The Mayor’s Office for Policing and Crime v Mitsui Sumitomo Insurance Co (Europe) Ltd and others

London suffered from serious rioting for four days from 6 to 9 August 2011. The rioters caused extensive damage to property. Property owners and insurers suffered significant losses. Several owners of uninsured property, including two of the respondents in this appeal, lost their businesses when they became insolvent as a result of those losses. Property owners and insurers, which had compensated their assureds, submitted claims for compensation from the appellant police authority (“MOPC”) under section 2 of the Riot (Damages) Act 1886 (“the 1886 Act”). The MOPC contested those claims initially on both liability to compensate and the quantification of loss. The liability of the MOPC to pay compensation is no longer in issue. The question is the quantification of the claims.

This appeal raises a question of statutory construction. It is whether persons who suffer loss when rioters destroy their property can in principle obtain compensation for consequential losses, including loss of profits and loss of rent, under section 2 of the 1886 Act, and if so on what basis.

This appeal is concerned with one riotous incident which occurred on the third night of the London riots. At about 11.40 pm on 8 August 2011 a gang of youths broke into the Sony DADC distribution warehouse, which is situated in a business park on Solar Way in Enfield. The youths stole goods from the warehouse and also threw petrol bombs which caused a fire. The fire destroyed the warehouse and the stock, plant and equipment within it.

The insurers of Sony DADC, which were the lessees of the warehouse, the insurers of the freehold owner of the warehouse, and companies which were customers of Sony DADC and whose stock in the warehouse had been destroyed, made claims against the MOPC.

In the Commercial Court of the High Court, Flaux J had to decide two preliminary issues. The first issue concerned liability and was whether the warehouse had been destroyed by persons assembled together “riotously and tumultuously” within the meaning of section 2(1) of the 1886 Act. In his judgment dated 12 September 2013 Flaux J held that it had been. The Court of Appeal (Lord Dyson MR, Moore-Bick and Lewison LJJ) in a judgment dated 20 May 2014 upheld that finding.

The second preliminary issue is the subject matter of this appeal. Flaux J held that section 2 of the 1886 Act provided compensation only for physical damage and not for consequential losses. The Court of Appeal reversed that finding. It held that section 2(1) of the 1886 Act provided a right to compensation for all heads of loss, including consequential loss, proximately caused by physical damage to property for which the trespassing rioter is liable at common law, save to the extent that they are excluded by the statute. The MOPC appeals to this court against that finding.

Section 2(1) of the 1886 Act as amended provides:

I can cover the other relevant provisions of the 1886 Act briefly. Section 2(2) allows the insurer which has indemnified its assured in whole or in part to claim compensation. Section 3(1) provides:

In support of its view that the 1886 Act provides for the recovery of consequential losses, the Court of Appeal began with a linguistic analysis of section 2(1). It pointed out that the words, which I have emphasised in para 7 above, compensated for loss “sustained … by such injury, stealing, or destruction”. This was loss that was caused by (i) damage to or destruction of a building, or (ii) damage, destruction or stealing of property in the building. Such loss could as a matter of linguistic analysis include consequential losses, such as the loss of rent while an owner repaired his building. Secondly, the other provisions in the 1886 Act, including the now-repealed preamble (which I discuss in para 31 below), did not militate against this view. Case law on predecessor legislation suggested that remedial statutes should be given a liberal interpretation. Thirdly, that case law, which I discuss in paras 20 to 23 below, also suggested a principle that the relevant community, which was then the hundred, stood as sureties for the trespassers. There was no reason to think that a rioter would not have been liable in tort for consequential losses before Parliament legislated in 1714. Thus the local authority incurred such liability under statute. The 1886 Act did not depart from what the Court of Appeal described as “the fundamental ‘standing as sureties’ principle”.

Fourthly, the court rejected any reliance on the regulations which the Secretary of State promulgated in the London Gazette in 1886 as an aid to the interpretation of the 1886 Act. Fifthly, the court rejected for lack of evidence a submission on behalf of the MOPC that there was a settled practice of interpreting the 1886 Act as excluding compensation for consequential losses. Sixthly, the court considered that there was an anomaly if the 1886 Act did not cover consequential loss. An owner of a commercial building which was damaged in a riot might choose to sell it in a damaged state and claim as his compensation the diminution in value caused by the physical damage. Where a building was valued by reference to its capacity to generate income, part of that diminution in value could be attributable to loss of rent or loss of profits that the purchaser would suffer pending the completion of remedial works. By contrast, if an owner decided to repair the building and suffered a loss of rent or a loss of profits while the remedial works were carried out, he could not recover such losses if the 1886 Act did not extend to consequential losses. The court said that there was no rational basis for imputing to Parliament an intention to allow recovery for such losses as part of a claim for diminution of value but to exclude a free-standing claim for losses of the same character. Seventhly and finally, the court derived no assistance from parallel Scottish legislation, namely section 10 of the Riotous Assemblies (Scotland) Act 1822 (3 Geo IV, c 33) because of its use of different language.

Lord Pannick QC for the MOPC submitted that Flaux J had reached the correct conclusion on the interpretation of section 2 of the 1886 Act and that his order on this point should be restored. In support of his contention he relied on what he called the purpose and the plain meaning of the words in section 2(1) and also on sections 3 and 7 and the repealed preamble of the 1886 Act. He also relied as a contemporaneous exposition of meaning on the first regulations under the 1886 Act which the Home Secretary promulgated on 28 July 1886. He departed from the argument of settled practice which had been included in his written case, accepting that evidence of such practice had not been adduced. But he submitted that the historical background to the 1886 Act and in particular the history of prior legislation and judicial pronouncements on that legislation supported the view that the legislation from the outset was a self-contained statutory scheme for compensation which was not co-extensive with the tortious liability of the trespasser. In the prior legislation the compensation was limited to physical damage to the premises or property in it. The 1886 Act did not materially alter the nature of that compensation scheme.

Mr Michael Crane QC for the first to third respondents presented the issue for this court as being whether the 1886 Act excludes in principle a head of loss caused by physical damage to property inflicted by rioters and otherwise compensable under the English law of tort. In advocating a negative answer to that question, he submitted that the words of the 1886 Act contained no such limitation and that the history of the legislation since the 1714 Riot Act (1 Geo I, c 5) was consistent with the ancient notion that the inhabitants of the hundred stood surety for the good behaviour of their fellow subjects. The principle was that the liability in damages of the rioter should be transferred to the hundred. That principle survived the transfer by the 1886 Act of that liability from the hundred to the police authority. The 1886 Act contained no clear language to limit the liability of the police authority by excluding the recovery of consequential loss. In short, the history of the legislation showed that the heads of loss recoverable from time to time in an action against the trespasser were recoverable as a matter of strict liability initially from the hundred and since 1886 from the police authority. The Court of Appeal had been correct in concluding that the police authority stood in the shoes of the trespasser save to the extent that the 1886 Act provided otherwise. The appropriate analogy in construing the 1886 Act was with a strict liability in tort, arising from the failure of the police to maintain law and order. He founded his argument also on the anomaly which had carried weight in the judgment of the Court of Appeal (its sixth reason which I have summarised in para 10 above). Mr Simon Pritchard for the fourth and fifth respondents, which had been trading companies, made submissions adopting and supporting those of Mr Crane. He also explained that those respondents were in large part uninsured and that Sony DADC’s liability as bailee had been restricted by contract to the manufacturing replacement cost of damaged stock. Their inability to recover the market value of their stock and their lost profits had precipitated their insolvency.

The appeal, as I have said, raises a question of statutory construction. While the arguments have been wide-ranging, the resolution of the dispute is to be found in the words of the 1886 Act, interpreted against the backdrop of the prior legislative history. In my view this is a case in which history rather than legal theory casts light, revealing the correct answer.

Linguistic analysis of the relevant provisions of the 1886 Act by itself does not provide a clear-cut answer. Section 2(1) speaks of compensation for “loss by” the injury or destruction of a building or the injury, stealing or destruction of property within the building. Those statutory words do not disclose whether the loss which the claimant has sustained by the destruction etc of his property is simply the damage to the property, to be compensated by payment of the cost of repair or the diminution in value of the building or other property, or extends to consequential loss, such as the loss of rent or loss of profit which the claimant would have derived from the property.

Section 6 of the 1886 Act provides that compensation will be payable in the same way for the injury or destruction of manufacturing or agricultural machinery and fixtures and for equipment in a mine or quarry. By providing that the Act will apply “in like manner” to such property, it casts no light on the scope of section 2.

What is striking, however, is that the 1886 Act does not expressly provide compensation for either (a) personal injury caused by rioters and resulting medical expenditure or (b) damage to property in the streets such as a parked car. We were referred to no jurisprudence to support the view that such losses could be claimed under the 1886 Act even where they resulted from damage to or the collapse of a building. On any view, therefore, the Act provides only partial compensation for damage caused by rioters. Further, those limitations show that it is not correct to interpret the words “sustained loss by such … destruction” as creating an unqualified causal test to which the normal rules of causation in tort can readily be applied.

I do not find the other provisions of the 1886 Act to be of any assistance in addressing the disputed question. Section 3 requires the compensation authority of the police area to “fix such compensation as appears to them just”, while section 4 allows persons who are aggrieved by the decision of the compensation authority to raise an action against it in order to obtain a judicial determination of their claims. Section 7 identifies who may be the claimants for damage to a church, chapel, school, hospital, public institution or public building by deeming them to have sustained “loss from such injury, stealing, or destruction”. It goes on to state that claims may be made “in relation both to the building and to the property therein”. I do not interpret its speaking of “loss from” destruction etc as altering the test in section 2. Nor do I construe the phrase “in relation both to the building …” as casting light on the scope of the claims that may be made “in relation … to” a building.

Such light comes in my view from the interpretation of the 1886 Act in the context of the prior legislative history since 1714, to which I now turn.

Parliament first provided for compensation for riot damage in 1714 in response to the public disorder which followed the succession to the throne of Great Britain of George, the Elector of Hanover, as George I. Section 1 of the Riot Act 1714 made it a felony punishable by death for an unlawful assembly of 12 or more persons to fail to disperse after a justice of the peace or other specified official had read a proclamation commanding them to do so. The procedure, which was a precondition of the felony, became popularly known as “reading the Riot Act”. Section 4 made it a felony punishable by death for rioters to demolish or pull down buildings for religious worship, dwelling-houses and farm buildings. Section 6, provided that when rioters had demolished or pulled down all or part of such buildings,

The 1714 Act did not specify the scope of the damages to be paid by the local community. Cases, which followed later riots, enabled judges to give some guidance. In Ratcliffe v Eden (1776) 2 Cowp 485 (98 ER 1200), which followed upon a riot by sailors in Liverpool, the Court of King’s Bench was concerned with the question of whether the victim of a riot could recover compensation not only for the damage to his house but for also the destruction of the furniture and household goods within his house. The hundred argued that the victim could not recover for the furniture and goods as their destruction was a separate and independent act from the damage to the house. The court rejected this defence. Lord Mansfield (at p 488) explained that the 1714 Act had altered the nature of the offence; rioters were no longer trespassers but felons and were to be hanged. Before the Act the trespassers would have been liable in damages. Under the Act the inhabitants of the hundred instead were liable in damages and this was an inducement to them to perform their duty of preventing or suppressing riots. He stated:

Other cases followed the anti-Catholic “Gordon Riots” in London in June 1780, which caused extensive damage and destruction of property, including Lord Mansfield’s house in Bloomsbury Square. In Hyde v Cogan (1781) 2 Doug 699 (99 ER 445) the court again considered whether the hundred was liable for the destruction of furniture in a house as well as the demolition of the house. In this case the argument advanced on behalf of the hundred was that the 1714 Act was penal against both the trespasser and the hundred and ought to be interpreted narrowly. Lord Mansfield, although present, declined to express an opinion, leaving Willes, Ashhurst and Buller JJ to decide the case. The judges rejected the contention that section 6, which provided for the compensation, was penal and held that it was remedial; Buller J said that, as a result, it should be interpreted liberally. In that bygone age when, according to Willes J, the furniture in a London house might be worth twice as much as the house itself, that liberal interpretation brought household goods within the scope of the statutory compensation scheme. The court also had before it a note of the judgment of Lord Loughborough in the Court of Common Pleas in the case of Wilmot v Horton, which had been decided earlier in the same year. In that case Lord Loughborough gave both the remedial nature of the Act and its substitution of the liability of the hundred for that of the offender as the reasons for allowing the recovery of compensation for the destruction of furniture within the house.

In Mason v Sainsbury (1782) 3 Doug 61 (99 ER 538) the question was whether insurers, who had indemnified the owner for the damage to his house in those riots, could maintain an action in the name of the assured against the hundred under the 1714 Act. In answering the question affirmatively, the Court of King’s Bench again explained that the Act put the hundred in the place of the trespassers. Lord Mansfield stated (at p 64):

Moving on over two centuries, in Yarl’s Wood Immigration Ltd v Bedfordshire Police Authority [2010] QB 698 Rix LJ at para 54 described the rationalisation of the liability of the hundred and now the police authority in these terms:

I recognise the force of the respondents’ emphasis on the statements of principle that the community (and now the police force) stood as sureties for the wrongdoer. But, for the following three reasons, I do not accept that the rationalisation can bear the weight that the respondents seek to place on it.

First, while the 1714 Act imposed on the hundred the obligation to compensate only for loss occasioned by the destruction of, or damage to, buildings, which the case law to which I have referred extended to furniture and household goods, the prior law of hue and cry imposed no such restriction. The obligation on the community to raise hue and cry (“hutesium et clamor”) when encountering an offender dates back to before the Norman Conquest, as Lord Mansfield said. For example, John Hudson, The Oxford History of the Laws of England, (2012) vol 2, p 171, refers to a statute of King Cnut (II Cn, c 29) imposing the obligation on someone who failed to raise hue and cry to make amends “at the rate of the thief’s wergeld”, in other words to pay compensation to the victim. Historically, wergild and bot, which had been features of law in England since at the latest the reign of the Kentish king, Aethelbert, in the late sixth and early seventh centuries, extended to payment of compensation for injuries or death and continued as part of the legal scene after the Norman Conquest at least into the 12th century, and afterwards in out of court settlements: Professor Anthony Musson, Wergeld: Crime and the compensation culture in medieval England, www.gresham.ac.uk . Codes were made from time to time establishing fixed values for specified types of injury and damage. The Statute of Winchester of 1285 (13 Edw I) made the hundred answerable for any theft or robbery if it failed to apprehend and deliver up the offender. Pollock and Maitland, The History of English Law before the Time of Edward I (1898) vol 1, pp 648-649, describe this as a form of joint and several liability to the victim. The Statute of Hue and Cry 1584-1585, to which reference was made in section 6 of the 1714 Act, allowed the victim to prosecute the hundred by way of special action on the case for “damages” where the offender was not apprehended. It also set up a system by which a Justice of the Peace and constables could recover the damages from the inhabitants of the hundred and pay the victims, thereby sharing the burden within the community.

Secondly, while under the 1714 Act the hundred incurred strict liability for the riot, the prior hue and cry legislation allowed the community to escape liability if hue and cry were raised and the offenders caught. Under the older law, therefore, the hundred were not sureties for the offender unless they failed to apprehend him. It may have been the intention of Parliament that because the 1714 Act made riot a felony punishable by death, with the result that the offender would not be around to pay compensation and as, like other felons, his assets would be forfeited to the Crown, the injured party should have a right of compensation against the hundred in substitution for his action of trespass. Be that as it may, it is clear that the principle on which the respondents founded could vary in its application.

Thirdly, and to my mind most importantly, the legislative history after 1714 undermines the respondents’ reliance on the general principle in the interpretation of the 1886 Act. The toughening of the criminal law which the 1714 Act represented was extended by the notorious Criminal Law Act 1722 (9 Geo I, c 22), commonly known as “the Black Act”. This introduced many new statutory felonies in response to the activities of poaching gangs (known as “blacks” because they blackened their faces) after the economic downturn caused by the South Sea Bubble. Section 7 of the Black Act provided for compensation from the hundred for “the damages” sustained by the killing and maiming of cattle, the cutting down of trees and the destruction of agricultural buildings and equipment. In the Malicious Injury Act 1769 (9 Geo III, c 29), in order to remove uncertainties as to the scope of the 1714 Act, Parliament made it a felony for any rioter to demolish, destroy or damage any mills or specified engines and equipment used in the mining industry or fences made for enclosing land by virtue of Acts of Parliament. Compensation for damage by rioters to mills and to works associated with mills was introduced by the Compensation for Injuries to Mills etc Act 1801 (41 Geo III, c 24).

In response to the developing industrial revolution, Parliament enacted the Malicious Damage Act 1812 (52 Geo III, c 130) which extended the compensation regime to protect industrial buildings and equipment by creating statutory felonies of (a) maliciously setting fire to commercial and industrial buildings and engines and (b) demolishing or beginning to demolish such buildings and equipment in the course of a riot. Section 3 of the Act provided that persons injured by the damage caused by rioters (in (b) above) were

Section 38 of the Seditious Meetings Act 1817 (57 Geo III, c 19) imposed on the inhabitants of the hundred the liability to pay full compensation for the destruction of or damage or injury to “any house, shop, or other building whatever” or for the destruction, taking away or damage of “any fixtures thereto attached, or any furniture, goods, or commodities” in those buildings in the course of a riot. Thereby it gave statutory effect to the 18th century decisions which included furniture and household goods within the scope of the compensatory regime of the 1714 Act. The Riotous Assemblies Act 1822 (3 Geo IV, c 33) introduced separate provisions for compensation in England and Wales on the one hand and Scotland on the other. Section 1 of the Act prohibited the raising of proceedings against the hundred under the legislation mentioned above if the damage sustained in the riot did not exceed £30. Section 10, which established a new compensation regime for Scotland, survived the repeal of the English provisions by the Act which I discuss next.

The Remedies against the Hundred (England) Act 1827 (7 & 8 Geo IV, c 31) is particularly important as it amended and consolidated the prior legislation and as it remained in force until repealed by the 1886 Act. Section 2 provided for compensation for the demolition or destruction in whole or in part of a wide range of buildings and industrial machinery, requiring the hundred to

I can detect nothing in the 1886 Act which removed that limitation. The now repealed preamble stated:

The 1886 Act made the following principal changes to the arrangements for statutory compensation:

As a result of changes in local government, it transferred liability to pay compensation from the hundred to the police authority (section 2(1));

The Secretary of State became responsible for creating and regulating the procedure by which claims could be made, the conditions for those claims and the circumstances in which they might be rejected (section 3(2));

The police authority was charged with inquiring into the claims and fixing compensation as appeared to it to be just (section 3(1));

The police authority was directed to have regard to the conduct of the claimant, such as any provocation of the rioters or failure to take proper precautions to protect his property, when deciding what compensation was due (section 2(1));

Compensation was payable not only if a building or property inside it had been destroyed or damaged by rioters but also if property in the building had been stolen by them (section 2(1));

Insurers were given a right to claim compensation in their own names and the right of the insured person who had received insurance payments was correspondingly reduced (section 2(2)); and

A claimant who was dissatisfied with the police authority’s decision could commence an action in the courts to recover compensation, which could not exceed the amount claimed from the police authority (section 4(1)).

In my view it is not correct to use a judicial rationalisation of a statutory scheme to override the words which Parliament has used. From 1714 to this day, the community, whether in the form of the hundred or the police authority, has not stood in the shoes of the offender for all purposes of compensation. As I have said (in para 16 above) the statutory provisions have given only partial compensation for the loss, injury and damage which a person may suffer as a result of rioting. I see no reason for inferring that Parliament intended that the statutory compensation should extend beyond the cost of repairing physical damage to property. When regard is had to the words of the statute, in the context of the prior legislative history, there is no reason to think that Parliament ever intended that the compensation scheme should mirror the offenders’ liability in tort or that its scope should develop as the law of damages for tort developed. While the adoption of a liberal interpretation, as enjoined by the 18th century case law, justified the inclusion of furniture and household goods within the scheme if they were damaged as a result of the demolition of the building or at the same time in the course of the same riot, it cannot alter the nature of the compensation scheme.

In summary, I consider that the words of the 1886 Act should be construed in the light of the prior legislation. The 1714 Act used open-textured wording, requiring the payment of “damages” to persons injured or damaged by the demolition of their houses. The courts’ liberal interpretation extended the hundred’s liability to cover physical damage to household goods and furniture but no further. This limited extension was incorporated into the 1817 Act. Over time, statutory innovations extended the scope of the compensation to cover agricultural buildings, mills, commercial and industrial buildings, the contents of those buildings, and mines and collieries. There is nothing in the wording of the 1886 Act that supports an intention to extend the scope of the compensation to cover consequential loss. Several provisions suggest a contrary intention. I refer in particular to the absence of compensation for personal injury, or for injury to property other than buildings and their contents, together with the unusual provision for compensation to be reduced according to very broad assessments of the conduct of the claimant. Together, they support the conclusion that the 1886 Act, like its predecessors, created a self-contained statutory scheme which did not mirror the common law of tort.

Further, I do not accept that there is any anomaly in this interpretation. A claim for loss of rent or loss of profits in addition to the cost of restoring or replacing a building is different from an estimation of the diminution in value of a commercial building, in which the valuation of the undamaged building had regard to its income earning potential. They are different heads of loss. A claim for the diminution in value of the building is a measure of the compensation available for the damage to the building itself, for example if the owner chooses to sell the damaged building instead of restoring it. If that diminution in value is greater than the cost of the restoration of the building, the claim will normally be capped at the latter figure. Even if there were an anomaly, that would not entitle the court to refuse to give effect to the words of the statute.

In the debate in this appeal counsel speculated on when the common law first recognised a claim for consequential loss. This court was referred to The Kate [1899] P 165, an Admiralty case concerning the collision of two vessels. The court held that the proper measure of damages was the value of the lost vessel at the end of the voyage and also the profits lost under the charter-party. In his judgment, the President, Sir F H Jeune, supported that conclusion by referring to The Columbus (1849) 3 Wm Rob 158. In the absence of further citation of authority, I am prepared to assume that by 1886 the common law of damages for tort would in principle include a claim for lost rent or lost profits arising from damage to a building. But that does not assist the respondents unless they could establish that the 1886 Act was intended to mirror the common law.

Mr Crane also referred to Bedfordshire Police Authority v Constable [2009] 2 All ER (Comm) 200, in which the Court of Appeal addressed the question whether a police authority’s liability under the 1886 Act for the damage to property caused by a riot in an immigration detention centre was covered by its insurance contract, which gave an indemnity in respect of all sums which the assured “may become legally liable to pay as damages”. The court, in the leading judgment of Longmore LJ, held that it was because the police authority was notionally in breach of its responsibility for preservation of law and order (paras 24-26). I have no difficulty with that conclusion, which is consistent with the thinking behind the medieval practice of hue and cry. But it falls far short of equating the statutory scheme with the wrongdoer’s civil liability in tort.

I can deal with the other submissions relatively briefly. First, in reaching my conclusion on the meaning of the 1886 Act I do not rely on the 1886 regulations which the Secretary of State promulgated in the London Gazette as an aid to the interpretation of the Act. The regulations were not laid before Parliament. But that of itself, while affecting their weight, would not exclude them from consideration as a guide to statutory meaning in accordance with Lord Lowry’s guidance in Hanlon v The Law Society [1981] AC 124, 193G-194G. They are consistent with the view which I have reached of the meaning of the Act by other means and might have been an important adminicle of evidence if the MOPC had produced evidence in support of a case of settled practice.

Secondly, section 10 of the Riotous Assemblies (Scotland) Act 1822 (3 Geo IV, c 33), formerly part of the Riotous Assemblies Act 1822 which I mentioned in para 29 above, gives only limited support to my view. Although my conclusion about the 1886 Act tallies with that reached by Temporary Judge, Morag Wise QC, in her opinion on the Scottish provision in the 1822 Act in Board of Managers of St Mary’s Kenmure v East Dunbartonshire Council 2013 SLT 285, there are, as she recognised, minor differences between the wording of the Scottish provision and that of both the English provisions in the 1822 Act and the 1886 Act, which might have supported a different interpretation of the English provisions. In any event, I do not need to rely on the Scottish provision in reaching my clear view on the meaning of the 1886 Act.

Thirdly, the MOPC advances an argument of public policy. The argument runs thus. The common law does not impose a duty of care on the police to prevent a third party injuring a person or damaging property: Michael v Chief Constable of South Wales Police [2015] AC 1732. The strict liability of the police under the 1886 Act is an exception to the common law principle of no liability. Therefore the court should be slow to widen the liability imposed by the Act. I am not persuaded by this argument. In my view, it is difficult to use the public policy of the common law as an interpretative tool because the statutory compensation has never sought to mirror the common law, but has created a self-contained regime for compensation for property damage caused by rioters.

For these reasons I would allow the appeal.