Optimizing Tax Administration Policies with Machine Learning

27 Pages Posted: 11 Mar 2020

See all articles by Pietro Battiston

Pietro Battiston

University of Pisa - Department of Economics and Management

Simona Gamba

Catholic University of the Sacred Heart of Milan - Department of Economics and Finance

Alessandro Santoro

Università degli Studi di Milano-Bicocca - Center for Interdisciplinary Studies in Economics, Psychology & Social Sciences (CISEPS); Università degli Studi di Milano-Bicocca - Department of Economics, Management and Statistics (DEMS)

Date Written: March 11, 2020

Abstract

Tax authorities around the world are increasingly employing data mining and machine learning algorithms to predict individual behaviour. Although the traditional literature on optimal tax adminis- tration provides useful tools for ex-post evaluation of policies, it dis- regards the problem of which taxpayers to target. This study identifies and characterises a loss function that assigns a social cost to any prediction-based policy. We define such measure as the difference between the social welfare of a given policy and that of an ideal pol- icy unaffected by prediction errors. We show how this loss function shares a relationship with the receiver operating characteristic curve, a standard statistical tool used to evaluate prediction performance. Subsequently, we apply our measure to predict inaccurate tax returns issued by self-employed and sole proprietorships in Italy. In our ap- plication, a random forest model provides the best prediction: we show how it can be interpreted using measures of variable importance developed in the machine learning literature.

Keywords: policy prediction problems, tax behavior, big data, machine learning

JEL Classification: H26, H32, C53

Suggested Citation

Battiston, Pietro and Gamba, Simona and Santoro, Alessandro, Optimizing Tax Administration Policies with Machine Learning (March 11, 2020). University of Milan Bicocca Department of Economics, Management and Statistics Working Paper No. 436, Available at SSRN: https://ssrn.com/abstract=3552533

Pietro Battiston

University of Pisa - Department of Economics and Management ( email )

Pisa
Italy

Simona Gamba

Catholic University of the Sacred Heart of Milan - Department of Economics and Finance ( email )

Via Necchi 5
Milan, MI 20123
Italy

Alessandro Santoro (Contact Author)

Università degli Studi di Milano-Bicocca - Center for Interdisciplinary Studies in Economics, Psychology & Social Sciences (CISEPS) ( email )

Piazza dell'Ateneo Nuovo, 1
Milano, 20126
Italy

Università degli Studi di Milano-Bicocca - Department of Economics, Management and Statistics (DEMS) ( email )

Piazza dell'Ateneo Nuovo, 1
Milan, 20126
Italy

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
255
Abstract Views
1,206
Rank
181,143
PlumX Metrics