One of the advantages of long-prediction-horizon model predictive control (MPC) is its applicability to processes with nonminimum-phase behaviour. Motivated by this attractive feature of MPC, a long-prediction-horizon MPC formulation is used to derive an approximate input-output-linearising nonlinear control law for hyperbolically stable, single-input single-output processes, whether nonminimum-phase or minimum phase. Indeed, the problem of nonlinear control of a class of nonminimum-phase processes is solved by exploiting further the connections between model predictive control and input-output linearisation. The derived control law has one single tunable parameter, and thus is very easy to tune. It is applied to linear processes, and the resulting linear control law is presented.