Direct link Share on

Tom Hickman QC analyses the case for primary legislation as the basis for Government powers needed to respond to the Covid-19 crisis. In this article he sets out his views on reliance to date on existing public health powers and on the relative merits – and limitations - of the Public Health (Control of Disease) Act 1984 and the Civil Contingencies Act 2004.

Summary: A central feature of the Government’s response to the coronavirus pandemic has been its reliance on existing public health powers under the Public Health (Control of Disease) Act 1984. This approach has been questioned by commentators who have suggested that the Civil Contingencies Act 2004 (“CCA”) should instead have been utilised, including most recently by Lord Sumption. This paper examines the Government’s choice and the potential availability of the CCA. It explains, firstly, that the consequence of existing public health powers being used has been that key coronavirus laws have lacked a clear legal basis and, moreover, that it has resulted in parliament being afforded inadequate opportunities to scrutinise the laws that have been introduced. Secondly, this paper explains that the CCA could indeed—contrary to the official Government position—have been invoked; but that the CCA is far from a panacea. The CCA offered at best a problematic stopgap solution. What has actually been required is a statute that sets out a clear basis for Government powers and a bespoke parliamentary oversight regime for scrutiny of the highly intrusive and constantly changing coronavirus regulations that the Government is relying upon.   

A central feature of the Government’s response to the coronavirus pandemic has been its reliance on existing public health powers under the Public Health (Control of Disease) Act 1984 (“PH(CD)A”). The Government has not taken what might be described as an “emergency powers” approach: it has not invoked the Civil Contingencies Act 2004 (“CCA”) and nor has it sought to use the emergency to derogate from the European Convention on Human Rights. The difficulty with a response to a national crisis that relies on existing non-emergency laws is that seeking to address the exigencies of an exceptional situation within an ordinary legal framework can involve ordinary laws being stretched and deployed in ways that were not foreseen or intended and it can result in reduced protections for individuals than were anticipated when those laws were enacted.  

This has proved to be the case with the Government’s response to the coronavirus pandemic.  There are two main problems with the Government’s use of the PH(CD)A. First, there is a lack of clear or explicit legal basis for the most intrusive criminal laws made under that Act and, second, the mechanisms for parliamentary oversight under the Act have proved unsuitable for proper scrutiny of the large volume of constantly morphing social-distancing laws that have been made under it. Such problems have led several legal experts to contend that the Government should have taken an emergency powers approach and used the CCA, which contains wide-ranging powers and enhanced parliamentary oversight. This view was recently expressed by Lord Sumption in his lecture, “Government by Decree”, but others have made similar points, such as Professor Keith Ewing in an early paper, “Covid-19: Government by Decree” and Dr Andrew Blick and Professor Clive Walker. The latter argued that by “discarding the CCA” the government “has evaded its important safeguards and adopted much weaker precautions…”.

There are therefore divergent views as to whether an “ordinary laws” or “emergency powers” response to the pandemic should have been taken. The question underlying this divergence of views is how laws should best be made to tackle the unprecedented national crisis that we find ourselves in. Addressing this issue is, in part, a backward-looking exercise: an assessment of decisions taken in March 2020 when the crisis was at its peak. But it is also a question about the appropriate legal regime for the future, assuming that the virus is likely to remain a threat to public health for some time to come. Addressing this question also provides insights for the design of emergency laws more generally.

This paper suggests that it is a mistake to approach this issue in terms of a choice between emergency powers on the one hand (the CCA) and ordinary laws on the other (the PH(CD)A). It contends that what is required in response to the pandemic—and what should have been done months ago—is that Parliament enact bespoke legislation which sets out on a clear basis the powers that the Government has at its disposal to combat the pandemic, tailored to the situation that we find ourselves in and which Parliament considers to be appropriate, together with bespoke provisions for parliamentary oversight suitable for scrutinising the ever-changing social distancing laws that have been and will no doubt continue to be introduced. We are no longer facing an emergency. We are now facing a prolonged crisis; and this has been the position for some time. An emergency powers narrative premised on a stark dichotomy between ordinary laws (the norm) and emergency powers (the exception) is unhelpful in such a context.  Once we appreciated what law was required to meet the crisis and what parliamentary safeguards were appropriate, this should have been translated into statute in a way which allowed Parliament to have a suitably robust degree of influence over the measures that were imposed.

    Neither the PH(CD)A nor the CCA are well tailored to the exigencies of the actual situation. In the case of the PH(CD)A this is unsurprising. The pandemic has given rise to unanticipated challenges that have resulted in unanticipated solutions. For example, the Government’s 2011 pandemic flu preparedness plan which contributed to the UK being ranked as the best in the world in 2019 for epidemic mitigation and response, stated that bans of gatherings, movement restrictions and border closures would not be effective. Lockdowns were not even contemplated. In terms of the CCA, it will be explained that the restrictions on when that Act can be used together with the swingeing powers it contains mean that its use is also problematic and certainly no panacea. It is extremely broadly framed precisely because it may be needed in a wide variety of situations covering the full spectrum of national security threats. To adopt the parable of Goldilocks and the three bears, the PH(CD)A is too cold and the CCA is too hot: what is required is a law that is just right.

Of course, enacting such a law would involve parliamentary time and effort. But if ever there was an issue that was deserving of parliamentary time and effort and a properly tailored legal basis it would surely be a pandemic that has claimed tens of thousands of lives and crippled the economy. To be sure, the Government is unlikely to be willing to bring forward new legislation and it would need significant political pressure, or a loss in the courts, for it to do so. Nonetheless, this is what ought to have occurred and what should still occur.

The Lockdown Regulations

When in March 2020 the Government invoked the PH(CD)A to enforce a national lockdown, a number of people, including the present writer, questioned whether the Act contained the requisite powers: see Lord David Anderson KBE QCHickman QC, Dixon and Jones, Robert Craig, Lord Sandhurst QC and Anthony Speaight QC, also Hickman, cf. King. In a recent lecture, Lord Sumption expressed fairly emphatically the view that the Act does not justify the powers claimed (see “Government by Decree”). The Public Health (Coronavirus, Restriction) (England) Regulations 2020 (“lockdown regulations”) which came into effect on 26 March 2020 set out criminal laws that required each person or household to remain in their homes until further notice. The specified grounds on which people could leave their home were very narrowly drawn. Government guidance reinforced, indeed as I have explored elsewhere it actually overclaimed, the strictness of the legal regime, stating that people could only leave home for four reasons: (1) for work (but only if the person could not work from home), (2) for exercise once per day, (3) for essential shopping, or (4) for medical need. For most people this meant they could not leave their homes for all or most of the day and night. People stocked up on food and some people did not leave home for days, sometimes weeks, on end. When people did leave home for a specified reason they were not permitted to “gather” with more than one other person at a time. Whilst people’s individual experiences of lockdown varied depending on their personal circumstances, the effect on many, particularly the infirm or those living alone, was severe social isolation.

The vast majority of people were no doubt willing to comply with such restrictions because of the need to slow the spread of the virus and “flatten the curve”, and would have done so whether they were embodied in guidance or law. The threat of criminal prosecution was not the reason such people stayed at home.  However, the threat of sanctions against a person if they do not stay in an identified place must be authorised by express statutory power or by “necessary implication” from express words. This is because common law regards compelling a person to stay in a particular place—for however short a period of time—as a an interreference with individual liberty. Indeed, it represents a coercive act that constitutes—unless willingly complied with—an “imprisonment” or “confinement” at common law. This was recently confirmed by the Supreme Court in Jalloh [2020] UKSC 4. Citing Street on Torts, Baroness Hale stated:

 “False imprisonment involves an act of the defendant which directly and intentionally (or possibly negligently) causes the confinement of the claimant within an area delimited by the defendant.” The essence of imprisonment is being made to stay in a particular place by another person.”

In that case a person who was subject to an immigration curfew and subject to prosecution if he left his residence without reasonable excuse between the hours of 11pm and 7am, was found to have been subject to a false imprisonment, as there had been no statutory basis for the curfew. It did not matter that he had not been guarded and had breached the curfew on numerous occasions.

Such restrictions on liberty can only lawfully be imposed when this is authorised by Parliament by express words or by necessary implication: eg Morris v Beardmore [1981] AC 446, 455; Ex p. Simms [1999] UKHL 33.  In the former case, the House of Lords held that were a statute to authorise the police to step foot onto private land in order to conduct a breathalyser test the statute had to contain express provision to this effect. Rejecting the argument that it was sufficient that there was no express provision in the statute precluding the police from entering private land, Lord Diplock explained,

“with respect, the boot is on the other leg; if Parliament intends to authorise the doing of an act which would constitute a tort actionable at the suit of the person to whom the act is done, this requires express provision in the statute… The presumption is that in the absence of express provision to the contrary Parliament did not intend to authorise tortious conduct: and this presumption, in my view, owes nothing to the European Convention for the Protection of Human Rights and Fundamental Freedoms (1953) (Cmd. 8969).”

In fact the principle applies slightly more broadly than this. In Gedi [2016] EWCA Civ 409 the Court of Appeal found that immigration curfews had to be authorised by express words or necessary implication given their impact on individual liberty whether or not they amounted to an imprisonment at common law, given the substantial interference with individual liberty involved.

During the first lockdown people were confined to a specific and identified place which on ordinary principles of statutory construction must therefore be authorised by primary legislation using express words or at least by necessary implication from express words. The fact that people could leave home if they had a “reasonable excuse” does not affect the analysis as this exception was lifted from the context of immigration curfews and terrorism restriction orders, both of which constitute common law confinements and can only be authorised by clear words (see Jalloh at [6], [28], JJ [2007] UKHL 45 at [6]). The fact that people could (assuming they were not too infirm to do so) leave home for exercise during the first lockdown is also not a material difference: if a night-time curfew—even one in place for a short period of time—requires clear statutory authorisation so must an indefinite stay-at-home requirement of the sort imposed in March. When people were not taking exercising or buying essential items then they had to stay at home. In addition, the lockdown regulations conferred on the police and other designated persons the power to “remove” a person to their home using force. This likewise had to be justified by express words or necessary implication.

Did the PH(CD)A therefore authorise confinement or removal expressly or by necessary implication?  

The PH(CD)A does not however authorise confinement or removal expressly or by necessary implication. Section 45C(1) and (3)(c)(d) of the PH(CD)A permits the Secretary of State to impose a “special restriction or requirement” by regulations in response to a threat to public health. A special restriction or requirement is something that a magistrate can impose under s.45G(2) on a person who is infected or who may be infected with disease.  The special restrictions and requirements that a magistrate can impose are as follows:

·       That a person, P, submit to medical examination.

·       That P be removed to a hospital or other suitable establishment.

·       That P be detained in a hospital or other suitable establishment.

·       That P be kept in isolation or quarantine.

·       That P be disinfected or wear protective clothing.

·       That P provide information.

·       That P’s health be monitored.

·       That P obtain training or advice sessions.

·       That P abstain from working or trading.

·       That P be subject to restrictions on where P goes or with whom P has contact.

Section 45D(3) precludes the Secretary of State from imposing the first four of these special restrictions and requirements in regulations made under s.45C. It is the last listed special restriction or requirement—restrictions on “where P goes or with whom P has contact” —which represents the purported legal basis for the lockdown as well as the restrictions on non-organised gatherings of persons (s.45G(2)(j)).

 If one takes a step back and looks at the relevant provisions of the PH(CD)A in the round it is striking how different the provisions look to the measures that have been imposed over the past months. Most of the provisions—such as those relating to decontamination, health monitoring or wearing protective clothing—contemplate entirely different kinds of public health problems. The extensive powers given to magistrates also represents a very different approach to disease control. To some extent a mismatch is to be expected because the Act is intended to provide a general legal foundation for public health measures which are not focused on specific diseases (a so-called “all hazards” approach). But the specificity of such provisions stands in marked contrast to the absence of any specific power to impose a general lockdown. If that had been contemplated by Parliament one would expect to see something more specific than s.45(2)(j). Lord Anderson QC made precisely this point in a very early comment on the lockdown regulations. He wrote:    

“For such a remarkable limitation of personal freedom to be contemplated by statute, one would have expected to find clear words in section 45G(2): something like “that P be required not to leave the place where P is living, save for specified purposes”.”

The crux of the problem is that the power to impose restrictions on where P goes or with whom P has contact does not authorise confinement expressly nor by necessary implication. The power has evident utility in precluding things that fall short of confinement like preventing people attending sports or social events or attending particular places like gyms or restaurants or pubs. It does not necessarily authorise the far more substantial confinement measures actually imposed and we cannot therefore have confidence that Parliament intended to confer such powers on the Secretary of State.

 Indeed, as mentioned above, section 45D(3) expressly precludes the Secretary of State from issuing regulations imposing “detention”, requiring people to be kept in “isolation” or “quarantine”, or requiring people to submit to medical examination. These can only be imposed by order of a magistrate on persons who are or may be infected. The statute therefore draws a rough but clearly discernible line between confinement and invasive touching, which can only be imposed on potentially infected persons by a magistrate, and other less intrusive measures (this broadly reflects the line drawn by the common law between actions that engage the trespassory torts of assault and false imprisonment and those that do not). It does not make sense for the Secretary of State to be precluded from confining or requiring the isolation of persons, even those who are infected or potentially infected with disease, and for such powers to be conferred on a magistrate (together with added safeguards for the individual), if the Secretary of State has power to confine everybody—or large categories of persons such as whole regions or towns—to their homes. 

There are variants on the vires concern articulated above, such as that since the Secretary of State has only had conferred on him powers that a magistrate could impose therefore, like a magistrate, he can only exercise the powers on infected or potentially infected persons. This appears to be Lord Sumption’s main objection. However, variants on the vires concern boil down to the same essential point, so crisply made by Lord Anderson QC in the quote above, that the extraordinary measures imposed by the Secretary of State are not clearly authorised by Parliament.

It will also be observed that the list of special restrictions and requirements includes a power for a person to be “removed” to a hospital or other establishment where they are being detained. This is one of the four special restrictions and requirements that the Secretary of State is expressly precluded from imposing by regulations (s.45D(3)). There is no other relevant reference to a removal power in the list of special restrictions and requirements or in relation to enforcement (on the contrary, the enforcement power is limited to constables returning persons to isolation or detention imposed by a magistrate: s.45O(4), (5)). Despite this, the lockdown regulations included power for authorised persons to remove people considered to be in breach back to the place where they are living (eg r.8(3)(b)). There is no authority for this in the PH(CD)A and it serves to reinforce the point that the Act does not contemplate confinement by regulations.

Such considerations led Emma Dixon, Rachel Jones and myself to conclude in April this year that, “whilst there is undoubtedly a plausible basis for the most intrusive restrictions in the English Regulations, section 45G(2) is by no means a clear or satisfactory basis for such extraordinary powers” (supra, para 45). Our conclusion was put to the Government by Mr Steve Baker MP in the first debate on the lockdown regulations on 4 May 2020. The Minister responded by confirming that the Government believed it had sufficient powers under the PH(CD)A but no further explanation was provided.  

The analysis of the core lockdown power undoubtedly becomes less straightforward and more contestable in relation to the later iterations of the lockdown regulations in May and June 2020 and in relation to the regulations which underpin the second English lockdown which commenced on 5 November. The ability for people to leave home for recreation in particular changes the character of the measures significantly. Moreover, the power to remove a person to their home is not found in the current lockdown regulations, there is instead only a power to direct a person to return home (Public Health (Coronavirus, Restriction) (England) No. 4) Regulations 2020 (r. 19(3)). Nonetheless, the impact on individual liberty is marked and one that could be expected to be clearly set out in primary legislation. The continued use of the PH(CD)A and the failure to put the social distancing laws on a clear statutory basis therefore remains unsatisfactory.

The Self-Isolation Regulations

That the problem of vires persists is most clearly apparent from The Health Protection (Coronavirus, Restrictions) (Self-Isolation) (England) Regulations 2020, which came into force on 28 September 2020 (“Self-Isolation Regulations”). The regulations require people to remain at home and isolate where they have been notified by a person acting on behalf of the Secretary of State that they, or a child for whom they are responsible, have been in close contact with an infected person. The measures also apply if a person is notified that they or their child has tested positive for Covid-19. The self-isolation measures are very restrictive: people cannot take exercise outside and can only go outside to buy food and drink, “where it is not possible to obtain these in any other manner” (r.2(3)(b)(vi)). It is a criminal offence not to observe these rules.

Prior to the Self-Isolation Regulations, people were asked to self-isolate and comply with guidance similar in terms to what is now legally binding under the regulations. It is of course extremely important that people observe the self-isolation rules. The translation of such rules from guidance into criminal law under the PH(CD)A nonetheless gives rise to a problem. The legal basis for the Self-Isolation Regulations is once again s.45C. Yet, as we have seen, s.45C does not expressly or by necessary implication authorise confinement. And there can be absolutely no doubt that the Self-Isolation Regulations subject people to confinement. Moreover, we have also seen that the PH(CD)A cannot be used to detain people in any establishment, to impose quarantine or to require a person to be kept in isolation (s.45D(3)). This can only be done by a magistrate. Given these express prohibitions, how can the Self-Isolation Regulations, which require people to isolate, possibly be compatible with the PH(CD)A?

If the analysis in the previous section is correct, then the answer is that they can’t. There is however a further dimension in relation to these regulations. If the lockdown regulations are in fact intra vires on the basis that there is an implicit legal basis for lockdown under the Act and lockdown is not expressly precluded, it nonetheless remains the case that the imposition of detention, isolation or quarantine is on the face of the Act expressly precluded from being imposed by regulations made by the Secretary of State.

It might be contended that the Self-Isolation Regulations do not require any person to be “detained” or “kept” in isolation or quarantine because people who are subject to the obligation to isolate are not locked up or guarded. This however is to strain the wording and purpose of s.45C beyond sensible limits. A magistrate’s exercise of the detention, quarantine or isolation powers would not necessarily involve persons being guarded round the clock. Moreover, to read the Act as precluding the Secretary of State from requiring infected or potentially infected persons to isolate or quarantine in a guarded environment but permitting him or her to compel large numbers of persons—whether healthy or infected—to isolate or quarantine on pain of arrest and criminal prosecution would not conform to the obvious statutory purpose of protecting individual liberty. Large numbers of people cannot be locked up or guarded. Therefore, provisions in the Act which are clearly intended to protect individual liberty would be circumvented where confinement is imposed on such a large number of people that they cannot be physically contained.  Yet from the perspective of the individual it makes no difference whether or not they have a policeman or community support officer patrolling their street, or even standing at their front door. They are still required to stay at home by legal command (and might get caught if they do not). The use of the terminology of “self-isolation” rather than simply “isolation” throughout the Self-Isolation Regulations appears designed to suggest that isolation is in some sense voluntarily or not compelled (as had been the case under the previous regime of unenforceable guidance). The terminology of “self-isolation” cannot however conceal the reality that people are being commanded to quarantine or isolate and have no choice in the matter.  

In any event, the Self-Isolation Regulations do in fact confer on police and other “authorised persons” powers to keep people in isolation.  Like the first lockdown regulations, they include power for an authorised person to “remove P to the place where they are Self-Isolating” if they are found outside, using force if necessary (r.10(1), (2)). The vires of the regulations cannot sensibly depend on how many “authorised persons” there are patrolling the streets and checking that people who have been told by the Secretary of State self-isolate are in fact staying at home. Persons subject to the Self-Isolation Regulations do in fact receive regular calls from officials to check they are obeying the rules.

A final point relates to children. Without wanting to descend too far into the detail, it is notable that the Self-Isolation Regulations require adults to keep children for whom they are responsible in isolation (r.2(2)(c)). This is clearly a requirement that children be “kept in quarantine or isolation” which is a measure the Secretary of State cannot impose. (It is not justified by the power of magistrates to order parents to ensure children comply with special restrictions and requirements (s.45G(7)), which is not conferred on the Secretary of State and could not override the express prohibition on the Secretary of State making regulations requiring people to be kept in quarantine or isolation).

The vires problem is not therefore limited to the original lockdown regulations. It is a problem that is of wider significance and at its root is the unsuitability of using the PH(CD)A to address the unique and unforeseen circumstances of the coronavirus pandemic. On any view, therefore, the PH(CD)A is being stretched and strained as a legal basis for the Government’s response to the pandemic.  

The tail that wags the dog?

The vires problem canvassed in this blog may be determined or way or the other by the Court of Appeal in Dolan v Secretary of State for Health [2020] EWHC 1786 (Admin), in which the claimant raises wide-ranging challenges to the coronavirus regulations. One recent blog has suggested that if the Court finds the lockdown to be ultra vires this could result in enormous liability to the Government in damages. There is not scope here to explore this issue in any detail. But three short points can be made in response. The first is that the correct interpretation of the PH(CD)A must be derived from the words used, not the consequences of the way the power has been used—against very large numbers of people—or the timing of the court’s consideration (see e.g. Duport Steels Ltd. v Sirs [1980] 1 WLR 142 at 157-158 (Lord Diplock)). The tail of possible liability cannot be allowed to wag the dog of principle. The second response is that for any person to obtain even nominal damages it would be necessary for them to show that they did not willingly comply with the lockdown or self-isolation rules. Thirdly, were the courts to find that the PH(CD)A was not a satisfactory basis for the most intrusive regulations, Parliament would surely take the opportunity to establish a solid basis for the social distancing regulations for the future and at the same time it would surely address what should happen were there to be any liability for past unlawful conduct. One option would be for Parliament to pass an Act of Indemnity or equivalent. After the Supreme Court in R (Reilly) v Secretary of State for Work and Pensions [2013] UKSC 68 found regulations that had imposed benefits sanctions to be ultra vires, Parliament placed the regime on a proper legal footing and extinguished the liability to repay up to £130m (see Reilly (No. 2) [2016] EWCA Civ 413).  

The issue is not about damages but about the role and authority of Parliament. It is about the importance of two interlocking principles: (a) the principle that Parliament must authorise interferences with individual liberty expressly or by necessary implication, and (b) the principle that laws speak with the same voice in times of calm and times of crisis. Adherence to such principles empowers Parliament and ensures it is not bypassed in an emergency: Parliament can then ensure that properly tailored laws are in place.


Alternatives: the Coronavirus Act 2020 and the Civil Contingencies Act 2004

Why has the Government placed the PH(CD)A at the centre of its response to the coronavirus pandemic? It is possible that when it resorted to the Act in March 2020 at the height of the crisis officials were simply following a pre-programmed equation: pandemic = PH(CD)A. But the legal risks must have been apparent. It is also puzzling that at precisely the same time as the lockdown regulations were being prepared the Government was pushing the Coronavirus Bill through Parliament on an expedited basis (it was enacted on 25 March 2020). The Bill was intended to provide a bespoke regime for addressing the coronavirus pandemic and many provisions, such as those now found in the Coronavirus Act 2020 (“CA”), Schs. 21 and 22 are directed at social distancing. Schedule 21, in particular, sets out powers for the Secretary of State to declare a “transmission control period” where the spread of coronavirus constitutes a serious and imminent threat to the public. This would allow persons to be screened and “to remain at a specified place in isolation from others for a specified period”. These powers are directed at infected or potentially infected persons and have never been used (and appear of little use). Why did the Government seek such bespoke but probably useless powers but not seek the wider powers that would have provided a clear and express legal basis for the nationwide lockdown?

According to the Government, the Coronavirus Bill included off-the-shelf provisions taken from a “draft Pandemic Influenza Bill”, which the Government had prepared as part of its pandemic preparedness planning. The measures in Schs.21 and 22 may represent measures that the Government had predicted it would need to enforce social distancing to meet a flu pandemic and which it had prepared in draft. It is possible that that the Government considered that it did not have time to table amendments after the decision to lockdown had been taken at some point over the weekend of 21-22 March 2020, a decision which, as we have seen, departed from the Government’s pandemic preparedness plan. The CA after all was enacted on an extremely expedited timetable the following week before Parliament went into recess on 25 March 2020.

We do not know why the Government did not seek to amend the CA to include lockdown and useful social distancing powers. It is notable however that at the time the legislation was proceeding through Parliament the lockdown was already in place. In his televised address on 23 March 2020 the Prime Minister announced that everybody had to stay at home, a message that was backed-up by published guidance and a text message instruction sent to everybody’s mobile telephone. But despite references to the police having power to enforce the lockdown—and reports of police seeking to do so— no regulations had been made. The first few days of lockdown were premised on a gigantic bluff. It is possible that tabling amendments to the CA to introduce lockdown powers might have been considered undesirable in this context, as drawing attention not only to the absence of regulations but the absence of any legal basis at all for the lockdown. The lockdown regulations themselves were made and came into effect at 1pm on 26 March after Parliament had gone into recess the previous evening. They were laid before an empty Parliament at 2.30pm the same day. The legal hole in the lockdown was thereby quietly plastered over before most people had identified the problem. All of this may very well have simply been a reflection of muddle, urgency and the chaos of the period, but foregoing the opportunity to make specific provision for lockdown powers in the CA meant that the Government committed itself to the PH(CD)A.   

As explained at the outset of this paper, there was also a second possible alternative to using the PH(CD)A. Concerns about the decision not to use the CCA were raised by several Members in debate on the Coronavirus Bill. It was suggested that the CCA would have given greater parliamentary oversight and greater judicial supervision than the provisions in the Bill allowed for. The Government’s response was that the CCA should only be resorted to if the Government does not have time to bring forward primary legislation (Penny Mordaunt MP, HC 23 March 2020, vol. 674 col. 132, also Chris Bryant MP, col. 177).  Michael Gove MP told the PACAC on 29 April 2020 (Q216) that the Government had clear advice that the CCA could only  be used “when you have an unexpected bolt from the blue”, leaving no time to enact new laws. That however did not explain why the PH(CD)A had been used in preference to the CCA. The explanation for that was provided by the Prime Minister on 4 November 2020. He stated:

“…the Civil Contingencies Act has a strict test known as the triple lock that must be met before emergency regulations under the Act can be made. One of these tests is that there must not be existing powers elsewhere, and the Public Health Act 1984 offers clear powers to impose restrictions on public health grounds. That is why, despite his very useful suggestion, the Public Health Act is the more appropriate route.” (HC vol. 683, col. 45).

The Government’s position is therefore that the CCA cannot be used where it is unnecessary; and it will be unnecessary where either (a) existing powers can be used, or (b) new powers can be enacted. Let us consider those propositions.

Section 21 of the CCA sets out three conditions for making regulations under the Act: the so-called “triple lock”. These are: that an emergency has occurred or is about to occur, that it is necessary to make provision and that the need for provision to be made is urgent. The necessity condition reads as follows:

“(3) The second condition is that it is necessary to make provision for the purpose of preventing, controlling or mitigating an aspect or effect of the emergency.”

In the abstract this makes a good deal of sense and reflects a common feature of emergency regimes. It is designed to restrict access to the very broad executive powers under the CCA, which include powers to override primary legislation, save where no other powers are available.

The availability of existing powers to make regulations is expressly addressed in s.21(6)(c) of the CCA. This provides that it is only necessary to utilise the CCA where, “the provision might be insufficiently effective if made under existing legislation”. In accordance with s.21 of the CCA, the Government was therefore required in law to utilise the powers in the PH(CD)A unless its provisions might have been insufficient for the powers that it introduced. For the reasons that have been explained above, it was the case that the powers contained in the PH(CD)A might have been insufficient: they were pregnant with legal risk and not obviously designed for the measures that the Government decided it needed to impose.

But what about the Government’s other contention: that the CCA cannot be used where it is possible to introduce new legislation? The Government suggested that the CCA could only be used where there was a “bolt from the blue” leaving no time to legislate. This analysis is however wrong in law.  Whilst it is generally desirable that bespoke legislation be utilised in preference to the CCA the triple lock contained in the CCA only applies to existing laws and powers. This is important because the position of the Government during the passage of the CA suggested that the CCA could not be used. That in turn suggested to Members of Parliament and Peers that there was no alternative but to enact the Bill which was before them.  In fact, the CCA could have been utilised.

The CCA would have brought with it additional political safeguards and enhanced parliamentary oversight to those contained in the PH(CD)A. The CCA requires regulations to be approved by Parliament within seven days and renewed by Parliament every thirty days (ss.26(1)(a), 27(1)(b)). Moreover, Parliament would have been able to amend the regulations (s.27(3)). The power to amend regulations would have been particularly valuable given the morphing nature of the coronavirus regulations and the fact that, whilst the core aspects of the regulations would undoubtedly have been approved by Parliament, certain non-core but nonetheless important elements of the rules have been controversial and other aspects have lacked any real scrutiny. The breadth of the exceptions to the stay-at-home requirement, the inclusion of young children in the “rule of six”, the broad nature of police enforcement powers and the very high level of the Fixed Penalty Notices for example are topics that Parliament might well have scrutinised with rigour and possibly changed if it had not been presented with the regulations on a take-it-or-leave-it basis. The presentation of regulations on a take-it-or-leave-it basis also provides little scope for those MPs and Peers (gradually increasing, it seems, in number) who are more deeply sceptical of the Government’s social-distancing strategy to probe and challenge the Government’s position.

Whilst regulations imposing special restrictions and requirements made under s.45C are usually subject to the draft affirmative process, which requires prior approval of each House, the Government has been able to exploit the urgency provisions contained in s.45R of the PH(CD)A to allow the regulations to be subject to later approval by Parliament. This, combined with the Government’s control over Parliamentary time, meant that that parliamentary scrutiny of the lockdown regulations and the amendments made to them routinely occurred a long time after the regulations were made, always after they had already been significantly amended, without any realistic possibility of them not being approved and with no power for Parliament to amend them.  The belated, restricted and stale nature of parliamentary scrutiny of the regulations made under the PH(CD) is addressed in more detail elsewhere (Hickman, “Abracadabra law-making”).

The CCA therefore does have advantages in terms of enhanced parliamentary scrutiny. But it is certainly not a panacea. The powers that it affords to the Government are extremely broad, allowing the Government to make provisions “of any kind that could be made by Act of Parliament or by the exercise of Royal Prerogative” (s.22(3)), including disapplying or modifying primary legislation (s.22(3)(j)). There is no detailed statutory framework such as one finds in the PH(CD)A or in Schedules 21 and 22 of the CA, to limit and control the use of the powers set out therein. Moreover, whilst the CCA could have been used to enact the provisions in the lockdown regulations (and other coronavirus regulations) where there was a risk that these exceeded the terms of the PH(CD)A, other restrictions—such as prohibitions on the holding of events—can be imposed under the PH(CD)A and so could not be the subject of regulations under the CCA. These other provisions would not satisfy the necessity criterion in s.21(3). Therefore if the CCA had been used, the Government would have had one foot in the CCA regime and one foot outside it. This would have been unsatisfactory and would have led to legal complexities. Legal arguments would surely have arisen over which measures were “necessary” under s.21(3). Whilst s.21(3)’s role as a gatekeeper provision to the CCA makes sense in theory, one of the lessons of the coronavirus pandemic is that it is problematic in practice.

The CCA is therefore not a panacea or even close to being one. It was at best a stop-gap solution until primary legislation tailored to the crisis was enacted. The principal significance of the CCA to the current debate is that it demonstrates that Parliament can have, and expects to have, a far greater role during a crisis than it has had under the PH(CD)A. In any legislation that Parliament enacted or which it comes to enact in the future, Parliament should ensure that it is afforded a similar level of scrutiny of coronavirus measures as it would have under the CCA. In my view, it would be appropriate for Parliament to have the following powers in respect of coronavirus regulations:

·       The ability to approve and amend the regulations before they come into effect (or at least within a short period of time, such as seven days, after they have come into effect). 

·       The ability to debate and approve the continuation of the regulations after thirty days, again accompanied by a right of amendment. 

·       Time should be made available for debate and this should exceed the normal time limits for debates on statutory instruments.

Drawing the threads together, it is clear that the Government in March 2020 was in an unenviable position. It faced a national emergency. It had a choice of three legal paths: the PH(CD)A, the CCA or the CA. It chose to follow the PH(CD)A path and place that Act at the centre of its response to the pandemic. It has been argued in this paper that this was the wrong choice because the Act is not tailored or suitable either to the powers required to meet the coronavirus pandemic or the level or form of accountability that Parliament needed. But whichever path the Government took in March 2020 it should have returned to Parliament in the Summer and introduced a Bill that set out bespoke powers for addressing the pandemic. After all, the Government was making clear at the time that society had to expect to adjust to a “new normal”. The country needed a “new normal” pandemic law.



This paper has advanced two main claims. First, that the consequence of the PH(CD)A being used as the central pillar of the Government’s response to the coronavirus pandemic has been that the laws that have intruded most deeply into individual liberty have lacked a clear or satisfactory legal basis and also that parliamentary accountability has been unduly restricted. Secondly, it has been suggested that although the CCA may have had a role to play during the initial emergency, the principal misstep from a constitutional perspective has been the failure to enact primary legislation setting out a bespoke basis for social distancing laws. Now that the emergency period has passed and we are confronted with an ongoing crisis, we know what powers are required and we also have a far better idea of the degree and form of scrutiny that Parliament should exercise over social distancing laws. Such matters should be set out in primary legislation. After all, Parliament’s first and foremost function is to enact the laws that are appropriate to govern society. It does not matter whether this takes the form of amendment to the PH(CD)A, amendment to the CA or a standalone Coronavirus II Act. Bespoke primary legislation would also have the advantage of giving the response to the pandemic a firmer democratic basis and therefore afford it greater legitimacy.

The political reality however is that the Government is unlikely to seek a legislative basis for coronavirus powers unless it is effectively forced to do so either by Parliament itself or by the courts. That however does not mean that it is not the correct course to follow.

 Tom Hickman QC, Barrister and Professor at UCL

This paper develops points made by the author to the Constitutional Affairs Committee in oral evidence on 18 November 2020

The author is grateful to Professor Gavin Phillipson, Professor Jeff King, Abe Chauhan and Robert Craig for valuable comments

+44 (0)207 5831770