Previous studies have investigated asymmetries in the effects of monetary policy on the real economic activity by using either vector autoregressive (VAR)‐based regime‐switching models with smooth transition technique or Gaussian functions to parameterise the dynamic effects of structural shocks on the economy. These kinds of VAR models assume asymmetry as a short‐run relationship between the series since the long‐run neutrality hypothesis of money states that monetary policy can only affect productive capacity of the economy in the short run, but not in the long run. The recent theoretical literature shows that this hypothesis is not quite right. Thus, this paper examines the extent to which monetary policy has a long‐run asymmetric effect on output in a number of Organisation for Economic Co‐operation and Development countries by using a nonlinear hidden cointegration analysis within a likelihood‐based panel framework. The findings indicate that there is a long‐run relationship between the real interest rate as an indicator of monetary policy and the growth rate of real output in five countries out of nine under review. This gives support for the view that output has responded asymmetrically to the real interest rate changes. The economic implication of our results is that monetary policy affects positive and negative output fluctuations differently.