Graph Analysis Uncovers Corporate Tax Strategies in Global Networks

Multinational corporations often minimize their tax bills through complex legal strategies, impacting government revenues and public services worldwide. A new study leverages knowledge graphs to map corporate structures and uncover these tactics, offering a tool for regulators and the public to understand and address tax avoidance. This research matters because it sheds light on how large companies legally reduce taxes, which can affect everything from infrastructure funding to economic inequality. The key finding is that companies frequently use specific structures, like having legal addresses in low-tax countries while headquarters are elsewhere, to lower their overall tax burden. For example, the study identifies patterns where subsidiaries in tax havens like the Cayman Islands are linked to parent companies in higher-tax nations, leading to significant tax savings. This was discovered by analyzing a vast network of corporate relationships and economic data. The methodology involved building a knowledge graph from public data sources, including the Global Legal Entity Identifier Foundation (GLEIF), which provides information on company ownership and locations. Researchers combined this with data on corporate tax rates and economic indicators from the World Bank and Wikidata. They used graph queries to detect known tax avoidance schemes, such as the 'Double Irish with a Dutch Arrangement,' where companies in Ireland and the Netherlands are used to shift profits to tax havens. Results from the paper show that in cases where a company's legal and headquarter addresses differ, the corporate tax rate in the legal address country is, on average, 10.5 percentage points lower. The analysis also revealed anomalies, such as high company densities in tax havens like Liechtenstein, where there is one company for every three inhabitants. Specific queries identified real-world examples, including a gaming company using a 'duck-rabbit' construct to exploit dual residency rules. In a broader context, this work provides a transparent way to scrutinize corporate tax practices, potentially aiding policymakers in closing loopholes and ensuring fair tax contributions. For regular readers, it highlights how data science can expose practices that might otherwise go unnoticed, promoting accountability in global finance. However, the study has limitations, as noted in the paper: the data may be incomplete due to reporting gaps, and not all aggressive strategies are detectable with current methods. Future work could expand the graph to include more entities and relationships for a fuller picture.