Forward guidance policies are often argued to stimulate economic activity by reducing nominal long term interest rates. We document why a lower nominal long rate is neither necessary nor sufficient for forward guidance to be successful. We determine the mechanisms behind widely varying long rate responses in existing empirical structural models. Imperfect information about the rationale for forward guidance can severely distort long rate effects and attenuate much of its expansive effect. These results suggest caution in interpreting event-studies of forward guidance.