Amend a Benefit and Build a Country Pilot

By the end you will have a reviewable amendment exercise and a pilot brief for your own country.

The question

“How would a proposed revaluation change our allowances, and who must decide whether it may be adopted?” Predict the Bernard household’s increase before opening Chapter 14’s solution.

Use the fictional Martin, Bernard and Petit cohort, without age supplements or forfait, with the April 2026 legal date. This is a policy experiment, not an enacted instruction. The Garniers’ housing file and the Fontaines’ notice file remain separate.

The source and amendment

The pinned DSS/2B/2026/46 instruction states:

le coefficient de revalorisation de la base mensuelle des allocations familiales (BMAF) … est fixé à 1,008

English gloss: the BMAF revaluation coefficient is set at 1.008. Projet → draft; revalorisation → revaluation. A departure from the statutory indexation formula requires its own legal basis.

Use the actual fictional amendment from Chapter 14. Labels, dates and source anchors are omitted:

rule CoeffNominal(e: Exercice, c: Decimal) strict {
    when coefficient_annee(e, c) and c >= 1.0;
    then coefficient_retenu(e, c * 1.05);
}

It replaces the canon’s retention of c with c * 1.05. Keep the baseline separately.

Your submission

1. Check: inspect the recorded draft-check request and response. Explain what EXPRESSIBLE establishes.
2. Evaluate: inspect the draft execution from BMAF €474.37 and coefficient 1.008. Identify the retained coefficient and new BMAF.
3. Impact: inspect which rule changes and what the report observed. Do not use its absence of scenario observations as evidence of no effect.
4. Three families: compare their baseline and draft answers, preserving the selected ArrondiLignesDss reading and all other facts.
5. Notice: draft an English demonstration explanation under Chapter 13’s four headings. Identify information the computation cannot supply.
6. Human question: name the competent decision-maker and one unresolved question, such as authority for the departure from indexation. If using the decency example, address the judgment request to the body identified in Chapter 9.
7. Reading: explain why the cent rounding is a named reading rather than a claim about the Cnaf’s operational practice.

The recorded draft scene preserves calls to law_draft_check, law_draft_eval and law_draft_impact, followed by separate questions for the three families. The results below come from the pilot recorded on 9 October 2026 through the local MCP recording server, version 0.1.0. This identifies the recording server, not the CLI. This exercise examines those saved results; it does not report a new run.

Worked solution

The draft is accepted as a formalisation, without establishing legal validity. It retains 1.0584 and produces BMAF €502.07. Impact identifies the semantic change to CoeffNominal; its recorded invocation evaluated no scenarios. The family questions supply the observations:

These are gross AF amounts under the named line-rounding reading, not net transfers or a population budget estimate. The draft first changes the retained coefficient and computes a BMAF of €502.07. Each family is then evaluated for 15 April 2026 with its BMAF input changed from €478.16 to €502.07. All other facts and the selected ArrondiLignesDss reading stay the same. The impact report does not perform these family calculations.

Demonstration explanation — Bernard household

Degree and manner of the algorithm’s contribution. The model calculates an illustrative allowance from supplied facts. The recording establishes no administrative decision or human approval.

Data processed and their sources. The fictional Bernard case supplies three dependent children, metropolitan residence, resources and published ceilings, with no age supplements or forfait.

Parameters and any weights applied. The calculation uses the supplied BMAF and the named ArrondiLignesDss reading, which rounds each allowance line to the nearest cent using HALF_UP. No population weights are used. This named reading does not establish the Cnaf’s operational practice.

Operations performed. Separate saved calculations give gross monthly AF of €349.06 before the change and €366.51 after it; a saved subtraction gives €17.45.

Question for the policy team’s legal reviewer: what legal basis would authorise this departure from statutory indexation, and which authority could adopt it?

This is an English teaching example. It has not undergone legal and editorial review by a native French legal reviewer and is not ready to send.

What this gives the fund

The submission brings the amendment, case consequences and explanation into a reviewable file. It distinguishes a well-formed draft, observed effects and administrative approval. Three families provide concrete checks; representative data and further review are needed for a policy assessment.

Where Catala stands on this

OpenFisca provides a reform mechanism, and Catala supports assertion tests, output tests and continuous integration (Catala book, chapter 3.3). Arxo provides built-in draft checks, draft execution and impact reports alongside case answers with recorded grounds. Chapter 3.3 of the Catala 1.3.0 book does not describe dedicated draft-check or impact-report forms. For the fund, this brings together the proposed rule change, its computed effects and the questions still requiring review.

Country-pilot brief

Complete this template for a programme in your own country:

Acceptance checklist

Your draft is separate from the baseline; each amount has a preserved question and grounds; each family card is identified; readings are explicit; the notice follows the four headings; and the unresolved human question is named. Reproduction, scenario expectations and legal approval remain separate checks.