Skip to content

Activate calibrated credit claiming behavior#3081

Open
martinholmer wants to merge 23 commits into
PSLmodels:masterfrom
martinholmer:activate-claiming
Open

Activate calibrated credit claiming behavior#3081
martinholmer wants to merge 23 commits into
PSLmodels:masterfrom
martinholmer:activate-claiming

Conversation

@martinholmer

@martinholmer martinholmer commented Jun 17, 2026

Copy link
Copy Markdown
Collaborator

This pull request represents step 2 of the two-step development plan described in issue #3080.

Calibrating Claiming Behavior

The value of the eitc_claim_prob_scale parameter is set to 1.04 and the value of the actc_claim_prob_scale parameter is set to 1.10 for all years in the policy_current_law.json file. These scaling factor values were calibrated so that Tax-Calculator, when using TMD input data, generates 2022 aggregate EITC and ACTC amounts roughly equal to the 2022 historical data reported by IRS/SOI.

Aggregate 2022 Credit Amounts ($B)

Credit
Type
IRS/SOI
actual amount
T-C estimate with
full claiming
T-C estimate with
calibrated claiming
EITC 60.1 73.5 60.0
ACTC 34.8 36.8 34.8

The 2022 T-C EITC estimates in the above table imply a participation (or claiming) rate of about 81.6%.
The 2029 T-C EITC estimates are 95.1 and 77.2, which imply a participation rate of about 81.2%.

Revising Tax-Calculator Tests

Because so many of the expected pytest results were derived using older versions of Tax-Calculator that assumed full credit claiming, most of the tests have been revised to switch back to the full-claiming assumption so that the expected test results remain unchanged. The exceptions to this approach are the behavioral responses tests generated by the make brtest command and the input data tests generated by the make idtest command. In those two cases, the new calibrated credit claiming assumptions are used, which means the expected results for those two sets of tests are now somewhat different.

@codecov

codecov Bot commented Jun 17, 2026

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.00%. Comparing base (8bb3c13) to head (83c7dbb).

Additional details and impacted files

Impacted file tree graph

@@            Coverage Diff            @@
##            master     #3081   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           13        13           
  Lines         2659      2659           
=========================================
  Hits          2659      2659           
Flag Coverage Δ
unittests 100.00% <100.00%> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines Coverage Δ
taxcalc/records.py 100.00% <100.00%> (ø)
taxcalc/taxcalcio.py 100.00% <ø> (ø)
🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant