GETTSIM: The German Taxes and Transfers Simulator

• Applicants: Professor Dr. Hans-Martin von Gaudecker, since 2/2024; Dr. Günter Kniesel-Wünsche; Professor Dr. Andreas Peichl
• Subject Area: Economic Policy, Applied Economics ; Software Engineering and Programming Languages
• Term: since 2021
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 443188837

This project seeks to establish the software GETTSIM as the reference implementation for modelling the German Taxes and Transfers system and to grow it into a community-driven package. Such software is a crucial building block for a wide array of economic modelling. Currently, any researcher who requires a depiction of the German taxes and transfers system needs to code it up by him- or herself unless he or she happens to have access to one of several in-house systems. Doing so is very inefficient, error-prone, and makes it impossible to compare simulations that are based on different software platforms. Moreover, it inhibits the widespread use of German microeconomic data. Modelling reforms is extremely hard even within the currently available proprietary code bases because their implementation language is ill-suited for incremental software evolution.On the basis of the recently developed Python package GETTSIM, the first step in this project aims to complete its features, improve its robustness, and enhance its performance. In order to grow it into a community package, we need to lower the bar regarding (Python-)programming proficiency. We will thus develop a Domain Specific Language capturing the semantics of the taxes and transfers system. This will allow domain experts who are not programmers to use it and even to code up complex policy reforms. It will also enable them to take advantage of highly performant parallel hardware for complex simulations, without any own knowledge of the intricacies of parallel programming. The open source nature of the code and high development standards ensure that modelling results are verifiable and reproducible.

• DFG Programme: Research Grants
• Ehemaliger Antragsteller: Professor Dr. Holger Bonin, until 2/2024