The hidden economy revealed: profiling informal workers through random forest classification

The phenomenon of informality, in all three of its forms: envelope wages, dependent workers, and workers without formal contracts, represents a complex aspect of the modern economy, having significant impact on regional development and society. The main objective of this study is to understand this phenomenon in both microeconomic and macroeconomic contexts, exploring the interactions between individual and structural factors that influence the evolution and scale of the informal economy across different regions. While the classical approach typically involves multilevel models to analyse such data, this study employs the Random Forest method due to its ability to handle data complexity and variability efficiently and robustly. The profile of informal workers, regardless of their employment form (no formal contract, envelope wages, or dependent workers), reveals both common traits and notable differences. While all share low fiscal morality, extended social networks, and a general distrust of authorities, they differ significantly in terms of company size, demographics, and political inclination, highlighting the complexity and diversity of informal labour and the need for tailored approaches to address these workers.