This section looks at the main inflation measures that are quoted in the UK, why they exist and what the differences between them are. The two key measures are:

  • RPI – Retail Prices Index
  • CPI – Consumer Prices Index

RPI is the original UK index with history dating back to World War I. CPI is an EU invention which was not intended to measure cost of living and was created for a completely different purpose. CPI’s method of calculation ensures it is usually around 1 per cent lower than RPI and significantly underestimates true inflation in the UK.

The history of price indices in the UK and the birth of RPI

The first cost of living index in the UK was published in 1914 and covered food alone. It was quickly expanded in 1916 to include clothing, fuel and some other items. Its main function was to help ordinary workers argue for fair cost of living rises – especially given the effects of the price rises that ensued due to World War I. The coverage of the index was very limited. It was purely meant to include working class people’s expenditure (though, bizarrely, it excluded things like beer) and was largely based on a 1904 survey of urban working class households’ expenditure.

This index and its antiquated construction continued until 1947, when it was relaunched as an “Index of Retail Prices” on the basis of a new survey. It was again radically overhauled in 1956 into the first Retail Prices Index (RPI). This index expanded the coverage to include many more products and services and, more importantly, most people’s expenditure. However, it excluded the top wage earners and certain low-earning households such as state pensioners – and still does to this day.

Since then, the RPI has continually evolved with changes to the method and coverage. Its basket of items that are surveyed every month is updated annually on the basis of household expenditure surveys.

What is CPI and why was it created

The Consumer Prices Index (CPI), on the other hand, was first collected in this country in 1996. When originally created, we used to know it more correctly by its original and more descriptive name: the Harmonised Index of Consumer Prices (HICP). It was so-called because it was developed by the EU with the sole purpose of allowing European comparisons to be made for the requirements of the Maastricht Treaty. The terms of this treaty determined whether you had a government that was fiscally responsible enough to be able to join the Euro.

Included among the six key criteria of the Treaty was that a country’s inflation rate had to be low and no more than 1.5 per cent higher than the three states with the lowest inflation rate. This caused a problem for the EU, because: (i) everybody had been using different inflation rate calculations; and (ii) inflation rates historically were volatile and often very high in some countries. They therefore came up with a new statistic that they asked members to calculate alongside their existing measures. It was recognised that it was not a cost of living index, and that it excluded some things that nations normally monitored, but at least it would allow comparisons.

Why CPI is now the preferred UK measure of inflation

In most European countries, HICP is not used as the preferred inflation measure by governments. For example, Germany continues to use its own bespoke CPI measure as its main indicator and only uses HICP for international comparisons.
So how has the UK government’s “preferred measure of inflation” become HICP (or CPI as we now call it), instead of the UK’s own measure (RPI)?

The answer could be quite simple. According to a report by the government’s own Office for Budget Responsibility, CPI is usually around 1 per cent lower than RPI. Therefore, merely by utilising the European CPI rather than the old UK RPI measure, inflation appears significantly lower. In March 2013 the UK Statistics Authority and ONS went as far as declassifying RPI as a “national statistic” apparently justified on the grounds that its method of calculation is not internationally acceptable.


The main take-out of this is that the way inflation is calculated affects the result. Inflation is a key statistic. Unfortunately because of this, it may be subject to political pressure to ensure methods are used that help it look low.

The differences between CPI and RPI

The table below summarises all the key differences between these two measures. They vary not only in the items included in the indices and their weightings but more importantly the method of calculation. CPI is almost always going to be lower than RPI because it uses a method of calculation called a geometric mean. It also leaves out a number of housing-related costs that often have a significant impact on the cost of living and the RPI index – see Coverage issues.


What are RPIJ and CPIH?

In response to criticisms of the large difference between CPI and RPI, in 2013 the ONS created some additional measures.

  • CPIH is similar to the CPI index but it includes household rents. Importantly it does not include any measure of house prices nor the cost of mortgages. It therefore typically creates an estimate of inflation that is lower than CPI. The H stands for Housing. See blog post on making CPIH the preferred inflation measure.
  • RPIJ is similar to RPI but instead of using ordinary mean scores to combine the prices it uses the geometric means which CPI uses. This ensures it will always produce a lower apparent level of inflation. The J stands for Jevons who was a statistician who promoted the use of geometric means.

More technical details on inflation measures

Measures such as CPI include over 600 individual items which aim to represent the typical household budget. The items covered are updated annually. Each month ONS either checks the prices of them directly (in the case of nationally listed prices e.g. newspapers, utilities, train fares, etc.), or sends out its team of researchers to look for them in around 150 locations spread around the UK. The ONS publish detailed information on how they do this and the items checked.