The Political Potential of Price Indices

The significance of inflation for the Coalition’s economic policies was made clear, from the opening paragraphs of the Executive Summary onwards, in the March 2011 Economic and Fiscal Outlook produced by the Office of Budgetary Responsibility (OBR):

1.1 The key economic developments since our November 2010 Outlook have been an unexpected fall in UK GDP in the final quarter of 2010, a rise in world oil prices, and higher-than-expected UK inflation. The labour market has performed much as expected, with unemployment rising after registering significant falls in the middle of last year
1.2 These data have on average prompted external forecasters to reduce their estimates of economic growth in 2010 and 2011. The average external forecasts for CPI and RPI inflation have risen significantly, again reflecting recent data.

RPI AND CPI TECHNICALITIES

Although they are among the most familiar of government statistics, the generation of the Retail Price Index (RPI) and the Consumer Price Index (CPI) is a complex business. The Office for National Statistics (ONS) CPI Technical Manual explains the derivation of the RPI and CPI figures and the differences between them. To understand these more fully, it is helpful to look at an Information Note produced by the ONS in August 2010, Differences between the RPI and CPI Measures of Inflation. This shows (extracts below) that, although “both track the changing cost of a fixed basket of goods and services over time and both are produced by combining together around 180,000 individual prices for over 650 representative items”, there are three significant methodological differences between the RPI and CPI: Coverage, Population Base and Index Construction.

COVERAGE

POPULATION BASE

INDEX CONSTRUCTION

As the chart below, also taken from the Information Note, shows, index construction is the most significant of the various factors which lead to changes in CPI being lower than in RPI over the same period. It is often referred to as the “formula effect” because of the difference between the geometric mean (GM) and the arithmetic mean (AM) of the same set of data.

The next most significant factor is housing.  There is a comprehensive explanation of the problems in taking account of this in CPI provided in Wikipedia.

IMPORTANCE OF THE CPI

In December 2003 the CPI, which is comparable internationally, became the basis of the inflation target of 2% set by the government for the Bank of England Monetary Policy Committee (MPC). The RPI continued to be used for other significant purposes – for example, pension increases and uprating of taxation allowances. However, the 2010 Coalition budget announced that the CPI will be used for the indexation of benefits, tax credits and public service pensions from April 2011. Later this was extended to be an option for private pension uprating. In the 2011 budget it was announced that most taxation allowances and thresholds would also be uplifted in future by reference to the CPI.

Not surprisingly given the beneficial impact of the formula effect on public expenditure, the OBR, in its March 2011 Outlook, paid close attention to the likely difference between RPI and CPI in the next few years (Box 3.5 page 70), and provided this chart:

Clearly, the impact of the formula effect increased during 2010. The cause, (clothing and footwear) had been the subject of a further ONS Information Note in January 2011, CPI and RPI: increased impact of the formula effect in 2010.

The OBR asserts that the CPI “is better-suited to accounting for the effect of substitution between goods and services when relative prices change”, echoing the ONS’s statement that “An advantageous property of the geometric mean is that it can better reflect changes in consumer spending patterns relative to changes in the price of goods and services.” However, the Royal Statistical Society (RSS) does not agree. In a recent letter to the Chair of the UK Statistics Authority (UKSA) (the ONS being its executive office) they state:

The UKSA's Monitoring Report concluded that the uses of consumer price indices both as macroeconomic measures of inflation and as "compensation indices" imposed different requirements on the indices. We believe that the need to address this and other issues is now pressing, and are keen to offer the Society's help in doing so. We are concerned, as I am sure you are, about the further damage that will be done to consumer confidence in official statistics if it is perceived that uprating to pensions and other benefits is being governed by an index perceived by many as inappropriate and unfair without a more relevant index being offered by statisticians.
There are good arguments for the CPI as a macroeconomic indicator (particularly once some indicator of owner occupier costs has been included) but, as you know, we do not feel that it currently serves the purpose of being a sufficiently good measure of price inflation as experienced by households to be used in uprating pensions and benefits or for use in wage negotiations …

Lord Hutton recently proposed a new basis for public sector pensions - Career Average Revalued Earnings (CARE) pension schemes, (good outline on the BBC website). A key feature during employment when the pension is being accrued, is the annual uprating, which Hutton thinks should be in line with earnings. Hutton proposed:

Recommendation 8: Pension benefits should be uprated in line with average earnings during the accrual phase for active scheme members. Post-retirement, pensions in payment should be indexed in line with prices to maintain their purchasing power and adequacy during retirement.

But a problem arises with people who leave public service before retirement age:

Ex.18 Regarding the indexation of deferred members’ benefits, there is a trade-off to be made. If the indexation measure were the same as for active members this would favour mobility. If it were lower, for example, if active members’ benefits were indexed by earnings and deferred members’ benefits by prices, this would favour retention. The Government should decide on whether pre-retirement indexation for deferred members is on an earnings based measure or prices based measure, as this decision will need to be based on the explicit objectives that government has about recruitment and retention versus mobility.

THE POLITICS OF RPI AND CPI

Could the gap between the two inflation measures, and the Coalition’s substitution of the CPI for the RPI, become a political issue? This very much depends on the progress of the UK economy in the next few years. Two broad scenarios can help illustrate the situation at the time of the next election (2015 if the Coalition holds).

In the first, Ed Balls’ and Paul Krugman’s predictions are borne out  (for the former's see an earlier post, for the latter's, see his Op-Ed, The Austerity Delusion, in the New York Times on 25 March):

And then there's the British experience. Like America, Britain is still perceived as solvent by financial markets, giving it room to pursue a strategy of jobs first, deficits later. But the government of Prime Minister David Cameron chose instead to move to immediate, unforced austerity, in the belief that private spending would more than make up for the government's pullback. As I like to put it, the Cameron plan was based on belief that the confidence fairy would make everything all right.
But she hasn't: British growth has stalled, and the government has marked up its deficit projections as a result.

If Balls and Krugman are right, whenever the election comes Labour can campaign on the basis of “we told you so” and “things can only get better”. There will no need for them to get over-involved in specific issues and Labour should easily win a working majority.

In the second scenario, the Coalition’s economic policies are to some extent successful, but the pain incurred is not so easily forgiven. The Coalition will try to hang the deficit they inherited around Labour’s neck. In response, Labour may well want to position themselves on the winning side of voter-friendly issues, taking a line clearly different from the Coalition’s while appearing responsible on public expenditure. The 2010 Budget changes involving CPI affected only a minority of electors, albeit a substantial one, pensioners and benefit recipients. The latest changes, extending CPI to allowances, makes the basis of inflation indexing something which affects most taxpayers and voters.  The general public's boredom threshold for a fairly dry mathematical subject can easily be moved up a notch when they realise it concerns money in paypackets.

On that basis, it is interesting to read Early Day Motion (EDM 1629) put to the House of Commons on 17 March under Ed Miliband’s name and sponsored by the shadow treasury team:

PENSIONS INCREASE (REVIEW) ORDER 2011
That this House considers that the Pensions Increase (Review) Order 2011 (S.I., 2011, No. 827) which was laid before this House on 17 March 2011 should be withdrawn because the Order requires the uprating of public sector pensions using the consumer price index, replacing the retail price index which the Government has indicated will be a permanent change; notes that the Government has refused to heed arguments that a temporary three year change to the index used would represent a fair contribution from benefits and pension recipients, at a time when wage growth generally is suppressed, to reducing the deficit whilst not unfairly impacting on their incomes over the longer term; but regrets that the Government has instead indicated that the change is permanent, leaving public sector workers and the poorest in society disadvantaged permanently, year after year, even once the deficit is gone and earnings growth has returned.

It would be naïve to think that before the 2010 election the RPI CPI change had never been an option for government. The Financial Times Westminster blog on 21 June 2010 posted the following:

Darling predicts coalition will link benefits to CPI not RPI

Alistair Darling [the former chancellor] has just briefed the lobby hacks on what to expect from tomorrow’s Budget. Top of his predictions was a change in index-linking for benefits from RPI (and the less well-known Rossi index) to CPI. The obvious benefit … is that the CPI measurement of inflation is usually lower. As a result the government could save about £1bn a year
...
Darling described the shift from RPI to CPI as one of several ideas lying around the Treasury which – having worked there until 6 weeks ago – he was obviously aware of. Some were more “fit for human consumption” than others, he joked.

The Coalition is not slow to claim that, however unpopular its measures may seem, a Labour government would have had to do the same. In this particular case, it looks as though Labour might have had more sense.  On the other hand, they may have got very close, which could explain, as well as the need to seem responsible on expenditure,  why the wording of EdM's EDM seems rather hedged about.