Mr. Robert Chote:

Yes.

It is an iterative process. For example, six to ten weeks ahead of a budget we would start working on a baseline forecast of what would happen to the economy and the public finances if there were no change in policy. Simultaneously, the Government will provide us with what is known as a draft scorecard, which is a list of measures that it is thinking of announcing, most of which would be specific tax and welfare transfer measures, plus decisions on the overall size of the spending envelope for public services. As I said, it would not come to us on issues such as the purchase and cost of aircraft carriers.

As the budget gets closer we go through successive iterations in finalising our pre-measures forecast. We also go through a process of examination of the list of measures provided by the Treasury. Typically, a lot of the measures provided eight weeks out will have dropped off the list as we get closer to the budget - in part, perhaps, because during the course of our discussions on a measure and the likely cost the Treasury may take the view that it is not something it would be happy with. The Treasury has never yet publicly pressed ahead with publishing a costing estimate that it knows we did not support. It is more likely that it would drop a measure rather than do that. Some measures will come on relatively late in the process, such that there is a nail-biting period in terms of the loss or gain of particular measures from or to the list. We set clear deadlines by which the Treasury needs to inform us of a measure that is substantive enough to affect the economic forecast and another deadline in respect of other new measures. We also allow a little wriggle room in relation to measures it has already told us about. For example, it may want to change a tax rate from, say, 11% to 12% or something like that, which involves no great analytical content.

The final process is a pre-measures forecast. We can then say what the post-measures forecasts would look like once all of this has been taken into account. At the end of the day, the Treasury can choose what forecast it wants us to publish for the public finances, or it can choose the policies, but it cannot choose both simultaneously unless we think the two are consistent. There is a choice to be made. We do not get up on budget day and surprise the Treasury. We will have had this discussion beforehand and will have set out our view of what particular measures will cost or raise. We then reach an aggregate view on whether a package is a net giveaway and is therefore likely to result in a boost to the economy and to activity in the short term, or whether it is something that is likely to push measured inflation higher or lower. As I said, this is done at an aggregate level rather than measure by measure.