Developing a Predictive Model to Estimate the Size of Illegal Gambling Markets

• In Event: Money laundering from a variety of criminal markets: the role of offenders and of intermediaries

Abstract

Illegal gambling constitutes a significant underground economy, generating substantial revenues that evade regulatory oversight. While enforcement efforts primarily focus on individual operators, the broader economic determinants influencing the scale of these markets remain largely unexplored. This study develops a predictive model to estimate the size and evolution of unregulated gambling markets by integrating macroeconomic indicators with advanced statistical techniques. Using a top-down approach, the model incorporates key socioeconomic variables such as gross domestic product (GDP) per capita, unemployment rates, and household expenditures. To enhance predictive accuracy, the study contrasts traditional forecasting methods with machine learning techniques, enabling the identification of complex interdependencies and nonlinear market trends. The model is empirically validated using accessible economic data, assessing its robustness and applicability across countries. Despite increasing recognition of illegal gambling as an economic force, few studies offer standardized, scalable methodologies for quantifying its scope. This research addresses that gap by proposing an adaptable predictive model that provides policymakers and regulators with a data-driven framework to assess and counteract unregulated gambling markets effectively. Beyond identifying key predictors of illegal betting volumes, the model facilitates estimations of market fluctuations based on economic trends. Furthermore, it establishes a standardized methodology for international comparisons, guiding the implementation of evidence-based regulatory policies.

Authors

• Julien Chopin, Université de Lausanne

• Lucas Tosi Rodriguez, Université de Lausanne

• Stefano Caneppele, University of Lausanne