Behind the times

Philip M D Grundy, Nick Martin, Edward J P Grundy and Matthew Smith explain why Roberts v Johnstone is no longer applicable

‘There is no doubt that several of the potential solutions identified provide both the tortfeasor and the claimant with a range of practical alternatives to the accommodation conundrum.’

At 38.204, McGregor on Damages states:

It is high time that the Roberts v Johnstone problem was tackled and a fair proper solution found and adopted. The Law Commission looked into the matter some time ago but found it too difficult to formulate an acceptable solution and so recommended that the Roberts v Johnstone method be retained. The Ogden Working Party is fully aware that the law needs to be righted and has it in mnd to investigate the issue in the near future.

The judiciary have attempted to identify a better solution for some time, but the reduced discount rate could be the spark to light change in this area to find a proper solution moving forwards. McGRegor continues (our emphasis):

What could trigger action on this front is further reduction in the discount rate, the possibility of which, as we have seen, is very much in the air. It is true that, as the discount rate lowers, the multipliers increase, but an examination of the figures in the tables in Ogden shows that the increases in the multipliers do not come anywhere near to balancing, or off-setting the effect of, the fall in the discount rate. Ironically the injured party will get more for care but less for special accommodation. Indeed should the discount rate move into the negative, which is highly unlikely but did happen in the Guernsey case in the Privy Council of Simon v Helmot [2012], the Roberts v Johnstone method becomes unworkable; it would produce a nil award.

McGregor opines that Roberts v Johnstone would be unworkable in a negative rate. That’s been shown to be the case at present, but it will also result in problems if -or when – the rate goes back up to 0%-1%.

Even McGregor did not foresee the discount rate moving to -0.75%! Nonetheless it has been recognised by all for a number of years that the valuation of accommodation claims needed to be revisited.

Relevant Cases

The Roberts v Johnstone method began in the case of George v Pinnock [1973] in which, at the time of trial, the claimant had moved into a suitable bungalow. The capital cost was £12,000 having been paid out of interim payments. The trial judge made no separate award for accommodation and, in fact, made no reference to it in his judgement at all. The Court of Appeal was persuaded by the parties to deal with the issue of accommodation. The claimant made two submissions:

  • She should receive either the whole or some part of the capital cost of acquiring the bungalow since it was acquired to meet the particular needs arising from the accident.
  • The claimant had been involved in greater annual expenses of accommodation than she would have incurred if the accident had not happened.

Orr LJ dismissed the first submission but found that the second was well founded. He found that the claimant was:

…entitled to be compensated to the extent that the loss of income or notional outlay by way of mortgage interest exceeds what the cost of her accommodation would have been but for the accident. She would also, in my judgment, have been entitled to claim the expenses of a move to a new home imposed by her condition and the expense of any new items of furniture required…

In Roberts v Johnstone [1989] the Court of Appeal reviewed the application of George. It is clear that the Court of Appeal approved the proposition that damages under the head of accommodation costs should not represent the full capital cost of the asset. This is because the asset would remain intact at the claimant’s death. We must remind ourselves that at the time Roberts v Johnstone was being considered the mortgage rate was just over 9%, and so if the court had used the annual mortgage interest to reach the appropriate figure, it would have led to recovery of a sum in excess of the capital value of the house. Therefore, sensibly, the court considered the lost income by reference to return and risk-free investment.

Stocker LJ said:

Where the capital aspect in respect of which the cost is incurred consists of house property, inflation and risk element are secured by the rising value of such property particularly in desirable residential areas, and thus the rate of 2% in risk reinvestment would not be a wholly unacceptable one.

Of course, the 2% has subsequently been increased to 2.5% until last year.

Tomlinson LJ made some helpful observations in the case of Manna v Central Manchester University hospitals NHS Foundation Trust [2017], albeit the facts are not necessarily relevant to most cases.

Damages in cases of this sort are notionally intended to provide a fund which will both meet the claimants lifetime needs and be exhausted contemporaneously with the termination of the claimant’s life expectancy…The resulting sum awarded will be wholly insufficient to purchase a property, but the theory is that the shortfall and thus the balance of the actual funds required in order to purchase a suitable property can be found in, or borrowed from, the awards made under other heads of damage such as pain, suffering and loss of amenity, loss of earnings, capitalised awards for therapy and other costs.

His Lordship continued by saying what we all know but are hesitant to state openly:

The exercise in which the court is thus engaged is in modern conditions increasingly artificial. The assumption underlying the approach is that the claimant will be able to fund the capital acquisition out of the sums awarded rubrics other than accommodation. But in modern times residential property prices have increased rapidly while general awards for pain, suffering and loss of amenity have remained at their traditional levels. Whilst Peter is no doubt robbed to pay Paul, it must often be the case that the accommodation assessed by the court as suitable is simply not purchased.

His Lordship also pointed out other potential issues:

Where the award is only for 50% of the sums regarded as necessary to meet the claimants reasonable needs.

His Lordship described the approach that we have been adopting for many years in Roberts v Johnstone as ‘imperfect but pragmatic’.

So how we are going to approach this in the future?

The case of JR v Sheffield Teaching Hospitals NHS Foundation Trust [2017] was the first major decision in this area following the change in the discount rate. In that case the defendant argued that:

The conclusion which must be taken from the Lord Chancellor’s statement is that there is at present no ability to obtain any positive return on a capital fund based on risk-free investments.

This means there is no need to compensate the claimant for the loss of that return. In the past the notion was that the sum expended on a house otherwise would have earned an income. This was the basis of the Roberts v Johnstone formula. That basis has no foundation. The one type of investment which will continue to yield a return in the long term is real property. The defendant’s argument is that ‘the cost of the accommodation can be borrowed….from the capitalised loss of earnings figure’.

William Davis J summarised the defendant’s argument as follows:

The negative yield on interest linked gilts means that the Roberts v Johnstone formula results in a negative sum for accommodation costs.

The defendant was not seeking credit for the negative sum (as others have hinted at), but simply a zero figure for the cost of the accommodation:

Applying Wells v Wells JR has suffered no loss by investing in the capital cost of accommodation from his other capital funds.

His Lordship made no award for accommodation, finding that:

The difficulty facing JR is that applying the rationale of Roberts v Johnstone in the current climate results in a nil award for the capital cost for accommodation.

We need to find a proper solution to the accommodation conundrum whatever the discount rate is moving forward.

Davis J appeared to want to proceed with a different approach and acknowledge that the situation was less than ideal, but noted that he had no evidence before him to consider some other approach.

He gave examples:

Given the current cost of borrowing, it might have been possible to say that the interest element on an appropriate mortgage (say £600,000 as the cost of a property less the amount of general damages) over a 25 year term would provide a reasonable figure, the cost of an annual mortgage interest being the alternative method of assessment suggested in George.

JR v Sheffield was the subject of an appeal. It was compromised prior to the determination of the Court of Appeal. It us our understanding that this was compromised on the basis that the lump sum is calculated by using the agreed capital cost of the accommodation less the agreed sum for general damages. This we understand is the approach being adopted in other such claims too.

Davis J continued:

[The alternative method] was rejected in Roberts v Johnstone because the rate of mortgage interest at that time was so high that an award on that basis would result in full recovery of the capital cost of the accommodation [if not more].

Davis J simply stated: ‘That is no longer the case.’

Therefore in the present climate Roberts v Johnstone and it’s calculation does not work.

The defendant could of course take a reversionary interest in the purchased property, in which it would therefore not result in any windfall benefit to the claimant but would provide the injured claimant with the accommodation they need during their lifetime.

It is obvious to all the present unfairness in the valuation of accommodation claims, particularly with the lower discount rates.

The box below shows an example of calculating an accommodation claim. You have to be creative to each individual set of circumstances: age, life expectancy, personal circumstances.

No longer will one calculation fit for all.

But if so which one is the best?

Example 1: Roberts v Johnstone calculation


Life expectancy / discount rate 37 year old male
Life expectancy 49.9
Life multiplier -0.75% 60.25
0% 48.93
0.5% 43.00
1% 38.07
2.5% 27.53

Using multipliers for a 37 year old man:

  • Capital cost : £500,000
  • 1% : £5,000
  • 2.5% : £12,500
  • 0.5% : £2,500
  • 2% : £10,000

Under the old discount rate of 2.5%, the calculation would be:

£12,500 x 27.53 = £344,125, a shortfall of approximately £150,000.

Now the shortfall will bein the range of £300,000-£400,000. Of course, the loss of earnings calculation will be higher (potentially and subject to the usual arguments), but pain, suffering, and loss of amenity damages remain as before.

Potential Alternatives

In our view, there are a range of approaches that the court could consider when dealing with accommodation issues, irrespective of the discount rate.

Declaration of trust

The first suggestion is that the tortfeasor is required to fund the full purchase price of suitable accommodation upfront. A declaration of trust could be established so that the claimant would be the life tenant, and the defendant would be entitled to the capital upon the death of the claimant. The full value of the property, plus any property price increases, would then revert to the defendant following an appropriate period after the death of the claimant. The tortfeasor would then gain on any capital growth on the full value of the accommodation. The extent of the benefit to the tortfeasor is of course a matter of conjecture and there will be costs associated with the establishment and running of such a trust, as well as wind-up costs following death. There may also be legal and taxation implications for tortfeasors and these would need to be carefully considered. This trust solution negates the need for any form of future loss quantum calculation as part of the legal process, rather it simply considers the suitability of the required accommodation and takes away the shortfall apparent under the Roberts v Johnstone calculation.

Although there may be an initial cost burden on the tortfeasor in setting up this arrangement, they may benefit from any appreciation in the property price over the lifetime of the claimant.

Interest-free loan

An alternative to the suggestion above would again involve the tortfeasor funding the full purchase price of the suitable accommodation upfront. The property would however belong to the claimant in it’s entirety and would form part of their estate on death. The capital to fund the accommodation is provided to the claimant as an ‘interest-free loan’ which could be secured, in some way, against the claimant’s general assets.

The interest-free loan would then be repayable after death from the value of the claimant’s estate. A form of liability would likely need to be created which would survive the claimant’s death. The claimant and their estate would have a liability for the capital cost of the accommodation purchase, i.e. the interest-free loan. If the claimant were a protected party, it is probable the Court of Protection would need to provide sufficient authority to the Property and Affairs Deputy to effect this type of arrangement. The tortfeasor would be able to recover the sum of the interest-free loan from the estate after the death of the claimant. In effect, the tortfeasor becomes the provider of an interest-free loan.

Residential mortgage funding

We have also considered the option of using residential mortgage funding, the repayments of which are dealt with by way of a periodical payment order. Suitable accommodation is initially purchased using a combination of upfront capital payment (a deposit) and a mortgage loan from a regulated mortgage lender. This could be established on an interest-only or capital repayment basis, subject to the lender’s specific criteria. Annual mortgage payments could be paid as part of a periodical payment order. The capital for a deposit would have to be found from other sources such as general damages and would likely be in the region of at least 30% of the valuation of the suitable accommodation. Equity value in the property would belong to the claimant and would therefore revert to their estate on death, less of course the outstanding mortgage liability.

Under this arrangement the tortfeasor would not have a charge over the accommodation. The annual periodical payment order could be linked to an appropriate inflation measure, i.e. bank base rate or London Inter-Bank Offered Rate (LIBOR) plus a premium (say 3.00% over) or standard variable rate (SVR), taking into account the actual cost of a residential mortgage loan, although this measure would likely need further analysis.

The primary difficulty with this approach is the lack of availability of mortgage lenders which are willing to offer residential mortgages to claimants on this basis. As we write we are not currently aware of any lenders willing to lend under these circumstances, although we understand that investigations with residential mortgage lenders continue.

Award of a sum for borrowing cost

The court may also wish to consider the merits of the award of a sum of damages based upon ‘the cost of borrowing’ ie the additional capital required in order to purchase a property. This was explored ona suggestive basis by William Davis J in JR v Sheffield. It was suggested that an approach based upon evidence of the cost of mortgage interest over a certain term might provide a solution to Roberts v Johnstone>. This would be based upon the cost of a 100% interest-only mortgage, which would be capitalised at the point of settlement. The suitable property is then purchased by the claimant using those damages, although potentially any shortfall would have to be made up out of heads of loss, such as general damages.

Under this cost of borrowing method, the tortfeasor would not have a charge over the accommodation. The notional capitalised multiplicand could be linked to an appropriate inflation measure, ie bank base rate or LIBOR plus a premium (say 3.00% over) or SVR to take into account the actual or historic current cost of residential mortgage loan. The adopted multiplier could be based upon term certain using the current -0.75% discount rate, derived from Table 28 of Facts & Figures (Facts & Figures 2017/18@Tables for the Calculation of Damages, edited by Simon Lavine et al), although this does require further analysis. It is accepted that this method of calculation may result in a figure which might be more or indeed less than the purchase price of suitable property.

The appropriate inflation measure will also need to be carefully considered. For example, historical SVRs on residential mortgages from 1995 to 2017 has been approximately 5.95% per annum, whereas currently the SVR is in the region of an average of 3.89% per annum. Using the cost of borrowing calculation, an example is set out in the box below.

Example 2: the cost of borrowing method

This assumes that the suitable accommodation purchase cost is £600,000 and the claimant is aged 20 years, with a normal life expectancy, although we have used a 25-year term to reflect the average term on a residential mortgage.

The long-term SVR approach:

£600,000 x 5.95% = £35,700 x 27.51 = £982,107

The current SVR approach:

£600,000 x 3.89% = £23,340 x 27.51 = £642,083

(Figure of 27.51 calculated using table 28 at -0.75% for 25 years).

It is accepted that this calculation results in a figure which in some cases would exceed the actual purchase cost of the accommodation and as such, certain credits could be applied to this or indeed this lump sum could be capped up to the sum of the appropriate accommodation. We also considered that where any property adaptations would enhance the property value, any potential enhancement value could be deducted from the capital sum before the cost of borrowing that sum is calculated. Such an approach may be appropriate, since the value of the enhancement has been added to the property and the claimant owns the accommodation, so theoretically enjoys that additional value even if they cannot enjoy that capital while they are alive.

Rental Accommodation

Rental of suitable accommodation could also be an option in certain circumstances.

Perhaps of limited use, and of course subject to a number of factors, rental may provide a solution. It is worth noting that a rental premium would likely be charged by a landlord for a claimant who requires specially adapted accommodation and there may also be cost implications associated with any adaptations at commencement and indeed at end of tenancy. For example, putting it back into original condition after the claimant vacates the property. Landlords would also likely be locked in for an uncertain period with a property that is desirable only for a relatively small group of people; again, good reason to expect a premium on the rental value. The difficulties associated with this option are therefore manifold and would only likely appeal to a small number of claimants.
Private equity investment

More recently, general enquiries have been made to explore the concept of a private equity investor, albeit further enquiries and research does need to be undertaken to progress this into a viable option. The theory is that a private equity investor buys the accommodation and the defendant pays the rent by the way of a periodical payment order. We are aware of one case, JM v Aylward [2015], which gives rise to an interesting issue which provides a structure for buying and adapting a property which avoids the disadvantages of Roberts v Johnstone in cases where there is a limited life expectancy.

In this case, it was determined that JM had capacity to decide where he wanted to live. As expected the Roberts v Johnstone calculation left a significant shortfall in the accommodation claim and there was no obvious way to make up the ‘missing’ balance, without compromising future equipment or therapies. The solution was that a commercial consortium agreed to buy the property in their name, and to grant JM a lifetime tenancy. On his death, the consortium would take back the property, along with any capital gain. JM would pay for the adaptations required using damages paid for by the tortfeasor. In return for his security of tenure, JM paid an enhanced rent. The rent and the increased cost of living were paid for via the periodical payment order, which we believe in this particular case was varied in-line with RPI indexation for simplicity. That said, perhaps more suitable inflation proofing might be possible by linkage to various measures tracked by the Office for National Statistics in relation to increases in rents in the private sector.


It is clear that the Roberts v Johnstone method of calculating accommodation claims is unworkable in a negative discount rate environment and we do not believe that the insurance industry can sensibly resist a change to the valuation of accommodation claims as this would reveal a clear and obvious lack of care for any claimant in such need. Certainly it would fundamentally undermine the stance being adopted by insurers; namely, claimants should still receive 100% compensation.

No easy solutions to the Roberts v Johnstone problem exist and while there are some interesting solutions which we have identified above, unless the court provides a consistent, fair, long-term and practical solution to the problem our alternatives outlined above will remain to be case specific. Perhaps a ‘one-size fits all’ solution may not even exist, given the changed economics.

There is no doubt that several of the potential solutions identified above provide both the tortfeasor and the claimant with a range of practical alternatives to the accommodation conundrum. That said, all of these potential solutions come with limitations and difficulties, and it is appropriate to have expert financial advice evidence to assist the court with some detailed explanation of how some of these solutions may be workable for our particular claimant. Our advice is to get some sound cogent advice! In our view a judge would not criticise the obtaining of such advice setting out the options in the individual case, and in fact would welcome it.

In the lowered discount rate ‘new world’, the claimant’s accommodation needs must still be adequately met. We can no longer ‘rob Peter to pay Paul’ and make up the unachievable shortfall in the Roberts v Johnstone calculation, and imagination from both sides is now required to resolve this long-standing and now urgent problem

George v Pinnock  [1973] 1 WLR118

JM (A Protected Party suing by his Sister and Litigation Friend, KR) v Aylward & ors (2015) out of court settlement (approved), Owen J, 2 March

JR v Sheffield Teaching Hospitals NHS Foundation Trust [2017] EWHC 1245 (QB)

Manna v Central Manchester University Hospitals NHS Foundation Trust [2017] EWCA Civ 12

Roberts v Johnstone [1989] QB878

Simon v Helmot [2012] UKPC 5