The idea that the ‘credibility’ of monetary policy is important receives almost universal assent. But the word can mean very different things in different contexts. In particular, the meaning that it has been given by the work of Robert Barro and David Gordon is a very special one. It should not be presumed that in ordinary discussion of ‘credibility’ theirs is always the meaning people have in mind. Consequently, Alan Blinder’s survey on attitudes to ‘credibility’, which failed to specify what he meant by the term, might be most misleading, and if unwarranted inferences are not to be drawn a most careful interpretation of it is required.