown-Khanna Grow American Incomes Now (GAIN) Act of 2017
// Reform_File_Author: Matt Jensen
// Reform_Reference: https://khanna.house.gov/media/press-releases/release-sen-sherrod-brown-and-rep-ro-khanna-introduce-landmark-legislation
// Reform_Description:
// - Increase EITC maximum amounts (1)
// - Increase EITC phase-in rates (2)
// - Increase EITC phase-out rate for childless filing units (3)
// - Increase EITC phase-out start for childless filing units (4)
// - Lower EITC minimim eligibility age for childless f.units from 25 to 21 (5)
// Reform_Parameter_Map:
// - 1: _EITC_c
// - 2: _EITC_rt
// - 3: _EITC_prt
// - 4: _EITC_ps
// - 5: _EITC_MinEligAge
{
  "policy": {
      "_EITC_c": {"2017": [[3000, 6528, 10783, 12131]]},
      "_EITC_rt": {"2017": [[0.3, 0.6258, 0.768, 0.864]]},
      "_EITC_prt": {"2017": [[0.1598, 0.1598, 0.2106, 0.2106]]},
      "_EITC_ps": {"2017": [[18340, 18340, 18340, 18340]]},
      "_EITC_MinEligAge": {"2017": [21]}
    }
}

$ cat BKee.json
{
    "consumption": {},
    "behavior": {
        "_BE_sub": {"2017": [0.3]},
        "_BE_inc": {"2017": [-0.5]},
        "_BE_subinc_wrt_earnings": {"2017": [true]}
    },
    "growdiff_baseline": {},
    "growdiff_response": {}
}

$ tc puf.csv 2017 --reform BK.json --assump BKee.json --tables

@MattHJensen @feenberg @Amy-Xu @andersonfrailey @hdoupe @GoFroggyRun @codykallen