Select Committee on Treasury Minutes of Evidence

Examination of witnesses (Questions 120 - 139)



  120. I accept that if that is the case, but what I am asking you is, what impact is the fact that they are inaccurate to a degree having on your ability to formulate regional policy?  (Mr Ritchie) To a degree. There is a review—

  121. It is not the view of somebody coming in from the street but the official watchdog.  (Mr Macpherson) I would argue that we are always seeking to improve statistics and we are keen, in order to perform the analysis we were talking about earlier, that we have the best possible data and we are working with the ONS and others in academia to ensure .that that data is available, but I think it is also worth being clear that we are not in a situation where, for example, education spending is being determined by regional GDP. I recognise that this is now far more important in terms of policy making than it was in the past, but a lot of the key formulae do not hinge on regional GDP data.

  122. That may be partly true, but one of the objectives that you have in regional policy now is to bring up regional productivity to nearer the national average. The accuracy of the GDP figures is central to estimating that.  (Mr Macpherson) It is. We will be setting out a technical note underpining our original PSA in the near future. As you say, data is very important, it concentrates the mind for all of us and it certainly concentrated our minds as we talked to the ONS.

  123. Can I move from GDP to estimates of how much public expenditure goes to the different regions. I quote your own memorandum, which says, "The Office of the Deputy Prime Minister is currently leading a feasibility study aimed at identifying the forward level of public expenditure, domestic and European, flowing to individual English regions and for each expenditure flow what determines the level of that expenditure flow". Plainly, if there is an exercise going on to estimate it you do not really know much public expenditure is flowing into the different regions, is that a fair interpretation?  (Mr Scotter) I think the answer to that is we have an estimate of the amount of money which is going into each region which gives us a good, broad picture of the amount going in, it may not be right to the individual pound per head. What we are about to do is improve the quality of those estimates so that we can make policy better. We are trying to get better evidence on which to base policy. I think it would be wrong to think that there is no value in the existing estimate, they have a substantial value but we want to refine them to make sure we have them right. There are some areas of public expenditure which are quite difficult to estimate their regional effects, some are very straightforward, social security spending has been mentioned, that goes to individuals, you know where the individuals live and you can count it for the region. Others bits of expenditure, let us say, the West Coast mainline, improvements on that, it is a bit more difficult to see who benefits from that development, whether it is spread evenly along the length, or whether it is London or Scotland. What our study is trying to do is see what we can do to improve those estimates of the difficult parts.

  124. The upshot of this discussion is that on the information on the amount of public expenditure going into regions it is probably inaccurate or you would not be having this review, and so is the estimate of the GDP, two basic parameters that must lie behind the formulation of regional policy. Is it not fair to say that your attempts to make regional policy in these circumstances is like making bricks with no straw.  (Mr Macpherson) You can exaggerate the problems. I think with things like regional GDP or spending you would not get very precise estimates. Okay, there may be margins of error. I think it is pretty clear, for example, when we are talking about London versus the North East there are quite big differences. That should inform policy. The fact is that the ONS are working very hard to extend and develop their data. I know that they have very good neighbourhood statistics which are due to go live in the near future, which will enable you to drill down to ward level, I think, in terms of a whole lot of indicators. They are working very hard to develop their data. We are happy with the quality of the data as it is but but we recognise that the world is moving on and we need more of it and we need it to be better and more accurate.

  125. On that ONS study I understand that the study that was being done by the ONS to look more carefully at regionalising government accounts, the pilot study, was due to be produced in June.  (Mr Macpherson) It was a pilot study on one small area.[3]

  126. What messages have come out of that pilot study?  (Mr Macpherson) It is due to be published now on 31 October.

  127. Could we have your conclusions on the significance of it when it is published.  (Mr Macpherson) Yes.

Mr Cousins

  128. Tomorrow in this building lots of my colleagues—sadly I will not be jointing them—will be debating the local government finance centre and I dare say this dreaded matter, the Area Cost Adjustment, will haunt the occasion. The Statistics Commission has told us, indeed everyone else that is interested, that ONS do not have the recourses to do a proper analysis of the different cost and price variations between regions to inform the regional GDP figures with some accuracy about price variations across regions. What is the Treasury doing about that?  (Mr Macpherson) The Treasury's view is that ONS is funded to do the very important job that it is required to do. I do not accept that the ONS funding situation is affecting the policy of statistics.

  129. That is not what the Statistic Commission say, let me read it to you, "ONS was reluctant to examine the need for producing regional GDP deflators because of resource constraints". Yet that difference in price factors between the regions underpins a proper sense of all of these regional spending allocation systems. People will be discussing here, in this building tomorrow, at enormous length and in huge complexity the Area Cost Adjustment and the fact of the matter is we are not at first base in understanding that, are we?  (Mr Macpherson) I agree with you that the Area Cost Adjustment is important. I think that we can always seek to improve the quality of price data.  (Mr Scotter) Specifically on the Area Cost Adjustment, of course the figures for that are derived from the new earnings survey. Clearly there is a matter of debate now about whether that is the right source, but there is actually some very good data about employment costs, which the Area Cost Adjustment is mainly intended to reflect in its current form, which the Office of the Deputy Prime Minister is able to analyse and produce the figures of our costs. I am not sure that the comments made by the Statistics Commission are actually making it more difficult to do that sort of analysis in that particular context.

  130. Have you asked ONS to improve the work on regional GDP deflators—it is crucial to understanding the reality of regional GDP—have you asked them to do that?  (Mr Scotter) There is work which is being undertaken jointly by my Department and by DTI to look at that to see whether there is anything that one can derive from the existing data to look at deflators in regional GDP in order to have some price adjustment. I am not sure it goes to the extent that it is an absolutely perfect piece of work. It will give a very good—

  131. When is it due?  (Mr Scotter) We hope to publish the figures early in the New Year. I am sure that the ONS has been revising its methodology for calculating nominal GDP in the regions and our piece of work will need to follow on after that.

  132. What really concerns me was Mr Macpherson's earlier comment that regional GDP, which is such a crucial baseline if we are looking at growth, productivity and economic reforms between regions, it does not underpin all of the government's resource allocation formulas. That is what you said.  (Mr Macpherson) I did say that.

  133. Is that something that Treasury is doing something about?  (Mr Macpherson) I think with some service specific allocations it may well make sense to consider allocations that are more at a subregional level, so, for example, considering from a local authority perspective information around local authorities, population or deprivation or, indeed, the Area Cost Adjustment should feed into those allocations. I do not think it is self-evident that applying a regional factor on top of that for local allocations is the best way forward.  (Mr Parkinson) It is one of the factors for the RDA allocations, which is the main regional allocation.

  134. It is true that the RDA allocation is the main regional allocation but that is only true because there are virtually no other regional spending allocations.  (Mr Parkinson) Most of them are locally based.

  135. We do not have, with great respect, GDP, which is the record of the performance on productivity and growth, for local authority areas, do we?  (Mr Macpherson) No.

  136. Are we ever likely to have?  (Mr Macpherson) I do not know.  (Mr Scotter) May I say something about the GDP deflator, most public expenditure is on specific services like health, local education, law and order and the price factors that affect those across the country are different. The GDP deflator is a general measure of the measure of price changes in the whole economy and that may not actually be the right indicator because it takes account of all private sector plus all of the activity of industry and business and that may not be the right set of figures to use for each of these allocation formulas. It is the cost to the Health Service that matters for the allocation of Health Service expenditure if there is a price factor to be taken into account. Similarly it is the cost for local government that matters when one is looking at the Area Cost Adjustment. I am not entirely convinced the regional GDP deflator would improve those because you would want to look at the specific factors that are affecting those particular activities.

  137. Is the government trying to analyse expenditure by region that is not presently identified by region, the principle one being defence?  (Mr Macpherson) There is almost a question of principle here round defence and/or for that matter things like the Foreign and Commonwealth Office and the Department for International Development.

  138. Stick to defence. Defence is strong on employment, it has a strong connection with manufacturing and industry.  (Mr Macpherson) The key question is, who benefits from a particular service. I think defence is a public good of a sort which if you are focusing on the outputs of defence we all benefit equally. I do not think you can argue that somehow the benefits of defence are greater in the north east say compared to the south west. I think that is one reason for not allocating it across the regions.  (Mr Ritchie) The figures that we published in public expenditure analysis for something like 15 years now have always tried to measure public expenditure both by territory and region according to who benefits. According to who benefits that is one way—and there is a lot of theology about this—in which you can measure the impact of public expenditure on that criteria. Expenditure that provides services for collective consumption at a national level, things likes defence, expenditure in the Foreign Office, even things like scientific research are non-identifiable and it does not make sense to say that the north east gets more or less benefit from expenditure on defending the country than any other region.


  139. Forgive me a moment, there may be a general collective benefit from defence but there is also, presumably, a specific economic benefit, from regimental headquarters or a naval base, or whatever?  (Mr Ritchie) There are different ways in which you can measure the impact of public spending by territory, by region. The way we always presented numbers in here is on a "for basis", expenditure for the benefit of a region. You can do it differently, you could measure public expenditure on an "in basis" which is where economic activity associated with the spending took place. You will measure defence capital spending by where the actual project took place. You will measure allocated pay by location of the staff, the people. That is a different way of measuring spending. These are just two different ways of measuring the impact of public spending by area, it is not a case of one is right and one is wrong.

3   Correction by witness: "It was, in fact, a pilot study for a single year." Back

previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2002
Prepared 10 December 2002