The Application of Artificial Neural Network Method to Investigate the Effect of Unemployment on Tax Evasion

Abstract

Although tax is an important component of government revenue which is used to finance government expenditures, there are many causes that lead people to avoid or evade from paying tax. Among the causes of tax evasion, main factors of them are low income of taxpayers, high tax burdens; increase the size of government, trade openness, high inflation and unemployment. This study postulates that unemployment has a high effect on tax evasion. The present study applied the Sensivity Analysis with Artificial Neural Network (ANN) methodology to investigate the effect of unemployment on tax evasion and also to determine the relative importance of unemployment among other causes of tax evasion for Malaysian data from 1963-2012. Results reveal that there is a positive relationship between unemployment and the extent of tax evasion and unemployment has a high effect on tax evasion among other causes of tax evasion.