A favourite film of one of the authors is Apollo 13. At one point, whilst trying to work out what had to happen to get the ill-fated space craft home, Commander Jim Lovell said: ‘All right, there’s a thousand things that have to happen in order. We are on number eight. You’re talking about number 692.’ The same must be the view of so many family lawyers regarding worrying about lifetime allowance (LTA) tax issues in the midst of financial remedy proceedings. In many cases, however, it is important that consideration of this subject is not relegated to the end – to do so may cost your clients many tens or hundreds of thousands of pounds in unnecessary tax.

In this article, we:

• briefly discuss the tax and how it is applied differently to defined benefit (DB) and defined contribution (DC) pension schemes;
• look at a number of case studies;
• look at how some of the HMRC-granted protection certificates may be affected by pension sharing orders (PSOs); and
• consider how LTA tax can be mitigated in some cases.

Background

The LTA is a limit to the total amount of tax-efficient pension benefits that an individual can accrue over their lifetime. It was introduced on 6 April 2006 (what was then known as ‘A-Day’). The LTA threshold – the amount of pension fund one could have before paying tax – started at £1.5m, rose to £1.8m in 2010, and then gradually fell back to £1.0m in 2016, before slowly climbing back to the current £1.073m.

Pension benefits are tested against the LTA as follows:

• DC arrangements with reference to their fund value when benefits are taken; and
• DB pension amounts are multiplied by 20 upon being put into payment to place a value on these (with any automatic lump sums been taken at face value).

Any excess over the LTA gives rise to a tax charge that is calculated depending on how these excess benefits are taken:

• if taken as a cash lump sum, the excess is subject to a 55% tax rate;
• if taken as an income, it is taxed at 25% in addition to the normal income tax regime.

These two approaches are often neutral to each other from a tax perspective.

HMRC provides several protections against the LTA, some of which are no longer available to new applications. Details of these can be found at www.gov.uk/guidance/pension-schemes-protect-your-lifetime-allowance. The impact of a PSO on the pension debit member (i.e. the one whose benefits are being reduced) can be affected where some such protections are held. It is also the case that the LTA position of the pension credit member (the one who is to receive the benefit of a PSO) can also be affected depending on certain specific factors that are outside the remit of this (relatively) brief note.

Between 2018/19 and 2020/21, the LTA was increased in line with inflation, as measured by the consumer prices index (CPI). In the 2021 spring budget, however, it was announced that the LTA would be frozen at the 2020/21 level of £1,073,100 until the end of tax year 2025/26. Thereafter, who knows what may happen, but in the current climate, expecting any relaxation of this limit which by definition will benefit only ‘millionaire pensioners’ is perhaps being unrealistic.

Case study 1

Let us start off with a straightforward case. H is looking to retire imminently and has a self-invested personal pension (SIPP) worth £2,000,000; this is a DC fund. DC funds are easy to value for LTA purposes; it is simply the value of the fund that is tested against the LTA. Unless any action is taken, H will have a tax bill of c. £250,000 when he retires, due to being c. £1,000,000 over the LTA threshold of £1,073,100.1

H and W are getting divorced, and a PSO of 50% is made in favour of W, who prior to the PSO, has no pension funds of her own, and the order is made prior to H’s retirement. H and W now each have £1m of pension funds, with their own individual LTA limit of £1.073m, and thus neither party has an LTA tax liability.

Case study 2

We are now going to take a simple DB case and develop the case with the addition each time of new features. This will hopefully bring out the differences between DB and DC pensions for LTA purposes and identify how the tax may be mitigated or unwittingly exacerbated.

Let us take the case of Mr Smith, aged 55, who is a member of a DB scheme. He has a deferred pension of £80,000 pa, payable ordinarily at age 60, and the cash equivalent value (CEV) is £2.2m. Mrs Smith is 5 years younger at age 50. We are asked to look at equality of income by means of pension sharing. The best way to look at the LTA issues is to consider the calculation for equality of pension income, which requires a PSO of 46.0%, but first totally ignoring the LTA issues:

So, what do we note from this calculation?

• First of all, we have ignored the instruction – we have assumed retirement at age 55, not age 60. And thus, if Mr Smith draws his pension at age 55, it is reduced from £80,000 pa at age 60 to £62,000 pa from age 55. The reason for this is that even at age 55, we have LTA issues. If we had assumed retirement at age 60, unless we are to assume that the LTA limit will increase each year by a rate greater than the anticipated investment return, deferral of the retirement until age 60 will simply exacerbate the LTA issues. Please bear in mind that the LTA limit is frozen until 2025/26 and in the current climate we are not sure it is realistic to assume tax regimes will become less onerous. Jonathan Galbraith (he of the Galbraith Tables) made just such a call in the case of W v H (Divorce Financial Remedies) [2020] EWFC B10, where, for practical considerations of LTA issues, he advised we should consider retirement at age 55, notwithstanding the letter of instruction.

In the (anonymised) written judgment (available at www.bailii.org/ew/cases/EWFC/OJ/2020/B10.pdf) HHJ Hess made explicit reference to Mr Galbraith’s report and acknowledged that it was appropriate therein for the expert to depart from the specifics of the letter of instruction with reference to assumed retirement ages in light of the facts of the case. The relevant section of the judgment is [63] (ii):

‘It has been suggested by Mr Galbraith from Mathieson Consulting Limited, the PODE instructed in this case, in his report of 3rd July 2019 … that (for reasons convincingly explained in detail by him which have been accepted by both parties, and which include a proper consideration of the Lifetime Allowance and Fixed Protection issues arising here) the appropriate equalisation age on the facts of this case is 60 (rather than the normal 65 or 67). I propose to adopt this recommendation.’

• Secondly, Mr Smith no longer has an LTA issue. Prior to the PSO, he had a pension of £80,000 pa if taken at age 60 or £62,000 pa if taken at age 55. DB pensions such as this one are valued for LTA purposes on a very simple multiple of 20 times the pension (plus a lump sum if there is an automatic one) and thus his pension would have been valued at £1.6m if taken at age 60 (20 × £80,000 pa) or £1.24m if taken at age 55 (20 × £62,000 pa). Thus prior to the PSO he would have breached the LTA of £1.073m, unless he had any protection. However, he has not crystallised his pension (by retiring) and thus the pre-PSO LTA issue is a hypothetical future liability. Once a PSO of 46.0% is made, reducing the pension to £33,461 pa at age 55, Mr Smith no longer has an LTA issue when he retires, with £669,000 (20 × £33,461) being well below the current threshold.
• Mrs Smith, however, does have an LTA issue. For DC funds, it is the CEV at the point of crystallisation which is used to value the pension for LTA purposes. With a pension credit now of £1,012,669, which is forecast to be worth £1,230,843 at age 55 (when she is assumed to retire), unless there is an increase on the current LTA limit of £1.073m, Mrs Smith will be £158,000 over the LTA limit, and thus face a tax charge of c. £40,000. We have ignored this tax liability in this calculation.
• This calculation highlights very clearly the lack of fairness in the way LTA rules are applied against DB and DC pensions. Both Mr and Mrs Smith are forecast to have the same guaranteed level of pension income, but Mr Smith is assessed as having funds below the LTA and Mrs Smith will require DC funds in excess of the LTA threshold. This is because using a factor of 20 to convert DB income into a notional value for a DB pension massively understates the true value of the pension, at least where an annuity purchase solution is used with DC funds for the purpose of matching incomes in retirement.

Case study 3

We now develop the previous case study and adjust the calculation such that the LTA liability faced by Mrs Smith is taken into account:

Here we can see:

• On Mrs Smith’s side of the equation, we have considered the tax she must pay, and deducted this from her fund, prior to the purchase of an annuity.
• As we have now considered Mrs Smith’s tax liability (unlike the previous calculation) the PSO must increase to compensate her, from 46.0% to 47.0%, and the pension credit increases from £1,012,000 to £1,032,911.
• But here we enter a vicious circle. To compensate Mrs Smith, via the PSO, for her tax liability, she needs to receive more pension monies which means the tax liability is increased, and thus she needs more pension credit to compensate her – and the only winner is the Exchequer. This is to the detriment of both parties, with incomes being equalised at a level that is c. £600 pa less each than what was shown in Case Study 2.

Case study 4

An alternative here is to recognise there is a sweet spot at which a PSO: (1) brings Mr Smith’s pension within LTA limits (must be below £53,655 pa, such that 20 times the residual pension does not exceed £1,073,100); but also (2) Mrs Smith at age 55 has no more than £1,073,100 of funds, and thus herself does not have an LTA liability. This can be achieved by a PSO of 40.1%:

But given a PSO of 46% is required to achieve equality of pension income (ignoring LTA liabilities), it follows that a PSO of 40.1%, which: (1) keeps Mr Smith’s pension within LTA limits; but also (2) gives Mrs Smith projected funds at age 55 of £1,073,100 will inevitably lead to an outcome where Mr Smith’s income is greater than Mrs Smith’s (£37,119 pa vs £29,173 pa).

So the proposal is that this settlement, which avoids losing £45,000 in tax from the joint pension pot, should be undertaken in conjunction with some offsetting for the income imbalance of £7,946 pa. It follows that there are various ways in which one might place a value upon the non-pension capital required by Mrs Smith to make up the difference here, and it is also widely accepted that adjustments in respect of tax and/or utility might be required in respect of any figure derived. The Pension Advisory Group (PAG) report2 can help with this.

Protection

At various stages in the life cycle of LTA (and it has only been with us since 2006), various protections have been made available – both transitional ones when the regime commenced, and thereafter primarily when the threshold has been reduced – to soften the blow for those who had funds that were within limits when thresholds were higher but would be in excess of a newly reduced threshold.

In case study 2 above, which evolved through case studies 3 and 4, what would have been the case had Mr Smith had Fixed Protection 2014, for example? Such protection would mean that Mr Smith could have pensions valued at £1.5m before LTA tax would become due. (It is worth noting that just because your client can produce a Fixed Protection certificate, this does not mean it is necessarily valid. If Mr Smith successfully obtained such protection in 2014, but had subsequently made pension contributions, such protection would have been lost.)

Assuming Mr Smith’s Fixed Protection 2014 of £1.5m was still valid, then he could draw a pension of up to £75,000 pa (20 × £75,000 = £1.5m) without incurring a tax charge. Thus:

• Ignoring any PSO, if Mr Smith were to retire at age 60 with the full pension of £80,000 pa, he would have a pension valued at £1.6m for LTA purposes, which is £100,000 more than his protection limit of £1.5m, thus giving rise to a tax bill.
• But if he were to retire at age 55, with a reduced pension of £62,000 pa, he would have no tax liability – a further reason why the PODE was right to have the temerity to suggest he should use age 55 instead of age 60 as instructed.

If the usual sequence of events were to be changed, such that before any PSO is made, Mr Smith were first to put his pension into payment, what then would be the outcome?

• First, Mr Smith would then be in receipt of a pension of £62,000 pa (assuming he took no tax-free cash) and he would have no LTA tax to pay on this, as he remains within the limits of Fixed Protection 2014.
• If following Mr Smith’s drawing of the pension a PSO is made, Mrs Smith will then receive a pension credit from funds which have already been tested, and even HMRC does not think it is right to test pension funds twice. Thus, Mrs Smith could apply for a ‘pension credit factor’ (note it has to be applied for and not assumed and can only be applied for if the pension being shared came into payment post-2006), which in effect means her own LTA limit will be increased to the extent of the pension credit she receives. The Pensions Tax Manual at PTM095200 refers (available at www.gov.uk/hmrc-internal-manuals/pensions-tax-manual/ptm095200).
• We can reconsider the calculation in case study 2 as an efficient solution if the PSO of 46% is made after Mr Smith has retired and tested his benefits, as in effect, Mrs Smith will find her LTA threshold increased from £1,073,100 by a further £1,012,669 – the extent of her pension credit, and thus no LTA tax is paid.

An alternative way in which protections may help would be if Mrs Smith had her own DC pension, to which she has not contributed post-2016, and thus Mrs Smith could retrospectively apply for Fixed Protection 2016, which would give her an automatic allowance of £1.25m. There are quite specific circumstances around the application for such protection, and to ensure that such protection is not then lost through receipt of a pension credit. Thus, if Mrs Smith had, say, a personal pension of £5,000, and she has not contributed to any pension post-2016, she could apply for Fixed Protection 2016, prior to receiving a PSO, and thus be allowed to have total pension funds of £1.25m without creating a liability to LTA tax. Again, if we refer to case study 2, if Mrs Smith could successfully apply for Fixed Protection 2016, then irrespective of whether Mr Smith retired prior to a PSO or not, Mrs Smith’s own funds would be within her newly created limit of £1.25m. Such are the subtleties of how the LTA tax regime and the pension sharing regime interact with each other: it is important for potentially affected individuals to tread carefully here and ensure that they are seeking the proper advice that they require to understand these issues.

Summary

• PSOs can be complex.
• LTA tax issues are complex.
• Where PSOs meet LTA issues, it is like the meeting of two great oceans. If there are issues regarding existing protections, it is like the meeting of two great oceans, with a violent storm overhead – all but the most competent of sailors should refuse to venture there.

There are only a handful of financial planners I know who are competent in this field – they must first and foremost be pension experts, but they must also be fully aware of the additional skills, knowledge and competencies of working within financial remedy proceedings. None of the information above is intended as formal advice and should not be used as such. It is also not intended that the above furnishes the reader with the skills to navigate these seas alone. It is intended to inform the legal practitioner of some of the issues, some of the pitfalls, and some of the very expensive errors that can be made if pensions are not handled correctly, so that they recognise when the appointment of a specialist financial planner is essential.

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 publication are not necessarily the views held throughout RBC Brewin Dolphin Ltd. Information contained in this document is believed to be reliable and accurate, but without further investigation cannot be warranted as to accuracy or completeness.