After an extensive consultation process, the calculation basis for Cash Equivalent Transfer
Values (CETVs) for final-salary pension schemes is changing. The new basis is due to come
into effect in April 2008, but this does not mean that thereafter CETVs will always produce a
fair value for dealing with pensions in a divorce or dissolution.
To understand why, it is necessary to look at when and why CETVs were introduced. With the benefit of this
perspective, lawyers will be better equipped to fulfil their duty of care obligations towards their
It is not necessary to be a pensions expert to appreciate the basic issues and avoid the main
pitfalls, as this article seeks to demonstrate.
The Social Security Act 1985 gave pension scheme members whose pensionable service
ended on or after 1 January 1986 a statutory right to have the 'cash equivalent' of their
benefits transferred to another pension arrangement. This legislation is now consolidated in
sections 93 to 101 of the Pension Schemes Act 1993, with associated regulations – the
Occupational Pension Schemes (Transfer Values) Regulations 1996.
A condition of earlier Regulations (under the 1985 Act) was that 'cash equivalents' had to be
calculated and verified by adopting methods and making assumptions that were certified by
an actuary as being consistent with "Retirement Benefit Schemes – Transfer Values (GN11)"
issued by the Faculty and Institute of Actuaries. GN11 applies to all pension schemes, both
private and public sector.
The Welfare Reform and Pensions Act 1999 introduced the possibility of pension sharing on
divorce. The Divorce etc (Pensions) Regulations 2000 and the Pensions on Divorce etc
(Provision of Information) Regulations 2000 set out how benefits under a pension
arrangement must be calculated. Regulation 2 of the Pensions on Divorce etc (Provision of
Information) Regulations 2000 imposes upon the person responsible for a pension
arrangement the requirement to supply pension information, including a valuation. Regulation
3 sets out how the valuation is to be made and calculated, and retains the requirement for the
valuation to be consistent with GN11.
It is important to recognise that these regulations concern the manner in which the benefits
are valued by the scheme (the CETV). They also provide that a pension sharing order is
made by reference to the CETV calculated and quoted (and if necessary because of the
elapsed time and implementation of the order, re-calculated) by the scheme. The CETV is
therefore a consistently calculated number. It is incorrect to regard the CETV as
unchallengeable. The relevance of this point will be discussed later.
Changes to the calculation basis
In May 2005, the Actuarial Profession issued a draft revised version of GN11, known as
Exposure Draft 54 ("EXD54"). This draft proposed a substantial alteration to the framework for
the calculation of transfer values. EXD54 provoked a great deal of comment within the
actuarial profession and the wider pensions community. Many commentators feared that the
proposed changes to GN11 would substantially increase Cash Equivalent Transfer Values.
The actuarial profession subsequently asked the Government to reconsider the legislative
basis for the calculation of transfer values. The result was a Consultation Document
"Approaches to the calculation of pension transfer values" published by the Department of
Work and Pensions (DWP) in June 2006. Consultation closed on Friday 11 August 2006.
In January 2007, the DWP published its "Response to the Consultation". In a Written
Statement by the Minister for Pension Reform (Mr James Purnell MP), the Government
announced the details of its decisions, which are published in the "Response to the
Consultation". The Government intends to regulate on the calculation basis, which is different
in a number of technical ways from the current GN11 basis. Broadly, the values will be
according to the expected cost of providing the benefits if the member remained in the
scheme until retirement. This is some way short of the proposals in EXD54 and, despite two
years of debate and consultation, in the values it produces it is broadly the same as the
current GN11 basis.
It should be noted that the new regulations will only apply to private sector pension schemes.
The Government is still considering arrangements for public service pension schemes and
states in the Response that it '… hopes to make an announcement soon. …'
The revised calculation basis only affects CETVs for defined benefit pensions. These are
likely to be the biggest concern for lawyers, largely because they can understate the true
value of the benefits by 30% or more. CETVs for defined contribution pensions (often referred
to as money purchase pensions) are unaffected by the change in the calculation basis but
that is not to say that the CETVs for these schemes represent a fair valuation of the benefits.
However, the true value of these benefits is much easier to calculate.
According to the Response Document, many respondents said that existing rules on the
calculation of cash equivalent transfer values were unsuitable for use in pension sharing on
divorce (and dissolution), because the values do not reflect the value of the promised benefits
to the member. The Government's response, published in the Response Document, is to note
the concerns and give them separate consideration.
Non-divorce CETV issues
Some understanding of the issues surrounding CETVs in a non-divorce context will assist an
understanding of why they are so problematic on divorce and dissolution. The challenge for
those responsible for calculating CETVs is to calculate a fair value to offer for the benefits
being transferred out that does not shortchange the transferee but also does not leave the
remaining scheme members short of funds for their own pension entitlements. This is difficult
because the cost of final salary pension promises cannot be known for certain, until people
actually retire. Therefore, it is necessary to make a number of actuarial assumptions.
If implemented, the EXD54 basis could have had the effect of increasing transfer values,
particularly for young people. That could have led to many more people deciding to transfer
out of final salary and into money purchase schemes. As a matter of public policy, this is
generally considered undesirable. It may also have had the knock-on effect of increasing the
required scheme funding levels thereby increasing scheme costs overall. With so many
pressures on final salary schemes, the likelihood is that even more employers would decide
to close their final salary schemes and move to the generally inferior defined contribution
The cost to schemes of actually calculating CETVs is considerable. Regulations require
schemes to provide one free valuation a year, but recommended charges for additional
calculations are £150-£200. These charges are clearly not a true reflection of the costs
because schemes have developed procedures and mechanisms for producing CETVs and
these are absorbed into the overall scheme administration costs.
In the DWP Response Document, the estimated cost to private sector final-salary schemes of
the proposed changes to the calculation basis is £8 million. This cost gives a possible clue as
to why the established CETV was adopted in the pension sharing legislation. Pension
schemes would be likely to object to the introduction of a different valuation basis for pension
sharing purposes, probably pointing to the number of CETVs produced for divorce cases
compared to the number of pension sharing orders made.
What is wrong with relying on CETVs in divorce?
So the CETV, whether it is on the existing or revised basis, is actually intended for a different
purpose. There are several reasons why this is the wrong number and we will look at some of
CETVs are calculated as if the member has left pensionable service. This may well be the
case, but if the member is still in pensionable service, the true value of the benefits will be
greater than the CETV.
Some schemes award discretionary increases in benefits such as inflationary increases and
have an established record of doing so. Schemes are free to ignore these discretionary
increases when calculating CETVs; in which case the CETV will be lower than the true value.
The present calculation basis requires that the values represent the expected cost within the
scheme of providing such benefits and should be assessed having regard to market rates of
return on equities, gilts or other assets as appropriate. Under the revised basis, the transfer
value should reflect a 'best estimate' of future returns having regard to the existing asset mix
of the scheme. The subtle and somewhat technical difference between the existing and new
basis is that the new calculation basis reflects the current, rather than theoretical, investment
asset mix of the scheme.
Generally, over a long period, equities (stocks and shares) have historically outperformed
gilts. If a scheme's assets are largely invested in equities (or they are assumed to be for the
purposes of calculating CETVs), long-term investment returns will be good compared with
those of gilts. Consequently, pension contributions from the employer can be lower than
would otherwise be the case. As the CETV is calculated on the cost to the scheme of
providing the benefits, actuarial assumptions on future investment returns can have a
significant effect on the calculated CETV.
UK mortality rates have changed significantly over recent years. People are living longer than
was expected even two decades ago. If people live longer, they draw their pension longer,
which makes providing pensions more expensive. This translates into higher pension
contributions from the sponsoring employer in order that the pension scheme can meet its
pension promises. Not using the most up to date mortality tables has implications for the
required level of employer contributions but it also depresses the quoted CETVs.
For private sector pension schemes, legislation requires that they are funded to specific
levels. At any one time, they are not required to have sufficient assets to meet all their current
and future liabilities. Instead, they are required to make assumptions about current and future
liabilities along with assumptions about future rates of investment returns on their assets. In
simple terms, the result is the total level of funding required to be reasonably confident that
the current and future pension liabilities can be met.
Not all employers are always able to make pension contributions to achieve the level that is
described as being fully funded. The business climate for the employer may affect its ability to
make the required level of contributions from time to time. Other economic and political
circumstances may also affect the pension scheme's assets. For instance, a significant fall in
the stock market may affect the value of the scheme's assets and political decisions such as
the abolition of favourable tax treatment on investment dividends can have a similar or
accumulative effect. So, there are various reasons why a pension scheme may become
technically under-funded. In many cases, the employer is able to redress the under-funding
over time; few are in a position to restore the pension scheme's funding position immediately.
An under-funded scheme may at its discretion, reduce its quoted CETVs proportionately. For
example, a scheme may be 30% under-funded and therefore reduce its CETVs by 30%.
Whilst this is equitable in the context of the intended purpose of CETVs, it has a distorting
effect when the CETV is required for divorce or dissolution purposes.
Other than a few notable exceptions, such as the Local Government Pension scheme, most
public sector pension schemes are not funded in the way private sector schemes are.
Instead, pension payments are met out of tax revenue. CETVs for public sector pension
schemes are therefore a theoretical value arrived at using factors supplied by the
Government Actuaries Department. As there are no underlying pension assets, the issues
concerning investment returns and scheme funding do not apply. Nevertheless, CETVs for
public sector pension schemes do not represent a fair value in the divorce and dissolution
Special considerations apply to some public sector pension schemes. In calculating the
CETV, it is usually assumed that pensions are taken between the ages of 60 and 65. In
practice, the rules of some public sector pension schemes allow for retirement at much
younger ages, with little or no reduction in pension. This is encountered when dealing with
people in the uniformed services, such as police officers and soldiers.
How is a fair valuation different?
A fair valuation of a pension is the cost of purchasing or replacing the benefit in the
commercial market. A CETV is a number for a different purpose.
As in most cases, one party in a divorce or dissolution will lose part or all of a pension benefit,
the value used should be a fair value, otherwise the outcomes will not be those that are
It is unlikely that you would accept a valuation of the family home based upon the cost of the
materials and labour that were required to build it, so is it fair and reasonable to make
decisions on valuable pension assets based on the wrong numbers? The analogy is not
perfect but the effects are the same.
Further changes to valuation regulations
I do not believe that we will see special values introduced specifically for family law purposes.
For private pension schemes the cost of introducing a new and additional calculation basis
would be very high – far greater than the estimate £8 million cost of changing from the current
to new CETV basis. It would also be difficult to defend having a (fair) valuation for family law
and a CETV that by implication is an unfair valuation.
The decision on the CETV basis has been taken already. It is not in the public interest to
increase CETVs, as one potential knock on effect would be to increase the number of
transfers. This in turn might increase the level of employer contributions required to achieve
adequate scheme funding. Final-salary pension schemes are already in decline as employers
seek to reduce cost by moving to inherently cheaper defined contribution schemes.
Increasing required funding levels and administration cost would accelerate the movement.
Public sector pension schemes are a sensitive area. Compared to private sector pensions for
people on similar salaries, public sector schemes provide superior levels of benefits. They
often have inflation protection that is unaffordable to most private sector schemes, and they
have the advantage of being effectively guaranteed by government. The cost of providing
these pensions is a very sensitive political matter. As the cost of providing final-salary private
sector pension schemes increases and they are gradually becoming less common, the value
and cost of providing pensions for public sector employees receives more attention.
It is not in the public interest to encourage transfers from unfunded public sector pension
schemes by improving the CETV calculation basis. This may be just one of the reasons why,
according to the Response to Consultation document, the Government is still considering
arrangements for these schemes.
How fair is fair?
A CETV may understate the value of a pension by a significant amount. The problem for
lawyers is that it is almost impossible to know how unfair a value is without doing some
detailed calculations. As an indication, it is not unusual for a public sector CETV to understate
the pension value by 30% or more and that means that it cannot be a scheme funding issue.
Public sector schemes are not unique in this respect and similar differences occur on private
Clearly, if the CETV itself is small, even a significant discrepancy is immaterial compared to
the cost and effort involved in getting a fair value calculated.
Why it matters that CETVs are not 'fair' values
It could be argued that if a pension sharing order is made, the fact that the CETV is not a fair
value is immaterial. If you slice an apple in half, neither the size of the whole apple or the two
halves matters. The only concern is that it really is cut in half. The same is true of final-salary
pension schemes. If the value of the assets allows all the pensions to be shared equally, by a
making pension sharing order, the fairness of the CETVs may be ignored. However,
difficulties arise where one asset is offset against others. If a fair valuation is made of each
asset, then offsetting is a simple matter but if a pension is offset solely on its CETV the result
could be very unfair.
The cost of making and implementing pension sharing orders may mean that it is uneconomic
to share all pensions equally, in which case an element of offsetting is inevitable. An
important point to remember is that the intention to make a pension sharing order does not
mean that the fairness of the pension value can be ignored. It is how the shares are arrived at
that is important.