Apples or pears? Pension offsetting on divorce...Apples or pears? Pension offsetting on divorce Rhys...

13
Apples or pears? Pension offsetting on divorce Rhys Taylor, Barrister and Arbitrator, 30 Park Place, Cardiff and 36 Bedford Row, London Hilary Woodward, Honorary Research Associate, Cardiff University Rhys Taylor is a barrister and arbitrator who specialises in matrimonial finance and family property disputes. He is based at 30 Park Place in Cardiff and 36 Bedford Row in London. Rhys was awarded the 2015 Family Law Commentator of the Year award. He sits part-time as a legal chair of the Residential Property Tribunal (Wales). Hilary Woodward is an Associate of the Cardiff School of Law and Politics. She practised for many years as a family solicitor and mediator and has extensive experience of empirical research into family law, most recently on financial remedy and pension orders on divorce. This article is based on a paper presented to the Family Law Bar Association annual conference, November 2015. This article addresses the divergence of approach to the question of pension offsetting in financial remedy proceedings. Case-law offers little guidance on the practice of pension offsetting. In JS v RS [2015] EWHC 2921 (Fam) Sir Peter Singer noted at para [74]: ‘I am aware from my general reading that there is at present debate but as yet no conclusion on precisely this topic of arriving at an offsetting figure. I am not aware from my own knowledge nor have I researched what the competing methods might be. I am thus left in the unsatisfactory position where I must alight upon an amount which will necessarily be arbitrary …’ A recent study Pensions on Divorce 1 found that financial disclosure and explicit reasoning in relation to pensions was unclear or inadequate in the majority of cases with relevant pensions. Where pensions were considered at all, offsetting was by far the most common way of dealing with them but was somewhat of a Cinderella remedy, with more professional and academic interest being directed at pension sharing orders. Very few court files, however, contained any express reference to offsetting or its calculation and there was little agreement amongst family practitioners either as to how to apply or value the offset or whether expert assistance was needed. This narrow topic raises much wider questions: Why does the family justice system tolerate such divergence and uncertainty? Might it be appropriate for there to be some authoritative guidance from the courts in which the differences are adjudicated upon? Is there a need for more interdisciplinary 1 Pensions on Divorce: An empirical study, by H Woodward with M Sefton (Cardiff Law School, 2014, funded by the Nuffield Foundation) included a survey of 367 divorce court files with a financial remedy order, interviews with 32 family solicitors, meetings with seven district judges and assessment of the court data by a pension expert, available at http://orca.cf.ac.uk/56700 (last visited 21 October 2015). December [2015] Fam Law 1485 Articles

Transcript of Apples or pears? Pension offsetting on divorce...Apples or pears? Pension offsetting on divorce Rhys...

Apples or pears? Pension offsetting ondivorce

Rhys Taylor, Barrister and Arbitrator, 30 Park Place, Cardiff and 36Bedford Row, LondonHilary Woodward, Honorary Research Associate, Cardiff University

Rhys Taylor is abarrister and arbitratorwho specialises inmatrimonial finance andfamily propertydisputes. He is based at30 Park Place in Cardiffand 36 Bedford Row inLondon. Rhys wasawarded the 2015Family Law

Commentator of the Year award. He sitspart-time as a legal chair of the ResidentialProperty Tribunal (Wales).

Hilary Woodward is anAssociate of the CardiffSchool of Law andPolitics. She practisedfor many years as afamily solicitor andmediator and hasextensive experience ofempirical research intofamily law, mostrecently on financial

remedy and pension orders on divorce.

This article is based on a paper presented tothe Family Law Bar Association annualconference, November 2015.

This article addresses the divergence ofapproach to the question of pensionoffsetting in financial remedy proceedings.Case-law offers little guidance on thepractice of pension offsetting. In JS v RS[2015] EWHC 2921 (Fam) Sir Peter Singernoted at para [74]:

‘I am aware from my general readingthat there is at present debate but as yetno conclusion on precisely this topic ofarriving at an offsetting figure. I am notaware from my own knowledge norhave I researched what the competingmethods might be. I am thus left in theunsatisfactory position where I mustalight upon an amount which willnecessarily be arbitrary …’

A recent study Pensions on Divorce1 foundthat financial disclosure and explicitreasoning in relation to pensions wasunclear or inadequate in the majority ofcases with relevant pensions. Wherepensions were considered at all, offsettingwas by far the most common way of dealingwith them but was somewhat of aCinderella remedy, with more professionaland academic interest being directed atpension sharing orders. Very few court files,however, contained any express reference tooffsetting or its calculation and there waslittle agreement amongst family practitionerseither as to how to apply or value the offsetor whether expert assistance was needed.This narrow topic raises much widerquestions:

• Why does the family justice systemtolerate such divergence anduncertainty?

• Might it be appropriate for there to besome authoritative guidance from thecourts in which the differences areadjudicated upon?

• Is there a need for more interdisciplinary

1 Pensions on Divorce: An empirical study, by H Woodward with M Sefton (Cardiff Law School, 2014, funded by theNuffield Foundation) included a survey of 367 divorce court files with a financial remedy order, interviews with 32family solicitors, meetings with seven district judges and assessment of the court data by a pension expert, available athttp://orca.cf.ac.uk/56700 (last visited 21 October 2015).

December [2015] Fam Law 1485

Article

s

engagement between lawyers andexperts working in this field?

• How appropriate are the Duxbury2

assumptions in the context of pensioncalculations and generally?

• How appropriate is the use of the utilitydiscount when offsetting? The utilitydiscount may be ordered by the court oragreed by the parties to reflect theperceived advantage of holding cashnow rather than waiting for pensionbenefits in the future. This includes anelement of discounting for acceleratedreceipt and a more amorphous judgmentabout the value of cash over other formsof wealth holding.

BackgroundIf you are a family money practitioner, youwill have probably had the followingexperience. You are busily preparing for ahearing and when reading that bit of thepension report which deals with thequestion of offsetting, you think, ‘I’m surethe last pension report I read approachedthis differently.’ However, life is busy, this isyour single joint expert report and when isit ever proportionate to go into extendeddebate over the expert’s proposed offsettingmethodology? When it comes to offsetting,it is tempting for the family lawyer to hidebehind the discretionary powers of s 25 tojustify nearly any solution which can becobbled together.

We are hugely indebted to 14 leadingpension experts who took part in thisproject by answering three short mockpension offsetting problems (small, mediumand bigger money cases) and to nine ofthem for attending a meeting on1 September 2015 to discuss their differingapproaches as if they were Part 25 experts.This paper would not have been possible

without their invaluable and expertcontributions. All experts who took partsigned confidentiality declarations as a basisfor sharing their reports with others in thegroup. As one expert did not wish to sharehis report he was not privy to any of theother reports or to the details of themeeting. It also follows that this expert’sresults are not recorded in the tables below.Nothing in this paper is intended as acriticism of any individual expert.

Whilst the individual approaches of eachexpert remain anonymous, the experts whotook part are as follows (in alphabeticalorder): Stephen Bridges (Bridges UKActuarial Services Ltd), Paul Cobley (OakBarn Financial Planning), Ian Conlon (IWCActuarial Limited), Peter Crowley (WindsorActuarial Consultants), Dani Glover andJulian Whight (Smith & WilliamsonFinancial Services), Miles Hendy (FraserHeath Financial Management Ltd),Peter Moore and John Riley (BDM),Mark Penston (Bluesky Chartered FinancialPlanners), Kate Routlege and Jim Sylvester(Collins Actuaries), Geoffrey Wilson(Excalibur Associates), and Paul Windle(Actuaries for Lawyers).

The differing approaches to pensionoffsettingIn their excellent treatise, Pensions onDivorce (Sweet & Maxwell, 2nd edn, 2013)Hay, Hess and Lockett look at the differentways in which a pension may be offset.They state:

‘There are many ways of calculatingset-off . . . It is right to say, however,that none of them has been adopted bythe courts as a universal or even afavoured approach. The authors wouldsuggest that the reason for this is thatthe value of the set-off will changedepending upon the nature of the case

2 Duxbury is described in At A Glance 2015–2016 (Class Legal) at Table 14 (cited with approval by Ryder LJ in H v H(Financial Remedies) [2014] EWCA Civ 1523, [2015] 2 FLR 447). It states, ‘Duxbury calculators are based on aniterative computation, seeking the amount which if invested to achieve capital growth and income yield (both at assumedrates and after income tax on the yield and CGT on realised gains) could theoretically be drawn down in equalinflation-proofed instalments over a period (usually, but not always, the estimated actuarial life expectancy of therecipient) but would be completely exhausted at the end of the period. It is not, and never has been, an attempt toidentify the sum necessary to guarantee a particular level of expenditure … The assumptions include three key financialpredications – an average income yield of 3% pa, an average capital growth of 3.75% pa and average inflation of3% pa.’ This results in a real rate of return (ie after inflation) of 3.75%.

1486 December [2015] Fam Law

Article

s

and the reason for it. Consequentially, arigid approach would prevent the courtfrom utilising its full discretionarypowers in accordance with therequirements of the Matrimonial CausesAct 1973 . . . The authors would suggestthat there can be no ‘one size fits all’solution to the offsetting problem. Thefair and equitable offset will dependupon a number of factors that will varyfrom case to case.’

Hay et al describe seven approaches tooffsetting, without consideration of theutility discount.3 These include:

• An offset using only the CashEquivalent (CE). This is the simplesttechnique.

• A more complicated approach assumesan outcome based on the presentpension benefits (eg an equalisation ofpension income at a set point in thefuture) and discounts that outcome backto a current lump sum, based uponrealistic actuarial assumptions. (Thelump sum can then be invested withincome drawdown and amortisation ofthe lump sum to produce the requiredoutcome.)

• A costlier outcome calculates the sumrequired to purchase an annuity at arate equal to the pension member’spension income. This approach providescertainty of outcome but at a price.

• The Duxbury algorithm might also beused to calculate the sum required but isnot based on actuarial assumptions.This produces the lowest figure for aparty seeking a pension offset and willbe commented upon in more detailbelow.

Upon reviewing the experts’ reports for thisexercise, we have found evidence of all theapproaches outlined above but notnecessarily the reasons for them. Rather, thediffering approaches appear to reflect aparticular expert’s personal approach andsubjective interpretation of the questionasked and not necessarily the facts of a

particular case. This is not to criticise any ofthe individual experts, all of whom haveproduced admirably reasoned reports whenviewed in isolation. However, when placedside by side, a divergent picture emergeswith experts defaulting to a variety ofdiffering assumptions and interpretations ofwhat they are being asked to do. Thisresults in a wide range of outcome ofoffsetting calculation for no compellingreason, as detailed below.

Computation or distribution?A difference of emphasis emerges betweenthe present authors and Hay et al. The latterdefend the status quo upon the basis thatthis allows for the maximum discretionpursuant to s 25 of the Matrimonial CausesAct 1973. However, as was made plain inCharman v Charman (No 4) [2007] EWCACiv 503, [2007] 1 FLR 1246, an applicationfor a financial remedy involves a two-stageprocess: first, computation and second,distribution. Whilst we accept that there isnot an airtight seal between each stage, thecomputational stage has less to do with s 25discretion and far more to do with simplyidentifying the correct figures. We suggestthat placing a present capital figure on aright to receive a future income is moreclosely related to the computational stagethan it is to distribution. Although withinthe computational stage, experts may alsobe applying professional judgment in thefigures and assumptions which they adopt,this is not to be confused with the s 25discretion.

Hay et al have eloquently systematised adivergent picture. Our respectful view is thatsuch systematisation is something of a figleaf for a situation which has arisenunchecked by the courts over many years ofpractice. The discretionary nature ofoutcome in financial remedy proceedingsmight act as something of a defence againstaccusations of woolly thinking, when anyparticular outcome becomes subject tochallenge.

3 Pensions on Divorce (Sweet & Maxwell, 2dn edn, 2013), esp. 13–008, and 13–010–13–018.

December [2015] Fam Law 1487

Article

s

The offsetting scenariosThe aim in each of the following scenarioswas to achieve a broadly equal division ofthe assets between husband and wife. Weinvited the experts to assume they werePart 25 experts dealing with financialremedy proceedings in May 2015. Inparticular, the instructions stated: ‘Make theassumption (which we accept does notpertain in real life) that the CE and/or“figure invested” represents the actuarialvalue of the pension and does not requirefurther consideration to ascertain its “true”value.4 This exercise is designed to focusonly on approaches to any offsetting oncethe actuarial value of the pension has beencalculated. If, however, you wish tocomment upon the respectiveadvantages/disadvantages of using the CErather than an actuarially calculated figure,please do feel free to comment. Further,ignore the incidence of any basic statepensions, assume there are no AdditionalState Pensions and do not make anyfractional adjustments for how long theparties have been married.’

Case A: Small money scenarioH is aged 42 and has a stakeholder pensionwith a CE of £100k, accumulated during the15 year marriage. W is 40 and has nopension of her own. They have two youngchildren. The net equity in the family homeis £100k. Their net incomes are similar;both pay basic rate tax. They both want tooffset in principle, so that W can retain asmuch of the current capital as possible andforego any pension sharing claim but theycannot agree on the amount.

Case B: Middle money scenarioH is aged 54 and a higher rate (40%) taxpayer. He contributes to a final salaryrelated pension scheme with a CE of £500k(which will allow H to transfer the fundinto a personal pension and permits internalpension sharing orders). W is also aged 54and is in part-time employment earning justbelow the minimum tax threshold, with no

pension of her own. They have been marriedfor 20 years and have one son at university.The net equity in the family home is £500k.Both wish to offset all of W’s pension‘entitlement’ against H’s share in the home:but what is the value of the offset?

Case C: Bigger money scenarioH is aged 56 and has £1.5m invested in aSIPP, which owns commercial property fromwhich he conducts his business. H hassuccessfully applied under the IndividualProtection 2014 provisions. W is aged 58.Both are higher rate tax (tipping into 45%rate) payers but W has no pension of herown (she is a barrister!). They have beenmarried for 30 years with two children, nowindependent. Their current net assets are£3m, half representing the net equity in theFMH and half in an investment portfolio.They are considering all options and wish toknow how they would value a pensionoffset (so that W receives as much currentcapital as possible in exchange forsurrendering any pension entitlement shemay have).

The resultsThe differing approaches in evidence madefor a fascinating but confused picture. Wereadily accept that these are only mockscenarios based upon necessarily sparse factsand that some matters might have beentidied up by a ‘real life’ letter of instructionwith more relevant information. We alsoacknowledge that complaint was made by anumber of experts that we had not providedenough information generally andparticularly for the middle money, definedbenefit pension case. This resulted in someexperts making their own individualassumptions about the benefits which wouldbe received in retirement. We accept that, asauthors of the questions, we are at leastpartly responsible for the divergence ofopinion. The range of assumptions, evenwith the imperfections of this exercise, arearresting nevertheless. We have confinedourselves to the headline issues whichemerge from the reports.

4 The invitation to assume the CE to be a reflection of true value elicited a number of critical comments from experts andmany did not apply this assumption.

1488 December [2015] Fam Law

Article

s

• The starting points employed by theexperts in their offsetting reports varied.For example, in the small moneyscenario some experts started by usingthe CE whereas others wished todiscount the receipt of future pensionincome back to a present capital sum.

• In those reports (whether small,medium, or bigger money) where anactuarial calculation was made todiscount a future income stream back toa present capital sum, different expertsapplied differing investment andmortality assumptions.

• All experts made an adjustment for taxbut the manner in which they did sodiffered, for example, in whether theyincluded the personal allowance in theircalculations or allowed for the tax freepension lump sum.

• Some expressly opposed the principle ofa utility discount; some applied anautomatic discount of between 1%–2%per annum to retirement; some applied agreater discount the greater the need forcash; others acknowledged that a utilitydiscount might be appropriate but didnot include it in their reports and left itto the parties or court to decide.

• Different retirement ages were assumed.

• Some made adjustments for the modestage differentials in the scenarios whereasothers thought this to beinconsequential.

• Some made adjustments to the CE toprovide for the costs of administrationand transfer whereas others did not.

• Some gave a single, definitive answer;

others a range of possible answerscommenting that there is no oneactuarially correct way to undertake thisexercise; others gave a range andaveraged the range to arrive at a singleresult.5

• Some allowed for ancillary benefits,such as a spouse’s pension, as part ofthe overall valuation exercise; othersignored them.

• In the bigger money case there was adifference in treatment of LifetimeAllowance Limits.

• Different experts used differentterminology.

The results are shown in tabular formbelow.6 One independent financial advisor(IFA) declined to answer the medium andbigger money scenarios, expressing the viewthat it was outside its competence to do so.The tables show the range of answersprovided to offset the whole value of apension (a figure which is not accessiblepresently and will be taxed upon receipt)against the non-pension assets. In each case,the wife’s cash offset amount is 50% of theamount shown.• The small money case had a pension CE

of £100,000, with offset suggestionsranging between £60,000 and £96,900.

• The medium money case had a pensionCE of £500,000, with offset suggestionsranging between £290,000 and£798,000.7

• The bigger money case had a pensionCE of £1,500,000, with offsetsuggestions ranging between £886,000and £1,425,000.

5 Readers are referred to PD 25D 9.1(g) which requires a summary of the range of opinion to be expressed.6 We are grateful to Stephen Bridges for allowing us to use the tables which he prepared for the experts’ meeting. ‘FIA’ is

Fellow of the Institute of Actuaries. Two experts (FIA3/4) collaborated to provide answers on two alternative bases.7 The fact that some offset calculations came in higher than the CE underlines just how misleading the CE may be in the

context of a defined benefit pension.

December [2015] Fam Law 1489

Article

s

Expert pension offset valuations: Small money case

Expert pension offset valuations: Mid-money case

Expert pension offset valuations: Bigger money case

1490 December [2015] Fam Law

Article

s

The Cash Equivalent valuationIt is submitted that the Cash Equivalent(CE) value occupies an unwarranted totemicposition. This is hardly a novel point tomake but one which is frequentlyoverlooked in practice. The CE is the sumone pension arrangement will transfer to analternative pension agreement to extinguishtheir liability under the scheme.8 The CE isgrafted into matrimonial law by virtue ofthe Pensions on Divorce etc (Provision ofInformation) Regulations 2000, whichrequire pension arrangements to provide thisinformation on divorce. It is merely astarting point in terms of information. Whilethe CE may provide a rough workingestimation for standard moneypurchase/defined contribution schemes, it isoften wholly misleading as to the true valueof a defined benefit scheme.9 Differentscheme actuaries apply different investmentassumptions and therefore the ‘true’ value ofdefined benefit pensions with similar CEscan be wildly different; further, the ‘true’value may be significantly higher than theCE.

A neutered adversarial system?We are not aware of any reported financialremedy case in which each side has beenpermitted to appoint a pension expert toallow for rival actuarial approaches to betested.10 Whilst it would rarely beproportionate in an individual case toappoint competing pension experts (and inthe truly big money cases pensions are oftena small component part of a much biggerpicture), the result is that the family courtshave never been asked to adjudicate on rivalapproaches to be found in differing reports,so far as we are aware.

In an analogous (Duxbury) context we noteH v H (Financial Remedies) [2014] EWCACiv 1523, [2015] 2 FLR 447, at para [27]with Ryder LJ’s commendation of the ‘wisewarning’ given by Holman J in F v F(Duxbury Calculation) [1996] 1 FLR 833 at849 cautioning against ‘… any standardpractice arising of experts being instructedon the point given that the assumed rate ofreturn was only a starting point forconsideration of the s 25 factors in the Act’.However, the result of this type of warningis that no cases ever seem to feature anadjudication of competing actuarialassumptions. Further, and with the greatestof respect, the warning in F v F againstexpert evidence appears to have contributedsomething to the fudge as betweencomputation and distribution issues.

This is in contrast to (an admittedlyimperfect comparison) leaseholdenfranchisement valuations (which give acurrent value of the freeholder’s futurereversionary interest). First-tier tribunalsroutinely allow for two experts and theassumptions behind competing expertevidence have been adjudicated upon in theUpper Tribunal in high profile test cases.11

Such cases provide a judicially-adjudicatedgoing rate which is then only departed fromfor good reason supported by expertevidence on particular facts.

Reflections on experts’ meetingAs with the detail of the reports, it is simplynot possible here to deal verbatim with allof the detail which was discussed at themeeting with the nine experts. The followingmust necessarily reflect only a summary ofthose parts which appear most noteworthy.

8 The Occupational Pension Schemes (Transfer Values) Regulations 1996 and The Occupational Pension Schemes (TransferValues)(Amendment) Regulations 2008.

9 One expert made his point graphically by giving two different answers to our middle money scenario. The first answerassumed an NHS pension, whose value was dependent upon Government Actuary Department factors, and the secondassumed a private sector final salary scheme where the CE was strongly market related. The difference in likely pensionincome ranged from £14,492 pa for the NHS pension scheme and £7,628 pa for the private scheme, notwithstandingthat both have the same CE of £500,000. When deferring these future benefits back to a present capital sum theresultant difference in value between two pensions, each with the same CE, was about £290,000.

10 In Martin-Dye v Martin-Dye [2006] EWCA Civ 858, [2006] 2 FLR 901 the parties were before the Court of Appealwithout a pension report. At para [70] Thorpe LJ recommended the use of a Single Joint Expert ‘where proportionate’.

11 For example, Earl of Cadogan v Sportelli EWLands [2006] LRA_50_2005, [2007] 1 EGLR 153http://www.bailii.org/ew/cases/EWLands/2006/LRA_50_2005.html

December [2015] Fam Law 1491

Article

s

Respective skills of the expertsThere was a good understanding, asbetween the actuaries and the IFAs, as to therange of their respective expertise. The IFAspresent accepted that there were somecalculations which necessarily involvedactuarial assumptions and were outside thescope of their particular skill set. Likewise,the actuaries present accepted that therewere certain pieces of work, such asconsidering the implications of LifetimeAllowance issues, which were far betterdealt with by IFAs.

However, there remained a tension. SomeIFAs asserted their competence to undertakecomplex pension reports whereas other IFAsdrew the line elsewhere and deferred toactuaries when actuarial assumptions wererequired. The present writers are clearly notin any position to adjudicate upon where todraw the line. We merely note that theredoes appear to be a difference of opinion asto where the line should be drawn and thatfamily lawyers are ill-equipped to determinethat issue.

The letter of instructionThe importance of the letter of instructionwas strongly emphasised. The experts wereat one in highlighting the frequentdeficiencies, unnecessary detail andover-complication presented in letters ofinstruction. The balance of preference in thegroup was for shorter letters which statedbasic information to include contact details,ages, realistically considered retirement dates(for which the parties may first wish to seean IFA) and letters of authority for theexpert to then approach the relevant pensionproviders to ask for the further detailrequired.

A helpful model of working, utilised by twoof the experts, involved an IFA seeing theparties, understanding their pension andfinancial situation, highlighting the issues atstake and then drafting a letter ofinstruction to an actuary, jointly with theparties. This ensured that the right level ofdetail was relayed to the actuary and thatthe questions were properly framed, withthe parties knowing the implications of what

they were asking and the experts’complementary skills being deployed tomaximum effect for the benefit of theparties. In practice, such an approach maynot be possible where the parties are unableto work together.

The starting pointThe group discussed the starting point foran offsetting calculation and whether anotional ‘equalisation of income’ pensionsharing order should be used as thebenchmark for the value to be offset. Inpractice, it is often the case that theoffsetting option is considered alongsidepension sharing in a composite report andso this approach may have a practicalmagnetism.

Many points discussed throughout themeeting hinged on whether this was the beststarting point or whether the starting pointwas a straight offset against 50% ofnon-pension capital. They could producevery different results. When only consideringoffsetting, the group, on balance, did notconsider that the notional ‘equalisation ofincome’ pension sharing assumption wasnecessarily the most straightforward startingpoint. However, all were agreed that itwould depend upon how the instructionswere framed.

The use of the CE as basis forvaluationThe group accepted that in a standarddefined contribution pension case, it wouldbe acceptable to use the CE as the startingpoint for calculations. Whilst this would notbe as actuarially accurate as deferring futurebenefits back to a present capital sum, thedifference between the two figures was notconsidered to be so material that itnecessarily justified a more complicated (andexpensive) starting point.

The group was broadly in agreement that a20% discount would usually be appropriatefor tax and would often represent the mainoffsetting calculation required in definedcontribution cases. Some were anxious toemphasise that the marginal rate of incometax for the party retaining the pension

1492 December [2015] Fam Law

Article

s

benefits was the necessary calculation tomake and that the standard rate of incometax when the pension is accessed could notalways be assumed.

However, it was noted that this only appliedto ‘standard’ situations and there may be anumber of situations where a definedcontribution pension scheme hadnon-standard features which would makethe CE an inappropriate starting point. Itwas agreed that lawyers were often illequipped to ‘look under the bonnet’ to see ifnon-standard features were present. It wasagreed that a quick review by an IFA, shortof a ‘full’ instruction, might be the way fora lawyer reasonably to determine whetherthe CE was acceptable to use.

Non-standard features which werehighlighted included: the presence ofguaranteed annuity rates or investmentreturns (a ‘hybrid’ pension which wasessentially a defined contribution scheme butwith some defined benefit characteristics),wildly differing ages, health issues, largesurrender penalties, non-standardinvestments and high value pensions whichmay have tax implications and/or LifetimeAllowance issues.

The group was also in agreement that theCE was not an appropriate starting point indefined benefit pension cases, for thereasons given above.

DuxburyUsing the Duxbury assumptions is includedby Hay et al as one possible way ofoffsetting a pension. It is of note that itprovides, by some considerable distance, thelowest cash offset figure when compared tothe other approaches.12

The group had strong views about Duxbury.In terms of its ability to produce a wholelife income stream, it was referred to as ‘not

fit for purpose’ with the experts being‘horrified’ by the courts’ continuingadherence to it. Experts were confused by itsprovenance and how it was reviewed ormight be challenged. The present authorsacknowledge that the proponents ofDuxbury do not seek to defend itsassumptions on the basis that it can be usedto provide a guaranteed income stream inplace of a joint lives’ maintenance order.13

The group appeared broadly in agreementwith the following:

• For pension assets prior to retirement, a2%–2.5% real rate of return (ie afterinflation) was realistic (cf. Duxbury at2.25% in year one and 3.75%thereafter14). It was noted that Duxburydoes not take into account charges madeto defined contribution schemes, whichwere assumed at up to 1%, wideningstill further the gap with the Duxburyoutcome.

• For assets not held within a pension, areal rate of return of 1.5% to 2% (ieafter inflation) was realistic.

• Post retirement it was considered thatinvestor risk profile changes and thatmany would wish to become risk free atthis stage and therefore experts assumean annuity purchase or other low riskstrategy when assessing likely investmentreturns.

Over the medium to long term, suchassumptions will produce an outcomesignificantly different from the Duxburyalgorithm. Of course, Duxbury is a ‘tooland not a rule’ and there are eloquentdefences to be made in its favour.15

However, the ‘only a tool’ mantra may ringhollow when most practitioners appearunwilling to seek the instruction of aPart 25 expert to challenge the assumptions.Interestingly, Mostyn J in JL v SL (No 3)[2015] EWHC 555 (Fam), [2015] 2 FLR1220 defended the Duxbury assumptions, inpart, upon the basis, ‘No evidence was

12 See Hay et al at 13–018.13 JL v SL (No 3) [2015] EWHC 555 (Fam) per Mostyn J; and Lewis Marks QC, ‘An Alternative view of Duxbury: A

Reply’ [2010] Fam Law 614.14 The reduction in year 1 is due to ‘credit crunch’ factors and is now in its seventh year, Capitalise, Version 20, and At A

Glance 2015–2016 Class Publishing.15 See fn 13.

December [2015] Fam Law 1493

Article

s

adduced as to why the views of theDuxbury committee should not befollowed …’16 He further defended theassumptions based upon the history of howthe market has performed.17

One of the defences of Duxbury is thatearned income has risks attached to it andthat the dampened assumptions fairly reflectthe risks associated with life’s vicissitudes. Incontrast, as noted by Wilson LJ (as he thenwas) in Vaughan v Vaughan [2007] EWCACiv 1085, [2008] 1 FLR 1108, at para [25],pension capital ‘… unlike any attemptedcapitalisation of earning capacity . . . doesnot depend on the application of futureeffort but, subject only to market and othervagaries, is in the bag …’. The difference inthe quality of income receipt as betweenearned income and pensionincome/drawdown may provide ajustification for not applying Duxburyassumptions in the pension offsettingcontext, in the authors’ respectful view.

The present authors do not presume to bequalified to adjudicate upon the rival viewsof a congregation of nine experts and theDuxbury proponents. However, they are atodds with one another and it would be aworthwhile exercise to see such positionsadjudicated upon by reference to evidence.

Utility discountThe discussion concerning the utilitydiscount was illuminating. There was a realtension as to whether this should beregarded as an expert computational issueor a court discretionary issue. As alreadynoted, it can reflect both an element foraccelerated receipt and a value judgmentabout cash over other forms of wealthholding. In many cases the time value ofmoney is not discounted back to the presentdate using any particular mathematical

assumptions, rather an instinctive ‘gutreaction’ as to a fair discount is made.18

The experts said that they were able tocrunch numbers if instructions invited themto do so, but none expressed any greatenthusiasm for the process. One stated thatthe utility discount starts with the expert butfinishes with the court; this is in contrast tocalculations relating to pension sharing,which the experts respectfully expected thecourt to accept.

The experts variously referred to it as beinga ‘finger in the wind’, ‘fudge factor’ and‘stick to beat the wife with’. Somequestioned both the principle and themechanism of discounting for utility.Confusion was expressed as to itsprovenance; guidance from the court as towhether and, if so, how it should beapplied, was considered needed in this area.

The notional benefit of holding currentassets rather than pension assets was alsoquestioned when it is often the case that theoffset is being required to assist a care-giverto retain the family home. The capitalcontained therein is not usually available fordiscretionary spending and there has been aconstriction in the market for re-mortgagespost 2008. It was also noted that equitiescan often outperform real property and alsothat the person with cash could be more atrisk than the person with the pension fund.

It was noted that some experts had providedgreater discounts for those claimants whowere perceived to be in greater need ofcurrent assets. Whilst it was accepted thatthere was a certain harsh logic to this froma computational perspective, it was agreedthat it sat very uncomfortably with thecourt’s overall duty to be fair. Thisparticular point highlighted thecomputation/distribution questions verystarkly.

16 Paragraph [17], and see also DR v GR (Financial Remedy: Variation of Overseas Trust) [2013] EWHC 1196 (Fam),[2013] 2 FLR 1534, at para [57] ‘… Of course the Duxbury figure is only ever a guideline but departure from it must berationally and numerically justified by reference to clear evidence …’

17 Paragraph [14].18 Hay et al note at 13–010, ‘The [offsetting] exercise will almost always reflect a reduction in CE to reflect taxation and

‘utility’ of realisable assets. The reduction is usually not more than 40 per cent and not less than 30 per cent (there is noauthority for this proposition; it merely reflects the authors’ experience and anecdotal reports).’

1494 December [2015] Fam Law

Article

s

Despite the reservations and criticism of thisapproach a minority view was expressedthat the parties’ financial situation and thepurpose of the offset remained relevant.

No going rate or approach was agreed. Theview of the group, however, was that theutility discount has an unhelpful prominencein the offsetting process.

April 2015 pension freedomsIt was considered that pension freedomsmay result in less need for offsettingcalculations. The reform allows for theconversion of future income into presentcapital, albeit with tax and financialplanning implications. Given the advantagesthat this may bring, the group was of theopinion that parties should be rationallymore enthusiastic about holding or receivingpension assets as part of a settlement. Theincreased flexibility and utility of pensionsfurther fed the question concerning therelevance of the utility discount.

The group took issue with some of theobservations of Nicholas Francis QC (sittingas a Deputy High Court Judge) in the caseof SJ v RA [2014] EWHC 4054 (Fam), atpara [83] where he opined that pensionfreedoms meant that pensions ‘… arevirtually to be treated as bank accounts forpeople over 55’. The deputy judge alsosuggested that courts are likely to be ‘mostreluctant in future’ to allow for actuarialreports on the issue of how to effect equalityof income outcomes. Further, ‘I suspect thatannuities will, in the overwhelming majorityof cases, become a thing of the past.’

The fundamental problem with treatingpensions as ‘bank accounts’ is knowingwhat, precisely, is the value contained withinthe bank account. While the group (as notedabove) were content in standard definedcontribution scenarios to use the CE as thestarting point, defined benefit schemesrequired a proper assessment of value.Second, unlike a bank account, there may besignificant tax issues depending on themanner in which value is drawn from apension.

It is clearly a matter for the courts as towhether the approach of ‘equalisation ofincome’ or ‘equal division of pensioncapital’ is deployed at the distribution stageof analysis.19 This does not, in ourrespectful opinion, remove the need forpension experts to be instructed, particularlyin the context of defined benefit schemes or‘off piste’ defined contribution schemes. Thejudgment in SJ v RA fails to make anydistinction between defined contribution anddefined benefit schemes. We respectfullysuggest that such comments, emanatingfrom the persuasive judgment of an eminentmatrimonial finance practitioner sitting as adeputy high court judge, may encouragecourts to make wrong decisions whenconsidering Part 25 applications inparticular circumstances.20

The IFAs present did not agree thatannuities were necessarily a thing of the pastand examples were cited of recent clientschoosing to eschew ‘pension freedoms’ (andthe lure of the Lamborghini) in favour ofthe security of an annuity.21

Following the experts’ meeting on1 September 2015, a sub-group of the

19 The practice of excluding pension capital acquired prior to cohabitation in ‘middle money’ cases is, respectfully, specious.Middle money cases are still, overwhelmingly, ‘needs’ cases and the exclusion of pre-acquired assets should only beconsidered in cases where assets exceed needs: N v F (Financial Orders: Pre-Acquired Wealth) [2011] EWHC 586 (Fam),[2011] 2 FLR 533, at para [15]. See also Harris v Harris [2001] 1 FCR 68, at para [26] where Thorpe LJ stated that thefractional approach, ‘… is an ingenious submission but one that, if followed with any sort of confidence might well leadto unrealistic results’. It is a curious accident of law reporting that such an important statement of principle neverreceived wider publicity with a FLR citation.

20 There is a particularly egregious example posted by Magnus Mill (Alexiou Fisher Phillips) in the LinkedIn discussiongroup ‘BDM Pension on Divorce Forum’. Registration on LinkedIn is required before viewing this pagehttps://www.linkedin.com/grp/post/4992123–6035020415552802816?trk=groups-post-b-title (last visited4 September 2015) where it is reported that a High Court Judge refused permission for an actuary where there was adefined benefit scheme with a CE of £2.6m and a DC pension with a CE of £1.5m. Pension sharing was not availableand the court, in refusing the application, opined that counsel’s submissions would be sufficient to deal with the issue.

21 Anticipating the April 2015 reforms the then pension minister, Steve Webb, made the memorable and ill-advised remarkthat people might choose to buy a Lamborghini with their pension pot and then live off their state pension. There areanecdotal reports that one individual cashed in his pension to purchase an old Routemaster bus!

December [2015] Fam Law 1495

Article

s

experts kindly drew up a glossary of agreedcommonly used terms. This is to be found inthe Appendix to this paper.

Key factorsWe suggest that there are six key factorswhich should be borne in mind whenconsidering different approaches tooffsetting:22

• Is the pension a defined contribution ordefined benefit scheme? The former hasa well-defined value, the latter does not.

• In the case of a defined contributionscheme, is it more appropriate to lookat what it would cost to replace thecurrent fund or the future benefits outof non-pension wealth? The answer willdepend mainly on the parties’ currentages and size of funds involved.

• In the case of a defined benefit schemeor the replacement of benefits under adefined contribution scheme, should theoffset value be the value to the memberof that pension, or the value to thenon-member of having an equivalentpension or share of pension?

• What is the fair rate to use fordiscounting to present values(recognising that there may bedifferences in the discount ratesappropriate to the pre-retirement andpost-retirement periods)?

• What account should be taken for tax,before and after retirement?

• What account, if any, should be takenfor the utility of holding cash?

ConclusionWhilst the broad s 25 brush is entirelyappropriate when considering assetdistribution on divorce, it is another matteraltogether to arrive at an arbitrary figure,

for want of evidence or lack of appreciationof the issues in play.23 The lack of explicitreasoning for pension offsetting outcomes,identified in the Nuffield report, Pensions onDivorce, is not a cause of these issues butmay be a symptom. How much morethought would be taken if lawyers wererequired to record the rationale for theiroffsetting decision?

To date, it has not been possible to arrive ata working formula which might be appliedin the valuation of pension offsets.24 Furtherinterdisciplinary discussion between lawyers,actuaries and IFAs of the above key factorsis needed, to achieve better mutualunderstanding and consistency, and grasp ofthe interrelationship between pensionoffsetting assumptions and Duxburycalculations. Interdisciplinary forums, sosuccessful in the child law context, are amodel which financial remedy lawyers couldconsider.

In addition to further interdisciplinarydiscussion, both lawyers and judges mightconsider whether some adjudication ofpension issues would be appropriate. This isnot a call to open the floodgates but for afew test cases to be judiciously selected toprovide some clarity in this confused andneglected corner of the law.

We are grateful to David Salter for assistingus generally and, in particular, for chairingthe meeting of experts on the1 September 2015. We are grateful to SirPeter Singer, His Honour Judge EdwardHess, Professor Gillian Douglas of CardiffUniversity and Andrzej Bojarski of 36Bedford Row for considering a draft of thispaper. We are grateful to Pranjal Shrotri of36 Bedford Row for assisting on the1 September 2015 and in particular for herexcellent note of proceedings. The mistakesremain our own.

22 We are grateful to Stephen Bridges for his helpful contributions here.23 Note NG v KR (Pre-Nuptial Contract) [2008] EWHC 1532 (Fam), [2009] 1 FLR 1478, at para [135] per Baron J,

declining to accept the argument that decisions in the financial remedy context are arbitrary and may therefore breachArt 1 of Protocol 1 of the European Convention of Human Rights etc. 1950. See also S Choudhry and J Herring,European Human Rights and Family Law (Hart Publishing, 2010), Ch 10.

24 David Burrows attempted to do so at [1999] Fam Law 556 ‘Pensions – How Much to Offset?’ This pre-White,pre-pension sharing order era analysis makes for a fascinating retrospective as to how far the law in this area haschanged in a relatively short period of time. In light of information gleaned in this exercise, we consider that DavidBurrows’ proposed discount ‘to compensate for … the accelerated benefit to the wife of offsetting the pension fundagainst the couple’s liquid capital’ may be too high at 2%–3% pa. Burrows himself describes it as ‘entirely arbitrary’.

1496 December [2015] Fam Law

Article

s

Appendix

Glossary of Terms/Abbreviations usedin pension offsetting reports

Abbreviation Stands for Meaning/Comments

CE/CETV/CEV/CEB Cash Equivalent (Transfer) Value (ofBenefits)

DB Defined Benefit pension, akaFinal-Salary or salary-related

DC Defined Contribution pension, akaMoney-Purchase

PLA Purchased Life annuity A ‘non-pension’ annuity purchased frompersonal funds.

SPA State Pension Age

UV Utility Value Value allowing for the lower attractiveness ofpension benefits due to deferment of lumpsums and pensions.

PCA Pension Commencement Age Age derived from instructions, or pensionanalysis.

LA, IP Lifetime Allowance, IndividualProtection

HMRC definitions indicating restrictions onmaximum pensions, with tax implications.

FV, FAV Fair (Actuarial) Value Value of future pension benefits discountedto present value using specified actuarialassumptions, (sometimes) the value of a DCfund reasonably equivalent to a specified DBpension.

MCCV Market Consistent Capital Values Value of future pensions using actuarialassumptions consistent with present financialmarket conditions.

OEV Offset Equivalent Value Value of future pensions using actuarialassumptions consistent with values of otherimmediately-realisable capital assets.

LV Liquidation Value Value if pension realised using new pensionfreedoms – for DB pensions, assumesimmediate transfer to a DC, then realisation.

NRA, NPA Normal Retirement (Pension) Age As defined in pension scheme rules, normallyearliest age at which DB pension can bestarted without early retirement discount.

FOS Financial Ombudsman’s Service Who publish actuarial assumptions for use inpension mis-selling calculations.

NISA New Individual Savings Account New form of ISA saving account

December [2015] Fam Law 1497

Article

s