BTW, it's not being means-based that causes the existing system to have perverse incentives; the NIT is also means-based by definition. The difference is mathematical: whether the subsidy is a step function of income, or a linear function. Step-function subsidies always create perverse incentives. Unfortunately, they are ubiquitous. I dream of the day when our legislators have sufficient mathematical ability and understanding to select appropriate linear functions instead.
Anyway, I came across a paper explaining why the NIT is a non-starter, politically, in the US. The basic problem seems to be that there are too many people -- especially young single men -- who currently receive no subsidy yet would be entitled to one under the NIT. This group would make the program too expensive.