Recent literature bas uncovered asymmetries in the response of real output to monetary policy variables. Nevertheless, it remains unclear whether such asymmetries relate to different responses to monetary policy or to the business cycle. This paper uses nonlinear models to examine the issues in the context of interest rate effects on quarterly UK GDP growth Strong evidence of nonlinearity is found, with asymmetry relating to the business cycle through lagged GDP regimes and interest rate changes. The results suggest that interest rate effects on GDP are larger when either lagged growth has been high or when interest rates have substantially increased in the past. However, the inclusion of interest rate regimes without taking account of GDP regimes yields an unsatisfactory model.