The results of two-photon interferometer experiments are nonclassical and have been interpreted as providing evidence against local realistic theories. But half of the events must be discarded, and it has been suggested that there may be local realistic theories that are consistent with the experimental results as a consequence. It is shown here that the predicted results of two-photon interferometer experiments, including the discarded events, are inconsistent with any local realistic theory in the limit of high detector efficiency if suitable shutters and randomly chosen phase shifts are included. The shutters and randomly chosen phase shifts are unnecessary for any theory in which it is assumed that the photons and their associated hidden variables must propagate away from the source at the speed of light.

• Received 7 June 1999

DOI:https://doi.org/10.1103/PhysRevA.61.012105