*3.2. Answer*

The above comment is apparently based on our phrase "so we derive its value from the estimation of the following translog cost function" (see Page 7 of our research), which should have rather been expressed in a more general manner (e.g., "so the most common approach is to derive its value from the estimation of a translog cost function"), since our only aim there was to show the steps that should have been followed if we had adopted the traditional approach.

In this context, we first state that the basis of our Lerner index estimations is our Equation (7) (see Figure 1).

**Figure 1.** A part of the Lerner index calculation process.

Second, we would like to emphasize that the Kumbhakar et al. (2012) stochastic frontier methodology does not require the estimation of a translog cost function per se, but only the estimation of its partial derivative with respect to output. As shown in Figure 1, we arrived at our Equation (7) after having taken the partial derivative of the translog cost function and "fitting" it into our Equation (5). This explains the reason why we have not included in our research any description at all, either in an equation or in a descriptive text form, of the calculation of marginal cost (*MC*).

#### **4. Di**ff**erences between Our Equation (7) and the Equation (7) in MT's Remarks**
