Mr. Robert Chote:

I thank the Chairman for the invitation. It is a great pleasure to be here. Perhaps the easiest thing to do would be to say a bit about the functions of the Office for Budget Responsibility, OBR, and how the costing role, in particular, fits in. The OBR was established in 2010. Our tasks are fourfold. Twice a year, we produce five-year-ahead forecasts for the economy and the public finances. In contrast to many other countries, the British Government no longer produces projections of its own. In some cases, independent fiscal bodies comment on the Government's numbers. In some instances, they produce parallel sets of numbers which can be compared. In the UK, the OBR produces the official numbers and it is then for Ministers to respond to those in setting out and explaining their policy decisions. We use those five-year-ahead forecasts to assess the British Government's progress against the domestic fiscal rules it has set. It is not our job to say whether those rules are sensible or whether the Government should aim to hit them at any given moment, it is our job to say whether the Government has a better than 50% chance of hitting them as they stand.

The third element, which leads directly to the issue of costings, is that we scrutinise the costing of individual measures included in budgets. We have a thing called the autumn statement, which is supposed to be different but, to all intents and purposes, is just a second bite at the same cherry and a mini-budget. The Treasury submits a list of policy measures that it is considering at the time of each of these statements. It submits costings to us so if it is a tax measure, it will be analysis done by the revenue and customs department and if it is a social security measure, it will be done by the Department for Work and Pensions. We look at that and have an iterative discussion. All of this takes place confidentially before the budget is announced. At the end of the day, we have to say publicly whether we agree or disagree with it or that the Government did not give us enough time or information to reach a considered judgment. It is important to remember that because we are producing the forecast and the forecast necessarily has to include the impact of any measures that have been announced, if we did not agree with the Treasury's estimate of what a particular thing is going to cost or raise, we would have to put our own number in to have a comprehensive forecast that stands in its own right.

Although formally speaking the process of scrutinising the costings is somewhat separate from the forecast, in fact we are producing a forecast. The Government is announcing policy measures and we must incorporate those in the package we produce.

The second round effects came up in earlier discussions and are the knock-on effects of measures on the economy and on the fiscal forecast. We tend to look at the second round effects of the package of measures as a whole. An average budget or autumn statement might have 50 to 80 measures and we would take an overall view, taking into account the fact that some are giving with one hand and some are taking with the other. There will be some measures that put inflation up and some that push it down. There might be some measures that affect the cost of capital in one direction or another. We do not provide a second round effect estimate for each and every measure. We look at that in the aggregate as part of the process.

The fourth and final thing we do is to produce long-term fiscal projections over a 50-year rather than a five-year rise, which many fiscal councils do. This brings in demographics and issues of that kind such as the evolution of a student loan system. We look at the measures of the public sector's balance sheet which allows one to bring in more elements of risk as well as the central forecast.

In the role we have been given, we are confined by legislation to analysing only the current policy of the current Government so we do not look at alternative policy measures that are put forward by Members of Parliament or by political parties. Since the OBR was created, there has been a long-running debate about whether we should take on the sort of role that the Dutch fiscal council has, which involves carrying out an explicit analysis of election manifestos and what their net effect on the public finances and the economy would be. That would be very largely a costing exercise and the decision so far has been that we should not play that role, although, for example, the chairman of the Treasury committee in the House of Commons still thinks that we should and is trying to push water up that particular hill. I think the letter that I sent to the chairman of the Treasury committee when this issue was raised has already been circulated to members of the committee. The letter dealt with some of the pros and cons and the practical issues that would be raised by that, some of which came up implicitly in the discussion the committee has just had with Professor John McHale and colleagues, for example, where ownership of these things lies or whether one is reliant on the expertise in Departments. We are a relatively small office of a maximum of only 30 people over the next few years.

In terms of the costings and some of the discussion about what we should look at, we do not look at the costing of individual measures, for example, a new aircraft carrier programme or whether we should build an exciting new railway line connecting the north and south of the UK. As a reflection of the fact that the Treasury keeps very tight control of the aggregate amount of expenditure on what one might think of as public services and capital investment, when we are doing our forecasts we reach a judgment on whether those aggregate totals are going to be overshot or undershot and by how much. This allows one to bring in some departmental intelligence, for example, it is a bad winter and the health service is having problems or the country is fighting a war and there is additional expenditure of that sort, but we do not do a bottom-up forecast regarding what we think everything the defence or transport Departments have said they are going to do will cost. There are elements of the costing function that some fiscal councils and parliamentary budget offices do which we do not even touch in our more limited scope. I will leave it there and I am happy to expand on any of it.