3. PENSION
DEBITS—SECTIONS
4-7
3.1 General
We can see the reason behind the negative deferred
benefits approach, as described in the explanatory notes to Clause
6. It appears to be designed as a way of ensuring that "leaver
profits" do not arise on divorce. We had previously expected
that the pension split would be effected once and for all as at
the date of the divorce, as your notes say will be the case for
money purchase schemes. The negative deferred benefit is quite
a different approach. It does raise the potential problem that
the negative deferred benefit might prove to be more than the
benefit that actually becomes payable. This could happen if the
benefit structure changed, or rates of pay did not keep pace with
inflation, or were even reduced. Such cases may be uncommon, but
there is the possibility of loss for the scheme, which is not
balanced by any possibility of profit.
It would seem that pensions in payment can be
split.
3.2 The Debit
The principle
We accept the principle of negative deferred
benefits for salary-related schemes, if that approach is considered
necessary for policy reasons. However, we find the drafting of
Clause 4 and Clause 6 convoluted and unclear, and we are concerned
by the lack of flexibility about the pension debit in Clause 6.
Flexibility
Firstly, we would suggest that the scheme be
explicitly allowed to adopt a general policy as to the method
of cancelling benefits, provided their value, in accordance with
regulations, equals the amount of the debit. This approach would
mirror the approach to the pension credit contained in paragraph
6 of Schedule 2. As currently drafted, Clause 6 may allow this
flexibility, as it would be a question of legal interpretation
as to what is a "relevant benefit". However, your note
makes it apparent that the intention is to be more prescriptive
for active members.
Current drafting
If flexibility is not to be allowed, the drafting
needs to be clearer and more specific. As currently drafted, it
is far from clear how Clause 6 would work. Following the drafting
through, starting from Clause 4, the pension sharing order will
specify a percentage—say 40 per cent. The parties will have
agreed on that figure as a percentage of the cash equivalent.
That percentage is then converted into an amount, and the debit
is expressed as an amount under Clause 4(1)(a). However, that
amount then needs to be converted back into an appropriate percentage
for the purposes of Clause 6, in calculating the debit. The appropriate
percentage is defined in Clause 6(5) as being the percentage specified
in the relevant order, so again it is the same 40 per cent. However
for an active member, that appropriate percentage then has to
be converted back again into an amount, for the purposes of Clause
6(2). Could this drafting structure not be simplified?
Timing
There is nothing about the timing of the reduction
of benefits in Clause 6. Clause 6(2) says "each . . . future
benefit . . . which corresponds to a relevant benefit" is
reduced. It is not clear whether the reduction is to be made when
benefits come into payment, nor whether it is to be made to benefits
which accrue over future service (or in relation to past service,
by augmentation or salary increases). Or is the definition of
relevant benefits in Clause 4(4) meant to extend to Clause 6?
"Relevant benefits"
You have used the cash equivalent as a mechanism
for identifying the benefits that are to be debited. A benefit
will be a "relevant benefit" and have to be reduced
by the appropriate percentage if the cash equivalent includes
"an" amount in respect of it, although presumably it
does not matter if the amount is less than the full value of the
benefit (for example if allowance has been made for the contingency
of the benefit becoming payable).
It seems that there could be a mismatch between
the benefits to which the cash equivalent relates, which have
to be reduced, and the benefits that actually become payable.
For example a cash equivalent may include allowance for death
in deferment benefits, which never become payable to the member,
whereas benefits that actually become payable to the member may
include a death in service lump sum or an ill-health pension or
bridging pension, for which no allowance has been made in the
cash equivalent. Or would an ill-health pension be treated as
a "relevant benefit" because it is simply a member's
pension, and is therefore covered by the cash equivalent?
If the cash equivalent includes allowance for
a spouse's pension on death in deferment and a spouse's pension
on death in retirement, are those two relevant benefits or one?
and is the spouse's death in service pension that may actually
come into payment on the member's later death the same kind of
relevant benefit? Lawyers, and we suspect actuaries, would regard
them as three different types of benefit, as each of them may
be calculated differently. Is the spouse's death in retirement
pension to be calculated on the basis of the member's pension
as reduced, or the member's pension without reduction but then
with a negative spouse's pension deducted? Clause 6 says it is
to be reduced by the appropriate percentage, but 50 per cent (if
that is the percentage) of a spouse's death in retirement pension
may be a lot more than 50 per cent of a spouse's death in deferment
pension.
Also, we are concerned at the idea that separate
tranches of benefit may need to have separate debits. Benefits
may consist of pre-contracted-out rights, GMPs, excess over GMP
and post April 1997 rights, AVCs, transfers in and augmentation
credits. In addition many schemes may have periods of different
benefit scales, for example final salary and then money purchase
benefits or vice versa, or periods when pensionable service carries
different survivors' pensions or pension increases. Is each of
these tranches a separate relevant benefit? It is the impact of
revaluation and pension increases that causes the problem, because
different tranches of benefit may need to be rolled forward separately.
If so, they will need to be separately shown on benefit statements,
each with its own separate debit. The possible impact of overall
Inland Revenue limits on the gross benefits, before deducting
the debits, will complicate matters further. If the limits bite,
which tranche of benefits will be treated as reduced first?
We wonder how easy it will be to communicate
the actual benefits to an individual member if the negative deferred
benefit approach is followed.
Examples
It is helpful that you have provided an example
in the explanatory notes. However the example only seems to reflect
one possible interpretation of Clause 6. If you want to say that
the benefits under the scheme can't be reduced until benefits
actually become payable, or that the benefits are reduced at the
time of divorce but that the negative benefits have to rolled
forward after the date of divorce (but only in salary-related
schemes?), it seems necessary to say something more than Clause
6 currently says.
We would refer you to the model of the Consumer
Credit Act 1974 which contains examples in Schedules. This approach
was, we believe, regarded as successful by practitioners and we
would recommend that you consider following it. As a general rule,
explanatory materials outside of legislation cannot normally be
relied upon in interpreting it. It would give insight to the DSS'
intentions in a way that would have legal weight and could be
relied on in Court.
Further discussions
We would welcome further discussion with the
DSS on how Clause 6 can be reworked to achieve the policy intention.
3.3 The CETV
The cash equivalent does not seem to be defined,
but is apparently to be governed by regulations under Clause 5.
This suggests that it may be different from the cash equivalent
transfer value (CETV) the pensions industry is used to. Your explanatory
note says the regulations for calculation in divorce cash equivalents
will "broadly reflect the principles" applying to normal
CETVs. We had thought that the aim was to use the normal CETV
as the measure of value because it is readily available. Your
explanatory note refers to a minimum consistent with MFR, and
we would have thought the CETV would already satisfy this test.
We wonder what sort of adjustments to the MFR basis you are referring
to in your explanatory note—for example, must the calculations
assume that the scheme's assets are entirely invested in equities
for a younger member, and in gilts for an older one?
We recognise that extra elements may need to
be built on to the CETV legislation, for example because it does
not currently apply to pensions in payment. However these divergences
should be kept to an absolute minimum in the interests of administrative
simplicity.
We note that Clause 4(4) defines this cash equivalent
as being the equivalent of the "benefits or future benefits
to which . . . the [member] is entitled . . . [other than as a
survivor]". The CETV under s.94 of the Pension Schemes Act
1993 ("PSA") relates to "any benefits which have
accrued to or in respect of [the member]". We note the assumption
in Clause 4(5) that the member has left pensionable service, which
helps to bring the cash equivalent in line with a normal CETV,
but we are puzzled by the reference to "future benefits"
in Clause 4(4). There should be consistency.
It would seem that you do not intend to allow
the cash equivalent to be reduced in the circumstances where the
normal CETV can be reduced, for example because the scheme is
underfunded.
3.4 The debit to contracted-out rights
Clause 7 covers reduction of protected rights
or of the GMP.
Where the member has protected rights, the effect
is to reduce the protected rights by the appropriate percentage
of the amount of the protected rights immediately before the pension
debit arose. The timing is clear, in contrast to Clause 6—it
seems the deduction can be made once and for all within the implementation
period.
Where the member has a GMP, that GMP will be
reduced, apparently with effect from State Pension Age (unless
the date on which he or she becomes subject to the pension debit
is later) by the appropriate percentage. The wording has the effect
that, where the GMP is revalued, the reduction is by the appropriate
percentage of the GMP, as revalued from time to time. The problems
which we identified with Clause 6 also apply to these provisions
concerning the GMP because the adjustment in the amount of a member's
GMP is in line with earnings factors, whereas the member's ultimate
benefit will be linked to final salary or investment returns.
Again there is a risk of a mismatch.
3.5 Minor points
In Clause 4(7), in the definition of "relevant
arrangement", we wonder whether "arrangement" should
be "pension arrangement".
