The Effects of Different Technology Assumptions in the Calculation of Symmetric IOTs on the Bias of Random Production Multipliers

Calculation of symmetric input-output tables is based on two assumptions: fixed structure of product sales and product technology assumptions. The first assumption guarantees the non-negative elements in the table but lacks an acceptable theoretical basis. On the other hand, the second assumption has an acceptable theoretical basis; that is, the emergence of some negative elements is inevitable. These negative elements should be removed using other methods. This issue raises a fundamental question of whether the choice of different technology assumptions in calculating symmetric table has any effect on the bias of random production multipliers or not. Considering the compiler's approach, we have used two IOTs with different technology assumptions for measuring the bias ofrandom production multipliers. Using Monte Carlo simulation for creating samples in 10, 1000, and 10000 sizes, the overall findings show the choice of technology assumptions has an effect on the bias of random production multipliers. Also, the findings of the present study have an important implication for the compilers; that not only product technology assumption is supported by a good theoretical foundation but also the bias of calculated random production multipliers is smaller than the bias of random production multipliers which are estimated based on the fixed structure of product sales.