We have models of the distribution of income and wealth in NZ - how many people earn what, and own what. They are of course proximate because the wealth data collected in NZ is very poor. But there are a number of ways we can get reasonable distributions and then run sensitivity experiments to see what difference if any it makes if we chose alternative income/wealth distributions. Then we specify how the tax would work, and for any one simulation you have to pick an income tax scale, decide what exemptions if any you’re going to allow, decide what rates you’re going to apply to this deemed capital income, what rate of capital income you’re going to deem, etc etc. And finally of course the whole package must be tax neutral. Now that just gives you an impact effect. From there you have to assess what if any behavioural changes you’re going to allow for, especially say with business investment and growth because business is a big beneficiary of this as its supply of capital changes now the property gravy train has been halted.
At the end of all that you get a feel for the numbers, you certainly have a range of reasonable outcomes. We found the number better off under various assumptions varied between 75% and 95%, dependent mainly on how we set the new income tax scale. generally though the more that benefited, the lower the benefit for those most of those who do.
Of course they’re models, not reality so nobody is going to say definitively this is what will happen. Economics just isn’t like that, predicting behaviour change is especially tricky. And working out