Estimating Factors Affecting Tax Evasion in Malaysia: A Neural Network Method Analysis

Abstract

Tax is one of the ways to finance government expenditures and it plays an important role in increasing the government revenue. The amount of tax collected depends on the structure of the economy. When the taxation system is ineffective, many people will use this opportunity to avoid paying tax and tax evasion will be popular. In the presence of tax evasion, the government cannot allocate revenue for their programs and cannot provide desirable social services. Realizing the significant impact of tax evasion on the economy, this study tries to determine factors that cause tax evasion and their relative contribution. The study employs an Artificial Neural Network methodology on Malaysian data from 1963-2010. The results show that the tax burdens, the size of governments and inflation rate have positive effect on tax evasion, but there is negative relationship between tax payers’ income and trade openness with tax evasion.