Proposing an Optimal Model ofTax Calculation Basis

As the main source of government revenue, the tax has always faced numerous challenges on the taxpayers’ part, such as tax evasion, late tax payment, and falsification of tax records. The problems with the calculation of taxable income cause these challenges. Therefore, a change in the calculation basis seems to serve as a solution to these challenges. This study aims to develop an optimal model for tax calculation, where the confirmable income act as the basis for tax calculation. Taxable income results from the subtraction of confirmable revenues and costs from extra-organizational financial resources. The study population consists of financial information of firms listed on the Tehran Stock Exchange from 1997 to 2016. Initially, instead of the taxable income, confirmable income is introduced as an optimal model; afterward, for the sake of accuracy, the tax rates for every firm and year are re-calculated using median, mean, regression, and artificial bee colony algorithm. Then the calculated taxes and real taxes were compared using paired t-tests. Finally, the rate that showed the slightest difference with the real rates, at the confidence level of 95%, is selected as the rate of the Industry/Firm-Year Group. The results show that this new method entails the use of the industry/firm group special rate every year; in other words, the tax rate of each group changes according to the inflation rate and economic conditions. This new method can also assist the government to gain more benefits from tax resources.