Get access

Measuring the Underground Economy with the Currency Demand Approach: A Reinterpretation of the Methodology, With an Application to Italy


  • Note: We wish to thank two anonymous referees, Fabio Bagliano, Mark Crain, Domenico Depalo, Fabio Padovano, Michael Pickhardt, Luca Salvadori, Alessandro Santoro, Jordi Sardà, Friedrich Schneider, Paolo Sestito, Brigitte Unger, Roberta Zizza, and all seminar participants at the Workshop on Macroeconomic and Policy Implications of Underground Economy and Tax Evasion (Bocconi University, 2012), the 2nd World Congress of the Public Choice Societies (Miami, 2012), the 52nd Annual Conference of the Italian Economic Association (Roma Tre University, 2011), the 2011 Conference on Shadow Economy, Tax Evasion and Money Laundering (Münster University, Germany), and the Lunch Seminar held at the Bank of Italy (Roma, 2011), for their helpful comments. The usual disclaimers apply.


We contribute to the debate on how to assess the size of the underground (or shadow) economy by proposing a reinterpretation of the traditional Currency Demand Approach (CDA) à la Tanzi. In particular, we introduce three main innovations. First, we take a direct measure of the value of cash transactions—the flow of cash withdrawn from bank accounts relative to total non-cash payments—as the dependent variable in the money demand equation. This allows us to avoid unrealistic assumptions on the velocity of money and the absence of any irregular transaction in a given year, overcoming two severe critiques to the traditional CDA. Second, in place of the tax burden level, usually intended as the main motivation for non-compliance, we include among the covariates two direct indicators of detected tax evasion. Finally, we control also for the role of illegal production considering crimes like drug dealing and prostitution, which—jointly with the shadow economy—contributes to the larger aggregate of the non-observed economy and represents a significant component of total cash payments. We propose then an application of this “modified CDA” to a panel of 91 Italian provinces for the years 2005–08.