A comment on Kerr–CFT and Wald entropy

We point out that the entropies of black holes in general diffeomorphism invariant theories, computed using the Kerr–CFT correspondence and the Wald formula (as implemented in the entropy function formalism), need not always agree. A simple way to illustrate this is to consider Einstein–Gauss–Bonnet gravity in four dimensions, where the Gauss–Bonnet term is topological. This means that the central charge of Kerr–CFT computed in the Barnich–Brandt–Compere formalism remains the same as in Einstein gravity, while the entropy computed using the entropy function gives a universal correction proportional to the Gauss–Bonnet coupling. We argue that at least in this example, the Kerr–CFT result is the physically reasonable one. The resolution to this discrepancy might lie in a better understanding of boundary terms.