This paper is concerned with how coupling can be used to enhance the e
fficiency of a certain class of terminating simulations, in Markov pro
cess settings in which the stationary distribution is known. We are ab
le to theoretically establish that our coupling-based estimator is oft
en more efficient than the naive estimator. In addition, we discuss ex
tensions of our methodology to Markov process settings in which conven
tional coupling fails and show (for Doeblin chains) that knowledge of
the stationary distribution is sometimes unnecessary.