So I would agree with this for the Schwarzschild metric. The swapping of signs in the space and time components of the metric does not have any real physical co

So I would agree with this for the Schwarzschild metric. The swapping of signs in the space and time components of the metric does not have any real physical consequences. But when you move to a Kerr metric it absolutely does because it permits the construction of closed timelike curves.