This page created 28 Aug 2002

This article should be considered in the light of recent Government proposals for new laws.

This is a lengthy page but it can be read in parts:

What is the financial health of our schemes?

Does a deficit matter?

What might IBM do to cheapen the schemes.

The Trust allows members to see Actuarial Reports at intervals of three years and we shall see the next late in 2004.  So any calculation of the funding of the plan has to be an approximation.  For this approximation it does not matter much whether the final salary fund or the aggregate funds are considered; the sums are more obvious for the aggregate.

At the end of year 2000, the funds had a (non-statutory) "surplus" of £585 million.  The proportion of the assets which were equities (as opposed to property, bonds etc.) was very close to 50%.  The proportion was the same at the end of 2001, the latest date we have facts for, so we will calculate on the basis it has stayed that way.

In December 2000 the 50% in assets had market value approximately £2.1 billion.  By December 2001 this had dropped to £1.86 billion. (As a sanity check that is a 0.87 ratio when the FTSE 100 dropped by 0.83 ratio.)

If we make the assumption that the drop in value continued along FTSE 100 lines then the £1.86 billion value will have dropped to £1.58 billion (FTSE at around 4400, Aug 2002)

So the drop in equity values alone will have taken away most of the "surplus".  It is perhaps worth an excursion at this point into the question of whether assuming the equities proportion at 50% is sound.  Some companies have reduced that proportion to close to zero.  Those companies have been saved from the effect of the market dip. Boots the chemist did it to avoid the volatile accounting of FRS17.  (A lucky move, even though the introduction of FRS17 has been delayed.)   Thorn did it to manipulate the statutory surplus calculation in a way that made it look more reasonable to have surplus paid out to the company - a lucky, even if dishonourable move.  But IBM was never threatened by FRS17 since it does not do UK accounting; it is hard to see a reason for a profound change in the 50%, although there could have been some normal investment reasons.  (For what it is worth Equitable has some 15% of its assets in equities.)

There are inflows to the funds (e.g. dividends) and outflows (e.g. pensions in payment).  These tend to balance one another in the £100M to £120M per year range.  There are also smoothing effects because all years that don't match the actuary's expectations have the unexpected difference allowed for over a period of years rather than instantly.

Such factors tend to be swamped by the changes in the calculation of "surplus" introduced by changes in the various parameters supplied by IBM, the Trust, and the actuary.  (e.g. salary predictions, pension degradation predictions, longevity tables, return on investment predictions.)  The most critical of these is the real return on investment, that is the actual return on investment minus inflation.

After a period approaching three years when the real return has been significantly below the figures assumed in the year 2000 actuarial report (4% real return), it seems likely that the actuary will have changed the figure downwards for more recent calculations.  For each one percent that he does this, the liabilities of the scheme go up some £500M. 

So the best-but-very-uncertain guess we have for the current fund status is a deficit of some £0.5 billion.  What the trust does not want the members to know, we cannot be sure of.

The obvious question is, if that is right, does it matter?

From an Armonk headquarters point of view, half a billion to be paid over several years is not that dramatic.  (IBM's post-tax profit in 2001 was $7.7 billion)   But there are other IBM pension schemes also under pressure.  The U.S. pension scheme was quoted as having 120,000 retirees and 140,000 workers (with the latter number dropping quickly).  Its pensions scheme has lost a once gigantic surplus. So there are billions to find.  We know that the company has taken a big hit from buying its own shares expensively (Newsletter 12).  The book value of the whole company is down to $23.6 billion.  (Hope you understand the significance of that better than we do.  Does it mean Bill Gates, net worth $59 billion, could buy IBM as a present for his wife?)

The message is the usual one - do you own research, form your own opinion.  Is it likely that Armonk will use its control of the UK Trust to further cheapen our schemes?

If the schemes are not changed, can IBM simply not make sufficient contributions?  Well, to an extent they can, since the deeds say that, subject to a legality constraint, the Employer shall pay "such other contributions into the fund as the Principal Employer, having considered the advice of the Actuary, decides."  Note that under this clause IBM only has to consider the advice of the Actuary; it can then ignore it.

However, there is a limit to the non-contributing, imposed by the legal Minimum Funding Requirement (MFR).  If you really want to know about the MFR and how upcoming legislation might change it then here is a good article. But all it is really necessary to know is that the MFR puts a floor to the scheme funding.  That floor is not enough to guarantee meeting all the obligations if the scheme terminates (see later) but it does limit the underfunding.  In particular it requires the actuary to agree with the Trustee each year on a schedule of contributions, which IBM must pay, to ensure that the plan is funded under the MFR calculation.

If we assume, as the calculations earlier indicate, that a surplus of 18% has become a similar sized deficit then we might reasonably deduce that the MFR ratio which was "more than 120%" funded has become less than 100% funded.  (The two calculations of assets versus liabilities are not identical, but are similar).  So the schedule of contributions probably has non-zero amounts in it.  The trust chooses not to tell members what the schedule is.

So the MFR limits Armonk's cheapening, unless changes are made to the scheme parameters or structure.

Pensions in payment (PIP) policy has an effect on cost.  You may believe that PIP policy was determined by the bargain established when you worked, which was for a competitive policy.  You may believe the policy is determined by the several statements over the years that our scheme is funded to degrade pensions by only 30% of inflation.  You may believe that any maintenance of value (of main service pension) is corporate charity.   On the first policy, your pension degradation will depend on what other trusts do, and it is relevant that other companies' schemes will also be less secure than they were.  (The Financial Times, 10/8/02 reports that, on average, funds are underfunded, at 88%.)  On the second policy, the automatic rule that has characterised more than a decade will continue.  On the third policy, do you expect Sam Palmisano to be more or less charitable than Lou Gerstner?  

How might IBM change the structure of the scheme?

The word "might" in this sentence has to be considered with care.  Ultimately, what IBM might do is constrained by UK and European law (not U.S. law).  As usual, the law is difficult to discern (there are 700 pages of company law) and every case is different, but commentators make the general point that the pension scheme expectation is part of the incentive for staff to remain working for the company. Having obtained that loyalty, failure of the company to follow through may be a breach of contract or the equivalent of a pay cut without consultation.  This is not just theory - the Big Food Group (formerly Iceland) is being faced with legal action from its employees.  Ernst & Young gave up plans for change when members threatened legal action.

(And for the unionised, industrial action is possible - workers at three plants run by Caparo Steel are in the middle of a series of one day strikes over pension changes.)

However, for the rest of this discussion we interpret "might do" in the sense of "what the trust deeds and 1995 Pensions Act do not expressly forbid".

Termination (or discontinuance or winding-up) is the term for what happens when a pension scheme, trust and all, ceases to exist.  Closure is the term used when the trust continues to operate but parts of the schemes in it do not.  IBM could terminate the I-Plan separately but can only terminate both C-Plan and M-Plan at the same time because they are combined by the deeds. Closure requires a scheme amendment and the approval of the trustees.  Termination can be imposed by IBM (or the trustees).

Closure of final-salary scheme comes in various degrees.  Closure to new members is what has already happened to the C-Plan.  Closure for the active members could occur in at least two forms.  One way is for an accountancy calculation to derive a single money figure for the value of what has been bought with your work (called the transfer value of your accrued rights) and use this much money as your initial credit in the M-Plan.  That is how it works for those who currently choose to move from C-Plan to M-Plan.

There is an alternative approach of turning the active member into both a deferred member of C-Plan and a from-scratch new member of the M-Plan.  (As if the member had left IBM and immediately rejoined as a new employee.)  That alternative is not so likely to be chosen.

Closure of the C-Plan to retirees cannot happen, except as part of termination.

At termination, the employer has to make a contribution, if necessary, to remove any underfunding according to the MFR test. (Although if the employer is bankrupt,  that debt may not be collectable.)  Termination requires all the assets to be turned into money. Money is then spent on trying to ensure the pensions get paid, by new arrangements.   This means buying an annuity or buying into another pensions scheme, for each individual.

The aim of the current MFR test is  that a scheme fully funded according to the MFR would, if the employer became insolvent, have sufficient assets to be able to protect fully pensions already in payment and to provide younger members with a transfer value that would give them a reasonable expectation of replicating scheme benefits if they transferred to an occupational or a personal pension scheme  In practice a scheme which is only just MFR funded will probably not be able to do these things.  (Since the MFR rest was introduced, facts such as the cost of annuities have worsened.)

So if the MFR test is not met, or otherwise proves inadequate, somebody will not get what they would have expected if the scheme had not terminated.  The pecking order is outlined by the 1995 Pensions Act and detailed in the deeds of our trust.  There are three levels of priority, roughly pensions in payment taking precedence over legally specified augments taking precedence over pensions not yet in payment.  In the top priority we have pensions in payment (including spouses but no future increases), AVCs and the M-Plan money.  The scheme would have to be horribly underfunded if these liabilities could not be fully met.

In the second priority are the Guaranteed Minimum Pensions.  That is a small part of your pension that arises under particular government regulations - the "Its your pension" leaflet you received in 2001 describes it.

Last in priority are the actives and deferreds. (Actives are treated as if they left the scheme on the day before it terminated, and thus became deferreds).  If there isn't enough for this priority, nobody knows what happens until it happens.  What is available "shall be apportioned by the Trustee amongst the individuals within that Priority, in such proportions as the Trustee having considered the advice of the Actuary shall decide."

Experience of other companies shows that this abrupt priority order can lead to anomalies.  Otherwise similar employees, one retiring just before termination and the other not, can get very different pensions.  There are sad tales of people who worked a long time but got little at termination.  This becomes an aspect of any early retirement decision, if you think termination is a significant possibility.

The summary above necessarily omits detail.  As usual, check out your own circumstance for yourself.