Skip to content

Commit

Permalink
Add aca_ptc_los_angeles integration test file
Browse files Browse the repository at this point in the history
  • Loading branch information
martinholmer committed Nov 12, 2023
1 parent 249a41e commit 02ba890
Showing 1 changed file with 198 additions and 0 deletions.
Original file line number Diff line number Diff line change
@@ -0,0 +1,198 @@
# Two 2022 integration tests (one for each LA rating area):

- name: aca_ptc_los_angeles test 22-15
absolute_error_margin: 0.00001
period: 2022
input:
people:
person1:
age: 50
employment_income: 40_000
is_aca_eshi_eligible: false
person2:
age: 45
employment_income: 30_000
is_aca_eshi_eligible: false
person3:
age: 15
is_aca_eshi_eligible: false
tax_units:
tax_unit:
members: [person1, person2, person3]
households:
household:
members: [person1, person2, person3]
state_fips: 6 # CA
county_fips: 37 # Los Angeles County has two zip3-defined rating areas
aca_zip3_ca_county_la: 911 # e.g., Pasadena (91105)
output:
tax_unit_medicaid_income_level: 3.0395136 # greater than 2.66 CA CHIP limit
is_medicaid_eligible: [false, false, false]
is_aca_ptc_eligible: [true, true, true]
aca_slspc_ca: 12_476 # same as 2022 KFF ACA calculator
aca_magi_percent: 318
aca_ptc_phase_out_rate: 0.0645
aca_ptc_ca: 7_961 # $7,947 per 2022 KFF ACA calculator
# annual contribution amount: 0.0645 * 70000 = 4515
# PEUS: 12476 - 7961 = 4515
# KFF: 12476 - 7947 = 4529
# 14 diff is a bit more than $1/month
# FPL = 21960 (for three in 2021), so exact MAGI/FPL = ~318.76%
# KFF implied applicable figure is 0.0647, which is between
# 318 ==> 0.0645 and 319 ==> 0.0648, so it appears as if KFF is
# not truncating the MAGI/FPL percent as called for on IRS Form 8962

- name: aca_ptc_los_angeles test 22-16
absolute_error_margin: 0.00001
period: 2022
input:
people:
person1:
age: 50
employment_income: 40_000
is_aca_eshi_eligible: false
person2:
age: 45
employment_income: 30_000
is_aca_eshi_eligible: false
person3:
age: 15
is_aca_eshi_eligible: false
tax_units:
tax_unit:
members: [person1, person2, person3]
households:
household:
members: [person1, person2, person3]
state_fips: 6 # CA
county_fips: 37 # Los Angeles County has two zip3-defined rating areas
aca_zip3_ca_county_la: 904 # e.g., Santa Monica (90405)
output:
tax_unit_medicaid_income_level: 3.0395136 # greater than 2.66 CA CHIP limit
is_medicaid_eligible: [false, false, false]
is_aca_ptc_eligible: [true, true, true]
aca_slspc_ca: 13_160 # same as 2022 KFF ACA calculator
aca_magi_percent: 318
aca_ptc_phase_out_rate: 0.0645
aca_ptc_ca: 8_645 # $8,631 per 2022 KFF ACA calculator
# annual contribution amount: 0.0645 * 70000 = 4515
# PEUS: 13160 - 8645 = 4515
# KFF: 13160 - 8631 = 4529
# 14 diff is a bit more than $1/month
# KFF implied applicable figure is 0.0647, which is between
# 318 ==> 0.0645 and 319 ==> 0.0648, so it appears as if KFF is
# not truncating the MAGI/FPL percent as called for on IRS Form 8962

# Two 2023 integration tests (one for each LA rating area):

# Read the NOTE below the following two tests to understand why they
# are specified the way they are and why an alternative specification
# of these two tests also causes differences between the results
# generated by PolicyEngine-US and the 2023 KFF ACA Calculator.

- name: aca_ptc_los_angeles test 23-15
absolute_error_margin: 0.00001
period: 2023
input:
people:
person1:
age: 44
employment_income: 30_000
is_aca_eshi_eligible: false
person2:
age: 42
employment_income: 30_000
is_aca_eshi_eligible: false
person3:
age: 12
is_aca_eshi_eligible: false
tax_units:
tax_unit:
members: [person1, person2, person3]
households:
household:
members: [person1, person2, person3]
state_fips: 6 # CA
county_fips: 37 # Los Angeles County has two zip3-defined rating areas
aca_zip3_ca_county_la: 908 # e.g., Long Beach (90805)
output:
tax_unit_medicaid_income_level: 2.413516 # less than 2.66 CA CHIP limit
is_medicaid_eligible: [false, false, true] # KFF ACA calculator misses this
is_aca_ptc_eligible: [true, true, false]
aca_slspc_ca: 8_569 # $10,977 per 2023 KFF ACA calculator (for three)
aca_magi_percent: 260
aca_ptc_phase_out_rate: 0.0440
aca_ptc_ca: 5_929 # $8,325 per 2023 KFF ACA calculator (for three)
# annual contribution amount: 0.044 * 60000 = 2640
# PEUS: 8569 - 5929 = 2640
# KFF: 10977 - 8325 = 2652
# 12 diff is $1/month
# FPL = 23030 (for three in 2022), so exact MAGI/FPL = ~260.53%
# KFF implied applicable figure is 0.0442, which is half way between
# 260 ==> 0.0440 and 261 ==> 0.0444, so it appears as if KFF is
# not truncating the MAGI/FPL percent as called for on IRS Form 8962

- name: aca_ptc_los_angeles test 23-16
absolute_error_margin: 0.00001
period: 2023
input:
people:
person1:
age: 44
employment_income: 30_000
is_aca_eshi_eligible: false
person2:
age: 42
employment_income: 30_000
is_aca_eshi_eligible: false
person3:
age: 12
is_aca_eshi_eligible: false
tax_units:
tax_unit:
members: [person1, person2, person3]
households:
household:
members: [person1, person2, person3]
state_fips: 6 # CA
county_fips: 37 # Los Angeles County has two zip3-defined rating areas
aca_zip3_ca_county_la: 905 # e.g., Torrance (90505)
output:
tax_unit_medicaid_income_level: 2.413516 # less than 2.66 CA CHIP limit
is_medicaid_eligible: [false, false, true] # KFF ACA calculator misses this
is_aca_ptc_eligible: [true, true, false]
aca_slspc_ca: 9_039 # $11,579 per 2023 KFF ACA calculator (for three)
aca_magi_percent: 260
aca_ptc_phase_out_rate: 0.0440
aca_ptc_ca: 6_399 # $8,927 per 2023 KFF ACA calculator (for three)
# annual contribution amount: 0.044 * 60000 = 2640
# PEUS: 9039 - 6399 = 2640
# KFF: 11579 - 8927 = 2652
# 12 diff is $1/month
# FPL = 23030 (for three in 2022), so exact MAGI/FPL = ~260.53%
# KFF implied applicable figure is 0.0442, which is half way between
# 260 ==> 0.0440 and 261 ==> 0.0444, so it appears as if KFF is
# not truncating the MAGI/FPL percent as called for on IRS Form 8962

# NOTE on the above two tests:
# In the above two tests, the three-person tax unit has an income
# low enough for the twelve-year-old child to be eligible for CHIP
# in California. If we had specified the tax unit to include just
# the two adults, PolicyEngine-US would have produced the same SLSPC
# amount as produced by the KFF ACA Calculator. But apparently
# there is no way to tell the KFF ACA Calculator that the two-person
# ACA health insurance unit is part of a three-person tax unit.
# Given that, the KFF ACA Calculator assumes a two-person tax unit,
# and hence, produces a MAGI/FPL percent that is too high. This
# leads to an overestimate of the annual contribution amount and
# leads to an underestimate of the PTC. We have decided to use the
# KFF ACA Calculator in a way that makes it easier to understand the
# differences in the results.
#
# It is important to understand that pointing out these differences
# is not meant to be a criticism of the KFF ACA Calculator. On the
# contrary, the collection of KFF ACA Calculators (going back almost
# to the start of the ACA) is a unique resource for historical data
# on the pricing of second-lowest-cost ACA silver plans. It would be
# only a modest exaggeration to characterize the KFF ACA Calculators
# as a national treasure.

0 comments on commit 02ba890

Please sign in to comment.