Minimising Selection Failure and Measuring Tax Gap: An Empirical Model

This paper presents an empirical model for minimising selection failure by tax departments in selecting cases for scrutiny assessment. This model also provides a new methodology for estimating tax gap from limited information that the department collects on a regular basis through scrutiny assessments. Using a maximum-likelihood procedure that corrects for sample selection bias, and the data on the scrutiny assessment exercise carried out by the income tax department, the model is estimated so that it relates the probability and extent of under-reporting to various inputs provided by the tax filer. The estimated model provides a mechanism to analyse the trade-off between two types of cases of failure - wrong selection of a case and failure to take up the potential underreporter.