| Previous Section | Index | Home Page |
The Economic Secretary to the Treasury (Miss Melanie Johnson): I join the hon. Member for Kingston and Surbiton (Mr. Davey) in congratulating you on your appointment, Madam Deputy Speaker. A wise choice has been made, and I am delighted that you have become a Deputy Speaker. I also thank the hon. Gentleman for raising this important and interesting issue, and congratulate him on securing the debate.
There is a public debate about whether price indices are fully taking account of quality improvements in new technology, particularly computers. Different methods of allowing for quality improvements can have a significant impact on measured economic growth. The Government, the Bank of England and the Office for National Statistics are actively involved in research on this complex statistical issue.
It is unfair for the hon. Gentleman to suggest that we have not been dealing with this measurement gap. All the bodies to which I have referred are significantly involved in dealing with this matter. That is a clear indication of the fact that we are well and truly engaged with the issue.
International comparisons of productivity have always been complicated by measurement difficulties, which introduce considerable margins of error. In recent years, these have been compounded by developments associated with the new economy. Information and communications technology has become an important driver of economic growth. Its importance has risen, not just because more ICT equipment is available, but because the quality of ICT has improved substantially.
Different countries' national statistical agencies have adopted different approaches to adjusting prices of ICT products and services for quality change. In particular, the method adopted in the US, known as hedonic price adjustment, has a significant effect on the measurement of output. Applying the same method to the United Kingdom would affect the size of the productivity gap.
Specifically, although both the UK and the US national accounts allow for changes in ICT quality, the hedonic price adjustment approach carried out in the US implies a much faster rate of decline in computer prices. That means that any given change in nominal sales or spending on computers will show up as a much larger real change in the US accounts than in the UK accounts. So there has been a growing perception that recently GDP--and hence productivity--growth in the UK may have been understated relative to that in the US. The hon. Gentleman has remarked as much.
However, quantifying such effects is less straightforward than it appears at first sight. Applying, for example, United States computer price indices in the United Kingdom accounts would increase measured gross domestic product growth by less than the effect on real investment and consumer spending on computers to the extent that they are imported rather than domestically produced. In other words, and more generally, the effects on measured GDP growth differentials of harmonising the measurement of ICT prices between countries would depend on the relative size of their ICT sectors as well as their spending on ICT.
None the less, it should be recognised that the US-UK productivity gap cannot be explained away, as the hon. Gentleman maintains, by statistical treatment. The gap is
too large and has persisted for too long to be a statistical artefact. It cannot be attributed only to methods of measurement.The hon. Gentleman seemed to be saying that the information in box 1.2 in "Productivity in the UK: the Evidence and the Government's Approach", a paper accompanying the pre-Budget report, had been hidden away. As I am sure he realises, as he has read the paper, much of what I have just said is a summary of that information. Nevertheless, it is worth making the points for the benefit of those who do not have direct access to the paper.
The hon. Gentleman was also being rather unfair by saying that we had not flagged up the issue. The issue is not only flagged up in the paper, but is the subject of an annexe to which he referred that addresses in considerable detail various technical issues.
Mr. Edward Davey: I do not wish to fall out with the Minister on this. I am merely trying to say that the issue is so important that the Chancellor should be talking about it, and that--although addressing it in a supplementary publication to the pre-Budget report is welcome--the Treasury should be addressing it more emphatically, to ensure that economic commentators and the wider public take notice of it.
Miss Johnson: It is important to recognise that debates in the House are followed closely by the Chancellor, who takes a keen interest in the comments of Opposition Members. Moreover, the fact that we have addressed the productivity issue in a paper accompanying the pre-Budget report demonstrates that we recognise its seriousness.
In the light of the hon. Gentleman's comments, I should emphasise that the first national statistics quality review, "Review of Short-Term Output Indicators"--the STOI review--found that the short-term economic indicators produced by the ONS were of good quality and fit for purpose. Various detailed studies were conducted as part of that review and formed the basis of that conclusion. One of the studies highlighted the importance of correctly measuring improvements in the quality, and hence the prices, of items in the information and technology sector, such as computers.
The quality review, which the hon. Gentleman mentioned, pointed out that there could be a significant impact on measured growth and investment in the United Kingdom simply by making different assumptions about the value to attribute to those quality improvements. That point was also made in the pre-Budget report publication that I mentioned.
The United Kingdom approach, contrary to what is sometimes asserted, does make quality adjustments for improved computer quality, but our approach differs from that of the United States. The United Kingdom uses two methods when considering the price of manufacturers' output.
One method is that, if an extra characteristic or option is added to the specification of a product, such as a computer, the producer is asked whether it is priced separately. If so, the overall price of the computer is adjusted downwards by half the cost of the new option.
The figure of one-half is used to allow for the fact that standardising an option allows manufacturers to exploit economies of scale.In many cases, however, a separate price for a new option or specification change will not exist. In such cases, the manufacturer is asked for his estimate of the cost of upgrading the specification, and that amount is fully taken into account in the quality adjustment of the price. That is used in a number of ways to upgrade a number of outputs.
A priority is for consistency of methods by which quality is taken into account in price indices in the UK. Thus, whether one is looking at producer price indices or export and import deflators, the user can be assured that the same or very similar methodologies are used, and any differences must be for sound methodological reasons.
Interest in this area is currently high, as policy makers look for evidence of the new economy and the associated sustained high productivity growth due to the input of increased information and communication technology--a phenomenon attributed in particular to the United States. Various research papers have taken United States producer price indices for information and communication technology and applied them to United Kingdom growth data.
The short-term indicators report came up with estimates of the maximum--I stress maximum--effect that use of alternative estimates for computer prices might have on output. The STOI study found that, had US price indices been applied to the output of the UK computer industry, the output of the production industries would have grown by a further 6 per cent. since 1995. I emphasise the fact that the figure is only illustrative, that it applies only to the index of production and that it is the upper limit. The hon. Gentleman referred to it as though it were a point on the scale, but it is at the far end of the scale.
Other effects could be offset in different ways. The report clearly states that the figure is illustrative. The fact that the topic is found in the report shows that we are engaging with the discussion, and indeed playing a central role in it. We are happy for there to be transparency and wider public awareness.
- this review hasn't investigated the complex issues related to computer prices, nor can it recommend using the US price indices within the UK Index of Production. The analyses given here simply demonstrate the potential impact of doing so.
I want to emphasise that the role of the review has been to identify areas for further development. By necessity, the analysis carried out during the review could not be a deep investigation covering all aspects of each area. In particular, on the impact of price indices on growth, there are some methodological issues that should be considered when assessing the studies.
It is misleading simply to take the output of the computer industry and look at how much the value added by that industry increases if prices are assumed to fall faster. Much of that output is bought by other companies in other industries, so if the value added by the computer industry rises when different price indices are used, the estimate of the value added contributed by the purchasers of the equipment will fall.
In other words, much of the effect of changing price indices is to redraw the picture of growth between sectors, with the computer industry growing faster than we first
thought and its customers growing more slowly. We do not yet know the effect on the overall growth of the economy.It is also very important to consider the situation regarding international trade in such products. A higher value assigned to computer products will increase the import bill for such goods as well. Of course, exports will also be revalued, but as the UK is a net importer of such products, the final outcome is likely to be a bigger boost to the measure of imports than to the measure of exports, which means that there could be a negative effect on overall growth to counter the positive effect highlighted by the commentators.
There is an important international context to this issue. One of the important uses of economic statistics is to make international comparisons, so it is important that methods are consistent across countries.
Earlier this year, the go-ahead was given for a European hedonic centre, which has been funded by Eurostat. In parallel, an OECD committee--to which the hon. Gentleman referred--is producing a handbook on computer quality adjustment, and a further OECD-sponsored group, which is investigating the international transferability of hedonic models, will produce useful conclusions for the European hedonic centre. The Office for National Statistics will be an active participant in that effort. The centre will begin work in January 2001, and its first year will be devoted to data collection and construction of various hedonic models. The models will then be tested, and we might see useful results in 2002. We and the ONS aim to be key participants in all the activities involved.
It is important to note the time scale, which makes it clear that there is much work to be done. It is also important to emphasise that the ONS is, as the hon. Gentleman readily granted, at the forefront of national accounts on the international scene. I feel that the hon. Gentleman's remarks about errors in national statistics were a little misplaced. The ONS leads Europe in terms of quality and timeliness, and presentation of an integrated set of national accounts. Much good work is going on, and to ensure that it continues we must make haste slowly--or, at least, with due care.
The hon. Gentleman suggested that a significant impact could have been made in a number of ways. He spoke of the loss of jobs. It is not true that interest rates would necessarily change: there is no evidence that rates rise according to different measurements on computers. Interest rates have been very low recently, and long-term interest rates are at their lowest for some 35 years.
The hon. Gentleman asked whether we were treating this as a priority. As I have said, we have produced formal papers, and formal things have been said in very public contexts. I assure the hon. Gentleman that the Treasury, the Government as a whole and the ONS take the issue very seriously, and the studies to which I referred are under way.
I have mentioned the need for methods that are consistent across countries. The United Kingdom has been active in the harmonisation of economic statistics. There are no international standards at present, and the hon. Gentleman suggested that a lack of uniformity undermined economic policy making. We are actively participating in the OECD committee's work--I mentioned the computer handbook earlier--and in that of the European Hedonic Centre.
The hon. Gentleman also suggested that inward investment might be affected by errors showing a smaller growth in gross domestic product than was the case, and made a comparison with the "golden halo" around the United States. I see no evidence of that. Inward investment is at record levels: we are second only to the United States in that regard.
The hon. Gentleman asked whether we were measuring inflation differently from the United States. Indeed we are, but that is not entirely true within Europe. For example, we now have a harmonised index of consumer prices across Europe, which represents our first attempt at an international standard for consumer price indices. It means that consumer inflation rates can now be compared across Europe. The ONS is actively participating in the work of the Eurostat taskforce, which developed the index.
The hon. Gentleman made comments about international comparability, some of which I have addressed. We think that comparability should be a priority, and will continue to urge our international colleagues to make progress--consistent, obviously, with thorough research within the necessary time scales. It is not currently clear how we should proceed.
I was not present when Sir John Kingman was here before I gave evidence to the Treasury sub-committee yesterday. As the hon. Gentleman said, he had left by the time I arrived, so we experienced different parts of the discussion. However, I have been given evidence that Sir John Kingman did not say that he had concerns in the evidence that he gave on behalf of the Statistics Commission. He said that the commission would investigate the concerns of the hon. Gentleman and others if it felt those concerns should be progressed in light of the hon. Gentleman's remarks.
The Office for National Statistics is giving priority to a programme of work to evaluate the different methods used to adjust for quality changes in the output of information technology industries. That project will also study the way in which computers have contributed to the growth in capital stock. The ONS will seek the advice of experts in the field, key users, its own methodologists, economists and national accountants, and will look closely at the approaches adopted in other countries. Wherever possible, the project will have as one of its aims international comparability, for the reasons that I have given. It will be essential that any adjustments made to current methods, and therefore to current estimates of growth and productivity in the United Kingdom, are based on thorough, soundly grounded research. The ONS plans to publish periodic progress reports, starting in Spring 2001 with an article in "Economic Trends".
To sum up, we have been actively involved in research and discussions on this topic. We are working with Eurostat and European counterparts on the issue, as is the Bank of England. The size of the effect suggested by the hon. Gentleman is speculation at this stage, and I hope that I have explained to the House how the issue is being appropriately addressed. At the moment, the degree of the hon. Gentleman's concerns is misplaced.
- Adjourned accordingly at six minutes past Two o'clock.
| Index | Home Page |