FAILURE MAP
← Case archive

FA-62531 / Tax bracket computation / Open access

The exemption phase-out is computed on total AMTI instead of the excess · case 01

The exemption disappears for filers barely over the threshold.

Verified by executionVariant 1 · 6 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

The 25% reduction is applied to all AMTI rather than AMTI above the threshold.

THE FAILURE

The 25% reduction is applied to all AMTI rather than AMTI above the threshold.

Unsuccessful approach: Measuring the excess over the exemption amount instead of the threshold still over-reduces.

Case contract

solve(amti, regular_cents, status): stipulated AMT, whole dollars. Exemption: single 85700 (phase-out threshold 609350), mfj 133300 (threshold 1218700). The exemption is reduced by 25% of AMTI above the threshold, with the reduction rounded down to whole dollars, never below 0. Base = max(0, amti - exemption). Tentative minimum tax = 26% of base up to 232600 plus 28% of base above 232600 (in cents: dollars times percent). AMT = max(0, TMT - regular tax). Return [exemption, tmt_cents, amt_cents].

Why this case matters

Tax computations hinge on which slice, threshold, ordering and rounding rule applies at each step; a misplaced boundary silently misstates liabilities.

1 / The failure

Exit 1
"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(amti, regular_cents, status):
    ex, thr = {'single': (85700, 609350), 'mfj': (133300, 1218700)}[status]
    if amti > thr: ex = max(0, ex - amti // 4)
    base = max(0, amti - ex)
    tmt = min(base, 232600) * 26 + max(0, base - 232600) * 28
    return [ex, tmt, max(0, tmt - regular_cents)]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
cases = [[('regression phaseout-base 1', (700000, 0, 'single'), [63038, 17369736, 17369736]),
  ('regression phaseout-base 2', (609353, 5000000, 'single'), [85700, 14197084, 9197084]),
  ('control: moderate single', (200000, 2500000, 'single'), [85700, 2971800, 471800]),
  ('control: no exemption left', (1000000, 99999999, 'single'), [0, 27534800, 0]),
  ('control: below exemption', (100000, 0, 'mfj'), [133300, 0, 0]),
  ('control: joint high', (1300000, 30000000, 'mfj'), [112975, 32771500, 2771500])],
 [('regression phaseout-base 1', (1300000, 30000000, 'mfj'), [112975, 32771500, 2771500]),
  ('regression phaseout-base 2', (832985, 0, 'single'), [29792, 22024204, 22024204]),
  ('control: below exemption', (100000, 0, 'mfj'), [133300, 0, 0]),
  ('control: odd excess', (609353, 5000000, 'single'), [85700, 14197084, 9197084]),
  ('control: rate break', (400000, 0, 'mfj'), [133300, 7002400, 7002400]),
  ('control: moderate single', (200000, 2500000, 'single'), [85700, 2971800, 471800])],
 [('regression phaseout-base 1', (633603, 10000000, 'single'), [79637, 15045848, 5045848]),
  ('regression phaseout-base 2', (869146, 10000000, 'single'), [20751, 23289860, 13289860]),
  ('control: rate break', (400000, 0, 'mfj'), [133300, 7002400, 7002400]),
  ('control: moderate single', (200000, 2500000, 'single'), [85700, 2971800, 471800]),
  ('control: partial phase-out', (700000, 0, 'single'), [63038, 17369736, 17369736]),
  ('control: no exemption left', (1000000, 99999999, 'single'), [0, 27534800, 0])],
 [('regression phaseout-base 1', (853491, 0, 'single'), [24665, 22741928, 22741928]),
  ('regression phaseout-base 2', (1388032, 0, 'mfj'), [90967, 35852620, 35852620]),
  ('control: no exemption left', (1000000, 99999999, 'single'), [0, 27534800, 0]),
  ('control: below exemption', (100000, 0, 'mfj'), [133300, 0, 0]),
  ('control: odd excess', (609353, 5000000, 'single'), [85700, 14197084, 9197084]),
  ('control: joint high', (1300000, 30000000, 'mfj'), [112975, 32771500, 2771500])],
 [('regression phaseout-base 1', (730275, 2500000, 'single'), [55469, 18429368, 15929368]),
  ('regression phaseout-base 2', (932132, 2500000, 'single'), [5005, 25494356, 22994356]),
  ('control: joint high', (1300000, 30000000, 'mfj'), [112975, 32771500, 2771500]),
  ('control: rate break', (400000, 0, 'mfj'), [133300, 7002400, 7002400]),
  ('control: moderate single', (200000, 2500000, 'single'), [85700, 2971800, 471800]),
  ('control: partial phase-out', (700000, 0, 'single'), [63038, 17369736, 17369736])]]
for label, args, expected in cases[N - 1]:
    check(label, solve(*args), expected)
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)
Boundary fixtureActualExpectedOutcome
regression phaseout-base 1[0, 19134800, 19134800][63038, 17369736, 17369736]Failed
regression phaseout-base 2[0, 16596684, 11596684][85700, 14197084, 9197084]Failed
control: moderate single[85700, 2971800, 471800][85700, 2971800, 471800]Passed
control: no exemption left[0, 27534800, 0][0, 27534800, 0]Passed
control: below exemption[133300, 0, 0][133300, 0, 0]Passed
control: joint high[0, 35934800, 5934800][112975, 32771500, 2771500]Failed

SHA-256 / 7aa8403c400983478899d4d3698d652ee8fced51054e5c8cc375d1e1baeb129f

2 / The unsuccessful fix

Exit 1
"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(amti, regular_cents, status):
    ex, thr = {'single': (85700, 609350), 'mfj': (133300, 1218700)}[status]
    if amti > thr: ex = max(0, ex - (amti - ex) // 4)
    base = max(0, amti - ex)
    tmt = min(base, 232600) * 26 + max(0, base - 232600) * 28
    return [ex, tmt, max(0, tmt - regular_cents)]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
cases = [[('regression phaseout-base 1', (700000, 0, 'single'), [63038, 17369736, 17369736]),
  ('regression phaseout-base 2', (609353, 5000000, 'single'), [85700, 14197084, 9197084]),
  ('control: moderate single', (200000, 2500000, 'single'), [85700, 2971800, 471800]),
  ('control: no exemption left', (1000000, 99999999, 'single'), [0, 27534800, 0]),
  ('control: below exemption', (100000, 0, 'mfj'), [133300, 0, 0]),
  ('control: joint high', (1300000, 30000000, 'mfj'), [112975, 32771500, 2771500])],
 [('regression phaseout-base 1', (1300000, 30000000, 'mfj'), [112975, 32771500, 2771500]),
  ('regression phaseout-base 2', (832985, 0, 'single'), [29792, 22024204, 22024204]),
  ('control: below exemption', (100000, 0, 'mfj'), [133300, 0, 0]),
  ('control: odd excess', (609353, 5000000, 'single'), [85700, 14197084, 9197084]),
  ('control: rate break', (400000, 0, 'mfj'), [133300, 7002400, 7002400]),
  ('control: moderate single', (200000, 2500000, 'single'), [85700, 2971800, 471800])],
 [('regression phaseout-base 1', (633603, 10000000, 'single'), [79637, 15045848, 5045848]),
  ('regression phaseout-base 2', (869146, 10000000, 'single'), [20751, 23289860, 13289860]),
  ('control: rate break', (400000, 0, 'mfj'), [133300, 7002400, 7002400]),
  ('control: moderate single', (200000, 2500000, 'single'), [85700, 2971800, 471800]),
  ('control: partial phase-out', (700000, 0, 'single'), [63038, 17369736, 17369736]),
  ('control: no exemption left', (1000000, 99999999, 'single'), [0, 27534800, 0])],
 [('regression phaseout-base 1', (853491, 0, 'single'), [24665, 22741928, 22741928]),
  ('regression phaseout-base 2', (1388032, 0, 'mfj'), [90967, 35852620, 35852620]),
  ('control: no exemption left', (1000000, 99999999, 'single'), [0, 27534800, 0]),
  ('control: below exemption', (100000, 0, 'mfj'), [133300, 0, 0]),
  ('control: odd excess', (609353, 5000000, 'single'), [85700, 14197084, 9197084]),
  ('control: joint high', (1300000, 30000000, 'mfj'), [112975, 32771500, 2771500])],
 [('regression phaseout-base 1', (730275, 2500000, 'single'), [55469, 18429368, 15929368]),
  ('regression phaseout-base 2', (932132, 2500000, 'single'), [5005, 25494356, 22994356]),
  ('control: joint high', (1300000, 30000000, 'mfj'), [112975, 32771500, 2771500]),
  ('control: rate break', (400000, 0, 'mfj'), [133300, 7002400, 7002400]),
  ('control: moderate single', (200000, 2500000, 'single'), [85700, 2971800, 471800]),
  ('control: partial phase-out', (700000, 0, 'single'), [63038, 17369736, 17369736])]]
for label, args, expected in cases[N - 1]:
    check(label, solve(*args), expected)
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)
Boundary fixtureActualExpectedOutcome
regression phaseout-base 1[0, 19134800, 19134800][63038, 17369736, 17369736]Failed
regression phaseout-base 2[0, 16596684, 11596684][85700, 14197084, 9197084]Failed
control: moderate single[85700, 2971800, 471800][85700, 2971800, 471800]Passed
control: no exemption left[0, 27534800, 0][0, 27534800, 0]Passed
control: below exemption[133300, 0, 0][133300, 0, 0]Passed
control: joint high[0, 35934800, 5934800][112975, 32771500, 2771500]Failed

SHA-256 / da463b7f4cbf6b3cfffde5a50ab30b25039d4764d23c69ef4a1600aef01c2f85

HELD IN THE MEMBER ARCHIVE

The verified repair and its recorded checks are member-only.

This mechanism has 6 recorded checks per implementation. The open-access tier publishes the failure and the unsuccessful fix; the repaired source that passes every check, and the observations that prove it, are available to members.

Every case sharing this mechanism uses the same contract and the same repair, so this one record is held back for all of them.

Member access is invitation-based. Sign in with your invited account to inspect the repair.

Sign in to the archive ↗

Verification & scope

A deterministic, bounded teaching model with a stipulated toy contract; it makes no claim of conformance to any real regulation, standard, or institution's rules. This reproducer isolates one failure mechanism. Results cover the supplied fixtures. Variants within a family share a test contract and should remain grouped when constructing evaluation splits. Related mechanisms with a shared evaluation_group must also remain together; these controlled models are not independent production incidents.

Observations recorded using Python 3.12.14 at 2026-09-29T14:47:05.399404+00:00.

Case digest / 9a112261f14518812d611c46c6971b4dfa9831c820d336de4f178bb1c9fa340c