Following the much anticipated decision of the Court of Appeal in Swift v Carpenter John Ross QC and Thomas Yarrow provide a comprehensive analysis of the difficulties accommodation claims present for the courts, the detail of the decision in Swift (including explaining the maths behind the formula) and the practical implications of the decision. The authors consider scenarios that are likely to cause complexities for the formula in Swift and propose ways to approach difficulties such as short life expectancies and cases where the claimant would have bought a property at some point in the future anyway.
The question of how best to approach properly compensating a victim of a tort, which results in the need for alternative accommodation which can be adapted to meet his/her special needs consequent on the injuries sustained by reason of the tort, has long troubled the Courts.
An account of the differing paths the Courts have taken over the years to address this problem can be found in McGregor on Damages 20th Edition. In essence the varying solutions have been: (i) George v Pinnock (1973) the additional cost of providing special accommodation over the cost of his existing accommodation which was now wholly unsuitable (if supported by medical evidence) was established as the recoverable loss; (ii) Moriarty v McCarthy (1978) and Roberts v Johnstone (1989) the amount awarded was the interest upon the difference in value between the claimant’s present home and the new or adapted accommodation which was needed. The reasoning was that to allow the full value of the alternative property would allow the claimant, or his estate, to acquire the extra value. Similarly, as regards adaptation costs, to allow the claimant to recover the full cost of these might result in an over-payment because they might enhance the value of the alternative property – in which case these costs are recoverable only to the extent that they do not enhance the value of that alternative property (of course, in some instances the installation of disability adaptations may reduce the value of the alternative property as and when it comes to be sold). This was done in Willett v North Bedfordshire Health Authority (1993).
Initially the Courts awarded interest on the differential capital costs at a rate which would allow the claimant to borrow to pay for the required accommodation and/or adaptation costs. This came to an end with the Court of Appeal decision in Roberts v Johnstone. The concern of that Court was that if the claimant was young with a long life expectation the resulting award under this head of claim could exceed the capital cost of the acquisition itself. In order to meet this “risk” the Court of Appeal went to the opposite end of the calculation scale and resolved that the award of interest should be only 2% – as contended for by the defendant insurer. The consequence was that the recoverable sum was often only 30% of the capital cost of acquiring the alternative accommodation. Matters improved for claimants following the House of Lords decision in Wells v Wells (1997) when a 3% discount rate was approved. If the claimant was a young person, say of 25, then the recoverable sum would often result in an award of 85% or so of the capital outlay. This sum fell to about 70% when the Lord Chancellor reduced the discount rate to 2.5%.
The real problem with the Roberts v Johnstone formula lay, however, in the realm of a catastrophically injured claimant who had only a short life expectation, as occurred in the case of Oxborrow v West Sussex Hospitals NHS Trust (2012), where the claimant (injured at birth) was only expected to live to the age of 21. There was an undoubted need for a specially adapted home but the approved formula would provide on 30% of the capital needed to acquire the appropriate accommodation. Some judges of the Court of Appeal began to observe that the assumption that the claimant will be able to fund the capital acquisition out of the sums awarded could not stand in the face of increasing property prices, when general damages awards were not increasing at a like rate.
The time came when Judges found themselves making nil awards for special accommodation claims when the discount rate descended into negative rates – see JR v Sheffield Teaching Hospitals NHS Foundation Trust (2017). The trial judge in Swift v Carpenter faced the same problem. In LP v Wye Valley NHS TRust (2018) the court effectively treated the formula as having no application when discount rates were negative.
Hence the need for the Court of Appeal to undertake a full review of the law on this issue. It is to be noted that the Law Commission examined it and found it too difficult to formulate an acceptable solution – see Law Commission No.262 (1999).
What is the solution that emerges from Court of Appeal judgments in Swift v Carpenter?
Defendant insurers have always been primarily concerned to ensure that the claimant should not be better off by reason of an award in respect of an accommodation claim than he/she would have been had the injury not been inflicted – the no windfall argument. As can be seen, this was the defendant’s submission in Roberts v Johnstone and it remained its submission before the Court of Appeal in Swift. This explains the concentration of effort in the judgments to respond to this argument and to endeavour to meet it. The solution proffered by the Court of Appeal is to produce a formula that will be likely to meet the defendant’s concern that at the end of life the claimant’s estate should not benefit from a windfall in the form of the capital value of the acquisition of the alternative adapted property acquired by the claimant to meet his/her tortuously induced disability needs. The solution is to deploy an actuarial approach to the calculation of the value of the reversionary interest which will arise at the end of the claimant’s anticipated life (and which will therefore accrue to his estate).
Reversionary interests: This may be a term of relative unfamiliarity to some practitioners in the field of personal injury and to explain the logic of the judgment necessitates a quick detour into the world of trusts.
The most simple example of a trust with a reversionary interest would be as follows: a property owner with a spouse and children decides in his/her Will to grant the surviving spouse a ‘life interest’ in a property and their children a ‘reversionary interest’. The property will, for all intents and purposes, belong to the surviving spouse until he/she dies, and then pass automatically to their children.
Both the life interest and the reversionary interest in the property will have an intrinsic value which can be realised in one of two ways. First, if during the surviving spouse’s lifetime the beneficiaries all agree that the trust should be wound-up and the property sold, the sales proceeds will need to be apportioned between the surviving spouse and the children in order to account for their respective life and reversionary interests.
The second example occurs where the reversioners decide they want to sell their interest without winding-up the trust. In such circumstances, a third party will buy them out of their beneficial interest in the property. The surviving spouse will keep his/her life interest in the marital home, but when s/he dies, the property will pass to the third party. For the children, this is a means of raising some capital on their home now, rather than waiting for their parent to die. For the purchaser, it is a form of long-term investment: they buy a property which they will not in fact own for a number of years at a price discounted to reflect the fact that their money is tied-up until the surviving spouse dies. It is a type of futures trading.
So how do the life interests and reversionary interests in such an example fall to be valued? In the first case – that of a winding-up – the interests are not being bought and sold, there is no market and consequently no market price. A formal valuation will typically be performed by an expert, which has as its aim obtaining a ‘fair and reasonable’ settlement as between the beneficiaries. In general terms, the closer the surviving spouse is to the end of life, the lower the value of their interest, and the smaller his/her share in the sales proceeds will be. In the second case – where the reversionary interest is sold to a third party – there are, of course, market transactions and consequently there does exist a market price for such products.
Returning to Swift, evidence was heard by the Court of Appeal from the director of the firm Foster and Cranfield, who attested to there being a very small market for such interests (four or five a year) to be sold, typically at auction. For a third-party purchaser, as with anyone making a long term investment they are looking for a good future yield on the capital they put in now. Mr Watson said that the auctions on these interests usually close at a price where the investor has effectively secured something in the region of 5-8% yield rate. His evidence, summarised at §158 of the Judgment was that eight out of the last ten cases handled by his firm had produced sales of the interests on the basis of a projected return between 6.2% and 7% per annum. One sale was based on a projected return of 5% and one was somewhat higher. His opinion was, therefore, that 6.6% was the appropriate return figure to use when calculating the reversionary interest price.
How does this work in practice in the current marketplace? Put simply, if an investor buys a reversionary interest for say £1,000,000 in a property which is held by a surviving spouse with a life interest and a life-expectancy of 10 years, the investor is looking at the end of the ten-year period to have a property which is worth the £1m plus ten years’ worth of 6.6% interest on that sum. Unfortunately for lawyers (who almost uniquely typically work in ‘simple’ interest) this calculation works with compound interest. The hoped-for return for the investor will be £1m x 1.066 x 1.066 … etc (i.e. ten times) which amounts to about £1.9m. This means, according to Mr Watson, when an investor attends an auction for a reversionary interest in a property with a value of around £1.9m currently held under a life interest predicted to last 10 years, it will be roughly £1m that s/he will be looking in order to secure the reversionary interest. That becomes the market price for a reversionary interest in the property with 10 years until it reverts. It follows that if the market value of the reversionary interest is £1m, the value of the life interest must be the remaining £900k.
The Court of Appeal considered that this method of splitting the value of a property between the life interest and reversionary interest could be cross-applied to personal injury accommodation claims. After hearing evidence, the Court considered that a property’s value, or the differential value between a property to be sold and the property to be acquired (as on the facts of Swift itself), should be notionally apportioned in this way. It should be emphasised that this creation of a trust for their purposes was purely ‘notional’ – i.e. the Court was not suggesting that the Defendant’s insurers themselves would actually purchase a reversionary interest in the property (although the future possibility of a market evolving is discussed below). They found that the appropriate method of valuation was to consider the ‘market’ and not the ‘fair and reasonable value’ (this, on the evidence, in practice providing lower figures).
The logic of the Court of Appeal was that if a Claimant’s damages award only covers the value of a notional life interest in the property, there is theoretically no windfall to his/her estate on death.
The formula chosen: So how is the market value obtained? The variables needed are (i) the value of the property (or share of the property) which is to be split into life/reversionary interests; (ii) the length of time of the life interest; (iii) the investor’s expected yield per annum. Variables (i) and (ii) will depend on the facts of the case, with (ii) being the expected life-expectancy of the Claimant from the date of damages award to death. Variable (iii) is a fixed number which was determined by the court to be 5%. Ultimately, this represented a cautious approach, and was made notwithstanding the evidence given by Mr Watson which had settled on 6.6%. The lower yield rate set by the Court reflected the fact that the market was very small and so pricing was not well-established, and it also brought it very nearly in line with the figure in Earl Cadogan and Another v Sportelli and Another (Lands Tribunal 15/9/06) which involved the analogous valuation of freehold reversions, and basis of awards in that context of 4.75%.
Once these three variables are in hand, one takes a simple compound interest formula and applies it backwards. As an example, using the Swift figures, one needs to work out what an investor would bid at auction on a £900,000 property in the knowledge ownership would revert to him/her after 45 years, if s/he wanted the value of his purchase to increase by 5% every year.
Or to put it quasi-algebraically: ? x 1.05 x 1.05… (45.43 times) = £900,000.
Multiplying by the same number 45.43 times can be written more simply with a “to the power of” function. The equation then becomes: ? x 1.0545.43 = £900,000.
Solving for ?: ? = £900,000 / 1.0545.43 = £98,087.27
Dividing by a number which is to the power of a positive number (as 1.05 is above) is the same as multiplying by the same number to the power of a negative number. So the above equation can also be written as: ? = £900,000 x 1.05 -45.43.
The ‘?’ above is the value of the reversionary interest. In the case of Swift it amounted to £98,087.27. When one considers dividing repeatedly by 1.05, it is easy to see how in cases of shorter life expectancies the reversionary interest value will be much closer to the full value of the property (you divide by 1.05 only a few times), while in cases of longer life expectancies the reversionary interest value will be much lower (you divide by 1.05 a large number of times). In mathematical terms, the reversionary interest value decays exponentially as life expectancy lengthens. Using 5% as a base, the ‘half-life’, or time it takes for the value to halve is just over 14 years, meaning the reversionary interest will be about half of the full value of the property or equity if the life expectancy is 14 years, about a quarter if it is 28 years, and about an eighth if it is 42 years.
In order to ascertain the damages award one needs to value the life interest, which will simply be the remainder of the value of the property (or differential in property values, as in Swift) once the reversionary interest is subtracted. In the case of Swift, that amounts to £900,000 – £98,087 = £802,812.73.
Practical implications: In practice, the Court of Appeal formula means a claimant will always find him/herself short of the money necessary to purchase the new accommodation. In the case of the claimant in Swift she was £98,000 short of being able to make the purchase. In theory, the Claimant could sell her reversionary interest at auction for that price to make up the shortfall. This would mean she would have the money to make the purchase, would live in the property for the rest of her life, but when she died the £900,000 equity would not pass to her estate but would revert to the investor who had bought the product at auction. In practice, in a case such as the Claimant’s where the shortfall was relatively modest (in comparison with other heads of damage), she would likely ‘borrow’ the money from, say, her PSLA award to make up the difference. In effect she would be buying her own reversionary interest on behalf of her estate. The full £900,000 would pass to her heirs upon her death, not because she had been overcompensated but because she had paid for it.
For Claimants with shorter life-expectancies there will be considerably more difficulty. They will have a much larger shortfall to make up – as above in cases where life expectancies are 14 years or less, this will be more than half of the property value. Nevertheless, the Court of Appeal augured that the result of their judgment might mean that a market in reversionary interests could grow considerably from the four or five transactions a year. If such a market did appear and if the Court of Appeal have broadly valued the reversionary interest correctly (two big ifs), and there did arise on the demand side a new cohort of investors eager to purchase products with relatively good yields (5% per annum) over relatively short terms (five, ten, fifteen years), then an injured claimant with a shorter life expectancy might well find that he/she could sell the reversionary interest to an investor, making up the shortfall needed to be secured in the property for the remainder of their life. If so, this could be quite an elegant solution to the problem of shorter life expectancies.
Complicated uninjured comparison scenarios: One area where the Court of Appeal’s judgment throws up more questions than answers is for types of cases where it is found that a Claimant would have purchased a property in the future if s/he had remained uninjured. How does that factor into the new reversionary interest analysis? Under the old Roberts v Johnstone approach this was relatively straightforward – at a certain point in life the Claimant was effectively becoming less deprived of capital to invest and so the multiplicand would reduce accordingly. The Court of Appeal judgment in Swift, however, does not provide an answer to the problem, despite referring to exactly this type of case in the annex to the Judgment as ‘Paradigm 1’.
Consider the scenario where a Claimant, “C”, is aged 10. They are given a life expectancy of 50 further years) to age 60. It is determined that he/she has an immediate need to purchase accommodation for £1m, but that if he/she had continued life uninjured a property worth £250,000 would have been purchased at the age of 35. How much should their damages award be? It seems there are three possible ways of dealing with this – all of which are compatible with the Swift Judgment.
In all three options, in order best to conceptualise the solutions, it is helpful to think of there being an investor willing to enter into partnership with C for the purchase. In reality there may not be, and C in all cases might make up the shortfall and purchase the reversionary interests herself. But as an exercise, it is helpful to think of the hypothetical investor’s share.
2020: C is awarded the Swift valuation of a life-interest (50 years) in the full £1m:
£1m – (£1m x 1.05-50) = £912,796.27.
The investor provides the £87,203.73 to secure the reversionary interest and together they buy the property.
2045: C purchases a second property for £250,000, or makes a capital investment, exactly as she would have done in her uninjured (counter-factual) state.
2070: C dies. The £1m property reverts to the Investor (C’s estate gets no equity). The £250,000 second property (or other investment) goes to C’s estate just as it would have done if she had remained uninjured.
Damages Award: £912,796.27
2020: C is awarded damages to purchase a 25-year fixed term interest in £250,000 worth of equity in the property, using the Swift formula; and to purchase a full life interest (50 years) in the remaining £750,000 of equity. The values of those two interests are:
£250,000 – (£250k x 1.05-25) = £176,174.31.
£750,000 – (£750k x 1.05-50) = £684,597.20.
The Investor puts up the remaining £139,228.49 to buy the two reversionary interests.
2045: The 25-year product matures and £250,000 worth of equity (25%) in the £1m property is released and goes to the investor. C then puts £250,000 of her own capital back into the property and now outright owns 25%.
2070: C dies. 75% equity in the property reverts to the Investor. 25% of the property goes to C’s estate (i.e. £250,000) just as it would have done if she had remained uninjured.
Damages award: £860,771.51
2020: The future £250,000 is immediately factored into the C’s finances and C is only awarded damages to buy a life-interest in £750,000 worth (75%) of equity in the £1m property.
£750,000 – (£750k x 1.05-50) = £684,597.20.
The Investor buys the reversionary interest in that 75% share for £65,402.80.
C then borrows the remaining £250,000 from other heads of damages to purchase outright 25% equity in the property.
2045: C settles her debt with her past self, using the £250,000 in capital she now has now amassed as she would have done if she had remained uninjured
2070: C dies. 75% equity in the property reverts to the Investor. 25% of the property goes to C’s estate (i.e. £250,000) just as it would have done if she had remained uninjured.
Damages Award: £684,597.20
All three options are compatible with the judgment in Swift and indeed use its formula. It is not difficult to see which options would be preferred by Defendants and Claimants.
Will there still be harsh outcomes for claimants? It is considered that there will still be some harsh outcomes. The problem remains the same as was recognised in Roberts v Johnstone and harsh outcomes are most likely to occur when the claimant has only a short future life expectancy. As the Court of Appeal has observed in other cases, the amounts awarded in respect of general damages will never be sufficient to enable a claimant to “rob Peter to pay Paul” (i.e. to deploy his/her general damages award to the acquisition of a suitably adapted alternative property, given the disparity between the sums so awarded and the cost of acquiring such an adapted property).
If the market does not develop, the Court of Appeal has left the door open for different guidance to apply to cases of shorter life expectancy, and one route which claimants may now explore is to seek permission to call expert actuarial evidence. This will be to seek to establish that in such a case the claimant will be significantly under-compensated and that some other solution requires to be found and damages awarded in a manner which is consistent with the requirement that the claimant is entitled to be so compensated that he can be placed in the same position that he would have been had the tort not been committed – the ‘full compensation’ concept was established in the 19th century and endorsed by Lord Scarman in Pickett v British Rail Engineering (1980). In many respects this introduces an unwelcome element of uncertainty and Masters (by whom such decisions are in practice going to be made) will have differing views on what should be the outcome of such applications, especially given the inevitable uncertainty that is inherent in the CPR’s overarching concept of “proportionality”.
One route that is likely to be explored is the solution of renting the alternative adapted property when the claimant’s life expectation is short. This will avoid the need for such a claimant to bear the burden of deploying other aspects of his/her damages award to the cost of acquiring that property. The costs of adapting the property (especially where they do not and will not enhance its sale value) will be recoverable as a separate head of claim. The remainder of the claim can be assessed on the basis of the differential cost of having to rent the adapted property (as opposed to the rental cost associated with his/her pre-accident accommodation or, if that was owned by the claimant, the rental value associated with that pre-accident accommodation) multiplied by the requisite life multiplier.
Will there be an appeal? It is understood that the defendant’s insurers are intending to seek permission to appeal to the Supreme Court. The usual outcome would suggest that the Court of Appeal (being unanimous in their views) would refuse permission and require the defendant to apply to the Supreme Court for permission to appeal. That would have the advantage of enabling 3 judges of the Supreme Court to examine the issue and decide whether or not it is a decision which merits further consideration by it or whether it considers that the issue is one of practice rather than of substantive law. The authors’ view is that the House of Lords Judicial Committee would have refused permission to appeal on that ground (Roberts v Johnstone did not proceed there and as Lord Scarman said in Pickett v BRE “Though arithmetical precision is not always possible, though in estimating future pecuniary loss a judge must make certain assumptions (based upon the evidence) and certain adjustments, he is seeking to estimate a financial compensation for a financial loss”). In short, an award of damages is but an “estimate” and not a perfect arithmetical outcome. However, it has to be recognised that the Supreme Court is a more interventionist entity than its predecessor. Accordingly, it must be a case of “watch this space”.