Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
The regulation of State Aid is crucial for a well-functioning European Union Single Market. However, both non-compliance of Member States and subsidies from abroad can jeopardize the level playing field. This paper uses machine learning techniques applied to financial statements data to detect potentially distortive public subsidies to companies in the European Union Single Market. We achieve high out-of-sample predictive accuracy and use the machine predictions to flag suspect cases of hidden recipients and explore the characteristics of these firms.