Although there has not been more information, there has been some shift of stance by the DWP.   Initially they were willing to acknowledge the CPI change was about scheme members getting less.  Thus we had the Pensions Minister quote "It [CPI] is a lower inflation rate, and it is important to be honest about that and to be clear about that".   The briefing to MP's from the DWP wrote "we need to balance protecting pensioner's incomes with the need to ensure that schemes remain viable".  Over the months, that stance has changed to pretending the liability savings are merely a consequence of choosing the most "appropriate" measure of inflation.  Thus we have this dissembling from the Work and Pensions Minister (the boss of the Pensions Minister):  “At cost of living are they [pensioners] losing? No. The CPI calculates the cost of living in a much more reasonable way”.

The view of professionals is that the change is "all about money".  (Slide25)

The reasons for the Ministerial change of stance may be:

a) The proposed change has not been as well received by employers as the DWP hoped.  Although it would mean a big reduction in liabilities, maybe £120 billion nationally, that isn't the primary concern for many CEOs since the cost saving is spread over a long future and the CEO expects to be in a different job by then.  The bigger immediate CEO concern is that final salary scheme pension liabilities appear on company accounts and are unstable because returns on the fund's investments are unstable.  That prevents company accounts giving the impression that the CEO has everything under control with a smooth progression for the company.  A change to CPI reduces the liabilities but does not do a lot to make them more stable.

b) The proposed change has not been as well received by trustees as the DWP hoped.  Although in theory using CPI for everything would be a simplification, it does not work out that way.  When new arrangements are introduced it is always likely that some scheme members will stay on the old arrangements.  (In the IBM case we see this particularly with the I-Plan).  So trustees have to implement both old and new variations.

Also trustees could be faced with having the power to adopt CPI but searching for a reason to adopt something so damaging to the scheme members.

c) The proposed change is not a good fit to legal opinion.  It comes up against the legal equivalent of the common sense view "Surely what I earned as future pension was well defined when I earned it, and cannot be changed later?"

d) The newspapers have been critical, e.g. front page banner "Pensions Cut Will Hammer Millions" in the Daily Express.

In short, there is now more reason to believe the worst scenarios will not happen, even though some damage to pensions received in future is inevitable (for deferreds in particular).

July 2010:

Newsletters 33, 35, and 38 have information on measures of inflation.

The Retail Prices Index is currently by far the most used in pensions calculations.  The government intends (from 2011) to make the Consumer Prices Index predominant, and the effect will be to lower some pension increases.  Steve Webb, the Pensions Minister, said on Money Box:  "It [CPI] is a lower inflation rate, and it is important to be honest about that and to be clear about that".

Neither RPI nor CPI is readily predictable, but there are technical reasons why the CPI will continue to be some 0.5% less than RPI.  Taken over a retirement lifetime this might aggregate to some 10%.   So not particularly quick or dramatic, but significant.   (A few percent cost difference makes a difference of millions of dollars for a scheme of IBM's size).  In many cases RPI increases are capped - Limited Prices Indexation.  When both RPI and CPI are higher than the cap it won't make any difference which is used.  If you expect high inflation then you will be expecting your pension to lose a lot of value, but not because of CPI replacing RPI.  Obviously, predicting inflation over pension timescales is difficult.  (In outline, the case for expecting high inflation goes as follows.  Inflation helps those with debts at the expense of those with savings, because the value of a fixed amount goes down.  Governments have taken on massive debts.  Hence governments will look kindly on inflation.  In outline, the case for expecting low inflation is that consumers unwilling/unable to pay higher prices will hold inflation down.)

You might reasonably ask "How can they do that - surely what I earned as future pension was well defined when I earned it, and cannot be changed later?"  That used to be the prevailing view.

Before the election, Steve Webb for the LibDems said  “We are clear that accrued rights should be honoured: a pension promise made should be a pension promise kept. Therefore we would not make any changes to pension rights that have already been built up. I have confirmed that I regard accrued indexation rights as protected.”  

Before the election, Philip Hammond said on behalf of the Conservatives "Indexation of pensions in payment is an established part of pensions legislation."  "We agree with the view that the right to indexation of pensions already accrued is part of the accrued pension rights and those rights will be protected."

The change of inflation measure is proposed both for public and private sectors.  These are different issues.  The cost saving for public sector schemes has direct effect on the national debt.  It is possible to reason that the magnitude of the national debt makes retrospective legislation (breaking promises) necessary.

The cost saving for private sector occupational pensions has direct effect on profits.  The national debt reason does not apply.  In fact, for the private sector, we do not know yet whether there will be retrospective legislation.   The Department of Work and Pensions says it is being considered.  Here is AMIPP's letter to Steve Webb on the subject.

IBMers will care about what will happen for our schemes.  The Regulator has suggested trustees should "give serious consideration to an interim communication, to assist members who may be faced with decisions on transfers or retirement planning, or may be concerned about press coverage".  The message our trust has given does not assist any member, and may have raised concerns.

If there is retrospective legislation (known in the pensions world as "an override of Section 67") then we can expect IBM to take advantage of it.  Although the legislation will not force increases to be lower, it will allow it.  With the trust funding level so low, the trustees would not have much of a case for keeping existing rules.

If there is no retrospective legislation:

- Lower increases associated with pension earned in the future will be allowed.  However, since our schemes are being closed to future accruals soon, this will not have much effect.

- The calculations for the Guaranteed Minimum Pension increase will change. (Since that was never directly linked to the RPI.)

- Other increases will depend on the interpretation of the deeds, and where that is not clear on interpreting what scheme members were told and what trustees intended.  It is too soon to go into detail on this but the key areas will be (a) pensions in deferment, (b) pension earned after 6 April 1997, (c) pension earned before that date.  For schemes nationally, research has shown that some 6/7ths have deeds that allow CPI for (a) and 2/5ths have deeds that allow CPI for (b).  Area (c) is a special trustee&IBM deal, with no national equivalent.  AVCs that were converted to pension are another special deal and you may have paperwork explicitly referring to RPI.