Retail Prices Index: Food Component (Hansard, 3 June 1991)

§ The Parliamentary Under- Secretary of State, Department of Social Security (Lord Henley)
My Lords, the food component of the retail prices index is calculated as the average change in price for the full range of available foodstuffs. It is based on 45,000 prices obtained from retailers throughout the country and is therefore soundly based and robust. It is not possible to state its accuracy in conventional statistical terms.

§ Lord Molloy
My Lords, I thank the noble Lord for that Answer. Can he give a little more detail on what types of food are taken into consideration and in how many towns across the country? Is it not fair to say that these types must vary from place to place? Is it 430 not the case that, regardless of whether one is dealing with poor areas or better off areas, the cost of food is going up and up?

§ Lord Henley
Yes, my Lords, the cost of food is going up. In the most recent figures for April, the food component went up by some 6 per cent., whereas the all-items index went up by some 6.4 per cent. The survey looks at food prices in some 180 different areas throughout the country and at some 175 different components. As I said in my original Answer, 45,000 different prices are looked at.

Lord Bruce of Donington
My Lords, is the noble Lord aware that this whole question was under review by the Public Accounts Committee in another place? Its report revealed, at paragraphs 20 to 29, that the whole system of taking prices was a complete farce and that in point of fact 80 per cent. of the prices were taken on a part-time basis by employees of the Department of Employment, that the samples taken were minute and that the results were not monitored in any way even by the RPI central office? Will he give some indication as to whether the Government will take steps to ensure that a reliable sample is taken over a wide area in order that the existing position can be changed? Everyone knows that the price of food is going up not by 6 per cent. per annum but by nearer 15 per cent.

§ Lord Henley
My Lords, I do not accept what the noble Lord is saying. I said that some 45,000 different items are looked at. That strikes me as quite a big sample. Obviously some prices rise faster than others. I do not know what specific foodstuffs the noble Lord himself buys in the shops. He will know, for example, that tea, cake, fish, sugar and preserves have all risen in price by more than the all-items retail prices index or even the food component, whereas coffee, vegetables, butter, beef and lamb have gone up by considerably less than the all-items retail prices index or the food component.

§ Lord Stallard
My Lords, will the noble Lord give me the name and address of his grocer, because where I go to shop all the prices have gone up irrespective of the stated rate of inflation? I should like the names and addresses of the places mentioned. Is he satisfied that the weighting given to the price of foodstuffs in the retail prices index adequately reflects the hardships being faced by pensioners and others on low and fixed incomes? Can he do something about that?

§ Lord Henley
My Lords, I do not think it would serve any purpose to tell the noble Lord where I buy my own food. The retail prices index covers a great many areas and a great many prices. As I said earlier in answer to the noble Lord, Lord Bruce, some prices rise faster than others. The weighting of food in the retail prices index is some 15 per cent. Twenty years ago or so the weighting was 25 per cent. As a result of surveys conducted every year by the family expenditure survey, the weighting is adjusted to reflect the average expenditure on food by the average family.

§ Lord Gibson-Watt
My Lords, does my noble friend agree that Welsh lamb is a particularly good buy at the present time?

§ Lord Henley
My Lords, I have the figure somewhere for the increase in the price of lamb. I understand that in April home killed lamb came down by 3 per cent.

§ Baroness Blackstone
My Lords, I accept that where the Minister buys his food, including Welsh lamb, is immaterial. However, is he aware that there has been some controversy over the best way to measure inflation? The controversy relates to questions of statistical validity as well as of accuracy. Is he further aware that the Governor of the Bank of England has argued recently that it would be helpful to make more use of a measure which reflects what is happening now? In the light of that, will he confirm that, by using the underlying rate of inflation or those measures that reflect what is happening now, the rate of inflation rose last month?

§ Lord Henley
No, my Lords, we are talking purely about the food component in the retail prices index. The year on year figure for April was 0.6 per cent. I admit that in the previous year it was 5.7 per cent. The underlying trend for the food component is still down. It came down from 6.9 per cent. last November, to 6.6 per cent. in December and to 6 per cent. this month.

§ Lord Jenkins of Hillhead
My Lords, in his original Answer the noble Lord used what struck me as an interesting and resonant phrase. He said that the statistic was "soundly based and robust". I have since been contemplating that. Can he tell us exactly what "robust" means in relation to a statistic?

§ Lord Henley
My Lords, I was trying to get over the point that no index will ever precisely reflect the spending of every individual. The noble Lord may go to a different grocer from the noble Lord, Lord Bruce. That grocer might charge different prices. The noble Lord might also buy more expensive food or more expensive wine, the prices of which might again increase faster than the all-items index. For that reason the index we have could be described as robust.

§ Lord Williams of Elvel
My Lords, the Minister has been kind enough to say that the foodstuffs index has no statistical accuracy. While accepting the fact that the Family Expenditure Survey gives a proper indication of weighting, will he not admit, as the testing of prices of foodstuffs in the shops as a result of that survey and of the CSO techniques has no statistical accuracy, that that is a false basis upon which to incorporate anything into the retail prices index?

§ Lord Henley
My Lords, I believe that the noble Lord misunderstood what I said. I did not say that it had no statistical accuracy; I said that it was not possible to state its accuracy in conventional statistical terms.

§ Lord Williams of Elvel
My Lords, what is the difference between the two statements?

§ Lord Henley
My Lords, I should have thought that that was perfectly clear.

Lord Bruce of Donington
My Lords, will the Minister correct himself? He said that the price increase over the past month was 0.6 per cent. whereas later, according to his own figures, he said that it was 6 per cent. Further, will he answer the question I put to him: are the Government acting on the recommendations of the Public Accounts Committee in another place?

§ Lord Henley
My Lords, I am delighted to correct myself; of course, it should have been 6 per cent. rather than 0.6 per cent. Moreover, I can tell the noble Lord that we have made improvements to price collection during the past year. There has been an increase of about 20 per cent. in the number of locally collected prices. There has also been greater training, supervision and quality control as regards those collecting prices.

§ Lord Molloy
My Lords, will the Minister address himself to the vital question? The Public Accounts Committee is highly dissatisfied with, and highly critical of, the present system. The RPI is of interest to many people, organisations, industries and trades. When the committee makes such a statement, would it not at least be reasonable for the Government to consider the question of whether there ought to be a thorough examination as to how the RPI works and how it is collated?

§ Lord Henley
My Lords, I believe I answered that point when responding to the noble Lord, Lord Bruce of Donington. I said that improvements had been made to price collection. Moreover, as I said earlier, the index will never reflect every individual's personal experience. It is a robust figure, but one which will represent the average for all people.