Mr. John Beggs:

Yes. I have to say I was very, very surprised to hear what Professor John FitzGerald had to say at the committee meeting. I've known him quite a long time. We've worked together in the Department of Finance in the 1970s. We've had some interactions, over the years. I ... I suppose, at my own initiative, I decided that what we needed to do ... I was doing the stress testing, shall I say. I may not be doing the stress testing, you know, indefinitely and I wanted to bring some consistency to the way in which these stresses and the scenarios were being presented to the risk analysts who had to carry out the exercise. A model was one way of doing it. So, I approached John FitzGerald with a view to, first of all, inputting the regulatory stress test, the data that we were given by, say, the Central Bank, which were very limited in number. I wanted to run those results through the model to generate a wider scene setting because the model generates quite a lot of information. And as Professor FitzGerald states in his ... in the letter to the committee, the model could do that and that, for me, was an important input into the stress test exercise to be carried out within the bank itself. I wanted the business units to get a full sense of ... the fullest possible sense as to what this stress involved. With the benefit of hindsight of what happened, it's quite clear that having the clearest and the most, you know, expansive version of a stress and what it would involve might have been useful.