A Modern Compiler for the French Tax Code

TL;DR

This paper introduces Mlang, a modern, open-source compiler that replaces the outdated legacy system used by the French tax authority, improving maintainability, correctness, and enabling advanced analysis of tax computation.

Contribution

The paper presents Mlang, a formally validated, modern compiler for French income tax algorithms, replacing legacy code and supporting new language features and analysis.

Findings

Mlang is thoroughly validated against DGFiP test suite.

It provides formal semantics and modern language compilation.

DGFiP is transitioning to Mlang for production use.

Abstract

In France, income tax is computed from taxpayers' individual returns, using an algorithm that is authored, designed and maintained by the French Public Finances Directorate (DGFiP). This algorithm relies on a legacy custom language and compiler originally designed in 1990, which unlike French wine, did not age well with time. Owing to the shortcomings of the input language and the technical limitations of the compiler, the algorithm is proving harder and harder to maintain, relying on ad-hoc behaviors and workarounds to implement the most recent changes in tax law. Competence loss and aging code also mean that the system does not benefit from any modern compiler techniques that would increase confidence in the implementation. We overhaul this infrastructure and present Mlang, an open-source compiler toolchain whose goal is to replace the existing infrastructure. Mlang is based on a reverse-engineered formalization of the DGFiP's system, and has been thoroughly validated against the private DGFiP test suite. As such, Mlang has a formal semantics; eliminates previous handwritten workarounds in C; compiles to modern languages (Python); and enables a variety of instrumentations, providing deep insights about the essence of French income tax computation. The DGFiP is now officially transitioning to Mlang for their production system.