The producer price indices include two main measures, covering input prices (the cost of raw materials and fuel used by manufacturing) and output prices (the price of goods as they leave the factory).

The RPI and the CPI are both designed to measure the prices paid by consumers at the final point of sale. Between them, therefore, the producer price and retail and consumer price indices cover the whole production process for goods.

The other important measure of consumer prices is the consumer expenditure deflator (see Consumer expenditure deflator) which is obtained indirectly by dividing consumer spending in current prices by spending in constant prices.

The national accounts data include a whole range of deflators in addition to the consumer spending deflator, including investment deflators and, of course, the GDP deflator which is the average of all prices in the economy.

Table 8.2: Annual inflation rates

One difficulty with all price measures is how to account for quality improvements in the goods and services which are covered in the index.

This point can readily be seen, for instance, by looking at motor cars and at electronics products in general and personal computers in particular.

Technological developments provide a constant improvement in the quality of the product being provided even though the price might remain static or even fall.

The statisticians do try to take account of this factor as far as they are able but it is still likely that consumer price indices, in general, overstate the rate of inflation. A further compositional problem arises from the introduction of new products.

Estimates of quality and new product bias exist for the UK and the US.

A Bank of England Working Paper (Cunningham A W F: Measurement bias in price indices: an application to the UK’s RPI, 1996) suggested an overstatement of inflation of 0.20% to 0.45% for the UK while the Final Report to the Senate Finance Committee from the Advisory Commission to Study the Consumer Price Index (1996) suggested a figure of 0.6% for the USA.

Price indices also suffer substitution effects where consumers switch their expenditure to cheaper products or to cheaper sources.

The effect of substitution bias can be minimised by regularly examining the weights attached to various products and by constantly ensuring that the survey sources match, as accurately as possible, spending behaviour.

From February 2009, for instance, new items were included in the basket to improve the coverage of goods.

Many of the changes reflected changing technology. Thus, Freeview boxes, M4 players and Blu-ray discs were included for the first time with MP3 players removed from the index.

But the changes also included the more mundane with parmesan cheese taking the place of imported cheddar cheese, fresh single cream being replaced by double cream and watch repair, clean and service dropped in favour of watch battery replacement.

Sometimes, items become less important to the consumer but remain a significant part of expenditure.

In 1914, when the Cost of Living Index (which became the RPI) was introduced, food accounted for 60% of expenditure. Now it accounts for around a seventh.

Such changes in expenditure patterns are allowed for by adjusting the weights attached to each product. The weights are based on the Family Expenditure Survey (FES) which takes information from around 7,000 households.

In working out these weights, two groups of people are excluded - those at the very top in terms of income and pensioners who depend mainly on state benefits. There are specific indices which seek to represent 'pensioner' expenditure.

Information on prices is obtained for more than 650 separate goods and services with price movements measured in about 150 areas of the country.

In total, 120,000 price quotations contribute to the compilation of each month’s index. The survey is normally conducted on the second or third Tuesday of the month.

The main components of the RPI and their relative importance are shown in the table below.

The RPI is used not only to track price changes but has traditionally been used as the basis for pay bargaining.

Since the 1970s when relatively high inflation was the norm, the RPI, moreover, has been used increasingly as a means of inflation adjusting various forms of tax revenue and some types of income.

Many savings products, including some National Savings products, as well as some gilt-edged securities are index-linked.

There are a number of variations to the general RPI measure. RPIX which is obtained by excluding the effects of mortgage interest payments from the basic RPI is the best known of these.

RPIX was for a long period the target measure of inflation used by the Government in the conduct of monetary policy, before being superseded by the consumer prices index. RPIX is often referred to as the ‘underlying’ rate while the All Items rate is known as the ‘headline’ rate.

The advantage in using RPIX rather than RPI as the inflation target was that the latter, since it includes mortgage interest payments, initially rose when interest rates were increased to counter inflationary pressure!

When interest rates were reduced, in response to lower inflationary pressures, the immediate effect was to bring the 'headline' rate down much faster than the 'underlying' rate, again because of the change in mortgage interest payments.

RPIX, therefore, provided a clearer indicator of how effective interest rate changes were.

RPIY goes one step further in the attempt to provide an indicator of underlying inflation by removing the influence of local and central government-induced price changes as well as the effects of mortgage interest payments.

The index, therefore, excludes council tax, VAT, excise duties, vehicle excise duty, insurance tax and airport duty in addition to mortgage interest payments.

The average index for the monetary union area calculated from these individual indices is the measure which is used by the European Central Bank in pursuing its target for price stability across the euro area. The HICP for the UK, renamed the Consumer Prices Index (CPI), is now the designated inflation target for the Bank of England (see Macroeconomic Policy 4.2 Monetary Policy - The inflation target).

UK inflation as measured by the CPI has been on a par with both the EU average and the average of the Monetary Union area over the past few years.

In general, the CPI is constructed in a similar way to the Retail Prices Index although there are a number of differences:

• In calculating the CPI, prices are aggregated using the geometric mean whereas the arithmetic mean is used in calculating the RPI
• The CPI excludes owner-occupier housing costs whereas the RPI includes them
• The CPI includes air fares, university accommodation fees, foreign students’ university tuition fees and payments by residents of nursing and retirement homes
• New car prices are calculated differently
• The weights used in calculating the CPI are based on a wider definition of expenditure and include foreign visitors and residents of institutional households as well as private households. The RPI takes into account private households only.

Since 1996 inflation as measured by the CPI in the UK has averaged one percentage point lower than RPIX inflation.

The two most important reasons for this gap between the two measures are:

• The exclusion of housing from the CPI
• The use of a geometric mean in its calculation. HSBC economists have calculated, for instance, that the use of the geometric mean has accounted for 0.4 – 0.5 percentage points, up to one half of the gap since 1996.

These two factors will ensure that future inflation as measured by CPI will be lower than the RPIX inflation rate, although the size of the gap may vary over time.

The use of the geometric mean will continue to bias CPI downwards. In addition, over the medium term, house price inflation should be approximately the same as nominal GDP growth, meaning that housing costs should rise by around 2.5 percentage points (the ‘real’ element of GDP growth) more than general inflation. The inclusion of housing in RPIX, therefore, would tend to bias it up relative to the CPI which excludes housing.

The implications for policy

When the CPI replaced RPIX as the monetary policy target, the structural gap between the two measures led to a great deal of debate about its effect on the operation of monetary policy.

If the gap were to remain at the one percentage point experienced on average since 1996, then it might be expected that the 2.5% RPIX target would be equivalent to a 1.5% CPI target. But the Government announced in December 2003 that the CPI target would be 2% to correspond to the ECB’s target.

As a result, this led to much discussion about whether the new CPI target represented a loosening of policy, an argument that the Bank of England has been keen to refute.

Conclusions

Why are there two indices to measure the prices which consumers face? Aren’t they just doing the same thing? The answer is yes they are. Both are obtained directly via survey data, although differences in content and construction give different rates of inflation over any given period of time.

The reason for two indices is simply that the relatively new CPI (see Consumer Prices Index) is based on the same methodology and content as the European harmonised index of consumer prices and is therefore directly comparable with the other European countries.

It has also become the official target for the Monetary Policy Committee at the Bank of England. The retail prices index which dates back to 1947 isn’t comparable across countries but is still the one which is used for practical purposes such as inflation-proofing government benefits and being used in wage negotiation.

Consumer spending is measured in two ways (see Growth 7.4 Expenditure Data). The first measure is what consumers actually spend on goods and services (sometimes referred to as spending ‘in nominal terms’ or ‘in value terms’ or ‘at current prices’).

It depends both on the number of goods and services which consumers buy and the price of those goods and services. The second is an artificial series which is designed to show how consumer spending would behave if prices were unchanged over time.

In other words, its aim is to measure the change in the number of goods and services bought. In the jargon of economics this latter series measures ‘real’ expenditure or expenditure ‘in volume terms’ or ‘at constant prices’.

It is built up from spending on individual goods and services. Where it is appropriate, this spending is deflated by suitable price indices such as the retail price index.

Where this is not possible, and this is true for about a third of spending, volume estimates have to be made. These individual ‘real’ series are added together to provide an aggregate ‘real’ measure of spending.

The consumer price index or consumer price deflator is obtained by dividing the ‘value’ series by the ‘volume’ series. By its nature, therefore, it is a price index which measures the whole range of consumer spending on goods and services.

Its indirect construction means that it can also be thought of as an average of the individual price index data used in the construction of the individual ‘real’ spending series.

The advantage of the consumer expenditure deflator over the RPI and the CPI, therefore, is that it avoids the need to select a basket of goods to be included in the index.

The input price index for materials and fuel purchased by manufacturing industry measures just that.

Being at the beginning of the production chain, the input price index provides an early warning of inflation pressures.

The output price index for home sales of manufactured products is sometimes also referred to as the index of factory gate prices. As the name suggests, it measures the prices at which manufacturers sell their goods.

The output index is often used in conjunction with the input series to illustrate how manufacturers’ profitability is changing.

For instance, the surges in oil prices through the 2000s meant that the input price index rose sharply on a number of occasions.

At the same time competition in the United Kingdom meant that it was sometimes difficult for manufacturers to pass those increased costs on.