Q&A: Inflation changes
- 10 January 2013
- From the section Business
The Office for National Statistics has decided not to make any fundamental change to the way it calculates the retail prices index (RPI).
It had been pondering whether or not to make the RPI rise more slowly, and bring it more in step with the slower moving consumer prices index (CPI).
Instead, in March 2013 it will launch an additional index called RPIJ, a new version of the RPI index but using the same methods as CPI for calculating average prices.
But it is not obvious that it will be used for anything.
Remind me, what is inflation exactly?
Inflation means that prices rise. There is nothing new about that. It has gone on for centuries. But it can be disastrously disruptive if allowed to get out of hand, or if it becomes a deliberate policy of a government.
The fundamental problem is that it devalues your money. For instance, inflation at 7% per year will halve the value of money in just 10 years.
Most governments and central banks have, for many years now, been obsessed with keeping some sort of lid on rising prices. But price stability is a difficult thing to achieve.
In the UK we do not even bother to try. The official inflation target is, in fact, for gentle inflation every year of (hopefully) just 2%.
How is it measured?
In the UK responsibility for measuring inflation lies with the Office for National Statistics (ONS), which is independent of the government. Each month it calculates and publishes a variety of inflation indexes.
The best known is the retail prices index (RPI), which has been published since 1947. Since 1996 another European-style index has been published, called the consumer prices index (CPI). This has become the government's preferred measure for almost all purposes, such as uprating state benefits, the state pension and public service pensions.
In essence, the ONS staff collect the prices of hundreds of goods and services each month, average them, and work out how that figure has changed in the past month. Few economic statistics are more important.
Why had the ONS been considering a change?
It is worried that the widening gap between RPI and CPI is undermining the credibility of these indexes, especially as most of the gap is due to the different ways of calculating the average level of prices in each index.
It is this so-called "formula" effect that the ONS was thinking of eliminating, in part or in whole.
So there is more than one way to calculate an average? Tell me more.
OK, here goes. If you want to find the average of a set of numbers, most people add up all the numbers, and then divide the total by the number of numbers they started with. That is the simple method we are all taught at school. It is called the arithmetic mean and is the method used in calculating the RPI.
There are other ways of doing it though. The CPI uses one of them. In this, you multiply all your numbers together, and then take the nth root of that total. That too produces an average, known as the geometric mean.
Thanks for the statistics lesson. But so what?
The point here is that with exactly the same set of numbers, the geometric mean will always rise more slowly than the arithmetic one. So by definition, the CPI will usually rise more slowly than the RPI. And that is why the topic is controversial. The ONS has now dodged this potential controversy by keeping the old RPI, but also launching new version - to be called RPIJ - that use the CPI's formula for working out average prices.
Why were some people suspicious?
The government has already replaced RPI with CPI for the uprating of public sector pensions. So those pension scheme members are worse off than they would have been with no change.
And this has had a knock-on effect on some private sector schemes too which copy the government's lead. Altering the calculation of the RPI so that it rose as slowly as the CPI would have saved pension schemes and the government money. But it would also have been to the disadvantage of millions of people if their incomes rose more slowly as a result. So some people smelt a rat.
I thought the difference between RPI and CPI was all to do with housing costs?
The fact that the various costs of buying and owning a home are deliberately excluded from the CPI does contribute to the difference with the RPI. But in fact it was always wrong to say that this was the main reason for CPI rising more slowly than RPI.
The formula effect has always been the single biggest reason for this, and recently has become far and away the dominant reason.
But reflecting the cost of home ownership, on which people spend lots of money, is very important for a realistic inflation index. So the ONS has already decided to launch a new measure of inflation in March 2013.
This will be the CPI index, but will include a proxy for housing costs. It will be called CPIH and may, in due course, become the government's new main measure of inflation. Just to confuse things even more.