A general problem in today’s Ethiopia is the failure or unwillingness to discover income tax base for a period of Pagume. The non existence of income tax base for Pagume means that there is neither factors payment (wage/salary to labor, house rent income to the owner of capital building, profit to entrepreneur) nor income tax revenue from the source side of factors’ income. Therefore, in terms of both short run and long run periods, the personal income tax revenue of Pagume remains undiscovered. Because the current income tax schedules do only cover income tax base of common period (360 days=12 months times 30 days).

Nevertheless, the current income tax law verbally directs every employer how to treat the income tax base of Pagume.

*“Employers have an obligation to withhold the tax from each payment to an employee, and to pay the withheld amounts to the Tax Authority the amount withheld during each calendar month, in applying preceding income attributable to the months ***of Nehassie** and **Pagume** shall be aggregated and treated as the income of one month.” Income tax law (2002) at 1873

The book also addresses concepts of universal fiscal year and income tax. Ethiopia’s fiscal year has a set of 12 months: 11 common months, each of 30 days and one special month of 35 and 36 days once in a leap year. Tax work is too technical and specialized. Therefore, operational concepts such as income tax brackets, optimum taxable income, marginal tax, total income tax revenue, average income tax and disposable income are explained in the book.

To derive the complementary income tax schedules for the special period of 35 and 36 days, the study multiplied the current income tax schedule by factors 1.1667(=35/30) and 1.2(=36/30) respectively. This exercise results in the first and second complementary income tax schedules. The first complements the common income tax schedule when the universal fiscal year has 365 days. But the second complementary schedule complements the common, when the universal fiscal year has 366 days. Therefore the study contributed two complementary income tax schedules to the current one, which could broaden the income tax bases. Broader and the broadest income tax bases yield more tax revenue to the government and disposable income to the tax payer.

Theories need to be tested by empirical work. Therefore, the second objective of the paper is to estimate and forecast universal personal income tax revenue based on the common and the new schedules. The actual personal income tax revenue data for a sample period of 1984-98 is collected from national budgetary accounts. The study generates the undiscovered personal income tax data form the actual data through dividing the actual by 360 days, and multiplying the result by 5 or 6 days. The study aggregates the actual and the generated undiscovered personal income tax revenue data to get universal data.

Therefore, the Common period PITR_{C}, Undiscovered PITR_{P }andUniversalPITR_{U }variables are linearly and separately regressed against time over a sample period of 1984-1998. The coefficient for each model is significant. The PITR forecast is made for the period 1999-2003. The findings show that the undiscovered personal income tax revenue increases as time increases. The direct forecasts for universal PITR rises steadily using the log linear growth compared to a less gradual increase predicted using simple linear trend. Therefore, if the government amends and uses these two complementary schedules together with the common schedule, the forecast result shows that more tax revenue can definitely be mobilized.