Well, I know that the platform used some prudence factor in estimating the cost. What I can tell you in more detail is what Finance has done.
We also factored in a prudence factor, but the factor was a bit more conservative and is based on the literature review of the typical behaviour of particularly high-income individuals facing an income tax rate increase. In technical terms, if you will, we used what we call an elasticity factor of 0.4. This by and large means that if you increase the tax rate by 10%, individuals will react by reducing their income tax base by 4%—that's the relationship. We used that factor to come up with our cost estimate.
Similarly for the second tax bracket cut, we were a bit more conservative. We were not assuming any actual behavioural effect, because there was not enough support in the literature to justify using a behavioural effect there. We just took what we call our static cost estimate—not cost, but—