The retail price index (RPI) represents the shopping basket of goods and services consumed by the average British household. Tomorrow's publication of the July RPI marks the climax of a monthly ritual that has been taking place since 1947 - the gathering of well over 100,000 prices to calculate the rate of inflation.
The CSO is expected to announce that the RPI rose 4.1 per cent in the year to July, up from 3.9 per cent a month earlier. By 20th century standards this is pretty dull: inflation was 27 per cent in August 1975 and minus 28 per cent in January 1922.
Inflation rates have varied widely over the past 70 years, but so has the shopping basket that the RPI reflects. Its make-up is based on the annual Family Expenditure Survey, in which 7,000 households throughout the country keep a record of their spending over a given fortnight, as well as describing any large purchases made over a longer period.
The results are used to determine the importance of particular items in the RPI. Items are 'weighted' according to the proportion of the average family budget for which they account. A 50 per cent rise in mortgage interest payments thus affects the index more than a 50 per cent rise in the price of a packet of tea.
The importance of particular items changes over time. Food accounted for 35 per cent of family spending in 1956, but 15.2 per cent by 1992. In the same period motoring has increased from 3 per cent to 14.3 per cent of the average family budget.
Items join and drop out of the index each year as spending on them rises or falls. Lard, sanitary towels, net curtain material and tinned rice pudding fell out this year, while multi-vitamins and child-minding charges were included for the first time.
In recent years condoms, compact discs and satellite dishes have joined the index, while cine film and black and white televisions have dropped out. Foreign holidays may soon join the index.
Two groups of people are excluded when weights are calculated: the richest 4 per cent of households, and old people dependent on benefits. (The latter have their own 'pensioner price index'.) Spending by these groups can differ widely from the average. For example, spending on audio-visual equipment fell 33 per cent among high-earning households between 1985 and 1989, but rose nearly 25 per cent for the average household.
The logistics of calculating the RPI are quite impressive. On a Tuesday in the middle of each month some 400 price collectors - largely Department of Employment staff - gather well over 100,000 prices for different types of goods and services in more than 180 parts of the country.
Most information is collected from personal visits to shops, although about 100 prices - for items like gas, electricity, newspapers and rail fares - are collected centrally.
Big chain stores that charge the same prices at each of their branches send their price lists direct to the CSO.
Price collectors return to the same shops each month and check the prices of the same goods, so the RPI compares like with like.
Prices for the same product can vary widely in the same month between different shops. For example, in June a 500g tub of soft margarine varied from 36p to 85p. An average is used in the index.
The inflation rate for the RPI as a whole masks widely differing rates for its various components. The most obvious difference is between prices for goods and services.
Services price inflation is about twice that for goods prices, and is slowing less quickly. Services are less exposed to international competition and the discipline of the exchange rate mechanism.
One such service - included in the index - is the price of sending a single red rose to Watford. As an official put it: 'We are not without romance in the CSO.'
(Graph omitted)Reuse content