I started writing expert witness reports around 2005 and, right from the start, a regular question was, ‘Should we seek equality of income or equality of capital?’ Letters of Instruction regularly ask the expert to calculate ‘The pension sharing order required to achieve (i) equality of income and (ii) in the alternative, equality of capital.’

I have always wondered what the solicitor means by ‘equality of capital’ and what they expect the expert to do. The judgment of Moor J in CMX v EJX (French Marriage Contract) [2022] EWFC 136 is very interesting in how he tackles equality of income/capital, and how equality of capital can be achieved. The judgment has provoked much commentary, and challenges much of the received wisdom in this area.

One such commentary is from Calum Smith, pupil barrister (at the time of writing) at 1 Hare Court. Another is from Paul Cobley – no doubt known to many readers as one of the true experts on pensions and divorce, although he does not produce PODE (Pension on Divorce Expert) reports. Paul has been one of the main contributors to both iterations of the Pension Advisory Group. Both commentaries can be found on the Financial Remedies Journal website.1

In [49] and [50] of his judgment, Moor J writes:

‘49. Turning to pension sharing, my attention was drawn to the decision of Nicholas Francis QC, sitting, at that point, as a deputy High Court Judge in SJ v RA [2014] EWHC 4054 (Fam) where he said:

“Why should someone receive more just on the basis of gender? There may have been an explanation when rules required the purchase of an annuity. However, to give the wife more than the husband, on account of either age or gender, would seem to me to be unacceptable discrimination unless it is a case which is governed solely by needs. If a person should receive more of a pension fund under the modern rules simply because she (or he in the case of a marriage where the husband is much younger) is likely to live longer, then such an approach would logically extend to all capital assets. Moreover, European Union judgments and rules are rapidly outlawing discrimination on account of gender. In cases where distribution is being made on a basis which is not guided by need it is, in my judgment, incorrect to distribute a pension fund on the basis of equality of income…”

50. I could not agree more. If assets are to be divided equally, they should be divided equally. In general, there is no justification for awarding more to one party because they are younger or have a longer life expectancy. Both parties should share the fruits of the marriage equally. Moreover, in my experience, the only thing that can be said is that life hardly ever goes to plan, whether it be one party living far longer than expected or another remarrying immediately. It follows that I have become very troubled by directions that ask a pensions actuary to calculate a division on the basis of equality of income in retirement. Apart from the fact that such reports tend to be very expensive, the simple fact is that such a direction almost enshrines the Duxbury paradox into practice. It cannot be right, in general, that the younger you are, the greater your award. In any event, it has no place whatsoever in equal division cases.’ (original emphasis)

What is not clear is whether the qualification made much earlier in the judgment, at [14], is also to be applied to [49] and [50]:

‘14. At this stage, all I say is that I cannot conceive of a more inappropriate direction [for there to be a PODE report considering equality of income] when the assets in the case are between £22 and £28 million.’

So, if we are not to look at equality of income in bigger money cases, it is equality of capital. But what is meant by equality of capital?

Option 1, Equality of Capital = Equality of CEVs

The simplest way to interpret equality of capital is to take the cash equivalent value (CEV) of the pensions, add them up and divide by two. Take a very simple case:

• H has a private sector defined benefit (DB) pension (which does not allow internal implementation of a pension sharing order (PSO)) with a CEV of £500,000, and this is the CEV of deferred, index linked, pension of £30,000 pa, payable at age 60.2
• W has a defined contribution (DC) fund of £100,000.
• Between them, H and W have CEVs of £600,000, therefore equality is achieved if both parties end up with CEVs of £300,000, which requires £200,000 of CEV to be transferred, via a PSO from H to W, which in turn requires a PSO of 40% over H’s DB pension.

But what does this lead to in practice? Well, it depends on two dominating factors:

(1) Is the CEV of the DB pension of good value?

(2) Is there a material age difference between the two parties (what I refer to as the age factor)?

Let’s start by taking the age factor out of the equation (by assuming both H and W are the same age, and, for the purpose of this exercise, I will assume they are both aged 55). Such an approach (add up the CEVs and divide by two) will enshrine a disadvantageous position to the spouse if the CEV of H’s DB scheme is less than the true cost of such pensions on the open market (which is an increasingly common position following the fall in DB CEVs during 2022). Looking at the above scenario, where after a PSO of 40% over H’s DB pension, both parties have equal CEVs of £300,000, the income position may be as set out in Table A.

Table A

We can see in Table A:

(1) Following a PSO of 40%, both parties have pensions with a total CEV of £300,000. Equality of CEVs has been achieved.

(2) However, the pension H expects at age 60 is £18,000 pa and for W it is £13,224 pa.

In this scenario, if we wish to achieve equality of income, a PSO of 49.2% is required, as we can see in Table B.

Table B

Following a PSO of 49.2%:

• The CEV of H’s pension reduces by 49.2% from £500,000 to £254,107.
• At the same time, H’s preserved pension per annum within the DB scheme reduces by 49.2% from £30,000 pa to £15,246 pa.
• W gets a pension credit of £245,893, this being 49.2% of the £500,000 CEV, which when added to her own DC funds of £100,000 leaves her with total DC funds (CEV) after pension sharing of £345,893.
• Such a fund may be worth £420,832 at age 60, allowing W to buy an annuity of £16,833 pa, which when adjusted for inflation may be said to be worth £15,246 pa in today’s money, thus matching H’s income.3

Why, in the above scenario, does W need more than half of the pension capital in order to equalise incomes? It is nothing to do with age – we have assumed both parties are aged 55. Age is not the issue. It is because, as is often the case, the CEV of the DB pension of H understates the true/open market value of H’s pension.

I said above that there are two dominant factors which influence the outcome of a solution where the CEVs are added up and divided by two: (1) Is the CEV of the pension of good value? and (2) Is there an age differential?

I have shown above that where age differentials are stripped out, by assuming both parties are the same age and where the CEV of H’s DB pension understates a true value, then W will need more than 50% of the CEVs to equalise incomes.

What happens if we make age the dominating factor, and remove any influence of an understated CEV? We can do this by assuming H’s CEV of £500,000 is not in respect of a DB pension, but instead in respect of a DC fund. And thus (absent any guaranteed annuity rates4) the CEV is the open market value. At the same time, we will assume W is aged 50 and H aged 55.

Table C sets out the calculation for equality of income, using such assumptions.

Table C

We can now see from Table C that where there is no influence of a CEV undervaluing a pension, W is the younger party and both parties are pre-retirement, in order to equalise incomes, W needs less than half of the available CEV (£285,000 out of a total of £600,000, or around 47.5%).

It has been asked in judgments, ‘Why should one party get more than the other party simply because he/she is younger?’ The truth is, all other things being equal, the younger party requires less than the older party. If the PODE calculates the younger party needs more by way of CEV than the older party in order to equalise incomes, this is because the CEV of the DB pension is understating its true value and is not a function of being younger.

Option 2, Equality of Capital = Equality of Open Market (or true or fair or actuarial) Value

So, having considered equalising capital values by reference to the CEVs above, and having found that approach perhaps to be wanting, not due to age but due to DB CEVs often understating the true open market value of a pension, is there a way in which we can look at equalisation of capital values which gets around this, and what does such a calculation achieve?

The way around the difficulties posed by the CEV of a DB scheme not representing full value, is to calculate for that pension a true open market value which could be the cost of buying an equivalent pension (in this case £30,000 pa from age 60) on the open market. In other words, we answer the question, ‘How much would H, in this case, need to have invested in a DC fund to provide this pension of £30,000 pa?’ In calculating such a figure, we are placing a value on H’s pension which is directly comparable to the CEV of W’s DC fund.

In this case, it may be that an open market value of £680,605 is placed on H’s DB pension. In other words, in this case, the CEV of £500,000 is only 73% of the open market value of the pension. Using this open market value, instead of the CEV, we can say that a PSO of 49.2% over the DB pension of H is required to equalise capital values, if we use the true open market value instead of CEVs, as set out in Table D.

Table D

What do we note from the calculation in Table D?

(1) Although we have calculated the open market value of H’s pension at £680,605, and this is the amount we use on his side of the balance sheet, W will still get a percentage of the CEV. Just because a PODE/actuary has said the true open market value is £680,605, the scheme is not going to change the CEV, and in this case W will receive by way of pension credit, 49.2% of the CEV.

(2) Both parties can then be said to have, post-pension sharing, pension assets with a capital value of £345,893.

(3) However, we can also see in this example, given the parties are the same age, that equality of capital, as measured by open market values, will also achieve equality of income.

In other words, if we use the open market value of pensions as a measure of capital, and if the parties are the same age, then a PSO to equalise incomes (49.2% in Table B) will be the same as the PSO required to achieve equality of capital values (Table D).

What happens, though, if we equalise capital values, as measured by open market values, and the parties are not of the same age? In Table E, we will revert to the assumption that H is 55 and W is 50.

Table E

We can see from Table E that where we use an open market value for DB pensions (instead of CEVs) and where W is younger than H (and both are pre-retirement) then a solution which provides equality of capital values will give W more by way of a PSO (49.2% of H’s pension) than if we sought to achieve equality of income (37.1% in Table C).

Conclusions

So, what conclusions can we draw from all of the above?

(1) If there is a DB scheme where the CEV understates the true value of the pension benefits, then a solution which uses the CEV as a measure of value will be to the detriment of the recipient of the pension credit.

(2) If there is a DB scheme where the CEV understates the true value of the benefits, then a PSO which is greater than that achieved by an equal division of CEVs will be required in order to equalise incomes. This is nothing to do with age.

(3) In fact, all other things being equal (if, inter alia, all pension funds are DC in nature and so there is no concept of a DB CEV understating the true value of a pension), and assuming the parties are pre-retirement, a wife who is younger than the husband will require less than half of the pension assets in order to equalise incomes, not more.

(4) If equality of capital is required, and it is required in a manner which is fair and not distorted by CEVs of DB schemes understating true values, then a PODE will be required to calculate a fair open market value of any DB pension.

(5) Only if you are prepared to accept the distortions and in-built unfairness created by a DB pension can you dispense with a PODE, and just use the CEVs of DB pensions without expert input.

(6) Whether or not it is appropriate to treat a pension as a capital asset, and to be divided based on capital values or as an income stream is a legal point and outside my area of expertise. I note, however, what the first Pension Advisory Group report says about this matter.5 But if it is to be treated as a capital asset, and divided based on capital values, and there are DB schemes (or DC funds with guaranteed annuity rates), a PODE will still be required.

(7) If you divide pensions based on fair capital values, then a wife who is younger than her husband will receive more from a capital-based division than she would from a division based on income.

NB: In all examples, it is assumed that both H and W are pre-retirement. If both parties are post-retirement, then the positions reverse: a younger wife would require more by way of PSO to equalise incomes than she would for equality of capital. It is also assumed that the DB schemes in question do not allow internal implementation of PSOs, which is the dominant position for private sector schemes.

Risk warnings

The value of investments, and any income from them, can fall and you may get back less than you invested. This does not constitute tax or legal advice. Tax treatment depends on the individual circumstances of each client and may be subject to change in the future. Information is provided only as an example and is not a recommendation to pursue a particular strategy. Opinions expressed in this article are not necessarily the views held throughout RBC Brewin Dolphin Ltd. Information contained in this article is believed to be reliable and accurate, but without further investigation cannot be warranted as to accuracy or completeness.