18 May 2006

A request:  If you are an "E-mail Buddy", please print this newsletter and give it to your buddy.

The story of the redesign of the IBM UK pensions schemes can now be summarised.  IBM wanted the schemes cheapened as part of a world-wide move away from final salary schemes and through a desire for IBM UK to be more of a low-benefits and short-term-profits company.  This was a choice rather than a necessity because both IBM UK and IBM worldwide were showing increasing profits.  IBM wanted the change quickly so that the effect on the future would appear on the immediate accounts and because if changes were decided after April 2006 they would require regulated consultation with the staff.  The package of changes was not in the interests of the scheme members.

In regard to employees the protection provided by the trust deeds was made irrelevant by the advice from the trust's legal advisers, Nathan Nabarro, that IBM might legally agree changes to the pension arrangements directly with the individual employees, in the form of changes to terms and conditions of employment.  The legal advice was that if IBM did that the trustee would have to follow, overruling what the deeds say.  In regard to the retirees and deferreds the trust chairman could, in theory, negotiate something to be put to the trust board, as part of the whole package.  The carrot/stick from IBM was the £500m immediate contribution to the fund; this was an advance on the money that was due to come from the Guarantee.  The carrot/stick available to the trust was the potential to call in the "debt on the employer". 

For December 2003 the "debt on the employer" (Actuarial Report on Main Plan, paragraph 3.17) was £1181m.  The I-Plan debt was £134m.  Rough calculations on changes (mostly the reduced yield on gilts) up to December 2005 suggest a debt then (for the final salary schemes together) of some £2000m.  IBM would not have enjoyed the prospect of finding an extra £1500m of contribution immediately, although that would have eliminated all future employer contributions.  Calling in the debt also has some disadvantages for the members.  A compromise that was less damaging to members than IBM's proposals, but required less from IBM than the £2000m, would seem to have been a possibility.  There is no evidence that the trust analysed and proposed such a compromise.

This was a once-only opportunity for the trust.  When the redesign is implemented the benefits will be lower and the debt from the company to the trust lower.  It will be illuminating to see how long it takes for the fund to get a further £1500m in employer contributions; it may never happen.  (The "Guarantee" is still in place but it will bring in much less cash than was expected when it was set up.)

Recent months have been spent in explaining the complicated new arrangements.  For retirees, and those in prospect of being retirees over some of the next fourteen years, this meant explaining how increases would be calculated.  In its description of the redesign AMIPP gave the algorithm which has been used for more than a decade, and in fact this algorithm was used for the 0.9% increase delivered in April 2006.  Pension Services however, gave a wrong description of the algorithm in the announcement of the increase, saying it would be "pro-rated" to ten months.  There is no pro-rating. Members who took the annual increase in RPI and pro-rated to ten months got a different figure than the actual increase and naturally thought there was a mistake.  The mistake was in the description; the increase matched the calculation IBM included in its package of proposals. 

Employees have been showered with documents, presentations, some help with using Independent Financial Advisers, and access to a program purporting to compare their options.  This communication has been poor and delayed but even if it had been better the employees would have faced considerations which couldn't be reliably predicted.  Given IBM's worldwide plans to get out of final salary schemes, will those who opt for final salary now find their scheme degraded further in the future?  Given IBM's trend to be the low benefits employer will those who opt for money purchase enhancements find the enhancements reduced (which would be fairer to those on the standard DC scheme)?  What will happen to inflation?  (See "Considerations" for some computed effects)  How long will retirement be?  What will IBM do to allow employees to take a pension while continuing to work for IBM?  (Other companies have introduced schemes which allow this "ramp-down" to retirement, as opposed to "cliff-edge" retirement).

Faced with these uncertainties, the sensible employee will be looking at a range of scenarios.  AMIPP has given some information on the variability of lifetimes.  The variability of IBM's future behaviour is obviously difficult to cover in scenarios but the long trend of delivering less than promised shows no sign of abating so prudence implies some gloomy scenarios should be included.

An opportunity was lost with the "Comparator" program.  It doesn't compare costs and it doesn't compare values.   The retiree MEDs, at the time the program was postulated, were both experienced software developers but it should not be assumed they had input to the program design.  The characteristics of the application suggest it was developed by actuaries and pension administrators without adequate input from the intended users.

It is not yet apparent what level of challenge there will be to the IBM package redesign.  The issues most likely to provoke challenge are (a) degrading the basis of converting AVC money to DB pension shortly after members had made their decisions on the better basis and (b) the soundness of the legal advice that the trustees received.  Some aspects of these are covered in AMIPP's description of the redesign.  Pension lawyers base advice that company agreements can override deeds on a case that looks to the layman very different from the IBM case.  In that case the redesign incorporated what was non-pensionable into the pensionable pay - IBM's goes the other way.  In that case there was agreement between the company, the trust, and the unions about the redesign and the agreement had been approved in a ballot of the union members - the disagreement was from some individuals who felt the agreement should have been even better for them.  IBM is the only approver of IBM's proposals.  Despite this apparent inapplicability, it can only be challenged if some employee challenges it.  From the point of view of the individual this is a bad deal - questioning IBM management decisions is not a good career move and the financial gain from success would be small compared with the financial aspects of the issue taken over all the membership.  Why would any employee be so self sacrificial?

One can reasonably ask how the individuals in the referenced case got their "day in court" when IBM complainants were unable to get the Pensions Ombudsman's dubious findings to court.  The answer is cost.  In the referenced case it was the trust that asked the court to clarify what it could do and hence the trust carried more of the cost.  Our trust chose to accept the legal advice without testing it in court. 

Over a period, AMIPP has lost contact with a number of people, listed in Lost Members.  They might have changed their email address and not told us, or they might have lost their jobs and no longer have an email address. If you can help by reminding them of the need to re-register, or by being an "e-mail buddy" for them, please do.  If you know of IBM leavers who might be receiving these newsletters at an IBM email address please remind them they will need to use the change of email address page.